Sabcs 2014 Allabstracts

2014 San Antonio Breast Cancer Symposium
Publication Number: S1-01

Title: TransCONFIRM: The correlative analysis of breast tumors from patients with advanced hormone receptor positive disease
identifies a genetic signature associated with decreased benefit from single agent fulvestrant
Rinath M Jeselsohn1, William T Barry1, Jin Zhao1, Gilles Buchwalter1, Cristina Guarducci2, Ilenia Migliaccio2, Chiara Biagioni2,3,
Martina Bonechi2, Naomi Laing4, Yuri Rukazenkov4, Eric P Winer1, Myles Brown1, Angelo Di Leo2,3 and Luca Malorni2,3.
1
Dana-Farber Cancer Institute, Boston, MA; 2Hospital of Prato, Istituto Toscano Tumori, Prato, Italy; 3"Sandro Pitigliani" Medical
Oncology Unit, IstitutoToscano Tumori, Prato, Italy and 4AstraZeneca Pharmaceuticals, Macclesfield, United Kingdom.
Body: Introduction: Several multi-gene expression based assays have been developed to assess the prognosis and predict
response to endocrine treatments in early stage hormone receptor positive (HR+) breast cancer. Although a significant number of
patients with metastatic ER+ disease will not respond to endocrine treatments, molecular assays to predict response in this
setting are limited. In addition, tissue specimens of metastatic lesions for molecular studies are not always available. In this study
we sought to identify a molecular profile in the primary tumors of patients who developed disease recurrence that could predict
response to endocrine treatment in metastatic disease.
Methods: We used the primary breast tumor samples from a subgroup of patients participating in the randomized phase III
CONFIRM trial, which compared 500mg versus 250mg of fulvestrant in post-menopausal women with HR+ advanced breast
cancer. Formalin- fixed paraffin embedded tumors were collected from 130 of the participants and were centrally reviewed for ER,
PR, HER2 and Ki67. RNA was sufficient for gene expression profiling in 112 of the cases using the NuGEN Ovation FFPE WTA
System and Affymetrix HTA 2.0 GeneChip. The majority of the patients in this analysis developed metastatic disease during
adjuvant endocrine treatment (N=55) or had de-novo metastatic disease (N=39) versus relapse after adjuvant treatment (N=18).
The association between gene expression and progression free survival (PFS) was investigated using a multivariate Cox
proportional hazard model adjusting for statistically significant clinicopatholgical factors. In addition we performed pathway-level
analysis and evaluated the PAM50 subtype predictor and Risk of Relapse (ROR) score.
Results: The median PFS was 8 months in our cohort. HER2 level by immunohistochemistry above 1+, high PR level, defined as
Allred score of above 6, and Ki67 of above 50% were significantly associated with PFS and were included in the multivariate
model. Dose of fulvestrant was not associated with PFS in this cohort. We identified a signature of 25 genes that is inversely
associated with PFS on fulvestrant treatment (FDR 20%). When compared to other published datasets of breast cancer tumors,
these genes are enriched in tumors with poor outcome and triple negative cancers. Pathway analysis revealed an association
between activation of the EGFR pathway and decreased PFS (P=0.01). PAM50 subtypes varied with the luminal subtype being
the most common (65%) and were generally concordant with the clinical subtype. However, we did not detect a significant trend
between PAM50 subtype or ROR score and PFS or overall survival.
Conclusions: In this cohort of patients with early and de-novo metastatic disease we identified a gene signature in the primary
tumors that is associated with decreased response to fulvestrant treatment in metastatic disease. This signature warrants further
validation to determine its predictive value and potential to assist in treatment decision making for patients with HR+ metastatic
disease.

Title: Prognostic effects of gene mutation in estrogen receptor positive breast cancer
Obi L Griffith1, Malachi Griffith1, Jingqin Luo1, Jasreet Hundall1, Christopher A Miller1, David E Larson1, Robert Fulton1, Richard K
Wilson1, Shuzhen Liu2, Samuel Leung2, Torsten O Nielsen2, Elaine R Mardis1 and Matthew J Ellis1. 1Washington University School
of Medicine, St Louis, MO and 2University of British Columbia, Vancouver, BC, Canada.
Body: Background: Relationships between recurrent somatic mutations and outcome in estrogen receptor positive (ER+) breast
cancer has not been extensively studied as the original discovery efforts were from either heterogeneously treated patients or
follow up was too brief. Targeted massively parallel sequencing (MPS) analysis was therefore conducted on DNA extracted from
archived formalin-fixed breast primaries from a cohort of over 600 patents from British Columbia treated with five years of
adjuvant tamoxifen monotherapy and followed for over 10 years (Nielsen et al CCR 16:5222, 2010).
Methods: Genes were selected for targeted sequencing by meta-analysis of five large-scale breast cancer sequencing studies
and manual review of breast cancer literature. In total 83 genes were identified and 3286 probes were designed to tile across all
known exons. Minimum starting input DNA was 50ng (mean=189.1ng). Illumina sequencing libraries were constructed, indexed,
pooled, and enriched for target sequences by hybrid-capture followed by paired-end 100bp reads. The Genome Modeling System
was used to perform single-tumor somatic variant prediction. Variant calls were filtered to include only targeted regions and
exclude variants with global mutant allele frequencies greater than 0.1% in 1000 genomes or NHLBI exome datasets.
Kaplan-Meier analysis and multivariable analysis (clinical features and intrinsic subtype by qPCR) was performed for
breast-cancer-specific and relapse free survival.
Results: A total of 638 samples met minimum quality controls of 80% targeted space covered at 20X or greater. On average each
sample had 332M of aligned bases and a mean coverage of 134.3X. In total 7,159 variants were identified including 3,696
missense, 494 nonsense, and 1,047 frameshift insertions or deletions. Preliminary results indicate significant associations
between mutation status and improved survival for PIK3CA, ARID1B, ERBB3, MAP3K1 and GATA3 or worse survival for PTEN,
DDR1, TP53 and JAK2. Five Y537N/C, two E380Q and 5 potentially novel ligand-binding-domain mutations were identified in
ESR1. Such mutations were recently reported to be associated with resistance to hormone therapy but were discovered here in
as much as 1.9% of pre-treatment samples. Analysis will be presented regarding the use of relapse events to differentiate
passenger from driver mutations.
Conclusion. Multiple recurrently mutated genes have both positive and negative associations with prognosis in tamoxifen
montherapy treated breast cancer populations. Associations with poor outcome suggest that PTEN, DDR1, and JAK2 are high
priorities for pharmacological interventions.
Table 1: Mutations and survival in ER+ breast cancer
Gene
P-Value
Hazard Ratio
PTEN
0.002
2.11
PIK3CA
0.007
0.69
DDR1
0.008
2.41
ARID1B
0.01
0.57
ERBB3
0.02
0.36
MAP3K1
0.01
0.5
TP53
0.04
1.48
GATA3
0.04
0.44
JAK2
0.05
2.1

Title: Identification of base pair mutations and structural rearrangements acquired in breast cancer metastases including a novel
hyperactive ESR1-DAB2 fusion gene specifically in hormone-resistant recurrence
Ryan J Hartmaier1,2, Shannon L Puhalla1, Steffi Oesterreich1,2, Amir Bahreini1,3, Nancy E Davidson1,2, Adam M Brufsky1 and Adrian
V Lee1,2,3. 1Women's Cancer Research Center, University of Pittsburgh Cancer Institute & Magee Women's Research Center,
Pittsburgh, PA; 2University of Pittsburgh, Pittsburgh, PA and 3University of Pittsburgh, Pittsburgh, PA.
Body: DNA structural variations (SVs) are a major source of genetic instability in cancer, but they remain understudied.
Large-insert mate-pair sequencing (MPS) is a powerful method designed to detect SVs, even in highly repetitive regions. Using
MPS and other methods, we performed a comprehensive analysis of genomic alterations in breast cancer progression.
Matched primary/recurrent frozen tumor samples from 6 patients, including two patients from our rapid autopsy program with
multiple metastatic tissues (20 total samples; average 5.5 years to recurrence) were examined by multiple large-insert library (3-5,
5-8, 8-12kb) MPS to identify metastatic acquired SVs. This was supplemented with RNAseq (n=15), whole exome sequencing
(n=18; 75x), whole genome sequencing (n=3; 40-65x), and SNP arrays.
A relatively small fraction (10%) of somatic single nucleotide variants (SNVs) in the primary tumor were identified in matched
metastatic samples, and the majority of metastatic SNVs were not found in the matched primary tumor. This indicates that a rare
sub-clone colonizes the metastatic site and evolves extensively before becoming clinically evident. For example, in one patient
with an ER+ tumor who initially declined anti-estrogen therapy, the recently described ESR1 Y537S mutation was not present in
the primary tumor or in metastatic disease 5 years later. However, after extensive anti-estrogen treatment for metastatic disease,
the mutation was identified at rapid autopsy, indicating that this mutation can be acquired even after initial metastatic spread.
We observed extensive patient-to-patient variability in the number and types of SVs. In general, the overall pattern of SVs was
remarkably similar between matched primary and metastatic samples, however, we identified a number of metastatic specific SVs
that likely contribute to disease progression. Specifically, in one patient with an ER+ primary tumor treated with adjuvant
Tamoxifen, we identified a novel fusion gene between ESR1 (estrogen receptor-, ER) and DAB2 (disabled-2) only in a lymph
node recurrence. RT-PCR and western blot analysis confirmed that the fusion RNA/protein was expressed/translated only in the
recurrent disease. The fusion retains the DNA-binding domain (DBD) and hinge region of ER while the ligand-binding domain
(LBD) is replaced with the majority of DAB2. We hypothesized that this is a functional genetic alteration conferring
ligand-independent ER-mediated signaling and growth. Confirming this, in vitro ERE-Tk-luc reporter assays showed that the
ESR1-DAB2 fusion has ligand-independent activity that is 13-290x higher than wild-type ER. Chromatin immunoprecipitation
assays in metastatic tissue from tumors with mutant ER show strong enrichment for ER at classical ER target genes. We are
currently assessing the genome-wide binding of ESR1-DAB2 and the functional contribution of DAB2 to the fusion protein.
This study represents the most comprehensive analysis to date of genomic changes in breast cancer progression and indicates
extensive changes occur during metastatic spread. A number of acquired changes likely represent therapeutically targetable
metastatic dependencies.

Title: Exome sequencing of post-menopausal ER+ breast cancer (BC) treated pre-surgically with aromatase inhibitors (AIs) in the
POETIC trial (CRUK/07/015)
Pascal Gellert1, Corrinne V Segal1, Qiong Gao1, Tiandao Li2, Christopher A Miller2, Elaine Mardis2, Lesley-Ann Martin1,
Christopher Holcombe3, Anthony Skene4, Judith Bliss1, John Robertson6, Ian Smith5, Mitch Dowsett5 and POETIC Trial
Management Group and Trialists7. 1Institute of Cancer Research, London, United Kingdom; 2Genome Institute at Washington
University, St Louis, MO; 3Royal Liverpool University Hospital, Liverpool, United Kingdom; 4Royal Bournemouth Hospital,
Bournemouth, United Kingdom; 5Royal Marsden Hospital, London, United Kingdom; 6University of Nottingham, Nottingham,
United Kingdom and 7POETIC Trial Management Group and Trialists.
Body: Aims
1. To determine the variability of mutational profiles and sub-clonality in core-cut biopsies from ER+ BC and the impact of
2-weeks AI therapy on these.
2. To identify mutations or patterns of mutations associated with poor anti-proliferative response to AI treatment.
Background
DNA alterations may lead to de novo and acquired resistance to medical therapies including AIs. Assessing this requires single
time-point or sequential sampling usually with core-cuts but there is little information on their ability to represent mutational
profiles or sub-clonal structure. We studied this in paired biopsies from ER+ BC primaries in 60 selected postmenopausal patients
from the Peri-Operative Endocrine Therapy for Individualising Care (POETIC) trial (CRUK/07/015) before and after 2-weeks
non-steroidal AI or no AI (randomised 2:1).
Methods
DNA was extracted from RNAlater-preserved diagnostic and surgical 14-gauge core-cut samples and peripheral blood from 20 no
AI (Control) and 40 AI-treated patients (15 poor and 25 good Ki67-responders [PR and GR, respectively]). Patients with low ER+
BC or unsuppressed estradiol on treatment were not considered. Exome sequencing (Illumina HiSeq 2000) achieved >60%
coverage across the exome at 15x depth. Variants were validated by re-sequencing (median >100x) together with 79 genes of
interest curated from COSMIC and selected publications. Statistically significant genes (SMGs) were determined using MuSiC.
Sub-clonality was analysed by SciClone.
Results
Good quality exomes were obtained on 102 samples including 44 pairs (control n=14; PR n=10; GR n=20). There were 5684
mutations (including 3616 missense and 1322 silent) affecting 3261 genes. SMGs in this series were PIK3CA (35.3%), TP53
(27.5%), CDH1 (13.8%), HEATR7B2 (8.8%), GATA3 (5.9%), CENPF (5.9%), MAP3K1 (5.9%), MAP2K4 (4.9%), HTR1A (2.9%)
and C22orf23 (1%). PR had more mutations than GR (median 65 vs 36, p=0.04). More PR than GR were HER2+ (5/14 vs 1/24,
p=0.019) and/or TP53-mutated (5/10 vs 3/20, p=0.08) but similar proportions were PIK3CA-mutated. The correlation of diagnostic
vs surgical variant allele frequencies was strong for the control (r=0.75) and treated (r=0.89) groups (for treated GR r=0.83; for
treated PR r=0.65). In the treated group there were fewer mutations at surgery vs diagnosis (p<0.026). PIK3CA and TP53
mutation status was identical between the paired samples in 41/44 and 40/44 cases; less frequently mutated genes showed lower
concordance. SciClone plots to infer sub-clonality were possible in 37 pairs; for 8 pairs (22%) there was clear evidence of at least
one sub-clone being present in only one core-cut sample.
Conclusion
This is the largest reported study of exome reproducibility in ER+ BC for mutation profiles based on core-cut biopsy. Multiple
sub-clones are identifiable in ER+ primary BC. In c.20% tumours, a single core-cut does not allow inference of all sub-clonal
populations, probably due to spatial heterogeneity. TP53 mutations but not PIK3CA mutations are associated with PR. Large
numbers of BC will be needed to identify any associations of lower frequency mutations with resistance. A trend to fewer
mutations after just 2 weeks AI needs confirmation.

Title: In-depth genomic analysis of ER+ breast cancers during development of endocrine resistance
J Michael Dixon1, Arran K Turnbull1, Chris Fan2, Joel S Parker2, Xiaping He2, Laura Arthur1, Carlos Martinez-Perez1, Lorna
Renshaw1 and Charles Perou2. 1University of Edinburgh, Edinburgh, United Kingdom and 2Comprehensive Cancer Centre,
Chapel Hill, NC.
Body: Background: Aromatase inhibitors (AIs) have an established role in the treatment of estrogen receptor alpha positive
(ER+) post-menopausal breast cancer. Response rates are only 50-70% in the neoadjuvant setting and up to 40-50% of all
adjuvantly treated patients will eventually relapse. Mutations in certain genes have been previously shown to confer resistance to
therapy, and molecular subtype has associations with poor outcome (i.e. LumB and HER2E). In order to improve the outcomes of
non-responders or patients who become resistant to endocrine treatment, the identification of key mutations, and their interaction
with subtype, is crucial. Dynamic profiling of the same tumour demonstrating de novo or developing resistance after responding to
one or more lines of endocrine treatment in the neoadjuvant setting provides a unique opportunity to identify such genomic
changes.
Methods: This series is unique in that it includes 17 post-menopausal women with ER+ breast cancer treated with neoadjuvant
letrozole. 13 of these patients progressed on treatment or initially responded to treatment and then developed acquired resistance
and 4 responded well. Dynamic clinical response was assessed for each patient using periodic 3D ultrasound measurements
performed during treatment. Fresh tissue was taken before treatment and when the tumor was resistant to treatment (4 patients
had 2 biopsies, 9 patients had 3 and 4 patients had 4 available biopsies taken). RNA and DNA were extracted from tumour and
normal DNA obtained from either matched blood or normal lymphatic tissue. In total, 51 tumour samples were available and have
completed RNA-Seq, with exome sequencing shortly to be completed.
Results: From the RNA-seq data, the intrinsic subtype distribution was 9 LumA, 7 LumB, and 1 HER2-enriched; when stratified
according to response, the "progressors" were 7 LumB, 5 LumA and 1 HER2-enriched, while "responders" were 3 LumA and 1
LumB. When examined in an unsupervised hierarchical clustering analysis along with >800 TCGA Breast tumor samples, 13/17
patients had all of their samples grouped immediately together, suggesting that the overall tumor phenotype was maintained.
Interestingly, the most dominant change in gene expression was the observation that there were 5 "progressor" patients where
the pre-treatment sample was LumB and all subsequent samples were LumA; we only observed one instance of a patient starting
as LumA and changing to a LumB, who was also labelled as a "progressor". Full exome sequencing is underway and these
results will be presented.
Conclusion: Genomic analysis of progression suggests that an apparent "subtype shift" appears in a number of patients where a
shift to LumA is seen; this apparent change may be reflective of decreased proliferation rates caused by therapy, or the
acquisition of a true LumA phenotype. We cannot differentiate between these two hypotheses at this time, but expect that the
exome sequencing will help to differentiate between these two hypotheses given the large number of mutations and copy number
alterations that can differentiate between LumA vs. LumB.

Title: Stromal tumor-infiltrating lymphocytes(S-TILs): In the alliance N9831 trial S-TILs are associated with chemotherapy benefit
but not associated with trastuzumab benefit
Edith A Perez1, Karla V Ballman2, S Keith Anderson2, E Aubrey Thompson1, Sunil S Badve3, Helen Bailey4 and Frederick L
Baehner4,5. 1Mayo Clinic, Jacksonville, FL; 2Mayo Clinic, Rochester, MN; 3Indiana University, Indianapoils, IN; 4Genomic Health
Inc, Redwood City, CA and 5University of California, San Francisco, CA.
Body: Background: Tumor-infiltrating lymphocytes (TILs) at diagnosis are reported to be prognostic in triple-negative breast
cancer (BC). Analysis of a small subset of 209 HER2+ patients (pts) with 49 events concluded that higher levels of S-TILs are
associated with increased trastuzumab benefit (Loi, 2014). Here we report the largest study to date evaluating S-TILs and their
prognostic and predictive association with clinical outcome in N9831 pts treated with either chemotherapy or chemotherapy plus
trastuzumab.
Methods: Samples assessed were from primary tumors of pts on N9831 arm A (standard ACT chemotherapy) and arm C
(concurrent chemotherapy with trastuzumab) (Perez, 2011). S-TILs were evaluated on H&E whole tumor slides by a single
pathologist with 10% of cases read by two pathologists in tandem. The percent of stromal lymphocytic infiltrates (S-TILs) was
quantitated in deciles; 60% S-TILs was used for the categorical cutoff (Denkert, 2010). The association between S-TILs,
treatment (tx) and recurrence-free survival (RFS) was studied and the interaction between S-TILs, trastuzumab benefit and RFS
was calculated.
Results: 489 pts from arm A (chemo) and 456 pts from arm C (chemo with trastuzumab) were assessed and were similar to pts in
the overall trial; all had RFS information and a median follow-up of 4.4yr. Tumors from 54% of pts in arms A and C were HR+;
14% were node-negative. Tumors with high S-TILs were more likely to be hormone receptor-negative (p< 0.0001). In
multivariable analyses including nodal status, hormone receptor status, tx arm, tumor size, tumor grade, and age, 60% S-TILs
was significantly associated with RFS (HR 0.20; 95%CI 0.0640.65, p=0.007) in arm A but not in arm C (HR 1.1; 95%CI
0.422.8, p=0.87); the interaction term of arm and 60% S-TILs was significant (p=0.042). Semi-continuous deciles were
associated with RFS in arm A (p<0.0002) but not in C (p=0.37). Hormone receptor status was an independent prognostic factor in
arm A (HR 0.61; 95%CI 0.41-0.93, p=0.02) but not in C (HR 0.79; 95%CI 0.44-1.41, p=0.42). In arm A the 10yr Kaplan-Meier
estimates for RFS were 90.9% and 64.5% for high S-TILs and low S-TILs pts, respectively (HR 0.23; 95%CI 0.0730.73,
p=0.013). In arm C the 10yr Kaplan-Meier estimates for RFS were 80.0% and 80.1% for high S-TILs and low S-TILs pts,
respectively (HR 1.26; 95%CI 0.53.2, p=0.63).
Arm A (N=489)
Variable
Arm C (N=456)
HR (95% CI)
p-value
HR (95% CI)
p-value
Lymph node(-)
1.00 (ref)
<0.0001
1.00 (ref)
0.013
1-3+
0.58 (0.25, 1.35)
2.99 (0.40, 22.70)
4-9
1.46 (0.64, 3.35)
3.29 (0.42, 25.80)
10+
2.42 (1.04, 5.64)
8.22 (1.06, 63.60)
Nodal status
HR status
negative
1.00 (ref)
positive
0.61 (0.41, 0.93)
0.0198
1.00 (ref)
0.42
0.79 (0.44, 1.41)
S-TILs status
<60%
1.00 (ref)
60%
0.2 (0.06, 0.65)
Tumor grade
0.007
1.00 (ref)
1.08 (0.42, 2.79)
0.87
Grade 1 or 2
1.00 (ref)
0.50
1.00 (ref)
0.51
Grade 3
1.18 (0.73, 1.91)
Tumor size
1.08 (0.98, 1.19)
0.12
1.00 (0.93, 1.07)
0.92
Age
0.99 (0.97, 1.00)
0.13
1 (0.98, 1.03)
0.87
1.24 (0.65, 2.39)
Conclusions: In exploratory analyses from this subset HER2+ population from N9831, S-TILs were associated with RFS in
patients treated with chemotherapy alone, and were not shown to be associated with RFS in patients treated with chemotherapy
plus trastuzumab.

Title: HER2 T cell dependent bispecific antibody (HER2-TDB) for treatment of HER2 positive breast cancer
Teemu T Junttila1, Ji Li1, Jennifer Johnston1, Maria Hristopoulos1, Robyn Clark1, Diego Ellerman1, Bu-Er Wang1, Yijin Li1, Mary
Mathieu1, Guangmin Li1, Judy Young1, Elizabeth Luis1, Gail Lewis Phillips1, Eric Stefanich1, Cristoph Spiess1, Andrew Polson1,
Bryan Irving1, Justin M Scheer1, Melissa R Junttila1, Mark S Dennis1, Robert Kelley1, Klara Totpal1 and Allen Ebens1. 1Genentech,
San Francisco, CA.
Body: Based on recent clinical success of tumor immunotherapies that block immune suppressive mechanisms to restore T cell
function, there is a profound interest in the clinical development of T cell targeted therapies. We have produced a
trastuzumab-based HER2 T cell dependent bispecific antibody (HER2-TDB) that conditionally activates T cells resulting in lysis of
HER2 expressing cancer cells at low picomolar concentrations. Due to its unique mechanism of action, which is unrelated to
HER2 signaling or sensitivity to chemotherapeutic agents, HER2-TDB can eliminate cells refractory to currently approved HER2
therapies. The potent anti-tumor activity of HER2-TDB was demonstrated using four model systems including MMTV-huHER2
and huCD3 transgenic mice. We demonstrate inhibitory effect of PD-L1 expression on the activity of bispecific T cell recruiting
antibodies. This resistance mechanism is reversed by anti-PD-L1 treatment and combination of HER2-TDB with anti-PD-L1
immune therapy resulted in enhanced inhibition of tumor growth, increased response rates and durable responses.
Significance:
This report presents a new immunotherapy for HER2+ breast cancer with an alternative, extremely potent mechanism of action
that is effective in cells resistant to current HER2 targeted therapies. Several significant advances are provided to bispecific T cell
recruiting antibodies: we characterize a critical resistance mechanism, a potential diagnostic marker, a novel transgenic efficacy
model and significantly improve the drug-like properties by using technology based on full length antibodies with natural
architecture. Finally we demonstrate the benefit of combining two immune therapies: direct polyclonal recruitment of T cell activity
together with inhibiting the T cell suppressive PD-1/PD-L1 signaling results in enhanced and durable long term responses.

Title: Reduced tumor lymphocytic infiltration in the residual disease (RD) of post-neoadjuvant chemotherapy (NAC)
triple-negative breast cancers (TNBC) is associated with Ras/MAPK activation and poorer survival
Justin M Balko6, Carsten Denkert2, Roberto Salgado3, Martin O'Hely4, Peter Savas1, Paul A Beavis1, Phil K Darcy1, Susan
Combs5, David L Rimm5, Jennifer M Giltnane6, Monica V Estrada6, Melinda E Sanders6, Rebecca S Cook6, Kai Wang7, Vincent A
Miller7, Phillip J Stephens7, Roman Yelensky7, Joseph A Pinto8, Franco Doimi8, Henry Gomez9, Carlos L Arteaga6 and Sherene
Loi1. 1Peter MacCallum Cancer Center, Sydney, Australia; 2Charit University, Berlin, Germany; 3Institute Jules Bordet, Brussels,
Belgium; 4WEHI University, Melbourne, Australia; 5Yale University, New Haven, CT; 6Vanderbilt University, Nashville, TN;
7
Foundation Medicine, Cambridge, MA; 8Oncosalud, Lima, Peru and 9Instituto Nacional de Enfermedades Neoplsicas (INEN),
Lima, Peru.
Body: Backgound: Tumor-infiltrating lymphocytes (TILs) are associated with improved prognosis in TNBCs, with several
retrospective analyses demonstrating that TNBCs with high baseline TILs have higher rates of pathologic complete response
(pCR) to NAC. Moreover, the TIL burden in the RD of patients who do not achieve pCR to NAC is also correlated with prognosis.
However, insight into the molecular pathways in TNBC which modulate heterogeneity in host anti-tumor immune responses is
lacking. To address this gap in knowledge, we analyzed TILs retrospectively in a cohort of clinically and molecularly characterized
TNBCs with RD after NAC.
Methods: TILs were scored in H&E stained slides by expert pathologists in the post-treatment tumors of 92 NAC-treated TNBC
patients with RD at the time of resection and in 44 matched baseline diagnostic biopsies. Genomic alterations in the RD were
assayed using targeted next-generation sequencing (tNGS) while selected transcriptional signatures were evaluated by
NanoString as previously published (Balko et al, Cancer Discovery 2014). Differences in pre- and post-NAC TILs were compared
between tumors harboring alterations in cell cycle, PI3K/mTOR, growth factor receptors, Ras/MAPK and DNA repair pathways.
Associations of TILs with transcriptional signatures were also tested.
Results: A strong positive association of TILs in NAC-treated specimens was observed with RFS (coxPH p=0.0001, relative risk
reduction of 3.4% for each % of TILs) and OS (p=0.0016; relative risk reduction of 2.8% for each % of TILs). In multivariate
analysis with stage, age, node status and RD tumor cellularity, TILs in the post-NAC disease remained a significant predictor of
RFS and OS (p=0.0008 and p=0.007, respectively). TILs tended to decrease with NAC in paired samples, although this decrease
was not statistically significant (p=0.07).
Genetic alterations in the Ras/MAPK (amplifications in KRAS, BRAF, RAF1 and truncations in NF1) and cell cycle pathway
(alterations in CCND1-3, CDK4, CDK6, CCNE1, RB, AURKA and CDKN2A) were associated with lower TILs in RD (p=0.005 and
p=0.05, respectively). A significant inverse linear correlation was detected between a transcriptional signature of Ras/MAPK
activation (Pratilas et al, PNAS 2009) and TILs in the RD (Spearmans r=-0.42; p=0.00028). Total number of alterations of likely
functional significance detected by tNGS showed no association with TILs, suggesting that the association of Ras/MAPK
deregulation and cell cycle alterations with TILs may be a pathway-specific effect.
In TNBC cell lines, chemical inhibition of MEK transcriptionally up-regulated MHC-I and MHC-II molecules, while simultaneously
down-regulating mRNA expression of the immune checkpoint inhibitor PD-L1 (MDA-231 p=0.00002, BT549 p=0.0003, and
SUM159PT p=0.009). In vivo experiments confirming these associations are underway.
Conclusions: Our data suggest a strong correlation of Ras/MAPK pathway activation with immune-evasion and outcome in TNBC.
With additional mechanistic understanding, rational design of clinical trials combining MEK inhibitors with PD-L1 antibodies in
TNBC may be warranted.

Title: A phase Ib study of pembrolizumab (MK-3475) in patients with advanced triple-negative breast cancer
Rita Nanda1, Laura Q Chow2, E Claire Dees3, Raanan Berger4, Shilpa Gupta5, Ravit Geva6, Lajos Pusztai7, Marisa Dolled-Filhart8,
Kenneth Emancipator8, Edward J Gonzalez8, Jennifer Houp8, Kumudu Pathiraja8, Vassiliki Karantza8, Robert Iannone8, Christine
K Gause8, Johnathan D Cheng8 and Laurence Buisseret9. 1University of Chicago, Chicago, IL; 2University of Washington, Seattle,
WA; 3UNC Lineberger Comprehensive Cancer Center, Chapel Hill, NC; 4Sheba Medical Center, Tel Hashomer, Israel; 5H. Lee
Moffitt Cancer Center and Research Institute, Tampa, FL; 6Tel Aviv Sourasky Medical Center, Tel Aviv, Israel; 7Yale School of
Medicine, New Haven, CT; 8Merck & Co., Inc, Whitehouse Station, NJ and 9Institut Jules Bordet, Universit Libre de Bruxelles,
Bruxelles, Belgium.
Body: Introduction: The PD-1 receptor-ligand pathway can be used by tumors to evade immune surveillance, thereby allowing
neoplastic growth. Pembrolizumab is a highly selective, humanized IgG4/kappa isotype mAb designed to block PD-1 interaction
with its ligands PD-L1 and PD-L2, thereby reactivating the immune system to eradicate tumors.
Methods: This is a multi-center, non-randomized trial of single agent MK-3475 treatment given intravenously at 10 mg/kg every 2
weeks, in patients with recurrent/metastatic triple-negative (ER, PR, and HER2 negative) breast cancer (TNBC). PD-L1
expression in tumor or stroma was required for study entry. PD-L1 status was determined by immunohistochemical analysis of
patients tumor tissues using the Merck proprietary 22C3 antibody. Primary objectives of this study were to determine the safety,
tolerability, and anti-tumor activity of MK-3475 in patients with PD-L1 positive, advanced TNBC. Secondary objectives included
assessments of progression-free survival, overall survival, and response duration. Adverse events (AEs) reported in any patient
receiving at least 1 dose of study treatment were monitored and graded using NCI CTCAE v. 4.0. Radiographic imaging was
obtained every 8 weeks and evaluated by both investigator and an independent radiologist to assess clinical responses as
defined by RECIST 1.1. This study (Clinicaltrials.gov: NCT01848834) is being conducted in conformance with Good Clinical
Practices.
Results: A total of 32 female patients with a median age of 50.5 years (range 29 72 years) with PD-L1 positive,
recurrent/metastatic TNBC were enrolled in the study. Most of these patients had received and progressed on multiple lines of
therapy for advanced disease. A preliminary analysis of data collected as of 23May2014 indicates that 5 patients (15.6%)
experienced at least one drug-related serious adverse event (SAE); each of 4 patients experienced one of the following: Grade 3
anemia, headache, aseptic meningitis or pyrexia, and a fifth patient experienced disseminated intravascular coagulation (DIC)
with thrombocytopenia and decreased blood fibrinogen. The patient who experienced DIC died. Preliminary analysis of data
collected from investigators as of 23May2014 indicates that no patient had a complete response, 16.1% of patients had a partial
response, 9.7% had stable disease, and 64.5% had progressive disease. As of 23May2014, all but one of the responders, in
addition to three patients with stable disease, remain on treatment.
Conclusion: This is the first report of clinical activity of an immune checkpoint inhibitor in TNBC. The preliminary results from this
study suggest that single agent MK-3475 is a well-tolerated and effective treatment with significant therapeutic activity in a subset
of heavily pre-treated patients with recurrent/metastatic triple-negative breast cancer.

Title: Theranostic multiparametric tests improve residual risk assessment in early luminal breast cancer
John MS Bartlett1,9, Vicky S Sabine1, Syed Haider1, Camilla Drake1, Cheryl Crozier1, Cindy Q Yao1, Cassandra L Brookes2,
Cornelis JH van de Velde3, Annette Hasenburg4, Dirk G Kieback5, Christos Markopoulos6, Luc Y Dirix7, Caroline Seynaeve8,
Daniel W Rea2 and Paul C Boutros1,9. 1Ontario Institute for Cancer Research, Toronto, ON, Canada; 2Cancer Research UK
Clinical Trials Unit, University of Birmingham, Birmingham, United Kingdom; 3Leiden University Medical Center, Leiden,
Netherlands; 4University Hospital, Freiburg, Germany; 5Klinikum Vest Medical Center, Marl, Germany; 6Athens University Medical
School, Athens, Greece; 7St Augustinus Hospital, Antwerp, Belgium; 8Erasmus MC Daniel den Hoed Cancer Center, Rotterdam,
Netherlands and 9University of Toronto, Toronto, ON, Canada.
Body: Background: Despite rapidly expanding availability of multiparametric tests which inform residual risk following adjuvant
therapy for early breast cancer current approaches provide minimal information on the appropriate targeted therapy to be
selected for patients at high risk of recurrence. We hypothesised that inclusion of key signalling nodes from driver molecular
pathways in early breast cancer in residual risk signatures would both improve risk stratification and identify candidate theranostic
targets for the next generation of clinical trials.
Methods: RNA was extracted from FFPE luminal breast cancers from the TEAM pathology study (Exemestane versus
Tamoxifen-Exemestane). Gene expression analyses were performed for 29 genes mapped across key signalling nodes within the
PIK3CA pathway. mRNA assessment for IHC4 (ER, PgR, HER2 and Ki67) was included in the model. Quantitative gene
expression was performed using the Nanostring platform. Novel signatures were trained in a randomly selected sub-set of the
TEAM pathology cohort (n=1700) and validated using the remaining 50% of patients (n=1700). Results presented represent
those from the validation cohort.
Results: The IHC4-protein and IHC4-mRNA risk scores were highly correlated (Rho=0.72, p=4.12x10-265), suggesting the mRNA
abundance-based classifier is able to serve as a good substitute for the protein-based model.
A gene signature including IHC4 markers assessed by mRNA performed significantly better (AUC 0.70 vs 0.66) than conventional
IHC4. A gene signature including 4 signalling modules from the PIK3CA pathway significantly outperformed both IHC4 and the 4
gene (ER, PgR, HER2, Ki67) classifier (AUC 0.75; p = 3.23x10-7 vs IHC4 and 1.39x10-3 vs "IHC4mRNA").
Conclusions: Inclusion of PIK3CA signalling modules identified key genes/nodes which are linked to early relapse in luminal
breast cancer and provided a significantly improvement in risk classification when compared to a currently validated
multiparameter test.

Title: The FERGI phase II study of the PI3K inhibitor pictilisib (GDC-0941) plus fulvestrant vs fulvestrant plus placebo in patients
with ER+, aromatase inhibitor (AI)-resistant advanced or metastatic breast cancer Part I results
Ian Krop1, Stephen Johnston2, Ingrid A Mayer3, Maura Dickler4, Vinod Ganju5, Andres Forero-Torres6, Bohuslav Melichar7, Serafin
Morales8, Richard de Boer9, Steven Gendreau10, Mika Derynck10, Mark Lackner10, Jill Spoerke10, Ru-Fang Yeh10, Gallia Levy10,
Vivian Ng10, Carol O'Brien10, Heidi Savage10, Yuanyuan Xiao10, Timothy Wilson10, Soo Chin Lee11, Katarina Petrakova12, Susanne
Vallentin13, Denise Yardley14, Matthew Ellis15, Martine Piccart16, Edith A Perez17, Eric Winer1 and Peter Schmid18. 1Dana-Farber
Cancer Institute, Boston, MA; 2Royal Marsden Hospital, London, United Kingdom; 3Vanderbilt-Ingram Cancer Center, Nashville,
TN; 4Memorial Sloan Kettering Cancer Center, NY, NY; 5Peninsula Oncology Centre, Melbourne, Australia; 6University of
Alabama, Birmingham, AL; 7Palacky University Medical School and Teaching Hospital, Olomouc, Czech Republic; 8Hospital
Arnau de Vilanova, Lleida, Spain; 9Royal Melbourne Hospital, Parkville, Australia; 10Genentech, San Francisco, CA; 11National
University Hospital, Singapore, Singapore, Singapore; 12Masaryk Memorial Cancer Institute, Brno, Czech Republic; 13Herlev
University Hospital, Herlev, Denmark; 14Sarah Cannon Research Institute, Nashville, TN; 15Washington University School of
Medicine, St Louis, MO; 16Institut Jules Bordet, Universit Libre de Bruxelles, Brussels, Belgium; 17Mayo Clinic, Jacksonville, FL
and 18Barts Cancer Institute, Queen Mary University London, London, United Kingdom.
Body: Background: Preclinical and clinical data indicate a key role for the PI3-kinase (PI3K) pathway in the pathogenesis of
resistance to endocrine therapies in hormone receptor-positive (HR) breast cancer (BC) and suggest that combining PI3K
inhibitors with endocrine therapy may partially overcome this resistance. FERGI is the first randomized Phase II study testing
pictilisib (GDC-0941), a PI3K inhibitor, in combination with fulvestrant to evaluate this hypothesis in MBC patients with and without
PIK3CA-mutant tumors.
Methods: 168 post-menopausal pts with ER-positive, HER2-negative MBC were randomized (1:1) to receive fulvestrant with
either pictilisib 340 mg QD (n=89, "combination" arm) or matching placebo (n=79, "control" arm). To be eligible, pts had to have
relapsed during or within 6 mos of completing adjuvant AI treatment or have progressed on an AI for MBC. Pts were stratified
based on tumor PIK3CA mutation status, resistance to prior AI therapy and presence of measurable disease. The primary
endpoint was PFS by investigator assessment in the intent-to-treat (ITT) group and in pts with centrally confirmed PIK3CA-mutant
tumors. The primary analysis was based on a 6 mo median duration follow up.
Results: Baseline disease and prior treatment characteristics were similar between study arms. Observed treatment-emergent
AEs were consistent with those previously described for single agent pictilisib and fulvestrant (primary toxicities were rash and GI
disorders). In the ITT population (84 events) the median PFS (mPFS) was 6.2 mo in the combination arm vs 3.8 months for the
control arm (HR, 0.77; 95% CI, 0.50-1.19). For pts with PIK3CA-mutant tumors (37 events), mPFS was 6.2 mo in the combination
arm vs 5.1 mo in the control arm (HR, 0.92; 95% CI, 0.48-1.76). For pts without a detectable PIK3CA mutant tumor (43 events),
mPFS was 5.8 months in the combination arm vs 3.6 months in the control arm (HR, 0.64; 95% CI, 0.35-1.17). Exploratory
post-hoc subgroup analysis suggested improvement in PFS in pts with ER+ and PR+ tumors (centrally confirmed) treated with
pictilisib plus fulvestrant. In the ER+/PR+ subgroup (57 events) mPFS was 7.2 mo in the combination arm vs 3.7 mo in the control
arm (HR, 0.46; 95% CI, 0.27 to 0.78). This improvement was independent of tumor PIK3CA mutation status. Multivariate analysis
suggests that this treatment effect in pts with ER+/PR+ tumors is maintained after adjusting for possible baseline imbalances. A
similar analysis on pts with luminal A tumors (per PAM50 analysis) was also consistent with the findings in pts ER+/PR+ disease.
Conclusions: This is the first report of a blinded, randomized clinical study evaluating a PI3K inhibitor in pts with MBC. In the ITT
population, the addition of pictilisib to fulvestrant was associated with a mPFS improvement of 3.8 mo to 6.2 mo. Exploratory
subgroup analyses suggested in pts with ER+/PR+ tumors are more likely to derive benefit from the addition of pictilisib to
fulvestrant irrespective of PIK3CA mutation status, though the subgroup analyses are limited by the sample size. Additional
biomarker analyses will be reported.

Title: Preoperative window of opportunity study of the PI3K inhibitor pictilisib (GDC-0941) plus anastrozole vs anastrozole alone
in patients with ER+, HER2-negative operable breast cancer (OPPORTUNE study)
Peter Schmid1, Sarah E Pinder2, Duncan Wheatley3, Jane Macaskill4, Charles Zammit5, Jennifer Hu6, Robert Price7, Nigel
Bundred8, Sirwan Hadad9, Alice Shia1, Louise Lim1, Shah-Jalal Sarker1, Patrycja Gazinska2, Natalie Woodman2, Darren Korbie10,
Matt Trau10, Paul Mainwaring11, Peter Parker12, Arnie Purushotham2 and Alastair M Thompson13. 1Barts Cancer Institute,
Charterhouse Square, London, United Kingdom; 2Guy's and St Thomas' Hospital, London, United Kingdom; 3Royal Cornwall
Hospital, Cornwall, United Kingdom; 4Ninewells Hospital, Dundee, United Kingdom; 5Brighton and Sussex University Hospitals
NHS Trust, Brighton, United Kingdom; 6St Barts Hospital, London, United Kingdom; 7King's College Hospital, London, United
Kingdom; 8University Hospital South Manchester, Manchester, United Kingdom; 9Royal Hallamshire Hospital, Sheffield, United
Kingdom; 10Australian Institute for Bioengineering and Nano-technology, St Lucia, Australia; 11ICON Cancer Care, Brisbane,
Australia; 12Kings College London, London, United Kingdom and 13MD Anderson Cancer Centre, Texas, TX.
Body: Background: Preclinical and clinical data support a key role for the PI3K pathway in resistance to endocrine therapy in
patients with ER+ breast cancer and suggest that combining PI3K inhibitors with endocrine therapy may overcome resistance.
Short-term preoperative window of opportunity (WOO) studies are a validated strategy for novel treatments to provide
proof-of-concept and define the most appropriate patient population by directly assessing treatment effects in tumour tissue
before and after treatment. This is the first WOO study with the PI3K inhibitor pictilisib (GDC-0941) in combination with
anastrozole (ANA).
Methods: 73 postmenopausal patients (pts) have been randomized (2:1 in favour of the combination) to receive 2-week
preoperative treatment with ANA plus pictilisib (n=50, "ANA+PIC" arm) or ANA alone (n=23, "ANA" arm). Pts had newly
diagnosed, operable, ER+, HER2-negative breast cancer of 1 cm size. Pts receiving HRT were excluded. Treatment effects and
correlative studies were assessed using FFPE and frozen tumour biopsies taken before and after 14 days of study treatment.
The primary endpoint was inhibition of tumour-cell proliferation, as measured by change in Ki67 expression, determined centrally
by 2 investigators. Secondary endpoints include induction of apoptosis (Caspase3) and safety. Comprehensive biomarkers
analyses include targeted NGS of a comprehensive cancer panel of >400 genes, copy number analyses, and pre- and
post-treatment reverse-phase protein arrays (RPPA) and RNA profiling.
Results: Baseline (BL) disease characteristics were similar between both study arms. PAM50 analysis showed that 53% and
47% of tumors were Luminal (Lum) A and B, respectively. 65% of tumors had >14% Ki67-positive cells. Observed
treatment-emergent AEs were consistent with those previously described for single-agent pictilisib and anastrozole.
Mean post-treatment percentage reduction of Ki67 was 84% (95% CI, 75%-89%) for ANA+PIC and 72% (54%-87%) for ANA.
Ki67-response (50% drop in % of Ki67+ cells) was 86% for ANA+PIC and 60% for ANA. By using the definition that pts with a
natural logarithm of %Ki67+ cells of 1 or 1-2 have a day 15 anti-proliferative response, 93% [ln(ki67): <1, 46%; 1-2, 46%] of
ANA+PIC were responders compared with 60% [<1, 47%; 1-2, 13%] of ANA-treated pts (p = 0.01).
Preplanned subgroup analyses showed a significant interaction of response to ANA+PIC with molecular subtype and Ki67 levels.
Patients with LumB tumors or high BL Ki67 (>14%) had a higher Ki67 response with ANA+PIC compared to ANA (LumB, 83% vs
38%; Ki67>14%, 94% vs 55%), whereas Ki67 response was similar for both treatments for LumA tumors (ANA, 75%; ANA+PIC,
73%) or tumors with low BL Ki67. Mean post-treatment % reduction of Ki67 in LumB tumors was 87% (95% CI, 49%-96%) for
ANA+PIC and 56% (16%-77%) for ANA (p=0.03).
Additional data on apoptosis and comprehensive pre- and post-treatment biomarkers analyses will be presented.
Conclusions: This first report of a preoperative WOO study evaluating a PI3K inhibitor in early breast cancer demonstrated
addition of pictilisib to ANA was associated with increased anti-proliferative response over single-agent ANA.

Title: Comprehensive molecular characterization of invasive lobular breast tumors
Giovanni Ciriello1, Michael L Gatza2, Katherine A Hoadley2, Hailei Zhang3, Suhn K Rhie4, Reanne Bowlby5, Matthew D Wilkerson2,
Cyriac Kandoth1, Michael McLellan6, Andrew Cherniack3, Peter W Laird4, Chris Sander1, Tari A King1 and Charles M Perou2.
1
Memorial Sloan Kettering Cancer Center, New York, NY; 2University of North Carolina, Chapel Hill, NC; 3Broad Institute of MIT
and Harvard, Cambridge, MA; 4University of Southern California, Los Angeles, CA; 5British Columbia Cancer Agency and
6
Washington University, St Louis, MO.
Body: Invasive lobular breast cancer (ILC) is the second most common histological subtype of breast cancer accounting for
10-15% of invasive breast tumors. ILC is typically ER+ and beyond the known mutation and/or loss of E-cadherin function, which
contributes to a highly discohesive morphology, little is known about the additional mechanisms driving ILC tumorigenesis, or
alterations that differentiate ILC from invasive ductal carcinomas (IDC).
Methods
A dataset of 817 breast tumors from the TCGA Project, including 490 IDC, 127 ILC and 88 samples with a mixed IDC-ILC
histology, were profiled on six genomic platforms to develop a comprehensive atlas of mutational, epigenetic, transcriptional and
proteomic data. Integrative genomic analyses, both supervised and unsupervised, of ILC tumors and across histological subtypes
were performed to identify genomic drivers of ILC oncogenesis.
Results
Comprehensive multi-platform analyses identified distinct molecular events associated with ILC tumors. As expected, lack of
E-cadherin protein, as determined by Reverse Phase Protein Array (RPPA), and CDH1 mRNA expression was uniformly
observed in ILC cases associated with distinct alterations targeting CDH1. In addition to previously reported CDH1 and PIK3CA
mutations, we identified a number of novel ILC-enriched recurrent mutations targeting PTEN, RUNX1, TBX3, and FOXA1.
An increased incidence of PTEN inactivating events, both mutations and copy number changes, were identified in ILC (13%)
compared to IDC ER+ (7%), which corresponded with altered PTEN protein expression. These alterations were largely mutually
exclusive with PIK3CA mutations and correlate with increased Akt activation as evident by increased Akt phosphorylation (pS473
and pT308), thus identifying a potential therapeutic opportunity for ILC patients.
GATA3 signaling, which regulates epithelial cell differentiation, is frequently altered in luminal/ER+ breast cancers. Our analyses
determined GATA3 mutations are more frequent in IDC luminal tumors as compared to ILC (19 % vs 5%). ILC luminal tumors
show significantly lower GATA3 protein expression, but a higher frequency of mutations in FOXA1 (9% vs 2% in Luminal IDC), a
transcription factor required to promote ER transcriptional programs. Within ILC tumors, FOXA1 mutations were found to cluster
into a specific region of the Forkhead (FK) DNA binding domain. A broader analysis of FOXA1 mutations in breast and prostate
cancer confirm two specific hotspots in the FK domain and the C-terminal transactivation domain. Interestingly, these mutational
classes are associated with distinct transcriptional changes suggesting different functional effects.
Finally, mRNA-seq analyses identified three robust molecular subclasses that are characterized by distinct genetic, genomic and
proteomic patterns, including an increased immune-related group (Class 2), as well as differences in prognosis.
Conclusions
In this study, we developed a comprehensive atlas of genomic alterations that reveals key molecular differences differentiating
ILC (FOXA1) from IDC (GATA3) tumorigenesis, a potential therapeutic target for ILC (Akt), and novel ILC subclasses based on
underlying biological events. These findings provide further insight into the molecular heterogeneity of ER+ breast cancer.

Title: Characterization and clinical relevance of the genomic alterations defining lobular breast cancer
Christine Desmedt1, Gunes Gundem2, Gabriele Zoppoli1, Giancarlo Pruneri3, Elia Biganzoli4, Marco Fornili4, Debora Fumagalli1,
Franoise Roth1, David Brown1, Peter Van Loo2, Sylvain Brohe1, Delphine Vincent1, Naima Keddoumi1, Samira Majjaj1,
Ghizlane Rouas1, Thomas Van Brussel5, Diether Lambrechts5, Otto Metzger6, Christine Galant7, Franois Bertucci8, Martine
Piccart1, Denis Larsimont1, Giuseppe Viale3, Peter J Campbell2 and Christos Sotiriou1. 1Institut Jules Bordet, Universit Libre de
Bruxelles, Belgium; 2Wellcome Trust Sanger Institute - The Cancer Genome Project, United Kingdom; 3European Institute of
Oncology, Milan, Italy; 4University of Milan, Fondazione IRCCS Istituto Nazionale Tumori, Italy; 5Vesalius Research Center, VIB,
Campus Gasthuisberg, Leuven, Belgium; 6Dana-Farber Cancer Institute, Boston, MA; 7Universit Catholique de Louvain, Belgium
and 8Institut Paoli-Calmettes, Marseille, France.
Body: Background:
Invasive lobular breast cancer (ILC) is the second most common histological subtype of breast cancer, is mostly estrogen
receptor-positive (ER+) and has distinct clinico-pathological features when compared to ER+ invasive ductal carcinoma (IDC). In
this study, we aimed to characterize the genomic alterations defining ILC in a large cohort of ILC patients with long-term follow up
(FU).
Material & methods:
In 499 centrally histologically confirmed ILC patients (median FU= 9.8 years) we analyzed mutational data gathered from targeted
sequencing of 360 cancer genes at an average coverage of 106X (alignment done with BWA, substitutions and indels further
referred to as mutations- called with Caveman and Pindel). Matched normal DNA was available for 242 patients. Genome-wide
copy number (CN) data were available for 178 patients. E-cadherin (CDH1) and beta-catenin (CTNNB1) stains were carried out
using the DakoTM antibodies. Invasive disease free survival (IDFS) was considered as primary survival endpoint.
Results:
A median of 6 [range:0-38] non-silent mutations was identified across the primary tumors of all patients. The most frequently
mutated genes (>3%) are listed in Table 1. Of those, CDH1, PIK3CA, TBX3, FOXA1 and the chromatin-related genes MLL2,
MLL3, ARID1A and ARID1B were more frequently mutated in our ER+/HER2- ILC (n= 451) compared to the ER+/HER2- IDC
(n=266) from The Cancer Genome Atlas, whereas GATA3, TP53 and MAP3K1 were less frequently mutated.
Samples with a CDH1 mutation were associated with changes at the protein level: 97.5% displayed a complete loss of the
protein, associated with cytoplasmic staining for CTNNB1, compared to only 63% of the CDH1 non-mutated tumors. CDH1
mutated tumors were further characterized by increased mutational frequencies of the ERBB-genes: 15.4% for the CDH1 mutated
tumors versus 3% in the CDH1 non-mutated tumors, most of those mutations being described in the literature as activating the
pathway. Almost all tumors (97%) with CN data had a heterozygous loss of CDH1.
The special alveolar, solid and trabecular lobular histotypes were associated with specific CN alterations and mutations.
Tumors with mutated ARID1A or ATM were associated with worse IDFS at the univariate level and ARID1A remained significant
in a multivariate analysis including standard parameters (HR =2.07, p=0.003). At the CN level, ATM and ARID1A losses, as well
as HER2 and VEGFA gains/amplifications were associated with decreased IDFS, all but ARID1A holding significance at the
multivariate level (HR_HER2=2.41, HR_VEGFA= 1.99, HR_ATM= 1.79, all p<0.05).
Conclusion:
This is the first and largest study to our knowledge to report genomic alterations present in ILC and their association with survival.
This work therefore opens new avenues for a better understanding of the disease and its clinical management.
Table 1: List of the most frequently mutated genes
Gene %
Gene %
Gene %
Gene %
CDH1 62.9
MAP3K1 7
USP9X 4.4
EPHB6 3.8
PIK3CA 44.9
BRCA2 6.6
ATR 4.2
MED12L 3.6
MLL3 15.8
ERBB2 6.4
COL22A1 4.2
PTEN 3.6
TBX3 13.2
ARID1B 5.6
MED13 4.2
ERBB3 3.2
FOXA1 9.6
ATM 4.8
NOTCH1 4.2
ROS1 3.2
MLL2 9.4
MLL 4.6
AKT1 4.2
BPTF 3.2
ARID1A 8.4
MYO5B 4.4
MYO3A 4
BRCA1 3.2
GATA3 8
EP300 4.4
NF1 4
IRS2 3.2
TP53 7.4
RUNX1 4.4
EP400 3.8
NOTCH4 3

Title: Survival advantage of anastrozol compared to tamoxifen for lobular breast cancer in the ABCSG-8 study
Michael Knauer1, Christine Gruber2, Otto Dietze3, Richard Greil3, Herbert Stger4, Margaretha Rudas5, Zsuzsanna Bago-Horvath5,
Brigitte Mlineritsch3, Werner Kwasny6, Christian Singer5, Peter Dubsky5, Raimund Jakesz5, Florian Fitzal2, Gnther Steger5,
Rupert Bartsch5, Martin Filipits5, Christian Fesl7 and Michael Gnant5. 1Breast Center St Gallen, Switzerland; 2Hospital of the
Sisters of Charity, Linz, Austria; 3Paracelsus Medical University, Salzburg, Austria; 4Medical University, Graz, Austria; 5Medical
University, Vienna, Austria; 6Landesklinikum, Wiener Neustadt, Austria and 7ABCSG.
Body: Introduction:
Invasive lobular cancer (ILC) is the second most common histological type of breast cancer, appearing to have a biology distinct
from ductal cancer (IDC). In the first 5 years, the prognosis of ILC seems to be more favourable compared to IDC. The aim of this
study was to investigate the differences in benefit from adjuvant Tamoxifen (Tam) and Anastrozol (Ana) for these histologic
subtypes.
Patients and Methods:
The ABCSG-8 (Austrian Breast and Colorectal Cancer Study Group 8) study was a randomized phase III clinical trial (n=3.714) in
hormone receptor-positive cancer addressing a sequence strategy with 3 years of Ana after 2 years of Tam in comparison with 5
years of Tam in a low- to intermediate-risk group (Grade 1/2) of postmenopausal patients not receiving adjuvant chemotherapy.
Univariate and multivariate comparisons for overall (OS) and disease-free survival (DFS) between endocrine treatments were
conducted per subgroup with respect to histology (ductal or lobular). Multivariate Cox analysis included endocrine treatment,
histology and their interaction additional covariates (age, T-stage, N-stage, grade, ER, PR). For the time-to-event analyses,
starting point for all patients was two years after randomization, i.e. when treatment changed from Tamoxifen to Anastrozole on
the experimental arm. To avoid crossover effects for post-study treatment, the total time for this analysis is 3 years.
Results:
The 3-year OS hazard ratio (HR) for Anastrozol versus Tamoxifen in ILC (n=694) was 0.24 (0.08-0.70) vs. IDC (n=2.739) HR 1.08
(0.75-1.58). In multivariate analysis, HR for ILC was 0.23 (0.08-0.68) vs. 1.02 (0.70-1.49) for IDC. The test for interaction of
treatment and histology was significant (p=0.01). In ILC, no other clinico-pathologic factor was significantly associated with
survival differences compared to IDC, where age, T- and N-stage maintained significance. In the models for DFS, our preliminary
analysis after 3-year follow-up did not show significant variations in treatment effect according to histology.
Discussion:
The magnitude of survival benefit from adjuvant Anastrozol vs. Tamoxifen varies by histology in this large phase III randomized
trial. A significant reduction in risk of death occurs only in patients with lobular cancer compared to ductal cancer after only 3
years of follow-up. An updated analysis with additional follow-up will be available at presentation.

Title: A randomized phase III trial comparing neoadjuvant chemotherapy with weekly nanoparticle-based paclitaxel with
solvent-based paclitaxel followed by anthracyline/cyclophosphamide for patients with early breast cancer (GeparSepto); GBG 69
Michael Untch1, Christian Jackisch2, Andreas Schneewei3, Bettina Conrad4, Bahriye Aktas5, Carsten Denkert6, Holger Eidtmann7,
Hermann Wiebringhaus8, Sherko Kmmel9, Jrn Hilfrich10, Mathias Warm11, Stefan Paepke12, Marianne Just13, Claus Hanusch14,
John Hackmann15, Jens-Uwe Blohmer16, Michael Clemens17, Serban Dan Costa18, Bernd Gerber19, Valentina Nekljudova20, Sibylle
Loibl20,2 and Gunter von Minckwitz20,21. 1Helios Klinikum Berlin-Buch; 2Sana Klinikum Offenbach; 3University Hospital, Heidelberg;
4
Elisabeth Krankenhaus, Kassel; 5Universittsklinkum Essen; 6Institute of Pathology, Charit University of Berlin, Berlin,
Germany; 7Universittsklinikum Schleswig-Holstein Kiel; 8St Barbara Kliniken Heesen; 9Kliniken Essen Mitte; 10Eilenriede Klinik
Hannover; 11Kliniken der Stadt Kln; 12Universitts-Frauenklinik rechts der Isar, Mnchen; 13Praxis Bielefeld; 14Rotkreuzklinikum
Mnchen; 15Marienhospital Witten; 16St Johannes Hospital Dortmund; 17Mutterhaus Trier; 18Universitts-Frauenklinik, Magdeburg;
19
Universitts-Frauenklinik, Rostock; 20German Breast Group, Neu-Isenburg, Hessen, Germany and 21Universitts-Frauenklinik,
Frankfurt am Main.
Body: Background: Anthracycline/taxane based regimen are standard of care for neoadjuvant therapy in breast cancer. A
reverse sequence of taxanes followed by anthracyclines was suggested to achieve higher pCR rates (H.Earl, Lancet Oncol 2014).
Dual HER2 blockade was shown to be superior to trastuzumab alone increasing the pCR rate by 20%. Nab-paclitaxel (nP) is a
solvent-free formulation of paclitaxel (P) encapsulated in albumin which might further improve the pCR rate in breast cancer
patients receiving neoadjuvant treatment and cause lower toxicity.
Methods: In the GeparSepto study (NCT01583426) patients were randomized to either nP (125 mg/m2) q1w or P (80mg/m2) q1w
for 12 weeks followed by 4 cycles of conventionally dosed EC (E, epirubicin 90mg/m2; C, cyclophosphamide 600 mg/m2) q3w.
The primary objective is to compare the pathological complete response rate (pCR, ypT0 ypN0). Further objectives are to
compare the pCR rate in predefined subgroups, pCR by other definitions, clinical response rate, rate of breast conserving surgery
and toxicity and compliance. Patients with untreated, histologically confirmed uni- or bilateral, cT2- cT4d carcinoma, and no
clinically relevant cardiovascular and other co-morbidities were included. HER2+ patients received trastuzumab (loading dose
8mg/kg; 6 mg/kg) plus pertuzumab (loading dose 840 mg; 420 mg) q3w concomitantly. HER2, estrogen receptor, progesterone
receptor, Ki67 and SPARC status were centrally assessed prior to randomization for stratification. To increase the pCR rate from
33% with P to 41% with nP, corresponding to an odds ratio of 1.41 with an alpha of 0.05 and a power of 80%, 1200 patients
would be needed. A window-of opportunity study was integrated to investigate response to anti-HER2 treatment without
chemotherapy, HER2+ patients were randomized to receive 6 weeks of either trastuzumab, pertuzumab or the combination with
biomaterial collection at the start and the end of the window.
Results: A total of 1204/1229 (window study n=71) recruited patients (7/2012 - 12/2013) from 69 German centers were
evaluable, 606 receiving nP. Baseline characteristics are well balanced; median age was 49/50 years (P/nP), 33/33% of the
patients presented with HER2+ tumors, 23/23% triple negative breast cancer TNBC (ER and PR <1 %), 87/85% ductal invasive,
56/52% G3 tumors; Ki67>20% 69/69%; SPARC positive 15.7/16%. 265 patients reported SAEs (119 P/146 nP) and 4 died on
study (1 P: cardiac decompensation; 3 nP: accident at home, multiorgan failure, sepsis during EC). The pCR rate (ypT0 ypN0) is
29% with P and 38% with nP, OR 1.5; p<0.01.
Conclusions: GeparSepto showed that the pCR rate is significantly higher with nab-paclitaxel compared to solvent-based
paclitaxel given weekly before anthracycline based chemotherapy. Subgroupanalyses and 2ndary endpoints will be presented at
the meeting.
The trial was financially supported by Roche and Celgene. The window-substudy was funded within the EU-FP7 project
RESPONSIFY No 278659.

Title: TNT: A randomized phase III trial of carboplatin (C) compared with docetaxel (D) for patients with metastatic or recurrent
locally advanced triple negative or BRCA1/2 breast cancer (CRUK/07/012)
Andrew Tutt1, Paul Ellis2, Lucy Kilburn3, Cheryl Gilett1, Sarah Pinder1, Jacinta Abraham4, Sophie Barrett5, Peter Barrett-Lee4,
Stephen Chan6, Maggie Cheang3, Mitch Dowsett12, Lisa Fox3, Patrycja Gazinska15, Anita Grigoriadis1, Alexander Gutin16,
Catherine Harper-Wynne7, Matthew Hatton8, Sarah Kernaghan3, Jerry Lanchbury16, James Morden3, Julie Owen1, Jyoti Parikh2,
Peter Parker9, Nazneen Rahman10, Rebecca Roylance11, Adam Shaw2, Ian Smith12, Rose Thompson13, Kirsten Timms16, Holly
Tovey3, Andrew Wardley14, Gregory Wilson14, Mark Harries2 and Judith Bliss3. 1Kings College London, London, United Kingdom;
2
Guy's and St Thomas' NHS Foundation Trust, London, United Kingdom; 3Institute of Cancer Research Clinical Trials & Statistics
Unit (ICR-CTSU), London, United Kingdom; 4Velindre NHS Trust Cancer Centre, Cardiff, United Kingdom; 5Beatson West of
Scotland Cancer Centre, Glasgow, United Kingdom; 6Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom;
7
Maidstone and Tunbridge Wells NHS Trust, Maidstone, United Kingdom; 8Sheffield Teaching Hospitals NHS Foundation Trust,
Sheffield, United Kingdom; 9Cancer Research UK London Research Institute, London, United Kingdom; 10Institute of Cancer
Research, London, United Kingdom; 11Barts Health NHS Trust, London, United Kingdom; 12Royal Marsden NHS Foundation
Trust, London, United Kingdom; 13BME Cancer Communities, Nottingham, United Kingdom; 14Christie NHS Foundation Trust,
Manchester, United Kingdom; 15Breakthrough Breast Cancer Research Unit, Guys & St Thomas NHS Trust, Kings College,
London, United Kingdom and 16Myriad Genetics, Inc.
Body: Introduction
Subgroups within triple negative breast cancers (TNBCs) appear to share impaired DNA repair mechanisms with BRCA1/2
germline mutation +ve (gBRCA+); hypothesised to confer sensitivity to platinum. TNT tested this hypothesis for women with
recurrent locally advanced (LABC) metastatic (MBC) gBRCA+ or TNBC.
Methods
Eligible patients (pts) had ER-, PgR-, HER2- BC or were known gBRCA+ (any ER/PgR/HER2). Pts were randomized to C (AUC 6
q3wk) or D (100mg/m2 q3wk) for 6-8 cycles or until disease progression if sooner with crossover possible on progression.
Ineligibility incl. 1) adjuvant taxane in 12mth 2) previous platinum 3) non-anthracyclines for MBC. Blood & archival tissue were
requested for BRCA1/2 genotyping, central analysis of TN subtype & DNA repair deficiency biomarkers. Primary endpoint was
ITT objective response rate (ORR) to cycle 6. Secondary endpoints incl. toxicity, PFS, OS & pre-planned subgroups (gBRCA+,
HR deficiency (HRD), basal-like (IHC & PAM50). TNT aimed to detect a 15% improvement in ORR with C compared to D
(=0.05, power=90%).
Results
376 (188 C, 188 D) pts were recruited (74 UK centres, 08/08-03/14). Median age 55.3yr (IQR 48-63). 29 (8%) were known
gBRCA+ (16 TNBC, 12 ER+ HER2-, 1 ER unk HER2-). 341 (91%) had MBC & 35 (9%) recurrent LABC. 34% had adjuvant
taxane. Median relapse interval 2.1yr (IQR 1.4-3.4). Tumour (primary 309 pts, recurrent 102 pts), blood DNA (288 pts) was
obtained.
ORRs were C 59/188 (31.4%, 95%CI 25-38) & D 67/188 (35.6%, 95%CI 29-42), abs. diff. -4.2% (95%CI -13.7-5.3) p=0.44.
Median PFS C 3.1 (95%CI 2.5-4.2) vs D 4.5 (95%CI 4.1-5.2) mth; abs. diff. -0.4 (95%CI -1.1-0.3) p=0.29. Median OS C 12.4
(95%CI 10.4-15.3) vs D 12.3 (95%CI 10.5-13.6) mth; abs. diff. -0.2 (95%CI -1.1-0.8) p=0.31.
For 43 gBRCA+ pts (29 known & 14 research test), ORR of 68.0% with C (17/25, 95%CI 46.5-85.1) & 33.3% D (6/18, 95%CI
13.3-59.0) p=0.03. 273 gBRCA- pts, ORR=28.1% C (36/128, 95%CI 20.5-36.8) & 36.6% D (53/145, 95%CI 28.7-44.9) p=0.16.
Interaction p=0.01.
195 pts had MYRIAD HRD scores. 81 high (42), ORR=38.2% with C (13/34, 95%CI 22.2-56.4) & 42.6% with D (20/47, 95%CI
28.3-57.8) p=0.82; 114 low (<42), ORR=29.2% C (19/65, 95%CI 18.6-41.8) & 34.7% D (17/49, 95%CI 21.7-49.6) p=0.55.
Interaction p=0.91.
By central ER/PgR/HER2/CK5/EGFR IHC 116 pts were core basal-like, ORR=32.2% C (19/59, 95%CI 20.6-45.6) & 28.1% D
(16/57, 95%CI 17.0-41.5) p=0.69 & 51 5-marker ve non-basal (5NP), ORR=24.0% C (6/25, 95%CI 9.4-45.1) & 42.3% D
(11/26, 95%CI 23.4-63.1) p=0.24. Interaction p=0.16.
By Prosigna PAM50 in pts entering as TN 174 had basal-like, ORR=32.6% C (28/86, 95%CI 22.8-43.5) & 35.2% D (31/88, 95%CI
25.3-46.1) p=0.75 & 36 pts non-basal subtypes, ORR=16.7% with C (3/18, 95%CI 3.6-41.4) & 73.7% with D (13/18, 95%CI
46.5-90.3) p<0.01. Interaction p=0.01.
Discussion
While TNT gives no evidence to support superior activity of C compared to D in unselected TNBC pts results for gBRCA+ pts
support their greater sensitivity to C than D. Neither a dichotomised MYRIAD HRD score nor 2 basal-like classifiers selected
sensitivity to C. The sensitivity to D of a non-basal-like subgroup by Prosigna PAM50 warrants independent investigation.

Title: NSABP B-36: A randomized phase III trial comparing six cycles of 5-fluorouracil (5-FU), epirubicin, and cyclophosphamide
(FEC) to four cycles of adriamycin and cyclophosphamide (AC) in patients (pts) with node-negative breast cancer
Jacobs A Samuel1,2, John W Wilson1,3, Hanna Bandos1,3, Richard M Elledge1,4, Andr Robidoux1,5, Louis Fehrenbacher1,6, Patrick J
Ward1,7, Johnathan Polikoff1,8, Adam M Brufsky1,9, Louise Provencher1,10, Alexander HG Paterson1,11, John T Hamm1,12, Robert L
Carolla1,13, Luis Baez-Diaz1,14, Priya Rastogi1,9, Thomas B Julian1,15, D Lawrence Wickerham1,15, Sandra M Swain1,16, Charles E
Geyer, Jr1,17, Eleftherios P Mamounas1,18 and Norman Wolmark1,15. 1National Surgical Adjuvant Breast and Bowel Project
(NSABP), Pittsburgh, PA; 2University of Pittsburgh Medical Center, Pittsburgh, PA; 3Graduate School of Public Health, University
of Pittsburgh, Pittsburgh, PA; 4Baylor College of Medicine, Houston, TX; 5Centre Hospitalier de l'Universit de Montral (CHUM),
Canada; 6Kaiser Permenente, Northern California, Vallejo; 7Onoclogy/Hematology Care Clinical Trials; 8Kaiser Permanente, San
Diego; 9University of Pittsburgh Cancer Institute, Pittsburgh, PA; 10Centre Hospitalier Affilie Universitaire de Quebec (CHA),
Canada; 11Tom Baker Cancer Centre, Canada; 12Norton Cancer Institute; 13CCOP, Ozark Health Ventures LLC-Cancer Research
for the Ozarks; 14MBCCOP, San Juan; 15Allegheny Cancer Center at Allegheny General Hospital; 16Washington Cancer Institute,
Medstar Washington Hospital, Washington, DC; 17Virginia Commonwealth University, Massey Cancer Center, Richmond, VA and
18
UF Cancer Center at Orlando Health.
Body: Background
NSABP B-36 was originally designed as a 2X2 factorial randomized study to compare 6 cycles of FEC-100 with 4 cycles of
standard AC with or without celecoxib in pts with node-negative breast cancer. The rationale for the trial was based on
observations from other adjuvant trials suggesting that longer duration of anthracycline-based therapy may result in improved
outcomes and also on accumulating evidence that prostaglandins may contribute to the malignant phenotype in breast cancer.
The trial opened in May 2004 but random assignment to celecoxib v placebo was terminated in December 2004 (after 327 pts
were enrolled) because of concerns for increased risk of cardiovascular disease with the use of COX-2 inhibitors. The trial
continued as a two-arm study and completed accrual in July 2008. The primary endpoint of disease-free survival (DFS), and
secondary endpoints of overall survival (OS), recurrence-free interval (RFI), distant recurrence-free interval (DRFI), and adverse
events comparing FEC-100 and AC are reported here. Analyses of quality of life indicators will be reported separately.
Methods
2,722 pts with T1-3, pN0 breast cancer were randomly assigned to either adriamycin 60 mg/m2 and cyclophosphamide
600mg/m2 every 21 days for 4 cycles (n=1361) or 5-FU 500mg/m2, epirubicin 100 mg/m2 and cyclophosphamide 500 mg/m2
every 21 days for 6 cycles (n=1361). Hormone- receptor positive pts were to receive physician's choice of hormonal therapy for a
minimum of 5 yrs. Trastuzumab for HER2-positive pts was at investigator's discretion but was not to begin for >3 weeks after the
last dose of chemotherapy. All women treated with lumpectomy were to receive breast radiotherapy; post-mastectomy chest wall
radiotherapy was at physician's discretion. The differences in DFS (OS, RFI, and DRFI) between two treatment arms were
assessed by log-rank test stratified by hormone receptor status and type of surgery.
Results
Median follow-up is 82.8 months. Pt and tumor characteristics were equally distributed between the two groups (<50 years old:
40%, lumpectomy: 68%, and hormone positivity: 65 %). Overall, Grade 3 and 4 expected toxicities were more frequent in the FEC
arm. Combined Grade 3/4 toxicities reaching statistical significance with a difference of 3% or more between AC and FEC arms
included fatigue 3.55% v 8.45%, febrile neutropenia 3.70% v 9.42%, and thrombocytopenia 0.74% v 4.41%, respectively. While
on treatment, Grade 3 left ventricular systolic dysfunction occurred in one pt in each arm. There were 2 toxicity deaths on AC and
5 on the FEC arm. Primary and secondary endpoint analyses at 8 ys did not reveal any significant differences in DFS, OS, RFI, or
DRFI.
Endpoint
AC (%)
FEC (%)
HR
CI
p-value
DFS
83.0
82.8
1.04
0.85,1.26
0.70
OS
91.2
92.0
0.94
0.71,1.24
0.65
RFI
90.0
90.4
0.97
0.76,1.26
0.84
DRFI
92.3
93.0
0.89
0.66,1.20
0.43
Conclusions
Six cycles of the FEC-100 regimen did not result in any efficacy advantage over 4 cycles of standard AC in node-negative breast
cancer pts. As anticipated, the FEC-100 regimen resulted in greater toxicity.
Support
NCI: U10-CA-12027, -37377, -69974, -69651, -44066-26; Pharmacia & Upjohn Co.

Title: Ten year update of E1199: Phase III study of doxorubicin-cyclophosphamide followed by paclitaxel or docetaxel given every
3 weeks or weekly in patients with axillary node-positive or high-risk node-negative breast cancer
Joseph A Sparano1, Fengmin Zhao2, Silvana Martino3, Jennifer Ligibel2, Thomas Saphner4, Antonio C Wolff5, George W Sledge,
Jr6, Edith A Perez7, William C Wood8 and Nancy E Davidson9. 1Montefiore Medical Center, Albert Einstein College of Medicine,
Bronx, NY; 2Dana-Farber Cancer Institute, Boston, MA; 3Angeles Clinic, Los Angeles, CA; 4Aurora Health, Two Rivers, WI; 5Johns
Hopkins, Baltimore, MD; 6Stanford University, Stanford, CA; 7Mayo Clinic; 8Emory University, Atlanta, GA and 9University of
Pittsburgh, Pittsburgh, PA.
Body: Background: We had previously reported after a median followup of 5.3 years that disease free survival (DFS) was
improved with either adjuvant weekly paclitaxel (hazard ratio [HR] 0.73, p=0.0006) or every 3 week docetaxel (HR 0.77, p=0.02)
compared with every 3 week paclitaxel (N Eng J Med 2008; 358; 1663), and that obesity and black race were independently
associated with inferior outcomes in estrogen receptor (ER)-positive disease after a longer followup (Cancer 2012: 118: 5937 &
JNCI 2012; 104: 406). We now report updated results after a median followup of 12.1 years.
Methods: Eligibility included axillary lymph node positive or high-risk (tumor at least 2 cm) node-negative breast cancer. All
patients received 4 cycles of AC (doxorubicin 60 mg/m2 and cyclophosphamide 600 mg/m2) every 3 weeks, followed by either:
(1) paclitaxel 175 mg/m2 every 3 weeks x 4 (P3), (2) paclitaxel 80 mg/m2 weekly x 12 (P1), (3) docetaxel 100 mg/m2 every 3
weeks x 4 (D3), or (4) docetaxel 35 mg/m2 weekly x 12 (D1). Patients with ER-positive breast cancer also received endocrine
therapy for at least 5 years, including either tamoxifen and/or an aromatase inhibitor. The primary comparisons included taxane
(P vs. D) and schedule (every 3 weeks vs. weekly), and the primary endpoint was DFS, defined as local, regional, and/or distant
relapse, second primary breast cancer, or death without recurrence.
Results: A total of 4954 eligible patients were accrued between October 1999 and January 2002. At the time of this analysis, 90%
of surviving patients were followed for at least 10 years, and there were substantially more DFS events (1639 vs. 1048) and
deaths (1283 vs. 686) than the original report. For the primary comparison, there remains no significant difference in DFS by
taxane (p=0.33) or schedule (p=0.88), although there was a significant interaction between taxane and schedule (p<0.007). When
comparing the standard arm (P3) to the other arms (with a hazard ratio [HR] < 1 favoring the experimental arms using a stratified
Cox model), the HRs for DFS were 0.84 (p = 0.011) for arm P1 and 0.79 (p=0.001) for arm D3 arm, and for overall survival (OS)
were 0.87 (p=0.09) and 0.86 (p=0.054), respectively. When evaluated by subtypes in exploratory analyses, the P1 arm (but not
the D3 arm) was associated with improved DFS (HR 0.69, p=0.01) and OS (HR 0.69, p=0.019) in the 1025 patients with triple
negative breast cancer (TNBC), and the D3 arm (but not P1 arm) was associated with improved DFS (HR 0.76, p=0.004) but not
OS (HR 0.87, p=0.20) in the 2785 patients with ER-positive, HER2-negative/unknown breast cancer (ERBC). In ERBC but not
other subtypes, black race (HR 1.60, p=0.002) and obesity (HR 1.23, p=0.009) were associated with inferior OS, and obesity was
associated with a higher recurrence risk between 3-8 years after diagnosis.
Conclusions: Improvements in DFS observed at 10 years for the P1 and D3 arms compared with the P3 arm are qualitatively
similar but quantitatively less than at 5 years, and the effects on OS are marginal in the overall population. The relative
effectiveness of weekly paclitaxel and every 3 week docetaxel may vary by subtype.

Title: The phase III ICE study: Adjuvant Ibandronate with or without capecitabine in elderly patients with moderate or high risk
early breast cancer
Gunter von Minckwitz1,2, Toralf Reimer3, Jochen Potenberg4, Bettina Conrad5, Ulrike Schrer6, Holger Eidtmann7, Marianne Just8,
Stefan Paepke9, Elmar Stickeler10, Georg Heinrich11, Michael Untch12, Volker Moebus13, Christoph Thomssen14, Christian
Jackisch15, Jens Huober16, Sibylle Loibl1,15, Valentina Nekljudova1 and Ulrike Nitz17. 1German Breast Group, Neu-Isenburg,
Hessen, Germany; 2Universitts-Frauenklinik, Frankfurt am Main; 3Klinikum Sdstadt, Universitts-Frauenklinik, Rostock;
4
Evangelisches Waldkrankenhaus, Berlin, Germany; 5Elisabeth Krankenhaus, Kassel; 6Klinikum Meiningen GmbH, Meiningen;
7
Universittsklinikum Schleswig-Holstein, Kiel; 8Onkologische Schwerpunktpraxis, Bielefeld; 9Universitts-Frauenklinik rechts der
Isar, Mnchen; 10Universitts-Frauenklinik Freiburg; 11Onkologische Schwerpunktpraxis, Frstenwalde; 12Helios Klinikum
Berlin-Buch; 13St dtische Kliniken Hoechst, Frankfurt; 14Universitts-Frauenklinik Halle; 15Frauenklinik Sana-Klinikum Offenbach;
16
Universitts-Frauenklinik Ulm and 17Bethesda Klinikum Moenchengladbach.
Body: Background: Although approximately 50% of newly diagnosed breast cancers arise in women above 65 years old they
are underrepresented in clinical trials. The ICE study was designed to investigate if a mono-chemotherapy with capecitabine in
addition to the third generation bisphosphonate ibandronate will improve the outcome compared to ibandronate alone in elderly
breast cancer patients with medium and high risk primary breast cancer not suitable for standard chemotherapy.
Methods: This is a prospective, multi-center, controlled, open-label, randomized phase III trial. Female patients 65 years with
unilateral or bilateral breast cancer who are either node-positive or high-risk node-negative (tumor size 2 cm, grade >I, and/or
ER-and PR-negative); and a Charlson Comorbidity Index (CMI) 2 received either ibandronate alone for 2 years, 50 mg p.o. daily
alternatively 6 mg i.v. every 4 weeks or the same ibandronate regimen together with capecitabine 2000 mg/m on day 1 14 q
day 22 for 6 cycles to be started within 6 months after axillary dissection. Patients with hormone-sensitive disease received an
endocrine therapy according to guidelines. The primary objective is disease-free survival. A total of 1,394 patients were needed
(5-year DFS improvement from 65% with ibandronate to 71.5% with ibandronate/capecitabine; =0.05, =80%) and we expected
497 events during a median 5 year follow up.
Results: Between 06/2004 and 08/2008, 1409 patients were randomized to ibandronate (N=702) or ibandronate/capecitabine
(N=668) in 172 German centers, of whom 1380 patients started treatment (689 and 691 respectively). Median age was 71 (range
74-80) years; 58.6% of patients had a CMI1, 48.2% had N+, 35.2% grade 3, and 19.2% had ER/PgR negative disease. 83.8% of
patients received all 6 capecitabine cycles. Capecitabine grade 3/4 toxicities were <2% except hand-foot-syndrome (6.8%). 3
patients died in relation with capecitabine treatment. Final DFS analysis will be presented at the meeting.
Conclusion: CALGB 49907 showed that AC or CMF was superior to capecitabine in patients aged 65 years but was associated
with twice as many moderate to severe toxic effects (64% vs 33%) (Muss H et al, NEJM 2012). The ICE study will provide
evidence if capecitabine monotherapy is as active as adjuvant treatment and if it might be an option for frail and/or elderly patients
where standard chemotherapy is considered to be too toxic.

Title: Intrinsic subtypes, PIK3CA mutation, and the degree of benefit from adjuvant trastuzumab in NSABP trial B-31
Soonmyung Paik1, Katherine L Pogue-Geile1, Nan Song1, Patrick G Gavin1 and Seong-Rim Kim1. 1NSABP, Pittsburgh, PA.
Body: Purpose: Considerable molecular heterogeneity exists among HER2-positive breast cancer regarding gene expression
and mutation profiling. Evidence from preclinical, clinical neoadjuvant, and metastatic clinical trials suggested PIK3CA mutational
status and PAM50 intrinsic subtype of a tumor were markers of response to anti-HER2 therapies. We evaluated the predictive
value of these two biomarkers in the adjuvant setting using archived tumor blocks from NSABP trial B-31.
Methods: Expression data for 49 genes using the nCounter platform was used to generate PAM50 intrinsic subtypes for 1,578
archived tumor blocks from B-31. Five PIK3CA hotspot mutations were examined by mass spectrometry of the primer extension
products in a randomly selected subset (N=671). We examined the heterogeneity of trastuzumab treatment effect across different
subsets defined by each marker using Cox regression and disease free survival as the endpoint.
Results: 741/1578 (47.0%) tumors were classified as HER2E subtype; and 166/671 (24.7%) had PIK3CA mutations. Hazard
ratios (HR) for trastuzumab in HER2E and other subtypes were 0.44 (95%CI: 0.34-0.58, p<0.001) and 0.47 (95% CI: 0.35-0.62,
p<0.001), respectively (interaction p=0.67). HRs for trastuzumab in PIK3CA wild type and mutated were 0.51 (95%CI: 0.37-0.71,
p<0.001) and 0.44 (95%CI: 0.24-0.82, p=0.009), respectively (interaction p=0.64).
Conclusion: Unlike results seen in the metastatic and neoadjuvant clinical trials, PIK3CA and PAM50 intrinsic subtypes were not
biomarkers for differential response to trastuzumab in NSABP B-31. These data suggest that results from the metastatic and
neoadjuvant setting may not be always applicable to the adjuvant setting.
Support: National Cancer Institute, Department of Health and Human Services, Public Health Service, Grants U10-CA-12027,
U10-CA-69651, U10-CA-37377, and U10-CA-69974, and by a grant from the Pennsylvania State Department of Health. The latter
Department specifically disclaims responsibility for any analysis, interpretations or conclusions.

Title: Mutational analysis of CALGB 40601 (Alliance), a neoadjuvant phase III trial of weekly paclitaxel (T) and trastuzumab (H)
with or without lapatinib (L) for HER2-positive breast cancer
Katherine A Hoadley1, William T Barry2, Brandelyn N Pitcher3, Joel S Parker1, Matthew D Wilkerson1, William Irvin, Jr1,4, Norah
Lynn Henry5, Sara M Tolaney6, Chau Dang7, Ian E Krop6, Donald A Berry8, Elaine R Mardis9, Charles M Perou1, Eric P Winer6,
Clifford A Hudis7 and Lisa A Carey1. 1University of North Carolina, Chapel Hill, NC; 2Alliance Statistics and Data Center,
Dana-Farber Cancer Institute, Boston, MA; 3Alliance Statistics and Data Center, Duke University, Durham, NC; 4Bon Secours
Cancer Institute, Midlothian, VA; 5University of Michigan, Ann Arbor, MI; 6Dana-Farber Cancer Institute, Boston, MA; 7Memorial
Sloan Kettering Cancer Center, New York, NY; 8Alliance Statistics and Data Center, MD Anderson Cancer Center, Houston, TX
and 9Genome Institute, Washington University School of Medicine, St Louis, MO.
Body: Background: In CALGB 40601, the HER2-Enriched (HER2-E) molecular subtype had significantly higher pathologic
complete response (pCR) rates regardless of treatment arm (single HER2-targeting with T+L or T+H, dual targeting with T+H+L)
(Carey et al, ASCO 2014). A TP53 mutation gene expression signature was significant in a multivariable analysis as were
treatment, molecular subtype, proliferation, and an immune cell genomic signature. Mutational analysis is now available for this
sample set.
Methods: 265 of 305 enrolled patients (pts) had RNA sequencing (RNAseq) of pre-treatment biospecimens; 181/265 had whole
exome sequencing (WES) available from tumor and matched normal blood. Somatic mutations were detected by the program
UNCeqR, which integrates WES and RNAseq. We examined the association of mutations with in-breast pCR, molecular
subtypes, and gene expression signatures.
Results: In this subset, there were 57 HER2-E, 58 Luminal A, 51 Luminal B, 9 Basal-like, 4 Normal-like, and 2 Claudin-low. The
pCR rate was 45% (51% THL, 47% TH and 34% TL), consistent with the entire study population. TP53 was the most frequently
mutated gene (56%); frequency varied by molecular subtype (Fisher p<0.0001) with the highest in the HER2-E (88%). Type of
mutation also varied by molecular subtype: 34% of TP53 mutations in HER2-E pts were nonsense or frame shift mutations
compared to 20% in Luminal B and 11% in Luminal A. The presence of a TP53 mutation was significantly associated with
achieving pCR (59% compared to 28% in wildtype; odds ratio=3.7, p<0.0001) which did not vary by treatment arm. TP53 mutation
status by WES was strongly associated with a gene-expression based predictor (AUC=0.85, p<0.001), suggesting the
RNAseq-based signature could be used as a surrogate measure of genotype. PIK3CA mutations were present in 36 pts (20%);
33/36 (92%) were in exons 9 and 20. PIK3CA mutations varied moderately among subtypes and were most prevalent in Luminal
B (31%) and HER2-E (25%). Rates of pCR did not vary by PIK3CA mutation status (39% vs 47% in wildtype, p=0.46). GATA3
mutations were identified in 7 pts (4 Luminal A, 3 Luminal B), but only 1 pt achieved pCR. ERBB2 mutations were found in 7 pts:
2 HER2-E, 2 Luminal A, 3 Luminal B. Two were previously identified (the lapatinib-sensitive activating mutation V777L and the
lapatinib-resistant mutation L755S), both were trastuzumab resistant in experimental models (Bose et al, Cancer Discovery
2013). Consistent with these results, the V777L pt achieved pCR on the THL arm; the L755S pt did not achieve pCR on the TL
arm.
Conclusions: TP53 mutation is a frequent, clinically important event in HER2-positive disease and predicts pCR to
chemotherapy plus HER2-targeting. Frequency and type of mutation was dependent on molecular subtype within this clinically
HER2-positive cohort. Ongoing analyses are comparing WES data between pre- and post-treatment samples as well as
investigating copy number and clonality. This research is supported in part by funds from GlaxoSmithKline and grants from the
Breast Cancer Research Foundation.

Title: 16 year long-term follow-up of the IBIS-I breast cancer prevention trial
Jack Cuzick1, Ivana Sestak1, Simon Cawthorn2, Hisham Hamed3, Kaija Holli4, Anthony Howell5 and John F Forbes6. 1Centre for
Cancer Prevention, Queen Mary University, London, United Kingdom; 2North Bristol NHS Trust, Bristol, United Kingdom; 3London
Bridge Hospital, London, United Kingdom; 4Tampere University, Tampere, Finland; 5Institute of Cancer Sciences, University of
Manchester, Manchester, United Kingdom and 6School of Medicine and Public Health, University of Newcastle, Newcastle,
Australia.
Body: Background: Several randomised clinical trials have shown the benefit of tamoxifen in healthy women to reduce their risk
of breast cancer. Here, we report the blinded median 16 year follow-up of the IBIS-I trial to update the long-term prevention of
breast cancer with tamoxifen treatment.
Methods: 7154 pre- and postmenopausal women were randomised to receive daily 20mg tamoxifen (N=3579) or matching
placebo (N=3575) for 5 years. The primary endpoint of this analysis was the occurrence of breast cancer (invasive and ductal
carcinoma in situ (DCIS)). Secondary endpoints included overall mortality, other cancers, and breast cancer specific mortality.
Cox proportional hazard models were used to assess occurrence of breast cancer and survival. All statistical tests were
two-sided.
Results: After a median of 16.2 years (IQR 14.4 to 17.7) of follow-up, a total of 589 breast cancers have been reported
(tamoxifen: 246 (6.9%) vs. placebo: 343 (9.6%)). Tamoxifen reduced the incidence of all breast cancer overall by 29% (HR=0.71
(0.60-0.83), P<0.0001) (Figure 1). Invasive ER-positive (ER+) breast cancers were reduced by 35% (HR=0.65 (0.53-0.80),
P<0.0001) (Figure 1), but no effect was seen for invasive ER-negative (ER-) breast cancers (HR=1.06 (0.71-1.58), P=0.8). A
non-significant 30% reduction in DCIS was seen with tamoxifen (36 vs. 51, HR=0.70 (0.46-1.07); P=0.1). The overall risk
reduction was similar in years 0-10 (HR=0.71) and years 10-20 (HR=0.70). Similar effects were seen in pre- and postmenopausal
women (HR 0.71 vs. 0.71). All-cause mortality was non-significantly increased in women randomised to tamoxifen (173 vs. 158,
OR=1.10 (0.88-1.38), P=0.4). The excess in deaths with tamoxifen is smaller than in the 96 month update. No differences in
breast cancer mortality was seen (24 tamoxifen vs. 27 placebo; OR=0.89 (0.49-1.60), P=0.7). A non-significant increase in other
cancers than breast were reported by women on tamoxifen (350 vs. 315, OR=1.12 (0.95-1.32); P=0.2). Specifically more
endometrial cancers (28 vs. 17), non-melanoma skin cancers (108 vs. 85), and lung cancer (32 vs. 20) were found in those
randomised to tamoxifen.
Conclusion: This updated analysis of the IBIS-I trial confirms the significant reduction in breast cancer occurrence with tamoxifen
in the post-treatment follow-up period. These results indicate tamoxifen has a long-term preventive effect on invasive ER+ breast
cancer in both pre- and postmenopausal women.

Title: Randomized comparison of adjuvant tamoxifen (T) plus ovarian function suppression (OFS) versus tamoxifen in
premenopausal women with hormone receptor-positive (HR+) early breast cancer (BC): Analysis of the SOFT trial
Prudence A Francis1, Meredith M Regan1, Gini F Fleming1, Istvan Lang1, Eva M Ciruelos1, Meritxell Bellet1, Herve Bonnefoi1,
Miguel A Climent1, Lorenzo Pavesi1, Harold J Burstein1, Silvana Martino1, Nancy E Davidson1, Charles E Geyer, Jr1, Barbara A
Walley1, Robert E Coleman1, Pierre Kerbrat1, Manuela Rabaglio-Poretti1, Alan S Coates1, Aron Goldhirsch1 and Richard D
Gelber1. 1SOFT Investigators, International Breast Cancer Study Group, Breast International Group, and North American Breast
Cancer Group, Bern, Switzerland.
Body: Background: The value of adding OFS to tamoxifen in premenopausal women with HR+ early BC is uncertain. SOFT was
designed to determine the value of adding OFS to tamoxifen and the role of adjuvant therapy with the aromatase inhibitor
exemestane (E) plus OFS in premenopausal women.
Methods: From Nov 2003 - Jan 2011, 3066 premenopausal women with HR+ BC were randomized to 5 years of tamoxifen vs
tamoxifen+OFS vs exemestane+OFS. OFS was by choice of GnRH agonist triptorelin, oophorectomy or radiation. SOFT was
stratified by the use of prior chemotherapy; 47% received no chemotherapy and 53% remained premenopausal after prior
chemotherapy, confirmed by estradiol within 8 months of completion. The comparison of T+OFS vs T alone was the primary
objective, to be tested at a 2-sided 0.05 level, when median follow-up was at least 5 yrs. The comparison of E+OFS vs T was a
secondary objective. The primary end point was invasive DFS. Secondary end-points included invasive breast cancer-free
interval (BCFI), distant recurrence-free interval (DRFI) and overall survival (OS).
Results:
As expected, the prior chemotherapy cohort represented a higher risk group (table).
T+OFS vs T
No chemo
(n=949)
Prior Chemo
(n=1084)
Primary Analysis
(n=2033)
Median age yrs
46
40
43
Node positive
9%
57%
35%
Tumor > 2 cm
14%
47%
32%
Grade 1
41%
14%
27%
Grade 3
7%
35%
22%
70/47/13/10
229/213/172/96
299/260/185/106
Events @ 67 months
DFS/BCFI/distant/deaths
After a median follow-up of 67 months, 5-yr DFS was 86.6% in the T+OFS group and 84.7% in the tamoxifen group (HR=0.83;
95%CI, 0.86-1.04; P=0.10). Multivariable allowance for prognostic factors suggested a greater treatment effect (HR 0.78; 95% CI,
0.62-0.98). Overall survival is not mature with 5-yr OS of 96.7% for T+OFS and 95.1% for tamoxifen (HR 0.74; 95%CI, 0.51-1.09;
P=0.13). In the no-chemotherapy cohort, one-third of events were not breast-cancer related and BCFI was > 95% with tamoxifen
alone. The cohort who remained premenopausal after chemotherapy had a 4.5% absolute improvement in 5-yr BCFI with T+OFS
vs T (table). In the chemotherapy cohort, 5-yr overall survival was 94.5% for T+OFS and 90.9% for T (HR=0.64; 95% CI,
0.42-0.96).
Prior
Chemo
Cohort
T
Events/Pts 5-yr BCFI
116/542
78.0%
Hazard Ratio (vs T)
T+OFS
97/542
E+OFS
80/544
82.5%
HR 0.78 (95% CI, 0.60-1.02)
85.7%
HR 0.65 (95% CI, 0.49-0.87)
5-yr DRFI
T
90/542
83.6%
T+OFS
82/542
84.8%
HR 0.87 (95%CI, 0.64-1.17)
E+OFS
67/544
87.8%
HR 0.72 (95% CI, 0.52-0.98)
Non-adherence with OFS reached 22% at 4 years. Targeted > grade 3 toxicities were reported for 31.3% of T+OFS group and
23.7% of tamoxifen group. Menopausal symptoms, depression, musculoskeletal complaints, hypertension and diabetes were
more frequent with T+OFS. Osteoporosis was reported in 5.8% for T+OFS and 3.5% for T.
Conclusions: Adding ovarian suppression to tamoxifen did not show significant benefit in the overall population in SOFT after 67
months median follow-up. However, for the cohort of women who received chemotherapy and who remained premenopausal, the
addition of ovarian suppression improved breast cancer outcomes, and further improvement was seen with the use of
exemestane plus ovarian suppression.

Title: Patient-reported endocrine symptoms, sexual functioning and quality of life (QoL) in the IBCSG SOFT trial: Adjuvant
treatment with tamoxifen (T) alone versus T plus ovarian function suppression (OFS) in premenopausal women with hormone
receptor-positive (HR+) early breast cancer (BC)
Karin Ribi1, Weixiu Luo1, Juerg Bernhard1, Prudence A Francis1, Meritxell Bellet1, Harold J Burstein1, Lorenzo Pavesi1, Vani
Parmar1, Carlo Tondini1, Marilena Visini1, Roberto Torres1, Per Karlsson1, Simon Spazzapan1, Antoni Avella1, Thomas Ruhstaller1,
Fabio Puglisi1, Meredith M Regan1, Alan S Coates1, Richard D Gelber1 and Gini F Fleming1. 1SOFT Investigators and International
Breast Cancer Study Group, Bern, Switzerland.
Body: Background: SOFT efficacy results reported at this SABCS show that T+OFS provides improved disease control
compared with T for the cohort of patients (pts) who received prior chemotherapy (chemo), but relatively little is known about
treatment-related endocrine symptoms, sexual function and QoL in premenopausal women receiving adjuvant T+OFS compared
with T.
Methods: From Nov 2003 to Apr 2011, 1,722 premenopausal pts with HR+ BC were enrolled and included in the QoL analysis of
the randomized phase III trial SOFT, to receive adjuvant treatment with 5 yrs T or T+OFS. A third group received
exemestane+OFS and is not included in this report. SOFT enrolled two cohorts: pts who received no chemo, and those who
received prior chemo with confirmed premenopausal estradiol levels within 8 mos of completing chemo. Pts completed a
questionnaire consisting of global and symptom-specific indicators at baseline, every 6 mos for the first 24 mos, and annually yrs
3 to 6. Differences in change of QoL from baseline between the two treatments were tested at short-, intermediate-, and long-term
(6, 24 and 60 mos post-randomization, respectively) for 13 symptoms and 4 global QoL indicators using mixed models with
repeated measures, overall and by chemo stratum.
Results: Changes of global QoL indicators (mood, physical wellbeing) from baseline were small and similar between treatments
over the whole observation period, but treatment differences were seen with respect to symptom-specific indicators, especially for
patients in the no chemo cohort. Overall, pts on T+OFS were substantially more affected by hot flushes than pts on T alone at
short- and intermediate-term (each p<.0001). The change of hot flushes from baseline improved for the T+OFS but not for the T
alone group over the period of 60 mos. Pts on T+OFS reported more loss of sexual interest (p<.0001) and sleep problems
(p<.0001) at short-term, and more vaginal dryness over the whole treatment period (each p<.01). Pts on T alone reported more
vaginal discharge up to 60 mos (each p<.05), but only among those pts who did not receive prior chemo. Symptom-specific
treatment differences (hot flushes, sleep problems, vaginal discharge) were less pronounced in pts who had received prior
chemo. Although pts receiving no prior chemo had less improvement in coping and greater treatment burden with T+OFS vs. T
(p<.05), no such treatment differences were seen for pts in the prior chemo cohort.
Conclusion: Global QoL (mood and physical wellbeing) did not differ between groups. Overall, pts receiving T+OFS experienced
worse endocrine symptoms and sexual functioning than those receiving T alone during the first two years of treatment; most
differences between treatments were no longer apparent thereafter. Differences between T+OFS and T with respect to impaired
symptom-specific QoL, being burdened by treatment, and having delayed adaptation during the first two yrs of treatment were
less pronounced for pts who received chemo prior to enrolling in SOFT, the cohort that benefited most from OFS in terms of
disease control.

Title: Macrophage-specific deletion of STAT5 disrupts normal mammary gland development and accelerates mammary
tumorigenesis
Nicholas J Brady1, Michael A Farrar1 and Kathryn L Schwertfeger1. 1University of Minnesota, Minneapolis, MN.
Body: Fibroblast growth factor receptor 1 (FGFR1) is amplified in 10% of human breast cancers and the ligands for FGFR1 are
overexpressed in 60% of triple negative breast cancers. Previous studies using a mouse model of FGFR1-induced mammary
tumorigenesis demonstrated that FGFR1 activation in mammary epithelial cells recruits macrophages to hyperplastic regions
where they promote angiogenesis and epithelial cell proliferation. Clinically, patients with increased numbers of tumor-associated
macrophages have increased risk of relapse and decreased overall survival. While these previous studies demonstrate a role for
macrophages during tumor initiation and progression the specific mechanisms that orchestrate the pro-tumor response are poorly
understood. Our work uses numerous in vitro and in vivo models to study the dynamic interactions between mammary epithelial
cells and macrophages. In an initial screen, we identified the transcription factor signal transducer and activator of transcription 5
(STAT5) as being rapidly activated in macrophages in response to tumor cell-derived factors and modulating pro-tumor functions
of macrophages. To assess the role of STAT5 in macrophages during normal mammary gland development, we created a
macrophage-specific conditional knockout mouse line with Cre recombinase expression driven by the Csf1r promoter. Mice
carrying a macrophage-specific deletion of STAT5 (Stat5 fl/fl ; Csf1r-Cre +) show increased epithelial cell proliferation and
impaired ductal elongation at 6 weeks of age compared to littermate controls (Stat5 fl/fl ; Csf1r-Cre -). Based on these data, we
hypothesized that the loss of STAT5 in macrophages would accelerate mammary tumorigenesis. To test this, we crossed the
macrophage-specific STAT5 knockout mice with a mouse model of FGFR1-induced mammary tumorigenesis. FGFR1 activation
was induced in the mice for 2 or 4 weeks beginning at 6 weeks of age. Consistent with previous reports, activation of FGFR1 in
mammary epithelial cells results in increased proliferation and lateral bud formation. Surprisingly, mice with STAT5-deficient
macrophages show signs of accelerated tumorigenesis, with increased proliferation and numerous instances of hyperplasia in the
first 2 weeks of FGFR1 activation. Future studies will focus on understanding the downstream effects of STAT5 deletion in
macrophages and the therapeutic potential of targeting this pathway in vivo. All together, these data demonstrate that STAT5 is a
critical factor that allows macrophages to regulate normal mammary gland development. In addition, this work illustrates that the
loss of STAT5 in macrophages can cooperate with a common genetic event in mammary epithelial cells during breast cancer
initiation.

Title: Principles governing A-to-I RNA editing in breast cancer transcriptome
Debora Fumagalli1, David Gacquer2, Franoise Roth1, Anne Lefort2, Frederick Libert2, David N Brown1, Naima Kheddoumi1,
Adam Shlien3, Tomasz Konopka2, Roberto Salgado1, Denis Larsimont1, Kornelia Polyak4, Karen Willard-Gallo1, Christine
Desmedt1, Martine Piccart1, Marc Abramowicz5, Peter J Campbell3, Vincent Detours2 and Christos Sotiriou1. 1Jules Bordet
Institute, Brussels, Belgium; 2Universit Libre de Bruxelles (ULB), Brussels, Belgium; 3Wellcome Trust Sanger Institute, Hinxton,
United Kingdom; 4Dana-Farber Cancer Institute, Boston, MA and 5Hospital Erasme, Brussels, Belgium.
Body: Background: Messenger RNA (mRNA) is the target of a series of post-transcriptional modifications that can affect its
structure and stability, one of the most relevant being RNA editing. The most common form of RNA editing in humans is of the
A-to-I type and is catalyzed by the adenosine deaminase acting on RNA (ADAR) family of enzymes. Currently, little is known
about how RNA editing operates in cancer. The main objectives of this study were to investigate and characterize the extent of
A-to-I RNA editing in breast cancer (BC) and to define the principles governing the editing process in this as well as other
cancers.
Material and Methods: The exome and transcriptome of 58 BC samples representing the four main known subtypes, namely TN,
HER2+, luminal A and luminal B, and 10 matched normal samples were sequenced using the Illumina platform. For the same
series, gene expression and copy number profiles were obtained using the Affymetrix platform. RNA-DNA single nucleotide
differences (RDDs) were called using a pipeline in line with the most updated bioinformatics tools and validated in an independent
cohort of 15 BC samples and breast cell lines.
Results: Overall, we detected 16,027 RDDs present in one or more samples, with all possible base changes represented. Among
these, 560 RDDs were located in Alu regions and were all of the A-to-I type consistent with the notion that A-to-I editing occurs
predominantly in forward-facing Alu forming double stranded RNA duplexes processed by ADAR.
We found that the same sites were edited in normal and tumor breast tissues. However, the editing frequency was significantly
higher in tumors compared to matched normal breast tissues. Moreover, high editing frequencies were observed in samples in
which more editing sites were detectable and/or in which ADAR expression was high.
We identified two key factors that independently determine ADAR expression and therefore A-to-I RNA editing in breast and the
majority of other human cancers: 1) the type-I interferon response in tumors and 2) gains in ADAR copy number. The mean
editing frequency was found to be significantly correlated with the expression of STAT1 and other type I interferon target genes,
both in our patient series, in a large pool of BC datasets and a panel of normal and breast cancer cell lines treated with interferon
(IFN) , , or for 1, 2 and 5 days in vitro. Moreover, the association between editing and STAT1 expression or ADAR
amplification was validated in 19 additional cancer types obtained from the TCGA dataset.
Conclusions: Our work, which represents the largest survey so far on RNA editing in BC, shows that A-to-I editing is a pervasive
and well-controlled phenomenon in cancer that can drive aberrant transcriptome expression in breast and potentially the vast
majority of cancers. Moreover, it suggests that the immune response can profoundly impact the transcriptome sequence in tumor
cells and thereby influence the internal mechanisms governing their behavior.

Title: Identification of a notch-driven breast cancer stem cell gene signature for anti-notch therapy in an ER+ presurgical window
model
Kathy S Albain1, Andrei Y Zlobin1, Kyle R Covington2, Brian T Gallahger1, Susan G Hilsenbeck2, Cheryl M Czerlanis1, Shelly Lo1,
Patricia A Robinson1, Ellen R Gaynor1, Constantine Godellas1, Davide Bova1, Kathy Czaplicki1, Barbara Busby1, Patrick J Stiff1,
Suzanne AW Fuqua2, Lucio Miele3 and Clodia Osipo1. 1Loyola University Chicago, Cardinal Bernardin Cancer Center, Maywood,
IL; 2Baylor College of Medicine, Houston, TX and 3Louisiana State University, New Orleans, LA.
Body: Background: Resistance to endocrine therapy (ET; tamoxifen or aromatase inhibitors, AI) for ER+ breast cancer is a
major cause of mortality and new treatment paradigms are needed. Cancer stem cells drive breast cancer growth and are
resistant to standard therapy. Notch signaling aids survival of these resistant stem cells and is inhibited by gamma-secretase
inhibitors (GSI). We showed combining GSI with ET in mice caused shrinkage of breast cancer tumors. A presurgical window
biomarker modulation model was used to confirm this discovery in humans.
Methods: The GSI MK-0752 was added to ET in patients before definitive surgery (ClinTrials.gov NCT00756717). There were 3
biopsies: day 0 (prestudy), day 14 (after ET alone), and day 25 at definitive surgery (after continued ET plus MK-0752, 350 mg
orally 3d on, 4d off, 3d on). Biopsies were analyzed for genes increased or decreased by GSI, to confirm that Notch and cancer
stem cell pathways were inhibited. Real-time PCR was used to validate expression of genes identified in pathway analyses of
microarray datasets generated from the biopsies. Mammosphere-forming assays were performed to confirm that ET+GSI impacts
breast cancer stem cells. The qRT-PCR data were evaluated using ANOVA with repeated measures and ANOVA was performed
on mammosphere results.
Results: The accrual goal was met and therapy well-tolerated in 20 evaluable women (PSABCS 2011, abs# S1-5). Of 33 genes
identified by analysis of expression microarrays, 19 genes (FDR<8%) were impacted significantly by GSI+ ET (3 increased, 16
decreased) compared to initial biopsy and/or ET alone. Genes with increased expression were DAXX, NOXA (both pro-apoptotic)
and LNFG (tumor suppressor). Six of 16 genes that decreased (NOTCH1, NOTCH4, HEYL, HES1, HES5, and HEY2) are Notch
pathway-associated genes. The GSI decreased expression of 3 genes from cell cycle and proliferative pathways (Ki67, CCND1,
CCNA2) and inhibited 2 genes expressed in cancer stem cells (RUNX1 and ALDH1A1). Five genes directly/indirectly regulated by
Notch were decreased by GSI (RICTOR, RPTOR, MMP7, ADAM19, and PRH). Estrogen deprivation for 3 days, mimicking short
exposure to an AI, increased mammosphere-forming ability of ER+ breast cancer cells more than 2 fold. The GSI MRK-003
blocked this mammosphere formation by 95%-98%.
Conclusions: A 7-day course of the GSI MK-0752 added to ET in the presurgical window had significant biomarker responses:
decrease in Notch signaling, cancer stem cell genes, proliferation-associated genes, the mTORC1 and 2 complex genes RICTOR
and RPTOR, metalloproteinases that promote metastasis, and PRH; as well as increase in 3 key genes that promote apoptosis
and tumor suppression. These results suggest that 1) GSI inhibited the intended Notch pathway, 2) putative breast cancer stems
cells can be targeted by this strategy, and 3) the biomarkers identified create a gene signature for anti-Notch therapy in ER+
breast cancer. Validation of efficacy of the GSI+ET therapy combination and this gene signature in a clinical trial is planned.
Support: Breast Cancer Research Foundation (research grant), Merck Oncology (drug/arrays), Swim Across America (clinical
trial costs), and DOD BC073237 (KRC).

Title: Predicting pre-surgical neoadjuvant chemotherapy response in breast cancer using diffuse optical spectroscopic imaging
(DOSI): Results from the ACRIN 6691 study
Bruce J Tromberg1, Zheng Zhang2, Anais Leproux1, Thomas D O'Sullivan1, Albert E Cerussi1, Philip Carpenter1, Rita Mehta1,
Darren Roblyer5, Wei Yang6, Keith D Paulsen4, Brian W Pogue4, Shudong Jiang4, Peter Kaufman4, Arjun G Yodh7, So-Hyun
Chung7, Mitchell Schnall7, Brad Snyder2, Nola Hylton8, David A Boas9, Stefan A Carp9, Steven J Isakoff9 and David Mankoff7.
1
University of California, Irvine, CA; 2Brown University, Providence, RI; 3MD Anderson Cancer Center, Houston, TX; 4Dartmouth
College, Hanover, NH; 5Boston University, Boston, MA; 6MD Anderson Cancer Center; 7University of Pennsylvania, Philadelphia,
PA; 8University of California, San Francisco, CA and 9Massachusetts General Hospital.
Body: Background: DOSI is an experimental imaging technique that employs risk-free near-infrared light for quantitative
measurements of breast tissue perfusion, metabolism, and composition without exogenous contrast. A multi-center ACRIN 6691
study was designed to evaluate whether changes from baseline to mid-therapy in a DOSI-derived Tissue Optical Index (TOI)
could predict pathologic complete response (pCR) in breast cancer neoadjuvant chemotherapy (NAC).
Methods: DOSI instruments were constructed at the University of California, Irvine and delivered to 6 participating sites. Bedside
measurements were conducted by scanning a handheld probe over a region of interest (up to 10 x 10 cm2) on both breasts.
Instruments were standardized and validated using a common set of tissue simulating phantoms and protocols. DOSI-derived
near-infrared absorption and scattering spectra (650-1000 nm) were used to calculate the tissue concentration of oxy- and
deoxyhemoglobin (ctO2Hb, ctHHb), water (ctH2O), %lipid and the tissue optical index (TOI=ctHHb x ctH2O/%lipid) in each probe
location. Baseline to mid-therapy changes in the tumor to normal (T/N) TOI ratio were evaluated from DOSI images as the
primary imaging endpoint for predicting clinical outcome (pCR). 60 female breast cancer patients (ages 28-69 years, mean
48.911), with locally advanced disease (tumors >2cm) were enrolled across the 6 institutions. DOSI measurements were
performed at baseline, during the first week of therapy, at midpoint, and at the completion of NAC. Logistic regression was used
to assess the association between pathologic complete response (pCR) and % change in T/N TOI from baseline to mid-therapy.
In addition, area under the receiver operating characteristic curve (ROC AUC) and its corresponding 95% confidence interval
were calculated.
Results: Of the 34 participants (mean age 48.4 10.7) with complete and evaluable data, 10 (29%) achieved pCR as determined
by central pathology review. The % change in TOI ratio ranged from -82% to 321%, with a median of -36%. Using -40% as a
threshold, we found that subjects in the group with a 40% or more decrease in T/N TOI were more likely to be pCR (p=0.0586,
OR=4.667, 95% CI: 0.945 to 23.038). The % change in TOI ratio from baseline to mid-therapy has an AUC of 0.604 (95% CI:
0.394 to 0.814) to distinguish pCR from non-pCR.
Conclusions: DOSI has been successfully implemented in a multi-center setting and changes in T/N TOI are a promising
predictor of NAC clinical outcome (pCR). A larger study population is needed to fully assess the utility of TOI and other DOSI
imaging endpoints for guiding therapies and predicting NAC response in individual subjects.
ACRIN receives funding from the NCI through U01 CA079778 and U01 CA080098.

Title: Impact of intrinsic subtype by PAM50 and other gene signatures on pathologic complete response (pCR) rates in
triple-negative breast cancer (TNBC) after neoadjuvant chemotherapy (NACT) +/- carboplatin (Cb) or bevacizumab (Bev):
CALGB 40603/150709 (Alliance)
William M Sikov1, William T Barry2, Katherine A Hoadley3, Brandelyn N Pitcher4, Baljit Singh5, Sara M Tolaney6, Charles S
Kuzma7, Timothy J Pluard8, George Somlo9, Elisa R Port10, Mehra Golshan6, Donald A Berry11, Olwen M Hahn12, Lisa A Carey3,
Charles M Perou3, Clifford A Hudis13 and Eric P Winer6. 1Alpert Medical School of Brown University, Providence, RI; 2Alliance
Statistics and Data Center, Dana-Farber Cancer Institute, Boston, MA; 3UNC Lineberger Comprehensive Cancer Center, Chapel
Hill, NC; 4Alliance Statistics and Data Center, Durham, NC; 5New York University Medical Center, New York, NY; 6Dana-Farber
Cancer Institute, Boston, MA; 7Southeast Cancer Control Consortium, Winston-Salem, NC; 8Washington University, St Louis
Medical Center, St Louis, MO; 9City of Hope Comprehensive Cancer Center, Duarte, CA; 10Mount Sinai Medical Center, New
York, NY; 11Alliance Statistics and Data Center, MD Anderson Cancer Center, Houston, TX; 12University of Chicago Medical
Center, Chicago, IL and 13Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background: Adding either Cb or Bev to standard NACT significantly increases pCR rates in TNBC (Sikov et al, SABCS
2013). Genomic analysis may help us to identify determinants of response within this clinical phenotype.
Methods: Patients (pts) with clinical stage II-III TNBC received weekly paclitaxel x 12 followed by ddAC x 4 +/- Cb and/or Bev.
Pre-treatment biopsies were collected in formalin, RNAlater and OCT; residual disease at surgery was biopsied when possible.
Illumina mRNA sequencing (RNAseq) was performed. Gene expression values were normalized to a TCGA subset of clinically
TNBC samples prior to downstream analysis. pCR was defined as the absence of residual invasive cancer in the breast (ypT0/is).
For each molecular signature, prognostic (effect on pCR in the overall study population) and predictive (effect of the addition of
Cb or Bev, separately, on pCR) relationships were explored with logistic regression models.
Results: PAM50 subtype analysis was performed on 367 pre-treatment samples (of 443 pts who started NACT); pCR results
were available for 360, comprising the analysis subset. 87% of these displayed a basal-like gene expression pattern, 2%
claudin-low, 4% HER2-enriched, <1% luminal A and 7% normal-like. In pts with basal-like tumors, pCRs rose from 47% to 61%
with the addition of Cb (p=0.014), an increment which did not differ significantly from the overall study population (adding in the
small number of non-basal-like tumors) (interaction p=0.93). In contrast, the addition of Bev increased pCRs in basal-like tumors
from 45% to 64% (p=0.0009), while reducing pCRs in non-basal-likes from 60% to 43% (interaction p=0.024); thus, a basal-like
gene expression pattern was predictive of benefit from Bev. Expression of a variety of immune signatures (B cell, T cell, IgG) was
positively associated with pCR, but not predictive of increased benefit from either Cb or Bev. High expression of the HER2
amplicon signature was uncommon and not prognostic for pCR overall but was associated with reduced benefit from Cb
(interaction p = 0.025). High proliferation, high p53 mutation and low IE (estrogen signaling) signatures were prognostic for higher
pCR rates and predictive of benefit from Bev (interaction p=0.031, 0.0017, 0.0002, respectively). In basal-like pts with residual
disease, surgical samples often (52%) displayed a normal-like PAM50 pattern, though this might be due to contamination in low
volume residual disease.
Conclusions: Selection criteria led to accrual of a high % of pts with basal-like tumors, limiting our ability to assess prognostic or
predictive impact of intrinsic subtype on pCR. Given that limitation, the magnitude of pCR benefit with Cb was consistent across
subtypes, while a basal-like pattern was predictive of greater pCR increment with Bev. Ongoing studies will test a large number of
other gene signatures and biomarkers, including the Lehmann et al subtypes. Recognition of clinically relevant subpopulations
within TNBC may distinguish pts likely to achieve a pCR from those for whom an investigational approach might be considered.

Title: The Connecticut experiment: 4 years of screening women with dense breasts with bilateral ultrasound
Jean M Weigert1. 1Hospitals of Central Connecticut, New Britain, CT and 2Mandell and Blau MD's PC, New Britain, CT.
Body: OBJECTIVE: To determine if the addition of screening breast ultrasound in women with mammographically normal but
dense breasts has increased breast cancer detection over the 4 years since the legislation was first enacted in 2009.
METHODS: The study utilized a retrospective chart review. Data collected included: (1) total number of screening mammograms;
(2) total number of dense breast screening ultrasounds; (3) screening ultrasound Breast Imaging Reporting Data System
(BI-RADS) code results; (4) biopsy results; and (5) demographic data on women with malignant biopsies. Data was obtained from
included sites from October 2009 through December 2013.
FINDINGS: Data was collected from 2 Radiology practices with a total of 5 sites.
In year 1, 30670 mammograms and 2706 screening ultrasounds were performed. In year 2, 32050 mammograms and 3351
ultrasounds; in year 3, 32230 mammograms and 4128 ultrasounds and in year 4, 27937 mammograms and 3331 ultrasounds
were performed.
In year 1, 151 ultrasounds were BIRADS 4 or 5 with 11 cancers detected. The PPV was 7.1%; the detection rate was 4/1000 and
22% of eligible women had the study (assuming 40% of women have dense breasts). In year 2, 180 ultrasounds were BIRADs 4
and 5 with 11 cancers detected. The PPV was 6.1% and the cancer detection rate was 3.2/1000 with 26% of eligible women
having the study. In year 3 148 ultrasounds were BIRAD 4 and 5 with 13 cancers detected. The PPV was 8.1% and the cancer
detection rate was 3.2/1000 with 32% of eligible women having the study. In year 4, 53 ultrasounds were BIRAD 4 and 5 with 11
cancers. The PPV was 17.2% with the cancer detection rate of 3.3/1000 and 28% of eligible women having the study.
Four Years of Screening Breast Ultrasound Data
Year 1
Year 2
Year 3
Year 4
Screening Mammograms
30670
32050
32230
27937
Breast Ultrasounds for Dense Breasts
2706
3351
4128
3331
BIRAD 4 and 5 Ultrasounds
151
180
178
53
Cancers
11
11
13
11
PPV
7.1
6.1
8.1
17.2
Number Cancers per 1000 Screened
4.0
3.2
3.2
3.3
% Eligible Screened
22.1
26.1
32.0
28.3
CONCLUSIONS: Based on the data collected from these sites, screening breast ultrasound in women with dense breast
parenchyma detects mammographically occult malignancy. Over the 4 years studied, the PPV improved from 7% to 17.2%
indicating that the selection of lesions biopsied was more accurate with fewer false positives. The rate of detection in the first year
was 4.0/1000 and then remained stable at 3.2/1000 in the three subsequent years. Of concern, the number of eligible women
who elect to have the additional test remains low at about 30% which is due to several factors including education and cost.

Title: Accelerated partial breast irradiation using intensity modulated radiotherapy versus whole breast irradiation: 5-year survival
results of a phase 3 randomized trial
Lorenzo Livi1, Icro Meattini1, Livia Marrazzo2, Stefania Pallotta2, Gabriele Simontacchi1, Calogero Saieva3, Vieri Scotti1, Carla De
Luca Cardillo1, Paolo Bastiani4, Jacopo Nori5, Lorenzo Orzalesi6 and Simonetta Bianchi7. 1Radiotherapy-Oncology Unit, Florence
University Hospital, Florence, Italy; 2Medical Physics Unit, Florence University Hospital, Florence, Italy; 3Cancer Research and
Prevention Institute, Florence, Italy; 4S. Maria Annunziata Hospital, Florence, Italy; 5Florence University Hospital, Florence, Italy;
6
Breast Surgery Unit, Florence University Hospital, Florence, Italy and 7Florence University Hospital, Florence, Italy.
Body: Background. Accelerated partial breast irradiation (APBI) has been introduced as an alternative treatment method for
selected patients with early stage breast cancer (BC). Intensity modulated radiotherapy (IMRT) had theoretical advantage of
further increase in dose conformity compared to 3D technique, with more normal tissue sparing.
We present the results of a randomized equivalence trial comparing local recurrence and survival of APBI using IMRT technique
to conventional WBI in early stage BC.
Methods. This study was performed at the University of Florence (Florence, Italy). Women aged more than 40 years affected by
early breast cancer, with a maximum tumor diameter of 25 mm, suitable for breast-conserving surgery, were randomly assigned
in a 1:1 ratio to receive either external whole-breast irradiation (WBI) or APBI using IMRT technique.
Patients in the APBI arm received a total dose of 30 Gy in 5 fractions to the tumor bed. Those in the WBI arm received 50 Gy in
25 fractions, followed by a boost of 10 Gy in 5 fractions. For this equivalence trial the prespecified equivalence margin was local
recurrence of 5% in the APBI arm. The primary endpoint was occurrence of ipsilateral breast tumor recurrences (IBTR); overall
survival (OS) and treatment toxicity were secondary endpoints. Treatment tolerance was assessed using the acute and late
radiation morbidity scoring scheme from Radiation Therapy Oncology Group (RTOG) and the European Organization for
Research and Treatment of Cancer (EORTC). Cosmetic outcome was scored on the Harvard Breast Cosmesis scale.
This trial is registered with ClinicalTrials.gov, number NCT02104895.
Results. 520 patients were randomized (260 to external WBI and 260 to APBI with IMRT) between 2005 and 2013.
At a median follow-up of 5 years (range 0.6-9.0), we recorded 6 IBTR: 3 sited in the quadrant index (true recurrence) and 6 in
other quadrants. The mean time to IBRT was 2.9 years (range 1-4). The 5-year IBTR rate was 1.5% (3 cases) in the APBI group
(95% CI 0.1-3.0), within the prespecified equivalence threshold of 5%, in comparison to that of WBI group (1.4%; 95% CI 0-2.8).
No statistically significant difference emerged between two groups (log rank test p=0.86).
OS at 5 years did not significantly differ between the groups (p=0.057). We identified 8 deaths (5-year event rate: 2.1%; 95% CI
0.9-3.3), 7 in the WBI group and only 1 in the APBI group. The 5-year overall survival was 96.6% for the WBI group and 99.4% for
APBI group.
Concerning acute adverse events, APBI group showed a statistically significant better safety considering any grade of skin toxicity
(p=0.0001). No grade 3 toxicity was observed for APBI group. Concerning late side effects, only two cases (0.8%) experienced
grade 2 toxicity in WBI group (skin fibrosis). In both treatment groups the cosmetic result was rated as excellent/good for more
than 90% of patients. Overall, APBI group showed better outcome to WBI group (p=0.045).
Conclusions. To our knowledge this is the first randomized study using IMRT technique for APBI delivery. No statistical difference
in terms of IBTR was shown between the two arms. APBI showed a significant better acute and late toxicity profile compared to
WBI.

Title: A large prospectively-designed study of the DCIS score: Predicting recurrence risk after local excision for ductal carcinoma
in situ patients with and without irradiation
Eileen Rakovitch1,2,3, Sharon Nofech-Mozes1,3, Wedad Hanna1,3, Frederick L Baehner4,5, Refik Saskin2, Steven M Butler4, Alan
Tuck6, Sandip Sengupta7, Leela Elavathil8, Prashant A Jani9,10, Michel Bonin11, Martin C Chang12,3, Elzbieta Slodkowska1, Joseph
M Anderson4, Farid Jamshidian4, Diana B Cherbavaz4, Steven Shak4 and Lawerence Paszat1,2,3. 1Sunnybrook Health Sciences
Centre, Toronto, ON, Canada; 2Institute for Clinical and Evaluative Sciences, Toronto, ON, Canada; 3University of Toronto,
Toronto, ON, Canada; 4Genomic Health Inc, Redwood City, CA; 5University of California, San Francisco, CA; 6London Health
Sciences Centre, London, ON, Canada; 7Kingston General Hospital, Kingston, ON, Canada; 8Henderson General Hospital,
Hamilton, ON, Canada; 9Thunder Bay Regional Health Sciences Centre, Thunder Bay, ON, Canada; 10Northern Ontario School of
Medicine, Thunder Bay, ON, Canada; 11Health Sciences North Sudbury, Sudbury, ON, Canada and 12Mount Sinai Hospital,
Toronto, ON, Canada.
Body: Background: DCIS patients need better tools to align the aggressiveness of treatment with the aggressiveness of their
disease. The DCIS Score (DS) was validated as a predictor of ipsilateral breast recurrence (IBR; DCIS or invasive) in 327 E5194
patients treated by breast-conserving surgery (BCS) without radiation (RT) (Solin,2013). This Ontario population based DCIS
study of 3335 women with DCIS from 1994 to 2003 (Rakovitch,2013) was conducted to test the DCIS Score as a predictor of
recurrence risk in patients treated with BCS alone and in patients treated with BCS+RT.
Methods: REMARK guidelines were followed. Breast pathologists centrally reviewed all H&E slides. The Oncotype DX DCIS
Score was obtained by standardized quantitative RT-PCR using fixed paraffin embedded tumor. The pre-specified primary
objective was to determine the relationship (hazard ratio (HR)/50 units) between the risk of an IBR and the continuous DS (using
Cox models) in patients treated with BCS alone with ER+ tumors and clear margins (CM, no ink on tumor).
Results: Tumor blocks were collected for 1569 patients (47% of parent cohort); 718 received BCS without RT (N=571 with CM)
and 846 received BCS+RT (N=689 with CM). Median follow-up was 9.4 years. Among 1260 pts with CM, 100 pts treated with
BCS alone had an IBR (DCIS, N=44; invasive, N=57); 86 pts treated with BCS+RT had an IBR (DCIS, N=32; invasive, N=55). In
the primary analysis, among 571 patients treated by BCS alone with CM the continuous DS was significantly associated with IBR
in ER+ patients (HR 2.26; 95%CI 1.41,3.59; P=0.001) and in all patients (HR 2.15; 95%CI 1.43,3.22; P=<0.001). The DS was
also associated with invasive IBR (HR 1.78; 95%CI 1.03,3.05; P=0.04); similar but non-significant results were noted in the ER+
subgroup (P=0.08). Among 689 pts with CM treated by BCS+RT, the DS was associated with IBR (HR 2.78; 95%CI 1.77,4.41;
P=<0.001). There was no interaction between the DS and RT (P=0.40). In multivariable analysis for IBR in CM cases, the HR/50
units for the DCIS Score among patients treated with BCS alone was 1.80(95%CI 1.17,2.74; P=0.008) and 2.86(95%CI 1.79,4.62,
P=<0.001) for those treated with BCS+RT adjusting for multifocality, tumor size and age.
Conclusions: DCIS Score quantifies recurrence risk for DCIS patients treated by BCS with or without RT. Integrating the DCIS
Score with established risk factors, such as multifocality, age, and tumor size, can help identify DCIS patients treated with BCS
alone with low 10 year risk (<10%) of recurrence and those who still have high 10 year risk of recurrence despite RT who may be
candidates for more aggressive treatment.
BCS Alone, CM
DCIS Score Risk
Group
BCS with RT, CM
10-Yr Kaplan-Meier IBR Rate (95%CI)
10-Yr Kaplan-Meier IBR Rate (95%CI)
All patients (N=571)
Unifocal DCIS (N=457)
All patients (N=689)
Low (<39)
12.7% (9.5%,16.9%)
N=355
9.7% (6.8%,13.8%)
N=298
7.5% (4.9%,11.2%) N=332 9.8% (4.6%,20.1%) N=91
Int (39-54)
33.0% (23.6%,44.8%)
N=95
27.1% (17.7%,40.2%)
N=72
13.6% (8.6%,21.2%)
N=155
High (>55)
Multifocal DCIS (N=177)
20.6% (10.3%,38.7%)
N=37
Log rank P-value
27.8% (20.0%,37.8%)
N=121
27.0% (18.2%,38.9%)
N=87
20.5% (15.1%,27.5%)
N=202
33.3% (21.9%,48.5%)
N=49
<0.001
<0.001
<0.001
<0.001

Title: Defective stalled replication fork repair and predisposition to hereditary breast cancer
Shailja Pathania1,2, Sangeeta Bade1, Morwenna Le Guillou3, Karly Burke2, Ying Su1, David T Ting4, Kornelia Polyak1,2, Andrea L
Richardson1,5, Jean Feunteun3, Judy E Garber1,2 and David M Livingston1,2. 1Dana-Farber Cancer Institute, Boston, MA; 2Harvard
Medical School, Boston, MA; 3Laboratoire Stabilite' Genetique et Oncogenese, Universit Paris-Sud, Gustave-Roussy, Villejuif,
France; 4Massachusetts General Hospital, Charlestown, MA and 5Brigham and Women's Hospital, Boston, MA.
Body: BRCA1 is a tumor suppressor gene, and germ line BRCA1 mutations increase the risk of breast cancer. While all cells with
BRCA1 mutations exhibit a heterozygous BRCA1mut/+ genotype, cancer develops primarily in females, often at young ages and
affects almost exclusively the breast and ovaries. Why BRCA1 shows such tissue specificity, and how a normal cell in a BRCA1
mutation carrier (BRCA1mut/+) gives rise to invasive tumor cells are largely unknown.
To determine whether BRCA1 heterozygosity in cells confers defect in any of the multiple, known, BRCA1 functions is a
potentially valuable step in achieving a better understanding of BRCA1 mutation-driven cancer predisposition. Thus, we have
analyzed a collection of primary mammary BRCA1mut/+ epithelial cells and skin fibroblasts obtained from BRCA1 mutation
carriers for such functions.
We, and others have recently shown that BRCA1 exhibits a new DNA damage repair function i.e. repair of stalled replication
forks (SFR). Stalled forks, when not resolved, lead to mutations, or collapse into double strand breaks (DSBs). Both outcomes
result in what is commonly referred to as replication stress (RS), which, when chronic, is a driving force behind cancer
development. To determine if SFR is defective in normal/healthy breast cells in BRCA1 mutation carriers, and whether this
haploinsufficiency results in the kind of genomic changes that lead to cancer, we have now generated 18 primary fibroblast
strains from skin punch biopsies and 10 primary mammary epithelial cell (MECs) strains from prophylactic mastectomies
performed on BRCA1 mutation carrying women. This collection includes N=23 different BRCA1 mutations, which, together, span
almost the entire BRCA1 gene. BRCA1+/+ control MECs were derived from tissue collected during reduction mammoplasties and
control fibroblasts were derived from skin punch biopsies from women with no BRCA1 mutation.
Our current data shows that BRCA1mut/+ strains exhibited multiple, normal BRCA1 functions, including the support of
homologous recombination- type double strand break repair (HR-DSBR), cell cycle- associated checkpoint functions, centrosome
number control, spindle pole formation, Slug expression and satellite RNA suppression. By contrast, nearly all strains were
defective in the repair of stalled replication forks and in the suppression of fork collapse, i.e. replication stress. These defects
were rescued by reconstituting BRCA1 heterozygous cells with wild-type BRCA1 cDNA, indicating that they are a product of
BRCA1 haploinsufficiency.
In addition, the development of sufficient replication stalling rendered BRCA1mut/+ cells defective in an otherwise intact BRCA1
function, HR-DSBR. No such conditional haploinsufficiency was detected in any of the other non-haploinsufficient functions,
noted above. Given the importance of replication stress in cancer development and of an HR defect in breast cancer
pathogenesis, these defects, when they develop serially, could contribute to the BRCA1 breast cancer development pathway.
Finally, given the important role of BRCA2, another hereditary breast cancer gene, in stalled fork stability, a similar analysis for
BRCA2mut/+ cells from BRCA2 mutation carriers is currently underway and will also be reported at the meeting.

Title: Large-scale identification of cell-specific PARP substrates
Yonghao Yu1 and Yajie Zhang1. 1UT Southwestern Medical Center, Dallas, TX.
Body: PARPs (poly-ADP-ribose polymerases) catalyze a protein posttranslational modification termed Poly-ADP-ribosylation
(PARylation). PARylation is composed of linear and/or branched repeats of ADP-ribose, whose lengths can reach up to 200 units.
The first gene encoding a poly-ADP-ribose polymerase, PARP1, was cloned in 1987. Numerous efforts have now led to the
identification of 16 additional PARP enzymes.
PARP1 is a nuclear protein that is activated as a result of sensing DNA strand breaks. PARylation levels in a quiescent cell are
usually very low. In response to genotoxic stress, PARP1 is recruited to nicked DNA and is rapidly activated. This triggers the
synthesis of a large number of PARylated proteins and the initiation of DNA damage repair mechanisms. Cancer cells with
defects in double-strand break (DSB) repair, such as BRCA1/2-mutated cells, are reliant on PARP1 activity for genome integrity.
These cells undergo unsustainable genetic damage, and eventually, apoptosis upon PARP1 inhibition. Indeed, recent late-stage
clinical studies revealed that PARP1 inhibitor treatment significantly prolonged progression-free survival of BRCA-deficient breast
cancer patients.
Contrary to the fruitful efforts in characterizing the upstream inputs regulating PARP1, its genuine downstream targets, however,
remain poorly defined. We recently developed the first mass spectrometry-based approach to a global mapping of the human
aspartic acid- and glutamic acid-ADP-ribosylated proteome (Zhang et al., Nature Methods 2013). This method allowed us to
identify 1,048 in vivo PARylation sites on 340 proteins. These modified proteins are involved in a wide array of nuclear functions
including DNA damage repair, transcription regulation, epigenetic modulation, and mRNA processing. Using a quantitative mass
spectrometry experiment, we also identified hundreds of novel PARP1 downstream effectors.
The central hypothesis of our current study is that ADP-ribosylation levels of PARP1 substrates can reflect the activity of PARP1
in a cell. We predict that a signature composed of multiple PARylated proteins can then be used to identify cells that are
"addicted" to PARP1 activity for genome integrity. To this end, we performed a large scale profiling of the Asp- and
Glu-ADP-ribosylated proteome of a benign breast epithelial cell line (MCF10A), and compared it to that of a panel of eight breast
carcinoma cell lines, including the ER+ (MCF7, T47D and ZR-75-1), HER2+ (SKBR3), and triple negative (MDA-MB-231,
MDA-MB-468, SUM159 and HCC1937) subtypes. Among these lines, MDA-MB-468 (PTEN null), SKBR3 (HER2+) and HCC1937
(BRCA1 null) are known to be sensitive to PARP inhibitors. We correlated the pattern of protein ADP-ribosylation to their IC50,
and found events of predictive value. In addition, we also observed that the differential PARylation pattern among these cells can
be divided into a "public" (proteins that are commonly modified across all cell lines) class, and a "private" one (proteins that
specifically modified in certain cell lines). We envision that this dataset will also serve as a valuable resource for the PARP1/DNA
damage research community to investigate cell line-specific Asp- and Glu-ADP-ribosylation events.

Title: A randomized, open-label, multicenter, phase 3 study of epoetin alfa (EPO) plus standard supportive care versus standard
supportive care in anemic patients with metastatic breast cancer (MBC) receiving standard chemotherapy
Brian Leyland-Jones1, Igor Bondarenko2, Gia Nemsadze3, Vitaliy Smirnov4, Iryna Litvin5, Irakli Kokhreidze6, Lia Abshilava7, Mikheil
Janjalia8, Rubi Li9, KC Lakshmaiah10, Beka Samkharadze11, Oksana Tarasova12, Ranjan Kumar Mohapatra13, Yaroslav Sparyk14,
Sergey Polenkov15, Vladimir Vladimirov16, Liang Xiu17, Bruce Kimelblatt17, Kris Deprince18, Ilya Safonov19, Els Vercammen20 and
Peter Bowers17. 1Avera Cancer Institute, Sioux Falls, SD; 2Dnepropetrovsk Medical Academy, Dnepropetrovsk, Ukraine; 3Institute
of Clinical Oncology (Ltd K. Madichi Mammological Center), Tbilisi, Georgia; 4Donetsk Regional Anticancer Center, Donetsk,
Ukraine; 5Dnepropetrovsk Regional Oncological Dispensary, Dnepropetrovsk, Ukraine; 6Martin D. Abeloff Laboratory Cancer
Research Center, Tbilisi, Georgia; 7Chemotherapy and Immunotherapy Clinic MEDULLA, Tbilisi, Georgia; 8Tbilisi Cancer Center,
Tbilisi, Georgia; 9St Luke's Medical Center, Quezon City, Philippines; 10Kidwai Memorial Institute of Oncology, Bangalore, India;
11
Research Institute of Clinical Medicine, Tbilisi, Georgia; 12Institute of Medical Radiology, Kharkiv, Ukraine; 13Apollo Specialty
Hospital, Chennai, India; 14Lviv State Oncology Regional Treatment and Diagnostic Centre, Lviv, Ukraine; 15Chernigov Regional
Oncology Center, Chernigov, Ukraine; 16Pyatigorsk Oncology Dispensary, Pyatigorsk, Russian Federation; 17Janssen R&D,
Raritan, NJ; 18Janssen R&D, Beerse, Belgium; 19Janssen R&D, Moscow, Russian Federation and 20Janssen R&D, High
Wycombe, United Kingdom.
Body: Background: Several investigational studies including MBC reported that erythropoiesis-stimulating agent (ESA)
treatment beyond correction of anemia decreased survival and locoregional tumor control, and/or increased adverse effects,
especially thrombotic vascular events (TVEs). These studies were reviewed at 3 FDA Oncologic Drugs Advisory Committee
Meetings (2004, 2007, and 2008). Other studies and well-conducted meta-analyses did not suggest adverse effects on tumor
outcomes when ESAs are used according to the prescribing information in subjects receiving chemotherapy, however, no study
rigorously evaluated tumor outcomes.
This Phase 3 study is the largest MBC trial in subjects receiving EPO for chemotherapy induced anemia specifically designed and
conducted to assess progression-free survival (PFS) as the primary endpoint. Conducted as a post-marketing requirement, the
study has been regularly and intensively monitored by an Independent Data Monitoring Committee (IDMC) of internationally
recognized experts in the fields of MBC treatment, clinical research, and statistics.
Methods: EPO-ANE-3010 (ClinicalTrials.gov #NCT00338286) is a multinational (19 countries and 223 participating sites) study
with 2,098 subjects who are anemic and receiving first- or second-line standard chemotherapy for MBC. Key inclusion criteria:
histologically confirmed MBC, stage IV disease and at least 1 measurable metastatic lesion according to Response Evaluation
Criteria in Solid Tumors (RECIST); hemoglobin (Hb) 11 g/dL, Eastern Cooperative Oncology Group performance score 0 or 1.
Key exclusion criteria: metastasis to bone only, receiving anticoagulants or endocrine therapy. Subjects were randomly assigned
(1:1) to receive either standard supportive care for treatment of anemia (SOC) plus 40,000 IU EPO given subcutaneously weekly
up to 4 weeks after the last dose of cytotoxic chemotherapy, or to SOC alone. Randomization was stratified by line of
chemotherapy and HER2/NEU status. EPO dose was held for Hb >12 g/dL. Disease was assessed every 8 weeks for the first
year, and then every 12 weeks until disease progression (PD) or death. Tumor response was determined according to modified
RECIST 1.0 criteria by Investigators and Blinded Central Review. Overall survival (OS) follow up continued after PD.
The primary endpoint is PFS. Secondary endpoints include OS, time to tumor progression, and overall response rate and safety
assessments (including incidence and severity of TVEs). For this non-inferiority study, a sample size of 1,650 disease
progression or death events was determined to provide over 80% power to rule out a 15% hazard rate increase (i.e., hazard ratio
of 1.15, epoetin alfa vs. control) in PFS with a 1-sided Type I error rate 0.025. The study is fully accrued with 2014 projected key
dates: clinical cut-off at 1650 PFS events, July; database lock, September; IDMC agreement on presentation of final results,
mid-November. Both primary and secondary endpoints will be fully reported.

Title: Final survival analysis from the randomized Women's Intervention Nutrition Study (WINS) evaluating dietary intervention as
adjuvant breast cancer therapy
Rowan T Chlebowski1 and George L Blackburn2. 1Los Angeles BioMedical Research Institute at the Harbor-UCLA Medical
Center, Torrance, CA; 2Beth Isreal Deaconess Hospital, Boston, MA and 3on behaf of the Women's Intervention Nutrition Study
Investigators, Torrance, CA.
Body: Between 1994 and 2001, 2,437 women between 48-79 years of age with early-stage breast cancer (73% node negative,
79% estrogen receptor positive, 70% = or < 2 cm) receiving standard cancer management (endocrine therapy +/- chemotherapy if
hormone receptor positive or chemotherapy if hormone receptor negative plus radiation therapy if clinically indicated) were
randomized, within 6 months of diagnosis, to a dietary intervention or control group from 39 US centers participating in the
Womens Intervention Nutrition Study (WINS). The dietary intervention, targeting fat intake reduction while maintaining nutritional
adequacy, was initiated during 8 biweekly individual counseling sessions by centrally trained registered dieticians implementing a
previously developed low-fat eating plan with subsequent every 3 month dietician contacts. During 5 years (median) intervention,
fat intake was significantly reduced (from 29.2% to 20.3% of calories, P< 0.0001) as was body weight (-2.7 kg, P=0.005) in the
dietary intervention group but not in the control group. Relapse-free survival, the primary study endpoint, was favorably impacted
in the dietary intervention compared to the control group (9.8% vs 12.4% with events, respectively, HR 0.78 95% CI 0.60-0.98, P
= 0.03 from adjusted Cox model) (J Natl Cancer Inst 2006:98;1767). Exploratory analyses suggested a greater dietary effect on
women with hormone receptor negative cancers.When intervention ended after 5 years median follow-up with a total of 171
deaths, there were somewhat fewer deaths in the intervention group (6.6% vs 7.3%, HR 0.89 95% CI 0.65-1.21). In the last WINS
update, based on national death registry information after 8.1 years median follow-up with a total of 251deaths, while there were
fewer deaths in the intervention compared to the control group (9.1% vs 11.1%), the difference was again not statistically
significant (HR 0.78 95% CI 0.59-1.03) (J Clin Oncol 2008:26;522). With this as background, the primary study objective is to
determine, with updated survival information after now 15 years median follow-up, whether a lifestyle intervention targeting fat
intake reduction associated with significant weight loss will improve overall survival in early stage breast cancer patients receiving
standard breast cancer management. Information on survival is being obtained regarding the status of the 2,081 WINS study
participants last known to be alive using the National Death Registry identified through DOBsearch.com. It is anticipated that
there will now be a total of approximately 430 deaths in the trial. The new information will be incorporated in updated survival
analyses, using time-to-event methods based on intention-to-treat principles, for the overall population and for subgroups of
interest. Results will be presented as a late breaking abstract.

Title: Phase 3, randomized, double-blind, placebo-controlled multicenter trial of daily everolimus plus weekly trastuzumab and
paclitaxel as first-line therapy in women with HER2+ advanced breast cancer: BOLERO-1
Sara A Hurvitz1, Fabrice Andre2, Zefei Jiang3, Zhimin Shao4, Silvia P Neciosup5, Max S Mano6, Ling-Min Tseng7, Qingyuan
Zhang8, Kunwei Shen9, Donggeng Liu10, Lydia M Dreosti11, Jifeng Feng12, Howard A Burris13, Masakazu Toi14, Marc E Buyse15,
David Cabaribere16, Mary-Ann Lindsay17, Tiffany Kunz18, Shantha Rao18, Lida B Pacaud18, Tetiana Taran18 and Dennis Slamon1.
1
University of California, Los Angeles, CA; 2Institut Gustave Roussy, Villejuif, France; 3Beijing 307 Hospital of PLA, Beijing, China;
4
Cancer Hospital of Fudan University, Shanghai, China; 5Instituto Nacional de Enfermedades Neoplasicas, Surquillo, Lima, Peru;
6
Instituto do Cncer do Estado de So Paulo, So Paulo, Brazil; 7Taipei Veterans General Hospital, Taipei, Taiwan; 8Tumor
Hospital of Harbin Medical University, Harbin, China; 9Ruijin Hospital, Shanghai Jiaotong University School of Medicine,
Shanghai, China; 10Sun Yat-sen University Cancer Center, Guangzhou, China; 11University of Pretoria, Gauteng, South Africa;
12
Jiangsu Provincial Tumor Hospital, Nanjing, Jiangsu, China; 13Sarah Cannon Research Institute, Nashville, TN; 14Graduate
School of Medicine, Kyoto University, Kyoto, Japan; 15International Drug Development Institute, Louvain La Neuve, Belgium;
16
Translational Research in Oncology (TRIO), Paris, France; 17Translational Research in Oncology (TRIO), Edmonton, Canada
and 18Novartis Pharmaceuticals Corporation, East Hanover, NJ.
Body: Background: Mammalian target of rapamycin (mTOR) is central to multiple signaling pathways regulating cell growth and
proliferation. In early studies, an mTOR inhibitor everolimus (EVE) showed antitumor activity in breast cancer and synergy with
both trastuzumab (TRAS) and paclitaxel (PAC). The BOLERO-1 study evaluated the addition of EVE to TRAS+PAC as first-line
therapy for HER2+ advanced breast cancer (ABC).
Methods: In this phase 3 randomized trial, women with HER2+ ABC without prior TRAS or chemotherapy for advanced disease
were randomized 2:1 to receive either EVE (10 mg/d) or placebo (PBO) and weekly PAC plus TRAS. The two primary objectives
were to compare the investigator assessed progression-free survival (PFS) between EVE+TRAS+PAC and PBO+TRAS+PAC in
the full population and the Hormone Receptor negative (HR) subpopulation. Secondary endpoints included overall survival
(OS), response rate, and safety.
Results: 719 patients were randomized to receive EVE (n=480) or PBO (n=239). Final PFS analysis was performed after 425
events in the full population. Baseline characteristics/prior therapies were balanced between two treatment arms; median age was
53 years, 70.5% had visceral metastases, 43.3% were HR; prior therapy included TRAS (10.8%) and taxane (24.9%); baseline
characteristics for the HR subpopulation were generally balanced between two treatment arms and similar to the overall
population. Median study follow-up at the time of analysis was 41.3 mo. The study did not meet its primary objective in the full
population; median PFS was 15 mo (95%CI: 14.6-17.9) in the EVE arm vs. 14.5 mo (95%CI: 12.3-17.1) in the PBO arm
(HR=0.89 [95%CI: 0.73-1.08]; p=0.1166). In the HR subpopulation (n=311), a clinically relevant 7.2 mo benefit in median PFS
was observed in the EVE arm (20.3 mo [95%CI: 15.0-24.1]) vs. PBO arm (13.1 mo [95%CI: 10.1-16.6]; HR=0.66 [95%CI:
0.48-0.91]; p=0.0049), which just fell short of crossing the protocol pre-specified level of statistical significance (p=0.0044).
Additional sensitivity analysis of PFS without censoring patients at the start of new antineoplastic therapy yielded hazard ratio
consistent with the primary analysis (p=0.0043). PFS based on central assessment corroborated investigator assessed PFS in
both the full population and HR subpopulation. OS data is immature. Safety profile of EVE was consistent with previous
observations in ABC; no new signals were identified. Most common adverse events (AEs) in the EVE arm vs. PBO arm were
stomatitis (66.5% vs. 32.4%), diarrhea (56.6% vs. 46.6%), and alopecia (46.8% vs. 52.5%); suspected drug-related serious AEs
were reported for 21.8% vs. 7.6%, and on-treatment AE related deaths were reported for 3.6% vs. 0% of patients, respectively.
Conclusions: First-line therapy with EVE+TRAS+PAC did not show PFS benefit in patients with HER2+ ABC; the HR
subpopulation derived a clinically robust benefit of 7.2 mo in median PFS suggesting that EVE may have a role in this patient
subpopulation. EVE+TRAS+PAC toxicity was generally manageable; no new safety signals were identified. (Funded by Novartis;
ClinicalTrials.gov number, NCT00876395.)

Title: TBCRC023: A randomized multicenter phase II neoadjuvant trial of lapatinib plus trastuzumab, with endcorine therapy and
without chemotherapy, for 12 vs. 24 weeks in patients with HER2 overexpressing breast cancer
Mothaffar F Rimawi1, Polly A Niravath1, Tao Wang1, Brent Rexer2, Andres Forero3, Antonio C Wolff4, Rita Nanda5, Anna M
Storniolo6, Ian Krop7, Matthew P Goetz8, Julie R Nangia1, Sao Jiralerspong1, Anne C Pavlick1, Carolina Gutierrez1, Rachel Schiff1,
Susan G Hilsenbeck1 and C Kent Osborne1. 1Baylor College of Medicine, Houston, TX; 2Vanderbilt University, Nashville, TN;
3
University of Alabama, Birmingham, AL; 4Johns Hopkins University, Baltimore, MD; 5University of Chicago, Chicago, IL; 6Indiana
University, Indianapolis, IN; 7Dana-Farber Cancer Institute, Boston, MA and 8Mayo Clinic, Rochester, MN.
Body: Background: We have previously shown in animal models that combination anti-HER2 therapy leads to complete tumor
regression of HER2+ breast cancer (BC). We translated these findings in a neoadjuvant trial of 12 weeks (wks) of L+T
(TBCRC006) that demonstrated a meaningful pathologic complete response (pCR) and near pCR (in ER+ tumors) in patients with
locally advanced HER2+ BC. In the present trial, we sought to determine whether longer treatment would lead to a higher rate of
pCR without the use of chemotherapy by converting near pCR to pCR especially in the ER+ subset.
Methods: TBCRC023 (NCT00999804) is a randomized phase II trial combining a Simon Phase 2 design in the experimental arm
with a pick-the-winner design, not powered for direct comparison. Women with HER2+ BC measuring 2 cm or larger were eligible
and were randomized in a 1:2 ratio to 12 vs. 24 wks of L+T. Letrozole (along with ovarian suppression if premenopausal) was
also administered in patients whose tumors were also ER+. Serial tumor biopsies were obtained at baseline, wk 1, wk 12, and at
the time of surgery. All evaluable patients were assessed for pCR, defined as no residual invasive carcinoma in the breast.
Patients did not undergo surgery, withdrew consent, or received additional neoadjuvant therapy were counted as non-responders.
Results: Ninety-seven patients were enrolled (33 in 12-wk arm and 64 in 24-wk arm), of whom 95 were evaluable. Seventy seven
percent of patients were white and 18% were black. Twenty percent were of Hispanic ethnicity. Median age was 51 and 55%
were postmenopausal. Median tumor size was 5 cm and 65% were ER+. Study treatment was well tolerated with grade 1-2
diarrhea (24% in 12-wk arm, 31% in 24-wk arm) and grade 1-2 acneform rash (12% in 12-wk arm, 19% in 24-wk arm) being the
most common toxicities. Grade 3 toxicities were uncommon and were mostly in the 24-wk arm (elevated liver function test: 9%,
diarrhea 2%, mucositis 2%), while the 12-wk arm had one grade 3 anemia and the studys only serious adverse event (acute
kidney injury). There were no grade 4 toxicities.
The experimental arm completed stage 2 accrual and pCR rate was numerically superior to control, entirely due to better results
in ER+ (Table 1), but lower than the expected pCR rate of 36% needed in the planned analysis to conclude in favor of enhanced
efficacy with extended therapy. pCR rates were also lower in the control arm than in the previous study.
Table 1. pCR rates
12-week arm % (n)
24-week arm % (n)
ER-positive
8.7% (2/23)
33.2% (13/39)
ER-negative
20% (2/10)
8.7% (2/23)
Overall pCR
12.2% (4/33)
24.2% (15/62)
Conclusions: Treatment with L+T (with endocrine therapy in ER+ tumors) for 24 weeks leads to doubling of the pCR rate in
women with HER2+ breast cancer without using cytotoxic chemotherapy. This approach is effective and well tolerated and
warrants study as part of a de-escalation strategy that may spare some patients the cost and toxicity of chemotherapy. Tissue
obtained on this trial will provide a valuable resource to validate correlative findings from our prior studies and discover new
biomarkers to help guide proper patient selection for treatment.

Title: Randomized phase II trial of fulvestrant alone or in combination with bortezomib in hormone receptor-positive metastatic
breast cancer resistant to aromatase inhibitors: A New York cancer consortium trial
Kerin B Adelson1, Bhuvaneswari Ramaswamy2, Joseph A Sparano3, Paul J Christos4, John J Wright5, George Raptis6, Miguel C
Villalona6, Cynthia X Ma7, Dawn Hershman8, Joseph Baar9, Paula Klein10, Tessa Cigler3, G Thomas Budd11, Yelena Novik12,
Antoinette R Tan13, Susan Tannenbaum14, Anupama Goel15, Ellis Levine16, Charles L Shapiro2, Eleni Andreopoulou4, Michael
Naughton7, Kevin Kalinsky8, Samuel Waxman17 and Doris Germain17. 1Yale Cancer Institute, Yale University School of Medicine,
New Haven, CT; 2Ohio State University, Columbus, OH; 3Montefiore Medical Center, Bronx, NY; 4Weill Cornell Medical Center,
New York, NY; 5Cancer Therapy Evaluation Program National Cancer Institute, Bethesda, MD; 6North Shore LIJ Cancer
Institute, Lake Success, NY; 7Washington University School of Medicine, St Louis, MO; 8Columbia University Medical Center,
New York, NY; 9University Hospitals of Cleveland, Cleveland, OH; 10Mount Sinai Medical Center Beth Israel, New York, NY;
11
Cleveland Clinic Taussig Cancer Institute, Cleveland, OH; 12New York University School of Medicine, New York, NY; 13Rutgers
Cancer Institute of New Jersey, New Brunswick, NJ; 14University of Connecticut Health Center, Farmington, CT; 15Mount Sinai
Medical Center Roosevelt, New York, NY; 16Roswell Park Cancer Institute, Buffalo, NY and 17Tisch Cancer Institute, Icahn
School of Medicine at Mount Sinai, New York, NY.
Body: Purpose: Fulvestrant (F) is a selective estrogen receptor downregulator (SERD) with activity in aromatase-inhibitor (AI)
resistant estrogen receptor (ER)-positive metastatic breast cancer (MBC). In preclinical studies, the proteasome inhibitor
bortezomib (B) enhances the antineoplastic effects of F, in part by promoting accumulation of large ER-aggregates that lead to
cell death (Ishii et al. Clin Cancer Res 2011 17:2292). The objective of this study was to determine if the combination of F+B was
more efficacious than F alone in MBC after AI progression.
Patients and Methods: Postmenopausal women with ER-positive MBC who had progressive disease after prior AI therapy were
eligible. They were randomized to F alone (500 mg IM days -15, 1, 15 in cycle 1, and day 1 of each subsequent cycle) or in
combination with B (1.6 mg/m2 IV on days 1, 8, 15). The primary endpoint was progression free survival (PFS), measured from
cycle 1, day 1 of starting F. A sample size of 118 was pre-specified in order to provide sufficient power to detect an improvement
in median PFS from 5.4 to 9.0 months, and compare PFS rates after 6 and 12 months (1-sided alpha=0.10, beta=0.10). Patients
with progression on F could cross over to the F+B combination.
Results: Of 118 patients enrolled, 59 received F alone (arm A), 57 received F+B (arm B), and 2 assigned to arm B never initiated
protocol therapy. There were no significant differences in patient characteristics between arms with regard to median age (57 vs.
59 years), ECOG performance status (0 and 1, 64% and 36%, respectively), prior chemotherapy for metastasis (25%), or liver
metastases (37%), although patients in arm A had longer median interval between diagnosis and metastasis (49 vs. 28 months)
and were more likely to present with metastasis (32% vs. 26%). Patients in arm B had more adverse events (all grades), including
nausea (63% vs. 29%), diarrhea (47% vs. 8%), sensory neuropathy (46% vs. 29%), and limb edema (37% vs. 19%), although
grade 3-4 events were uncommon, and only 11% discontinued B due to toxicity. At 12 months, the PFS proportion in Arm A and
Arm B was 13.6% vs. 28.1%, respectively (P=0.03, 1-sided chi-square test) (95% CI for difference [14.5%] = -0.06%, 29.1%).
Although median PFS was similar in the two arms (2.69 vs. 2.73 months, respectively), the hazard ratio for Arm B vs. Arm A
(referent) was 0.73 (95% CI = 0.49, 1.09, P=0.06, 1-sided log rank test). Both results were significant at the pre-specified 1-sided
0.10 alpha level. Of 27 patients on arm A who crossed over to F+B at progression, 4 (15%) were progression-free for at least 24
weeks and had periods of disease control that were longer than when treated with F alone.
Conclusion: Adding bortezomib to fulvestrant in AI-resistant ER-positive MBC enhances its effectiveness by delaying acquired
fulvestrant resistance. These results support additional evaluation of proteasome inhibitors in combination with SERDs.
Acknowledgement: Supported by contract N01-CM-62204 to the New York Cancer Consortium (P.I. J. Sparano) and grant P30
CA013330 (P.I. D. Goldman) from the National Institutes of Health, and by a grant from Millennium, Inc.

Title: Fulvestrant 500 mg versus anastrozole as first-line treatment for advanced breast cancer: overall survival from the phase II
first study
John FR Robertson1, Antonio Llombart-Cussac2, David Feltl3, John Dewar4, Marek Jasiwka5, Nicola Hewson6, Yuri Rukazenkov6
and Matthew J Ellis7. 1University of Nottingham, Derby, Nottingham, United Kingdom; 2Hospital Arnau de Vilanova, Lrida, Spain;
3
FNsP Ostrava, Radioterapeutick Klinika, Ostrava-Poruba, Czech Republic; 4Ninewells Hospital and Medical School, Dundee,
United Kingdom; 5Centrum Onkologii, Instytut im Marii Sklodowskiej-Curie, Krakw, Poland; 6AstraZeneca Pharmaceuticals,
Macclesfield, United Kingdom and 7Washington University School of Medicine, St Louis, MO.
Body: Background: Fulvestrant 500 mg showed a clinically significant improvement in median overall survival (OS) vs
fulvestrant 250 mg (26.4 vs 22.3 months, respectively; hazard ratio [HR] 0.81; 95% confidence interval (CI) 0.69, 0.96; nominal
p=0.02) in the Phase III CONFIRM study, for patients (pts) with hormone receptor positive (HR+) disease following failure on prior
endocrine therapy. Further evidence for OS effects of fulvestrant 500 mg was sought in the Fulvestrant fIRst-line Study comparing
endocrine Treatments (FIRST), which compared fulvestrant 500 mg with anastrozole as first-line treatment for postmenopausal
pts with HR+ locally advanced (LA) or metastatic breast cancer (MBC). In the primary analysis, fulvestrant 500 mg was as
effective as anastrozole for clinical benefit rate (primary endpoint) and significantly better for time to progression (TTP; secondary
endpoint). In a follow-up analysis, median TTP was 23.4 months for fulvestrant 500 mg vs 13.1 months for anastrozole (HR 0.66;
95% CI 0.47, 0.92; p=0.01). Here we report the only scheduled FIRST OS analysis.
Methods: FIRST, a Phase II, randomized, open-label study (NCT00274469), compared fulvestrant 500 mg (im on Days 0, 14 and
28, and every 28 days thereafter) with anastrozole (1 mg/day po). Pts had not received prior endocrine therapy for advanced
disease. OS (time from randomization to death) was compared by unadjusted log-rank test after approximately 65% of deaths.
Effect of treatment on OS was examined across subgroups (including age, hormone receptor status and visceral disease). Pts
alive or not known to have died were right-censored at last known date alive, including 20 pts in centers invited but who did not
join the OS follow-up phase. Serious adverse events (SAEs) were recorded.
Results: 205 pts (median age 67.0 years) were randomized from 62 centers in 9 countries (fulvestrant 500 mg: n=102;
anastrozole: n=103). The first pt enrolled on Feb 6, 2006. As of July 2014, 33/205 pts (16.1%) were known to be alive across both
treatment groups and 137/205 (66.8%) pts had died. Median OS was significantly greater for fulvestrant 500 mg (54.1 months) vs
anastrozole (48.4 months; HR 0.70; 95% CI 0.50, 0.98; p=0.041). OS analyses in pre-specified subgroups demonstrated a
consistent treatment effect for fulvestrant 500 mg vs anastrozole (global interaction test p=0.755). The frequency of SAEs was
similar between fulvestrant 500 mg (23.8%) and anastrozole (21.4%).
Conclusions: HR+ pts receiving first-line fulvestrant 500 mg lived significantly longer than pts on anastrozole (median OS
difference of 5.7 months). A consistent OS treatment effect was observed across predefined subgroups. FIRST is therefore the
second randomized trial to show an OS advantage for fulvestrant 500 mg over another endocrine therapy. No new safety signals
were identified with longer-term treatment. Improved OS data provide further support for superior efficacy of fulvestrant 500 mg
over anastrozole as first-line endocrine therapy for postmenopausal women with HR+ LA or MBC. If confirmation of superiority for
fulvestrant 500 mg is seen in the Phase III FALCON study (NCT01602380), fulvestrant 500 mg should be considered for approval
as a first-line agent in this setting.

Title: Characterization of male breast cancer: First results of the EORTC10085/TBCRC/BIG/NABCG International Male BC
Program
Fatima Cardoso1, John Bartlett2,3, Leen Slaets4, Carolien van Deurzen5,6, Elise van Leewen-Stok6, Peggy Porter7,8, Barbro
Linderholm9, Ingrid Hedenfalk10, Carolien Schroder6,11, John Martens6,12, Jane Bayani2, Christi van Asperen6,13, Melissa Murray14,
Clifford Hudis15,16, Lavinia Middleton17, Joanna Vermeij18, Stephanie Peeters19, Judith Fraser20, Monica Nowaczyk21, Isabel Rubio22,
Stefan Aebi23, Catherine Kelly24, Kathryn Ruddy25, Eric Winer26, Cecilia Nilsson27, Lissandra Dal Lago28, Larissa Korde29, Kim
Benstead30, Danielle Van Den Weyngaert31, Oliver Bogler32, Theodora Goulioti33, Nicolas Dif34, Carlo Messina35, Konstantinos
Tryfonidis35, Jan Bogaerts4 and Sharon Giordano36. 1Champalimaud Cancer Center, Lisbon, Portugal; 2Transformative Pathology,
Ontario Institute for Cancer Research, Toronto, Canada; 3University of Edinburgh, Scotland, United Kingdom; 4European
Organization for Research and Treatment of Cancer, Brussels, Belgium; 5Erasmus Medical Center, Rotterdam, Netherlands;
6
Dutch Breast Cancer Research Group (BOOG), Netherlands; 7Fred Hutchinson Cancer Research Center; 8University of
Washington, Seattle; 9Sahlgrenska University Hospital, Gothenburg, Sweden; 10Lund University, Lund, Sweden; 11University
Medical Center, Groningen, Netherlands; 12Breast Cancer Genomics and Proteomics Lab, Erasmus Medical Center, Rotterdam,
Netherlands; 13Leiden University Medical Center, Leiden, Netherlands; 14Memorial Sloan Kettering Cancer Center, New York;
15
Breast Medicine Service, Memorial Sloan Kettering Cancer Center, New York; 16Weill Cornell Medical College, New York, NY;
17
University of Texas, MD Anderson Cancer Center, Houston; 18ZNA Jan Palfijn, Belgium; 19UZ Leuven, Leuven, Belgium;
20
Beatson West of Scotland Centre, Glasgow Scotland, United Kingdom; 21Specialist Hospital St Wojciech, Gdansk, Poland;
22
Breast Surgical Unit, Hospital Universitario, Vall d'Hebron, Barcelona, Spain; 23Swiss Group for Clinical Cancer Research
(SAAK), Switzerland; 24All Ireland Cooperative Oncology Research Group (ICORG), Ireland; 25Mayo Clinic, Rochester;
26
Dana-Farber Cancer Institute, Boston, MA; 27Vastmanlands Hospital, Sweden; 28Jules Bordet Institute, Brussels, Belgium;
29
University of Washington, Seattle, WA; 30Cheltenham General Hospital, United Kingdom; 31ZNA, Middelheim, Belgium; 32Global
Academic Programs, University of Texas, MD Anderson Cancer Center, Houston, TX; 33Breast International Group,
Headquarters, Brussels, Belgium; 34European Organization for Research and Treatment of Cancer (EORTC), Brussels, Belgium;
35
European Organization for Research and Treatment of Cancer (EORTC), Brussels, Belgium and 36MD Anderson Cancer Center,
Houston, TX.
Body: Intro: Male BC is a rare disease (<1% male tumors); knowledge is limited and management extrapolated from female BC.
An international consortium, coordinated by EORTC and TBCRC, was created to better characterize and manage this disease,
with 3 parts: 1) retrospective joint analysis 2) prospective registry 3) clinical trial(s). We report 1st results of part 1. Methods: Joint
analysis of male BC pts with available FU & FFPE samples, treated in 1990-2010, in 23 centers from 9 countries. Clinical data,
long term outcomes, local pathology were centrally analyzed at EORTC. FFPE samples were analyzed at 3 central labs (UK, NL,
US) for histology, grade, ER, PR, AR, HER-2, Ki67. Cut-off for positivity: Allred score 3 for ER/ AR/PR; 20% for Ki67;
ASCO-CAP for HER2. Outcome data (OS & RFS) analyzed per Kaplan Meier, p-values correspond to logrank test, CIs by
Brookmeyer & Crowley. Reported % use number of non-missing values as denominator. Results: 1822 pts enrolled; 349 (19%)
ineligible (no valid central lab assessment); of 1473 eligible pts (1384 from EU, 89 from US) 63% diagnosed in 2001-2010.
Median age at diagnosis (Dx) was 68.5 ys. Of pts with known M status at Dx, 56 (5.1%) were M1; of 1046 M0 cases, 60% were
N0 and 51% had T1 tumors at Dx; 4% of pts had BCS and 18% had SLNB; half received adjuvant RT. (Neo)adjuvant CT was
used in 30% of M0 pts, most (44%) anthracycline-only; 77% of M0 pts received adjuvant ET, most tamoxifen (88.4%). Central
pathology: 697 cases with central histology & grade (full series ongoing): 87% ductal, 53% grade 2. In 1473 pts, at least 1
biomarker centrally assessed; 92% ER highly+ (Allred 7-8), PR had wider variation (35% highly+); 87% AR highly+; 25% Ki67
high; 5% HER2pos. Using IHC surrogates: 58% Luminal A-like, 35% LuminalB-like/HER2neg, 6% LuminalB-like/HER2pos; 0.1%
HER2pos/non-Luminal; 1% TNBC (16% not classified). Outcome: For 1046 M0 pts, median FU was 5.7 yrs (0-19.2); 63% alive at
analysis; in 88% cause of death was reported, mainly 33% non-cancer and 28% progression(PD)/toxicity. Significant OS
improvement over time is seen. 75% (42/56) M1 pts died, mainly due PD. In M0 pts, median OS (yrs) was significantly correlated
with ER+ p=0.001 [Allred 0-2: 3.9 (0.1, 6.0); 3-6: 7.1 (4.7, 11.7); 7-8: 8.8 (8.3, 10.0)] & with PR+ p=0.022 [Allred 0-2: 7.3 (6.1,
10.3), 3-6: 8.0 (7.0, 9.3), 7-8: 9.5 (8.4, 12.8)]; less correlation for AR; no major differences for HER2 or Ki67. Median OS & grade
(394 cases; full series ongoing): Grade1: 12.8 yrs (6.1-15.6); Grade2: 7.9 (6.5-10.7); Grade3: 9.3 (4.9-21.1). Median OS for IHC
surrogates: 8.7 (7.8-9.7) for Luminal A; 8.3 (6.9-9.4) for Luminal B/HER-2neg; 9.3 years (5.9-21.1) for Luminal B/HER-2+. Similar
results were seen for RFS. ER/PR/AR using histoscores will be presented. Conclusions: a) 56% pts had T1 tumors at Dx but only
4% had BCS; b) ER was highly + in >90% but adjuvant ET given in only 77% pts; c) Male BC is usually ER+, PR+ & AR+ and of
Luminal A-like subtype (5% HER2pos & 1% TNBC); d) Significant improvement in OS over time; e) ER and PR (Allred) are
prognostic (high expression/better prognosis), less for AR, not for Ki67 nor IHC surrogates; f) In-depth characterization of
samples is ongoing. Funding: BCRF, EBCC Council, Pink Ribbon NL, BRO.

Title: The genomic landscape of male breast cancers
Salvatore Piscuoglio1, Melissa Murray1, Charlotte KY Ng1, Elena Guerini Rocco2, Luciano G Martelotto1, Francois-Clement
Bidard3, Carey A Eberle1, Nicola Fusco2, Rita A Sakr4, Leticia De Mattos-Arruda1, Raymond Lim1, Timour Baslan5, James Hicks5,
Tari A King4, Edi Brogi1, Larry Norton6, Britta Weigelt1, Clifford A Hudis6 and Jorge S Reis-Filho1. 1Memorial Sloan Kettering
Cancer Center, New York, NY; 2School of Pathology, University of Milan, Milan, Italy; 3Institut Curie, Paris, France; 4Memorial
Sloan Kettering Cancer Center, New York, NY; 5Cold Spring Harbor Laboratory, Cold Spring Harbor, NY and 6Memorial Sloan
Kettering Cancer Center, New York, NY.
Body: Background: Male breast cancer (MaBC) accounts for <1% of all breast cancers and its genomic landscape has yet to be
characterized. The majority of MaBCs are ER-positive invasive ductal carcinomas of no special type and, unlike female breast
cancers (FBC), rarely display HER2 gene amplification or a triple-negative phenotype. Given the relative rarity of MaBCs,
treatment decisions for MaBC patients are often extrapolated from trials carried out with FBC patients. Here we employed
targeted capture massively parallel sequencing to define the repertoire of somatic mutations and gene copy number alterations
(CNAs) in MaBCs.
Methods: Subtyping of the 64 MaBCs included in this study was performed by means of immunohistochemistry using the
definitions described in the latest St. Gallens consensus report. DNA extracted from microdissected tumor and adjacent normal
tissue were subjected to massively parallel targeting sequencing of all exons of 273 genes most frequently mutated in FBCs or
directly related to DNA repair. Somatic mutations were defined using a combination of MuTect, SomaticSniper, MutationSeq and
Haplotype Caller. Selected mutations were validated with Sequenom MassARRAY. CNAs were identified using Varscan2 and
GISTIC2.0. Pathway and network analysis of mutations/CNAs was performed using Ingenuity Pathway Analysis and HOTNET.
The genomic landscape of MaBCs was compared with that of FBCs of the same subtype analyzed as part of The Cancer
Genome Atlas project.
Results: All MaBCs were ER-positive and HER2-negative. Using the St. Gallens criteria, 37.5% and 62.5% were classified as
luminal A-like or luminal B-like, respectively. The genes most frequently mutated in MaBCs were PIK3CA, GATA3, FLG and
PLEC, with PIK3CA being the only significantly mutated gene as defined by MutSigCV (q=0.003). CNA analysis revealed
recurrent gains of 1q and 8q and loss of 16q. GISTIC2.0 identified significantly recurrent high-level amplifications in 1q25.3, 8p11
(FGFR1, ZNF703), 8q24.3 (DEPTOR), 17q23 (PPM1D) and 15q26 (IGF1R) and deletions in 11q22 (ATM) and 21q22.12
(RUNX1). HOTNET analysis of the genes mutated and/or targeted by gene amplifications in MaBCs revealed significantly altered
subnetworks involving genes related to DNA repair, PI3K and FGF signaling pathways. The most frequently mutated genes in
luminal A-like MaBCs were PIK3CA and MLL3, while those of luminal B-like MaBCs were PIK3CA and GATA3. MLL3 and GATA3
mutations were only found in luminal A-like MaBCs (p=0.039) and luminal B-like MaBCs (p=0.048), respectively. Although the
mutational landscapes of MaBCs and luminal FBCs were qualitatively similar, PIK3CA and TP53 were less frequently mutated in
MaBCs (p<0.001), whereas genes found to be recurrently mutated in FBCs, such as MAP2K4 and NCOR1, were not mutated in
MaBCs.
Conclusions: MaBCs are preferentially of luminal subtype and are characterized by recurrent mutations in PIK3CA, GATA3, FLG
and PLEC. Genetic alterations in MaBCs often target DNA repair and FGF signaling pathways. The known drivers of luminal
FBCs appear to be less frequently altered in MaBCs. Given these important differences between MaBCs and FBCs, caution
should be exercised in the extrapolation of biologic and clinical implications from studies in FBCs to the management of MaBCs.

Title: Detrimental effects of sequential compared to concurrent treatment of pertuzumab plus T-DM1 in HER2+ breast cancer cell
lines
James E Korkola1, Moqing Liu1, Tiera Liby1, Laura Heiser1, Heidi Feiler1 and Joe W Gray1. 1Oregon Health & Science University,
Portland, OR.
Body: Background. Pertuzumab and T-DM1 are two recently approved monoclonal antibody based therapies targeting HER2+
breast cancer. Pertuzumab interferes with dimerization of HER family members, while T-DM1 binds to HER2 and interferes with
its oncogenic function while also specifically delivering a cytotoxic agent (emtansine). One arm of the I-SPY 2 clinical trial is to
investigate the efficacy of a combination Pertuzumab plus T-DM1 in HER2+ breast cancer patients. Methods. We performed
pre-clinical screening of response to each agent alone and in combination in a set of 21 HER2+ breast cancer cell lines, with an
end goal of identifying markers of response to the therapies. There were five treatment regimens employed in the initial screen: i)
pertuzumab alone for 72 h; ii) T-DM1 alone for 72h; iii) pertuzumab plus T-DM1 concurrently for 72h; iv) pertuzumab for 24h
followed by addition of T-DM1 for 48h more; and iv) T-DM1 for 24h followed by addition of pertuzumab for 48h more. Response
was assessed using the Cell Titer Glo assay as a measure of cell viability. To assess the effects of drug combinations, we used a
stringent measure of synergy and antagonism employing the median effect method of Chou and Talalay that included 95%
confidence intervals to determine significance. Results. Initial screens showed that concurrent treatment of cells with pertuzumab
plus T-DM1 gave significant synergistic interactions in 15/21 cell lines as measured by the median effect method, with
combination indices (CI) less than 0.5 (and 95% upper confidence levels less than 1.0) for at least one drug concentration.
However, 24h pretreatment with pertuzumab followed by T-DM1 significantly diminished the response of cells to T-DM1, resulting
in significant antagonism in 17/21 cell lines test (CI>1.5, lower confidence level greater than 1). Since this could be due to a
shorter exposure time to T-DM1, and since patients are scheduled to be treated with pertuzumab first followed by T-DM1 one
hour later, we repeated the experiment with one hour between pertuzumab and T-DM1 rather than 24h. While the inhibitory effect
was diminished, this treatment regimen still resulted in significant antagonism when T-DM1 was given 1 hour after pertuzumab in
5/5 cell lines tested, in contrast to concurrent pertuzumab plus T-DM1 treatment, which showed synergy. Conclusions.
Pertuzumab plus T-DM1 appears to be beneficial when given concurrently, but pretreatment with pertuzumab appears to blunt
the efficacy of T-DM1. This has important potential ramifications for patient treatment, and may further elucidate mechanisms of
action for both compounds. Further testing will be necessary to determine whether these timing effects are operational in vivo and
whether immune effects mitigate the antagonism.

Publication Number: P1-01-01
Title: Withdrawn by Author

Title: Mammographic breast density and magnetic resonance (MR) imaging in women with mammographically occult breast
cancer
Chun Jennifer1, Freya Schnabel1, Shira Schwartz1, Jessica Billig1, Jennifer Gillman1 and Linda Moy1. 1NYU Langone Medical
Center, New York, NY.
Body: Background: Recent data suggests a correlation between mammographically-occult breast cancer (MOBC) and
increasing breast density. There is a dearth of information on the assessment of the amount of fibroglandular tissue (FGT) with
contiguous MR images through both breasts and background parenchymal enhancement (BPE) in women with MOBC. The
purpose of this study was to evaluate the relationship between mammographic breast density (BD), BPE, and FGT in women with
MOBC and mammographically evident breast cancer (MEBC).
Methods: The Breast Cancer Database at our medical center was queried for all women who were newly diagnosed with breast
cancer between 2010 and 2014. Variables included age, BD, BPE, FGT, tumor characteristics. Statistical analyses included
Pearsons Chi Square Tests.
Results: Out of a total of 1781 women, there were 110 (6%) with MOBC and 1671 (94%) with MEBC. The median age of the
cohort was 59 years. Majority of the patients had early stage disease (72%) and invasive ductal carcinoma (61%). There was a
higher proportion of women with DCIS with MEBC (22%) compared to women with MOBC (16%) and a higher proportion of
invasive lobular carcinoma in women with MOBC (17%) compared to the women with MEBC (10%). Increased mammographic
BD was significantly associated with mammographically occult breast cancer (p<0.001). However, BPE (p=0.29) and FGT
(p=0.28) were not associated with MOBC. In the MOBC group, there was a higher proportion of women with dense breasts on
mammography (74%) compared to women with dense breasts on MRI (48%).
Conclusions: For women who develop mammographically occult breast cancer, there are currently no guidelines regarding
recommendations for future screening. Increased breast density reduces the sensitivity of mammography, which compromises
the ability to detect meaningful findings. Our study shows that MRI may be less affected by obstacles of sensitivity that is usually
associated with mammography. Our results support the use of MRI in women with dense breasts who are at higher risk of
developing MOBC. By better understanding the patient population who are more likely to be diagnosed with
mammographically-occult breast cancer, we can more effectively select patients who should be screened with breast MRI.
Table 1. Mammographic breast density, BPE, and FGT in study population
VARIABLES
TOTAL (N=1781)
MOBC (N=110)
MEBC (N=1671)
P-VALUES
p<0.001
MAMMOGRAPHIC BREAST DENSITY

Predominantly fatty
94
91
Scattered fibroglandular
587
38
25
23
562
40
Heterogeneously dense
715
47
67
61
648
46
Extremely dense
135
14
13
121
BACKGROUND PARENCHYMAL ENHANCEMENT

Minimal
74
16
15
24
59
15
Mild
225
49
30
48
195
49
Moderate
116
25
13
21
103
26
Marked
47
10
42
11
p=0.29
FIBROGLANDULAR TISSUE
Predominantly fatty
91
21
12
84
23
Scattered fibroglandular
160
37
24
41
136
37
Heterogeneously dense
121
28
20
34
101
27
p=0.28
Extremely dense
57
13
14
49
13

Title: Value of breast MRI for preoperative axillary assessment of breast cancer patients
Sean C Dupont1, Judy C Boughey1, Tanya L Hoskin1, Katrina N Glazebrook1 and Tina J Hieken1. 1Mayo Clinic, Rochester, MN.
Body: Background: Many newly diagnosed breast cancer patients undergo preoperative breast MRI which includes visualization
of some degree of the axilla. The impact of patient characteristics and tumor biology on the fidelity of MRI imaging of the axilla in
these cases is not well-studied. We sought to examine the correlation between nodal findings on breast MRI and pathologic nodal
status in newly diagnosed breast cancer patients.
Methods: With IRB approval, we identified 1868 consecutive breasts with invasive cancer in 1803 patients undergoing primary
operation with axillary surgery from 1/2010-7/2013. Patients undergoing neoadjuvant chemotherapy (348) were excluded leaving
1510 breasts with cancer in 1455 patients. Preoperative MRI was performed in 763 patients (52%). We evaluated patient,
imaging and pathology data.
Results: Patients evaluated with MRI were median age 57 years. The majority of tumors were T1 (63%) and T2 (28%). MRI
identified suspicious axillary nodes in 240 cases (31%), of which 123 (51%) were node positive at operation. Suspicious axillary
findings on MRI predicted both nodal status and pN stage at operation (both p<0.0001). The sensitivity, specificity, accuracy,
positive (PPV) and negative predictive value (NPV) of MRI for nodal disease across tumor types is shown.
Pathologic Node
Positive
All Cases (n=763) 234 (31%)
Sensitivity (95%
CI)
52.6 (46-59.1)
Specificity (95%
NPV (95% CI) PPV (95% CI)
CI)
Accuracy (95%
CI)
77.9 (74.1-81.3)
78.8 (75-82.1)
51.3
(44.8-57.7)
70.1 (66.7-73.3)
Approximated Biologic Subtype (missing=24)

ER+/HER2(n=609)
194 (32%)
51.0 (43.8-58.2)
80.0 (75.8-83.7)
77.8
(73.4-81.5)
54.4
(46.9-61.7)
70.8 (67-74.3)
HER2+ (n=69)
20 (29%)
65.0 (40.9-83.7)
71.4 (56.5-83)
83.3 (68-92.5)
48.1
(29.2-67.6)
69.6 (57.2-79.8)
ER-/HER2- (n=61) 17 (28%)
58.8 (33.5-80.6)
75.0 (59.4-86.3)
82.5
(66.6-92.1)
47.6 (26.4-70) 70.5 (57.3-81.1)
p value
0.43
0.33
0.56
0.73
0.98
0.74
Tumor Grade (missing=5)

1 (n=259)
58 (22%)
32.8 (21.4-46.5)
79.6 (73.2-84.8)
80.4
(74.1-85.5)
31.7
(20.6-45.1)
69.1 (63.0-74.6)
2 (n=351)
124 (35%)
55.6 (46.5-64.5)
79.7 (73.8-84.6)
76.7
(70.7-81.8)
60.0
(50.4-68.9)
71.2 (66.1-75.9)
3 (n=148)
51 (34%)
66.7 (52.0-79.0)
70.1 (59.8-78.8)
80.0
(69.6-87.6)
54.0
(41.0-66.4)
68.9 (60.7-76.1)
p value
0.001
0.0003
0.11
0.72
0.02
0.93
Diagnostic performance did not vary significantly based on patient body mass index (BMI) or approximated biologic subtype, but
specificity was better for patients >age 50 (p=0.007) and sensitivity and PPV both were worse for grade 1 vs grade 2/3 tumors.
Node positivity rate at operation was 48% (51/106) for patients with a solitary abnormal node on MRI and 54% (72/134) when >1
suspicious node was seen (p=0.39). However, multiple vs solitary MRI suspicious nodes correlated with 3 positive nodes at
operation (40 [30%] vs 19 [18%], p=0.03) and pN2/pN3 disease (29 [22%] vs 12 [11%], p=0.03).
Conclusion: Axillary lymph node findings on MRI for breast cancer predict nodal status and disease volume in invasive breast
cancer patients. Tumor biologic subtype did not affect performance characteristics of preoperative MRI, but tumor grade did
influence the sensitivity and PPV of MRI. When MRI is performed in the evaluation of newly diagnosed breast cancer, axillary
findings inform pathologic nodal stage at operation regardless of tumor subtype.

Title: Evaluation of tumor response after neoadjuvant chemotherapy in breast cancer patients: Correlation between dynamic
contrast-enhanced MRI and pathologic tumor cellularity
Hak Hee Kim1, Won Kyung Kim1, Joo Hee Cha1, Hee Jung Shin1, Eun Young Chae1 and Woo Jung Choi1. 1Asan Medical Center,
University of Ulsan College of Medicine, Seoul, Korea.
Body: Purpose: The objective of the study was to analyze whether the measurement of changes of the tumor on magnetic
resonance imaging (MRI), could be applied to predict pathologic response after neoadjuvant chemotherapy.
Methods: Between January 2013 and October 2013, 155 patients (age range, 31-74 years; mean age, 47.6 years) with
pathologically proven breast cancer who underwent neoadjuvant chemotherapy followed by surgery were retrospectively enrolled.
Pre-and post-treatment measurements and changes in tumor size and dynamic curve pattern of MRI were analyzed. Histological
regression was scored on the residual overall tumor cellularity based on Miller-Payne grading system.
Results: The change of maximal diameter ( = 0.681, p < 0.001), angio-volume of tumor ( = 0.683, p < 0.001) and washout ratio
of dynamic curve ( = 0.607, p < 0.001) measured on MRI showed strong correlation with histological regression. Among them,
the correlation factor was highest in the change of angio-volume. The change of plateau ratio of dynamic curve showed moderate
correlation ( = 0.574, p < 0.001) with histological regression. However, the change of persistent ratio of dynamic curve showed
weak correlation ( = 0.206, p = 0.011).
Conclusions: The change of maximal diameter, angio-volume of tumor and washout ratio of dynamic curve measured on MRI
showed strong correlation with residual cellularity in lesions after neoadjuvant chemotherapy and could be applied to predict
pathologic response.

Title: Effects of preoperative MRI on rate of ipsilateral and contralateral recurrence of breast cancer
Amanda L Amin1, Irene B Helenowski1, Thomas E Kmiecik1, Shruti R Zaveri1, Nora M Hansen1, Kevin P Bethke1 and Seema A
Khan1. 1Northwestern University, Chicago, IL.
Body: Introduction: Preoperative MRI of the breast is the most sensitive imaging modality in the detection of multifocal or
multicentric breast cancer, as well as simultaneous contralateral breast cancer. The aim of this retrospective review was to
evaluate the effect of preoperative MRI on local control for patients with breast cancer.
Methods: The Enterprise Data Warehouse of Northwestern Medicine was searched for women who underwent breast conserving
surgery for ductal carcinoma in situ (DCIS) or primary invasive breast cancer in the interval of 2004-2010. The use of preoperative
MRI, and the clinical and therapeutic details of the patients thus identified were extracted by direct review of the electronic
medical record. A breast event was defined as a local recurrence in the treated breast more than six months after completion of
treatment (ipsilateral) or a new breast cancer in the untreated breast (contralateral). Differences in the frequency of all events
(local and distant), for ipsilateral breast events, and for contralateral breast events was evaluated with Cox proportional hazards
model, adjusting for patient age, tumor size, nodal status, the presence of triple negative disease, and the use of radiotherapy
and systemic therapy.
Results: In our cohort of 1097 patients, 526 had preoperative MRI and 571 had no MRI. The patients who had preoperative MRI
were younger (59 vs. 66 years, p<0.0001), were more commonly premenopausal (37.8% vs. 27.3%, p=0.0004), were more likely
to present with palpable tumors (34.8% vs. 26.6%, p=0.004), were more likely to have invasive lobular disease (16% vs. 11.6%),
and less likely to have DCIS (16.5% vs. 29%, p=0.001 for differences in histologic pattern). Mean tumor size was equivalent in the
two groups (17.5 and 17.3 mm), but nodes were more frequently positive in the MRI group (23.9% vs. 19.1%, p=0.045). Triple
negative tumors were more frequent in the MRI group (14.1% vs. 7.5%, p=0.0003). Mean follow up was 51.5 months in the MRI
group and 59.4 months in the no MRI (p<0.0001). The number of events was 49 in the MRI group and 68 in the no MRI group.
The Cox hazard ratio (HR) for all events (adjusted for follow-up duration and factors described in the Methods) was equivalent
between the two groups (HR 0.90, 95% CI 0.59-1.36, p=0.61). The HR for ipsilateral (HR 0.93, 95% CI 0.57-1.51, p=0.76) and
contralateral events (HR 1.22, 95% CI 0.57-2.62, p=0.61) was equivalent between the two groups.
Conclusions: In analyses adjusted for important prognostic features, the use of preoperative breast MRI was not associated with
a reduced hazard of any breast cancer event, or of in-breast events (ipsilateral or contralateral). However, the MRI group had a
more adverse tumor and patient profiles; a propensity score analysis will be performed, to further adjust for these differences.
These findings add weight to the position that routine use of preoperative MRI for all breast cancer patients is not beneficial.

Title: Quantitative breast DCE-MRI risk biomarkers and breast tumor immunohistochemistry phenotypes: A preliminary
assessment
Shandong Wu1, Margarita L Zuley1,2, Brenda F Kurland1, Rachel C Jankowitz1,2, Jules Sumkin1,2 and David Gur1. 1University of
Pittsburgh, Pittsburgh, PA and 2UPMC Magee-Womens Hospital, Pittsburgh, PA.
Body: Purpose
Fibroglandular tissue (FGT) and background parenchymal enhancement (BPE) estimated from breast dynamic contrast
enhanced MRI (DCE-MRI) have been correlated with breast cancer risk. We performed a preliminary study assessing in a small
cohort of breast cancer patients the relationship between FGT and BPE both quantified from cancer-unaffected breasts and
breast tumors immunohistochemistry phenotypes, including testing of estrogen (ER), progesterone (PR), and human epidermal
growth factor receptor 2 (HER-2/neu) receptors.
Materials and Methods
We retrospectively identified 51 breast cancer patients who were diagnosed with unilateral breast cancer from 2009-2011. Six out
of the 51 were excluded for analysis due to missing data in ER, PR, or Her-2 testing in their pathology reports. The 45 cancer
patients (mean 47.37.6 YO; 24 IDC, 4 ILC, 16 mixed of IDC and DCIS, and 1 papillary carcinoma) consist of 25 premenopausal
and 20 postmenopausal women. Breast DCE-MRI scans acquired at time-of-diagnosis were analyzed using fully automated
computer algorithms. From the contralateral cancer-free breast, four quantitative variables were computed for each patient: the
absolute volumes of FGT and BPE (i.e., |FGT| and |BPE|) and their relative amount over the whole-breast volume (i.e., FGT%
and BPE%). BPE were assessed over the FGT region of the breast, and separately from each of the three sequential DCE
post-contrast-subtracted sequences (i.e., SUB1, SUB2, and SUB3). Multivariable logistic regressions were performed to assess
the quantitative FGT/BPE variables for distinguishing/predicting the status (+/-) of ER, PR, and HER-2 testing. Area under the
curve (AUC) of the receiver operating characteristic (ROC) was used to evaluate the regression models discriminative
performance. Odds ratios (ORs) were reported with 95% confidence intervals (CIs).
Results
In this cohort, the most common phenotypes were ER+/PR+ (14/45 each HER2+ and HER2-) and triple negative (9/45).
Premenopausal status was associated with tumors that were HER-2+ (AUC=0.66; p=0.04), ER+ (AUC=0.73; p=0.02), and PR+
(ACU=0.66; p=0.04). FGT% (AUC=0.74; p=0.017; OR=5.3 [95% CI: 1.3, 21.1]) and BPE%-SUB3 (AUC=0.69; p=0.02; OR = 1.4
[95%CI: 1.1, 1.9]) were discriminative of HER-2 status; after adjusting for menopausal status, FGT% remained the association
(AUC=0.77; p=0.04; OR=4.4 [95% CI: 1.1, 18.5]) but BPE%-SUB3 did not (p=0.06). BPE%-SUB1 was discriminative of PR status
(AUC=0.70; p=0.03; OR=2.3 [95%CI: 1.1, 4.8]) but did not contribute to predict when adjusting for menopausal status (AUC=0.75;
p=0.06). All FGT and BPE variables were not associated (p>0.05) with ER status.
Conclusions
In this preliminary study, FGT% and BPE% were associated with HER-2 and PR status, and none of FGT and BPE was
associated with ER status, of breast tumor.
Clinical Relevance
Quantitative MRI variables of FGT and BPE that are predictive of breast cancer risk are related to breast tumor
immunohistochemistry phenotypes. While FGT and BPE cannot replace the testing of these phenotypes, they may contribute to
assess the risk of molecular subtypes of breast cancer, with potential implications for improving breast cancer screening
strategies.

Title: Quantitative assessment of early- and delayed DCE-MRI background parenchymal enhancement in breast cancer risk
prediction
Shandong Wu1, Wendie A Berg1,2, Margarita L Zuley1,2, Brenda F Kurland1, Rachel C Jankowitz1,2, Jules Sumkin1,2, Robert M
Nishikawa1 and David Gur1. 1University of Pittsburgh, Pittsburgh, PA and 2UPMC Magee-Womens Hospital, Pittsburgh, PA.
Body: Purpose
Background parenchymal enhancement (BPE) estimated from breast dynamic contrast enhanced MRI (DCE-MRI) has been
correlated with breast cancer risk. Multiple time-point post-contrast-subtracted sequences are acquired in typical clinical breast
DCE-MRI protocols. We quantitatively assessed BPE using fully automated computer algorithms and investigated the relationship
between BPE quantified from three time-point post-contrast sequences and breast cancer risk prediction.
A retrospective case-control study was performed using breast DCE-MRI scans from 102 patients (mean 47.27.3 YO) who
underwent either open surgical biopsy or core biopsy from 2009-2011: 51 women had unilateral breast cancer and 51 were ageand date-of-MRI matched controls with a unilateral biopsy-proven benign. For each MRI scan, three post-contrast-subtracted
sequences (i.e., SUB 1, SUB 2 and SUB 3) acquired over a total of 7 minutes were analyzed. BPE was quantified from each of
the sequences using fully automated computer algorithms on the breasts contralateral to the cancers and the contralateral
(negative) breasts of the controls. For each sequence, two quantitative BPE measures were generated: the absolute BPE volume
(|BPE|) and its relative amount over the whole breast volume (BPE%). Volumetric absolute and relative amounts of fibroglandular
tissue (|FGT| and FGT%) were also automatically quantified, from the pre-contrast sequence. BI-RADS breast density
assessment was retrieved from the mammography report (< 6 months) prior to cancer diagnosis (for cases) or MRI (for controls).
Multivariable conditional logistic regression was performed to assess BPE measures, quantified respectively from SUB 1, SUB 2,
and SUB 3, as predictors of breast cancer risk.
Results
Breast cancer risk odds ratios (ORs) were reported by |BPE| and BPE% (Table 1), after adjustment for breast density, |FGT|, and
FGT%. Breast density did not correlate with risk for breast cancer in this study cohort: OR for BI-RADS density alone was 0.75
(95% CI: 0.35, 1.59; p=0.5). OR was 1.14 (95% CI: 0. 72, 1.81; p=0.6) for |FGT| alone and 0.70 (95% CI: 0.19, 2.52; p=0.6) for
FGT% alone.
Table 1: Breast cancer risk odds ratios (ORs) with 95% confidence intervals [CIs] for BPE quantified from SUB 1, SUB 2, and
SUB 3, respectively.
BPE from SUB1
BPE from SUB2
BPE from SUB3
|BPE|
2.0 (95% CI: 1.1-3.7; p=0.02)
2.0 (95% CI: 1.2-3.3; p=0.01)
1.7 (95% CI: 1.1-2.7; p=0.02)
BPE%
3.6 (95% CI: 1.3-10.1; p=0.01)
2.8 (95% CI: 1.3-6.2; p=0.01)
2.4 (95% CI: 1.2-5.0; p=0.02)
Conclusions
Increased BPE quantified from DCE-MRI are predictive of breast cancer risk, independent of measures of breast density and
FGT. BPE estimated from three time-point SUB sequences has a similar predictive effect of breast cancer risk, while an early
sequence (e.g., SUB 1) appears to have a larger magnitude of effect than that of a delayed sequence (e.g., SUB 3).
Clinical Relevance
Quantified BPE in breast DCE-MRI has potential for use as a biomarker of breast cancer risk and may be included to improve
breast cancer risk prediction. A single post-contrast sequence (i.e., SUB 1) may be adequate for use to estimate breast cancer
risk using breast MRI in the context of breast cancer screening for high-risk women.

Title: Changes in breast density with mammography and breast MRI among high-risk postmenopausal women
Parijatham S Thomas1, Mary Beth Terry1, Xiaotao Guo1, Binsheng Zhao1, Richard Ha1, Swathi Namburi1, Tong Xiao1, Dawn
Hershman1, Susan Refice1 and Katherine D Crew1. 1Columbia University Medical Center, New York, NY.
Body: Background: Mammographic density (MD) represents one of the strongest predictors of breast cancer risk. Tamoxifen, a
selective estrogen receptor modulator (SERM) and proven chemopreventive agent, causes reductions in MD which correlate with
subsequent breast cancer risk. Current methods for measuring MD, such as the Cumulus technique, are labor intensive and not
fully automated, which limits utility in the clinical setting. Breast MRI is being used with increasing frequency in high-risk women
for breast cancer screening and yields 3-dimensional views of breast volume. Breast MRI fibroglandular tissue volume (MRIV)
has been correlated with MD in cross-sectional studies, but less is known about changes in breast density over time on MRI and
its relation to breast cancer risk. One study reported a significant reduction in breast MRIV but not MD with the SERM, raloxifene,
in high-risk women. We implemented a novel MRI segmentation method to measure MRIV compared to MD in 20 high-risk
postmenopausal women that received high-dose vitamin D supplementation for 1 year.
Methods: Twenty high-risk postmenopausal women [defined as a 5-year breast cancer risk per the Gail model of 1.67%, lobular
or ductal carcinoma in situ (LCIS/DCIS)] were assigned to a 1-year intervention of vitamin D3 20,000 IU or 30,000 IU weekly.
Other eligibility criteria included baseline MD 25% (as assessed qualitatively by BIRADS category) and no current SERM use.
Women underwent a digital mammogram and breast MRI at baseline and 12 months. MD was evaluated by the Cumulus
technique and MRIV was measured using a novel segmentation method on noncontrast T1-weighted images from breast MRIs.
The method incorporated two well-established image analysis techniques, the region-based active contour model and watershed
algorithm, into a unified framework. Specific anatomical landmarks were introduced into the algorithm to consistently identify
breast boundaries on longitudinal MRI slices. For fibroglandular tissue segmentation, an algorithm combining region-based active
contours and level-set approach was used to allow for easy initialization and fast speed. Correlation coefficients and paired t-tests
were used to compare MD and MRIV and changes from baseline to 12 months.
Results: From Nov 2007 to Jan 2011, 20 postmenopausal women were enrolled and 16 were evaluable at 12 months. Mean
baseline breast density measurements were lower for MD compared to MRIV, 12.89% (SD 11.99) vs. 26.64% (SD 6.95),
respectively. After 1 year of high-dose vitamin D, there was no significant change in MD or MRIV, mean absolute changes of
-0.14% vs. -1.12%, respectively. We did not observe a significant correlation between MD and MRIV density measurements
(correlation coefficient= -0.31, p=0.08).
Discussion: We observed greater changes over time in breast density on serial breast MRI compared to mammography,
however, no significant changes in MD or MRIV was noted after a 1-year intervention of vitamin D. Longer follow-up may be
required in order to detect changes in breast density with non-hormonal interventions.

Title: Optimization of magnetic resonance imaging predictive performance by breast cancer subtypes for predicting response to
neoadjuvant chemotherapy
Wen Li1, Wei-Ching Lo1, Ella F Jones1, David C Newitt1, John Kornak2, Lisa J Wilmes1 and Nola M Hylton1. 1University of
California San Francisco (UCSF), San Francisco, CA and 2UCSF, San Francisco, CA.
Body: Objective: Change in functional tumor volume (FTV) measured by dynamic contrast-enhanced magnetic resonance
imaging (DCE-MRI), has been shown to be associated with response to neoadjuvant chemotherapy (NACT) for patients with
stage II/III breast cancer. FTV reflects the vascularized volume of tumor and is calculated by applying minimum thresholds PEt for
the initial percent enhancement and SERt for the early-to-late signal enhancement ratio following gadolinium contrast injection. In
this retrospective study, we investigate the association of FTV influenced by PEt and SERt values with recurrence-free survival
(RFS) in breast cancer subtypes defined by hormone receptor (HR) and HER2 status. Materials and Methods: 64 patients with
locally advanced breast cancer were imaged by DCE-MRI before treatment (MRI1), after one cycle of adriamycin-cytoxan (AC)
therapy (MRI2), inter-regimen (MRI3, taxane receivers only) and at the completion of chemotherapy prior to surgery (MRI4).
Because treatment length and regimen varied among patients, the MRI exam performed after all NACT and before surgery was
designated as final MRI (MRIf). FTV was calculated with varying PEt (30200% in steps of 10%) and SERt (02 in steps of 0.2)
values. A Cox proportional hazard model was used to analyze association with RFS, defined as the time between surgery and
disease recurrence or last disease-free follow-up, for the following imaging predictors: early percent change in FTV (FTV2: MRI1
vs. MRI2), final percent change (FTVf: MRI1 vs. MRIf), and final FTV in MRIf (FTVf). RFS association was evaluated for the full
cohort and within subsets defined by tumor subtype (HR+/HER2-, HER2+, triple negative). Estimated hazard ratios per unit
change in predictors, associated 95% confidence intervals (CI) and p-values were evaluated. The cutoff of p<0.05 was used to
differentiate PEt/SERt values with higher predictive performance. Results: For the full cohort, FTVf showed the most robust
association with RFS among the three predictors, with p<0.05 across all measured combinations of PEt/SERt. The PEt/SERt
combination with smallest p-value was found for PE=30% and SER=0.2, with hazard ratio 1.09 (95% CI 1.061.13, p<0.001).
P-values less than 0.05 for the FTVf predictor was found for the full range of SERt values tested, but was mostly confined to PEt
in 30110% range; FTV2 resulted in p<0.05 for very few PEt/SERt combinations. When analyzed by tumor subtypes, different
prediction profiles were observed. FTVf showed p<0.05 for the lower PEt range only in the HER2+ (n=15) and triple-negative
(n=11) groups. A limited number of PEt/SERt pairs showed p<0.05 for FTVf and FTV2, with the exception of the FTVf in the
triple-negative group, for which p <0.05 for many PEt/SERt combinations in the low PEt range. Conclusions: This study was
undertaken to explore the significance of FTV for predicting breast cancer recurrence following NACT. While this study is
retrospective in nature and had small sample size, the findings nevertheless suggest that performance of imaging predictors
based on FTV may be improved if threshold optimization is performed separately for the clinically-relevant subtypes defined by
HR and HER2 receptor expression.

Title: A preoperative PET/MR imaging for diagnosing malignant axillary lymph nodes in women evaluated for breast cancer
Kyung Jun Yeu1, Jeong Yeong Park1, Jung Eun Choi1, Su Hwan Kang1, Eun-Jung Kong2, Ihn-Ho Cho2 and Soo Jung Lee1.
1
Yeungnam University College of Medicine, Daegu, Korea and 2Yeungnam University College of Medicine, Daegu, Korea.
Body: Objective: We determined whether and how malignant node could be diagnosed preoperatively with simultaneous
PET/MR imaging.
Materials and methods:Two hundred thirty-six consecutive women with breast cancer were recruited to undergo preoperative
PET/MR imaging. 52 patients were excluded; 31 DCIS, 7 neoadjuvant chemotherapy and 14 previous excisional biopsy. Thus,
184 women (51.9 10.2 years old) were included. Axial T1w without fat saturation, contrast enhanced image and
diffusion-weighted images with simultaneous acquired PET images were analyzed by nuclear medicine physicians who were
blinded the histologic findings. Visual FDG avidity, long axis and cortical thickness of axillary lymph node (ALN), morphologic
feature-loss of fat hilum, cortical thickening, round shape or irregular shape- and ADC value were analyzed in most suspicious
ALNs.
Results:In total, 70 patients (15 patients have only micrometastasis) exhibited ALN metastasis. Mean size of breast mass was 2.1
1.5 cm. Macrometastatic ALN showed high FDG uptake, longer axis, thicker cortex, more frequent morphologic abnormalities,
higher signal intensity at DWI and higher ADC values with statistical significance. No significant difference between
micrometastatic ALN and benign ALN in PET/MR imaging. With pathologic diagnosis as the reference standard, the sensitivity,
specificity and accuracy of PET for determining ALN metastasis were 84%, 58% and 68%, respectively. Those are 77%, 84% and
82% in considering both PET with morphologic change and 81%, 66% and 72% in considering PET, morphologic change and
DWI, respectively.
Conclusion:PET/MR imaging techniques showed high accuracy in the preoperative evaluation of axillary status in patients with
breast cancer. Additional information by DWI is unlikely to be useful in predicting metastatic ALN.

Title: The impact of a critical look at the consequences of preoperative MRI in breast cancer patients
I Ching Yeung1, Mehrzad Namazi2, Valerie Deslauriers1, Fatima Haggar2 and Angel Arnaout1,2. 1Ottawa Hospital, Ottawa, ON,
Canada and 2Ottawa Hospital Research Institute, Ottawa, ON, Canada.
Body: BACKGROUND: Despite the fact that routine use of preoperative breast MRI for breast cancer has not been shown to
improve oncologic outcomes, it is still an exceedingly popular test. Due to low MRI specificity, patients may be subjected to
additional invasive test. There have been few studies specifically evaluating the outcomes from these additional tests and its
implications on the health care system. The objective of this study was to critically evaluate the impact of performing a
preoperative MRI on breast cancer patients at our institution.
METHODS: A retrospective chart review was performed on all female breast cancer patients diagnosed and awaiting surgery
(2010-2012). Based on extracted data, indications for preoperative breast MRI were established for our institution (2013).
Adherence to these guidelines was then assessed over the next year.
RESULTS: In 2010-2012, 1159/1674 breast cancer patients underwent a preoperative breast MRI. The MRI group was younger
(p<0.0001), but not different in histologic subtype (p=0.06) or biomarker status (0.61). 421/1159 (36%) of MRI patients underwent
at least one additional MRI induced imaging test (ultrasound, mammogram, 6 month MRI) following the MRI. 35% (415/1159) of
patients underwent an additional MRI-induced biopsies, 52% of which were benign in the breast and 62% of which were benign in
the axilla. Post-MRI biopsies resulted in upstaging (DCIS to invasive cancer; node negative to node positive) in 25/1159 (2%) of
patients.
Preliminary data assessing local guideline adherence demonstrated that 188/349 (54%) of new breast cancer patients underwent
a preoperative MRI. Locally accepted indications for MRI use included: dense breast (26%); assessment of tumor extent; (19%);
lobular carcinoma (16%); assessment of locally advanced cancer (16%). 71% of MRI studies were ordered by the radiologist.
22% of patients had no obvious indication for MRI according to our local guidelines. MRI-induced biopsies occurred in 76/189
(40.2%) of patients, 60% of which were benign in the breast and 71% of which were benign in the axilla.
CONCLUSION: We have critically evaluated the impact of preoperative breast MRI on a large volume of patients. Often, MRI
results did not result in significant treatment change. Barriers to guideline evidence based care implementation continue to exist in
the setting of multidisciplinary breast cancer care. We must continue to work together to best counsel our patients and effectively
manage our health care cost, in keeping with the Choosing Wisely Campaign Canada.

Title: Preoperative breast MRI does not affect survival outcomes in T1-2 breast cancer patients who underwent
breast-conserving therapy
Jegyu Ryu1, Sanghwa Kim1, Jee Ye Kim1, Hyung Seok Park1, Seho Park1 and Seung il Kim1. 1Yonsei University College of
Medicine, Seoul, Korea.
Body: Purpose: The aim of the study is to evaluate the efficacy of preoperative breast MRI in T1-2 women with breast cancer.
Methods: Total 1179 patients, who had T1-2 breast cancer and underwent definitive surgery between 2007 and 2010, were
reviewed in this study. Preoperative breast MRI was performed in 981 women with breast cancer. We divided patients into two
groups: Patients underwent preoperative MRI and breast-conserving therapy (BCT), those who did not undergo preoperative MRI
but BCT. Patients underwent preoperative MRI and total mastectomy were excluded. Clinicopathological features were analyzed
using Chi-square or Fishers exact test if indicated. Survival analyses were examined using Kaplan-Meier method and log-rank
test. Cox-proportional hazard model was accessed as multivariate analysis.
Results: In 873 patients who underwent BCT, 675 patients (77.3%) received preoperative MRI, and 198 (22.7%) patients did not.
Clinicopathological features including T-stage, Nodal status, histologic type, progesterone receptor, histologic grade, age group (
35 vs. > 35 yr), adjuvant chemotherapy, radiation, and hormone therapy were not significantly different between two groups.
Patients aged 50 or younger received more preoperative MRI than those over aged 50 years (42.8 % vs. 51.5%, p=0.02).
Estrogen receptor positive-tumors were more common in patients with preoperative MRI (74.9% vs. 67.2%, p=0.03). HER2
over-expression was frequently found in women without preoperative MRI (7.9% vs. 14.6%, p<0.001). In univariate analyses,
there was no significant difference in recurrence-free and overall survival between patients who underwent breast MRI and those
who did not (p for RFS = 0.75, p for OS = 0.48). In multivariate analyses, preoperative MRI did not influence on RFS and OS (HR
for RFS = 0.82, 95%CI= 0.35-1.91, HR for OS = 1.52, 95% CI= 0.40-5.76). Age group and HER2 were not associated with RFS
and OS (all p>0.05). ER was an independent prognostic factor for RFS and OS (all p<0.01).
Conclusions: Routine use of preoperative MRI in women with T1-2 breast cancer may not translate into better RFS and OS.

Title: Triple negative breast cancer targeting paramagnetic nanoparticle for non-invasive tumor imaging
Meser M Ali1, Nadimpalli RS Varma1, Branislava Janic1, Li Zhang1 and James R Ewing1. 1Henry Ford Hospital, Detroit, MI.
Body: Triple-negative breast cancer (TNBC) refers to a subtype of breast cancer that is negative for estrogen receptors (ER)
and/or progesterone receptors (PR) , and lacks HER2 overexpression. Therefore, this subtype of breast cancer lacks the benefits
of specific therapies which target these receptors. TNBC has been characterized by an acidic extracellular environment. In human
aggressive breast tumors, pHe has been measured by microelectrode and the values are in good agreement with the values
observed in animal systems i.e. that the pHe is significantly acidic in the range from 6.2 to 7.0. However, tumor intracellular pH is
either neutral or alkaline. Interestingly, a similar pH gradient is not observed in normal tissues. Therefore, this acidic extracellular
pH (pHe) within tumor tissues can be exploited for targeted delivery of drugs and imaging agents. Recently, A pH low insertion
peptide (pHLIP) derived from the protein bacteriorhodopsin has been found to target tumor acidic pH. The peptide inserts across
cell membranes as an -helix when extracellular pH (pHe) is acidic, but does not form the helix at normal or alkaline pH. Since
aggressive TNBC has an acidic environment, the pHLIP will insert into the cancer cell membrane, but the pHLIP will not insert
into the cell membranes of normal tissues, providing excellent specificity for targeting TNBC. Here, we demonstrate that
pHLIP-tagged nanoparticles bind to and are internalized by TNBC cells in vitro. Systemic delivery of the Gd-G5-pHLIP leads to
accumulation of the nanoparticles in a flank mouse model of TNBC tumor that are detected by optical and MR imaging.
We have synthesized pH-responsive MRI nanoprobe, phosphonate G5-(GdDOTA-4AmP) by following our published synthetic
method. The MW of the conjugated G5 dendrimer was estimated at 79,082 g/mole by maldi-tof analysis. This corresponds to a
G5-dendrime with an average of 44 chelated Gd3+ ions per dendrimer. Gd44-G5 dendrimer was reacted with heterobifunctional
cross-linker (sulfo-SMCC) to form reactive maleimides and then maleimide- Gd44-G5 dendrimer was coupled with C-terminus
cysteine group of biotinylated Bt-pHLIP (AEQNPIYWARYADWLFTTPLLLLDLALLVDADEGTCG-pegBiotin). The HABA assay
with biotin and avidin revealed that on average 3.1 molecules of biotin are present in Gd44-G5-Bt-pHLIP dendrimer. Finally,
Rhodamine dye was conjugated to amines surface of preloaded Gd44-G5-Bt-pHLIP3 in order to achieve final conjugate
Rhodamine-Gd44-G5-Bt-pHLIP3.
To study pH-dependent translocation of molecules across the cell membrane, we have added Rho-Gd-G5-Bt-pHLIP to the cells
and incubated for 3 h at pH 7.4 and 6.5. The cellular uptake of Rho-Gd-G5-pHLIP was significantly higher at pH 6.5. When
Rho-Gd-G5 was used, the cellular uptake was considerable lower at both, pH 6.5 and 7.4. Hence, we have shown the ability of
pHLIP peptide for intracellular delivery of Gd-G5 nanoparticles in vitro at pH 6.5 but the same ability is attenuated significantly at
neutral pH. We have created a mouse model of TNBC MDA-MB-231 tumor. The pharmacokinetics of Gd44-G5-pHLIP was
visualized in the MDA-MB-231 tumor over the course of 105 min post-contrast administration.

Title: Evaluation of second-generation photoacoustic mammography in detecting the breast cancer vasculature and hypoxic
status; a preliminary study
Masahiro Kawashima1, Iku Yamaga1, Masae Torii1, Mariko Tokiwa1, Fakhrejahani Elham1, Masako Kataoka1, Shotaro Kanao1,
Masahiro Takada1, Yasufumi Asao1,2, Tsuyoshi Shiina1 and Masakazu Toi1. 1Kyoto University, Kyoto, Japan and 2Canon Inc,
Tokyo, Japan.
Body: Background: Functional imaging of tumor vasculature and oxygenation status is essential for monitoring the therapeutic
response to the manipulation of abnormal vasculature. Moreover, it could be also applicable for the detection and risk
assessment of breast lesion with borderline malignancy since hypoxia and angiogenesis is known to be associated with the
malignant potential of precursor lesion of solid tumor. Photoacoustic mammography (PAM) is a novel optical imaging technology
that can visualize the hemoglobin distribution and its oxygen saturation (SO2) noninvasively. We have previously reported a
promising clinical result of a prototype model of PAM (Canon Inc., Tokyo, Japan) in breast cancer patients. However, the
improvement of spatial resolution and the identification of signal origin are still big challenges when considering its application for
clinical settings.
Materials and methods: We developed the second-generation model of PAM (PAM-02). This instrument has achieved the
improved spatial resolution (1.3mm) and enhanced detectability by carrying a high-sensitive detector. Moreover, it is equipped
with B-mode ultrasound, which enables us to identify the tumor location in PAM images more precisely. The distribution of
hemoglobin within breast tissue carrying solid tumor was evaluated by using PAM-02 under the approval of the ethics committee
in Kyoto University Hospital, Japan. Contralateral breast without tumor was also evaluated as a control if possible. Calculated
SO2 from photoacoustic (PA) signals were illustrated by using color scale.
Results: Seventeen breast lesions from 15 patients were analyzed including 4 ductal carcinoma in situ (DCIS), 12 invasive ductal
carcinoma (IDC) and one usual ductal hyperplasia. Tumor locations were successfully identified in 14 out of 17 lesions (82.3%) by
B-mode ultrasound imaging. The location of 3 lesions undetectable by B-mode ultrasound imaging could be identified by
comparing with corresponding MRI images. B-mode ultrasound imaging made it easy to distinguish intra-tumoral PA signals from
peri-tumoral PA signals. Intra-tumoral PA signals were detectable in 68.7% of malignant lesions (11 out of 16 lesions).
Peri-tumoral PA signals, which were suggested to be from feeding vessels, were detectable in 81.3% of malignant lesion (13 out
of 16 lesions). In the case of benign UDH, PA signals were not detected in either intra- or peri-tumoral region. Intra-tumoral SO2
was estimated to be lower than peri-tumoral SO2 in malignant lesion. While peri-tumoral PA signals were often described as
continuous vasculature, intra-tumoral PA signals often showed the spotty patterns. In addition, PA signal density was relatively
higher in DCIS compared with IDC. These findings was supposed to reflect the decreased hemoglobin perfusion within solid
structure of breast cancer. The minimum detectable lesion was DCIS with a diameter of 8mm.
Conclusion: Improved spatial resolution and combination with B-mode ultrasound imaging facilitate the region-specific evaluation
of PAM imaging. PAM-02 was supposed to be feasible for evaluating the hypoxic status within small breast tumor and its
microenvironment.

Title: Locoregional assessment by FDG PET/CT in stage IIIII breast cancer patients: A multivariate analysis
Suzana C Teixeira1, Bas B Koolen1, Wouter V Vogel1, Marcel P Stokkel1, Marie-Jeanne Vrancken-Peeters1, Vincent van der
Noort1, Emiel J Th Rutgers1 and Renato A Valds-Olmos1. 1Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital,
Amsterdam, Noord-Holland, Netherlands; 2Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam,
Noord-Holland, Netherlands; 3Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland,
Netherlands; 4Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands;
5
Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 6Netherlands Cancer
Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 7Netherlands Cancer Institute, Antoni van
Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands and 8Netherlands Cancer Institute, Antoni van Leeuwenhoek
Hospital, Amsterdam, Noord-Holland, Netherlands.
Body: Aim : Previously we demonstrated the additional value of PETCT in prone position for breast cancer detection. The
purpose of the present study was to evaluate, with a multivariate analysis, the factors influencing detection and quantification of
FDGavid primary tumors (PT) and regional lymph node metastases (LN) detected with PETCT in patients with stage IIIII breast
cancer.
Materials and methods : From August 2010 to April 2012 we included 198 patients (mean 51 yr, range 2682), with stage IIIII
breast cancer. A dose of 18FFDG between 180240 MBq, was administered intravenously. After 6010 minutes a PETCT of
the thorax in prone position was acquired, scanning 3.00 min per bed position and reconstructed in highresolution using 2x2x2
mm voxels and 2 mm CT slice thickness. Subsequently, a standard PETCT was performed in supine position from skull base to
thighs, scanning 1.30 min per bed position and using a standard reconstruction with 4x4x4 mm voxels, a standard 5mm CT slice
reconstruction and an extra 2 mm CT slice reconstruction. On all PETCT images we quantitatively assessed the SUVmax of
FDGavid PTs and LNs. We qualitatively assessed tumor multifocality, PT visibility, LN detection in defined regions and the
occurrence of anatomical mismatch between PET and CT. The obtained results were then evaluated with a multivariate analysis
for scanning position, patient age, breast size, tumor volume, number of FDGavid lesions, lymph node location and anatomical
mismatch between PET and CT.
Results:[/bold Prone position imaging positively influenced the visualization of tumor multifocality (p<0.001), the total
number of lymph nodes (p<0.001) and of axillary LNs (p<0.001). PT visibility was not significantly influenced by any of
the parameters.
A higher SUVmax of the primary tumor was found solely with increased tumor volume (p=0.001) or breast size (p<0.001).
The standard 5mm CT slice reconstruction of the supine PET/CT was the only factor causing an increase in anatomical
mismatch between PET and CT for axillary lymph nodes (p=0.004).
Conclusion: Prone position for PET/CT influences the visualization of primary tumor multifocality and the number of
FDG avid loco-regional lymph nodes. Tumor FDG-uptake appears to be most influenced by tumor volume and breast
size. Both results can be important for adequate staging and subsequent breast cancer treatment.

Title: The effect on shortterm progression free survival of the detection of 4 FDG-avid nodes or occult N3disease in breast
cancer patients with PETCT
Suzana C Teixeira1, Bas B Koolen1, Paula HM Elkhuizen1, Vincent van der Noort1, Marie-Jeanne Vrancken-Peeters1, Marcel P
Stokkel1, Emiel J Th Rutgers1 and Renato A Valds-Olmos1. 1Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital,
Amsterdam, Noord-Holland, Netherlands; 2Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam,
Noord-Holland, Netherlands; 3Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland,
Netherlands; 4Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands;
5
Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 6Netherlands Cancer
Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 7Netherlands Cancer Institute, Antoni van
Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands and 8Netherlands Cancer Institute, Antoni van Leeuwenhoek
Hospital, Amsterdam, Noord-Holland, Netherlands.
Body: Aim: The risk for locoregional recurrence (LRR), after neoadjuvant chemotherapy (NAC), is based on staging before NAC
as well as the final pathology after NAC. Especially the number of tumourpositive axillary nodes, which is also an important
selection factor for postoperative radiotherapy, is not adequately assessed by ultrasound before NAC or axillary lymph node
dissection after NAC. PETCT has a high positive predictive value for the detection of lymph node metastases. Newly found lymph
node metastasis on PETCT incentivized us to change the radiotherapy plan for patients with primary breast cancer scheduled for
NAC in our institute.
Koolen et al. reported an upstaging of 23% stage IIIII breast cancer patients to the radiotherapy requiring highrisk group (4
FDGavid axillary nodes or detection of occult N3disease) due to new lymph node metastasis detected with FDG PETCT
imaging. In this study, we report the effect of this upstaging with PETCT shortterm progression free survival (PFS).
Materials and methods:
Between 2007 and 2011 a total of 278 breast cancer patients (mean age 48.9y, range 1975y), with a tumour of at least 3 cm and
without metastases, received a baseline PET/CT for staging purposes and subsequent response monitoring to NAC. The group
was divided in three groups: a low (T2N0), intermediate (T02N1 and T3N0) and a highrisk group (T03N23, T3N1 and T4).
We classified LRR, distant metastases and death as an "event"; including all patients in the PFS analysis of the first 3 years.
Results:
With a median followup (FU) of 37 months and the upstaging as depicted in table 1
Upstaging of breast cancer risk-group after PET/CT
Total group before PET/CT
Changed After PET/CT
Complete after PET/CT (FU-events)
Low-risk: N=47
N=5
Low-risk (2) N= 42
Intermediate-risk: N=144
N=38
Intermediate-risk N= 106 (14)
High-risk: N=87
High-risk N= 130 (27)
Total: N=278
Total: N= 43
Total N= 278
Table 1: The group was divided in three groups: a low- (T2N0), intermediate- (T0-2N1 and T3N0) and a high-risk group
(T0-3N2-3, T3N1 and T4). The table shows the upstaging of breast cancer patients from low- and intermediate- to the high-risk
group, and thus, after PET/CT, requiring radiotherapy.
: The patients not upstaged by PETCT showed no difference in PFS between the highrisk, intermediaterisk and lowrisk
groups (Logrank p=0.18). Due to the migration of the 43 patients from the low and intermediate group to the highrisk group,
based on PETCT findings, the PFS differed significantly between the riskgroups (Logrank p=0.04). No difference in
locoregional recurrence was seen between the lowrisk and the highrisk group (P=0.18).
Conclusion:
After upstaging with PETCT, into the highrisk group requiring radiotherapy, a significant difference is seen between the three
riskgroups. PETCT restaging may more adequately predict progression free survival. The detection occult lymphatic metastasis
with PETCT leads to upstaging in clinically unsuspected patients with primary breast cancer, enabling adequate radiotherapy
treatment.

Title: Impact of perflubutane-enhanced ultrasonography for evaluating malignancy grade of breast cancer
Noriko Yoshimura1, Norio Masumoto1, Ai Amioka1, Keiko Kajitani1, Hideo Shigematsu1, Akiko Emi1, Takayuki Kadoya1, Tsuyoshi
Kataoka1, Rumi Haruta1 and Morihito Okada1. 1Hiroshima University, Horoshima, Japan.
Body: Objectives
This study aimed to determine whether or not signal intensity caused by the contrast effects of Contrast-enhanced
ultrasonography (CEUS) using perflubutane could predict malignancy grades of invasive breast cancer.
Methods
Fifty-four patients with clinical stages I III breast cancer between April 2013 and April 2014 underwent CEUS using
perflubutane. We analyzed the assciation between contrast-effect intensity and contrast time in CEUS and the prognostic factors
based on tumor size, nodal status and immunohistochemical markers (ER, HER-2 status, nuclear grade, Ki-67) in breast cancer.
Results
Time to washout of time required to reach plateau intensity from the start of the maximum intensity was significantly associated
with the Ki-67 value (p = 0.03). Also, A parameter of intratumoral blood perfusion, peak intensity (PI), was significantly associated
with the Ki-67 value (p = 0.006) and ER status (p = 0.002) (Table 1), but not with tumor size (cT; p = 0.25, pT; 0.96), node status
(p = 0.99), HER-2 status (p = 0.32) and nuclear grade (p = 0.61).
Table1. Relationship between changes in temporal contrast over time as a perfusion parameter and clinicopathological factors
Factor
n(%)
Peak intensity
Mean SD
p
0.50
Mean age SD (y)
53.8 13.5
< 50 y
30 (55.6)
65.5 25.4
50 y
24 (44.4)
61.1 21.5
1 or 2
19 (35.2)
65.8 28.0
35 (64.8)
62.4 21.2
< 30
24 (44.4)
53.8 22.4
30
30 (55.6)
71.4 21.9
Negative
13 (31.7)
58.2 20.9
Positive
41 (68.3)
80.5 24.6
Negative
47 (87.0)
62.3 24.0
Positive
7 (13.0)
71.9 20.5
NG
0.61
Ki67
0.006
ER
0.002
HER2
0.32
ER: estrogen receptor, HER2: human epidermal growth factor receptor 2, NG: nuclear grade.
Also, Thirty-eight, 7, and 9 patients had luminal, HER-2-positive and triple-negative tumors, respectively. The PI values for these
tumors were 56.8 20.9, 71.9 20.5, and 85.7 23.2, respectively. And, the PI value was significantly greater in the
triple-negative, than in luminal tumors (p = 0.001).
Table2. Relationship between signal intensity as a perfusion parameter and tumor subtypes
Tumor subtype
Subtype
Peak intensity
Mean SD
p*
Luminal
38
56.8 20.9
HER-2-positive
71.9 20.5
0.09
Triple-negative
85.7 23.2
0.001
Furthermore, PI significantly correlated with the Ki-67 value (Spearman r = 0.54, P = 0.00002).
Conclusions
These findings indicated that PI has excellent predictive value for grade malignancy in breast cancer and might help to determine
appropriate therapeutic strategies.
Key points
Contrast-enhanced ultrasonography (CEUS) enables the real-time evaluation of detailed hemodynamics in breast cancer.
Peak intensity (PI) was significantly associated with Estrogen Receptors and Ki-67 assessed by immunohistochemistry.
PI significantly correlated with the Ki-67 value, indicating that PI reflects the grade of proliferative activity in tumors.
Analyses of contrast-effect intensity will be applied to evaluate grades of malignancy and determine treatment strategies.

Title: An analysis of lesions diagnosed on screening breast ultrasound in women with dense breasts: Four years of data
Jean M Weigert1. 1Hospitals of Central Connecticut, New Britain, CT and 2Mandell and Blau MD's PC, New Britain, CT.
Body: Objective: To analyze the type of cancers and high risk lesions diagnosed in women with normal mammograms with dense
breasts with the addition of bilateral breast ultrasound and determine whether these lesions make an impact on clinical outcomes.
Methods: Four years of ultrasound data from two sites with five offices in Connecticut was analyzed. The type of lesion including
size, nuclear and histologic grade, ER/PR/Her2 status, node status, patient age and risk factors was reviewed.
Results: A total of 532 Ultrasounds with Birads 4 or 5 were reported with 46 cancers or high risk lesions. There were 14 Invasive
Ductal Carcinoma, 10 Invasive Lobular Carcinoma, 8 Mixed type, 1 Mucinous, 1 Tubular, 6 Ductal carcinoma in situ, 3 Atypical
Ductal Hyperplasia with papilloma and 3 Lobular Carcinoma in situ. Of the invasive cancers and DCIS, 9 were nuclear grade 1,
25 nuclear grade 2 and 7 nuclear grade 3. They ranged in size from .3 to 8 .0 cm and the patient age was 45-77 years. Four
patients had positive metastatic lymph nodes. 3 were in tumors that were nuclear grade 3 and one with micro metastasis was in a
nuclear grade 2 tumor. 29 cancers had known hormonal status of which 27 were ER/PR+ and 2 were ER+/ PR-. 5 patents had a
family history of breast cancer and 2 patients had prior history of their own breast cancer, one had uterine cancer. In the third and
fourth year of screening usg, 4 cancers were diagnosed in patients who had a negative usg either one year or two years prior.
They ranged in size from .4 to 1.2 cm and were all node negative.
LESION Characterization
Lesion Type
Number Average Size Sentinel Node Family History Prior Self Cancer Prior Screening Usg
Invasive Ductal Grade 1
1.2 cm
1.7 cm
1 micro-met
1.4 cm
3 macromet
Invasive Lobular Grade 1
0.7 cm
Invasive Lovular Grade 2
1.5 cm
Mixed IDC/ILC Grade 1
n/a
1.2 cm
1.3 cm
DCIS Grade 2
1.3 cm
DCIS Grade 3
1.9 cm
LCIS
n/a
ADH /Pappiloma
0.5 cm
Tubular
0.4 cm
Mucinous
8.0 cm
Discussion: In our first four years of tracking patients with dense breasts, the addition of bilateral breast ultrasound diagnosed a
variety of cancers and high risk lesions. These were not visible on screening mammography and were not palpable. There were 4
with node positive disease diagnosed in high nuclear grade tumors. Most patients had no additional risk factors other than dense
breasts. In the third and fourth year, cancers were found in patients who had a prior screening usg. Their mammograms remained
negative. The cancers were small and node negative.
Conclusion: The successful treatment of breast cancer is based on several variables including type of cancer, grade, size at
diagnosis, and stage at diagnosis. We do not know at what point these lesions diagnosed with screening breast usg would have
been clinically evident, either on mammography or physical examination. One might consider that the treatment outcome would
be altered if the lesions were larger and with more positive lymph nodes.

Title: Retrospective comparison of sensitivity and positive predictive value (PPV) of contrast enhanced spectral mammography
(CESM) to contrast enhanced breast MRI (BMRI) in 50 malignant breasts
Luna Li1, Lydia Liao1, Pauline Germaine1, Elizabeth Tinney1, Kristin Brill2 and Karen Hendershott2. 1Cooper University Hospital
Women's Imaging, Voorhees, NJ.
Body: Objective: Contrast enhanced spectral mammography (CESM) is a new study using contrast for early breast cancer
detection. It utilizes dual energy digital mammographic technology to detect contrast enhanced cancer that may be invisible on
conventional mammogram. Limited studies have shown that adding CESM to diagnostic workup adjunct with mammogram and
breast ultrasound does increase sensitivity for breast cancer detection. More studies are needed to compare the sensitivity and
positive predictive value (PPV) of CESM to BMRI to further define the role of CESM in breast cancer diagnosis.
Methods: This study involved 50 malignant breasts in 48 women retrospectively chosen from of 960 patients in our institution
during the period of October 2012 to March 2014. Both CESM and BMRI were done for each patient within 30 days. The cancer
diagnoses were confirmed by tissue diagnoses. The number of malignant lesions was quantified in each breast. The size of
lesions was classified into three categories based on standard of breast cancer stages: 1 (0.2cm -<= 2cm), 2 (2cm<lesion<=5cm),
3(>5cm). The enhancement intensity of both lesions and background has been quantified based on a scale of 0-3. Statistical
significance was analyzed using T test for mean size of index cancer and mean score of enhancement intensity of background
and lesions on CESM and BMRI. Sensitivity and PPV were calculated for both CESM and BMRI. Morphology consistence was
calculated on both studies.
Results: Both CESM and BMRI have sensitivity of 100% for breast cancer detection. CESM has a PPV of 98% versus 93% for
BMRI. No statistical significance was identified on mean size of index cancer (3.7cm for CESM and 3.8cm for BMRI). The
enhancement intensity of background and lesions is significantly higher on CESM than on BMRI (p<0.01). The smallest cancer
can be detected by both CESM and BMRI is less than 0.5 cm. Morphology consistence was 46/50 (92%). Of the 50 breasts, there
was one false-positive finding at CESM mammography, and four false-positive findings at MR imaging.
Conclusions: Our study indicates that CESM and BMRI have comparable high sensitivity on breast cancer detection. CESM has a
higher PPV than BMRI that may indicate a better specificity (no significant difference due to the small sample size). Significantly
less background enhancement intensity on CESM than on BMRI reflect an increased specificity. More studies need to be
conducted for further evaluation.

Title: Contrast-enhanced ultrasound for evaluation of therapeutic efficacy of radiofrequency ablation for primary breast cancer
Toshikazu Ito1, Jyunji Okayama1, Kumiko Uji1 and Masaaki Izukura1. 1Rinku General Medical Center, Osaka, Izumisano, Japan;
2
Rinku General Medical Center, Osaka, Izumisano, Japan; 3Rinku General Medical Center, Osaka, Izumisano, Japan and 4Rinku
General Medical Center, Osaka, Izumisano, Japan.
Body:
To evaluate therapeutic efficacy of percutaneous ultrasound (US) guided radiofrequency (RF) ablation therapy for primary breast
cancer using contrast-enhanced US.
The study was approved by the institutional ethics committee, and written informed consent was obtained. Between January 2012
and April 2014, 25 patients with biopsy-confirmed T1 breast cancer underwent RF ablation therapy in one institution. We
examined 25 patients with 25 T1 breast cancers by contrast-enhanced MRI/CT, and contrast-enhanced US before and after RF
ablation therapy. US guided RF abltion was performed using a 17-gauge internally cooled electrode (Cool-TipTM, Valleylab,
Boulder, CO, USA) under general anesthesia. After injection 0.015mL/kg body weight of Sonazoid(perflubutane),
contrast-enhanced US was performed at a low mechanical index (0.17-0.24). Therapeutic success was defined as a lack of
contrast enhancement by contrast-enhanced MRI, US and non-viable cancer tissues by US guided vacuum assisted biopsy
(VAB) or core needle biopsy (CNB).
Contrast-enhanced US and MRI four weeks or more after RF ablation therapy showed ablation zones and adequate tumor
necrosis in all 25 cancer lesions treated. Contrast-enhanced US made it possible to see that tumor vessels of treated lesions had
disappeared after treatment. Contrast enhanced MRI showed a lack of stain with ring-like enhancement of ablation zone and
non-viable tissues were seen by VAB and/or CNB in 25 ablation zones.
Contrast-enhanced US using Sonazoid was useful modalities for evaluating the efficacy of RF ablation therapy in breast cancer
lesions.

Title: 18F-fluoro-2-deoxy-glucose positron emission tomography/ computed tomography evaluation shows the utility for predicting
postoperative outcomes on Luminal types breast cancer patients
Kenjiro Aogi1, Yoshifumi Sugawara1, Sachiko Kiyoto1, Takayuki Kadoya2, Hideo Shigematsu2, Norio Masumoto2 and Morihito
Okada2. 1NHO Shikoku Cancer Center, Matsuyama, Ehime, Japan and 2Hiroshima University, Hiroshima, Japan.
Body: Background: Hormonal receptor-positive, i.e. Luminal types, breast cancer has better prognosis compared with other
phenotypes, but predictive factors for postoperative outcomes have been in research besides hormonal receptor expressions.
18F-fluoro-2-deoxy-glucose positron emission tomography/ computed tomography (18F-FDG PET/CT) using maximum
standardized uptake value (maxSUV) has been reported its utility as one of predictive markers for relapse-free survival (RFS) on
breast cancer patients. In this retrospective study, the utility of 18F-FDG PET/CT was evaluated on predicting postoperative
outcomes especially on Luminal types breast cancer patients. Methods: Three hundred forty four breast cancer patients with
clinical Stage I-III who underwent preoperative 18F-FDG PET/CT were enrolled between January 2006 and December 2011 in two
institutes. Patients classified into Luminal types, Luminal A(166 patients): Estrogen receptor(ER)(+) or Progesterone receptor
(PR)(+), human epidermal growth factor receptor type2 (HER2) (-), nuclear grade(NG)1-2 or Luminal B(96 patients): ER(+) or
PR(+), HER2(-), NG3, were evaluated in this study. Moreover, patients were also divided in two groups according to cut-off
maxSUV on receiver operating characteristic (ROC) analysis (4.0 group vs >4.0 group, AUC=0.830). Statistical analyses were
performed using Student t test and log-rank test, and p values of less than 0.05 were considered to be statistically significant.
Results: Significant correlations were shown between maxSUV and clinical T-factor, nuclear grade, vascular invasion and
presence or absence of recurrence (p<0.0001, p<0.0001, p=0.0181, and p=0.0056, respectively). In uni- and multivariate
analyses using Cox model for recurrence, only maxSUV was shown to be significant (p=0.021 and p=0.043, respectively). In
RFS, the maxSUV4.0 group showed significantly better prognosis compared to that of the maxSUV >4.0 group (p=0.003).
Conclusion: 18F-FDG PET/CT using maxSUV was concluded to be useful for predicting postoperative prognosis on luminal
types breast cancer patients.

Title: Clinical studies of palpation imaging of the breast on over 1000 patients
Cary S Kaufman1, Jae S Son2, Eli Yered2 and Armen Sarvazyan3. 1Bellingham Regional Breast Center, Bellingham, WA; 2Medical
Tactile, Inc, Los Angeles, CA and 3ARTANN Laboratories, Inc, West Trenton, NJ.
Body: INTRODUCTION
Palpation Imaging (PI) of the breast is based on the principles of tactile imaging, a method for measuring the elasticity lesions
using a tactile sensor array. PI provides objective data beyond subjective manual palpation by quantifying the shape and
hardness of a lesion as well as creating a permanent digital image. PI is also known as "Tactile Imaging", "Mechanical Imaging",
"Stress Imaging" or "Computerized Palpation". It quantitatively evaluates mechanical and structural properties of a breast lesion
that is altered by tumor growth, and provides a quantitative characterization of the detected lesion. PI is a radiation-free and an
inexpensive in comparison to other imaging devices. The purpose of this paper is to evaluate the clinical efficacy of PI by
summarizing all clinical studies conducted to date with this technology.
METHODS
The two step process to identify the clinical presence of a breast cancer include identifying the presence of an abnormality and
secondarily to determine if that abnormality is malignant. Not all studies examined each portion of this two-step process. There
have been 9 clinical studies involving 1,155 symptomatic patients using PI in the U.S., China, and UK. To examine the ability of PI
to detect an abnormal breast mass, we reviewed the sensitivity comparisons with CBE. To determine whether a breast lesion was
malignant, we reviewed the analysis of the sensitivity, specificity and accuracy for the standard imaging modalities;
mammography, breast ultrasound, clinical breast examinations (CBE) and MRI using pathology as the gold standard.
Calcifications were not included because they represent DCIS which does not present clinically and is the subject of over
diagnosis.
RESULTS
All studies included in this initial meta-analysis were on symptomatic patients.
To identify the presence of an abnormality, the sensitivity of PI to detect breast masses was 89% in comparison with CBE which
was 83%.
To identify the presence of breast cancer,
The sensitivity of PI in detecting breast cancer was 85%, in comparison to CBE (59%), mammography (76%) and ultrasound
(90%).
The specificity of PI in detecting breast cancer was 79% in comparison to CBE (78%), mammography (71%) and ultrasound
(79%).
The Accuracy of PI in detecting breast cancer was 82% comparison with CBE (67%), mammography (74%) and ultrasound
(84%).
Breast Cancer Detection Comparision
Palpation Imaging
CBE
Mammogram
Ultrasound
Sensitivity for Cancer Detection
85
59
76
90
Specificity for Cancer Detection
79
78
71
79
Accuracy for Cancer Detection
82
67
74
84
CONCLUSIONS
A summary of clinical studies conducted on symptomatic patients to date suggest that PI is (a) sensitive in the detection of breast
masses and b) identifies breast cancer with similar sensitivity and specificity compared with other standard imaging modalities
while being cost effective. PI may fill a gap in developing countries where little breast screening infrastructure exists. Ongoing
clinical research with PI is warranted to evaluate its potential in the early diagnosis and screening for breast cancer in the U.S.

Title: Contrast enhanced spectral mammography (CESM) is a new breast imaging contrast study: A review of indications,
protocol, interpretation and pitfalls
Lydia Liao1, Luna Li1, Pauline Germaine1, Elizabeth Tinney1, Kristin Brill2 and Karen Hendershott2. 1Cooper University Hospital
Women's Imaging, Voorhees, NJ.
Body: Purpose: This presentation is to discuss the indications, protocol, interpretation and pitfalls of CESM using illustrations of
breast malignant and benign cases performed in our institution, after a brief literature review. CESM characteristics for both
benign and malignant lesions are defined after comparison to other imaging studies and correlation to final tissue diagnosis.
Methods: 1000 CESM cases including 100 breast cancer and 200 benign lesions were retrospectively analyzed. These cases
were performed in our institution from October 2012 to March 2014. GE Senographe Essential Full Field Digital System
(SenoBright) was utilized for CESM studies. Both low and high energy images were obtained after intravenous 75-100cc Isovue
370 administration at flow rate of 1.5-2cc per second. Both low energy and subtracted images were reviewed on GE Centricity
Imagecast on 5 megapigxel monitor. Abnormal CESM lesions underwent further evaluation and tissue sampling for a final
diagnosis to determine the presence of breast cancer.
Results: CESM is one of the contrast studies. It provides both morphologic and functional information of breast lesions. The
sensitivity of CESM for cancer detection was 93-100% per patient in contrast to 95-100% of breast MRI. There are morphological
and enhancement features for both malignant and benign lesions on CESM. CESM could pick up lesions when regular digital
mammography and ultrasound are negative. CESM is a part of diagnostic study on selected patients. It has a short duration exam
time and results are easily accessible. Some CESM artifacts are unique and commonly occurred.
Conclusion: CESM is a new breast imaging contrast study. The sensitivity for breast cancer detection is comparable to breast
MRI. With its easy accessibility and cost effectiveness, CESM may play an important role in breast cancer detection in conjunct
with diagnostic mammography and breast ultrasound. Recognition of CESM artifacts helps to increase diagnostic accuracy.

Title: Role of FDG-PET/CT in prediction of underestimation of invasive breast cancer in cases of ductal carcinoma in situ
diagnosed at needle biopsy
Hideo Shigematsu1, Takayuki Kadoya1, Noriko Yoshimura1, Keiko Kajitani1, Akiko Emi1, Noriko Masumoto1, Rumi Haruta2,
Tsuyoshi Kataoka2 and Morihito Okada1. 1Research Institute for Radiation Biology and Medicine, Hiroshima University, Hiroshima,
Japan and 2Hiroshima University, Hiroshima, Japan.
Body: Background: The aim of this study was to evaluate the significance of 18F-fluorodeoxyglucose positron emission
tomography/computed tomography (FDG-PET/CT) for predicting the underestimation of invasive breast cancer in cases of ductal
carcinoma in situ (DCIS) at needle biopsy.
Patients and Methods: Of 83 consecutive cases with diagnoses of DCIS at primary needle biopsy who underwent curative
surgery between 2010 and 2013, the association between the maximum standardized uptake value (SUVmax) on FDG-PET/CT
before excision and the underestimation of invasive breast cancer was examined.
Results: There were 29 (34.9%) cases diagnosed to have invasive breast cancer at excision. Receiver operating characteristics
(ROC) curve analysis showed the cutoff value of SUVmax to predict underestimation of invasive breast cancer was 1.6. The rates
of underestimation were 61.5% for patients with a tumor of SUVmax > 1.6 and 11.4% for patients with a tumor SUV max 1.6 (p
<0.001). High value of SUVmax was significantly associated with symptomatic presentation (p < 0.001), palpability (p < 0.001),
mass formation (p = 0.013), high Breast Imaging Reporting and Data System category (p = 0.011) and core needle biopsy (p =
0.007). In multivariate analysis, high SUVmax was only significant predictive factor of underestimation of invasive breast cancer
(HR 11.7, 95% CI 3.70-37.0, p < 0.001).
Association between each preoperative variable and underestimation of invasive component in multivariate analysis in cases of
DCIS diagnosed at needle biopsy
Variables
HR (95% CI)
P value
Presentation
symptomatic vs. screening detected
1.64 (0.43-6.22)
0.47
Palpable
yes vs. no
1.54 (0.36-6.69)
0.58
Lesion size
< 2cm vs. 2cm
1.23 (0.34-4.40)
0.76
Mass formation
yes vs. no
1.28 (0.36-4.44)
0.7
BI RADS category
5 vs. 3, 4
2.92 (0.80-10.7)
0.11
Biopsy device
mammotome biopsy vs. CNB
2.16 (0..64-7.30)
0.21
Nuclear grade
high vs. not high
high vs. not high 2.24 (0.51-9.78)
0.29
SUVmax
higher vs. lower
11.7 (3.70-37.0)
< 0.001
BI RADS, Breast Imaging Reporting and Data System; DCIS, ductal carcinoma in situ; IBC, invasive breast cancer; SUVmax,
maximum standardized uptake value
Conclusion: SUVmax on FDG-PET/CT is useful for predicting the underestimation of invasive breast cancer in cases of DCIS at
needle biopsy.

Title: High-resolution specimen-positron emission mammography (s-PEM) indicates the spread of cancer in breast-conserving
surgery
Gou Watanabe1, Masatoshi Itoh3, Dan Kyokutou3, Mika Watanabe2, Takanori Ishida1, Akihiko Suzuki1, Minoru Miyashita1, Hiroshi
Tada1, Naoko Mori4 and Noriaki Ohuchi1. 1Tohoku University, Sendai, Japan; 2Tohoku University, Sendai, Japan; 3Sendai Medical
Imaging Center, Sendai, Japan and 4Department of Diagnositic Radiology, Sendai, Japan.
Body: Background: Breast cancer surgery, including breast-conserving surgery or sentinel lymphnode biopsy, has become
minimally invasive. Recent consensus guidelines on the margins for breast-conserving surgery have defined a negative margin as
one with no ink on tumor, regardless of the distance from the tumor, whereas a positive margin has ink on the tumor. Therefore,
the extent of required tumor resection will decrease for cosmoses and expand the indications for breast-conserving surgery.
Therefore, there is a real need for a more accurate imaging of tumor spread. In particular, a functional imaging of breast cancer
specimens will help determine negative margins in breast-conserving surgery. We collaborated with Sendai Medical Imaging
Center (SMIC) who have high-resolution positron emission mammography (PEM) with the novel scintillator Pr3+-doped
transparent ceramic Lu3Al5O12 (Pr:LuAG) for the imaging of breast cancer specimen with 18F-fluodeoxyglucose (FDG).
Methods: With the approval of the hospital ethics committee, positron emission tomography (PET) was conducted preoperatively
on the day of surgery with 18F-FDG. After the PET results were explained to the patients, breast-conserving surgery was
performed in 11 patients from February to July 2014. In the operating room, medical staff exposures were calculated with a
portable dosimeter. All of the cases were carried out intraoperative frozen sections that were taken from all sides of the outside of
the specimens. The specimens were serially sectioned in 5-mm slices for permanent histology. After the specimens were
removed, they were sent to the SMIC and high-resolution specimen-PEM (s-PEM) was conducted. We evaluated the detection
rate of s-PEM in in situ or invasive lesions (extent of locations were tolerated up to a 10% error), the predictive value of margin
status between s-PEM and frozen section analysis and medical stuff exposure.
Results: In 11 breast specimens, eight invasive lesions and 15 in situ lesions (excluding the foci that exist in less than three
carcinoma ducts) were confirmed by permanent histology. s-PEM detected all of the invasive lesions and 14 out of 15 in situ
lesions. Three breast specimens had positive margins in permanent histology. Two out of three cases with positive margin
accumulated 18F-FDG at the edge of the specimens, which were considered positive margins with s-PEM, whereas positive
margins were only one out of three of the frozen section analysis. In contrast, a false positive margin was found in only one case
with s-PEM, which confirmed the preoperative core-needle biopsy that was diagnosed as an intraductal proliferative lesion.
Medical staff exposure was determined for the operator, first assistant, second assistant, anesthesiologist, scrab nurse and
circulating nurse as 31, 34, 25, 11, 21 and 6 SV, respectively.
Conclusion: s-PEM detected not only invasive lesions but also in situ lesions with great accuracy. It might accelerate the
decrease in the extent of resection required in breast-conserving surgery. However, further studies are needed to elucidate the
characteristics of the false positive cases in s-PEM.

Title: Cardiac function in BRCA1/2 mutation carriers with a history of breast cancer (BC) treated with anthracyclines (anthra)
Filipa Lynce1, Ana Barac2, Karen L Smith4, Mihriye Mete5, Lynette Wray5, Madeline Nardacci3, Pia Herbolsheimer1, Raquel
Nunes1, Sandra M Swain1, Robert Warren3 and Claudine Isaacs3. 1Washington Cancer Institute, MedStar Washington Hospital
Center, Washington, DC; 2MedStar Heart Institute, MedStar Washington Hospital Center, Washington, DC; 3Lombardi
Comprehensive Cancer Center, MedStar Georgetown University Hospital, Washington, DC; 4Johns Hopkins Sidney Kimmel
Cancer Center, Washington, DC and 5MedStar Health Research Institute.
Body: Introduction: Animal data suggest that cardiac-specific loss of BRCA1/2 genes leads to increased cardiac dysfunction in
the setting of stressors such as ischemia or anthra chemotherapy (CT). Women with BRCA1/2 mutations who develop BC may
receive anthra-based CT. It is not known whether BRCA1/2 mutation carriers treated for early stage BC with anthra-based CT are
at higher risk for anthra-induced cardiac toxicity than similarly treated women with sporadic BC. This study investigates left
ventricular ejection fraction (LVEF) and myocardial global longitudinal strain (GLS), a novel echocardiographic measure of cardiac
contractility, in women with and without BRCA1/2 mutations who were treated with anthra-based CT.
Methods: We identified 41 women with a history of BRCA1/2 mutation-associated BC and 52 women with a history of sporadic
BC, all of whom received anthra-based adjuvant or neoadjuvant CT. We excluded women who received HER2 therapy or who
had stage IV disease. BC CT was completed at least 6 months prior to study enrollment. All participants completed a cardiac
questionnaire and underwent a comprehensive 2D and 3D echocardiographic exam with assessment of myocardial GLS using
speckle-tracking software (QLAB10, Philips, Andover, MA). Abnormal LVEF was defined as less than 55% (50-54.9% borderline
reduced and 40-49% mildly reduced) and abnormal GLS was defined as absolute value of less than 18.9%. Patients with history
of hypertension (HTN) were excluded from this analysis to avoid confounding effects of HTN on GLS.
Results: We present data on 57 normotensive participants, 34 BRCA1/2 mutation-carriers and 23 non-mutation carriers. Mean
age was 49.5 +/- 8.7 years in the BRCA mutation-carrier group and 51.1 +/- 8.7 years in the non-mutation carrier group
(p=0.496). Women with BRCA1/2 mutations were more likely to be Caucasian (p=0.026) and have undergone oophorectomy
(p<0.001). The average time from anthra treatment to study enrollment was 5 +/- 4.4 years. Groups were well balanced with
regards to age, diabetes, hyperlipidemia, history of coronary disease, smoking and cumulative dose of anthra. Four BRCA1/2
mutation-carriers (12%) and 4 non-mutation carriers (17%) had reduced GLS (p= 0.450). Mean LVEF was similar between the
groups: LVEF was borderline reduced in one BRCA1/2 mutation-carrier (3%) and borderline or mildly reduced in 4 non-mutation
carriers (18%)(p= 0.091).
Conclusions: In this population, reduced LVEF and myocardial GLS were present in a small percentage of women treated with
anthra-based CT for early stage BC. However, reduction in LVEF and myocardial GLS did not differ between the BRCA1/2
mutation carriers and the women with a history of sporadic BC. Larger studies are needed to definitively rule out a difference in
cardiac toxicity by BRCA mutation status in BC survivors treated with anthra. This project is supported by Fisher Center for
Familial Cancer Research Award at Lombardi Comprehensive Cancer Center. AB is supported by Georgetown-Howard
Universities Center for Clinical & Translational Science (GHUCCTS) post-doctoral KL2 Award (5KL2TR000102-04).

Title: CYP2D6 intermediate metabolizers includes patient groups with distinct metabolic activity
Daniel L Hertz1, Anna C Snavely2, Howard L McLeod3, Christine M Walko3, Joseph G Ibrahim2, Steven Anderson4, Karen E
Weck2, Peter Rubin5, Oludamilola Olajide6, Susan Moore6, Rachel Raab7, Daniel R Carrizosa8, Steven Corso9, Gary Schwartz8,
Jeffrey M Peppercorn10, James P Evans2, Zeruesenay Desta13, David A Flockhart13, Lisa A Carey2 and William J Irvin, Jr14.
1
University of Michigan, Ann Arbor, MI; 2University of North Carolina, Chapel Hill, NC; 3Moffitt Cancer Center; 4Laboratory
Corporation of America; 5Moses Cone Regional Cancer Center; 6Rex Hematology Oncology Associates; 7Brody School of
Medicine, East Carolina University; 8Carolinas Medical Center Hematology-Oncology Associates; 9Palmetto Hematology
Oncology; 10Duke Cancer Institute; 11Waverly Hematology Oncology; 12Carolina Breast Care Specialists; 13Indiana University,
Indianapolis, IN and 14Bon Secours Cancer Institute, Midlothian, VA.
Body: Background: Tamoxifen is a selective estrogen receptor modulator that is the most commonly used and cost effective
hormonal agent for pre-menopausal hormone-receptor positive breast cancer patients. CYP2D6 activity phenotype, which is
classified by genotype, predicts the extent of metabolic activation of tamoxifen to endoxifen. We previously reported that
increasing the daily dose to 40 mg/day in intermediate metabolizers (IMs), but not poor metabolizers (PMs), achieves target
endoxifen concentrations, defined as that of extensive metabolizers (EMs) on 20 mg/day. There was substantial endoxifen
variability in the IM phenotype group, which is composed of several discrete diplophenotypes (EM/IM, EM/PM, IM/IM, IM/PM). We
enrolled a second, larger cohort of patients in order to determine whether these diplophenotypes should be combined into a
single IM phenotype or segregated.
Methods: 380 patients on tamoxifen 4 months and not on potent CYP2D6 inhibiting medications enrolled in Lineberger
Comprehensive Cancer Center (LCCC) trial 0801. Genotyping was performed using the Amplichip CYP450 test (Roche
Diagnostics) for CYP2D6, followed by systematic assignment of phenotype based on diplophenotype. Tamoxifen was increased
from 20 to 40 mg/day in PMs and IMs. Endoxifen concentrations in IM diplophenotypes were compared with EM/EMs and
PM/PMs at baseline and at 4 months (after dose increase in patients with IM and PM phenotypes).
Results: After exclusion of UM patients and patients missing endoxifen data at baseline and/or 4 months, 295 patients were
included in this analysis. At baseline the EM/IM patients had similar endoxifen level to the EM/EM patients while the IM/IM and
IM/PM patients had similar levels to the PM/PMs. After 4 months on 40 mg/day the endoxifen concentrations in EM/IM patients
were significantly greater than EM/EMs; EM/PM and IM/IM patients were similar to EM/EMs; but IM/PM patients remained
significantly lower than EM/EMs and similar to PM/PMs (See Table 1 for results).
Conclusions: The large group of patients currently defined as CYP2D6 intermediate metabolizers is comprised of four distinct
CYP2D6 diplophentoypes. The most metabolically active diplophenotype (EM/IM) are very similar to EM/EMs while the least
active diplophenotype (IM/PM) are similar to PM/PMs. A more accurate CYP2D6 activity classification system may be necessary
if genetic association testing and genotype-guided therapy are pursued.
Endoxifen Level at Baseline and 4 Months by CYP2D6 Diplophenotype
Baseline
Endoxifen
Diplophenotype n
4-Month
Endoxifen
P-val vs.
EM/EM
Median (SD)
P-val vs.
PM/PM
P-val vs.
EM/EM
Median (SD)
P-val vs.
PM/PM
EM/EM
103 8.67 (6.01)
NA
p=0.0001
8.23 (5.09)
NA
p=0.007
EM/IM
56 8.02 (4.75)
p=0.09
p=0.002
13.11 (9.38)
p<0.0001
p<0.0001
EM/PM
74 5.72 (4.45)
p=0.0001
p=0.02
8.91 (5.28)
p=0.42
p=0.003
IM/IM
17 4.29 (4.10)
p=0.001
p=0.26
6.52 (5.53)
p=0.27
p=0.24
IM/PM
32 3.90 (3.17)
p<0.0001
p=0.48
5.82 (3.47)
p=0.0009
p=0.77
PM/PM
13 3.33 (2.89)
p=0.0001
NA
6.08 (2.57)
p=0.007
NA
Diplophenotype classified as extensive metabolizer phenotype, continued on 20 mg/day. Diplophenotypes classified as

intermediate metabolizer phenotype, changed to 40 mg/day. Diplophenotype classified as poor metabolizer phenotype, changed
to 40 mg/day.

Title: Experience in the community oncology practice with a 25-gene hereditary cancer panel
Sami Diab1, Patricia Rodriguez2, Anna Leininger3, Lisa Clark4, Mike F Janicek5, Ellen B Smith6, John Sandbach6, Jennifer Saam7
and Lucy Langer4. 1Rocky Mountain Cancer Centers and Colorado Integrative Care, Aurora, CO; 2Virginia Cancer Specialists,
Arlington, VA; 3Minnesota Oncology, Woodbury, MN; 4Compass Oncology, Portland, OR; 5Arizona Oncology, Scottsdale, AZ;
6
Texas Oncology, Austin, TX and 7Myriad Genetic Laboratories, Inc, Salt Lake City, UT.
Body: Introduction:
The identification of patients with an inherited cancer syndrome is becoming increasingly relevant in the treatment and prevention
of cancer. Testing with a panel of genes provides the opportunity to rapidly identify or rule out deleterious mutations in several
genes simultaneously and, thus, has the potential to streamline the testing process and more efficiently and accurately provide
results. We report on the experience of panel-based testing in the community oncology setting.
Objectives:
To evaluate the performance of a 25-gene hereditary cancer test in the community oncology setting and describe the patient
characteristics and test findings.
Methods:
We retrospectively evaluated the 25-gene panel testing data obtained between September 2013 and February 2014 in 6 large
community oncology practices. The gene panel was based on next generation sequencing and rearrangement analysis of 25
genes with cancer risk data: BRCA1, BRCA2, MLH1, MSH2, MSH6, PMS2, EPCAM, APC, MUTYH, CDKN2A, CDK4, PALB2,
CHEK2, SMAD4, BMPR1A, STK11, TP53, CDH1, PTEN, ATM, NBN, BARD1, BRIP1, RAD51C, and RAD51D. Personal and
family history was obtained by health care provider report on test requisition forms.
Results:
The panel test was performed on 359 individuals during the time period. 69.9% of these patients had a phx of at least one of the
eight panel cancers: breast, ovarian, colorectal, endometrium, pancreas, melanoma, prostate, and/or stomach. cancer diagnosis.
97.8% of the patients tested met the 2013 NCCN guideline criteria for HBOC, 2012 NCCN guideline criteria for Lynch Syndrome,
or both. 37 pathogenic mutations were found in 34 distinct patients. The variant rate in this population was 35.4%. Mutations were
identified in 16 different genes, and only 20 of the 37 mutations (54.1%) were found to be in the 6 genes included in HBOC and
Lynch Syndrome testing (BRCA1, BRCA2, MSH6, MLH1, MSH2, PMS2).
Conclusion:
The 25-gene hereditary cancer panel increased the identification of deleterious mutations which would otherwise not have been
detected. The variant rate in this setting is similar to the rate reported previously in both community and academic centers where
panel testing is used. Use of a hereditary cancer panel in the community oncology practice may improve detection rates and
provide an opportunity for enhanced cancer management.

Title: The impact of Angelina Jolie (AJ)'s story on genetic referral and testing at an academic cancer centre
Jacques Raphael1, Sunil Verma1, Paul Hewitt1 and Andrea Eisen1. 1Sunnybrook Odette Cancer Centre, Toronto, ON, Canada.
Body: Background
In May 2013, AJ revealed to the media that she had undergone preventive double mastectomy. The actress had a family history
of breast and ovarian cancer and tested positive for the BRCA1 gene mutation. Media coverage has been extensive, but it's not
clear what messages the public and professional medical staff took from this personal story.
Methods
We conducted a retrospective review using data from the clinical database of the Familial Cancer Program in a tertiary care
cancer centre. The impact of AJ's story on genetic counseling referrals was assessed by comparing the number of referrals made
before and after the story. In addition, the appropriateness of referrals was reported by comparing the number of patients who
qualified for genetic testing as defined by the Ontario Ministry of Health and Long Term Care and the ones who were found to
carry a BRCA1/2 mutation before and after the media release. We previously reported the impact of AJ's story, 6 months before
and after, and showed that all the numbers practically doubled. This is an update of this assessment 1 year before and after the
story.
Results
The increase in the number of women referred for genetic counseling was persistent after one year with an increase by 88%
compared to 90% originally: 1042 referrals before versus 1962 referrals after. There was still an increase in the number of women
who qualified for a genetic testing, but it was was less than previously reported. Originally there was an increase of 105% after 6
months. After 12 months, the increase is 68%: 493 women qualified for the testing before versus 827 women after. Furthermore
the preliminary increase in the number of BRCA1/2 carriers identified is also less pronounced. The preliminary increase was by
110%. After 12 months, it is 42%: 80 BRCA1/2 carriers identified (45 BRCA1, 35 BRCA2) before versus 114 after (63 BRCA1, 51
BRCA2).
Conclusion
In our preliminary report we showed that after AJ's story, the number of referrals doubled and the quality of referrals remained the
same. In this updated analysis, the number of referrals after the story remained as high as previously, however the quality of
referrals did not remain the same. With the longer period of observation, fewer referred women qualified for genetic testing and
fewer BRCA1/2 carriers were identified. If the positive "celebrity effect" reported previously will begin to subside, our next
challenge will not only be to meet the increased demand for cancer genetic services but also to ensure that referrals to our clinic
are appropriate by working to reduce the referrals of low risk women triggered by media attention.

Title: Genetic variant in the OPG gene is associated with aromatase inhibitor-related musculoskeletal toxicity in breast cancer
patients
Anneleen Lintermans1, Kathleen Van Asten1, Annouschka Laenen1, Thomas Van Brussel1,2, Johan Verhaeghe1,3, Dirk
Vanderschueren1,3, Diether Lambrechts1,2 and Patrick Neven1,3. 1KU Leuven, Leuven, Belgium; 2VIB and 3University Hospitals,
Leuven, Belgium.
Body: Background
Aromatase inhibitor (AI) therapy is associated with musculoskeletal toxicity, which adversely affects quality of life and therapy
adherence. Our objective was to evaluate genetic predictors for endocrine therapy-related musculoskeletal pain in a prospective
cohort study, which was previously published.
Patients & methods
254 early breast cancer patients starting AI (n=159) or tamoxifen therapy (n=95) were included in this genetic biomarker study.
Musculoskeletal symptoms were assessed at baseline and at 3, 6 and 12 months thereafter. AI-induced musculoskeletal pain
was defined as an increase in arthralgia or myalgia relative to baseline. Single nucleotide polymorphisms (SNP) in candidate
genes involved in estrogen signaling or reported to have associations with AI-related musculoskeletal pain or estrogen levels
were selected. Of the 40 selected SNPs, 7 failed genotyping and were excluded from further analysis.
Results
Twenty SNPs exhibited a minor allele frequency <0.05 and were omitted from the final analysis. Overall, 13 SNPs in CYP19,
CYP17, osteoprotegerin (OPG) and estrogen receptor 1 were included.
74% of AI-users reported new or worsened musculoskeletal pain compared to 33% of tamoxifen-treated patients. The OPG SNP
rs2073618 was significantly associated with AI-induced musculoskeletal toxicity after correction for multiple testing (p=0.039).
None of the other SNPs showed significant associations, both in AI and tamoxifen-treated patients.
Conclusion
The SNP rs2073618 in OPG predicts the risk of musculoskeletal symptoms with AI therapy, which has not been reported so far.
Validation of this finding and further functional studies are needed.

Title: BRCA1 and BRCA2 mutations in ethnic Lebanese Arab high risk women for hereditary breast cancer
Nagi El Saghir1, Nancy Uhrhammer2, Hussein Assi1, Katia Khoury1, Stephanie Decousous2, Yannick Bidet2, Sara Jaber1, Raghid
Charara1, Rania Farhat1, Ziad Salem1, Ali Shamseddine1, Arafat Tfayli1, Jaber Abbas1, Faek Jamali1, Muhieddine Seoud1,
Deborah Armstrong3, Yves-Jean Bignon2 and Nathalie Zgheib1. 1American University of Beirut Medical Center; 2Centre Jean
Perrin, Clermont-Ferrand, France and 3Johns Hopkins Kimmel Cancer Center.
Body: Background: Breast cancer is the most common malignancy in women in Lebanon and Arab countries with 50% of cases
below age 50. The incidence of hereditary breast cancer in Lebanese and Arab women is unknown.
Methods: 250 Lebanese women with breast cancer, of young age with or without family history, were recruited at the American
University of Beirut Medical Center (AUBMC) between 2009 and 2012. Study was approved by IRB. All signed an informed
consent. Risk assessment questionnaire, medical chart review, and whole blood were collected. Coding exons and intron-exon
boundaries of BRCA1 and BRCA2 were sequenced. Full BRCA gene sequencing was performed at Institut Jean Perrin, France.
Study was funded in part by an Ethnic Research Initiative (ERI) grant awarded by GSK.
Results:
14 out of the 250 patients (5.6%) had a deleterious BRCA mutation (7 BRCA1, 7 BRCA2) and 31 (12.4%) had a variant of
uncertain significance (VUS). Table 1 shows deleterious BRCA mutations based on age group and FH.
Deleterious BRCA mutations based on age group and family history
Total
Age 40 (with no Age 40 (with

FH)
positive FH)
Age 41-50 (with

positive FH)
Age 51 (with
positive FH)
Number of patients
250
74
74
75
27
Patients with deleterious

mutations (%)
14
(5.6%)
1 (1.4%)
8 (10.8%)
4 (5.3%)
1 (3.7%)
BRCA 1
BRCA 2
All 7 BRCA1 mutation carriers had a positive family history, were between 32 and 48 years of age, and had Grade 3 IDC with
negative ER, PR, HER2 receptors (TNBC). Six BRCA2 mutation carriers had IDC with positive hormone receptors (HR) and 2
had HER2-positive disease. We found 31 VUS. One VUS (BRCA2) was seen in two sisters with breast cancer. One VUS
(BRCA2) was seen in 4 patients and another in 2 patients, while 2 VUS (BRCA1) mutations were seen in 2 sets of 2 patients. The
significance of these VUS cannot be ascertained at this time. Haplotype analysis is ongoing.
Conclusions:
This is the first large study of ethnic Lebanese Arab women with breast cancer. The prevalence of BRCA deleterious mutations in
women with breast cancer who are considered high risk of carrying a BRCA mutation is 5.6% in our total cohort, while in patients
40 with positive FH it is 10.6%. Those numbers are lower than expected from US and European populations. Tumor grade and
pathology characteristics in this patient population correlated with that previously documented for BRCA1 (TNBC) and BRCA2
(positive HR) associated breast cancers. Our data supports use of young age together with positive FH should be used to select
patients for counseling and BRCA testing in Lebanon and Arab countries with resource-sensitive guidelines. Several VUS were
found in patients and sisters with breast cancer. The finding that 94.4% of high risk patients had no deleterious BRCA mutations
suggests the need to look for alternate gene mutations and other factors that may contribute to the development of breast cancer
in these high risk patients. Conclusions regarding haplotypes and diversity will be reported at the meeting.


Title: Different genomic rearrangements account for 17% of BRCA1/2 mutations in Greece
Angela Apessos1, Eirini Papadopoulou1, Vassiliki Metaxa-Mariatou1, Konstantinos Agiannitopoulos1, Christos Markopoulos2,
Vasileios Venizelos3, Grigorios Xepapadakis4, Maria Vasilaki-Antonatou5, Antonios Keramopoulos6, Nikolaos Bredakis6, Aristeidis
Tsiftsoglou7, Georgios Kesisis7, Stylianos Kakolyris8, Nikolaos Touroutoglou9, Ioannis Natsiopoulos9, Konstantinos Papazisis10 and
Georgios Nasioulas1. 1Genekor MSA, Athens, Greece; 2Medical School, University of Athens, Athens, Greece; 3Metropolitan
Hospital, Athens, Greece; 4Rea Maternity Hospital, Athens, Greece; 5IASO General Hospital, Athens, Greece; 6IASO Maternity
Hospital, Athens, Greece; 7St Luke's Hospital, Thessaloniki, Greece; 8University General Hospital of Alexandroupoli,
Alexandroupoli, Greece; 9Interbalkan Medical Center of Thessaloniki, Thessaloniki, Greece and 10Euromedica General Clinic of
Thessaloniki, Thessaloniki, Greece.
Body: AIM: The aim of this study was to further delineate the extent and nature of mutations in the BRCA1 and BRCA2 genes,
responsible for hereditary breast and ovarian cancer in Greek families.
MATERAILS & METHODS: Genomic DNA was isolated from whole peripheral blood of patients referred to our center for
mutation analysis of the BRCA1 and BRCA2 genes. Patients were included on the basis of affected family members, types of
cancer present in the family and age at diagnosis of breast cancer in the proband. Families were subdivided into high, medium
and low risk depending on the number of affected family members, types of cancer diagnosed in the family and age at diagnosis
of affected family members. In total, 675 families have been analyzed by our group in the past 4 years. Mutation analysis in all
cases included sequencing of the coding region and the splice sites of the two genes. In addition, MLPA analysis was carried in
585 of the patients.
RESULTS: In total, a pathogenic mutation has been identified in 12% of the 675 patients analyzed. Of the 78 mutations identified
in total, 13 (17%) were large genomic rearrangements. These were deletions of exons 8, 20, 23, 23-24 and the entire BRCA1
gene, in addition to a duplication of exons 3-8 of the BRCA1 gene. As far as BRCA2 is involved deletions of exons 3, 15 and the
entire BRCA2 gene were detected. All deletions were confirmed by use of other MLPA probe sets and/or relative quantitation by
Real Time PCR. Of the rearrangements identified, two, namely deletions of exon 20 and exons 23-24 of the BRCA1 gene were
identified in more than one unrelated families. In addition, the recurrent mutations 5382insC and G1738R, which have been
previously identified as founder mutations in the Greek population, were identified in multiple unrelated analyzed families.
CONCLUSIONS: Our results indicate that different large genomic rearrangements account for an important proportion (17%) of
the mutations in the BRCA1 and BRCA2 genes, in Greek families at risk of carrying a germline mutation as judged by family /
personal history. The use of the available technologies for the identification of such mutational events is therefore necessary
when carrying out complete analysis of the genes in high risk families of Greek background.



Title: Exome sequencing identified emergence of HER2 kinase domain mutations in trastuzumab-resistant breast cancer
Yizhou Jiang1, Ke-Da Yu1, Wen-Jia Zuo1, Yu-Jie Wang1 and Zhi-Ming Shao1. 1Fudan University Shanghai Cancer Center,
Shanghai, China.
Body: Purpose
Human epidermal growth factor receptor 2 (HER2)-positive breast cancer is primarily treated with trastuzumab and other
HER2-targeted therapies. However, approximately 30% of HER2-positive tumors exhibit de novo resistance, whereas 40%
acquire resistance to these therapies. The mechanisms underlying drug resistance are unclear, and a better understanding of
acquired resistance to HER2-targeted therapy is essential for the development of more effective therapeutics.
Methods
We enrolled 11 patients with HER2-positive breast cancer, all of whom had been treated with trastuzumab for one year and
developed lung metastasis in the follow-up. We conducted exome sequencing of the primary and paired metastatic tumors from
four patients. We identified somatic mutations in HER2 (K753E and L755S) affecting the kinase domain only in the metastatic
lesions of two cases. Sanger sequencing was performed to analyze all exons of the HER2 gene in the paired primary and
metastatic tumors of the remaining seven cases. Furthermore, we enrolled 48 patients with HER2-positive disease who
underwent trastuzumab-based neoadjuvant therapy. We performed Sanger sequencing of the HER2 gene in paired tumor DNA
before and after neoadjuvant therapy. To determine the phenotype of these mutations, we functionally characterized the HER2
mutations using protein structure analysis, in vitro kinase assays, cell culture, and xenograft experiments.
Results
We identified somatic mutations in HER2 affecting the kinase domain in four of 11 metastatic tumors (one case with K753E and
three cases with L755S). None of the primary tumors harbored HER2 mutations. Furthermore, Sanger sequencing in the
neoadjuvant setting revealed two of 48 cases harboring HER2 mutations (K753E and L755S) in the post-treatment samples, none
of whom had any HER2 mutation in the paired pre-treatment tumor. Protein structure visualization hinted that the HER2 L753 and
L755 side chains were in close proximity to the binding site for small-molecule inhibitors, indicating mutations at this residue might
produce drug resistance. In vitro kinase assays revealed that HER2 K753E had greater tyrosine kinase-specific activity than
wild-type. HER2 autophosphorylation, phosphorylation of HER2s dimerization partners (EGFR or HER3) and downstream
signaling proteins were significantly upregulated in MCF10A and MCF7 cells overexpressing HER2 K753E or L755S.
Trastuzumab or lapatinib treatment did not decrease the number of colonies formed in MCF10A cells transduced with K753E or
L755S. MCF7 cells bearing K753E or L755S mutant could reverse the inhibition of lapatinib in xenografts. The MCF10A-HER2
K753E and L755S cells were resistant to trastuzumab (IC50>1000g/mL) and lapatinib (IC50>10mol/L) but could be readily
inhibited by neratinib (IC50 of 15 nmol/L), an irreversible kinase inhibitor of HER2.
Conclusion
These data implicate that mutations in HER2 kinase domain are a key mechanism in resistance to HER2-targeted therapy. The
selection of functional HER2 mutations might act as potential drivers of trastuzumab resistance during the progression of
HER2-positive disease. Furthermore, more potent HER2 inhibitors such as neratinib might be of therapeutic benefit for
trastuzumab-resistant breast cancer.

Title: Prognostic and predictive implications of monosomy 17 in curable HER2-amplified breast cancer
David Page1, Yong H Wen1, Dana Dure1, Clifford Hudis1, Edi Brogi1 and Heather McArthur1. 1Memorial Sloan Kettering Cancer
Body: Background: Monosomy of chromosome 17 is identified in approximately 6% of HER2-amplified breast cancers.
Monosomy 17 has been retrospectively associated with inferior response to trastuzumab in the metastatic setting. The adverse
impact of monosomy 17 could be related to the loss of putative genes encoding DNA-binding regulatory proteins, such as ROX or
TOP2A. To date, the prognostic or predictive significance of monosomy 17 in women with potentially curable HER2-positive
breast cancer has not been fully evaluated.
Methods: Through a search of institutional databases, we identified women with HER2+ breast carcinoma with monosomy 17 by
fluorescence in situ hybridization (FISH) and treated at our center between 1/1/00 and 6/1/11. HER2-amplification was defined as
a HER2/CEP17 copy ratio 2.0, and monosomy 17 was defined as the presence of 1.4 CEP17 copies/cell. Clinical outcomes
were determined by chart review, defined by Standardized Definitions for Efficacy (STEEP). Differences in treatment and other
disease characteristics were compared by the log-rank test.
Results: We identified 99 women with stage I-III HER2-positive breast cancer showing monosomy 17 by FISH: 51% (n=50) were
treated with trastuzumab plus chemotherapy (tras/chemo), 31% (n=31) with chemotherapy (chemo alone), and 18% (n=18) with
neither (no tras/chemo). An adjuvant anthracycline was administered in 82% of women in the tras/chemo group and 97% of
women in the chemo group. The 3 treatment groups were balanced for age (median: tras/chemo 49y; chemo 53y; neither 56y),
but were not balanced for hormone receptor (HR)-status (HR+: tras/chemo 96%; chemo 81%; neither 78%, p=0.04) or TNM stage
(stage III: tras/chemo 16%; chemo 27%; neither 0%, p=0.01). With a median follow-up of 7y (0.16-13.4y), 4y distant relapse free
survival (DRFS) was 100% for the tras/chemo group, 87% for the chemo alone group, and 80% for the no tras/chemo group
(p=0.018); and 4y overall survival (OS) was 100% for the tras/chemo group, 93% for the chemo alone group, and 81% for the no
tras/chemo group (p=0.005).
Conclusions: Here we report the largest retrospective series of patients with curable HER2-amplified breast cancer and
monosomy 17. HER2-amplified monosomy 17 does not appear to adversely affect survival in the curative setting, with outcomes
comparable to those reported in large phase III adjuvant trastuzumab trials. Despite receiving less anthracycline-based
chemotherapy, better OS and DRFS were observed in women treated with tras/chemo compared with women treated with chemo
alone, thereby supporting the critical role of trastuzumab in this patient population, despite data to the contrary from the
metastatic setting.

Title: Activating HER2 mutations promote oncogenesis and resistance to HER2-targeted therapies in breast cancer
Wen-Jia Zuo1, Yi-Zhou Jiang1, Ke-Da Yu1 and Zhi-Ming Shao1. 1Fudan University Shanghai Cancer Center, Shanghai, China.
Body: Purpose: Somatic mutations in the tyrosine kinase domain of human epidermal growth factor receptor2 (HER2) have been
reported to lead to resistance to HER2-targeted therapies in HER2-positive breast cancer, while activating mutations of HER2
have been described in HER2-negative breast cancer. The prevalence, clinicopathological characteristics, and phenotypes of
HER2 mutations are not well established, thus we sought to describe the HER2 mutation profile of Chinese breast cancer
patients.
Methods: DNA samples were gathered from breast cancer patients undergoing neoadjuvant (N=102) or adjuvant therapy (N=498)
at Fudan University Shanghai Cancer Center between January 1, 2006 and December 31, 2012. Sanger sequencing was
performed to analyze all exons of HER2 to identify somatic mutations. To determine the phenotypes of novel HER2 mutations, in
vitro kinase assays, protein structure analysis, cell culture, and xenograft experiments were conducted.
Results: 10 HER2 somatic mutations were observed in 17 patients (17/600, 2.83%). 7 novel HER2 mutations were uncovered, 4
in the transmembrane domain and 3 in the kinase domain. Kinase domain mutations L768S and V773L were detected in
HER2-negative tumors, while K753E was found in HER2-positive disease. In vitro kinase assays found that L768S and V773L
exhibited a significant increase of tyrosine kinase-specific activity, while Western blots showed that L768S and V773L strongly
increased phosphorylation of all signaling proteins in both MCF10A and MCF7cell lines, indicating that they were activating
mutations. In Matrigel cultures, L768S and V773L formed acini when seeded in vehicle, but maintained spherical morphology
when seeded in culture containing trastuzumab. The addition of lapatinib in Matrigel culture inhibited the growth of all except
K753E, which was successfully inhibited by neratinib. Similarly, L768S, V773L and K753E increased the number of cell colonies
formed in soft agar, trastuzumab and lapatinib treatment decreased the number of colonies formed by L768S and V773L, but only
neratinib could inhibit the colony growth of K753E. Xenograft showed L768S and V773L displayed a more rapid growth, while
K753E showed resistance to lapatinib in vivo. MCF10A cells bearing K753E mutation were found to be resistant to lapatinib
(IC50>10,000 nmol/L), but could be inhibited by neratinib, though requiring a relatively higher dosage (IC50 of 32 nmol/L) than
HER2 WT (IC50 of 480 nmol/L for lapatinib, <2 nmol/L for neratinib) and other HER2 mutations. Meanwhile, clinical follow-up
showed that the 2 patients with K753E mutation who received adjuvant trastuzumab treatment presented with either brain or bone
metastasis, in their 3rd and 5th year after initial cancer diagnosis, suggesting K753E mutation may have a role in trastuzumab
resistance as well.
Conclusions: HER2 somatic mutations were found in 2.83% of patients in this study. HER2-positive tumors harboring certain
HER2 kinase domain resistance mutations may not benefit from trastuzumab or lapatinib treatment, and neratinib may offer an
alternative treatment option for these patients. HER2-negative disease with activating mutations may benefit from HER2-targeted
therapies, and may be of interest in prospective clinical trials.


Title: Prevalence of PIK3CA mutations in tumor tissue of a consecutive cohort of breast cancer patients (n=700)
Martina Vetter1, Kristin Reinhardt1, Silke Wegener1, Juergen Dittmer1, Tilmann Lantzsch2, Christoph Uleer3, Susanne Peschel4,
Jutta John5, Volker Hanf6, Joerg Buchmann7, Karl-Friedrich Brrig8, Edith Weigert6, Christoph Thomssen1 and Eva J Kantelhardt1.
1
GYN Universitaet Halle-Wittenberg, Halle (Saale), Germany; 2GYN St Elisabeth & St Barbara Hospital, Halle (Saale); 3GYN
Praxis Uleer, Hildesheim; 4GYN St Bernward Krankenhaus, Hildesheim; 5GYN Stdtisches Klinikum, Hildesheim; 6GYN Klinikum
Frth, Frth; 7Institut fuer Pathologie Martha Maria, Halle (Saale) and 8Institut fuer Pathologie Hildesheim, Hildesheim.
Body: Introduction:
The AKT / mTOR pathway is activated by phosphorylation of the lipid kinase phosphatidylinositol 3-kinase (PI3K) and regulates
proliferation, apoptosis, survival and adhesion of tumor cells. In up to 30% of breast cancers, dysregulation of PI3K is reported,
resulting from mutations in the PIK3CA gene that encodes the catalytic subunit (p110a). Hotspots of mutations were described in
exon 9 (helical domain, E542K, G > A) and exon 20 (kinase domain, H1047R, A > G). Prevalence and the prognostic impact of
the PIK3CA mutations as well as the predictive value with regard to endocrine therapy are controversially discussed. In this study
we describe the prevalence of PIK3CA mutations in a consecutive cohort of breast cancer patients and its association to tumor
characteristics and known prognostic factors.
Material & Methods:
Fresh frozen tumor samples were obtained from a consecutive cohort of 700 breast cancer patients (inclusion criteria: pTx pNx
M0 Gx HRx HER2x) who were primarily operated in 6 centres between 2009 and 2011. Tumor DNA was extracted and analyzed
by conventional and quantitative PCR (exon 9: cosmid 763 and exon 20: cosmid 775). Patient data, tumor characteristics and
prognostic factors were obtained from patients charts. ER, PgR and HER2 results were based on local pathology, uPA and PAI-1
values were determined centrally.
Results:
All isolated DNA (n=700) were analyzable by PCR and showed a mutation rate of 22% in the entire cohort; including two tumors
with both mutations. Among the tumors with a positive steroid hormone receptor (HR) status (n=606; 87%), we found a mutation
rate of 24% for PIK3CA, whereas only 9 (10%) of 94 HR-negative tumors were mutated. Tumors with a PIK3CA mutation were
significantly associated with HR positivity / HER2 negativity (p=0,001). 14% of the cohort was HER2 positive. In 98 cases with a
HER2 positive tumor 17 DNAs were mutated (17%) and 139 DNAs of the 602 HER2 negative tumors had a positive PIK3Ca
mutation status (23%). Of the mutated tumors, 94 % were HR positive, and 11% HER2 positive. Only three samples were HER2
positive and HR-negative. 6% of the mutated tumors were triple negative.
No significant association was found to age, menopausal status, tumor stage, nodal status, grading and uPA/PAI-1 status. Most
mutant tumors were histological grade 2 (p=0,001). A quarter of the intermediate risk group according to St. Gallen risk estimation
had a somatic PIK3CA mutation.
Conclusion:
In our cohort, nearly all mutated tumors are hormone receptor positive and a minority of all mutated samples is HER2 positive.
This data adds important information to the heterogeneous results of other previously published patient cohorts. It suggests a
possible role of PI3K-dysregulation with regard to resistance against endocrine therapy and anti-HER2-treatment.

Title: No Show

Title: Histone deacetylase HDAC7 is a putative therapeutic target in invasive lobular carcinoma
Matthew J Sikora1 and Steffi Oesterreich1. 1University of Pittsburgh, Pittsburgh, PA.
Body: Invasive lobular carcinoma (ILC) is the second most common histological subtype of breast cancer, accounting for 10-15%
of cases. Though ILC represents a minority of breast cancer patients, ILC affects over 30,000 women annually in the US, making
ILC the 6th most common womens cancer. Despite this incidence, no ILC-specific therapeutic options exist. The majority of ILC
are estrogen receptor (ER)-positive (>90%); this and other biomarkers suggest that ILC patients are ideal candidates for
endocrine therapy. However, the efficacy of endocrine therapy in ILC is poorly understood since no ILC-specific prospective
clinical trial data exist. Recent retrospective analyses suggest that a subset of ILC patients may not benefit from endocrine
therapy. Understanding the unique mechanisms of endocrine response and resistance in ILC is critical to improving ILC patient
outcomes.
We recently identified estrogen-mediated gene expression specific to the ILC cell lines MDA MB 134VI and SUM44PE.
ILC-specific ER target genes were enriched for repression by estrogen. Further, tamoxifen regulated estrogen-repressed genes
as an agonist in MDA MB 134VI cells, in parallel with tamoxifen-induced growth. Thus, ER-mediated repression of target genes
may be critical in maintaining cell growth and survival. Our laboratory previously identified that ER-mediated gene repression in
breast cancer cells is mediated by histone deacetylase 7 (HDAC7). E2-induced repression requires specifically HDAC7, not
HDAC1-6/8-10. In a panel of breast cancer cell lines, HDAC7 was most strongly expressed in MDA-MB-134VI cells, the only cell
line among the panel derived from an ILC. Increased HDAC7 expression in ILC versus invasive ductal carcinoma (IDC) can also
be observed in the Cancer Genome Atlas; mean HDAC7 expression in ER-positive tumors is 2-fold higher in ILC (n=148) versus
IDC (n=508) (p<0.0001). We hypothesize that HDAC7 mediates the unique ER-mediated gene repression in ILC cells, and may
serve as a novel therapeutic target in conjunction with endocrine therapy for ILC.
To model HDAC7 inhibition, we generated stable cell lines carrying inducible HDAC7 shRNA constructs from MCF-7 (IDC) and
MDA MB 134VI (ILC). These lines are being used to examine the effects of HDAC7 depletion on estrogen- and
tamoxifen-mediated gene expression, cell proliferation, anoikis resistance, and endocrine resistance. Parallel studies using the
class IIa HDAC inhibitor MC1568 are being used as proof-of-principle for using HDAC7-specific inhibitors. Additionally, we
identified novel splice variants of HDAC7 in breast cancer cells. Based on the mouse homolog Hdac7, these splice variants may
alter the ability of HDAC7 to interact with transcription factors, or change its activity as a transcriptional repressor.
Our observations suggest that HDAC7 may mediate ER-mediate gene repression in ILC cells, which may be necessary for cell
proliferation and endocrine resistance in this breast cancer subtype. Preclinical studies using ILC model systems will identify the
role of HDAC7 and specific splice variants in controlling ER-driven gene expression. Further studies using HDAC7-specific
inhibitors will determine whether targeting HDAC7 in ILC patients can improve endocrine response and inhibit endocrine
resistance.

Title: Identifying hypermethylated tumor suppressor genes in breast cancer with an in vivo total genome knockdown screen
Margaret Thomas1, Krysta Coyle1, Mohammad Sultan1, Luzhe Pan1, Dae-Gyun Ahn1, Patrick Lee1, Carman Giacomantonio1 and
Paola Marcato1. 1Dalhousie University, Halifax, NS, Canada.
Body: Changes in gene expression are required for the progression of breast cancer; and DNA methylation, which silences tumor
suppressor genes, is often responsible for these changes. Several de-methylating/DNA synthesis inhibiting drugs (such as
decitabine) are potential breast cancer therapies; however, for their application to be successful, we need to identify which
patients would most benefit from these treatments. To accomplish this, the critical genes, which must be silenced by
hypermethylation in order for breast cancer to progress, need to be identified. For this purpose, we have performed an in vivo
total genome knockdown screen.
A genome-wide lentiviral-based shRNA screen was performed with breast cancer cell line MDA-MB-231 tumors implanted in the
mammary fat pads of female NOD/SCID mice, with or without decitabine treatment (50mg/kg). The resulting tumors were
harvested and shRNA sequences retrieved and hybridized to a Decode microarray, allowing for the identification of shRNA
sequences that were enriched or depleted under decitabine treatment. The enriched shRNA sequences likely target genes of two
categories: 1) they are methylated in the tumor and when expressed have tumor suppressive function, or 2) they are required for
decitabine-mediated DNA synthesis inhibition and apoptosis. To determine if the shRNA-targeted genes identified in the screen
fall within category 1 or 2, further analysis was done using GEO datasets for methylation in breast cancer cell lines, primary
tumors, and normal breast tissues.
In replicates of 6 mice, 111 shRNA sequences were enriched more than 2-fold in 5/6 mice. Of the 111 genes, 29 genes showed
significant enrichment (p<0.05) in the decitabine-treated tumors. Using the GEO datasets, 19 of those significantly enriched
genes fit a tumor suppressor methylation profile and merit further investigation of their methylation status in the MDA-MB-231 cell
line. The remaining enriched genes may play a role in DNA synthesis inhibition and apoptosis- this will be confirmed with in vitro
apoptosis assays. DAVID analyses did not reveal any functional clustering of the genes of interest, but did implicate several
pathways that involved a single potential hypermethylated tumor suppressor gene.
We have identified several genes that are hypermethylated in breast cancer and are potentially involved in tumor progression.
Confirmation experiments will reveal a list of genes that when found hypermethylated in patient tumors would identify candidate
breast cancer patients who would benefit from decitabine treatment.

Title: Maintenance of a lean phenotype is associated with increased ER expression and ER gene intron methylation in murine
MMTVneu luminal mammary cancer
Emily L Rossi1, Sarah M Dunlap2, Nikki A Ford2, Laura W Bowers1 and Stephen D Hursting1. 1University of North Carolina, Chapel
Hill, NC and 2University of Texas, Austin, TX.
Body: Background: Obesity is associated with increased breast cancer mortality in premenopausal women, who predominately
develop estrogen receptor (ER)- negative breast tumors. New mechanism-based targets and intervention strategies for
offsetting the procancer effects of obesity are urgently needed given the rising rates of obesity in women throughout the world and
the lack of effective targeted therapies for ER-negative breast cancer. Unfortunately, the mechanisms by which obesity impacts
ER-negative breast cancer prognosis remain unclear. Methods: In this study, we utilized the MMTV-neu transgenic mouse model
of Her2+ breast cancer to test the hypothesis that dietary energy balance modulation alters mammary tumor development and
progression through epigenetic regulation of ER in the mammary epithelium of these mice. MMTV-neu transgenic mice form
spontaneous mammary tumors that progress from an ER-positive hyperplasia to aggressive ER-negative ductal
adenocarcinomas. Female MMTV-neu transgenic mice were randomized to 3 diet regimens (30/regimen), resulting in either an
obese, overweight or lean phenotype. A subset of mice were killed at baseline, 1, 3, and 5 months following diet initiation, and
tissues were collected for analysis; remaining animals were followed for up to a 22 month survival study. Results: Gene and
protein expression analysis revealed that mammary ER expression, known to be lost by 8 weeks of age in normoweight
MMTV-neu transgenic mice, was maintained in lean mice but lost in the overweight and obese mice (which did not statistically
differ in any parameter studied). Additionally, we found lean mice had significantly increased mammary ER gene expression and
delayed onset of hyperplasia relative to the overweight/obese mice. ER and ER gene intron methylation increased from 5
months to 12 months on diet in lean mice and conversely decreased in overweight/obese mice, which parallels the ER gene
expression data. After 22 months of feeding, lean mice had significantly increased tumor-free survival in relative to the
overweight/obese mice. Interestingly, human and mouse HER2+ (but not but not HER2-) breast cancer cell lines, when treated
with serum from obese (BMI>30), women resulted in a significant down-regulation of ER gene expression. Conclusions: In
MMTV-neu transgenic mice, dietary energy balance modulation impacts spontaneous mammary tumor development and ER
and ER levels in normal and tumor tissue, possibly through epigenetic mechanisms. Moreover, increased mammary ER
expression may represent a novel target for interventions to offset the enhancing effects of obesity on breast cancer.

Title: Super-enhancer analysis defines breast cancer subtype and identifies tumor dependencies
Matthew G Guenther1, David A Orlando1, Matthew L Eaton1, Cindy A Collins1, Mei Wei Chen1, Sneha Solanki1, Jakob Loven1,
Christian C Fritz1 and Eric R Olson1. 1Syros Pharmaceuticals, Watertown, MA.
Body: Epigenomic modifications define gene regulatory features that control transcription and disease cell state. Recent studies
of these regulatory features have identified large clusters of enhancers, termed super-enhancers, which define key cell identity
and disease genes. Using ChIP-seq and RNA-seq analysis, we have discovered Super-enhancers in breast cancer cell line
models and in primary tissue and have characterized their roles in establishing tumor cell state. We find that Super-enhancers
recapitulate clinical subgroups in both breast cancer cell line models and in invasive ductal carcinoma.
Super-enhancer-associated genes encode known and novel therapeutic targets including kinases, phosphatases, chromatin
regulators and transmembrane proteins. Such genes include key drivers such as ERRB2 in HER2+ patient samples, ESR1 in
estrogen receptor positive samples, and CCND1 in samples of luminal subtype. We describe the biological and disease relevance
of Super-enhancer-associated genes in the context of tumor cell state and drug target discovery.

Title: DNA methylation-based classification are mostly concordant with intrinsic subtypes of breast cancer
Natsue Uehiro1, Fumiaki Sato1, Sunao Tanaka1, Fengling Pu2, Junji Itou1, Shigehira Saji2 and Masakazu Toi1. 1Graduate School of
Medicine, Kyoto University, Kyoto, Japan and 2Graduate School of Medicine, Kyoto University, Kyoto, Japan.
Body: Backgrounds:
Breast cancer (BC) is a heterogeneous disease and usually divided into 5 subtypes according to clinicopathologic features. These
subtypes have concordance with mRNA-based subtypes called as "intrinsic subtypes". Further gene expression analysis
categorized triple negative BCs into 7 sub-subtypes which relate to clinical behaviors. DNA methylation is one of the epigenetic
systems to regulate gene transcription. DNA methylation in CpG islands (CGIs) associated with the gene promoter region affects
gene expression levels. Recently genome-wide analyses suggested that methylation status not only in CGIs, but also in CpG
shores (within 2000 bps from CGIs) and CpG shelves (2000-4000 bps from CGIs) contribute to gene silencing. Thus, we
hypothesized that intrinsic subtypes may possess specific methylation profiles especially in regions surrounding CGIs (islands,
shores and shelves). To date, several reports demonstrated each intrinsic subtype has specific methylation patterns in CGIs.
They indicate that DNA methylation is promising markers for cancer detection, prediction of therapeutic response and clinical
outcomes. However, there were no report that conducted a genome-wide high-resolution epigenetic analysis for BC.
Objectives:
This study aims to determine whether specific DNA methylation patterns exist in CGIs, shores and shelves in BC cells, and
whether the DNA methylation-based classification correlates with intrinsic subtypes, furthermore, to identify the subtype specific
epigenetic markers using high-resolution methylation array.
Materials and Methods:
DNA was extracted from 31 samples of 28 BC cell lines (8 luminal, 18 basal, 2 unclassified) and 3 samples of 3 non-BC cell lines.
We performed genome-wide methylation analysis using Illumina Infinium HumanMethylation450 BeadChip. We did unsupervised
hierarchical clustering analysis of cell lines using top 3000 variably methylated (VM) loci among cancer cell lines. We also
analyzed top 3000 VM CGI-related loci. Gene functional annotation clustering was performed using DAVID.
Results:
In top 3000 VM loci, CGIs were most frequently observed (43.0%). According to clustering analysis using top 3000 VM loci, cell
lines were roughly classified into 4 clusters. There were trends that cell lines of same subtype fell into the same cluster, but not
definitely. In case of clustering using top 3000 VM CGI-related loci, cell lines were clearly divided into 2 clusters, high and low
methylated clusters. High methylated cluster contained almost all luminal cell lines and low methylated cluster contained most of
basal cell lines and all non-BC cell lines. In low methylated cluster, cell lines of the same sub-subtype, such as Basal-like 1 and 2,
were clustered closely. Gene functional annotation clustering suggested that genes related to "transcription" and "differentiation"
were more frequently regulated by epigenetics.
Conclusions:
High-resolution methylation analysis revealed methylation-based clusters were concordant with intrinsic subtypes in BC cells.
Transcriptional factors and differentiation-related genes could be differently regulated by epigenetics among different subtypes of
BC.

Title: Meta-analysis reveal novel mechanism of bisphenol A effect on mammary gland
Xiang Gu1, Hui Gao1,2, Danhan Wang2, Abhik Bandyopadhyay1, Qiaoxiang Dong1,2 and LuZhe Sun1. 1University of Texas Health
Science Center, San Antonio, TX and 2Institute of Environmental Safety and Human Health, Wenzhou Medical University,
Wenzhou, Zhejiang, China.
Body: Pubertal exposure to bisphenol A (BPA) has been shown to cause abnormal mammary gland development and neoplastic
transformation in adults. However, the mechanism of this BPA effect still remains largely unknown. We exposed 21-day-old
Balb/c mice to BPA by gavage (25 g/kg/day) during puberty for 3 weeks. Primary mammary cells were isolated at 6 weeks, and
2 and 4 months of age and subsequently sorted into Lin-CD49f-CD24-stromal cells, Lin-CD49fhighCD24med MaSCs-enriched
basal cells and Lin-CD49flowCD24high luminal progenitor-enriched cells with FACS for gene expression analysis. The Gene
Ontology (GO) Enrichment Analysis shows that upon BPA treatment, the biological processes (BPs) significantly enriched in
luminal compartment were related to response to organic substance and hormone stimuli at early stage (2 month old) whereas to
cell cycle regulation at late stage (4 month old), consistent with published study in mammary gland morphology where BPA
promoted the expansion of luminal population. However, the BPs enriched in spheres derived from basal cells (highly enriched
with MaSCs) collected at late stage (4 month old) were mainly related to cell adhesion, epithelial differentiation and in stromal
cells were related with vascular development for tissues collected either immediately after BPA treatment or 6 weeks later at
4-month of age. More interestingly, the Downstream Function Effects Analysis performed by Ingenuity Pathway Analysis reveal
the changes in any compartments will all lead to significant activation of cancer-related downstream effects including hyperplasia,
neoplasm of mammary tumor, proliferation of tumor cells and invasion of breast cancer, although there are few overlapping
common genes in these compartments and conditions upon BPA treatment, indicating the different mechanism in response to
BPA treatment and tumorigenesis. Network analysis revealed that the potential upstream regulators for these enriched BPs are
TGF1, VEGF, PPARG, Estrogen/ Estrogen Receptor, glucocorticoid, ERK and NOTCH signaling. Together with our recently
published data, this study reveals the difference in immediate and delayed response to pubertal BPA treatment and indicates that
cross-talk between several growth factor and hormonal receptor signaling pathways may contribute to the altered cellular
processes by BPA, consistent with the role of BPA as an endocrine disruptor.

Title: Modeling breast cancer dormancy and re-emergence
Amanda M Clark1, Sarah E Wheeler1, Donald P Taylor1,2, Carissa L Young3, Venkateswaran C Pillai7, Donna B Stolz1,3,5,8, Raman
Venkataramanan7, Douglas A Lauffenburger3, Linda G Griffith3 and Alan Wells1,2,5,6. 1University of Pittsburgh, Pittsburgh, PA;
2
University of Pittsburgh, Pittsburgh, PA; 3University of Pittsburgh, Pittsburgh, PA; 4Massachusetts Institute of Technology,
Cambridge, MA; 5McGowan Institute for Regenerative Medicine, University of Pittsburgh, Pittsburgh, PA; 6Pittsburgh VA Medical
Center, VA Pittsburgh Healthcare System, Pittsburgh, PA; 7University of Pittsburgh, Pittsburgh, PA and 8University of Pittsburgh
Cancer Center, University of Pittsburgh, Pittsburgh, PA.
Body: Most breast cancer (BrCa) mortality results from distant metastases. Current evidence strongly suggests that in some
instances these disseminated cells remain dormant for long periods of time. Both the non-proliferative state and protective
microenvironment of the metastatic niche likely contribute to the observed resistance of metastases to chemotherapies that are
otherwise effective against the primary tumor. Although significant interventional progress has been made on primary tumors, the
lack of relevant accessible model systems for metastases has hindered the development of therapies against this stage. To
address this gap, we developed an innovative all-human 3D ex vivo hepatic microphysiological system (MPS) to faithfully
reproduce human physiology and thereby facilitate the investigation of BrCa behavior in a micrometastatic niche. The liver is a
major site of metastasis for carcinomas and is also the primary site of drug metabolism (activation and/or detoxification), which is
a significant factor in determining efficacy and limiting toxicities of cancer therapies.
The MPS incorporates hepatocytes and nonparenchymal cells (NPC) isolated from fresh human liver resections. BrCa cells
(RFP+) are seeded on day 3 and afforded time to intercalate into the hepatic tissue until treatment with chemotherapy on day 7
for 72h. Surviving BrCa cells are stimulated on day 13 with LPS/EGF and cultured through day 15. Proliferation is monitored by
RFP quantification, Ki67 staining and EdU incorporation. Physiological function of the hepatic tissue is monitored throughout the
experiment by protein catabolism (urea), active metabolism (glucose, CYP P450) and injury markers (AST, ALT, A1AT,
fibrinogen). Luminex assays (55 analytes) were used to provide insights into the communication networks in the hepatic
metastatic milieu during different stages of dormancy and progression, and identify potential metastatic biomarkers via
computational approaches.
The MPS maintains the physiologic function of the hepatic niche through 15 days and BrCa cells effectively integrate into the
established niche. Spontaneous dormancy is observed amongst a subpopulation of BrCa cells, indicated by the absence of Ki67
staining and EdU incorporation after 12 days of culture. Further, we demonstrate that the BrCa cells surviving chemotherapy
(doxorubicin) are non-proliferating (Ki67-/EdU-). Notably, re-awakening of the surviving non-proliferating cancer cells is observed
in the presence of physiological inflammatory stressors (LPS/EGF). Luminex analyses of the milieu effluent identified signaling
molecules from NPC influenced the metastatic cell fraction entering dormancy.
This MPS provides unprecedented insights into the tumor biology of dormant micrometastases. We demonstrate the recreation of
spontaneous, rather than engineered, BrCa dormancy in an all-human ex vivo hepatic MPS. Mimicking the dormancy and
outgrowth observed in patients, we found that dormant breast cancer cells that are resistant to chemotherapy can be stimulated
to re-emerge following an inflammatory insult. Ultimately, this MPS provides an accessible tool to identify new therapeutic
strategies for metastasis during initial seeding, dormancy and re-emergence, while concurrently evaluating agent efficacy for
metastasis, metabolism and dose-limiting toxicity.

Title: Mesenchymal stem cell regulated microRNAs converge on the speech gene FOXP2 and regulate breast cancer metastasis
Benjamin G Cuiffo1, Antoine Campagne1,2, George W Bell3, Antonio Lembo4, Francesca Orso4, Evan Lien1, Manoj K Bhasin1,
Monica Raimo4, Summer E Hanson5, Andriy Marusyk6, Peiman Hematti5, Kornelia Polyak6, Odette Mariani2, Stefano Volinia7,
Anne Vincent-Salomon2, Daniela Taverna4 and Antoine E Karnoub1,8,9. 1Beth Israel Deaconess Medical Center, Harvard Medical
School, Boston, MA; 2Institut Curie, Paris, France; 3Whitehead Institute for BioMedical Research, Cambridge, MA; 4University of
Turin and MBC Via Nizza, Torino, Italy; 5University of Wisconsin-Madison, School of Medicine and Public Health, Madison, WI;
6
Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA; 7University of Ferrara, Ferrara, Italy; 8Harvard Stem Cell
Institute, Cambridge, MA and 9Broad Institute of MIT and Harvard, Cambridge, MA.
Body: About 90% of breast cancer mortalities are due to the spread of breast cancer cells (BCCs) from a primary tumor to distant
organs, a process known as metastasis. However, the molecular mechanisms underlying metastasis remain poorly understood.
Substantial evidence now supports a major role for the tumor microenvironment (TME) in catalyzing breast cancer metastasis.
Indeed, observations indicate that proximal interactions between BCCs and cells of the TME induce altered gene expression
programs in BCCs, allowing for the navigation of the various steps of the metastatic cascade. Our group and others observed that
breast tumors recruit mesenchymal stem cells (MSCs): multipotent fibroblasts that normally exert tissue maintenance functions.
We and others have observed that physical interactions of MSCs with BCCs are sufficient to drive their metastatic dissemination
in murine xenograft models, via the induction of epithelial-mesenchymal transition (EMT) and dedifferentiation into stem cell-like
states (cancer stem cells, or CSCs), states tightly associated with the capacity to seed new tumors (for example in foreign
tissues) and with chemotherapeutic resistance. However, the TME-induced molecular pathways regulating such mechanisms
remain poorly understood.
MicroRNAs (miRNAs, miRs) are small noncoding RNAs that regulate gene expression via base-pair interactions with messenger
RNAs (mRNAs), resulting in mRNA degradation or translational inhibition. Due to their ability to interact with large numbers of
target mRNAs simultaneously, miRNAs are major regulators of cell identity, and thereby serve critical roles in metastasis.
We performed miRnome-wide screening of MSC-stimulated BCCs to determine if TME interactions might contribute to BCC
metastasis via the deregulation of miRNAs. We observed that proximal MSCs induce aberrant expression of a specific set of
miRNAs in BCCs, which had not been previously implicated in breast cancer pathogenesis. These miRNAs, led by the
transcriptionally co-regulated miR-199a-3p and miR-214, were sufficient to actuate the metastasis of weakly metastatic human
BCCs in xenograft models. We observed that exogenous expression of the miRNAs provided BCCs with phenotypes and gene
markers characteristic of CSCs, including enhanced tumor initiation capacities. Interestingly, we found that the MSC-induced
miRNAs function as an interrelated network, and converge upon a common novel target: the speech associated gene FOXP2.
Knockdown of FOXP2 phenocopied the metastatic phenotypes observed in MSC-induced miRNA expressing BCCs. Importantly,
elevated levels of the MSC-induced miRNAs or depressed levels of FOXP2 could predict patient prognosis in the clinic.
Altogether, our results incriminate FOXP2 and its MSC-induced miRNA regulatory network as novel determinants of breast
cancer metastasis.

Title: Dual inhibition of IGF1R and insulin receptor in estrogen receptor positive and triple negative breast cancer and monitoring
blockade of metastasis using novel MRI
Deepali Sachdev1, Joel D Winer1, Naoharu Kobayashi2, Michael Garwood2 and Joseph Weber1. 1Masonic Cancer Center,
University of Minnesota, Minneapolis, MN and 2Center for Magnetic Resonance Research, University of Minnesota, Minneapolis,
MN.
Body: Insulin-like growth factors (IGFs) and insulin acting via the type I IGF receptor (IGF1R) and insulin receptor (IR)
respectively regulate biology of estrogen positive (ER+) and triple negative (TN) breast cancer cells. In animal models, inhibition
of IGF1R alone with antibodies has demonstrated significant inhibition of tumor growth and/or metastasis of several types of
cancer cells. Unfortunately, the results of initial clinical trials with anti-IGF1R antibodies have been disappointing. One reason for
this is that inhibition of the related IR through which insulin and IGF-II signal, may be necessary for optimal efficacy of this
targeted approach. Second, there is a need to develop markers that can be used to stratify patients who may benefit from these
drugs and/or monitor response to these drugs.
To examine the effectiveness of dual inhibition of IGF1R and IR, we evaluated the effects of BMS-754807, a small molecule dual
tyrosine kinase inhibitor of IGF1R/IR on ER+ and TN breast cancer cells. In ER+ cells (MCF-7, T47D and ZR-75-1), BMS-754807
inhibited IGF-I, IGF-II and insulin stimulated activation of downstream PI3K and MAPK pathways, proliferation and
anchorage-independent growth in vitro. BMS-754807 also blocked signaling in MCF-7 tumors and inhibited xenograft growth of
MCF-7 tumors (n=10 per treatment) compared to vehicle. Interestingly, while tumor growth was suppressed over a period of five
weeks, eventually the tumors displayed resistance to BMS-754807. In TN cell lines (MDA-MB-231; MDA-MB-231-LM2 and
MDA-231-BoM, lung seeking and bone specific metastatic variants respectively of MDA-MB-231; and MDA-MB435A/LCC6)
BMS-754807 also inhibited activation of the PI3K pathway and motility in vitro. In contrast to ER+ cells, BMS-754807 did not
inhibit primary tumor growth of TN breast cancers cells injected into the mammary fat pad of mice. But BMS-754807 (50 mg/kg
daily by oral gavage) inhibited metastasis of TN cells, MDA-231-LM2 and MDA-MB435A/LCC6 cells, in the orthotopic and tail vein
models of metastasis compared to vehicle (n=10/group). Our data indicate that regulation of metastases and tumor growth by
IGF1R can be discrete events and functional imaging to identify biological properties of metastatic breast cancer regulated by
IGF1R/IR are needed to better define treatments. Therefore, we used a novel MR sequence called sweep imaging with Fourier
transformation (SWIFT), where the data is acquired quasi-simultaneously with the radiofrequency pulse, to monitor response to
BMS-754807. We used MDA-231-LM2 cells with the tail vein injection model of metastasis. Mice injected with breast cancer cells
were treated with either vehicle or BMS-754807 and metastasis was monitored by BLI and SWIFT MRI weekly. SWIFT was
sensitive in detecting inhibition of metastasis by BMS-754807.
Our results suggest that dual inhibition of IGF1R and IR is effective in blocking growth of ER+ and metastasis of TN breast
cancers. However, combination of this therapeutic strategy with other agents may be necessary to prevent or delay onset of
resistance. Further, noninvasive biomarkers of response to IGF1R/IR targeted drugs can be developed with SWIFT imaging.


Title: Triple negative breast cancer metastases: Protein signal transduction networks within the tumor-stromal microenvironment
complement genomic analysis and stratify local versus distant metastasis
Virginia Espina1, Maren K Levin2, John D Carpten3, David W Craig3, Daniel Von Hoff3, Lance A Liotta1 and Joyce
O'Shaughnessy4. 1George Mason University, Manassas, VA; 2Baylor Sammons Cancer Center, Dallas, TX; 3Translational
Genomics Research Institute, Phoenix, AZ and 4Baylor Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, TX.
Body: Background: Genomic profiling of primary TNBC may not reflect the lethal metastatic (mTNBC) lesions that constitute a
selected subclone of the primary tumor. Moreover, while the stroma microenvironment of primary tumors has been the subject of
study, almost nothing is known about the stromal microenvironment of metastasis. We conducted proteomic signal pathway
network analysis of microdissected tumor cells and stroma from mTNBC samples for which whole genome sequencing was
known. We addressed the following questions: 1) Can potentially actionable driver mutations in the metastasis be inferred from
the activated/ suppressed protein signaling pathways in the tumor cells? 2) Are the tumor cell signaling pathways different
between local and distant metastasis? 3) Does the proteomic signaling profile of the stroma provide strategies for stromal therapy
of mTNBC?
Methods: Enriched populations of mTNBC cells and adjacent stromal cells were collected by Laser Capture Microdissection from
13 fresh frozen mTNBC samples. Signal pathway profiling of 130 protein and/or phosphoprotein endpoints were quantified by
reverse phase protein arrays (RPPA). The samples represented local (regional lymph node and chest wall) (n=7) and distant
(n=6) metastasis. Spearman rho correlation analysis identified pair-wise protein linakges (rho=0.8, p<0.01). Whole genome
sequencing was also performed on the mTNBC samples (Craig DW. Mol Cancer Ther, 2012).
Results: We compared the protein signaling network of the tumor and matched stroma to investigate tumor-stroma crosstalk, and
also to the genomic alterations (GAs) in the same tumor. GAs in mTNBC fell into two major categories: DNA repair or cell
cycle/growth factor signaling. Proteomic analysis confirmed the down-regulation of endocrine (ER, AR, PR) and HER2 pathways.
PhosphoJAK1/2 and phosphoHer3 were markedly down-regulated in mTNBCs, while EGFR and Her4 signaling were not.
Spearman rho pairwise correlations revealed differences between local and distant metastases as well as between local and
distant stroma. Cyclin D1 in locally metastatic tumor cells was linked with Autotaxin, PLCgamma and RUNX1, whereas in distant
metastatic tumor cells, Cyclin D1 was linked with HIF-1alpha, phosphoJAK1, and HSP70. Growth factor signaling via
phosphoEFGR was more prevalent in distant mTNBC tumor samples (322 linkages) compared to local mTNBC tumor samples
(146 linkages). Distant stromal cell linkages were dominated by growth factor (phospho mTOR) and immune cell crosstalk, e.g.
21 linkages between CD63 in distant stromal samples compared to 1 CD63 linkage in local stromal cells.
Discussion: Proteomic signal pathway data can stratify metastatic lesions into functionally important groups based on a) clinical
phenotype (distant versus local metastasis) and b) GA subtype (DNA repair vs growth factor signaling), thereby providing new
actionable strategies for novel therapies, e.g. anti-JAK1 and anti-HSP70, and providing a means to prioritize potential driver
mutations. Profiling the signal pathway network of mTNBC stroma, for the first time, provides novel strategies for targeting
immune cell and mTOR pathways, for new classes of stromal therapy.

Title: Activation of oncogenic pathways by mitochondrial reprogramming in triple negative breast cancer
Jun Hyoung Park1, Santhosh Kumar1, Sajna Vithayathil1, Kavisha Arora1, Nagireddy Putluri2, Efrosini Tsouko3, Taraka R Donti1,
Daniel E Frigo3, Chad J Creighton2, Michael T Lewis2, Arun Sreekumar2, Lee-Jun Wong1,2 and Benny Abraham Kaipparettu1,2.
1
Baylor College of Medicine, Houston, TX and 2Dan L. Duncan Cancer Center, Baylor College of Medicine, Houston, TX.
Body: Triple negative breast cancer (TN BCa) Driver pathway is still poorly understood. Thus, it is important to identify the
underlying mechanisms of triple negative breast cancer progression. Mitochondria, a semiautonomous organelle in cells, play an
important role in cellular energy metabolism, free radical generation, and apoptosis. Mitochondria-nuclear crosstalk is a
bidirectional pathway of communication between mitochondria and nucleus that influences many cellular and organismal
activities. This crosstalk can regulate several oncogenic pathways involved in tumorigenesis. Using transmitochondrial cybrid
(cybrid) technology, we generated different cybrid models under common nuclear background of TN BCa. Mitochondria from cells
of different cancer potential including benign breast epithelium, moderately and highly metastatic breast cancer cell lines were
used to understand cancer mitochondria regulated tumor pathways. Tumor and gene expression analysis suggested that among
different cancer pathways, c-Src signaling pathway is one of the most consistently activated pathways in cybrids with TN BCa
cancer mitochondria. Further analysis in parental cells and other tumor models suggested that autophosphorylation of c-Src is
regulated by mitochondrial tumor characteristics. Our preliminary analysis also suggest that mitochondria targeted drugs are
promising combination therapy for the management of Src-driven TN BCa. This finding is particularly important while considering
the poor response rate observed after single drug therapy with Src family tyrosine kinase inhibitor Dasatinib in unselected TN
BCa patients.

Title: Recurrent ESR1-CCDC170 rearrangements in an aggressive subset of estrogen-receptor positive breast cancers
Jamunarani Veeraraghavan1, Ying Tan1, Xi-Xi Cao1, Jin-Ah Kim1, Xian Wang1, Dean P Edwards1, Alejandro Contreras1, Susan G
Hilsenbeck1, Eric C Chang1, Rachel Schiff1 and Xiao-Song Wang1. 1Baylor College of Medicine, Houston, TX.
Body: The crucial role of gene fusions in epithelial tumorigenesis has been recently appreciated by several milestone
discoveries, but no significant recurrent translocations have yet been found in the vast majority of breast cancers that express the
estrogen receptor (ER). While a majority of ER+ tumors can be effectively treated by endocrine therapy, tumors of the luminal B
subtype are more aggressive and endocrine therapy-resistant. Further, the molecular blueprint of these aggressive tumors is
poorly understood. Thus, characterizing the genetic alterations underlying the more aggressive forms of ER+ breast cancer is of
critical significance in breast cancer management. In this study, by large-scale analyses of breast cancer transcriptome and copy
number alterations, we identified recurrent rearrangements between estrogen receptor gene, ESR1 and its neighbor gene,
CCDC170, which are enriched in the more aggressive and endocrine-resistant luminal-B tumors. Further screening of 200 ER+
breast tumor tissues by RT-PCR identified eight ESR1-CCDC170 positive tumors. The genomic rearrangements underlying these
fusions were verified by genomic PCR and capillary sequencing. CCDC170 encodes a protein with unknown function. The
observed fusion joins the 5-untranslated region of ESR1 upstream to the coding region of CCDC170, enabling the expression of
N-terminally truncated CCDC170 (CCDC170) under the promoter of the ESR1 gene. Consistent with the behavior of luminal B
tumors, forced expression of CCDC170 in ER+ breast cancer cells leads to markedly increased cell motility, invasiveness and
anchorage-independent growth, and reduced endocrine sensitivity in vitro, as well as enhanced xenograft tumor formation and
reduced endocrine sensitivity of the tumors in vivo. When introduced into benign breast epithelial cells, CCDC170 impairs acini
morphogenesis and enhances cell motility and invasiveness. Further, Knockdown of the endogenous ESR1-CCDC170 fusion
variant expressed in HCC1428 cells potently inhibited cancer cell proliferation and migration, which can be rescued by forced
expression of this fusion. Mechanistic studies suggest that CCDC170 engages Gab1 signalosome to potentiate growth factor
signaling and enhance cell motility. The augmented growth factor signaling driven by CCDC170 appears to be sustained even
after withdrawal of serum, and is not affected by endocrine treatment. Together, this study identified neoplastic ESR1-CCDC170
fusions in a more aggressive subset of ER+ breast cancer, which also suggests a new role of ER in breast tumorigenesis by
contributing its promoter to an oncogene.

Title: TP53 alteration in morphologically benign breast tissue in patients with ER/PR negative high grade breast carcinomas:
Marker for early detection of high risk patients for high grade carcinomas?
Xi Wang1, Moritz Stolla1, Todd S Laughlin1, Paul G Rothberg1, Kristin Skinner1 and David G Hicks1. 1University of Rochester
Medical Center, Rochester, NY.
Body: Background: TP53 alterations have been identified in approximately 20% of breast carcinomas, especially in ER/PR
negative high grade carcinomas. It is believed that TP53 alteration is an early event in the carcinogenesis of high grade breast
carcinomas. TP53 alteration could even be identified in morphologically "benign" appearing breast tissues. Nevertheless, the
precursor lesion of high grade ductal carcinoma in situ and/or invasive carcinoma is not well recognized morphologically.
Design: We retrospectively analyzed the TP53 status of 108 cases with high grade (modified Bloom-Richardson grade 3) invasive
ductal carcinoma, 51 cases with non-high grade invasive ductal carcinoma, and 50 cases with benign breast from mammoplasty.
The high grade carcinomas were all estrogen receptor (ER) and progesterone receptor (PR) negative and 96.3% of them were
HER-2 negative. The non-high grade carcinomas were all ER and/or PR positive, 92.2% Her2 negative. Blocks with tumor and/or
benign tissue were selected from each case. Immunohistochemical stain for p53 was performed. TP53 gene sequencing was
performed on selected tumors with inconclusive p53 staining. Further ER and Ki67 immunohistochemical stains were performed
on the blocks with p53 positive benign tissue.
Results: Of the 108 high grade carcinomas, 48 (44%) were positive for TP53 alteration. Seventeen of the 48 (35.4%) cases also
showed focal p53 staining in the adjacent benign appearing tissue. Only one case was negative for TP53 alteration in tumor but
with focal p53 staining in benign tissue. Of the non-high grade carcinomas, 5 (9.8%) were positive for TP53 alteration, one of
which had focal p53 staining in adjacent benign breast tissue. No p53 staining positivity was identified in the mammoplasty
specimens. The p53 staining positive benign glands were ER negative, and did not show an increase in Ki67 labeling index. The
morphology of these cells appeared to be slightly enlarged nuclei, with a mildly increased nuclear/cytoplasm ratio, slightly
irregular nuclear contours, open chromatin, and prominent nucleoli, but still within the limits of "benign" glands.
Conclusion:
TP53 is already altered in "benign" appearing breast glands in patients with high grade breast carcinomas. These glands show
the same ER status as their counterpart carcinomas, but remain inactive with low Ki67 labeling index, indicating further genetic
changes are needed to trigger the unlimited tumor growth and fully malignant transformation. The special morphology and p53
positivity of the "benign" appearing cells could serve as an indicator for future potential risk of TP53 positive high grade breast
carcinoma.

Title: SMAD nuclear staining patterns in inflammatory breast cancer suggest non-canonical TGF-signalling establishing
collective, sheet-like invasion SMAD nuclear staining patterns in inflammatory breast cancer suggest non-canonical
TGF-signalling establishing collective, sheet-like invasion
Melike Marsan1, Gert Van den Eynden2, Patrick Neven1, Ignace Vergote1, Peter Vermeulen2, Luc Dirix2 and Steven Van Laere1.
1
KU Leuven, Leuven, Belgium and 2Antwerp University, Antwerp, Belgium.
Body: Introduction: Inflammatory Breast Cancer (IBC) is an aggressive and highly metastatic form of breast cancer. Recent
expression profiling studies revealed IBC-specific modulation of TGF-signalling. In the present study, we aim to validate these
findings by evaluating nuclear SMAD staining patterns.
Materials and Methods: Immunohistochemistry (IHC) was performed for SMAD2, -3, -4 and -6 on tissue sections from patients
with (N=79) and without (N=133) IBC. Protocols for IHC were established on cell blocks from breast cancer cell lines with and
without SMAD staining identified by western blotting on nuclear protein extracts. A staining score was assigned by multiplying the
percentage of stained cancer cell nuclei by the staining intensity evaluated on a three-scale basis (1=weak, 2=intermediate,
3=strong). Expression data were subjected to unsupervised hierarchical cluster analysis (UHCA) and statistical assessment of
staining differences was done using uni- and multivariate models. Additionally, we looked at the correlation between nuclear
SMAD expression and gene expression in 11 patients. For each gene, we generated a spearman correlation coefficient, and
selected only those genes that were positively correlated with nuclear SMAD expression in 11 patients (p<0.05).
Results: UHCA for SMAD2, -3 and -4 nuclear protein expression data identified two sample clusters downstream of the first
bifurcation. IBC samples were significantly (OR=60.34, P<0.001) enriched in the cluster characterized by increased nuclear
SMAD2 protein expression and absence of both nuclear SMAD3 and -4 protein expression. Univariate analysis revealed that
these staining patterns are significant (all Ps<0.01). Multivariate analysis demonstrated that gain of SMAD2 nuclear expression
and loss of SMAD3 nuclear expression in IBC are unrelated observations and independent of histological grade, hormone
receptor expression, ErbB2 amplification and tumour stage. SMAD correlation analysis showed us that genes positively
correlated with nuclear SMAD2 expression included CD44 and REL, both markers that represent crucial pathways in IBC.
Interestingly, nuclear SMAD3 expression was positively correlated with ZEB1.
Discussion: In line with the expectations, our data show that TGF-signalling indeed differs between samples from patients with
and without IBC. In addition, this study does offer novel insights on IBC biology. First, SMAD2 nuclear staining is gained in IBC in
the absence of its canonical binding partners SMAD3 and -4, suggesting non-canonical TGF-signalling in IBC. Second, it has
been shown that SMAD2 induces an invasive response program in cancer cells without affecting their epithelial morphology, while
SMAD3 accounts for invasion by induction of Epithelial-to-Mesenchymal transition (EMT). Thus our results indicate that IBC cell
invasion occurs in a collective, sheet-like fashion instead of EMT, which is probably responsible for the invasion of non-IBC cells.

Title: Immunohistochemical (IHC) marker discordance between primary breast cancer biopsy and recurrent cancer: Would IHC
testing of the surgical breast or lymph node have altered treatment?
Michele M Gage1, Martin Rosman2, Charles Mylander2, Crystal Tran2, Rubie S Jackson2 and Lorraine Tafra2. 1Walter Reed
National Military Medical Center, Bethesda, MD and 2Anne Arundel Medical Center, Annapolis, MD.
Body: Objective: Based on emerging data on tumor heterogeneity and the evolutionary branching of tumor cells, tumor cells in
the lymph node may represent more virulent clones with the inherent capability of metastasis. IHC discordance from original
cancer diagnosis to recurrence is documented to occur in up to 20% of cases, raising the question if characterization of these
likely more virulent cells would more accurately guide treatment and predict prognosis. Our pilot study sought to determine if
crucial clinical information is gained by IHC testing of the surgical breast or lymph node specimens at the time of initial surgery.
Methods: Using the cancer registry and oncology records, all invasive breast cancers diagnosed after 2001 with subsequent
recurrence were identified. Cases missing all IHC data were disqualified. We then evaluated ER and HER2 of the primary cancer
biopsy and recurrence biopsy to identify discordances. Those with discordances who had surgical breast and lymph node
specimens available were accessed, tested, and evaluated by our breast cancer pathologist.
Results: A total of 128 recurrence cases with partial or complete primary and recurrence IHC data were identified. Of the 95
initially ER positive cases with recurrence IHC available, 13/95 had discordant, or ER negative, recurrence. Additionally, 5/27
initially ER negative tumors, 3/14 initially HER2 positive tumors, and 6/69 initially HER2 negative tumors had discordant
recurrence results. In 128 cases, 27 cases were identified to have ER or HER2 discordance from primary biopsy diagnosis to
recurrence. Of all cases with original surgical breast or positive lymph node specimen available, 9 markers on 7 patients were
performed for our pilot study. One of seven surgical breast specimens and one of two lymph node specimens were concordant
with the recurrence, but not the initial biopsy. The tested surgical breast was ER positive, while the surgical lymph node was
HER2 positive, concordant to their recurrences, but discordant with initial biopsy.
Breast Biopsy
Recurrence Concordant
Recurrence Discordant
ER Positive
98/127
82/95
13/95 (14%)
ER Negative
29/127
24/27
5/27 (19%)
Total ER Discordance
18/122 (15%)
HER2 Positive
19/119
11/14
3/14 (22%)
HER2 Negative
100/119
63/69
6/69 (9%)
Total HER2 Discordance
9/83 (11%)
Conclusion: Tumor discordance of the original cancer biopsy and recurrence is not uncommon. Our pilot study demonstrated that
ER and HER2 discordance occurred in 15% and 11% of cases, respectively. Though our pilot study was limited by small sample
size, we found that IHC testing of the surgical breast and lymph node specimen may provide additional clinical information and
affect management. Of the two cases that had a positive lymph node available, one was HER2 positive and concordant with the
recurrence. Of the seven breast specimens tested, one was ER positive and concordant with the recurrence. Had IHC testing
been performed at that time of surgery, adjuvant treatment management would have been altered. Further testing of our IHC
discordant recurrence patient population will be pursued to investigate the potential benefits of surgical breast and lymph node
IHC testing.

Title: Promotion of lymphangiogenesis by postpartum breast tumor cells
Traci R Lyons1, Sarah A Black1, Bernard W Futscher2, Virginia F Borges1 and Pepper J Schedin1. 1University of Colorado
Anschutz Medical Campus, Aurora, CO and 2Arizona Health Sciences Center, Tucson, AZ.
Body: Multiple studies report poor survival rates in postpartum women with breast cancer[1, 2]. A recent study from our group
has specifically shown that incidence of recurrence and death from breast cancer is increased approximately 3-fold in postpartum
patients compared to nulliparous patients[3]. We define postpartum breast cancers as those diagnosed within 5 years of birth and
as a highly metastatic subset of young womens breast cancer[4]. We have recently shown increased incidence of lymph node
involvement in patients diagnosed less than 2 years postpartum and have utilized tissues from our clinical cohort to show
increased lymphatic vessel density in stage II postpartum patient samples compared to nulliparous. Using xenograft and isograft
murine models we have modeled postpartum breast cancer to show that postpartum tumors are larger[5], exhibit increased
lymphatic vessel density, and more frequently spread to mouse lymph node and lung tissues[5] compared to tumors in nulliparous
mice. In order to determine whether postpartum tumor cells specifically promote lymphangiogenesis, we utilized postpartum
tumor cells ex vivo to show that lymphatic structure formation is promoted via increased postpartum tumor cell secretion of PGE2,
which acts on the EP2 receptor on lymphatic endothelial cells. We have also performed expression analyses on postpartum
tumor cells isolated from our xenograft model to reveal increased expression of COX-2, VEGF-C, and Semaphorin 7a; all three of
these molecules have reported, and possibly interrelated, roles in promotion of lymphangiogenesis[6, 7]. Importantly, both in vivo
inhibition and knockdown of COX-2 decrease postpartum tumor associated lymphangiogenesis in pre-clinical models. These
results, along with our results indicating that COX-2 inhibition during the postpartum period decreases metastasis in our
pre-clinical models, have led us to propose that COX-2 dependent promotion of lymphangiogenesis may facilitate, in part,
metastasis of postpartum tumors.
1. Stensheim, H., et al., Cause-specific survival for women diagnosed with cancer during pregnancy or lactation: a registry-based
cohort study. J Clin Oncol, 2009. 27(1): p. 45-51.
2. Johansson, A.L., et al., Increased Mortality in Women with Breast Cancer Detected during Pregnancy and Different Periods
Postpartum. Cancer Epidemiol Biomarkers Prev, 2011. 20(9): p. 1865-72.
3. Callihan, E.B., et al., Postpartum diagnosis demonstrates a high risk for metastasis and merits an expanded definition of
pregnancy-associated breast cancer. Breast Cancer Res Treat, 2013. 138(2): p. 549-59.
4. Lyons, T.R., P.J. Schedin, and V.F. Borges, Pregnancy and breast cancer: when they collide. J Mammary Gland Biol
Neoplasia, 2009. 14(2): p. 87-98.
5. Lyons, T.R., et al., Postpartum mammary gland involution drives progression of ductal carcinoma in situ through collagen and
COX-2. Nat Med, 2011. 17(9): p. 1109-15.
6. Timoshenko, A.V., et al., COX-2-mediated stimulation of the lymphangiogenic factor VEGF-C in human breast cancer. Br J
Cancer, 2006. 94(8): p. 1154-63.
7. Bender, R.J. and F. Mac Gabhann, Expression of VEGF and semaphorin genes define subgroups of triple negative breast
cancer. PLoS One, 2013. 8(5): p. e61788.

Title: Defining the role of the kynurenine pathway in mediating anoikis resistance in triple negative breast cancer
Thomas Rogers1, Nicholas D'Amato1, Travis Nemkov2, Michael Gordon1, Kirk Hansen2 and Jennifer Richer1. 1University of
Colorado- Anschutz Medical Campus, Aurora, CO and 2University of Colorado-Anschutz Medical Campus.
Body: Background: Anoikis resistance is thought to be a critical trait of metastatic cancer cells, enabling them to leave the
primary tumor and travel through extracellular matrix, intravasate, and survive in the vasculature or lymphatics in transit to a
metastatic site. This is particularly important for the triple-negative breast cancer (TNBC) subtype, which has a peak risk of
recurrence within the first three years post-diagnosis and an increased mortality rate in the first five years as compared to other
subtypes. We performed global profiling of TNBC cells in attached versus forced suspension culture conditions (using poly-HEMA
coated plates) for 24 hours. These data revealed that TNBC cells surviving in suspension upregulate multiple genes involved in
tryptophan catabolism, also known as the kynurenine pathway (KP), including the rate limiting enzyme tryptophan
2,3,-dioxygenase (TDO) and kynureninase (KYNU). A key metabolite of this pathway has been found to activate the aryl
hydrocarbon receptor (AhR), which was also up-regulated in suspended cells.
Hypothesis: We hypothesize that the ability to upregulate the kynurenine pathway (KP) facilitates TNBC cell survival in
suspension and mediates the migratory/invasive potential of TNBC.
Methods: We assessed mRNA and protein levels of TDO2, KYNU and AhR by RT-qPCR and western blot. AhR luciferase
reporter activity, as well as known AhR regulated genes, were measured in suspension compared to adherent conditions. TNBC
cells were treated with small molecule inhibitors of AhR and TDO2. Additionally, secretion of endogenous kynurenine was
measured by high performance liquid chromatography (HPLC). Purified kynurenine was added to rescue AhR activity following
TDO2 inhibition. Finally, anoikis sensitivity and migratory potential were measured following pharmacological inhibition of TDO2
and AhR.
Results: Relative mRNA levels of TDO2, KYNU, and AhR increase 9, 7, and 2 fold respectively in suspended TNBC cells
compared to adherent conditions (P<0.0001). Estrogen receptor positive breast cancer cells lines do not significantly upregulate
these genes. AhR reporter activity (P<0.0001) and nuclear localization increase in suspended TNBC cells. Additionally, AhR
reporter activity (P<0.0001) and AhR target gene expression (CYP1A1, CYP1B1: P<0.0001) decreased in TNBC cells treated
either AhR antagonist (CH223191) or selective TDO2 inhibitor (680C91). Conversely, when purified kynurenine was added to
TNBC cells in vitro, AhR activity increased (P<0.05). Furthermore, kynurenine secretion, as measured by HPLC, increased 2.5
fold in suspended TNBC cells and this increase in kynurenine was reduced by addition of 680C91 (P<0.0001). Finally, targeting
TDO2 or AhR increased anoikis sensitivity and decreased migration of attached and suspended TNBC cells.
Conclusions: Collectively, these results suggest that the kynurenine pathway may play a critical role in metastatic TNBC. Further
mechanistic studies will focus on how the kynurenine pathway is mediating these tumorigenic properties either through the de
novo synthesis of NAD+ and/or activation of AhR by kynurenine. Targeting the kynurenine pathway in the clinic may provide a
therapeutic strategy to reduce TNBC mortality rates.

Title: Extramedullary hematopoiesis aids initiation of cancer metastasis
Kenji Yokoi1, Tomonori Tanei1, Megumi Kai1, Yuki Saito1, Yan Ting Liu1 and Mauro Ferrari1. 1Houston Methodist Research
Institute, Houston, TX.
Body: Despite the recent advances in treatment of primary tumors, metastatic disease is resistant to current therapies and
remains the primary cause of cancer-related death. Therefore, prevention and improved therapy of cancer metastasis are the
ultimate goals of cancer research.
Although tumor cells are the driving force of metastasis, novel findings suggest that the tumor microenvironment also plays a key
role. Microenvironments distant from the primary tumor are primed for future metastatic growth by the recruitment of bone
marrow-derived myeloid cells, creating "pre-metastatic niches", and ready for arrival of circulating cancer cells.
The number of platelets is often increased in the late stage of cancer patients and experimental evidences show that platelets
facilitate tumor metastasis. Within the circulatory system, platelets surround and guard tumor cells from immune elimination and
produce various kinds of growth factors and cytokines to aid establishment of metastasis. But the role of platelets within the
pre-metastatic niches has not yet been explored, particularly with respect to their ability to aid circulating cancer cells to arrest,
survive and grow.
In our study, 4T1 murine breast cancer cells, implanted into mammary fat pad (mfp) of mice were spontaneously metastasized to
the lung and liver around 6 weeks after the implantation. Interestingly, splenomegaly was found in these mice bearing 4T1 tumor
in the mfp. Pathological and immunofluorescence analysis of the spleen revealed extramedullary hematopoiesis including
megakaryopoiesis. Consequently, platelets accumulated into the other organs of the tumor bearing mice as early as 2 weeks after
the implantation of 4T1 cells to the mfp. Interestingly, platelets accumulated into lung and liver, but not to the other major organs.
To validate the effect of extramedullary hematopoiesis, 4T1 cells expressing luciferase (4T1-luc) were implanted into the mfp of
normal mice or mice resected with spleen in advance. Then, development of spontaneous metastasis was monitored by
luminescence imaging of the live mice. There was a significant reduction in the number of metastasis in the mice without spleen
as compared to that in the normal mice, indicating crucial role of extramedullary hematopoiesis in the spleen in spontaneous
metastases. Next, to evaluate biological effect of accumulated platelets into the lungs, normal mice or mice bearing 4T1 in the
mfp (2 weeks after the implantation) were injected intravenously with 4T1-luc cells through tail vein. The amount of the arrested,
survived and growing tumor cells in the lung was imaged and quantified using luminescence imaging of the live mice. There was
a significant increase in these parameters in the mice bearing 4T1 tumor in the mfp as compared to those in the normal mice.
These data strongly indicates that accumulated platelets aid circulating tumor cells to arrest, survive and grow at the lungs.
Therefore platelets accumulated into the secondary sites can serve as the pre-metastatic niches. To validate these findings, mice
bearing 4T1 cells in the mfp will be pre-treated with CD42 antibodies to delete circulating platelets or glycoprotein GP IIb/IIIa
inhibitors which can reduce binding of activated platelets to endothelial cells.


Title: No Show

Title: Galectin-7 increases resistance of breast cancer cells to drug-induced apoptosis and promotes tumor escape by killing T
cells
Yves St-Pierre1, Andre-Anne Grosset1, Marilyne Labrie1, Donald Gagn1, Maria-Claudia Vladoiu1, Louis Gaboury2 and Nicolas
Doucet1. 1INRS-Institut Armand-Frappier, Laval, QC, Canada and 2Institute for Research in Immunology and Cancer, Montreal,
QC, Canada.
Body: Resistance to apoptosis induced by anti-cancer drugs is a major obstacle for the treatment of aggressive forms of breast
cancer. Galectin-7 was recently shown to be specifically expressed in basal-like and HER-2 positive but not in luminal subtypes of
human breast cancer. Galectin-7 also increases the metastatic behavior of breast cancer cells since overexpression of this
protein in poorly metastatic breast cancer cells increases their ability to metastasize to the bone and to the lung. This pro-tumoral
activity of galectin-7 in breast cancer is surprising since galectin-7 had been previously associated with activation of p53 and
breast cancer cells often harbor a transcriptionally inactive form of p53. We have recently reconcile this apparent contradiction by
showing that elevated levels of galectin-7 in breast cancer cells occurs via a gain-of-function mechanism of mutant p53. In
p53-null breast cancer cells, we have shown that C/EBP-2 (also known as LAP2), the most transcriptionally active of the
C/EBP isoforms, is responsible for the upregulation of galectin-7. How galectin-7 contributes to the progression of breast cancer
remains unclear. Up to now, regulation of apoptosis by intracellular galectins has been largely attributed to their ability to
translocate to mitochondria, possibly following their interaction with bcl-2. Here, we report that a mutant of galectin-7 that is
unable to translocate to mitochondria induces resistance of human breast cancer cells to apoptosis induced by etoposide or by
hypoxia-mimicking conditions. Surprisingly, this mutant and the wild-type form of galectin-7 bind equally well to bcl-2 in vitro and
in vivo. Interestingly, both forms decreases the translocation of p53 to the nucleus and reduce the expression of p21 following
treatment with doxorubicin. We also found that galectin-7 was released by breast cancer cells and that recombinant galectin-7
increased apoptosis of monocytes, T CD4+ and T CD8+ cells. This suggests that galectin-7 contributes to the establishment of an
immunosuppressive tumor microenvironment by killing helper and effector T cells. Taken together, these results challenges the
current paradigm that mitochondrial galectins are important for resistance to apoptosis and call for a greater focus on the role of
galectin-7 in breast cancer. They also indicate that targeting both cytosolic and extracellular galectin-7 could improve the efficacy
of anti-cancer drugs for the treatment of aggressive forms of breast cancer.

Title: Transcription factor Fra-1 modulates the adhesive properties of breast cancer cells contributing to a metastatic phenotype
Leticia Oliveira-Ferrer1, Melanie Krschner1, Vera Labitzky2, Daniel Wicklein2, Volkmar Mller1, Udo Schumacher2, Karin
Milde-Langosch1 and Christine Schrder2. 1University Medical Center, Hamburg, Germany and 2University Medical Center
Hamburg-Eppendorf, Hamburg, Germany.
Body: Fra-1, a component of the transcription factor AP-1 family, has been found to be overexpressed in carcinomas with high
metastatic potential such as in triple-negative breast cancer. The effect of Fra-1 on the morphology, motility and invasive potential
of breast cancer cells has been previously described. Since tumor cell adhesion plays an essential role in the metastatic process,
especially for extravasation from blood vessels, we investigated the influence of Fra-1 on breast cancer cell interactions with the
endothelium.
Using the human breast cancer cell lines MCF7 (weakly invasive, estrogen receptor (ER)-positive) and MDA MB 231 (strongly
invasive, ER-negative) with stable Fra-1 overexpression, we performed dynamic cell-flow adhesion assays on surfaces coated
with E-selectin or with human pulmonary microvascular endothelial cells (HPMEC). Our analyses revealed increased adhesion of
Fra-1 overexpressing MCF7 cells to E-selectin but also to activated endothelial cells, whereas the per se highly invasive and
Fra-1-positive MDA MB 231 cell line showed enhanced cell rolling and tethering on E-selectin and endothelium-coated surfaces,
but no increased firm adhesion after Fra-1 overexpression. These different behaviors correspond to an up-regulation of various
adhesion-related proteins such as CD44, Integrin 5 and CEACAM6 in Fra-1 overexpressing MCF7 cells measured by microarray
analysis and flow cytometry in comparison to weaker differences in the expression of adhesion molecules found in the Fra-1
overexpressing MDA MB 231 cell line. In line with these results and based on cDNA microarray data of breast cancer patients
(n=175) high Fra-1 expression significantly correlates with shorter overall survival and higher rate of lung metastasis in
ER-positive breast cancer patients (n=130), but has no impact on the prognosis of patients with ER-negative tumors.
Thus, in addition to its pro-invasive and pro-migratory effect, Fra-1 might influence the metastatic potential of breast cancer cells
by changing the expression of adhesion molecules resulting in increased adherence to endothelial cells under flow conditions.

Title: Loss of Rab1B promotes triple-negative breast cancer metastasis by activating TGF-/Smad signaling
HongLin Jiang1 2, Hefen Sun1 2 and Wei Jin1 2. 1Fudan University Shanghai Cancer Center, Shanghai, China and 2Shanghai
Medical College, Fudan University, Shanghai, China.
Body: Background: Triple-negative breast cancer (TNBC) is a highly aggressive tumor subtype associated with poor prognosis.
The onset of metastasis in organs such as the lung,bone and brain is a major cause of mortality in TNBC patients. Thus,
identification of novel targets for the treatment of triple-negative breastcancer is urgent for improved outcomes in patients with this
disease.
Methods and Results:Here we conducted quantitative proteomic analyses in breast cancer cell lines, which include a normal,
primary tumor and a metastatic tumor that were isolated from a single patient. Stable isotope labeling of amino acid in cell culture
followed by LC-MS/MS analysis was performed and deregulated proteins were identified using statistical analysis. We identified
for the first time that Rab1B, a member of RAS oncogene family, was down-regulated in high metastatic human breast cancer
cells. In a panel of breast tumor tissues, Immunohistochemical analysis demonstrated an inverse correlation between metastatic
propensity and the expression of Rab1B. Compared with primary tumors, Rab1B expression was notably decreased in lymph
node metastases. Low Rab1B expression also correlated with poorer survival in triple-negative breast cancer patients. Through in
vitro assays, we confirmed that silencing of Rab1B expression by RNAi in MDA-MB-231 and MDA-MB-468 breast cancer cells,
can potentiate the proclivity to metastasize in Transwell assay and enhance lung colonization by tail vain injection in mice. In a
reverse-complimentary approach, we determined that elevated Rab1B expression in highly metastatic breast cancer cell lines
(MDA-MB-231HM and MDA-MB-231Bo) can suppress the ability of invasion and metastasis in vitro and in vivo. Mechanistically,
we establish that the metastatic behavior strongly correlates with increased phosphorylated Smad3 levels and overexpression of
TGF- type I receptor (TRI). Conversely, restoration of Rab1B can inhibit the activation of this pathway in MDA-MB-231HM and
MDA-MB-231Bo cells.
Conclusion:By utilizing a series of mammary tumor cell lines, animal models, and clinically sourced tissues, we demonstrate that
Rab1B acts as a metastasis suppressor in triple-negative breast cancer through regulating TGF-/Smad signaling pathway.

Title: Effects of JAK pathway inhibition in pre-clinical models of breast cancer bone marrow metastases
Alessia Bottos1, Jason Gill1, Thomas Radimerski2, Alexandar Tzankov3, Aleksandra Wodnar-Filipowicz1 and Nancy E Hynes1.
1
Friedrich Miescher Institute for BioMedical Research, Basel, Switzerland; 2Novartis Institutes for BioMedical Research, Basel,
Switzerland and 3University Hospital Basel, Basel, Switzerland.
Body: Bone marrow (BM) metastases are an important problem in metastatic breast cancer. Metastatic-relapse occurs even
decades after the first diagnosis and can show different clinical features compared to the primary disease. The cross-talk between
cancer cells and the microenvironment influences tumor behaviour and impinges on therapeutic efficiency.
To uncover new targets for the inhibition of BM metastases, we performed a transcriptional analysis of BM stromal cells using
in-vivo models of breast cancer bone metastases. We identified IL6, and its downstream JAK kinases pathway, as major
up-regulated signals in the bone stroma of tumor-bearing mice. Moreover, we found that a significant proportion of BM
metastases from breast cancer patient biopsies are positive for pSTAT3 staining, attesting to activation of JAK-STAT signaling.
These results provide a rational to investigate the therapeutic potential of a JAK inhibition in bone metastases. We evaluated the
effect of two JAK inhibitors: the BSK805, a JAK2 inhibitor, and ruxolitinib a JAK1-JAK2 inhibitor currently in clinical trials for
metastatic breast cancer. In pre-clinical models of breast cancer metastases (the triple negative MDA-MB231 SCP1833 and the
ER+ EO771), treatment with the JAK inhibitors decreased pSTAT3 signaling in the bone, but surprisingly increased metastastic
tumor burden.
Considering the importance of the JAK-STAT pathway as a therapeutic target, we are investigating the possible bystander effects
of blocking JAK kinases. Analysis of the BM stoma showed that JAK inhibition decreased TRAP+ osteoclasts and affected the
mesenchymal compartment, with a significant increase in the BM fat compartment in tumor-bearing animals. The reasons for and
the consequences of these changes in the bone stroma are under investigation. As the JAK-STAT signaling pathway is also a key
regulator of the hematopoietic compartment, we are studying how JAK inhibitors influence the immune response in tumor bearing
mice. Further dissection of the immunomodulatory effect of JAK inhibition and how this influences tumor dissemination and
expansion is warranted.

Title: Melatonins inhibitory effect on breast cancer metastasis mediated by ROCK-1
Debora C Zuccari1, Thaiz F Borin1, Ali S Arbab3, Lvia C Ferreira2, Marina G Moschetta1, Gustavo R Martins1, Larissa B Maschio2,
Vanessa A Fabri1 and Verena B Coimbra1. 1Faculdade de Medicina de So Jos do Rio Preto, FAMERP, Sao Jose do Rio Preto,
So Paulo, Brazil; 2Universidade Estadual Paulista (UNESP/IBILCE), Sao Jose do Rio Preto, So Paulo, Brazil and 3Cancer
Cener, Georgia Regents University, Augusta, GA.
Body: The metastasis occurrence, an important prognostic factor, depends on peculiarities such as cellular invasiveness and cell
migration, mechanisms controlled by regulatory and effector molecules such as Rho-associated kinase protein (ROCK-1). An
increased expression of this protein promotes tumor growth and metastasis, a mechanism which can be restricted by the use of
the effectors inhibitors. Melatonin, a hormone secreted by the pineal gland, has shown oncostatic action and anti-metastatic
effects and can modulate the ROCK-1expression. The objective of this study was to investigate the anti-metastatic response
mediated by ROCK-1 and through the action of melatonin and its specific inhibitor (Y27632) in vitro and in vivobreast cancer
models. Cells from metastatic (MDA-MB-231) and non-metastatic (MCF-7) breast cancer lines were treated with melatonin and
Y27632. Cell viability was verified by MTT assay, cell migration/invasion assays in Boyden chamber, ROCK-1 protein and gene
expression by western blot and quantitative real time PCR, respectively. In addition, the in vivometastasis study was performed
using female athymic nude mice induced by injection of 2x105 MDA-MB-231 viable cells by tail veinfor lung metastasis and
byintracardiac for bone metastasis, during 3 weeks. The animals were treated with melatonin and Y27632 for 2 weeks. The
metastasis developments were evaluated by single photon emission computed tomography (SPECT). Treatment with melatonin
reduced cell viability and migration of both cell lines (p<0.05). The use of melatonin and Y27632, in association or not, decreased
the ROCK-1 protein expression in metastatic cells, not significantly altering its expression in the non-metastatic line (p>0.05). An
statistically significant reduction of ROCK-1 gene expression was observed in all treatment groups (p<0.05). ROCK-1
downexpression was more efficient in the group with associated treatments for both lines (p < 0.05). In vivoSPECT images
showed multiple foci in the lungs (on Tc-99-tetrofomin images) and in the vertebrae (on Tc-99-MDP images). The numbers of
"hot" spots were significantly higher in lung metastasis of control animals compared to treated groups. Semi quantitative analysis
showed significantly lower activity in animal lungs that received treatment (p<0.05). The bone metastasis did not show difference
between control and treatment groups. Melatonin and Y27632 are effective drugs of metastatic breast cancerin vitro treatment,
confirming their effects in decreasing cell viability, invasion, migration and protein expression of ROCK-1 in these cells. In vivo,
melatonin and Y27632 treatments seem to be effective reducing lung metastasis but not bone metastasis. Melatonin, in particular,
appears to be more effective when combined to ROCK-1 inhibitor.
Support: FAPESP.

Title: N-glycosylation changes in tumour cells are associated with vascular invasion and metastasis in breast cancer patients
Karin Milde-Langosch1, Tosca Karius1, Dina Schtze1, Harriet Wikman2, Isabell Witzel1, Leticia Oliveira-Ferrer1, Volkmar Mller1
and Udo Schumacher3. 1Hamburg University Medical School, Hamburg, Germany; 2Hamburg University Medical School, Institute
of Tumor Biology, Hamburg, Germany and 3Hamburg University Medical School, Institute of Anatomy and Experimental
Morphology, Hamburg, Germany.
Body: Background: In a recent study based on cDNA microarray data of breast cancer patients, we could show that the mRNA
expression of some glycosylation enzymes involved in synthesis and trimming of N-glycansis significantly associated with
metastasis and recurrence in breast cancer patients suggesting an important role of N-glycosylation in tumour progression
(Milde-Langosch et al., Breast Cancer Res Treat 2014). The aim of the present study was to further investigate this issue on a
protein and glycoprotein level (1.) by western blot analysis of selected glycosylation enzymes and (2.) by lectin histochemistry
using lectins which bind to O- and/or N-glycans.
Methods: In order to confirm the role of two selected glycosylation enzymes, ribophorin (RPN1) and mannosidase (MAN1A1) on
protein level, we analyzed their expression in a cohort of 196 breast cancer samples by western blot analysis. In addition, we
studied tumour cell binding of various lectins, which recognize diverse O- or N-glycans (UEA1, HPA, PNA, GNA, PHA-L), using a
tissue microarray with up to 293 evaluable breast cancer samples. These results were correlated with recurrence-free survival
(RFS) and overall survival (OS) and data on vascular and lymphatic invasion.
Results: By western blot analysis, the negative prognostic significance of ribophorin (RPN1), a key enzyme of N-glycosylation,
was corroborated on a protein level. For MAN1A1 which is involved in trimming of N-glycan structures, two isoforms with different
molecular weight could be detected on western blots, showing contradictory associations with OS. Protein levels of RPN1 and
total MAN1A1 showed significant correlations to the respective mRNA expression data. By lectin histochemistry, we found
significant correlations of lectin binding to lymph node metastasis, vascular invasion and lymphangiosis in breast cancer samples.
Binding of PHA-L and GNA (binding to N-glycans) and PNA (binding to O- and N-glycans) was significantly associated with
vascular invasion, whereas PNA and PHA-L binding correlated with lymph node involvement. In addition, PNA and HPA reactivity
was associated with microscopically detected lymphangiosis. PNA binding also correlated with shorter RFS and OS.
Conclusion: While the role of O-glycans in breast cancer metastasis has already been described, our data point to an additional
strong impact of N-glycosylation on metastasis and progression of mammary carcinomas.

Title: Investigating the role of Duffy antigen receptor for chemokines (DARC) isoforms in breast cancer metastasis
Rupali Hire1, Andrea Walens1, Brianna Bennett1, Kauthar Mumin1, Hailey Campbell1, Michael Lou2, Michele Monteil2, Elizabeth
Howerth3 and Melissa Davis1. 1University of Georgia, Athens, GA; 2Health Sciences Campus, University of Georgia, Athens, GA
and 3College of Veterinary Medicine, University of Georgia, Athens, GA.
Body: Duffy Antigen Receptor for chemokines (DARC) is known to regulate the biological activities of chemokines. It has been
implicated in the progression of various diseases including different types of cancer, affecting survival, proliferation and
metastasis. Genetic variants in the DARC genes are strongly associated with African ancestry and are linked to malaria
resistance. There are at least two isoforms of DARC, which have not been appreciably investigated in these contexts, that may
have differing chemokine activity functions. To fill this knowledge gap, we have been characterizing the impact of DARC gene
variants on DARC isoform expression in specific tissue types in order to define relationships between the polymorphic DARC and
its effect as an immune regulator in breast cancers. We hypothesize that DARC variants, which are known to impact expression
of the gene on erythrocytes, also are associated with functional activity related to DARCs control of metastatic potential in breast
cancer. Specifically, we have determined that one DARC polymorphism which is well defined in ancestral groups (i.e. Duffy null rs2814778; -541T>C) is associated with differential expression of DARC isoforms in epithelial cell types, including lymphoblast
and breast tissues.
We investigated the erythroid silencing Duffy Null allele of DARC using DNA sequencing and quantitated DARC isoform
expression by real time PCR in breast cancer cell lines, ancestral cell lines and breast cancer tumor tissues. We will show that
DARC isoforms have significantly variant expression among ancestral groups that is associated with the Duffy Null allele. We will
also show that certain types of breast tumors have variable DARC expression that is also associated with this genotype.
Additionally, we visualized the subcellular localization of DARC and known interacting chemokines breast cancer cells with IHC
and IF confocal microscopy.
Specifically, our results indicate breast tumors differentially express DARC in which may be affecting the chemokine pool and
immune response within the tumor and its microenvironment. In our Hapmap ancestral populations, all DARC genotypes
expressed the gene in lymphoblast, with significantly different levels of isoforms among African, African American and European
individuals. Breast cancer cells also show different levels of specific DARC isoform mRNA levels and distinct levels and patterns
of binding chemokines, related to tumor subtype origins (i.e. ER status).
Based on these findings we conclude that DARC isoforms have differential expression levels in breast cancer cells within tumor
and non-tumor regions that may drive both pro and anti tumorigenic events with varying chemokine binding or localizations during
tumorigenesis of specific breast cancer subtypes. Different levels of isoforms in epithelial cells may contribute to its essential
function as atypical chemokine receptor. We hypothesize that immune responses in breast cancer patients with Duffy Null
genotypes will differ and may express more severe metastasis due to altered epithelial regulation of alternate DARC isoforms.

Title: Pathway activation mapping of metastatic breast cancer identifies potential organ-specific signatures: Implications for
patient stratification to targeted treatment
Mariaelena Pierobon1, Stephen P Anthony2, K Alex Reeder1, Nicholas J Robert3, Donald W Northfelt4, Mohammad Jahanzeb5,
Linda Vocila6, Julia D Wulfkuhle1, Bryant Dunetz7, Lance A Liotta1 and Emanuel F Petricoin1. 1George Mason University,
Manassas, VA; 2Evergreen Hematology & Oncology, Spokane, WA; 3Virginia Cancer Specialists, Fairfax, VA; 4Mayo Clinic,
Scottsdale, AZ; 5University of Miami, Deerfield Beach, FL; 6Translational Drug Development, Scottsdale, AZ and 7Sideout
Foundation, Fairfax, VA.
Body: Background: The development of distant metastases is the strongest prognostic factor associated with cancer mortality.
Customization of treatment based on molecular profiling of a patients primary tumor has yielded promising results and has
opened a new paradigm for treating patients with advanced disease. Nevertheless profiling the primary tumor may not reflect the
actionable molecular drivers of the metastatic lesions. In the last few years a large number of genomic and proteomic studies
have demonstrated that at the molecular level metastatic lesions significantly differ from matched primary tumors. The impact of
metastasis site organ microenvironments on breast cancer primary tumors has been only partially explored. Since most targeted
therapies work by modulating aberrantly-activated protein kinase signaling, the aim of this study was to utilize reverse phase
protein arrays (RPPA) to explore whether metastatic lesions derived from different patients, but invading the same host target
organ showed similarities in their signaling architecture and presented with organ-specific targetable signatures. Methods: Snap
frozen material collected from 14 metastatic breast cancer patients enrolled in a prospective phase II trial ("Side Out II") were
used for this analysis. Sites of metastasis were: liver (n=7), skin/chest wall (n=4), and lung (n=3). All samples were subjected to
Laser Capture Microdissection (LCM) and RPPA to measure the activation/phosphorylation levels of 12 FDA approved drug
targets and linked substrates. Results: Among the 7 patients with liver metastases, p70S6K, c-Abl, ERK 1/2 were highly
phosphorylated in 5 cases (71.4%), ErbB2/HER2 in 4 cases (57.1%), and AKT in 3 cases (42.8%). On the contrary, of the 4
patients with skin/chest wall lesions 3 showed high activation of ErbB2/HER2, ErbB3/HER3, and ERK 1/2 (75.0%) and 2 of EGFR
(50.0%). Of note, none of the skin/chest wall lesions showed activation of AKT and only 1 case presented with activated p70S6K
(25.0%). Lung metastasis showed an overall low activation of all 12 analytes measured. Discussion: Although this data is based
on a relatively small number of samples and needs further validation, our initial analysis revealed potentially important trends that
could have an impact in the prioritization and selection of targeted agents for metastatic breast cancer trials. We found that the
distribution of the 12 FDA approved drug targets and downstream substrates varies between different metastatic sites. Thus, the
emergent hypothesis from this work is that breast cancer patients with liver metastases may be more susceptible to AKT-mTOR
directed targeted therapeutics and patients with skin/chest wall metastasis may benefit from therapies that target EGFR and/or
downstream ERK signaling.

Title: Modeling breast cancer metastasis in the mouse via measurement of circulating tumor cells
Jennifer L Gorman1, Michael Parsons1, Eldad Zacksenhaus2, Sean Egan3, Martin Chang4 and Jim R Woodgett1.
1
Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada; 2Toronto General Research Institute
University Health Network, Toronto, ON, Canada; 3Program in Developmental and Stem Cell Biology, The Hospital for Sick
Children, Toronto, ON, Canada and 4Mount Sinai Hospital, Toronto, ON, Canada.
Body: Objectives: Advances in the detection of circulating tumor cells (CTC) in the blood of metastatic breast cancer patients
have indicated that patients with greater than 5 CTC/7.5mL of blood have poorer prognosis; however, little is known about the
biology of these cells and what characteristics are critical for breast tumor cell dissemination. We are developing and evaluating
several mouse models of breast cancer to characterize important markers that distinguish CTCs from other cell types in the blood
as well as to assess whether silencing of various metastasis supporting genes leads to reductions in CTC counts and formation of
metastatic tumors.
Methods: This study involves two separate approaches to characterize CTCs. The first involves xenograft models using human
breast cancer cells injected into the fat pad of immunocompromised mice. These cells are tagged with both mCherry, for
detection in blood, and luciferase, to track metastatic spread of the primary tumor cells to other sites via intravital imaging. Blood
was collected from tumor bearing mice at end point, processed and stained for characterized CTC markers. The second
approach involves endogenous mouse models to characterize CTC counts and relevant markers in a variety of metastatic and
non-metastatic murine breast cancer models using negative depletion to remove contaminating normal and hematopoietic cells.
In both approaches CTCs have been assessed using the Imagestream by Amnis.
Results: Initial xenograft experiments with MDA-MB-231 cells injected into the mouse mammary gland fat pad revealed a low
concentration of mCherry+/cytokeratin+/Hoechst+/CD45- cells in the blood of mice without confirmed metastatic dissemination.
Pilot studies on endogenous tumor models have demonstrated the utility of our negative depletion approach to remove
contaminating hematopoietic cells as well as the detection of a population of EpCAM+/Hoechst+/CD45- cells.
Conclusions: Initial experiments have provided feasibility for detailed characterization of mouse CTCs, which will be presented.
This work complements our study being carried out on characterizing CTCs in metastatic breast cancer patients. Increasing
understanding of the biology of circulating tumor cells and relating their characteristics to those of both the primary and metastatic
tumors may reveal mediators common and functionally important to all 3 tumor cell populations and may therefore serve as better
targets for treatments that are effective against both the primary and metastatic tumors.

Title: SSA, a novel sulindac sulfide derivative, inhibits tumor cell growth and invasion in breast cancer
Bin Yi1, Xiangling Feng1, Hong Chang1, Ruixia Ma1, Xiaoguo Zhang1, Gary A Piazza1 and Yaguang Xi1. 1University of South
Alabama, Mitchell Cancer Institute, Mobile, AL.
Body: Nonsteroidal anti-inflammatory drugs (NSAIDs), such as sulindac sulfide (SS), have been reported for striking
chemopreventive activities in various types of human malignances, including breast cancer. However, the toxicities related to
cyclooxygenase (COX) inhibition resulting in suppression of physiologically important prostaglandins limit their clinical use for
chemoprevention in human cancer. We recently developed a novel amine derivative of SS named as SSA, which lacks inhibitory
effect on COX-1 or -2, yet shows 10 times greater suppressive effect than SS on growth of a panel of breast cells. Moreover, SSA
treatment at sub-cytotoxic concentration (4M) for 36 hr can significantly inhibit both migration and invasion of highly aggressive
breast tumor MDA-MB-231cells. To understand the molecular mechanism accounting for this activity, we examined 4 oncogenic
miRNAs, including mir-17-92, mir-9, mir-10b, and mir-21 that were previously reported by our group to be associated with SS
anti-invasive activity in breast and colorectal cancer. We found SSA could significantly down-regulate these miRNAs; whereas
their forced expression was able to counteract the anti-invasive activity of SSA in MDA-MB-231 cells. There results imply that
significance of SSA with non-COX inhibitory properties in suppression of tumor cell invasion could provide novel insights into
development of safer and more effective strategies for prevention of breast cancer progression and metastasis. In addition, after
inducing mobility of non-invasive breast MCF-7 cells by using TGF-1, we treated these invading cells with SSA and found that
their mobility was significantly decreased. These results support anti-invasive activity of SSA in human breast cancer cells and
inhibition of TGF-1 as well as oncogenic miRNAs may be responsible for the mechanistic basis by which SSA prevents breast
tumor cell invasion.
This study is supported by an American Cancer Society Research Scholar Grant (RSG-13-265-01-RMC) and NIH/NCI R21
Grants (5R21CA160280 and 1R21CA182754) to Yaguang Xi.

Title: Global analysis of the transcriptome in matched primary and metastatic tumours defines ER specific gene alterations
Damir Vareslija1, Ailis Fagan1, Jean McBryan1, Patrick Buckley2, Michael Farrell2, Arnold DK Hill1 and Leonie Young1. 1Royal
College of Surgeons in Ireland, Dublin, Ireland and 2Neuropathology, Beaumont Hospital, Dublin, Ireland.
Body: The incidence of brain metastasis (BM) among women with metastatic breast cancer (MBC) is currently on the rise, and it
is estimated that between 10-30% of patients diagnosed with breast cancer may also develop BM (Weil 2005). BM is considered
a late complication in the clinical setting and is an infrequent occurrence even in mice studies that readily metastasise to other
organs (Valiente 2014). While pre-menopausal status and negative estrogen receptor (ER) status are seen as main risk factors
(Gil-Gil 2013), it is now recognised that BM is emerging as an increasing clinical problem across the various MBC subtypes.
BM patients have very poor overall survival and as a result it is crucial to identify the factors that promote the survival and
adaptation of breast cancer cells to allow them to colonise distant sites such as the brain. The ability to predict the BM potential
could be of great clinical importance and in this study we decided to investigate the mechanisms underlying BM across the
various MBC subtypes.
To our surprise, in a cohort of breast cancer patients we found a comparable number of ER positive and ER negative BM. We
established that ER positive patients (n=10) had a substantially longer disease latency period (median disease free survival
(DFS) 36 months, range 8-85 months) in comparison to ER negative patients (n=8, HER2 positive, n=2, triple negative, n=6),
(median DFS 24 months, range 9-60 months). To understand and elucidate the gene alterations required for successful
colonisation of the brain we undertook RNAseq of sequential primary and brain metastatic tumours from ER positive (n=4) and
ER negative (n=3) breast cancer patients. Significant elevations in gene expression were observed in the brain metastatic
tumours in comparison to the matched primary. Though a large degree of patient heterogeneity was observed, common
ER-specific metastatic pathways were identified.
This study highlights the requirement of unique gene sets for the colonisation to the brain and that functional characterisation of
the differentially expressed genes will enable the identification of novel molecular targets for prevention and treatment of breast
cancer BM.

Title: Neutrophil elastase modulates breast cancer progression by fostering collective cell detachment and tumor emboli
dissemination
Caterina Marchio1, Laura Annaratone1, Davide Balmativola1, Maria Stella Scalzo2, Stefania Bolla2, Silvia Grasso3, Isabella Russo3,
Federica Fusella3, Luigia Macr2, Mara Brancaccio3, Paola Defilippi3 and Anna Sapino1. 1University of Turin, Turin, Italy; 2Azienda
Ospedaliera Citt della Salute e della Scienza, Turin, Italy and 3University of Turin, Italy.
Body: Proteases constitute a large family of enzymes involved in several processes, including degradation and remodeling of the
extracellular matrix to drive dissemination of cancer cells into adjacent tissue. Within this large family, the serine protease
neutrophil elastase (NE) has been proven clinically meaningful in breast cancer. Indeed, high levels of NE correlate with poor
outcome and endocrine resistance in breast cancer patients.
In an experimental model it has been demonstrated that estrogen receptor positive (ER+) MCF7 breast cancer cells grow in
suspension as 3D-spheroids in NE-addicted medium, thus resembling the micropapillae of a human breast cancer characterized
by high propensity to metastasize, i.e. the micropapillary carcinoma. We hypothesized that NE may produce disarrangement of
tumor cell adhesion to the substrate fostering neoplastic lymphovascular invasion (LVI) and metastasis. ER+/E-Cadherin
(E-CAD)+/HER2- (MCF7, T-47D, ZR-75-1), ER+/E-CAD+/HER2+ (BT-474), ER-/E-CAD-/HER2+ (SK-BR-3) and
ER-/E-CAD-/HER2- (MDA-MB-231) cells were grown with serine proteases (NE, cathepsin-G), hyaluronidase and collagenase.
NE and cathepsin-G led to 3D-spheroid formation of ER+/E-CAD+ cells only. In 3D-spheroids from MCF7 cells grown with NE the
luminal Epithelial Membrane Antigen (EMA) lined the external border of cell clusters, which faced cancer associated fibroblasts in
co-cultures experiments, thus recapitulating the inverted polarity of MPCs. MCF7 3D-spheroids were tamoxifen resistant. Injection
of NE in tumors of MCF7 cells in SCID mice triggered neoplastic cluster detachment, micropapillae, vascular emboli similar to
3D-spheorids and metastases. In a cohort of human breast carcinomas with LVI the MCF7-3D-spheroid-alike pattern was the
most prevalent in tumor emboli. Immunohistochemical NE expression was mainly detected in polymorphous neutrophilic
granulocytes (PMNs) within vessels and in the stroma and PMNs were significantly higher in breast carcinomas with LVI. In fully
developed metastases within lymph-nodes, which reverted the EMA expression as in the primary tumor, no NE+ PMNs were
observed. Our results may explain why high levels of NE negatively act on patient prognosis by creating a favorable environment
for breast cancer invasion and metastasis.

Title: Expression of S100A14 promotes cancer cell invasion and metastasis and is associated with poor prognosis in breast
cancer
Takashi Sugino1, Mizuko Tanaka2, Takuma Oishi1 and Takashi Nakajima1. 1Shizuoka Cancer Center Hospital, Nagaizumi-cho,
Sunto-Gun, Shizuoka Prefecture, Japan and 2Fukushima Medical University School of Medicine, Fukushima, Fukushima
Prefecture, Japan.
Body: Introduction: Identifying key molecules involved in cancer metastasis is important in anticancer drug discovery. To identify
such molecules, we used a dual approach involving: 1) gene screening and in vivo confirmation of metastasis-promoting
molecules using a mouse mammary tumor model; and 2) functional and clinicopathological analyses of the candidate molecules
in a human cancer. The products of several candidate genes promoted metastasis of mouse mammary tumors. A protein
designated S100A14 promoted metastasis in the mouse model and was found to be associated with poor prognosis in human
breast cancer patients.
Methods and Results: 1) Identification of metastasis-related molecules: We established two highly metastatic cell lines, 66Lu10
and 66HM, which formed multiple metastatic colonies in general organs, from MCH66 mouse mammary tumor cells. Two
low-metastasis cell lines, 66C8 and 66LM1, were also established. DNA microarray analysis was used to identify candidate
metastasis-related genes based on differential expression in 66Lu10 and 66HM cells compared with 66C8 and 66LM1. An in vivo
metastasis assay in which the highly metastatic cells were transfected with gene-specific siRNAs and orthotopically inoculated
into mice was performed to confirm the metastasis-promoting activity of the candidate genes. Several novel molecules that
promoted metastasis in the mouse model were identified, including S100A14. 2) Significance of S100A14 in human cancer:
Analysis of S100A14 protein expression in various human cancer cell lines and tissues revealed that S100A14 expression is
upregulated in breast cancer. Functional analysis using MCF7 and SK-BR-3 human breast cancer cells showed that S100A14
colocalizes with F-actin on the cell membrane at cell-cell attachment sites and on focal adhesions at the leading edge.
Immunoprecipitation analysis also demonstrated an interaction between S100A14 and F-actin. Boyden chamber and
wound-healing assays showed that S100A14 knockdown substantially suppresses the invasive activity of both cell lines.
Immunohistochemical analysis of archival specimens of primary tumors from 167 breast cancer patients showed strong
membranous staining of S100A14 (53%). Elevated expression of this protein was significantly associated with younger age (<60
years), ER-negative status, HER2-positive status, and poorer prognosis (P=0.0002).
Conclusions: Using our mouse mammary tumor model for identifying metastasis-related genes, we identified S100A14 as a
strong candidate metastasis-promoting molecule in vivo. Functional analyses using human breast cancer cell lines indicated that
S100A14 promotes invasive activity via interaction with cytoskeletal components. Upregulated expression of S100A14 is
associated with poor prognosis in breast cancer patients. Therefore, S100A14 is a potential prognostic biomarker and breast
cancer therapeutic target.

Title: The role of cytoplasmic polyadenylation element binding protein-2 in breast cancer
Asma Hasan1, Mousumi Majumder1 and Peeyush K Lala1. 1University of Western Ontario, London, ON, Canada.
Body: Background: We had shown that over-expression of the inflammation-associated enzyme cyclo-oxygenase (COX)-2
promotes breast cancer progression and metastasis and sustains stem-like cells (SLC), believed to evade traditional therapy and
promote tumor reoccurrence. Stable transfection of the COX-2 gene into COX-2, HER-2, non-metastatic human MCF-7 breast
cancer cell line (named MCF-7-COX-2) led to a highly aggressive phenotype in vitro, where we observed epithelial-mesenchymal
transition, increased migration, invasion, clonogenic tumorsphere forming ability (surrogate of SLC) and expression of SLC
marker ALDH. Furthermore, we observed an increase in the tumorigenic and metastatic capacity of the cells in
immuno-compromised mice in vivo. Combined gene expression and microRNA (miRNA) micro-array analysis of MCF-7 and
MCF-7-COX-2 cells revealed COX-2 induced up-regulation of two miRNAs, miR-526b and miR-655, and down-regulation of their
common gene target, cytoplasmic polyadenylation element binding protein (CPEB)-2. While both miRNAs were found to be
oncogenic in functional assays, the function of CPEB-2 in breast cancer remains untested. Hypothesis: miR-526b and miR-655
down-regulate CPEB-2 to promote breast cancer cell aggressiveness and the SLC phenotype. Approaches & Results: (1) To
test if expression levels of COX-2 and CPEB-2 are inversely related, we measured both gene and protein expression in multiple
COX-2 disparate breast cancer cell lines. (2) We also tested if there is an inverse relationship between miRNA expression and
the expression levels of their target gene CPEB-2. We found that high COX-2/miRNA expressing cell lines such as MDA-MB-231
and MCF-7-COX-2 had significantly lower expression of CPEB-2 than MCF-7 cells (low COX-2/low miRNA). (3) To test the
functional effects of CPEB-2 gene manipulation in vitro and in vivo, we knocked down CPEB-2 in CPEB-2-high MCF-7 cell line.
MCF-7-CPEB-2-KD cells were found to be more migratory and invasive than control cells in transwell assays. Ongoing studies
are to test (a) whether there is a cause and effect relationship between the expression of the miRNAs and CPEB-2 expression,
that is, whether knocking down or knocking in the miRNAs respectively up-regulates and down-regulates CPEB-2 expression;
and (b) whether CPEB-2 expression is lower in breast cancer tissues than in adjacent non-tumor tissues, and inversely correlated
with the miRNAs in breast cancer tissues. Conclusion and Significance: If CPEB-2 is proven to be tumor suppressor, these
oncogenic miRNAs and putatively anti-oncogenic CPEB-2 may serve as potential biomarkers in personalizing breast cancer
therapy with COX-2 inhibitors as adjuvant. (Supported by funds of the OICR to PKL. AH and MM are CaRTT/TBCRU scholars).

Title: FOXC1/FOXA1 transcriptional balance in breast cancer: From acquisition of mesenchymal and stem cell traits to occult
lymph node independent breast cancer metastasis
Partha S Ray1,2, Tor W Jensen2, Tania Ray3, Omid Gholamalamdari4, Connie Y Tsai2, Bomy Kim2, Rohit Bhargava5 and K V
Prasanth4. 1University of Illinois College of Medicine, Urbana, IL; 2Carle BioMedical Research Center, Urbana, IL; 3Oracle
Biosciences, Champaign, IL; 4University of Illinois at Urbana Champaign, Urbana, IL and 5University of Illinois at Urbana
Champaign, Urbana, IL.
Body: Background: Distant metastatic spread of cancer cells to other organs from the primary site of origin currently constitutes
the most significant contributor to cancer-related morbidity and mortality. Epithelial-to-mesenchymal transition (EMT) is a biologic
transformation of cancer cells from a non-migratory phenotype to a migratory one, and is thought to initiate the metastatic
cascade in cancer. EMT has also been reported to trigger acquisition of stem cell traits in breast cancer. Transcription factor (TF)
Forkhead box C1 (FOXC1), strongly associated with the basal-like and claudin-low breast cancer molecular subtypes, is a
powerful EMT inducer and is also a marker of stem/progenitor cells. In contrast, TF forkhead box A1 (FOXA1), strongly
associated with luminal subtypes, is an EMT repressor and a luminal differentiation marker, thus seemingly exerting reciprocally
opposite transcriptional effects to that of FOXC1. We hypothesized that effective EMT program activation status in breast cancer
might be better predicted by examining the expression ratio of an EMT inducer and EMT repressor, such as FOXC1/FOXA1,
theoretically being more reflective of net transcriptional effect than either component alone. Methods: Herein we utilize RNA-Seq
profiling of the HRAS-transformed MCF10A cell series, a well characterized and widely accepted in vitro model of breast cancer
progression and metastasis, to correlate measured FOXC1/FOXA1 ratios to dynamic shifts in EMT marker expression in 3D
matrigel cultures and to stem cell traits observed in primary and secondary mammosphere suspension cultures. We further test
the ability of the FOXC1/FOXA1 expression ratio to predict lymph node independent breast cancer metastasis and death in
independent human breast cancer gene expression datasets. Results: RNA-Seq and qRT-PCR profiling confirmed progressive
increase in FOXC1/FOXA1 ratio to correlate with a progressive loss of E-cadherin expression and synchronous gain of EMT
markers N-cadherin, Fibronectin, and Vimentin. FOXC1/FOXA1 ratio was found to be directly proportional to mammosphere
formation efficiency, a surrogate indicator of stem cell enrichment. In patients without any evidence of nodal metastasis, elevated
FOXC1/FOXA1 ratio was associated with significantly decreased 10 year Overall Survival (HR 2.58;95%CI 1.39 to 4.80, p =
0.003, 295 patient Van de Vijver dataset), 10 year Disease-specific Survival (HR 1.74;95%CI 1.16 to 2.61, p = 0.008,1992 patient
Curtis dataset) and predicted the development of lung metastasis. Conclusion: Elevated FOXC1/FOXA1 expression ratio
indicates EMT program activation in breast cancer, and predicts the associated occurrence of lymph-node-independent distant
metastasis and death in human patients. These findings may allow for the early (pre-symptomatic) diagnosis of clinically occult
(node negative) metastasis by using the FOXC1/FOXA1 ratio as a biomarker of metastasis and permit institution of appropriate
therapy earlier than currently possible. The current study improves our understanding of EMT and highlights the importance of
future studies geared towards unraveling mechanisms involved in regulating FOXC1 and FOXA1 expression in breast cancer.

Title: No Show

Title: Activated thrombin activatable fibrinolysis inhibitor is a novel anti-metastatic factor in breast cancer
Zainab A Bazzi1, Deborah Rudy1, Lisa A Porter1, Dora Cavallo-Medved1 and Michael B Boffa1. 1University of Windsor, Windsor,
ON, Canada.
Body: Thrombin activatable fibrinolysis inhibitor (TAFI) is a plasma zymogen initially known for its role in attenuating fibrinolysis.
Activated TAFI (TAFIa) is formed through proteolytic cleavage by thrombin, plasmin or, its most effective activator, thrombin in
complex with the endothelial cofactor thrombomodulin (TM). TAFIa is a carboxypeptidase, which acts by cleaving carboxyl
terminal lysine and arginine residues from protein and peptide substrates, including plasminogen binding sites on cell surface
receptors. Carboxyl terminal lysine residues play a vital role in accelerating plasminogen activation to plasmin on the cell surface.
Plasmin has many critical functions including cleaving components of the extracellular matrix (ECM), which enhances invasion
and migration of cancer cells. In addition, plasmin can activate matrix metalloproteinases (MMPs), which also play a role in
degrading the ECM. Furthermore, the expression of TM in tumours is inversely correlated to metastasis. Studies have shown that
the anti-metastatic effects of TM result from its ability to bind thrombin. Given that the thrombin/TM complex is responsible for the
activation of TAFI, the anti-metastatic effects of TM may be modulated by TAFIa. We therefore hypothesize that the activation of
TAFI on the cell surface inhibits plasminogen activation and decreases breast cancer cell invasion and migration. Expression of
TAFI and TM were assessed in breast cancer cells with varying degrees of metastatic potential. Although TAFI mRNA levels did
not correlate with malignancy, TM mRNA levels were found to be inversely correlated with breast cancer cell malignancy.
Moreover, cell invasion and migration of MDA-MB-231 and SUM149 cells were assessed upon treatment with potato tuber
carboxypeptidase inhibitor (PTCI), which is a specific inhibitor of TAFIa. Inhibition of TAFIa resulted in a significant increase in cell
invasion and migration of both cell lines. Cell invasion and migration of MDA-MB-231 and SUM149 cells were also assessed
upon treatment with TM. Treatment with TM significantly decreased cell invasion and migration of both MDA-MB-231 and
SUM149 cells. Additionally, experiments using a fluorogenic collagen substrate showed an increase in extracellular collagen
cleavage after PTCI treatment, using both MDA-MB-231 and SUM149 cells. Furthermore, the ability of TAFIa to inhibit pericellular
plasminogen activation was evaluated. Plasminogen activation was significantly decreased on the surface of MDA-MB-231 and
SUM149 cells following treatment with various concentrations of TAFIa. Taken together, these results indicate a vital role for
TAFIa in regulating pericellular plasminogen activation and ultimately ECM proteolysis. Enhancement of TAFI activation in the
breast cancer tumour microenvironment may be a therapeutic strategy to inhibit invasion and prevent metastasis of breast cancer
cells.

Title: No Show

Title: No Show

Title: Fatty acid binding protein 5 promotes metastatic potential of triple negative breast cancer
Catherine A Powell1, Mohd W Nasser1, Jacob C Wochna1, Konstantin Shilo1, Xiaoli Zhang1 and Ramesh K Ganju1. 1Ohio State
University, Columbus, OH.
Body: Triple negative breast cancer (TNBC) is an aggressive subtype of breast cancer that has a poor prognosis for patients, not
only due to its aggressive nature, but also due to the lack of effective therapeutic strategies. High mortality is associated with
metastasis of the primary TNBC. We analyzed the role of fatty acid binding protein 5 (FABP5), an intracellular fatty acid transport
protein, in aggressive breast cancers, especially TNBC. FABP5 expression was analyzed in a tumor microarray (TMA) containing
423 breast cancer patient samples and found that FABP5 expression is associated with tumor grade, triple negative status, and
disease-free survival. Next, a mechanistic approach was taken and found that FABP5 knockdown TNBC cell lines express lesser
EGFR protein compared to control, and upon EGFR activation express less phospho-FAK and phospho-Pyk2. We found that
EGFR was not down regulated at the transcriptional level in FABP5 knockout cells. Further analysis indicated the role for Cbl, an
E3 ubiquitin-protein ligase, as a potential mechanistic target for the down-regulation of EGFR protein in FABP5 knockdown cell
lines. EGFR is over-expressed in aggressive breast cancers, especially TNBC. Though EGFR is highly expressed in aggressive
subtypes of breast cancer, targeted therapeutic strategies have not been successful in clinic. EGFR has been shown to be
involved in distant metastasis in aggressive breast cancers. We found a correlation between FABP5 and EGFR expression and
metastasis-free survival using publically available datasets from The Cancer Genome Atlas. We next studied the functional
consequence of FABP5 knockdown in TNBC cell lines. We showed that EGF-induced FABP5 knockdown cells migrated less
compared to control. Additionally, FABP5 knockdown cells were significantly less able to migrate into a wound in response to
EGF stimulation. EGF-induced cell attachment of FABP5 knockdown cells is significantly decreased compared to control. Our
findings suggest FABP5 modulates metastatic potential of TNBC through alterations in EGFR expression and downstream
migratory signaling molecules.

Title: Signaling consequences and rational therapeutic combinations with glutaminase inhibitor, CB-839, in basal breast cancer
Jennifer B Dennison1, Zhenlin Ju1 and Mills B Gordon1. 1UT MD Anderson Cancer Center, Houston, TX.
Body: Purpose: Proliferation rates of basal breast cancers are dependent on glutamine metabolism as previously demonstrated
by glutamine withdrawal or glutaminase inhibition by CB-839 in cell line models. Early and late adaptive signaling and metabolic
responses to glutamine stress may enable cells to proliferate or evade cell death. We hypothesized that inhibition of these
pathways would be synergistic in combination with glutaminase inhibition by CB-839.
Methods: Protein signaling responses in HCC1806 cells (basal A breast cancer) were quantified using reverse phase protein
arrays (RPPA) at early (4 h) and late (24 h) time points in response to low glucose or low glutamine concentrations as well as
treatment with CB-839, a selective glutaminase inhibitor. Signaling pathways evaluated included mTORC1/2, PI3K/AKT, MAPK,
and AMPK. ADP and ATP were quantified by ion exchange chromatography. Drug combinations predicted by RPPA were
evaluated at therapeutically relevant concentrations using proliferation or apoptosis assays after 72 h of continuous treatment.
Results: In agreement with previous reports of glutamine withdrawal, low glutamine and glutaminase inhibition resulted in
inhibition of proliferation and early inhibition of mTORC1 signaling including pmTOR, pP70S6K, p4EBP1, and pS6. In contrast, for
cells incubated in low glucose, proliferation and mTOR signaling were maintained at early time points despite robust
pAMPK/pACC signaling and increased ADP/ATP ratios (2-orders of magnitude higher than baseline); mTOR signaling was only
inhibited after 24 h when ATP concentrations were reduced. Bioenergetic stress as detected by pAMPK and pACC was not
observed with glutaminase inhibition ATP concentrations were increased by CB-839 treatment. Importantly, 2-deoxyglucose, a
compound used to reduce ATP levels and activate AMPK signaling, was antagonistic in combination with CB-839. In contrast,
hydroxychloroquine, an inhibitor of autophagy activated by mTOR inhibition, was synergistic with CB-839.
Consistent with mTORC1-mediated negative feedback loops, CB-839 treatment also increased AKT and ERK signaling.
Combinations of CB-839 with AKT or MEK inhibitors (AZD5363 and GSK1120212B) were at synergistic and additive,
respectively, for cell proliferation. The MEK inhibitor in combination with CB-839 resulted in supra-additive effects using apoptosis
as an endpoint.
Conclusion: Glutaminase inhibition by CB-839 acutely inhibited mTORC1 signaling independently of AMPK signaling in
HCC1806 cells. MEK and AKT signaling increased, likely in response to negative feedback loops with mTORC1. Combinatorial
approaches with CB-839 targeting these compensation pathways resulted in increased responses supporting further preclinical
and clinical studies.

Title: Metformin blocks de novo synthesis of cholesterol in triple negative breast cancer cells
Ann D Thor1, Zeying Fan1, Reema Wahdan-Alaswad1, Steve M Anderson1, Jennifer K Richer1 and Susan M Edgerton1. 1University
of Colorado Anschutz Medical Campus, Aurora, CO.
Body: Background: Metabolic shifts in cancer and stem cells are well recognized (the so called Warburg Effect). Tumor cells
frequently demonstrate alterations in lipid metabolism. The anti-diabetic drug metformin has been shown to have broad
anti-cancer activity, reducing both the incidence and mortality of several types of cancer. Metformin inhibits pro-oncogenic cell
signaling and metabolic pathways, and has demonstrated unique anti-stem cell activity. We have reported that metformin is
particularly active against triple negative (TN) breast cancer cells, inducing S phase arrest and apoptosis. Metformin down
regulates de novo fatty acid synthesis, significantly reducing fatty acid synthase (FASN) and FASN protein expression.
Mechanistically, these shifts appear secondary to a metformin associated up regulation of miR-193a, which directly targets the
FASN 3"UTR (In press, Wahdan-Alaswad et al, Hormones and Cancer).
Cancer cells require high cholesterol for membrane synthesis, rigidity, the formation of lipid rafts (membrane regions which
organize critical receptor/signaling molecules including the epidermal growth factor receptor (EGFR)), and to provide precursor
molecules for hormones and sterols. While de novo cholesterol biosynthesis and high systemic cholesterol levels have been
associated with an increase in breast cancer risk and a worse prognosis for women with the disease, clinical studies of statins
(drugs to inhibit the mevalonate synthesis pathway and thus block cholesterol synthesis) have shown mixed results.
Silvente-Poirot S et al have suggested that the contradictory effects of anti-cholesterol strategies may be due to the complexity of
the genes and enzymes controlling the "sterolome" (Science, 2014).
Results: The effects of metformin on de novo cholesterol biosynthesis in the TN breast cancer cells, MDA-MB-231 and
MDA-MB-468, were studied using expression profiling and miRNA microarrays. Metformin treatment for 24 h down regulated the
expression of 21 genes encoding enzymes in the cholesterol biosynthesis pathway at high levels (up to 9 fold), which has been
confirmed by qRT-PCR. Metformin treatment for 6 h also resulted in the upregulation of 17 miRNA, which are predicted to control
many of the down regulated genes. These data suggest a post-translational inhibition of cholesterol biosynthesis, similar to what
we have reported for fatty acid biosynthesis (see above). The down regulated enzymes include critical steps in complex
cholesterol biosynthesis pathway occurring in the cytosol, endoplasmic reticulum and the nuclear envelope. Western blots have
confirmed that miRNA 141 and miR192 reversed the metformin induced reduction in p-MEK1/2, p-Akt and pMAPK.
Conclusions: Metformin regulates cancer cells by multiple pathways including reduction in cell growth, increases in apoptosis and
alterations in cellular metabolism. The effects on the cholesterol pathway are both extensive and affect multiple downstream
pathways including EGFR, Akt, and MAPK.
Partial funding was provided by the Susan G. Komen Foundation.

Title: Targeting lactate metabolism as a novel therapeutic target in platinum-resistant triple negative breast cancer (TNBC)
Narumi Harada1, Vidhya Varghese1, Lingmin Xie1 and Laura M Kenny1. 1Imperial College London, United Kingdom.
Body: Introduction TNBC is a heterogeneous disease and is often associated with expression of a stem cell signature (CD44+,
CD24-). Recently platinum agents have been shown to improve the pathological response rates in the neo-adjuvant setting;
however a significant proportion of patients with TNBC will relapse. The study objective was to evaluate the potential of targeting
glycolysis, mitochondrial function and lactate metabolism in platinum resistant TNBC cell lines.
Methods Stem cell signature and glycolytic phenotyping: CD44 and CD24 antibodies were used to measure the percentage
of stem cells in each of the cell lines using flow cytometry(FACS). Protein expression of hexokinase, Glut1 transporter, lactate
dehydrogenase A/B(LDHA and LDHB), and monocarboxylic transporter 1 and 4 (MCT1 and MCT4) was determined in
MCF7(control), MDA-MB231, MDA-MB468, and HCC38 cells. Proliferation and apoptosis assays: IC50s for platinum,
metformin, and MCT1 inhibitor AR-C155858, Tocris (MCT1I) were calculated by the sulforhodamine B (SRB) assay. Propidium
iodide and AnnexinV staining intensity used to measure apoptosis by FACS. Metabolic analysis: The change in lactate and
glucose levels in culture media post-treatment with metformin and MCT1I was quantified by 1H NMR. Intracellular lactate and
ATP concentrations were determined by lactate(BioVision) and ATP determination assays(Invitrogen). Mitochondrial membrane
potential and Reactive oxygen species (ROS) were measured by MitoTracker Red and MitoSOX Red. 2NBDG was used to
assess cell glucose uptake. Combination index SRBs: Cells were treated with increasing concentrations of MCT1I and
metformin for 72h. The combination index values were calculated using CalcuSyn software.
Results MDA-MB231 and HCC38 cells were resistant to cisplatin and associated with a stem cell signature (CD44+, CD24-)
which was not detected in the other cell lines. MCT1 was highly expressed in HCC38 cells but not at all in MDA-MB231. In
contrast, MCT4 was abundantly expressed MDA-MB231 but minimally in HCC38 cells.
In these two cell lines, only MDA-MB231 was sensitive to metformin which reduced ATP production, induced ROS generation,
and increased the percentage of apoptotic cells. HCC38 also produced ROS, but ATP production and apoptosis were unaffected.
HCC38 cells were more sensitive to MCT1I than other cells, which was associated with increased intracellular lactate.
The NMR data showed significant increases in lactate in culture media following treatment with metformin in MDA-MB231 and
HCC38. Only HCC38 had increased 2NBDG uptake and reduced of glucose levels in culture medium by metformin.
Metformin significantly enhanced the anticancer efficacy of MCT1I in HCC38 cells, as a synergistic effect of the two agents was
observed across the range of different concentrations; whilst an antagonistic effect was shown in MDA-MB231 cells with almost
all concentrations.
Conclusions Targeting lactate metabolism selectively inhibits a platinum-resistant stem cell population of TNBC, this effect is
enhanced when combined with metformin.
The inhibition of glycolysis and lactate metabolism may represent a new therapeutic strategy for this population and warrants
further study.

Title: No Show

Title: An association between obesity and more aggressive breast cancer subtype
Saranya Chumsri1, Paula Rosenblatt1, Candace Mainor1, Gauri Sabnis1, Olga Goloubeva1 and Angela H Brodie1. 1University of
Maryland Greenebaum Cancer Center, Baltimore, MD.
Body: Background: Obesity has become epidemic in the United States with over one third of the adult population being obese.
The incidence of obesity is even higher among African American population with slightly more than half of the population being
obese. Elevated body mass index (BMI) has been associated with an increased risk of breast cancer in postmenopausal women.
This is believed to be due to higher levels of estrogens, inflammation, and insulin. Given that adipose tissue is the major source of
estrogen production via the aromatization of androgens into estrogens in postmenopausal women, there is a concern that obese
women may have inadequate estrogen suppression with a fixed dose of aromatase inhibitors (AIs). Multiple recent retrospective
analyses of large phase III adjuvant endocrine therapy trials demonstrated that obese patients had worse outcome when treated
with AIs. However, it remains unclear whether there is also a difference in the tumor subtype among normal weight, overweight,
and obese patients.
Method: A retrospective review of breast cancer cases diagnosed from July 2008 and August 2011 was performed. Patient
characteristics, BMI, and pathologic findings were collected. An immunohistochemistry (IHC)-based molecular subtypes was
assigned based on St. Gallen criteria: Luminal A (ER/PR+, HER2- and Ki67 14%), Luminal B (ER/PR+, HER2+ or Ki67 > 14%
or grade 3), HER2 enriched (ER/PR-, HER2+), and Triple negative breast cancer (TNBC; ER/PR-, HER2-). Using the WHO
classification, normal weight was defined as BMI < 25, overweight BMI 25-30, and obese BMI 30.
Result: 143 patients were included in the analysis. The median age was 55 years. The majorities of our patients were African
American (63%) and presented with early stage disease (stage I 29%, stage II 45%, stage III 22%, and stage IV 5%). IHC-based
molecular subtypes were luminal A 20%, luminal B 48%, HER2 enriched 7%, and TNBC 25%. 21% were normal weight, 30%
overweight and 49% obese. As shown in the table below, comparing between normal weight, overweight, and obese patients,
there is a possible association between tumor subtype and BMI status (p = 0.03). The majority of TNBC patients (94.4%) were
overweight or obese. Specifically, there is significantly more aggressive luminal B subtype compared to luminal A (52.86% vs.
15.71%) among obese patients (p = 0.017). We further evaluated the effect of obesity and tumor growth in xenograft model of
MCF7Ca. Comparing between obese mice fed with high fat vs. lean mice fed with chow diet, there is a significant increase in
tumor growth among obese mice (p = 0.04) which corresponds to the high proliferation index measured by Ki67 seen in patients
with luminal B subtype.
IHC-based molecular subtypes and distribution by weight category
Weight Category
Luminal A
Luminal B
HER2 Enriched
TNBC
Normal Weight
8 (26.67%)
15 (50%)
5 (16.67%)
2 (6.67%)
Overweight
9 (20.93%)
17 (39.53%)
2 (4.65%)
15 (34.88%)
Obese
11 (15.71%)
37 (52.86%)
3 (4.29%)
19 (27.14%)
Conclusion: Our study suggests that obesity may provide a microenvironment that support and accelerate tumor growth,
particularly in luminal breast cancer subtypes. Furthermore, the poor outcome seen in obese patients may also be in part due to
more aggressive tumor subtype.

Title: No Show

Title: A novel pharmacodynamic assay to measure glutaminase inhibition following oral administration of CB-839 in triple
negative breast cancer biopsies
Andy MacKinnon1, Mark Bennett1, Ethan Emberley1, Mathew Gross1, Julie Janes1, Evan Lewis1, Alison Pan1, Mirna Rodriguez1,
Peter Shwonek1, Taotao Wang1, Jinfu Yang1, Frances Zhao1 and Francesco Parlati1. 1Calithera Biosciences, South San
Francisco, CA.
Body: Triple negative breast cancer (TNBC) cell lines are highly dependent on glutamine (Gln) for growth and survival. A critical
step in Gln utilization is its conversation to glutamate (Glu) by the mitochondrial enzyme glutaminase (GLS). CB-839 is a potent
inhibitor of GLS that has anti-proliferative activity in TNBC cell lines and antitumor activity in TNBC xenograft models (Gross et al.,
Mol. Cancer Ther. 13:890). Across a panel of breast cancer cell lines derived from both receptor positive and TNBC tumors,
sensitivity to CB-839 was associated with (i) elevated GLS expression, (ii) elevated GLS activity, and (iii) the TNBC subtype.
Importantly, many of the determinants of CB-839 sensitivity in cell lines are also present in primary tumor samples, including high
mRNA and protein expression of GLS and a high Glu to Gln ratio in TNBC tumors as compared to receptor positive tumors.
These observations motivate the Phase 1 clinical study of CB-839 in TNBC patients. To aid in the selection of a recommended
Phase 2 dose, we sought to develop a pharmacodynamic (PD) assay to directly measure the GLS activity in breast tumor lysates
in order to determine the extent of GLS inhibition in tumor biopsies from CB-839 treated patients.
To develop a robust PD assay, we first identified conditions that maintain the GLS:CB-839 inhibitory complex during preparation
of lysates from CB-839 treated samples. High concentrations of KCl (150 mM) and low concentrations of K-phosphate (15 mM) in
the lysis buffer, as well as maintaining the lysate at a low temperature stabilized the inhibited complex. Following gel filtration of
the lysate to remove unbound CB-839 and exchange the buffer, GLS activity was immediately measured with a coupled enzyme
assay. The GLS activity measured at this step reflects the residual activity present in a sample that was exposed to CB-839. To
quantify the amount of total GLS present in the sample, we incubated the same lysate for 3 hours at room temperature under
conditions of low KCl and high phosphate to allow the the GLS:CB-839 complex to fully dissociate prior to measuring activity. This
assay format allows quantitation of the % GLS inhibition from a single tumor lysate sample and eliminates the requirement for
multiple biopsies as well as any assay variability due to tumor heterogeneity.
We utilized this tumor PD assay to determine the plasma drug levels required for maximal tumor GLS inhibition in a preclinical
TNBC model. Mice bearing HCC1806 TNBC tumors were first treated with a range of CB-839 doses. Four hours after oral
administration, a 10 mg/kg dose of CB-839 resulted in >90% inhibition of tumor GLS. CB-839 plasma concentrations of 100 nM
corresponded to 50% inhibition of tumor GLS, while maximal inhibition occurred at plasma concentrations 300 nM. In xenograft
studies, maximal anti-tumor efficacy was achieved with BID dosing at 200 mg/kg, a dose and schedule that resulted in trough
plasma levels of CB-839 of 300 nM and sustained GLS inhibition in tumors. As part of an ongoing Phase 1 trial, this assay will
be utilized to monitor tumor PD responses in TNBC patients undergoing treatment.

Title: Reduction of HER2-associated lipogenic phenotype by docosahexaenoic acid (DHA) induces apoptosis in breast tumor
cells harboring HER2 overexpression
Graziela R Ravacci1, Maria M Brentani1, William Festuccia2, Tharcisio Tortelli1, Angela F Waitzberg3 and Dan L Waitzberg1.
1
University of So Paulo FMUSP, Faculty of Medicine, So Paulo, SP, Brazil; 2Institute of BioMedical Science University of
So Paulo ICB/USP, So Paulo, SP, Brazil and 3Federal University of So Paulo (UNIFESP), So Paulo, SP, Brazil.
Body: The lipogenic phenotype is associated with oncogenic transformation and malignancy in the cancer setting. Our
hypothesis is that the oncogenic nature of lipogenesis depends on the expression of HER2 oncogene, and its modulation by
docosahexaenoic acid-DHA, might induce apoptosis in breast tumor cells with HER2 overexpression. To evaluate if the lipogenic
phenotype might be induced by HER2 overexpression, we used a oncogenic transformation cell model in which, cells were
engineered to overexpress HER-2 receptor (HB4aC5.2), but are identical to their parental strain (HB4a) in all other aspects,
permitting an specific analysis of enhanced HER-2 effects. Toward this end the lipogenesis profiling was characterized,
evaluating several molecular features, including synthesis (FASN), uptake(CD36), transport(FABP4) and storage(DGAT) of FA by
RT-PCR and lipogenic regulatory pathways (mTOR, DEPTOR, SREBP1 and PPAR) in both cells. Lipogenic contribution to lipid
rafts formation, which is necessary to HER2 receptor location and activation in the cell membrane, was evaluated by gas
chromatography and confocal microscopy. The influence of HER2 overexpression and lipogenic phenotype on proteins activated
by HER-2 (AKT, ERK1/2 and FASN) was analyzed by western blot. Next, HB4a and HB4aC5.2 cells were treated, alone or in
combination, with DHA, Trastuzumab (anti-HER2), and GW9662 (PPAR inhibitor) for 72h, and the above experiments were
repeated. Cell death was analyzed by flow cytometry and confocal microscopy. Results: In HB4aC5.2 cells, the oncogenic
transformation by HER2 overexpression was associated with a lipogenic phenotype, which contributed to increase of lipid rafts
formation, activation of survival and proliferation signals, as compared to HB4a normal cells (p<0.001). It is believed that PPAR
is the main regulator of cell lipogenesis. However, our data have shown that mTOR/SREBP pathway, exclusively mTORC1, was
decreased concomitantly to increase of DEPTOR gene expression and AKT activation, in HB4aC5.2 cells. Also, the treatment
with GW9662 did not change the expression of lipogenic genes, suggesting that lipogenesis seems independent of PPAR
activation, and that HER-2 overexpressing cells may use alternative mechanisms to maintain the lipogenic phenotype. DEPTOR
is overexpressed in white adipose tissue and is associated with regulation of lipogenesis. Moreover, DEPTOR overexpression
suppresses S6K1 but, by relieving feedback inhibition from mTORC1 to PI3K signaling, activates Akt. In HB4aC5.2 cells, only the
DHA treatment decreased the lipogenic phenotype, DEPTOR expression, disrupted lipid raft, inhibited HER-2 signaling, and
induced apoptosis (p<0.001). In addition the combined treatment using DHA and Trastuzumab increased cell death. Conclusion:
We show that lipogenic phenotype was associated with HER-2 in breast cancer, can be induced early in the oncogenic
transformation, and it seems important to promote survival and proliferation. Otherwise, Its modulation by DHA increased death in
tumor cells, suggesting this kind of FA may represent a useful tool for controlling the HER2-positive breast cancer.

Title: Beyond fatigue The impact of breast cancer treatment on body composition and energy expenditure
Kathy D Miller1, Jessica L Sollars1, Sandra K Althouse1, Linda K Han1, Stephen J Ventura2 and Jeffrey S Sledge2. 1Indiana
University Melvin and Bren SImon Cancer Center, Indianapolis, IN and 2University of Wisconsin, Madison, WI.
Body: Background: Patients (pts) frequently report fatigue as a consequence of breast cancer therapy but the true impact of
cancer treatments on body composition and energy expenditure has not been measured.
Patients and Methods: Pts with stage 0-III were evaluated prior to any therapeutic intervention to assess baseline health status,
at 6 months to assess acute impact of therapy, and at 12 months to assess chronic effects and spontaneous recovery.
Measurements included body composition, maximum watts produced and power envelope (watts/kg) based on laboratory testing,
energy expenditure over 7 days in real world conditions, insulin resistance (HOMA), fatigue (BFI), quality of life scales (FACT-B),
and serum for insulin-like growth factor-1 (IGF-1), adiponectin, leptin, markers of bone formation (P1NP, BSAP, osteocalcin),
bone resorption (sCTX), and bone mineral metabolism (PTH, 25-OH Vit D).
Results: 45 patients have been enrolled; 25 have completed the 6 month assessment to date. Median age was 52 (30-68).
Patients were analyzed separately based on therapies received.
Chemotherapy +/- HRT
Hormone Therapy
Local Therapy Only
Baseline (n=23) 6 months (n=12) Baseline (n=14) 6 months (n=11) Baseline (n=8) 6 months (n=2)
BMI (kg/m2)
27.8
26.3
28.7
29.3
32.0
28.2
Lean mass (kg)
39.5
36.3
42.0
41.1
43.0
36.1
Body fat (%)
43.9
47.3
46.5
50.6
49.4
45.6
HOMA-IR*
1.75
1.77
1.21
1.81
2.10
1.18
Peak power (Watts)
68.2
44.1
70
53
97
88.5
Watts/kg lean muscle
1.7
1,2
1.7
1.3
2.3
2.4
Time to zero power (min.sec)^
12.3
6.39
9.5
6.9
11.2
10.6
*HOMA-IR was not assessed in patients known to have diabetes. ^Time to zero power quantifies the time over which patients can
sustain generating motive power as a measure of endurance.
Week long bi-axial pedometry studies at baseline and six months revealed a decrease in pt activity between 17 and 23 percent,
confirming the decrease in motive power seen in the power envelope and body composition tests. Enrollment continues to a
planned total of 75 pts. Additional follow-up with 12 month assessment, as well as QOL and serum data will be presented.
Conclusions: Pts in this study were more deconditioned than expected based on population normative data. Systemic therapy
had a profound impact on body composition with a substantial decrease in lean muscle mass with a corresponding increase in %
body fat and decrease in the ability to generate and sustain motive power. Our data suggest that most patients would not be able
to participate in or comply with commonly proposed exercise interventions and physical activity recommendations. Future
interventions need to be restructured, following the model of individualized rehabilitation with careful assessment of the pts
baseline status to improve the health of breast cancer survivors.

Title: BMI and metabolic factors in long-term breast cancer survivors: Changes from diagnosis and comparison to non-breast
cancer controls
Ana Elisa Lohmann1,2, Marguerite Ennis3 and Pamela J Goodwin1,2. 1Samuel Lunenfeld Research Institute, Sinai Hospital,
Toronto, ON, Canada; 2Division of Medical Oncology and Hematology, University of Toronto, Toronto, ON, Canada and 3Appplied
Statitician, Toronto, ON, Canada.
Body: Background: The metabolic syndrome is associated with poor breast cancer (BC) outcome. We evaluated changes from
diagnosis in metabolic factors and BMI in long-term survivors and compared their status at long-term follow-up (LTFU) to that of
age-matched women with no history of BC.
Methods: A total of 535 women with early breast cancer were enrolled between 1989 and 1996 and followed prospectively. From
2005 to 2007, those alive without distant recurrence were re-contacted to participate in a long-term followed-up study and 285
agreed. A control group of 167 age-matched women without cancer history was recruited from women presenting for screening
mammograms. Mean changes in metabolic factors from diagnosis to long-term follow-up were assessed with paired t-tests. In
spite of matching, controls were younger and had higher income than survivors and the comparison to controls was made using
age-adjusted regression models. Variables were transformed to normality before statistical testing.
Results: With a median follow-up of 12.5 years, BC survivors gained on average 2.35 kg and BMI, waist and hip circumference,
waist-hip ratio, glucose, insulin, HOMA, total cholesterol and its components (but not triglycerides) increased significantly. After
age adjustment, waist circumference, glucose, HOMA and total triglycerides were significantly higher in BC survivors compared to
controls
Comparison of LTFU BC survivors to non-cancer controls
Unadjusted mean standard
deviation
Unadjusted mean standard

deviation
BC patients
controls
P-value for age-adjusted

difference
waist circumference
(cm)
85 12
81 10
.01
Glucose (mmol/L)
5.5 1.0
5.2 0.9
.01
Insulin (pmol/L)
60.9 50.5
47.1 28.3
.06
HOMA
2.25 2.24
1.64 1.24
.03
Triglycerides (mmol/L)
1.28 0.64
1.10 0.57
.01
BMI (kg/m2)
26.3 4.9
25.4 4.5
.26
Despite exclusion of BC patients with diabetes at study entry, 24.9% of BC survivors self reported diabetes or pre-diabetes
(1.99%/year) versus 12.6% in controls (OR 2.3, p= .0017).
Conclusion: The metabolic status of long-term BC survivors deteriorated over time and age-adjusted at LTFU were worse with
respect to a number of factors compared to the control group.

Title: Prevention of aromatase Inhibitor (AI)-induced joint symptoms with omega-3 fatty acid supplementation: A randomized
placebo-controlled pilot study
Maryam B Lustberg1, Tonya Orchard2, Xueliang Pan3, Raquel Reinbolt1, Amanda Logan1, Joanne Lester2, Rachel M Layman1,
Erin Macrae1, Ewa Mrozek1, Bhuvaneswari Ramaswamy1, Robert Wesolowski1, Michael Berger1, Michael Knopp2, Charles
Loprinzi4, Charles L Shapiro1 and Lisa Yee1. 1Ohio State University Comprehensive Cancer Center & The Stefanie Spielman
Comprehensive Breast Center, Columbus, OH; 2Ohio State University Comprehensive Cancer Center, Columbus, OH; 3Ohio
State University, Columbus, OH and 4Mayo Clinic, Rochester, MN.
Body: Introduction: AI-induced joint symptoms negatively impact drug adherence and quality of life. Based on observations that
n-3 polyunsaturated fatty acids (PUFAs) have anti-inflammatory effects and that the mechanism of AI-induced joint symptoms
may be partly due to inflammation, we hypothesized that women taking more n-3 PUFAs are less likely to develop AI-induced
joint symptoms.
Methods: We conducted a randomized, double-blind, placebo-controlled study comparing n-3 PUFA vs placebo in
postmenopausal breast cancer patients starting adjuvant AIs. Participants were randomized to n-3 supplements [2.58 g
eicosapentaenoic acid + 1.74 g docosahexaenoic acid/day; Marine Nutriceuticals, Mt. Bethel, PA] vs matched placebo for 24
weeks (wks). Primary endpoints was feasibility; secondary outcomes were self-reported symptoms as assessed by the Brief Pain
Inventory short form (BPI-SF), Functional Assessment of Cancer Treatment, Breast & Endocrine Symptoms (FACTB-ES), and
Stanford's Health Assessment and Disability Index (HAQ) at baseline prior to AI receipt, 12 and 24 wks. Compliance and toxicity
were evaluated monthly. Serial peripheral blood n-3 PUFA levels and inflammatory cytokines (IL-6, TNFR2, IL-17) were drawn.
MRI of hands/wrists was performed in selected patients using a 3 Tesla dedicated wrist coil at baseline and treatment end.
Results: Forty-four women were enrolled and randomized to study drug; 42 received 1 cycle (4 wks) of treatment; 36 had 1
post treatment evaluation at wk 12 or 24. Median age was 59.5 (range 43-76); history of prior taxane (n=15, 34%). The two
groups baseline characteristics were similar. Overall, 93% and 88% of patients took >80% of the placebo and n-3 PUFA doses,
respectively. Baseline erythrocyte n-3 PUFA was similar for both groups (6.6% 1.6%, 7.2% 1.9%, p=0.20), but higher in the
n-3 PUFA arm by wk 24 (6.5%1.0% vs 15.0%3.3%, p<0.001). Most toxicities were grade 1; the n-3 PUFA arm had only 1
(2.5%) grade 3 toxicity (diarrhea). The n-3 PUFA arm reported lower mean BPI-SF scores after treatment [(-.0.28/ -0.25 at week
12/24); but not statistically significant compared to placebo (p=0.494 and 0.601)]. Based on BPI-SF, the n-3 PUFA arm reported
less interference of pain symptoms compared to placebo at 12 weeks (-.72, p=0.08). This arm also had a decreased walking,
activity and working (WAW) score on BPI-SF at 12 weeks (-.81 p=0.05), and reported significantly greater pain relief from
medications at 12 (p=0.043) and 24 weeks (p=0.011). Both arms had similar baseline and wk 24 serum IL-6 levels; levels
decreased from baseline to wk 24 in the n-3 PUFA arm (-0.540.25, p=0.048). There was a non-significant trend (p= 0.2) toward
decreased wrist inflammation by MRI imaging at 24 wks in the n-3 PUFA arm.
Conclusions: This is the first randomized pilot study to show that n-3 PUFA supplementation to prevent AI-induced joint
symptoms is feasible and well tolerated. There is preliminary evidence that this intervention may help reduce the burden of
AI-induced arthralgias.
OSU Study #11022; ClinicalTrials.gov Identifier: NCT01478477. Grants from the National Cancer Institute (CA037447-26) to the
Alliance for Clinical Trials in Oncology supported this pilot study.

Title: Longitudinal evaluation of taxane-induced neuropathy in early stage breast cancer
Maryam B Lustberg1, Scott Monfort2, Janani Singaravelu4, Raquel E Reinbolt1, Xueliang Pan2, Bhuvaneswari Ramaswamy1,
Rachel M Layman1, Robert Wesolowski1, Ewa Mrozek1, Erin Macrae1, Charles Shapiro1, Robyn Patrick3, Charles L Loprinzi5 and
Ajit Chaudhari2. 1Ohio State University Comprehensive Cancer Center & The Stefanie Spielman Comprehensive Breast Center,
Columbus, OH; 2Ohio State University, Columbus, OH; 3Ohio State University Comprehensive Cancer Center, Columbus, OH;
4
Ohio State College of Medicine, Columbus, OH and 5Mayo Clinic, Rochester, MN.
Body: Introduction
Chemotherapy-induced peripheral neuropathy (CIPN) is a dose-limiting toxicity of several chemotherapy drug classes, including
taxanes. Peripheral neuropathies have been shown to lead to pain, falls, and difficulty in walking and performing activities of daily
living in a variety of patient populations. Although the prevalence of CIPN has been noted in cancer patients, the development of
self-reported symptoms, gait changes and balance changes during treatment have not been well explored to date. We
hypothesized that the use of taxane-based chemotherapy will result in significant changes in spatiotemporal gait and balance
parameters, as well as self-reported quality of life and function.
Methods
We characterized the alterations in gait and balance that occur in non-metastatic breast cancer patients during taxane
chemotherapy. We evaluated (1) spatiotemporal gait parameters, including cadence and step length, and (2) balance parameters,
including time-to-contact and 95% ellipse area, using each patient as her own control. Laboratory assessment of gait and balance
was conducted at baseline and at completion of therapy in selected patients. We compared the natural history of changes in gait
and balance parameters with changes in CIPN status as measured by validated patient reported outcomes, including EORTC
QLQ-C30, CIPN-20, and Brief Pain Inventory Short Form (BPI-SF), and the Duke Activity Status Index (DASI). Time points
included pre-chemotherapy, after each cycle of chemotherapy, and one month after the end of therapy to collect information on
neuropathy, pain and functional capacity. The preliminary data were illustrated using individual plots; trend lines (changing over
time) were based on least square means at each time point, which were estimated using the linear mixed models for repeated
measures.
Results
To date, 15 patients with localized breast cancer have been enrolled; patient recruitment is ongoing. The median age is 42 years
(range 25-67). Ten patients (67%) received weekly paclitaxel, 1 patient (7%) received paclitaxel every 2 weeks, and 4 patients
(27%) received docetaxel every 3 weeks. Preliminary results with these 15 patients, based on least square means at each
timepoint, showed trends in several parameters. As treatments progressed, patients tended to develop more difficulty in quiet
balance and in their ability to actively shift weight in the sagittal and frontal planes. From the CIPN-20, they also tended to
develop increased difficulty with sensory and motor systems. From the QLQ-C30, their global health status also tended to
worsen. For most of these parameters, the largest changes were observed between the 2nd and 3rd treatments, though some
changes were not observed until the 4th treatment. From the BPI-SF, no trends in pain symptoms or pain interference were
observed within this preliminary cohort.
Conclusions
Gait and balance testing is feasible in the clinical setting. Preliminary observations suggest that balance, function and quality of
life may all be affected by taxane therapy, even without pain symptoms. The findings of this study will enable us to better
characterize the neurotoxic effect of taxanes and to ultimately test the effectiveness of preventative measures and interventions.
Funding by NCI R03CA182165-01.


Title: A prospective longitudinal study of the impact of breast cancer treatment with adjuvant chemotherapy or tamoxifen alone on
ovarian reserve
Shari B Goldfarb1,2, Giuliano Bedoschi3,4, Eli Grunblatt5, Sumanta Goswami5, Tessa Cigler2, Jessica Quistorff1, Fernanda
Pacheco3, Robert Stobezki3, Mark Robson1, Fred Moy3, Clifford Hudis1,2, Sujata Patil1, Maura Dickler1,2 and Kutluk Oktay3,4.
1
Memorial Sloan Kettering Cancer Center, New York, NY; 2Weill Cornell Medical Center, New York, NY; 3New York Medical
College, Valhalla, NY; 4Innovation Institute for Fertility Preservation and In Vitro Fertilization, New York, NY and 5Yeshiva
University, New York, NY.
Body: Background: Many women with breast cancer (BC) of reproductive age would prefer to preserve their fertility and ovarian
function, if possible. However, available information is insufficient to predict the likelihood and extent of ovarian damage that will
be suffered by an individual woman. In prior studies, menses was used as a surrogate for fertility, but it is a poor marker of future
fertility. Preliminary work suggests that anti-mullerian hormone (AMH) may be a better surrogate for ovarian reserve. The goal of
this study is to delineate the extent of ovarian damage from specific treatment regimens by using serum AMH. Here we report
results from the largest longitudinal study with a pretreatment evaluation to measure the impact of both chemotherapy (chemo)
and tamoxifen (tam) on ovarian reserve in BC patients. Methods: A multi-institutional longitudinal IRB-approved study was
performed in 207 premenopausal women with stage 0-III BC. AMH levels were evaluated at baseline and 1 yr post completion of
adjuvant chemo or 1 yr post initiation of tam. After the exclusions (failure to follow-up, sample inadequacy, presumed PCOS and
consent withdrawal), 115 women ages 26-46 (median 38) were available for analysis at 1 yr post treatment (primary endpoint).
AMH levels were measured in frozen sera with an in-house ultrasensitive ELISA assay and were log transformed due to
non-normal distribution. Results were analyzed with Wilcoxon rank sum test and repeated measures ANOVA to adjust for age,
tam use and stage. Results: In the 115 evaluable women, 98 pts had HR positive BC & 17 had triple negative BC. 77 pts
received anthracycline-based (ddAC-T/EC-T) chemo, 26 non-anthracycline-based chemo (TC/CMF/TH) & 12 tam only. In the 103
women who received chemo, compared to baseline (median 0.21 ng/ml, 0.001-3.9 ng/ml), AMH levels declined significantly
(median 0.11 ng/ml, 0.001-4.47 ng/ml; p<0.0001) at 1 yr post chemo. The type of chemo, stage and adjuvant tam use did not
significantly impact the results. In the 12 women who received tam only, compared to baseline (median 0.505 ng/ml, 0.010-1.27
ng/ml), AMH levels declined significantly (median 0.155 ng/ml, 0.010-0.91 ng/ml; p=.0049) at 1 yr. Despite a significant decline in
ovarian reserve by AMH, 69% (66/95) of pts who received chemo had return of menses by 1 yr post treatment. Conclusion: This
longitudinal study shows that BC chemo is detrimental to ovarian reserve. Although a decline in AMH was also seen in pts treated
with tam only, this is a small cohort and the results are hypothesis generating. The decline in AMH may be due to ovarian
stimulation from tam, which can cause predominance of larger follicles that produce less AMH rather than actual reduction in
ovarian reserve. This needs to be studied in a larger cohort. The fact that the majority of pts resumed their menses despite a
significant decline in their ovarian reserve by AMH indicates that resumption of menses is not the best measure of ovarian
normalcy/fertility. This new info can be used for counseling pts in childbearing ages so they can make better informed decisions
on fertility preservation, if desired. Supported by NIH RO1 HD053112, Jodi Spiegel Fisher Cancer Foundation and Susan G.
Komen Foundation.

Title: Patient centered support in the survivorship care transition: Outcomes from the patient-owned survivorship care plan
intervention
Elizabeth A Kvale1,2, Wendy Demark-Wahnefried2, Karen Meneses2, Sejong Bae2, Chao-Hui S Huang5, Casey B Azuero5 and
Christine S Ritchie4,3. 1Birmingham VA Medical Center, Birmingham, Al; 2University of Alabama, Birmingham, Al; 3University of
California, San Francisco, CA; 4San Francisco VA Medical Center, San Francisco, CA and 5University of Alabama, Tuscaloosa,
Al.
Body: Background: Survivorship care plans will be mandated for use by 2015. Few studies have focused on the care transition
process, and few show an impact on the patient experience or health outcomes. This study examines the effect of a
patient-owned survivorship care plan (POSCP) intervention on patient outcomes and care coordination. Based on a conceptual
model that incorporates IOM survivorship priorities and the chronic care model, the POSCP intervention is a single coaching
encounter utilizing motivational interviewing to engage patients in the development of a POSCP that incorporates health goals
and strategies related to surveillance, symptom management, and health behavior.
Methods: 79 recent breast cancer survivors Stage I-IIIB were randomized to receive POSCP intervention (n=40) or usual care
(n=39). Patient outcomes were assessed using SF-36 Health Survey (SF-36), Social/Role Activities Limitations, Self-Efficacy for
Managing Chronic Disease Scale, and Personal Health Questionnaire Depression Scale (PHQ-9) at baseline and 3 month follow
up. Care coordination was assessed using confirmed primary care physician (PCP) visits and reported discussion of survivorship
plan at follow up. Logistic and linear regressions were conducted to examine the effect of intervention on care coordination,
patient experience, and health outcomes.
Results: In this sample age M= 58.35, SD = 10.62, 81% (N=64) non-hispanic white, 16.5% (N=13) African American. A greater
proportion of participants in the intervention group reported a confirmed appointment (59.5% vs. 51.4%) and discussed
survivorship issues (47.4% vs. 35.1%) with their PCPs compared to those in the usual care group, however these differences did
not achieve statistical significance (p= .70 and .65). After controlling for education and perceived financial adequacy, participants
in the intervention group demonstrated significantly higher self-reported health (F (3, 70) = 3.57, p< .05), lower social role
limitation (F (3, 70) = 3.82, p < .05), as well as a trend toward greater self-efficacy (F (3, 69) = 2.51, p = .07) compared to those in
the usual care group. Results of paired-samples t-tests revealed a significant decrease in depression scores from baseline to 3
months follow up in the intervention group, t (36)= 3.21, p < .01, ^2 =.23. No between group differences were identified in
symptom burden or patient activation.
Conclusion: This pilot study of a POSCP process utilizing motivational interviewing techniques to enhance patient engagement
has a positive impact on patient outcomes and demonstrates promise as a strategy to improve survivor patient experience, care
coordination, and health outcomes.

Title: Risk factors for lymphedema are dependent on level of axillary surgery
Sharon L Kilbreath1, Kathryn M Refshauge1, Jane M Beith2, Leigh C Ward3, Owen A Ung5, James R French4, Louise Koelmeyer4,
Katrina Kastania1 and Jasmine Yee1. 1University of Sydney, Sydney, NSW, Australia; 2Chris O'Brien Lifehouse, Camperdown,
Sydney, NSW, Australia; 3University of Queensland, Brisbane, Queensland, Australia; 4Westmead Hospital, Westmead, NSW,
Australia and 5Royal Brisbane and Women's Hospital, Brisbane, Queensland, Australia.
Body: Background: Differentiation of lymphoedema (LE) risk factors for those who have undergone a sentinel node biopsy (SNB)
from those who have undergone axillary node dissection (AND) is not considered, even though the incidence rates for the two are
vastly different. In addition, events women are typically cautioned against have not been well investigated.
Methods: A prospective study was conducted in which women were recruited and assessed prior to surgery, and then seen within
4 weeks following surgery, and at 6, 12 and 18 months following surgery. Women were categorised as having LE if their
bioimpedance interlimb ratio exceeded previously established thresholds. Following post-surgery assessment, women were
asked to complete weekly diaries regarding events that occurred in the previous week. Risk factors were broadly grouped into
demographic, lifestyle, breast cancer treatment-related, arm swelling-related, and post-surgical activities (eg, airplane travel).
Crude association between each potential predictor and presence of arm swelling was then identified using unadjusted logistic
regression. Those variables with P<0.2 at this initial screen were considered for inclusion in a logistic regression model. The final
multivariable model retained all variables with P<0.1 or odds ratio> 2.0, taking into account biological plausibility. The final
multivariable models were developed without and with consideration of the presence of swelling in the first year.
Results: 450 women (SNB group: 241; AND group: 209) were recruited and attended the final assessment; a subgroup of 243
women, of whom 112 had AND completed >70% of the weekly diaries. The incidence of LE for the SNB group was 3.3% (n=8)
and 18.2% (n=38) for the AND group. The unadjusted risk factors for LE at 18 months for SNB included high BMI and absolute
body weight, living alone and presenting at diagnosis with three or more other medical conditions. The final model for the SNB
group included a high BMI and not living with a partner, explaining 21.3% of the variance. Inclusion of post-operative swelling in
the model explained 48.4% of the variance. The unadjusted risk factors for LE at 18 months for AND included being older, low
education, Stage 3, high number of nodes removed and involved, and radiotherapy to the axilla, and receiving taxane
chemotherapy. The final risk factors model for the AND group included clinical stage 3, being older, low education, and receiving
taxane-based chemotherapy, explaining 20.4% of the variance. The addition of any swelling within the first 6 months following
surgery explained 36.8% of the variance. Notably none of the factors related to air travel, arm trauma, medical procedures (eg,
blood pressure, injections, blood drawn on the affected side), or exercise which women are typically cautioned against
differentiated women who had and did not get LE at 18 months.
Conclusion: Advice to women undergoing SNB should differ to that provided to those undergoing AND, and for both, we should
not be burdening them with range of behaviors to avoid. Importantly, for women at high risk, periodic assessment in the first year
should occur to identify and manage any arm swelling.
Acknowledgement: Cancer Australia and NBCF.

Title: Determination of the first evidence-based diagnosis of secondary upper limb lymphedema
Sharon L Kilbreath1, Elizabeth S Dylke1, Geoff P Schembri3, Leigh C Ward2, Dale L Bailey1, Deborah Black1, Elizabeth A Bailey3,
Jane M Beith4 and Kathryn M Refshauge1. 1University of Sydney, Sydney, NSW, Australia; 2University of Queensland, Brisbane,
Queensland, Australia; 3Royal North Shore Hospital, St Leonards, NSW, Australia and 4Chris O'Brien Lifehouse, Cameprdown,
NSW, Australia.
Body: Background: The incidence of secondary upper limb lymphedema after treatment for breast cancer is unclear due to the
wide variety of measurement tools and diagnostic thresholds that are used in both the literature and clinical practice. Furthermore,
this lack of clarity in what constitutes lymphedema or not has prevented the progression of the field of lymphedema. Many of the
thresholds have been chosen for ease of use only and have no evidence base to support them. The aim of this study, therefore,
was to determine which clinical diagnostic threshold for unilateral upper limb lymphedema has the best sensitivity and specificity
when compared to diagnosis by lymphoscintigraphy.
Methods: Women with and without a history of secondary upper limb lymphedema were assessed using lymphoscintigraphy,
bioimpedance spectroscopy (BIS) as well as volume and circumference measurements using the perometer. Dermal backflow
score was determined as the diagnostic criteria for the lymphoscintigraphy and was assessed by an experienced nuclear
medicine physician. Determination of the presence of lymphedema by lymphoscintigraphy was compared with diagnosis by both
commonly-used and normatively-determined diagnostic thresholds for circumference, volume and bioimpedance.
Normatively-and commonly-used diagnostic thresholds examined
Whole arm volume
Circumference
Whole arm BIS
Normatively-determined thresholds examined

3SD perometry threshold*
Single elevated circ (3SD threshold)*
3SD Cornish, 2001
2SD perometry threshold*
Adjacent raised circ (3SD threshold)*
2SD Cornish, 2001
3SD frustrum threshold*
Single raised threshold (2 SD)*
3SD, Ward 2011
2SD frustum threshold*
Adjacent raised circ (2SD threshold)*
2SD, Ward 2011
3SD SOAC*
2SD SOAC*
Commonly-used thresholds examined
200 ml interlimb difference
2 cm single interlimb difference
10% difference
Adjacent 2 cm interlimb difference

5 cm SOAC
* Dylke et al, 2012; Sum of arm circumferences (SOAC); Circumference measure (Circ)
Results: For those with widespread dermal backflow, any clinical diagnostic criteria could differentiate between those with and
without lymphedema. In contrast, for those with mild to moderate dermal backflow, only the normatively-determined threshold, set
at 2 standard deviations above the norm, for arm circumference and full arm bioimpedance (Cornish et al 2001) had adequate
sensitivity and specificity. Both of these thresholds had clinically relevant positive (23 and 10 respectively) and negative (0.2 and
0.3) likelihood ratios.
Conclusion: Evidence-based diagnostic thresholds have been established for the diagnosis of secondary upper limb
lymphedema. In determining if lymphoedema is present in those with mild lymphedema, normatively-determined circumference
and bioimpedance thresholds that account for limb dominance should be used. Adoption of these evidence-based criteria will
allow, for the first time, comparison between studies, clarifying the incidence and risk factors for lymphedema, allowing the field to
make meaningful progress forward in determining who is at-risk for lymphedema and how to prevent it from developing.
Acknowledgements: Cancer Australia and National Breast Cancer Foundation.

Title: Impact of radiation on cardiovascular outcomes in breast cancer patients
Angel Qin1, Paula Silverman2, Janice Lyons3 and Cheryl L Thompson4. 1University Hospitals Case Medical Center, Cleveland,
OH; 2Case Comprehensive Cancer Center, Case Western Reserve University, Cleveland, OH; 3University Hospitals Case
Medical Center, Cleveland, OH and 4University Hospitals Case Medical Center, Cleveland, OH.
Body: Background: Radiation therapy is a key component in the treatment of breast cancer. Recent studies have attempted to
detail the potential impact that radiation may have on the development of cardiovascular complications in patients with breast
cancer. The results of these studies, however, are conflicting regarding the magnitude and clinical significance of the
cardiovascular effects with the use of modern radiation techniques. The aim of our study was to examine cardiovascular
outcomes in breast cancer patients who received radiation therapy compared to those who did not.
Methods: After IRB approval, all breast cancer patients who received systemic chemotherapy at University Hospitals Case
Medical Center between 1998 and 2006 and with known radiation data were included in this retrospective study. A complete
patient medical history including risk factors, comorbidities, and tumor and treatment characteristics were obtained from the
institutional tumor registry and were confirmed by the medical record. Billing data provided dates of diagnosis for heart failure
(HF), coronary artery disease (CAD), and arrhythmia. The patients were divided into three groups: no radiation, right sided
radiation, and left sided radiation. The groups were compared to one another and then further subdivided and analyzed based on
timing of diagnosis of CAD and HF. Multivariate analyses using a Cox proportional hazards model were conducted for each
outcome adjusting for age and comorbidities.
Results: 1277 patients were included in this study with average follow-up of 7.3 years (SD =3.5). CT based planning was used for
the majority of the patients. Radiation start dates were not available for 186 patients; these were estimated at 5.5 months after the
initiation of chemotherapy. There was no significant difference between groups in the incidences of the cardiovascular outcomes
(p>0.05). Further analysis after accounting for timing of diagnosis of CAD and HF showed there was a significantly shorter time to
onset of CAD in the left sided radiation group compared to the right sided radiation group; 4.5 years (SD=2.2) compared to 6.5
years (SD=3.6) respectively, p<0.001. However, multivariate analyses were not statistically significant (p=0.47). The onset to HF
was shorter in the right sided radiation group compared to the left sided radiation group; 3.5 years (SD 3.1) compared to 4.3 years
(SD=3.5) respectively, p<0.001, but again, multivariate analyses were not statistically significant (p=0.43).
Discussion: In our study, we did not find a difference in the incidence of adverse cardiovascular outcomes among patients who
did and did not receive radiation, both before and after correcting for timing of diagnosis. We did find a difference in the time to
development of these adverse cardiovascular outcomes between the two radiation groups, but no difference was found in either
the incidence or time to development when controlled for age and co-morbid conditions. This study suggests that there is little
impact of radiation on cardiovascular outcomes using modern techniques that limit radiation exposure. Prospective longitudinal
studies focused specifically on cardiac endpoints may be needed to fully understand the cardiovascular impact of radiation.

Title: Objective markers of fatigue in women undergoing adjuvant chemotherapy for breast cancer
Carolyn Behrendt1, Sarah Waliany1, Paul Frankel1, Sunita Patel1 and Joanne Mortimer1. 1City of Hope Comprehensive Cancer
Center, Duarte, CA.
Body: Background. In breast cancer patients, fatigue is a leading detriment to quality of life. Objective markers of fatigue are
needed to advance research into its treatment and prevention. To that end, we evaluated physical and neurocognitive tests as
potential markers of fatigue before and after chemotherapy.
Methods. Women about to begin adjuvant chemotherapy for breast cancer were studied prospectively. Before and after
chemotherapy, subjects were assessed on self-reported fatigue (Brief Fatigue Inventory, BFI), depression (CES-D), Pittsburgh
Sleep Quality Index, body mass index (BMI), and 18 objective measurements: grip strength in dominant and nondominant hands,
6-minute walk, daily total energy expenditure (TEE) per accelerometer averaged over 1 week, and examiner-administered
neurocognitive tests: grooved peg board in dominant and nondominant hands; digit symbol coding and symbol search (Wechsler
Adult Intelligence Scale, 4th edition); 2 trail-making, 4 color-word interference, and 4 verbal fluency tests (Delis-Kaplan Executive
Functioning System, D-KEFS). Due to correlation between markers, each was evaluated individually as a continuous variable in
generalized linear models of BFI. Models used generalized estimating equations with independent correlation matrix. Covariates
(age, tumor stage, BMI) and potential interaction between marker and time of assessment (pre- or post-chemotherapy) were
considered.
Results. Of subjects enrolled (n=28), 3 withdrew before wearing the device; these were similar to evaluable subjects except for
lower scores on digit symbol (p<0.002) and color inhibition (p<0.02). Another subject (a vigorous athlete age 36) was excluded,
because she alone did not experience the primary endpoint (fatigue) at either assessment. Thus, the final sample was n=24 (age
50.3+9.5 years). BFI before chemotherapy was median 1.17 (interquartile range 0.33-3.50) and increased +2.11(+2.64) after
chemotherapy. Worse sleep quality (p=0.003), greater grip strength in dominant (p=0.001) and nondominant hand (p=0.003), and
higher daily TEE (p=0.03) were each associated with BFI after chemotherapy but not before. In contrast, better performance on
each of 4 neurocognitive tests (D-KEFS color patch p=0.04, color word p<0.001, switch correlation p<0.001, switch accuracy
p<0.001) was inversely associated with BFI before chemotherapy but not afterwards. Other neurocognitive tests and 6-minute
walk were not associated with BFI. Only depression score was associated with BFI before (p<0.0001) and after (p<0.0001)
chemotherapy; moreover, a given level of depression was associated with greater BFI after chemotherapy than before
(interaction with time, p<0.0001). None of the models fit BFI data better when covariates were taken into account or when
markers were considered simultaneously.
Conclusions. Unlike neurocognitive tests, certain physical measures (grip strength in either hand, TEE, sleep quality) can serve
as marker of the fatigue that follows cancer treatment. Grip strength has the advantage of being both objective and efficient to
measure. No marker currently studied is suitable for comparing fatigue pre- versus post-chemotherapy, however.

Title: Effect of a 12-week supervised physical activity and healthy eating program on body weight, functional capacity and serum
biomarkers in survivors of triple-negative breast cancer: A randomized, controlled trial
Linda Vona-Davis1, Jame Abraham2, Daniel Bonner1, Diana Gilleland1, Gerald Hobbs1, Sobha Kurian1, Mary Anne Yanosik1 and
Anne Swisher1. 1West Virginia University, Morgantown, WV and 2Cleveland Clinic, Cleveland, OH.
Body: Regular physical activity and healthy body composition are important predictors of good outcomes for breast cancer
survivors, especially in overweight/obese individuals. However, individualized exercise and healthy eating programs have not
focused on lifestyle changes and outcomes among women recovering from triple-negative breast cancer. Our purpose was to
examine the associations between baseline levels of inflammatory cytokines and obesity-related adipokines after weight loss, diet
and physical activity intervention in survivors of triple-negative breast cancer. We enrolled overweight/obese survivors (average
time since diagnosis, 4 years) and randomly assigned them to a 12-week supervised exercise and low-fat diet (n=18) or a usual
care group (n=10). The program consisted of supervised exercise 3 times/week, as well as 2 unsupervised sessions per week.
The goal of our Get Fit for the Fight program was to complete 150 min/week of moderate-intensity aerobic exercise. Participants
completed a 3-day diet record during baseline testing and the dietitian recommended ways to cut calories from these typical
eating patterns. The goal of caloric restriction was to decrease dietary fat in order for the participant to consume 200 kcal/week
less. Assessments included aerobic fitness, body composition, and self-reported physical activity, mood and quality of life. Blood
cytokines and obesity-related adipokines were also analyzed before and after the intervention period. The intervention group lost
an average of 2% body fat compared with the control group (p<0.01). Significant (p<0.05) decreases were seen in the
intervention group for body weight, BMI, and all skinfold measures except mid-axillary. Self-reported physical activity and
breast-cancer specific quality of life also improved significantly in the intervention group from baseline to 12 weeks, indicating a
shift from inactivity toward increasing time spent in moderate physical activity, primarily during weekends. No significant
associations were observed between the intervention group and serum levels of inflammatory cytokines. However, BMI was
significantly correlated (p<0.05) with serum leptin/adiponectin ratios, an indicator of insulin resistance. These findings indicate that
a pragmatic lifestyle intervention with physical activity and healthy eating were consistent with improvements in body composition,
functional capacity and quality of life for triple-negative breast cancer survivors. The intervention also evoked favorable changes
in blood leptin/adiponectin ratios which are linked to reductions in central adiposity and improved insulin sensitivity.

Title: Risk of breast cancer related lymphedema after treatment with taxane-based chemotherapy: A prospective cohort study
Meyha N Swaroop1, Cynthia L Miller1, Nora Horick2, Chantal M Ferguson1, Melissa N Skolny1, Jean O'Toole3, Lauren S
Jammallo1, Michelle C Specht4 and Alphonse G Taghian1. 1Massachusetts General Hospital, Boston, MA; 2Massachusetts
General Hospital, Boston, MA; 3Massachusetts General Hospital, Boston, MA and 4Massachusetts General Hospital, Boston, MA.
Body: Background: Taxane-based chemotherapy is routinely used in the treatment of breast cancer and has been shown to
improve both disease-free survival (DFS) and overall survival (OS). A common side effect of taxane-based chemotherapy is fluid
retention in the extremities, which may increase the risk of breast cancer related lymphedema (BCRL). BCRL is a chronic swelling
of the arms, breast, or trunk due to accumulation of lymphatic fluid in the interstitial tissues, which has a profoundly negative
impact on quality of life. Little data exists regarding the impact of taxane-based chemotherapy and fluid retention on risk of
developing BCRL. We sought to determine whether receipt of taxane-based chemotherapy for the treatment of breast cancer
increases the risk of BCRL development in a large, prospective cohort of breast cancer patients.
Methods: We identified 569 patients diagnosed with unilateral breast cancer between 2005-2012 who underwent surgery and
prospective screening for BCRL at our institution. All patients included in this analysis had 18 months of post-operative
follow-up. Bilateral arm volume measurements were performed using a perometer preoperatively and every 3-7 months
postoperatively. BCRL was defined as a relative volume change (RVC) of 10%. Clinicopathologic characteristics and treatment
details were obtained by medical record review. Cox proportional hazard analyses were performed to analyze risk of BCRL. Arm
measurements obtained after contralateral prophylactic surgery or diagnosis of metastasis were excluded to avoid potential
confounding.
Results: Arm volume measurements from 569 patients were included with a median post-operative follow-up of 28 months
(range 18-75.1). 33% (187/569) of patients received taxane-based chemotherapy in the neoadjuvant and/or adjuvant setting, and
92% (172/187) of these patients received pre-medication with dexamethasone to prevent hypersensitivity and reduce edema. 3%
(18/569) received non-taxane based chemotherapy and 64% (364/569) received no chemotherapy. 23% (131/569) had axillary
lymph node dissection (ALND), 61% (346/569) had sentinel lymph node biopsy (SLNB), and 16% (92/569) had no nodal surgery.
At 24 months, the cumulative incidence of BCRL was 5.0% (95% CI: 3.15-7.81%) among patients who did not receive
taxane-based chemotherapy, compared to 13.4% (95% CI: 9.17-19.4%) in the taxane-based chemotherapy group. On univariate
analysis, taxane-based chemotherapy was associated with increased risk of BCRL (HR=2.2, p=0.0037), in addition to ALND,
higher body mass index, greater number of lymph nodes (LNs) dissected and greater number of positive LNs (p<0.05 for all).
However, taxane-based chemotherapy was not associated with increased BCRL risk on multivariate analysis (HR= 0.78, p=0.43),
in which only ALND (HR=4.9, p<0.0001) and number of positive LNs (HR=1.1, p=0.02) remained significant.
Conclusion: Our results suggest that patients who receive taxane-based chemotherapy are not at an increased risk of BCRL
compared with patients who received non-taxane or no chemotherapy. This data can be used to improve patient education and
counsel those who experience temporary fluid retention while on taxane-based chemotherapy.

Title: Improved outcome of breast cancer survivors participating in a multidisciplinary cancer survivorship program at Texas Tech
University Health Sciences in El Paso, TX
Zeina Nahleh1, Rebecca Pasillas1, Alok Dwivedi1, Azadeh Nasrazadani1, Rosalinda Heydarian1, Luis Sanchez1, Indika
Mallawaarachchi1, Cecilia Ochoa1 and Edward Saltzstein1. 1TTUHSC-Paul L.Foster School of Medicine, El Paso, TX.
Body: Introduction: Breast Cancer (BC) survivors in El Paso, TX include a majority of Hispanics. We have previously reported
that these survivors have decreased mental and physical health related Quality of Life (QOL). We sought to determine whether
BC survivors would benefit from participating in a comprehensive multidisciplinary BC Survivorship Program at the Garbar Breast
Care Center (GBCC), as determined by improvement of performance on the following validated questionnaires: 1) Patient Health
Questionnaire 9(PHQ9), 2) General Anxiety Disorder 7 (GAD 7), and SF36 QOL questionnaires. Methods: After IRB approval,
we recruited consecutive patients at our institution over 6 months starting October, 2013, and who are within the first 5 years
post-diagnosis with Stages I-III BC and have completed surgery, chemotherapy and/or radiation therapy. Survivors were enrolled
in the survivorship program staffed by an oncologist, a nurse practitioner, a nutritionist and a clinical psychologist. They
participate in individual as well as group sessions. Survivors are initially screened for depression and generalized anxiety disorder
using PHQ9 and GAD 7, in addition to assessing the patients QOL at baseline, and every 3 months. Survivors are provided with
a personalized summary of all received treatment and follow up care plans; dietary counselling and individual meal plans, in
addition to in-depth psychological assessment and an intervention using Mindfulness Based Stress Reduction. Results: 47
patients have been recruited so far and all but one completed baseline questionnaires;17 and 7 respectively had 3 and 6 months
follow up visits. 94% were Hispanics. Mean age 54 years. 38% of participants were younger than 50 years of age; Stage 1 BC
(36%), Stage 2 (36%), and stage 3 (28%) . 80% received chemotherapy ; and 68% had hormone receptor positive BC and
received endocrine therapy. The scores at baseline were : PCS representing the mean for the SF-36 QOL Physical Health was
46.0, and the MCS for Mental Health was 44.0, both below the population norm (50.0). Mean scores for GAD 7 was 7.54 and for
PHQ 9 7.33, both abnormal (<5 on PHQ9 and GAD 7 are considered normal scores). At 3 months the scores were as follows:
PCS 49.71, MCS 45.34, both improved (the higher the better); GAD7 5.18; and PHQ9 6.82 (both improved, the lower the better).
At 6 months, same favorable trend continues: PCS 50.07 [SD 5.54, 44.29-53.35]; MCS 49.66 [SD 7.73, 38.93-62.41]; GAD7 5.00
(SD 3.70); and PHQ9 4.57 (SD 3.64).
Conclusion: BC survivors are benefiting from participating in the new BC Survivorship Program launched at Texas Tech in El
Paso, TX. They have experienced improved mental and physical health related QOL, improved anxiety and depression by GAD7,
and PHQ9, with scores reaching near normal values at 6 months into the program which continues to enroll BC survivors. This
program represents a culturally appropriate model for Hispanic BC survivors. It is expected to continue to improve the quality of
life of these patients, empower them in their transition from cancer treatment to survivorship and lead to improved psychosocial
adjustment and normal social functioning with significant implication not only on survivors, but also on their families and the
community.

Title: Randomized clinical trial assessing the impact of survivorship care plans on survivor knowledge
Amye J Tevaarwerk1, Kevin A Buhr1, Kari B Wisinski1, Mark Burkard1, Mindy Gribble2, Willam Hocking2, Wenjun Sun1, SarahMaria
Donohue1, Jamie Zeal1, Abigail Terhaar1, Douglas A Wiegman1 and Mary E Sesto1. 1University of Wisconsin, Madison, WI and
2
Marshfield Clinics, Marshfield, WI, United Kingdom.
Body: Purpose: Efforts to inform survivors about long-term risks and planned follow-up after cancer treatment have increased.
Survivorship Care Plans (SCPs) and care planning sessions have been recommended since 2005, yet the benefits of both SCP
provision and care planning are only now being assessed. The impact of SCPs separate from care planning sessions is unclear,
however SCPs alone require less time and cost to provide than SCPs combined with care planning sessions. We hypothesized
that SCPs alone might enhance patient knowledge. In a randomized trial, we assessed change in patient knowledge of diagnosis,
treatment, late/chronic side effects and followup care pre and post receipt of SCP.
Methods: Patients who completed primary treatment within the past 2 years for Stage 0-3 breast cancer were consented,
enrolled, and randomized to immediate (intervention) versus delayed receipt (control) of an individualized SCP without care
planning. All participants completed the Wisconsin Survey of DiagnOsis and Management in Breast cancer (WiSDOM-B), a test
of knowledge about ones own breast cancer diagnosis, treatment and late effects scored out of 40 points. The survey was
completed at both baseline and at four weeks (prior to delayed receipt); the primary outcome was change in score. The study was
designed with 90% power to detect a between-group difference of 4 points (out of 40 possible).
Results: Between November 2013 and March 2014, we recruited 64 women aged 36-78 with Stage 0-3 breast cancer. Most
participants were Stage 1-2 (n=50, 79%) with n=7 (10.9%) in Stage 3. Most had received chemotherapy (62.5%), endocrine
therapy (89.1%), or radiation (76.6%).
At baseline, the average WiSDOM-B score was 28.3 out of 40 (70.8%), range 15.5-38.5. There was no evidence of change in
score from baseline to four weeks for either the immediate SCP group (+0.41, 95% CI [-0.97,+1.79]) or delayed receipt control
group (+0.95, 95% CI [-0.48,2.38]). Observed variation was consistent with sample size assumptions, and the observed treatment
difference (-0.55, 95% CI [-2.53,1.44], p=n.s.) ruled out the prespecified clinically significant effect size of +4.0 (+10%).
Overall, participants scored better on questions testing knowledge of diagnosis (side, year of diagnosis, lymph node test results,
receptor status, and stage) and surgical side effects than on questions testing knowledge of other treatment (chemotherapy,
radiation or endocrine therapy) side effects.
Participant satisfaction with knowledge and care team communication was collected pre and post SCP, as well as feedback on
SCP content and use of the SCP. These and analyses of selected questions shown to change post SCP by Nissen et al will also
be presented.
Conclusion: In a controlled, randomized clinical trial, receipt of a SCP without care planning sessions did not appear to
significantly increase survivor knowledge about cancer diagnosis, treatment and followup as assessed by the WiSDOM-B survey.
Efforts to improve survivor knowledge should investigate the impact of care planning sessions or other interactive health
information tools. A second study of patients (n=64) randomized to immediate or delayed receipt of SCPs, in conjunction with
care planning sessions, is currently enrolling.

Title: Exercise and the influence of endocrine therapy on cardiac and metabolic outcomes in breast cancer survivors
M Tish Knobf1, Lyndsey Harris2, Karl Insogna1, Barbara Smith3 and Sangchoon Jeon1. 1Yale University, New Haven, CT; 2Case
Western, Cleveland, OH and 3Michigan State University, Lansing, MI.
Body: Exercise and the Influence of Endocrine Therapy on Cardiac and Metabolic Outcomes in Breast Cancer Survivors
Background: Breast cancer survivors (BCS) are at higher risk for decreased cardiovascular fitness and illnesses due to excess
weight and sedentary lifestyle. Exercise has been shown to improve fitness, body composition and helps to maintain weight but
sparse data exists on the influence of endocrine therapy on these outcomes.
Methods: The Yale Fitness Intervention Trial (Yale FIT) was a 12 month RCT comparing aerobic-resistance exercise to home
based physical activity. Female cancer survivors who were < 3 years of completing primary and/or adjuvant chemotherapy and
within the first 5 years of menopause therapy were eligible. The aerobic-resistance intervention was conducted at a community
fitness center 3 times/week and included 30 minutes of aerobic and 20-30 minutes of resistance exercise, supervised by an
interventionist for first 6 months. The home based group received a flyer outlining recommendations for 30 minutes moderate
intensity physical activity most days of week. Data were collected at baseline, 6 and 12 months.
Results: The majority of the 154 women enrolled were white, married, employed and average age was 52 years. At baseline,
mean BMI was 29 (+ 6.8), waist circumference 86.3 (+ 15.3) and subjects reported expending a median of 773 MET minutes per
week. A graded exercise stress test was used to measure cardiovascular function. The fitness center group increased time on
treadmill (p=0.05) and had improved heart rate recovery (p=0.01) when compared to the home based group. No significant
differences were found across time for metabolic outcomes by fitness center or home based group. However, when analyzed by
type of endocrine therapy (30% were on an Aromatase Inhibitor (AI), 31% on Tamoxifen (TAM) and 39% were on no endocrine
(E) therapy), there was a significant difference in serum insulin and insulin resistance between the fitness center and home based
group for women taking TAM at 6 months (p=0.03) and 12 months (p=0.01). Specifically the insulin levels in the fitness center
group remained stable while the home based group demonstrated increased serum insulin and insulin resistance. There was a
trend for women in the home based groups who gained weight (> 5 lbs) to have higher insulin levels at 12 months (p=0.06). There
were no significant changes at 6 or 12 months by group (fitness center or home based), by type of endocrine therapy or by weight
change for total lipids, triglycerides, or HDL/LDL. To provide additional insight into the findings, analysis of cytokines and
inflammatory markers is in progress.
Conclusions: This study confirms the cardiovascular fitness benefits of exercise but the relationship between exercise, endocrine
therapy and weight on insulin levels is more complex. Exercise (aerobic-resistance and aerobic) combined with stable weight over
12 months resulted in relatively unchanged insulin levels. In contrast, only aerobic-resistance exercise (fitness center group)
appeared to stabilize insulin levels in women who gained weight.
Funding: NIH/NCI R01CA122658.

Title: No Show

Title: Perceptions and gaps of women living with advanced breast cancer: Results from the "Count Us, Know Us, Join Us" online
survey in Latin America
Maira Caleffi1. 1Hospital Moinhos de Vento, Porto Alegre, Brazil.
Body: Background: Patients with advanced breast cancer (ABC) represent a subgroup of BC patients who have a unique set of
needs in their multiprofessional care, as the focus is more concerned with surviving the challenges of the disease, rather than
being a survivor. However, little is known about these specific needs. A global Internet-based survey, Count Us, Know Us, Join
Us," was performed to identify and raise awareness of unmet needs of women living with ABC. Herein, the outcomes of
respondents in Argentina, Brazil, and Mexico are presented.
Methods: The survey, conducted by Harris Poll between Oct 2012 March 2013, sponsored by Novartis Oncology, comprised 40
questions to assess respondent demographics and baseline characteristics, diagnosis, metastatic spread of disease, overall
health, family history, communication with the health care practitioner (HCP), information sources and resources, treatment,
support, quality of life (QOL), satisfaction, emotional state, and economic impact. It was distributed online through locally based
patient advocacy groups. Women 21 years old with BC that had spread to distant parts of the body beyond the breast and its
local lymph nodes (stage IV) were eligible to participate.
Results: Overall, 1,273 participants from 12 countries completed the survey, including 302 respondents from Latin America
(n=100, Argentina; n=100, Brazil; n=102 Mexico). Of these, 62% were between 4059 years old and 59% had metastatic disease
at diagnosis. The majority of patients in Brazil (64%) and Mexico (74%), but not Argentina (22%), were very satisfied with the
communication they had with their HCP, although overall 74% wanted their HCP to also address their emotional needs. Just over
a quarter (28%) felt isolated from the early stage BC community, and two-thirds (67%) felt no one understood what they were
going through; 41% experienced a decrease in support from family and friends after being diagnosed. In total, 74% of women felt
their QOL had been impaired moderately or strongly following diagnosis. Around half of respondents expressed dissatisfaction
with aspects of their mental health, including outlook on life (43%), sense of control over their life (48%), emotional state (50%),
and self-esteem (47%) following diagnosis. Moreover, moderate or strong negative impacts on jobs (67%) and finances (71%)
were reported by applicable respondents. An impact on work was indicated by 89% of employed women and overall 66% stated
that ABC interfered with their ability to work such that they suffered a loss of personal income, with the highest figure for
respondents in Mexico (85%).
Conclusions: This survey has identified that ABC has a major impact on many aspects of a sufferers life. One of the key findings
is the need for a larger psychosocial component in the care of women with ABC in Latin America. Engaging the commitment and
collaboration of a multiprofessional care team, caregivers, patient organizations, and governments to improve the quality and
breadth of care beyond therapeutic aspects may help to fulfill this unmet need and improve the general well-being of this often
invisible patient population.

Title: Patients views of follow-up after treatment for breast cancer. A comparison of two approaches
Samantha T Muktar1, Paul TR Thiruchelvam1 and Dimitri Hadjiminas1. 1Imperial College Healthcare NHS Trust, London, United
Kingdom.
Body: Background: Routine follow up of patients' post treatment of breast cancer is standard practice in most countries. Follow
up involves regularly scheduled clinical appointments with the aim of detecting early breast cancer recurrence and provision of
psychological support to the patient. In the United Kingdom, financial constraint has led individual hospital trusts to revaluate the
need for lengthy follow up schedules. Development of novel, less time intensive follow-up services, such as open access follow
up favour a more patient-led approach.
Aim: To assess patients' views on breast cancer follow up, as well as the effect on patient satisfaction of transferring current
clinical follow up to an 'open access follow up'.
Method: We report the patients views on the basis of pooled data of a detailed survey performed in a large Breast Cancer
Centre. All patients receiving regular clinical follow up care over a 6-month period were invited to participate in this prospective
study. Patients were provided with a flow-chart, illustrating the current follow up, as well as a proposed open access follow up
process.
Results: Between November 2013 and April 2014, 304 patients were recruited into the study. 39% of patients were within the first
year of their diagnosis with 18% more than 3 years into their follow up. Caucasian women made up the majority of our population
group (81%), with 7% Indo-Asian, and 7% Afro-Caribbean.
The main expectation from follow up was surveillance for early detection of recurrence as expressed by 92% and anxiety of
treatment side effects. 93% were satisfied with the current follow up they were receiving (satisfaction scores 7-10) and of those,
84% would choose to continue current follow up rather than move toward an open access approach.
92% of patients favoured current clinical follow up over open access follow up, with 66% highlighting open access follow up as
an ineffective method of follow up. In stark contrast, 94% of patients reported current clinical led follow up to be effective. 91% of
patients requested their follow up to be led by a breast surgeon and oncologist, rather than their primary care physician or
community nurse. Interestingly, no significant correlation was established between age, ethnic background, distance from hospital
and time from diagnosis with the type of follow up preferred.
Conclusion: Following treatment for breast cancer, patients prefer a more regular clinician-led service to a patient-led open
access follow up process. This may be explained by the observation that patients seek the reassurance of regular clinical review
to identify recurrence early, however the psychological support of a clinical consultation cannot be underestimated. It may be
inferred that patients who are satisfied in the follow up they were currently receiving are more likely to appreciate the current
follow up system rather than moving to an open access' approach. Further study is warranted which investigates the impact of
intensified surveillance on survival based on identification of recurrence and to repeat this study to seek consistency of opinion.

Title: Evaluating alcohol and tobacco use in an ethnically diverse sample of breast cancer patients: Implications for survivorship
Vivian J Bea1,2, Joan E Cunningham2,3, Dana R Burshell2,3 and Marvella E Ford2,3. 1Medical University of South Carolina,
Charleston, SC; 2Medical University of South Carolina Hollings Cancer Center, Charleston, SC and 3Medical University of South
Carolina, Charleston, SC.
Body: Background/Objective: Data suggest that modifiable risk factors such as alcohol and tobacco use may increase the risk
of breast cancer (BC) recurrence and reduce survival. According to 2012 data from the Centers for Disease Control (CDC) and
Prevention Behavioral Risk Factor Surveillance System (BRFSS), 4.8% of women in South Carolina (SC) are heavy drinkers (>1
drink/day) with 9.6% having 4 drinks at least once in the past month (US state medians are 5.2% and 11.4%); 19.1% are current
smokers (US state median 17.4%). Female BC mortality in SC is almost 50% higher in African Americans (AAs) than in European
Americans (EAs; 27.5/100,000 and 19.1%, respectively). In SC, there is an AA subpopulation, the Sea Islanders, who carry the
lowest AA rate of European genetic admixture. Given the substantial racial survival disparity we examined the patterns of alcohol
and tobacco use in an on-going, ethnically diverse statewide study of women with recently diagnosed invasive breast cancer.
Methods: Participants were identified within 18 months post-diagnosis through the SC Central Cancer Registry (SCCCR).
Women who opted into the study were interviewed via telephone, self-reporting data including race/ethnicity, educational status,
alcohol consumption and tobacco use during the past 30 days. Published CDC guidelines were used to categorize alcohol and
tobacco use.
Results: During the first 24 months of recruitment, 162 women have opted into our study; 132 have been interviewed and results
analyzed (37 EAs and 95 AAs). Age at interview ranged from 38.9 to 90.2 years, with AAs slightly younger (p=0.071). Over half
had more than a high school education with no difference by race (p=0.203).
Alcohol: The minority of participants self-identified consuming alcohol (28.8%). Heavy use was infrequent: two (1.5%) reported
consuming on average >1 drink/day,and eight (6.1%) consumed 4 drinks on any one day. Consumption was less prevalent
among AAs than EAs (p=0.063). Use tended to be lower among AAs (statistically not significant).
Tobacco: Smoking (daily or occasional) was reported by 12.1% of participants (AA vs EA: p=0.771).
Alcohol or Tobacco: Use of alcohol and/or tobacco was 1.6 times as prevalent among EAs compared to AAs (p=0.053).
Conclusions: Compared to self-reported state data few participants reported heavy alcohol consumption or current tobacco use,
particularly AAs, but 35.6% of participants do use alcohol or tobacco. While these findings suggest that alcohol and tobacco may
not contribute to the racial disparities in breast cancer mortality observed in SC, it is nonetheless imperative to reduce these
modifiable risk factors and improve breast cancer outcomes for all breast cancer survivors, regardless of race and ethnicity.
Table 1
African American (AA) N=96
European American (EA) N=43
p-value
Age (years): mean (st dev*)
58.8 (11.6)
62.9 (13.2)
0.066
Education: more than high school
51 (53.1%)
29 (67.4%)
0.114
22 (23.2%)
18 (41.9%)
0.025
1 (4.6%)
1 (5.6%)
1.000
Yes (vs No)
7 (7.6%)
3 (7.1%)
Among smokers: Daily use:
4 (57.1%)
3 (100%)
27 (28.7%)
20 (47.6%)
2 (2.2%)
1 (2.4%)
Alcohol in past 30 days

Yes (vs No)
Among those who drink:
Heavy: average >1 drink/day
Current tobacco use
Alcohol and/or Tobacco

Both Alcohol and Tobacco
1.000
0.475
0.032
1.000
*standard deviation

Title: Metabolic syndrome is prevalent in women with newly diagnosed breast cancer
Louise C Wong1, Carolyn Behrendt1, Robin Smith1 and Joanne E Mortimer1. 1City of Hope, Duarte, CA.
Body: Purpose: We estimate the prevalence of metabolic syndrome (MetSyn) among women with newly diagnosed breast
cancer and explore potential risk factors.
Methods: From January 2012 to May 2014, women with newly diagnosed localized breast cancer were enrolled into a
longitudinal survivorship study. At baseline, results of physical exam, use of concomitant medications, serum lipid profile and
vitamin D were obtained from the medical record, while quality of life variables were measured via questionnaires in English.
MetSyn was defined as at least 3 of 5 criteria: central obesity (BMI >30 or waist >=88 cm); triglycerides >=150 mg/dL; HDL
cholesterol <50 mg/dL or cholesterol-lowering treatment; systolic >=130 or diastolic >=85 mm Hg blood pressure or treatment for
hypertension; fasting glucose >=100mg/dL or diagnosis of Type 2 diabetes. Risk factors for MetSyn (age, education, serum
vitamin D, self-reported energy in past month) were evaluated using multivariate logistic regression.
Results: Subjects (n=154) were age 52.0 +10.1 years. Prevalence of MetSyn at baseline was 46.1(+4.0)%; for another
1.9(+1.1)% (n=3), MetSyn status was indeterminate due to lack of data on HDL and triglycerides. Among subjects with MetSyn
(n=71), most had been diagnosed with hypertension (81.7%) and diabetes (61.4%). Without MetSyn (n=80), the latter conditions
were much less prevalent (31.3% and 22.5%, respectively, both p<0.0001). Adjusted for race and year of accrual, the likelihood
of MetSyn at baseline was higher per year of age (Odds Ratio (O.R.) 1.05, 95% CI 1.01-1.10, p=0.015), lower per increasing
educational attainment (from less than high school to high school/some college, to bachelors, to graduate degree: O.R. 0.58,
0.36-0.93, p=0.024), and higher per increasing "time in past month with a lot of energy" (from little/no time to most of the time, to
all of the time: O.R. 2.30, 1.11-4.75, p=0.025). An inverse association with serum Vit D (p=0.14) did not reach statistical
significance.
Conclusions: Metabolic syndrome is present in nearly half of women newly diagnosed with breast cancer and is associated with
age, less education, and more energy in past month.

Title: Improving quality of life among Latino breast cancer survivors: A national randomized control trial of patient navigators
using LIVESTRONGs Cancer Navigation Center
Amelie G Ramirez1, Edgar Muoz1, Sandra San Miguel1, Kip Gallion1, Arely Perez1, Leo Castillo2, Sarah R Arvey3 and Frank
Penedo2. 1University of Texas Health Science Center, San Antonio, TX; 2Northwestern University, Feinberg School of Medicine,
Chicago, IL and 3LIVESTRONG, Austin, TX.
Body: Purpose: Latina breast cancer survivors experience an unequal burden of unmet needs after treatment, which compromise
their health care and wellness, experts recommend providing psychological services as an integral part of quality cancer care.
Methods: Redes En Accion: The National Latino Cancer Research Network and LIVESTRONG partnered to conduct a
randomized control trial utilizing trained, bilingual, bicultural patient navigators to improve wellness and access to psychosocial
services among non-metastatic Latino breast cancer survivors from Texas and Chicago. The trial tests the efficacy of patient
navigation (PN) in improving general and disease-specific quality of life (QoL), treatment compliance, and identification of
mechanisms that may promote quality of life. The study involves a 2 X 4 randomized repeated measures design with an
experimental condition (combined PN over three months with access to the LIVESTRONG Cancer Navigation Center [LCNC]
services [PN+LCNC]) versus a control condition (PN only) as the between-groups factor, and time-point
(baseline/pre-randomization [T1]; post-PN [3-months post T1; T2], and 6 [T3] and 12 months [T4] follow up after T2) as the within
groups factor. LCNC provides free, bilingual support to U.S. cancer survivors throughout the cancer journey. LCNC also refers to
survivors to services addressing their medical, economic and psychosocial needs, and monitors client contact and access
outcomes.
Preliminary Results: A total of 128 Latina women with early stage breast cancer have been randomized into either our control
(Standard Patient Navigation; PN) or our experimental (Enhanced Patient Navigation; EPN) conditions. Complete data for
multiple regression analysis on 76 randomized Latina women who completed baseline and first follow up assessment is available.
At baseline, we identified major stated needs by our participants. These included fear of recurrence, more information regarding
their disease, and assistance in communicating with the medical team. The top three physical functioning concerns stated by our
participants included lack of energy/fatigue, interference with daily routines and poor physical well-being. Consistent with prior
work with Hispanic cancer survivors, our sample reported significantly lower (over 1.5 standard deviation below) general health
related quality of life, physical well-being and emotional well-being as measured by the FACT-G. Exploratory analyses have
begun to reveal that there are several significant moderators of the effects of the intervention across several outcomes. At
conference, we will present quality of life data on breast cancer participants, exploratory analysis, and lessons learned.
Discussion: Limited work has addressed the psychosocial needs of Latina breast cancer survivors. Culturally sensitive patient
navigation (PN) could address these needs and significantly improve breast cancer survivorship.

Title: Symptoms, desire for help, and quality of life among recent breast cancer survivors
Steven C Palmer1, SarahLena L Panzer1, Karen Glanz1, Marilyn M Schapira1, Angela M DeMichele1 and Linda A Jacobs1.
1
University of Pennsylvania, Philadelphia, PA.
Body: Introduction:
Advances in detection and treatment have led to improved survival among breast cancer (BC) patients and 89% can expect >5
year survival. Progress comes at a cost, however, and patients experience lasting effects from disease and treatment. It is not
clear how these effects impact quality of life (QoL) achieved by BC survivors. We examined relationships between symptoms,
desire for help, and QoL among recent BC survivors.
Method:
Eligibility included non-metastatic BC treated < 1 year prior and attendance at a survivorship clinic. QoL was assessed with the
SF-12. 19 common symptoms were assessed on a 0-5 bother scale. Participants reported desire for help item for each symptom
rated > 1.
Results:
171 primarily white (73%), middle aged (M=54.9 yrs), and Stage I (58%) BC survivors were recruited. Both Physical (M=48.1) and
Mental (M=53.8) QoL were similar to national norms. Survivors reported an average of 12 symptoms, most commonly Fatigue
(90%), Insomnia (75%), Hot Flashes (73%), and Joint Pain (70%). Most bothersome symptoms included Joint Pain (M=2.9),
Decreased Sexual Drive (M=2.8), Hot Flashes (M=2.8), and Vaginal Dryness (M=2.7). Participants desired help with few
symptoms (M=2.3), primarily Weight Gain (50%), Joint Pain (45%), and Numbness in Hands/Feet (44%). Both Physical and
Mental QoL were negatively associated with number of symptoms experienced (r = -.46 and -.41, respectively) and number of
symptoms for which help was desired (r = -.16 and -.41, respectively).
Conclusion:
Symptoms are common among recent breast cancer survivors and negatively impact Physical and Mental QoL. Desire for help is
less common, though similarly associated with impairment in QoL. Assessing QoL in isolation from symptom burden may miss
important areas for which remediation is possible and could result in improved QoL. Patient education regarding the potential for
and value of symptom reduction may be needed in BC survivors.

Title: Oral health-related quality of life in women with early stage breast cancer
Linda S Taichman1, William G Giannobile1, Thomas M Braun1, Marita R Inglehart1 and Catherine H Van Poznak1. 1University of
Michigan, Ann Arbor, MI.
Body: Introduction: Aromatase inhibitors (AI) are a well-established component of the adjuvant therapy in postmenopausal (PM)
women with hormone receptor positive (HR+) early breast cancer (BCA). AIs are associated with side effects that may adversely
affect the quality of life (Qol). The impact of AIs and on oral health-related QoL (OHRQoL) in women with BCA is unknown. To
generate data on the impact of adjuvant AIs on OHRQoL we performed a prospective, longitudinal, cohort study assessing patient
reported outcomes.
Methods: PM women with early BCA were eligible if they were on an adjuvant AI for 3 to 12 months. AI patients were recruited
from the Breast Medical Oncology Clinic. The Control group consisted of PM women without BCA who were not on AI therapy
and were recruited at the time of screening mammography. Study participants provided socio-demographic information and
completed questionnaires OHRQoL which included the Michigan Oral Health-Related Quality of Life (MOHRQoL) Scale, and Oral
Health Impact Profile (OHIP-14) which both measure discomfort, dysfunction and disability resulting from oral problems on a 5
point scale. Presence of saliva status as a measure of oral health was measured by the saliva flow rate. Data was collected at
baseline and at 6, 12 and 18 months. The baseline data is presented here. The student t-test and chi-square test were used to
analyze outcomes. The Pearson correlation coefficient was used to evaluate the correlation between the OHIP-14 and the
amount of saliva.
Results: The study met its target accrual of 58 PM women; 29 with BCA on AI and 29 controls. Median time on AIs at study entry
was 5.7 months. Demographics were similar regarding age, education, income level, frequency of dental visits, and dental
insurance status across both groups. The OHIP-14 score varied by AI use (p=0.02) and duration of AI use (p=0.03) where the
lowest poor OHRQOL was found in AI users less than 6 months. Patients receiving AI therapy had significantly lower perception
of OHRQoL for the individual OHIP-14 items: "I have had pain in the mouth", "I have had to limit my diet", "I have had painful
aching in my mouth", "I have had to limit foods I eat" and "I have felt tense because of problems with my mouth" compared to
controls.
Conclusions: This study is the first to report on the OHRQoL among PM women with early BCA using AIs. The baseline data
demonstrates a significantly lower OHRQoL in those on AIs compared to controls. Dimensions that were particularly relevant
were physical pain, psychological discomfort and physical disability. Analysis of serial time points is ongoing.

Title: Determinants of weight gain after breast cancer diagnosis: Results from the prospective SU.VI.MAX cohort
Laurent Zelek1,2, Fabien Szabo1, Serge Hercberg1, Pilar Galan1 and Mathilde Touvier1. 1Sorbonne Paris Cit, Nutritional
Epidemiology Research Team (EREN), Epidemiology and Biostatistics Center, Inserm U1153, Inra U1125, Cnam, Universities
Paris 13, Paris 5, and Paris 7, Bobigny, France and 2Avicenne Hospital, Bobigny, France.
Body: PURPOSE To identify correlations between lifestyle related risk factors and weight gain after breast cancer.
MATERIAL AND METHODS The SU.VI.MAX randomized trial was initially designed to assess the effect of a daily antioxidant
supplementation on the incidence of cardiovascular disease and cancer. A total of 13,017 subjects were enrolled in 19941995.
Health events occurring during the follow-up (1994-2007) were self-reported by participants. Pathological reports were used to
extract cancer characteristics Anthropometric measurements were taken one year after inclusion in the study and repeated every
two years during follow-up. Participants were invited to provide a 24-h dietary record every two months.
Among the 258 invasive breast cancers, 169 had complete weight data before and after cancer diagnosis and were used for
analysis. When measured weight and height were not available before diagnosis (n=28), self-reported weight and height were
used instead of measured data. After cancer diagnosis, all anthropometric data were measured by study nurses. Women with
weight gain (WG) 5% (moderate WG) were compared to those with WG <5% by multivariate logistic regression, according to the
following factors: menopausal status, educational level, physical activity (irregular, <1h equivalent walking/day, 1h equivalent
walking/day), smoking status, supplementation group of the trial, weight status before diagnosis, chemotherapy, endocrine
therapy, tumor size, pN, mitoses and grade, ER and PR status, mean daily energy intake before diagnosis (<, median=1865
Kcal/d). The same analysis was performed to compare women with WG 10% (severe WG) to those with WG <10%. We also
calculated the difference between weight before diagnosis and weight at the age of 20 ("lifetime weight change") and we tested its
association with the risk of moderate or severe WG.
RESULTS Mean age at diagnosis was 54.4 y. Mean duration between pre- and post-diagnosis weight measurements was 5.8
years. Among the 169 women included, 59.8% were post-menopausal at diagnosis and 28.4% were already overweight before
cancer diagnosis. 66 women had a WG 5% and 25 had a WG 10%. Women who practiced more physical activity at baseline
were less incline to face moderate WG (OR=0.42 (0.18-0.97), P=0.04). Moderate WG was also directly associated with hormone
receptors status. All other studied factors (including treatment type or energy intake) were not associated with the risk of
moderate WG. Surprisingly, tumor size was directly associated with the risk of severe WG (mean tumor size = 14.6mm among
women with <10% WG versus 20.1mm among women with 10% WG, P=0.04). All other studied factors were not associated with
the risk of severe WG. Lifetime weight variation was not associated with the risk of moderate or severe WG.
CONCLUSION Physical activity seems to have a protective effect against moderate WG. Severe WG raises risk for recurrence
(Caan B, et al. 2012), but since it is also linked with tumor size, we hypothesize that this relation is more complex than a direct
causal effect of WG on recurrence risk.

Title: Surgery for breast cancer in the elderly is associated with better outcomes
Yasmin A Jauhari1, Pam Chaichanavichkij1 and Richard J Sainsbury1. 1Isle of Wight NHS Trust, Isle of Wight, United Kingdom.
Body: Background
Breast cancer often presents differently in the elderly with more neglected disease. There is patchy evidence to support surgery
as a preferred management tool with a positive survival outcome. The objective of this study was to 1. Understand disease
presentation and management of patients diagnosed with breast cancer over the age of 80, and 2. To review survival outcome
between conservative and surgically managed elderly breast cancer patients.
Method
A retrospective single centre case series was conducted from 2008 to 2014 on patients aged 80 and over at breast cancer
diagnosis. Survival outcome between conservative and surgical treatment groups were reported by Kaplan Meier analysis, with
comparison by log-rank test.
Results
There were 156 patients, with a median age of 86.6 years (range 80 107). 67.9% had invasive ductal carcinoma.
At presentation, 9.6% had T4 fungating lesions and 11.9% had distant metastatic disease. The median tumour size was 25mm
(range 4 85mm) and 81.5% were oestrogen receptor positive.
Conservative treatment (44.9%, n = 70) was accounted for in 80% of cases by patient choice and/or pre-morbid status precluding
surgery. 55.1% of patients underwent surgical resections, the majority of which were performed with curative intent (96.4%). 11
patients had surgery under local anaesthetic with no complications. Median duration of hospital stay was 1 day.
Conservative vs. surgical cohorts were comparable in co-morbidities aside from age (89.6 +/- 5.1 vs. 85.2 +/- 4.2 years, p =
0.005)) and dementia (18 vs. 9 (p = 0.015). The overall survival for those treated conservatively was 29 +/- 4.1 months compared
to 40 +/- 6.1 months (p = 0.004) for those having surgery.
Conclusion
The elderly are more likely to present with advanced or neglected breast cancer. Those undergoing surgery had a significant
survival advantage and this modality should be offered unless co-morbidities are severe.

Title: High serum levels of 25-hydroxyvitamin D are associated with better quality-of-life, and lower levels of perceived stress,
depression, and fatigue among breast cancer survivors
Chi-Chen Hong1, Song Yao1, Yali Zhang1, Shicha Kumar1 and Christine Ambrosone1. 1Roswell Park Cancer Institute, Buffalo, NY.
Body: Background: Recent evidence suggests a role for vitamin D in breast cancer etiology and survival. Although data are not
definitive, breast cancer survivors might derive additional benefits from adequate vitamin D levels, including better quality-of-life
(QoL) and psychosocial well-being.
Objective: Using a prospective study design, we examined associations between QoL and psychosocial functioning among
breast cancer survivors and circulating 25(OH)D levels at the time of initial breast cancer diagnosis and 1 year post diagnosis.
Methods: Women (n=504) with incident breast cancer (stages 0-III) were recruited prior to initiating breast cancer treatment at
Roswell Park Cancer Institute (RPCI), with 372 patients also providing data 1 year later. Demographic and clinical data were
obtained from the RPCI Surgical Oncology Breast Database. At each time point, participants provided blood samples and
completed surveys on mental and physical health (SF36 Health Survey), perceived stress (Perceived Stress Scale), depression
(Center for Epidemiologic Studies Depression Scale), and their experiences with fatigue (Multidimensional Assessment of Fatigue
Scale). Serum 25(OH)D levels were measured by immunochemiluminometric assays.
Results: Median 25(OH)D levels at diagnosis and 1 year post were 23.8 ng/ml (range 0.1-67.4) and 26.6 ng/ml (range 3.0
80.0), respectively. At the time of diagnosis, multivariate logistic regression analyses indicated that women with sufficient levels of
25(OH)D ( 30 ng/ml) had 50% lower odds of reporting poor physical health (OR=0.52, 95% CI: 0.29, 0.90) or poor mental
functioning (OR=0.54, 95% CI: 0.31, 0.95) compared to women with deficient levels (<20 ng/ml). Women with higher 25(OH)D
were also less likely to report high levels (>median) of perceived stress (OR=0.45, 95% CI: 0.26-0.80), depression (OR=0.46,
95% CI: 0.25-0.83) or fatigue (OR=0.54, 95% CI: 0.31, 0.92). Findings were similar 1 year later, with reduced odds of poor
physical health (OR=0.47, 95% CI: 0.24-0.93), poor mental health (OR=0.62, 95% CI: 0.33, 1.13), high levels of perceived stress
(OR=0.67, 95% CI: 0.36, 1.25), depression (OR=0.33, 95% CI: 0.15, 0.71), and fatigue (OR=0.45, 95% CI: 0.24, 0.84). Changes
in vitamin D levels over the 1 year period were also assessed. While none of the relationships were significant, women in the
upper tertile of vitamin D change (>3.99 ng/ml) were more likely to report low mental health and higher levels of perceived stress
at 1 year compared to patients with declines in 25(OH)D of >0.95 ng/ml, but were less likely to report poor physical health or
fatigue. This may have been due, in part, to women being somewhat more likely to use vitamin D supplements if they reported
lower mental health QoL scores. Changes in circulating 25(OH)D levels during this 1 year period were strongly associated with
use of supplements (p=0.001).
Conclusions: Breast cancer survivors with adequate vitamin D levels were less likely to report poor QoL compared to women
with deficient vitamin D levels. Maintenance of optimal vitamin D levels may help improve breast cancer patients QoL as they go
through breast cancer treatment.

Title: What is a healthful diet? Cancer survivors' interpretations of their own dietary recall data
Ann C Klassen1,2, Katherine C Smith2, Michelle Brosbe1, Kisha Coa2, Susan Hannum2 and Laura Caulfield2. 1Drexel University
School of Public Health, Philadelphia, PA and 2Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.
Body: Background
The number of long-term cancer survivors will continue to grow, and developing strategies for tertiary prevention of recurrence,
cancer-related comorbidities, and general chronic disease-related health burden is critical. Nutrition and energy balance are
increasingly recognized to play important roles in both decreasing recurrence risk, and improving health and well-being
throughout survivorship. To inform strategies for nutrition education and behavior change for cancer survivors, we used a mixed
methods approach to explore concepts about healthful diet and eating patterns among long-term cancer survivors.
Methods
We collected three ASA-24 dietary recalls from 53 breast, prostate and non-Hodgkins lymphoma survivors. Utilizing qualitative
interviews, we provided participants with feedback on how their eating patterns compared to dietary guidelines, and asked
respondents to reflect upon their diet and dietary changes.
Results
Like most American adults, respondents were not eating in accordance with many dietary recommendations. In terms of
adequate intake of recommended nutrients, 68% met fruit and vegetable guidelines, but only 15% reported adequate intake of
fiber. In terms of following recommendations for limits on intake, 89% reported diets that met guidelines for limiting alcohol intake,
but only 55% met limits for cholesterol, 40% for saturated fat, 19% for caloric intake, and only 2% limited their sodium intake to
within recommended levels. Respondents were often surprised by their own dietary patterns, and had differing conceptualizations
of a "healthful" diet.
Conclusions
Cancer survivorship may provide motivation for dietary change, but survivors need nutrition education to identify specific
strategies for dietary improvement.

Title: A new model of care for breast cancer survivorship
Moira Rushton1, Robin Morash2, Gail Larocque2, Carrie Liska2 and Roanne Segal2. 1University of Ottawa and 2Ottawa Hospital
Cancer Program, Ottawa, ON, Canada.
Body: Background
Breast cancer (BC) is the most common cancer diagnosed in women. With advances in diagnostics and therapy, early stage
disease has an estimated 85% 10-year overall survival. Cancer survivors have unique needs including screening for recurrence,
psychosocial concerns, as well as routine health maintenance. There is growing evidence to support the safe transition from
oncologist to primary care. This is challenging in the era of complex adjuvant and extended adjuvant programs, which are
standard of care for women with hormone receptor positive BC.
In February 2013 the Wellness Beyond Cancer Program (WBCP), developed at the Ottawa Regional Cancer Program, started
accepting BC patients. A risk stratification system enables the treating oncologist to determine the most appropriate care provider
stream based on patient complexity and/or risk of relapse. There are three possible streams: primary care provider (PCP), nurse
practitioner (NP), or oncologist. The program offers education on general and breast specific cancer survivorship issues. Unique
to our program is a hormone re-assessment review (HRR), scheduled in advance, to discuss switching hormonal therapies and/or
extended adjuvant strategies. A rapid re-entry system is in place for those with evidence of disease recurrence or a new primary.
Specific aims
The aims of our study are:
1. To describe unique aspects of WBCP developed for early stage BC patients
2. To assess efficacy of the WBCP
3. To evaluate safety of the HRR
4. To examine patient and primary care giver satisfaction with the program
Methods
All BC patients eligible for discharge through the WBCP will be included in the evaluation. We will evaluate patients by examining
needs assessments completed at time of referral and one year later, as well as completed patient and PCP satisfaction surveys.
The proportion of patients with changing scores, increasing or decreasing, will reflect improvements or worsening of needs. The
proportion of both patients and PCP that are/are not satisfied with the program will be recorded and examined further.
For the HRR we will code the proportion of patients where the consult was requested and was "scheduled vs. not scheduled".
Finally, the number/proportion of patients and reasons for re-entry will be captured.
Results
February 2013June 2014, 1339 BC patients were referred to the program, 584 patients to PCP, 740 to NP, and 15 remaining
with their oncologist. Early results are reported here with results of the surveys expected by December 2014. 7 patients have
required re-entry for disease recurrence and 30 HRR have been booked. No cancer specific negative outcomes have been
reported. Survey results addressing patient needs, empowerment and concerns are being collated, with PCP satisfaction results
to be reported.
Conclusion
Our institution has endorsed evidence-based recommendations for an organized survivorship program and has taken it a step
further by developing a strategy to address the unique long term needs of hormone positive early stage breast cancer. Early
feedback supports that this program has increased patient and caregiver knowledge of survivorship issues and anticipates
improved patient empowerment, without compromising cancer specific outcomes.

Title: Relationship of dietary and red blood cell polyunsaturated fatty acids to inflammatory markers in breast cancer survivors
taking aromatase inhibitors
Tonya S Orchard1, Xueliang Pan1, Joanne Lester1, Amanda Logan1, Charles L Shapiro1, Rebecca D Jackson1, Martha A Belury1,
Lisa Yee1 and Maryam B Lustberg1. 1Ohio State University, Columbus, OH.
Body: Introduction: A large body of literature supports modulation of inflammation by polyunsaturated fatty acids (PUFA).
Inflammation may play a role in the painful joint symptoms commonly experienced by breast cancer survivors taking aromatase
inhibitors (AI). AI-induced joint symptoms negatively impact drug adherence and quality of life in many breast cancer survivors.
N-3 PUFAs, especially eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) have anti-inflammatory effects; in contrast,
excess very long chain n-6 PUFAs may promote peripheral and joint inflammation. We hypothesized that breast cancer survivors
with greater EPA+DHA or n-3/n-6 PUFA in their diet would have decreased inflammation and be less likely to develop AI-induced
joint symptoms.
Materials and methods: This is a secondary analysis of PUFA intake self-reported from food frequency questionnaires and red
blood cell (RBC) PUFA data from a randomized placebo-controlled pilot study comparing n-3 PUFA supplementation vs placebo
in 44 postmenopausal women with breast cancer initiating first line adjuvant AIs. Primary outcomes were feasibility and
tolerability; secondary outcomes were differences in clinical symptoms and inflammatory markers. Participants were randomized
to 4.8 g/day DHA+EPA vs matched placebo for 24 weeks (wks). Serial peripheral blood samples were drawn for RBC PUFA
levels and inflammatory cytokines (IL-6, TNF-R2 and IL-17). Clinical symptoms were assessed by the Brief Pain Inventory short
form (BPI-SF), FACTB-ES, and Stanford Health Assessment -Disability Index (HAS) at 12 and 24 wks. Compliance and toxicity
were evaluated monthly.
Results: Median age of all 44 women enrolled and randomized was 60 years (range 43-76); breast cancer stage I (n=29), II
(n=10), III (n=5); prior taxane (n=15, 34%).Baseline mean dietary intakes of n-3 PUFAs were: total n-3 PUFAs = 1.43 0.86 g/d;
EPA = 0.05 0.05g/d; and DHA = 0.10 0.11 g/d. There was a trend toward lower TNF-R2 with higher dietary intake of EPA
(p=0.09) and DHA (p=0.10). Baseline mean RBC EPA+DHA (n-3 Index) was 4.36% and the RBC n-3/n-6 ratio was 0.20. RBC
total n-3 PUFAs were similar at baseline between n-3 treatment and placebo groups (6.6 1.6%, 7.2 1.9%, p=0.20). RBC
PUFAs were not predictive of baseline IL-6 or IL-17. However, the baseline RBC n-3 Index trended toward an inverse relationship
with TNF-R2 (r = -0.23, p 0.14)
Mean dietary intake of n-3 PUFAs did not change over 24 wks for the 30 women with complete FFQ data and blood samples:
total n-3 = 1.38 0.70 g/d; EPA = 0.06 0.07 g/d; and DHA = 0.13 0.14 g/d. Dietary EPA and DHA were significantly
associated with lower TNF-R2 (r = -0.40, p=0.02 for both) at 24 wks. RBC total n-3 PUFAs were significantly higher in the n-3
treatment vs placebo group at wk 24 (6.5%1.0% vs 15.0%3.3%, p<0.001). RBC EPA+DHA and RBC n-3/n-6 ratio were not
predictive of inflammatory cytokines at 24 wks.
Conclusions: There was a trend toward lower TNF-R2 in breast cancer survivors with higher dietary and RBC EPA and DHA
before starting AIs, but only dietary intake of EPA and DHA was significantly related to lower TNF-R2 after 24 wks of AI therapy.

Title: Crowdsourcing the collateral damage from breast cancer treatment
Susan M Love1, Amaka Obidegwu1 and Christine A Fischetti1. 1Dr. Susan Love Research Foundation, Santa Monica, CA.
Body: INTRODUCTION: Although survivorship research focuses on quality of life after breast cancer treatments, it is usually
initiated by providers and focused on previously reported side effects and consequences of treatment. Patients are often reluctant
to report side effects they may experience during and after cancer treatment and providers are often reticent to ask. In an effort to
explore the patient perspective, the Dr. Susan Love Research Foundation convened a collaboration of advocacy groups to
"crowdsource" womens questions regarding collateral damage from treatment. These concerns will then be incorporated into the
Health of Women [HOW] Study, an ongoing, online cohort study of breast cancer open to anyone aged 18 or older.
METHODS: Emails to current HOW participants were sent out July 2013 to solicit questions about collateral damage. In October
2013, we undertook the design of a new website landing page for the submission of questions, partnered with other breast cancer
advocacy organizations, and began collecting responses. Responses were then categorized to guide questionnaire development.
RESULTS: Sixteen breast cancer and other advocacy organizations came together to support this project. Emails to current HOW
participants resulted in 1191 responses. The website landing page resulted in an additional 3000 responses. Overall, 16.8% of
respondents complained of fatigue, 16.3% of memory problems, 15.7% of anxiety and/or depression, and 14.0% of
numbness/neuropathy. Other less frequently reported problems included problems with nail growth, vision, hearing, urinary tract
infections, and allergy-like symptoms.
CONCLUSION: Through this approach, we received an overwhelming number of responses about collateral damage from
treatment. Many of the issues are known side effects, while others are less commonly reported, but debilitating nonetheless. We
will next compare responses to previously validated questionnaires, develop a comprehensive collateral damage questionnaire
for inclusion in the HOW Study, and generate a report for publication to be distributed to the participants.

Title: A novel candidate for chemotherapy induced alopecia (CIA) through local modulation of apoptosis and inflammation
JiaWei Liu1,2, Saad Harti1, Shiiba Tadafumi1, Reiko Kondo1, Angelo Mello1 and Geert Cauwenbergh1. 1Legacy Healthcare,
Lausanne, Switzerland and 2Harvard Medical School, PPCR, Boston, MA.
Body: INTRODUCTION
CIA, with an incidence of 65%, is considered by the sufferers as the most emotionally distressing side effect of cancer therapies.
Cooling devices, the only option for CIA, have variable efficacy, and requires trained staff. To improve patients quality of life, the
solution needs to address not only prevention, but also recovery (faster, non-patchy, in original quality) of hair, nails, eyebrows
and eyelashes, in a self-medication setting.
Normal hair cycle mainly consists of growth phase (anagen), regression phase (catagen) and resting phase (telogen) that
precedes the shedding phase (exogen). The rapidly dividing anagen cells are damaged by chemotoxic agents and undergo
apoptosis, inducing premature onset of the apoptosis-driven catagen phase. The intrinsic apoptosis pathway is essentially
mitochondria dependent and executed by members of the Bcl-2 family. Androgenetic alopecia (AGA) subjects show remarkably
low level of Bcl-2; Moreover p53 knock-out mice do not undergo CIA.
These indicate that pharmacological inhibition of apoptosis pathways is key to effectively manage excessive hair loss, including
CIA.
In addition, apoptotic cells or cell debris will trigger and sustain scalp inflammatory condition by stimulating production of
pro-inflammatory mediators, delaying hair regrowth process. Dampening inflammation is therefore essential to stimulate hair
regrowth.
APOPTOSIS MODULATION (STUDY N1)
OBJECTIVE: To demonstrate the molecular basis for proof of concept and safety through assessment of the candidates potential
in modulating Bcl-2 towards normal level in AGA subjects.
METHOD: After 3 months topical application, analyze, via immunohistochemistry, Bcl-2 level in scalp biopsies of male AGA
subjects (n=20), and compare with Bcl-2 in non-AGA volunteers (n=25).
RESULT: AGA subjects' Bcl-2 level is depressed (1.7). The candidate restored Bcl-2 (3.2) towards normal level (4.7).
ATTENUATION OF INFLAMMATION (STUDY N2)
OBJECTIVE: To test, in vitro, the anti-inflammatory potential of the candidate.
METHOD: To evaluate TNF-induced expressions of the pro-inflammatory mediators (E-selectin, ICAM-1 and il-8) in endothelial
cells (HUVECs) using specific Antibody Binding Capacity (sABC) technology, measured by flow cytometry.
RESULT: The candidate is able to inhibit TNF-induced expression of inflammatory marker/cytokine: E-selectin, ICAM-1 and il-8
in HUVECs.
OPEN CLINICAL INVESTIGATION (STUDY N3)
OBJECTIVE & METHOD: To assess the candidates efficacy and tolerance, through topical application on female cancer
patients, under supervision of oncologists/nurses (n = 11)
RESULT: Candidate showed promising results in faster / non-patchy / original-quality recovery and prevention of CIA. Nail
damages were generally prevented.
CONCLUSION AND DISCUSSION
In-vivo and in-vitro studies demonstrate that the candidate normalizes scalp apoptosis, a key to prevent CIA, and attenuates scalp
inflammation, a key to promote faster/non-patchy/original-quality hair recovery.
In cancer patients under chemotherapy, the candidate showed promising results in faster/non-patchy/original-quality recovery and
prevention of CIA, although a higher dose may be required. Nail damages were prevented.

Title: Weight loss and bone health in postmenopausal breast cancer survivors
Adetunji T Toriola1, Jingxia Liu1, Patricia A Ganz2, Graham A Colditz1, Lin Yang1, Sonya Izadi1, Anna L Schwartz3 and Kathleen Y
Wolin4. 1Washington University School of Medicine, St Louis, MO, Minor Outlying Islands; 2UCLA School of Medicine, Los
Angeles, CA; 3Idaho State University, Pocatello, Idaho and 4Loyola University College, Chicago, IL.
Body: Background:
The importance of bone health among postmenopausal breast cancer survivors cannot be overemphasized. In addition to bone
loss due to aging, secondary bone loss resulting from treatment is a major concern. Although weight loss has adverse effects on
bone health in postmenopausal women without cancer, the impact of weight loss on bone health in postmenopausal women with
breast cancer is not known. Our objective in this study was to evaluate the effect of weight loss on bone health among
overweight/obese postmenopausal breast cancer survivors participating in a weight loss trial.
Methods: We conducted this study in a subset of women (postmenopausal, N=81) enrolled in the multi-site Exercise and Nutrition
to Enhance Recovery and Good health for You (ENERGY) study. The ENERGY study is a randomized controlled clinical trial
designed to achieve a sustained 7% loss in body weight among overweight/obese breast cancer survivors with stage I, II or IIIA
disease. Weight loss was achieved largely through dietary modification with the addition of physical activity. Bone health was
assessed in a subset of women at a single site using bone turnover markers (BTMs) and dual energy X-ray absorptiometry
(DXA). Markers of bone formation: osteocalcin, bone-specific alkaline phosphatase (BALP), procollagen type I N-terminal
propeptide (PINP) and bone resorption: N-telopeptides of type-I collagen (NTX), C-terminal telopeptide (CTX) were quantified in
blood samples collected at baseline, 6 month and 12-month follow-up. DXA T-score was used to measure bone mineral density at
baseline and at 12-month follow-up. Generalized estimating equations were used to assess differences in mean values over time
as well as from baseline to 6- and 12-month follow-up. Data from intervention and control arm women were used and treated as a
cohort.
Results: The mean age and body mass index (BMI) of women enrolled in our sub-study were 56 years and 31.6kg/m2,
respectively. Majority (54.3%) had stage 2 disease. Mean weight decreased by 5.6 pounds between baseline and 6-month
follow-up and 4.3 pounds between baseline and 12-month follow-up. There were statistically significant decreases in mean
osteocalcin (2.0ng/ml, p-value<0.001), PINP (7.9ng/ml, p-value<0.001) and NTx (3.65nM BCE/L, p-value<0.001), but not BALP
and CTX-1 levels between baseline and 12-month follow-up. No significant changes were observed in mean T-scores, pelvis and
lumbar spine bone mineral densities between baseline and
12-month follow-up. Weight change from baseline to 12-month follow-up was weakly inversely associated with changes in DXA
T-score (r=-0.25, p-value=0.04) and CTX (r=-0.22, p-value=0.07) but not with other BTMs over the same time period.
Conclusion: Weight loss over a 12-month period was associated with decreases in markers of bone formation and resorption but
no changes in BMD in overweight/obese postmenopausal women with stage I, II or IIIA breast cancer. Further studies on the
association of weight loss and the possible modifying effect of physical activity are needed in order to personalize weight control
strategies in overweight/obese breast cancer survivors.

Title: Are physicians "choosing wisely" when imaging for distant metastases in women with early stage (1 or 2) breast cancer? A
population study
Demetrios Simos1, Christina Catley2, Carl vanWalraven3, Ian D Graham4, Dean Fergusson4, Angel Arnaout5, Susan Dent1 and
Mark Clemons1. 1The Ottawa Hospital Cancer Centre, University of Ottawa, and Ottawa Hospital Research Institute, Ottawa, ON,
Canada; 2Ottawa Hospital Research Institute, Ottawa, ON, Canada; 3University of Ottawa, and Ottawa Hospital Research
Institute, Ottawa, ON, Canada; 4Centre for Practice-Changing Research, University of Ottawa, and Ottawa Hospital Research
Institute, Ottawa, ON, Canada and 5University of Ottawa, and Ottawa Hospital Research Institute, Ottawa, ON, Canada.
Body: Background: The probability of detecting radiologically evident distant metastases in asymptomatic patients with early
stage (stage 1 and 2) breast cancer (BC) is low. Published guidelines, including the 2012 American Society of Clinical Oncology
(ASCO) "Top-5" recommendation for "choosing wisely" in oncology, recommend against imaging in these patients. Imaging is
also associated with significant false positive results and the need for additional confirmatory imaging and/or invasive procedures.
Despite this evidence-base, imaging continues to be over-utilized. We decided to quantify the rates of staging imaging in women
with early breast cancer in Canadas largest province to determine whether provincial practice patterns are in keeping with the
spirit of the guidelines and the 2012 ASCO "Top-5" recommendation.
Methods: Provincial registry data available through the Institute for Clinical Evaluative Sciences were used to identify all patients
with a first diagnosis of stage 1 and 2 breast cancer who underwent breast cancer surgery. For each patient, all imaging of the
most common metastatic sites (i.e. skeleton, thorax and abdomen) was captured from the date of diagnosis up until 3 months
after definitive breast cancer surgery. Patients who received neoadjuvant therapy or with a prior history of any invasive
malignancy were excluded.
Results: Between 2007 and 2012, 27,236 patients with stage 1 and 2 BC were identified. Overall data are shown in the table:
No. of
Stage
patients
% having at least
one imaging test
(total)
Total no. of imaging

tests (mean per patient
imaged)
% of patients imaged who

No. of confirmatory imaging
required confirmatory imaging tests performed (% of total
(total)
imaging)
14,113
83% (11,713)
40,464 (3.5)
55% (6,420)
11,202 (28%)
13,123
94% (12,330)
53,729 (4.4)
65% (7,983)
16,292 (30%)
The most common initial (i.e. not confirmatory) imaging modalities of the skeleton, thorax and abdomen, for both groups were,
bone scan (19%; 12,799/66,699), chest x-ray (28%; 18,789/66,699) and abdominal ultrasound (20%; 13,387/66,699) respectively.
Use of advanced imaging (isotope bone scans, CT, MRI, PET) to look at potential sites of metastasis represented 38%
(11,063/29,262) and 46% (17,180/37,437) of initial imaging tests in stage 1 and 2 BC patients respectively.
Conclusion: Imaging for distant metastatic disease in patients with stage 1 and 2 BC is commonly performed in Ontario despite
multiple guidelines recommending against this. Advanced imaging (isotope bone scans, CT, MRI, PET) is commonly used as the
initial imaging modality of choice. Of particular concern is the fact that 60% (14,403/24,043) of BC patients with early stage
disease who were imaged required some form of confirmatory imaging. These results show the importance of the recent ASCO
"Top-5" recommendation against staging imaging in such patients. The reasons for this disconnect between evidence and
practice are not fully understood, but knowledge translation strategies beyond publishing guidelines or recommendations are
required if we are to elicit a meaningful and sustained change in physician practice.

Title: Adjuvant radiation after lumpectomy: A cost comparison of treatment patterns in 43,247 women from the National Cancer
Data Base
Rachel A Greenup1, Rachel Blitzblau1, Kevin Houck1, Janet Horton1, Lynn Howie1, Manisha Palta1, Aimee Mackey1, Randy
Scheri1, Julie A Sosa1, Alphonse G Taghian2, Jeffrey Peppercorn1, Barbara L Smith2 and E Shelley Hwang1. 1Duke Cancer
Institute, Durham, NC and 2Massachusetts General Hospital, Boston, MA.
Body: BACKGROUND: Breast cancer treatment contributes the greatest proportion of cancer-related health care spending in the
United States. Locoregional therapy comprises a significant share of these costs. We hypothesized that among eligible women
with early-stage breast cancer treated with lumpectomy, evidence-based utilization of hypofractionated whole breast radiation or
omission of radiotherapy could substantially reduce cancer-related treatment costs.
METHODS: Using the National Cancer Data Base which captures approximately 70% of all newly diagnosed cancers in the
United States, we identified 43,247 women with clinically node-negative, T1-T2 invasive breast cancers treated with lumpectomy
during 2011. Women with DCIS, those treated with mastectomy, accelerated partial breast irradiation (APBI), or unknown or
questionable radiation regimens were excluded. Adjuvant radiation was categorized into the following regimens: conventionally
fractionated whole breast irradiation therapy (CF-WBI) [25-40 fractions, 45-66 Gy], hypofractionated whole-breast irradiation
(HF-WBI) [15-24 fractions, 40-58 Gy), and lumpectomy without radiation (no RT). Women were considered eligible for no RT if
70 years with T1N0, ER+ breast cancers, and for HF-WBI if 50 years, with T1-T2 N0 invasive breast cancer. Treatment costs
were calculated using Medicare Physician Fee Schedule payment information for 2011, and based on average current procedural
codes billed per regimen. Costs per patient were estimated as follows: CF-WBI $13,358.37, HF-WBI $8,327.98, and lumpectomy
without RT $0. Actual treatment costs were compared to evidence-based, reduced-cost radiation regimens for which patients
were potentially eligible.
RESULTS: Median patient age was 63 years (range 19-90). Median tumor size was 1.2 cm. Of the total study cohort, 84.5% was
eligible for HF-WBI, and 22.3% for no RT. Among the 36,562 (84.5%) patients eligible for treatment with HF-WBI, 28,383 (77.6%)
received radiation therapy. Of these, 22,653 (79.8%) received CF-WBI, 5,289 (18.6%) received HF-WBI, and 441 (1.6%) received
accelerated partial breast irradiation (APBI). Among 9,651 women 70 years with ER+ tumors eligible for no-RT, 4,245 (44.0%)
received CF-WBI, 1,768 (18.3%) received HF-WBI, 153 (1.6%) received APBI, and 3,485 (36.1%) received no RT. 26% of
women received the least expensive evidence-based radiation regimen for which they were eligible, while 67% of patients were
treated with more costly radiation regimens. Estimated costs of actual treatment were $420.2 million during 2011, compared to
$256.2 million had women been treated with the least expensive radiation regimen for which they were eligible. This translates
into an annual cost savings of $164.0 million, a 39% reduction in costs.
CONCLUSIONS: Utilization of evidence-based adjuvant radiation following lumpectomy is associated with reductions in
cancer-related costs in the locoregional treatment of early-stage breast cancer. Although treatment decisions should not be driven
by health care costs alone, consideration of hypofractionated regimens or omission of radiotherapy for patients that fit
evidence-based eligibility criteria could translate into dramatic reductions in annual health care spending.

Title: NSABP B-43 is unlikely to produce a cost-effective treatment strategy for HER2+ DCIS
Michael D Alvarado1, Mitchell T Hayes1, Rajni Sethi2 and Elissa Ozanne3. 1UCSF, San Francisco, CA; 2UCSF, San Francisco, CA
and 3Dartmouth Institute for Health Policy and Clinical Practice, Hanover, NH.
Body: Background. National Surgical Adjuvant Breast and Bowel Project (NSABP) B-43 is an ongoing randomized clinical trial
that compares breast conserving surgery, radiation therapy, and two infusions of trastuzumab to breast-conserving surgery and
radiation only for HER2+ ductal carcinoma in situ (DCIS). We sought to identify, from a societal perspective, the level of risk
reduction observed in this trial that would be required to make treating HER2+ DCIS patients with trastuzumab cost-effective.
Methods. We developed a Markov decision-analytic model to evaluate these treatment strategies in terms of life expectancy,
quality-adjusted life years (QALYs), costs, and the incremental cost-effectiveness ratio (ICER) over 15 years. The results of
NSABP B-43 are pending. We therefore assumed the hazard reduction to be 36%, based on the statistical considerations of the
trial protocol and explored this in sensitivity analyses. A series of one and two-way sensitivity tests were conducted.
Results. The baseline ICER for switching from a breast-conserving surgery and radiation strategy without trastuzumab treatment
to a breast-conserving surgery and radiation strategy with trastuzumab treatment is $1.57 million per life-year gained and
$299,616.73 per QALY gained. Compared to a societal willingness-to-pay threshold of $100,000 per QALY gained, adding
Herceptin treatment is not cost-effective in the treatment of DCIS under these baseline conditions. In a one-way sensitivity test,
we found that the Herceptin strategy only became cost-effective at an 88.3% risk reduction for local recurrence. The model was
not sensitive to any costs or probabilities within 20% of baseline values. The model was sensitive to the utilities of having had
breast-conserving surgery and radiation with and without trastuzumab treatment. It was not sensitive to other utilities within 0.05
of baseline values.
Conclusions. High risk DCIS treated with breast-conserving surgery and radiation has a low risk of local recurrence as
demonstrated in multiple randomized trials. The addition of a costly drug in order to reduce this small risk further is very unlikely to
be cost-effective and could be considered overtreatment of a non-fatal disease that is sufficiently managed with local treatment.
Prospective cost-effectiveness analyses have the potential to guide allocation of valuable resources for breast cancer research,
and should be regularly integrated into the protocol design of large clinical trials.

Title: Post-operative imaging after atypical ductal hyperplasia excision: The findings and costs
Jennifer K Plichta1, Adrienne N Cobb1, Gerard J Abood1, Constantine Godellas1 and Claudia B Perez1. 1Loyola University Health
Systems, Maywood, IL.
Body: Introduction: With a reported incidence of 2-12% in breast biopsy specimens, the appropriate management of atypical
ductal hyperplasia (ADH) remains in evolution. At present, the optimal screening guidelines for patients with high-risk breast
lesions such as ADH remain unclear. Current practices often parallel the surveillance of cancer patients and include a 6 month
interval mammogram prior to resuming annual screening, which may result in unnecessary procedures and financial costs. This
interval mammogram is typically a diagnostic study, which is an additional cost to the patient and healthcare system. The purpose
of this study was to identify interval pathology following initial surgical resection and review associated costs.
Methods: Following institutional review board approval, the pathology database from a single institution was queried for patients
who underwent surgical excision for atypical ductal hyperplasia from 2008 to 2013. Those who did not have follow-up data
available were excluded. Subsequent clinical care was reviewed, including interval imaging and need for additional intervention.
Based on a review of hospital charges from 2013, the average charge for a unilateral diagnostic mammogram (out-patient, digital)
was $382.
Results: There were 55 patients who underwent an excisional biopsy that were diagnosed with ADH and had subsequent
follow-up. The median age was 57 years (range 38-82 years), and the median breast cancer risk assessment score was 2.3% at
5 years (range 0.5-17.9%) and 12.5% lifetime risk (range 2.2-37.6%). Pathology included concurrent lobular carcinoma in situ
(n=1), atypical lobular hyperplasia (n=3), flat epithelial atypia (n=14), and papillary lesions (n=19). In addition to a routine clinical
breast exam, a short-term follow-up diagnostic (ipsilateral) mammogram was performed in 35 patients. Of the 35 interval
mammograms obtained, 31 yielded benign findings on initial imaging, while 4 patients required additional imaging that ultimately
resulted in benign findings. The overall hospital charges for the 35 short interval mammograms alone during this 6 year period
were roughly $13,370. For the patients that resumed annual surveillance, 3 had abnormal mammograms requiring additional
imaging, and no malignancies were identified in this subset of patients. To date, the median physician follow-up is 3 years, and 52
patients have undergone at least one mammogram since their initial imaging; all subsequent findings have been benign for all
patients. When extrapolated to national data, cost savings to the healthcare system from eliminating short interval mammograms
would exceed $12 million annually without compromising clinical outcomes.
Conclusions: Based on our findings, a 6 month follow-up mammogram is not recommended and incurs unnecessary costs to the
patient and healthcare system. In the post-surgical breast, imaging may be misleading and result in additional procedures and
significant charges that ultimately do not affect clinical outcomes. Although a clinical exam is still recommended at 6 months
following surgical excision for a diagnosis of ADH, patients should forego short interval (6 month) imaging and resume annual
mammogram surveillance.

Title: A cost utility analysis from a Chinese health care perspective of Nab-paclitaxel or docetaxel, both as alternatives to
solvent-based paclitaxel in metastatic breast cancer (MBC)
George Dranitsaris1, Bo Yu2, Jennifer King3, Adams Zhang3, Satyn Kaura3 and Zhai Qing2. 1Augmentium Pharma Consulting;
2
Fudan University Shanghai Cancer Center and 3Celgene Corporation, Summit, NJ.
Body: Background: Paclitaxel and docetaxel are commonly used for the treatment of MBC in China. However, one important
drawback, particularly with docetaxel, is the potential for dose-limiting toxicity. To improve the side effect profile and efficacy of
paclitaxel, an albumin-bound formulation (nab-paclitaxel) is currently available in China (Abraxane). To provide health economic
data for China, a cost utility analysis comparing nab-paclitaxel to docetaxel, both as alternatives to paclitaxel was conducted.
Methods: Clinical data was obtained from a meta analysis of randomized trials comparing either nab-paclitaxel (260 mg/m2 q3wk)
or branded docetaxel (100 mg/m2 q3wk) to solvent-based branded paclitaxel (175 mg/m2 q3wk). Health care resource use for the
delivery of chemotherapy and the management of grade 3/4 toxicity was collected from a time and motion study in three Chinese
cancer centers, from a survey of clinicians and from the oncology literature. Using the Time Trade-off technique, treatment
preferences and utility estimates were obtained from interviewing 28 cancer patients from two centres in China. All costs were
reported in 2014 $U.S.
Results: Nab-paclitaxel had the most favourable safety profile characterized with the lowest incidence of grade 3/4 neutropenia,
febrile neutropenia, anemia, emesis and stomatitis. This translated to lower costs for managing the grade 3/4 side effects of
nab-paclitaxel relative to both docetaxel and paclitaxel ($21 vs. $166 vs. $81) In the preference assessment, 22 of 28 (78.6%)
patients selected nab-paclitaxel as their preferred agent. As an alternative to paclitaxel, the incremental cost per quality adjusted
life year (QALY) gained was determined to be more favourable with nab-paclitaxel than docetaxel ($57,900 vs. $130,600).
Conclusions: Nab-paclitaxel is an economically attractive alternative to both paclitaxel and docetaxel in MBC patients, providing a
substantially lower cost per QALY. Additionally in the patient preference survey, the majority of patients surveyed selected
nab-paclitaxel as their preferred agent.

Title: Health economic evaluation of: Bevacizumab plus paclitaxel vs. bevacizumab plus metronomic cyclophosphamide and
capecitabine as first-line therapy in patients with HER2-negative advanced breast cancer: A multicenter, randomized phase III
trial - SAKK 24/09
Klazien W Matter-Walstra1, Martin Bigler2, Matthias Schwenkglenks1, Daniela Bertschi2, Jrg Brechbhl2, Ursula Hasler-Strub3,
Roger von Moos4, Andreas Mller5, Ralph Winterhalder6 and Christoph Rochlitz7. 1ECPM / University Basel, Basel, Switzerland;
2
SAKK Coordinating Center, Bern, Switzerland; 3Cantonal Hospital, St.Gallen, Switzerland; 4Cantonal Hospital, Chur, Switzerland;
5
Cantonal Hospital, Winterthur, Switzerland; 6Cantonal Hospital, Luzern, Switzerland and 7University Hospital, Basel, Switzerland.
Body: Background: Bevacizumab combined with chemotherapy has been shown to improve response rate and progression free
survival in metastatic breast cancer. The aim of the health economic analysis (HEA) of this study was to demonstrate that the
combination regimen of bevacizumab with cyclophosphamide and capecitabine (metronomic chemotherapy) compared to
bevacizumab and paclitaxel treatment decreases overall treatment costs in patients with breast cancer without suffering any
losses in effectiveness.
Methods: In this multicenter, randomized phase III trial, we compared bevacizumab (10 mg/kg i.v. q 2 weeks) with either
paclitaxel (90 mg/m2) i.v. on days 1, 8, and 15 of a 4 week cycle (arm A) or daily oral capecitabine (3x500 mg) and
cyclophosphamide (50 mg, arm B) as first-line treatment in patients with HER2-negative advanced breast cancer. Primary
endpoint of the health economic analysis was the total incurred treatment costs until patients stopped trial treatment (time to trial
treatment stop (TTS)). TTS was defined as treatment stop due to progressive disease, symptom deterioration, unacceptable
adverse events, patient refusal, death or other reasons for withdrawal. The HEA adopted a health system perspective including all
substantial direct medical costs incurred in the treatment of the patient. Health-related quality of life was measured by means of
the EQ-5D utility instrument. Statistical differences between costs in the treatment arms were tested by the Wilcoxon rank-sum
test. A global multivariable linear model, with a gamma distribution and a logarithmic cost transformation was used to analyze the
costs for the two arms controlled for age.
Results: Between September 2010 and December 2012, 147 patients were included at 22 centers in Switzerland, 73 (intention to
treat (ITT) n=71) in arm A and 74 (ITT n=68) in arm B. The clinical study results will be presented at ASCO 2014. In January
2014, 66 patients in arm A and 63 in arm B of the ITT patients had reached TTS and were analyzed. Mean TTS was 7.3 month in
arm A (95%CI 6.38.2, median 5.9; quality adjusted mean 5.9, median 5.1) versus 8.5 month (95%CI 6.710.2, median 6.8;
quality adjusted mean 6.5, median 5.0) in arm B. Total incurred mean costs per patient were CHF 69474 in arm A (95%CI
6062478324, median CHF 61815; mean cost per month CHF 10044) versus CHF 80324 in arm B (95%CI 6297597672,
median CHF 61751; mean cost per month CHF 10229). There were no significant differences in costs between the treatment
arms and age had no significant effect on the results.
Conclusion: Metronomic chemotherapy plus bevacizumab compared to bevacizumab and paclitaxel treatment showed no
substantial reductions in treatment costs. In view of the clinical results the HEA does not favor the metronomic approach. An
incremental cost-utility analysis is planned.

Title: Patients perspectives on using a genomic test for their breast cancer treatment decisions: Disparities in information
exchange
Catherine Chanfreau-Coffinier1, Ninez A Ponce1, Patricia A Ganz1,2, Michele Toscano3, Joanne Armstrong3 and Jennifer S Haas4.
1
UCLA Fielding School of Public Health, Los Angeles, CA; 2UCLA Geffen School of Medicine, Los Angeles, CA; 3Aetna, Hartford,
CT and 4Harvard Medical School, Boston, MA.
Body: Research Objective
Genomic tests predicting tumor recurrence risk may help inform treatment decisions and improve quality of care. Yet differences
in information exchange and patient decision style about genomic tests may further exacerbate disparities in care, particularly if
recommendations conflict with patients preferences or other prognostic information. This study looks at the information exchange
between patients and physicians and patients preferences in decision-making in the context of receiving a genomic test for
early-stage breast cancer.
Study Design
Retrospective study of women covered by a large national health insurer, who had received the most common, commercial
genomic test for early-stage breast cancer prognosis. All eligible minority women (n=705) and a random sample of eligible white
women (n=719) matched by age category, geography and diagnosis year, were invited to a mailed survey later linked to claims
data. We used random-effects logistic regression models (cluster=state) to examine the association of patient self-reported
race/ethnicity with 3 outcomes: 1) knowledge of being tested; 2) patient decision style (assessed with the Control Preference
Scale and categorized as active (patient-based), shared, or passive (physician-based); 3) amount of information received on the
test. Models were adjusted by patient age, education, income, family structure/support, year of testing, and plan type.
Population Studied
Privately insured women aged 64 years old or less, enrolled for at least 9 continuous months, with an insurance claim for the
genomic test in 20092012.
Principal Findings
The 62% response rate yielded 896 respondents, including 108 Hispanics, 112 Black, 97 Asians, and 549 whites. 10% of the
respondents were unaware that they had received the genomic test, with significant differences in this knowledge by
race/ethnicity after controlling for age, education and income (OR= 3.8 for Hispanic and OR = 2.7 for Black women compared to
White, both p < .010). Decision style was reported predominantly as active (47%) and shared (40%), and varied significantly with
age, income, education and race. In particular, Blacks were more likely than Whites to report a passive rather than shared
decision (OR = 1.6; p = .065). Black women were also more likely to rate the information received on the tests clinical
significance as insufficient (OR=2.1; p = .001).
Conclusions
While most women reported more active styles of decision-making, we found significant differences in information exchange that
may lower the odds of a well-informed decision for minority patients. Variation in patient reports may reflect variation in
understanding and retaining information about the genomic test, or in information delivery by physicians.

Title: Telemedicine: Expanding access to cancer genetic services to underserved populations
Linda Patrick-Miller1, Diana Harris2, Evelyn Stevens2, Brian Egleston3, Linda Fleisher2, Rebbeca Mueller2, Amanda Brandt2, Jill
Stopfer2, Susan Domchek2 and Angela Bradbury2. 1University of Chicago, Chicago, IL; 2Unniversity of Pennsylvania, Philadelphia,
PA and 3Fox Chase Cancer Center, Philadelphia, PA.
Body: Background: Given the increasing demand for genetic services and limited genetic workforce, many patients do not
receive recommended pre- and post-test genetic counseling. Telemedicine has been used to expand specialized medical
services to low access populations. The feasibility and outcomes of telemedicine in clinical genetics are not well described.
Methods: Patients at 3 community sites without genetic counseling services received real-time pre-test (V1) and post-test (V2)
counseling for cancer susceptibility with a genetic counselor (GC) at a center of expertise via community sites and host
institutions computers equipped with web cameras and videoconferencing software. Mixed-methods surveys assessed patient
knowledge, satisfaction, psychosocial responses and experiences at baseline (BL), post-V1 and post-V2. We used paired T-tests
to assess change between time points and linear regressions.
Results: Of 100 patients approached, 83% consented to telegenetic services. To date, 57 have completed BL and V1, and 70%
proceeded with genetic testing, 31 patients have received results, including 3 carriers (BRCA2, MSH2, PMS2). Patient
characteristics did not differ between those who agreed to and declined telegenetics. 4% of sessions were aborted due to
technology failures. 30% experienced disconnections but were completed. Nearly all (94%) were satisfied with their telegenetic
experience. Knowledge and satisfaction with telegenetic services significantly increased and general anxiety and depression
significantly decreased. Event related (state) anxiety did not change significantly.
Table 1.
Outcome
BL Mean (sd)
Post-V1 Mean(sd)
Knowledge (6-28)
20.9(2.8)
22.0 (3.0)
20.8 (3.3)
21.8(3.2)
7.4(4.1)
6.6(4.1)
6.6 ((4.0)
5.7 (3.8)
3.9 (3.9)
3.5 (3.4)
3.6 (3.7)
3.4(3.5)
36.0(15.2)
35.7(13.7)
34.6(15.0)
34.5(13.1)
General Anxiety (0-21)

Genarla Depression(0-21)
State Anxiety(20-80)
Satisfaction with Genetic Services
p
0.007
21.5(3.1)
NSS
0,02
5.7 (3.5)
0.06
0,05
2.9 (3.5)
0.07
NSS
32.1(12.5)
NSS
42.2(3.6)
0.002
53.0(5.3)
0.008
39.5(3.(
39.8(4.0)
Satisfaction with Telemedicine
Post-V2 Mean(sd)
51.3(5.6)
51.5(5.7)
Patients reported several advantages to telegenetics (e.g. decreased travel burden) and few disadvantages (e.g. audio
challenges and technical glitches).Conclusions: Telemedicine delivery of cancer genetic services is feasible, identifies genetic
mutation carriers, increases knowledge, decreases anxiety and depression and is associated with high satisfaction, suggesting an
innovative model for delivery of genetic services for patients and community practices without access to local genetic providers.

Title: Receipt of risk reduction strategies among an underserved population of BRCA mutation carriers
Romy Thekkekara1,2, Christina Seelaus1, Maria O"Connell1, Elizabeth Marcus1,2 and Pamela S Ganschow1. 1Stroger Hospital of
Cook County, Chicago, IL and 2Rush University Medical Center, Chicago, IL.
Body: Introduction: Risk reduction strategies for women with deleterious BRCA mutations include risk-reducing mastectomy
(RRM) or screening with annual mammograms and MRIs, chemoprevention and prophylactic bilateral salpingo-oophorectomy
(RRBSO). The objective of the study was to describe the receipt of various risk reduction strategies in an underserved and largely
understudied minority population of BRCA mutation carriers.
Methods: BRCA mutation carriers detected in our center between October 2005 and January 2014 were included in the study.
Participants who had bilateral mastectomies prior to genetic testing were excluded from the analysis of breast cancer reduction
strategies. Women age 40 or older, with at least one ovary intact at baseline and no history of ovarian malignancy or metastatic
cancer were included in the analysis of uptake of RRBSO. Compliance with annual screening mammogram and MRI was defined
as receipt of 2 screening mammograms and 2 MRIs within 2 years of the disclosure of the test results. Medical records were
reviewed to collect demographic data, personal history of cancer and receipt of mammography, MRI, RRM and RRBSO. Uptake
of chemoprevention was not assessed.
Results: Of the 92 BRCA mutation carriers, 89 were women and 3 were men. 35% were Caucasian, 34% were African American
and 26% were Hispanic. Sixty eight percent were uninsured and 24% had public insurance. The mean age at the time of testing
was 44 years (range: 18-71 years). BRCA1 positivity was seen in 54% and BRCA2 positivity was seen in 46%. Of the 89 women,
7 had bilateral mastectomies prior to testing and 15 were lost to follow up. Among the remaining 67 women, 53 (79%) had a
personal history of breast and/or pelvic cancer (affected) and 14 (21%) had no personal history of breast or pelvic cancer
(unaffected). Of the 53 affected women, 14 (26%) received RRMs. The median duration from disclosure to RRM was 104 days
(IQR: 21- 477 days). Among the 39 affected women who did not undergo RRM, 31 had follow up care at our institution for at least
2 years after disclosure, during which 25 (81%) had at least 2 screening mammograms and 13 (42%) had at least 2 screening
MRIs. Among the 14 unaffected women, only 1 (7%) woman underwent RRM. Ten unaffected women had follow up care at our
institution for at least 2 years after testing, during which 4 (40%) had at least 2 mammograms and 2 (20%) had at least 2 MRIs. Of
the 44 women eligible for RRBSO, 8 (13%) women were lost to follow up. Among the 36 remaining, 23 (64%) underwent RRBSO.
Twenty-one out of the 23 who underwent RRBSO had a previous history of cancer. The median duration between disclosure to
RRBSO was 170 days (IQR: 90-454 days). Thirteen women (36%) have not yet received the recommended RRBSO after a
median follow up of 976 days (IQR:465-1591 days).
Conclusions: Among a diverse group of medically underserved minority women who tested positive for a deleterious BRCA
mutation, the receipt of surgical risk reduction strategies and MRI screening is low, especially among women who do not have a
personal history of cancer. Additional studies are needed in this population to examine the factors associated with acceptability
and compliance with standard of care recommendations as well as institutional barriers to receipt of these risk reduction
strategies.

Title: Implementation of cancer risk assessment and genetic testing in underserved patients
Ian K Komenaka1, Lisa M Winton1, Jesse N Nodora2, Lisa Madlensky2, Meredith A Heberer1, Richard Schwab2, Marcia E Bouton1,
Jeffrey N Weitzel3 and Maria Elena Martinez2. 1Maricopa Medical Center, Phoenix, AZ; 2University of California, San Diego, La
Jolla, CA and 3City of Hope National Medical Center.
Body: PURPOSE:
There is great disparity in genetic testing for breast cancer. Hispanic/Latina women with breast cancer are more likely to have
adverse clinical features and have a high prevalence of BRCA mutations. We propose that Academic-Community clinic
partnerships offer great potential to provide access to genetic cancer risk assessment (GCRA) for underserved communities,
including Hispanic/Latina women. The study also evaluated the willingness of these patients to participate in biospecimen
collection.
METHODS:
This study assessed the implementation of a limited GRCA and testing service at a safety net institution from July 1, 2011 to
December 31, 2013. In the 10 years prior, only two breast cancer patients had undergone genetic testing and both were insured.
The inability to perform GCRA was recognized as a critical area of need. Therefore, a breast surgical oncologist received training
with City of Hope National Medical Center. The goal is to provide clinicians the appropriate skills to provide GCRA services in
areas where these are not available.
Three generation pedigrees and sociodemographic information were collected including health literacy, education, self-reported
income, employment status, and insurance status. We conducted a comparison of the patient characteristics along the continuum
of GCRA, genetic testing, and mutation carriers, and for the latter group, we describe the BRCA mutation profile.
RESULTS:
125 patients were offered GCRA and all accepted, of which 70% of this patient population was Hispanic and 66% did not have
health insurance. Of the 125 patients, 84 (67%) were recommended to undergo genetic testing and 81 (96%) agreed. Of the 81
patients who underwent genetic testing, 68 were also asked to participate in the City of Hope Cancer Screening and Prevention
registry and all but one (94%) agreed.
Significant differences between patients who had genetic testing and those who did not were shown for race/ethnicity, insurance,
and family history. A higher percentage of Hispanic patients and patients with no insurance underwent testing. Additional trends in
differences between patients who were tested vs. those who were not were observed for education and health literacy but these
were not statistically significant. Few differences were observed between women who had genetic testing and mutation carriers;
however, the number of carriers was too small to merit statistical testing. Twelve of 81 (15%) patients were found to have
deleterious mutations, seven BRCA 1 and five BRCA 2. Of the 12 mutation carriers, one patient had ovarian cancer and therefore
had already undergone bilateral salpingo-oophorectomy and two others underwent RRSO. Six are either considering RRSO or
getting financial assistance for the operation. The last three are still undergoing breast cancer treatment.
CONCLUSION:
Results of our experience at a safety net hospital with a largely minority and uninsured population show that limited GCRA and
testing can be successfully implemented. The great majority of patients agree to undergo counseling, testing, and participate in
biospecimen research registries. Current recommendations for genetic counseling are far from being met across the country and
this model could be considered for similar safety net populations.

Title: Womens opinion on when to start screening mammography and reasons for not undergoing screening
Ian K Komenaka1, Meredith Heberer1, Jesse Nodora2, Chiu-Hsieh Hsu3, Lisa Winton1, Marcia Bouton1 and Maria Elena Martinez2.
1
Maricopa Medical Center, Phoenix, AZ; 2University of California, San Diego, La Jolla, CA and 3University of Arizona, Tucson, AZ.
Body: Background: In late 2009 significant controversy arose when some screening recommendations were changed to
advocate screening mammography starting at age 50 rather than the long standing recommendation of starting at age 40 years.
More recently some studies have called into question the benefit of screening mammography. The current study was performed
to evaluate patients opinion on when women should start screening mammography and reasons for not undergoing screening.
Methods: Maricopa Medical Center is the safety net hospital in Phoenix, Arizona. 1,157 consecutive patients were seen at the
Breast Clinic from May 2013 to May 2014. Sociodemographic variables were collected including health literacy assessment using
the Newest Vital Sign (NVS) validated screening instrument. Patients were asked when they felt women should start screening
mammography. In addition, in women at least 40 years of age, if they did not undergo screening mammography, they were asked
for the primary reason for not undergoing screening. Differences in patient characteristics were evaluated based on a Fishers
exact test for categorical variables and one-way ANOVA for continuous variables.
Results: Thirteen of the 1,157 consecutive patients were male and excluded. The average age of the 1,144 consecutive female
patients was 45 years. Most patients were Hispanic, underinsured, and had limited health literacy. Overall use of screening
mammography was poor at only 24%. 402 women (35%) felt that age 40 years was the most appropriate time to start screening.
Only 30 women (3%) felt that age 50 years was the most appropriate age. More women, 470 (41%), chose an age younger than
age 40 to start screening. More than half (55%) of these women who chose an age 50 years or younger, however, did not
undergo screening because they felt they had "no problems" or "didnt know" they should get a mammogram despite choosing an
age to start that was below their current age. Only 187 women (32%) cited cost as the reason for not undergoing screening. Other
reasons for not undergoing screening were: physician did not recommend and other medical problems/forgot/too busy. Few
patients (4%) cited problems with mammograms for not undergoing screening and none cited concerns about false positives.
Multivariate analysis showed that patients with adequate health literacy and insurance were more likely to use screening
mammography than patients who were uninsured or had limited health literacy. Family history of breast cancer was not
associated with use of screening mammography.
Conclusions: Use of screening mammography was poor in this underinsured population. Most women felt that screening
mammography should start at age 40 years or younger. More than half of women who did not undergo screening did not do so
because they had "no problems" or "didn't know" they should. Although many women feel that screening should start at age 40,
most women in this population do not understand the concepts of screening and early detection. Interventions to increase use of
screening mammography should focus of the concept of screening as well as the age.

Title: Mammograms on-the-go: Predictors of repeat visits to mobile mammography vans in St. Louis, Missouri
Bettina F Drake1, Salmafatima S Abadin1, Sarah Lyons1, Su-Hsin Chang1, Lauren T Steward1, Susan Kraenzle2 and Melody S
Goodman1. 1Washington University, St Louis, MO and 2Joanne Knight Breast Health Center, The Alvin J. Siteman Cancer Center,
St Louis, MO.
Body: Background: Among women, breast cancer is the most common noncutaneous cancer and second most common
cancerous cause of death. The American Cancer Society recommends that all healthy women over age 40 have a mammogram
annually because early detection and treatment of tumors has been associated with a 15% decrease in breast cancer mortality.
African American women have higher mortality rates than white women and, in general, uninsured women have low rates of
screening. To increase screening rates, mobile mammography has been implemented in many cities. Previous studies have
investigated womens self-reported adherence to screening guidelines at the time of participation in mobile mammography, but no
study has examined if women use it as an annual screening tool.
Objective: The purpose of this study was to determine if women are using mobile mammography vans as their established source
of medical care for breast cancer screening and the factors that predict repeat visits to these vans.
Methods: A prospective cohort study was conducted from 2006 to 2013 in which 8450 women who received a mammogram as
part of Siteman Cancer Centers Breast Health Outreach Program responded to surveys and provided access to their clinical
records. Only visits on the mammography van were included. The predictor variables explored in this study were: urban status,
insurance coverage, age group, race, marital status, mammography experience at baseline visit, employment status, and year of
screening. Data were analyzed using chi-square tests, logistic regression, and negative binomial regression.
Results: Among the study participants, 25.3% (N=2134) had multiple visits to the mobile mammography van. Of these women,
57.2% had good mammography experiences at baseline, 48.2% were from urban settings, 70.6% were uninsured, 51.2% were
ages 50-65, 69.7% were Black, 76.4% were not currently married, and 63.3% were unemployed. Women who were ages 50-65,
uninsured, or Black had a higher odds of a repeat visit to the mobile mammography van compared to women who were ages
40-50, insured, or White (OR=1.135, 95% CI: 1.013-1.271; OR=1.302, 95% CI: 1.146-1.479; OR=1.281, 95% CI: 1.125-1.457),
respectively. However, the odds of having a repeat visit to the van was lower among women who reported a rural zip code or
were unemployed compared to women who provided a suburban zip code or were employed (OR=.503, 95% CI: .411-.616;
OR=.868, 95% CI: .774-.972), respectively.
Conclusion: This study has identified key characteristics of women who are either more or less likely to use mobile
mammography vans as their primary source of medical care for breast cancer screening and have repeat visits. It is important
that mobile mammography is maintained and remains easily accessible to women who continuously use the service. Further
research should be done to discover ways to make mobile mammography a more effective resource for those more likely to use it
for routine screening.

Title: Multivariable analysis of repeat utilization of mobile mammography units: 10 year analysis of a comprehensive cancer
center
Tezo Karedan, Elizabeth Riley1, Lane Roland1, Laura Barkley1, Jianmin Pan1, Shesh Rai1 and Sarah Mizuguchi1. 1University of
Louisville, Louisville, KY.
Body: Introduction: Mobile Mammography Units (MMU) are a model of outreach. This study analyzes the association of race,
insurance and location to MMU utilization. Our group previously reported these variable as independent predictors of MMU repeat
use. This study explores if race, age, insurance and location jointly can predict the utilization of MMU. Methods: From Jan 2001Dec 2010, 48,324 screening mammograms were performed with 21,587 unique subjects. Demographic data was retrospectively
reviewed to identify race/ethnicity, insurance and location for each encounter. Locations were grouped as Corporate (C),
Partnership Clinic (PCl,) Partnership Community (PCo.) Insurance type was classified as Private Insurance (PI,) Medicaid
(MCaid), Medicare (MCare) and Uninsured (UI.) Insurance type was classified based on the primary insurer. Utilization was
defined as 1x or more than 2x in 10 yrs. Descriptive statistics related to different predictors were produced and statistical
comparisons were conducted. P-values were calculated using Chi-square test for comparison between the two groups. Odds ratio
and its 95% confidence interval were provided. Logistics regression analysis is used to jointly model the effect of independent
factors on repeat unitization of MMU. Reference range for race, insurance and location were chosen based on prior work. Results
were declared significant at significance level of 5%. Results: Univariable analysis of race, insurance status, location and age
were predictive of repeat utilization of the MMU over a 10 yrs. (p= value <.001, <.001, <.001 <.002) respectively. All variables,
race, insurance, location and age remained statistically significant for repeat use in multivariable analysis. Blacks (B) were more
likely to repeat compared to Whites (W) and Hispanics (H). MCare patients remained most likely to repeat compared to PI, UI or
MCaid patients in multivariable analysis. Location, PCl and Age 50-64 continued to predict for highest utilization rate within the
respective cohorts. The estimated OR are shown in Table 1.
Table 1
Univariable
pvalue
Multivariable
OR (95% CI)
pvalue
Reference
OR (95% CI)
Race
<.001
<.001
Black
White
<.001
0.880 (0.829-0.934)
<.001
0.897 (0.843-0.955)
Hispanic
<.001
0.643 (0.555-0.745)
<.001
0.701 (0.604-0.813)
Insured
<.001
Private
<.001
0.834 (0.770-0.904)
<.001
0.782 (0.700-0.874)
Medicaid
0.003
0.804 (0.698-0.926)
<.001
0.688 (0.587-0.806)
Uninsured
0.006
0.828 (0.828-0.968)
<.001
0.799 (0.718-0.890)
Location
<.001
Corporate
<.001
0.856 (0.805-0.911)
0.004
0.891 (0.822-0.964)
PCo
<.001
0.714 (0.666-0.766)
<.001
0.732 (0.680-0.787)
Age
0.002
25-39
0.150
0.849 (0.679-1.061)
0.150
0.848 (0.678-1.061)
40-49
<.001
0.895 (0.844-0.949)
<.001
0.901 (0.849-0.956)
65-89
0.083
0.928 (0.654-1.010)
<.001
0.809 (0.724-0.905)
<.001
Medicare
<.001
P CL
<.001
50-64
Conclusions: To our knowledge this is the reported largest database of MMU. Race, insurance status and location remained
independent predictors of repeat utilization. Interestingly an insured subset (MCare) was more likely to repeat utilize the van
regardless of race and location. This may reflect the average age of screening mammography however a better understanding of
the uninsured subset may be important given the intended outreach of the MMU.

Title: The impact of social inequalities on breast cancer mortality in Brazil
Ruffo Freitas-Junior1, Carolina M Gonzaga1, Maria-Paula Curado2, Ana-Luiza L Sousa1 and Marta R Souza1. 1Federal University
of Goias (UFG), Goiania, Goias, Brazil and 2International Prevention Research Institute (IPRI,), Ecully, France.
Body: Introduction: Female breast cancer mortality has decreased considerably in developed nations. In contrast, an increase
has been observed in developing countries. Objective: To describe the impact of social inequalities in female breast cancer
mortality in Brazil, between the years of 1990 and 2011. Methods: Breast cancer mortality data and estimates for the resident
population were obtained from the Brazilian National Health Service (SUS) database for the 1990-2011 period. Age-standardized
mortality rates were calculated (20-39, 40-49, 50-69 and 70 years) by direct standardization using the 1960 standard world
population. Trends were modeled using linear regression, with mortality rates as the dependent variable and the year of death as
the independent variable. The Social Exclusion Index (SEI) and the Human Development Index (HDI) were used to classify the 27
Brazilian states. Pearsons correlation was used to describe the association between the SEI and the HDI and the variations in
mortality rates in each state. Results: Age-standardized mortality rates in Brazil were found to be stable (annual percent change
[APC] = 0.32; 95%CI: -0.1 0.7). Statistically significant decreases in mortality rates were found in the states of Rio Grande do
Sul, Rio de Janeiro and So Paulo. Increases in mortality rates were most notable in the states of Maranho (APC = 11.2;
95%CI: 5.8 16.9), Piau (APC = 9.8; 95%CI: 7.6 12.1) and Paraba (APC = 9.3; 95%CI: 6.0 12.8). There was a statistically
significant correlation between SEI and a change in female breast cancer mortality rates in the Brazilian states between 1990 and
2011 and between HDI and mortality between 2001 and 2011. This reduction was most notable in the Brazilian states with better
socioeconomic conditions. Conclusions: It was observed a direct impact of social inequalities in female breast cancer mortality
rates in Brazil. Reductions in these rates were found in the more developed states, possibly reflecting a better local healthcare.

Title: Early discontinuation of adjuvant chemotherapy in women with early stage breast cancer: The BQUAL study
Alfred I Neugut1,2, Grace C Hillyer2, Lawrence W Kushi3, Lois Lamerato4, Jinjoo Shim2, Dana H Bovberg5, David Nathanson4,
Christine B Ambrisone6, Jeanne S Mandelblatt7, Carol Magai8, Wei Yann Tsai2, Judith S Jacobson2 and Dawn L Hershman1,2.
1
Columbia University Medical Center, New York, NY; 2Kaiser Permanente of Northern California, Oakland, CA; 3Mailman School
of Public Health, New York, NY; 4Henry Ford Health Center, Detroit, MI; 5University of Pittsburgh Cancer Institute, Pittsburgh, PA;
6
Roswell Park, Buffalo, NY; 7Georgetown University, Washington, DC and 8Long Island University, Long Island, NY.
Body: BACKGROUND
Adjuvant chemotherapy for early stage breast cancer decreases recurrence and increases survival. However, early
discontinuation of chemotherapy occurs frequently and has a negative influence on patient outcomes.
METHODS
The Breast Cancer Quality of Care Study (BQUAL) is a prospective cohort study designed to investigate factors associated with
early discontinuation of adjuvant chemotherapy among women diagnosed with non-metastatic breast cancer at three sites in the
U.S between 2006 and 2010 (Columbia University Medical Center, Kaiser-Permanente of Northern California, Henry Ford Health
System). Chemotherapy regimens were classified based on NCCN guidelines. Regimens were further categorized as standard
and non-standard/experimental. Early discontinuation for standard treatments was defined as missing 20% of the recommended
number of treatments for the prescribed regimen. We used multivariate analysis to examine the association between early
discontinuation and sociodemographic factors, tumor characteristics, and baseline psychosocial factors.
RESULTS
Of 1157 women recruited, 478 patients initiated chemotherapy; 35 women received non-standard/experimental chemotherapy
and an additional 17 did not complete all interviews and were excluded from the analysis. Of the remaining 426 patients, 59
(13.9%) did not complete the full course of prescribed chemotherapy. In multivariate analysis, compared to those who completed
their full prescribed course of adjuvant chemotherapy, those who discontinued were more often >50 years of age (p=0.04). Early
discontinuation of chemotherapy was less likely among Asian women (OR 0.12, 95% CI 0.01-0.96), those who held positive
beliefs related to the efficacy of chemotherapy (OR 0.43, 95% CI 0.22-0.81), and those who were more optimistic (OR 0.93, 95%
CI 0.86-0.99). Women prescribed chemotherapy regimens that had more cycles (>5 cycles) or contained paclitaxel/docetaxel
were significantly more likely (OR 7.54, 95% CI 2.68-21.20 and OR 5.02, 95% CI 1.59-15.83, respectively) to discontinue
chemotherapy treatment early than regimens with 6 or less cycles.
CONCLUSIONS
Women prescribed longer regimens were significantly more likely not to complete the full course. Positive beliefs about the
efficacy of treatment were associated with continuation of treatment. Educational interventions focused on the importance of
completing therapy may increase chemotherapy adherence.

Title: Young women with breast cancer: Needs and experiences
Punam Rana1, Jonathan Sussman1, Jenna Ratcliffe1, Margaret Forbes1, Mark Levine1 and Nicole Hodgson1. 1McMaster
University, Hamilton, ON, Canada.
Body: Background
About 11% of all breast cancers in the United States are diagnosed in women younger than the age of 45. Young women with
breast cancer (YWBC) may experience unique physical and psychosocial problems. Meeting the needs of these young women is
often dependent on a patients oncology team. There is a lack of data about the experiences of YWBC in our region including
whether their needs are met and which resources are available and offered to them.
Objectives
To explore the unique physical and psychosocial needs of YWBC in our region (Hamilton and Niagara) and to describe what
information YWBC receive, and which decisions are made, regarding: fertility, breast reconstruction, mastectomy, sexuality and
emotional support.
Methods
This study includes women 40 years of age or less who were diagnosed with invasive breast cancer within 12 months. Using a
qualitative design, participants attended a 1 hour focus group or a 30 minute interview moderated by the research coordinator,
JR. Recordings were transcribed verbatim, with anonymization of any identifying information. Fourteen participants (ages 30 to
40) were included. Co-PI (PR) and JR reviewed the transcripts, complied recurring categories, and discrepancies were resolved
via consensus.
Results
Most participants underwent a timely work-up and diagnosis (64%) but some experienced a delay in diagnosis (37%) as they
were seen as unlikely to have cancer due to age. A majority of the participants felt they were provided enough information for
informed decisions about treatment. However, over one third of the women were not provided any information about social or
community supports and a similar proportion also felt that fertility and sexuality issues were not adequately addressed. Over one
third planned on having bilateral mastectomy and/or immediate reconstruction, regardless of physician recommendation. The
most distressing aspects of the cancer experience included: telling children and parents about the diagnosis; waiting for test
results; fear of recurrence; and re-starting life after breast cancer.
Recurring Themes From YWBC
Recurring Themes
FG1
FG2
FG3
Timely Diagnosis
Delay in Diagnosis
Poorly Communicated
Undergoing B/L mastectomy
Undergoing reconstruction
INT1
INT2
INT3
INT4
Initial Diagnosis
Treatment
Communication
Well informed about Rx plan
Not well-informed about Rx plan
Well informed about fertility

Not well-informed about fertility
Needs met re: sexuality
Needs unmet re: sexuality
Info provided re: supports
Info not provided re: supports
Supports Utilized
Social Worker
Community Services
None
FG=Focus Group; INT=Interview

Conclusion
YWBC feel that they are provided with adequate medical information for treatment decision-making. However there is a gap in
supports for psychosocial, fertility, and sexuality-related needs. Interventions targeted to address these unmet needs should be
evaluated. In the second phase of this study, we will create an instrument to further assess the needs of young women with
breast cancer in our region.

Title: Enhancing compliance with national nutrition recommendations in breast cancer survivors. Experience in an
underprivileged community
Laurent Zelek1,2, Chloe Bodere1, Delphine Bourlier1 and Anne Festa1. 1Oncologie 93, Bobigny, France and 2Avicenne Hospital,
Bobigny, France.
Body: BACKGROUND: To enhance compliance with national nutrition recommendations in breast cancer survivors (BCS), a
3-year program granted by the Regional Health Authority began in 2013 in an area (Seine-Saint-Denis, SSD) which is among the
poorest in France. Median household income is 68% lower than in Paris (+68%), a gap growing with time. Whereas it is widely
admitted in France that 25% of patients are faced with financial difficulties after breast cancer, this proportion reaches 40% in
SSD. We present here intermediate 1-year results.
PATIENTS AND METHODS: Oncologie 93 is a non-profit organization whose aim is to provide supportive care, health education
and individualized assistance to patients and families, and to facilitate timely access to quality medical and psychosocial care.
Vulnerability was evaluated using an 11-item standardized score (EPICES) previously investigated by French Health Examination
Centers. Strictly speaking this score was aimed at measuring precarity, a concept referring to a social condition assumed to face
worsening. This score is more strongly related to health status than the administrative classification of poverty (Sass, Sante
Publique 2006). Vulnerability was defined by a score >30 and considered as severe when >40. Between March 2013 and
December 2013, 27 BCS were enrolled in a 3-month education program including 3 sessions of a professionally led support
group (with dieticians and social workers). Given the high level of poverty in the area and the incidence of financial difficulties in
cancer survivors, a particular attention was paid to comparing the costs of different foods in order to promote affordable dietary
changes. Mean age was 52. Median vulnerability score was 51.8 (0-93.48) and 64% of patients had a score >30. Dietary intakes
were assessed at baseline, and 1 and 6 mos. after the last session. At baseline, although 13 BCS correctly answered the
questionnaire, 17 BCS had difficulties to prepare meals and fatigue was the main reason for 10 of them; 4 BCS were lost for
follow-up before the end of the program. At 1 mo. 15 BCS had knowledge of healthy dietary choices and 11 were ready to
translate it into practice. Of note, 9 BCS decided to enroll in a tailored physical activity program or planned to do it. At 6 mos. only
10 BCS still had knowledge of healthy diet but all of them turned it into practice. Barriers were reported in 10 BCS and included
anxiety or depression (n=8), family (n=6) and asthenia or other treatment side effects (n=5).
CONCLUSION: A short-term dietary intervention is feasible in vulnerable BCS living in an underserved area and improves
adherence to higher quality diet in a meaningful number of patients. In spite of the attention paid to the affordability of dietary
modifications, numerous barriers still exist in this population, mainly related to psychosocial reasons. Furthermore, fatigue
negatively influences the ability to prepare meals. A total of 100 BCS will be included over 3 years, and further studies will
evaluate the impact of the vulnerability level and psychosocial comorbidities on BCS compliance.

Title: Religion-related factors and breast cancer screening among American Muslims
Aasim Padela1, S Murrar1, S Mallick1, Z Hosseinain1, C Liao1, C Ajax1, F Marfani1 and M Peek1. 1University of Chicago, Chicago,
IL.
Body: Background: Cancer disparities research often overlooks the influence a shared religion may have across race and
ethnicity, and thereby misses opportunities to leverage shared religious networks to promote screening. While American Muslims
have low rates of mammography and their health behaviors are strongly influenced by religion, a description of such relationships
is lacking in the literature. Our study fills this gap.
Methods: We conducted a mixed-methods exploration of how religion-related factors impact breast cancer screening practices.
We sampled an ethnically and racially diverse group of Muslim women frequenting mosques and community sites. A survey
incorporated measures of fatalism, religiosity, religious discrimination, and Islamic modesty, while subsequent focus groups
elicited perspectives on how religious beliefs, values and identity impact breast cancer screening intentions. Survey analyses
involved logistic regression models, while focus groups were analyzed using a team-based framework content analysis approach.
Results: Of 240 survey respondents, 72 were Arabs, 71 S. Asians and 59 African Americans. Seventy-five percent had insurance,
while 85% had a PCP. 77% reported ever having a mammogram while only 37% a mammogram within the past 2 years. In
multivariate models, positive religious coping (OR= 0.21;P <.05) and perceived religious discrimination in healthcare (OR=0.74;P
<.05) were negatively associated with having a mammogram in the past two years, while having a primary care physician
(OR=20;P<.01) was positively associated. Ever having a mammogram was positively associated with increasing age
(OR=1.1;P<.05), years of US residency (10-20 yrs OR=11; 20 yrs OR=4.3;P<.05) and knowing someone with breast cancer
(OR=3.5;P<.01). Importantly, ethnic/racial affiliation did not influence mammography rates.
Of 50 focus group participants there were nearly equal numbers of S. Asians, Arabs, and African Americans, 74% reported ever
having a mammogram, with 56% having a mammogram within the past two years. Focus group data revealed that family support
and encouragement strongly impacted screening intentions, and that obtaining screening in a way that accommodated notions of
religious modesty was paramount and prior experiences with such accommodations influenced subsequent intentions. Focus
group participants believed that the mosque is a critical community venue for setting religious mores but is underutilized for health
education and for motivating theological responses to illness. Participants expressed the need for mosque administration and
religious leaders to openly discuss breast health and mammography screening among the Muslim community.
Conclusions: Aspects of religion appear to influence cancer screening behaviors similarly across the socioeconomic, ethnic, and
racial diversity of American Muslims. Promoting biennial mammography screening among American Muslims requires addressing
ideas about religious coping as related to preventive cancer screening and empowering women to combat perceived religious
discrimination. Mosques are underutilized in breast cancer screening interventions but are a ripe setting for religiously-tailored
programming that can address barriers to screening and promote a culture of health in this community.

Title: An evaluation of mobile mammography outreach in urban and rural communities
Lauren Steward1, Susan Kraenzle2, Bettina Drake1, Sarah Lyons1 and Melody Goodman1. 1Washington University School of
Medicine, St Louis, MO and 2Alvin J. Siteman Cancer Center/Barnes Jewish Hospital, St Louis, MO.
Body: Background: Mobile mammography has been used to reach underserved women in a diverse number of settings. In this
work, we demonstrate similarities and differences between rural and urban communities served by a single mobile mammography
unit (MMU) affiliated with a comprehensive cancer center in Missouri.
Methods: An outreach registry of patients serviced by the MMU was created and includes data from medical records and
responses to a brief questionnaire completed at each visit. Data was examined by point of care (urban/rural) to assess the
efficacy of mobile mammography as an outreach strategy in each of these environments. Bivariate analyses were used to
examine the relationships between demographic characteristics such as age, income, race/ethnicity, education, employment
status, marital status, insurance status, and living environment proxy.
Results: Between 2006 and 2013, 9480 women received their care on the mobile mammography van. The sample was stratified
by point of care (urban vs. rural) served, with majority of the women (86%) residing in urban/suburban St. Louis City/County, and
14% in rural regions. Urban zip codes had a lower percentage of women with income greater than $20,000 (12% v. 21%) and
higher percentage of women with income less than $10,000 (49% v. 37%) in comparison to rural communities(p=0.01). There
were higher proportions of black women in urban communities (63%) compared to rural communities (10%; p<0.001). Almost half
(47%) the women that received mobile mammography in rural zip code were married compared to less than a quarter of women
from urban zip codes (24%; p<0.001). Rural communities (83%) had a higher percentage of uninsured women compared to urban
communities (67%; p<0.001). Women in urban and rural communities were similar in respect to age, employment status, and
education.
Conclusions: Mobile mammography has the potential to reach a large population of women with limited educational, financial, and
healthcare resources. Future studies will be needed to determine if increasing the range and extent of mobile units, such as this
one, would be effective in increasing the screening rates of not only women in Missouri, but also in other portions of this country.

Title: Evidence-based programs to reduce breast cancer mortality in Panama City
Anna Cabanes1, Maria Roquebert2, Tauane Cruz1, Blanca Benaglio1 and Becky Royer1. 1Susan G. Komen, Dallas, TX and
2
Gestion Social, Panama City, Panama, Panama.
Body: In Panama, as in other Latin American countries, breast cancer is the most common cause of cancer-related deaths
among women, and is becoming a major public health problem. Breast cancer mortality rates have steadily risen since 2002 and
reflect not only breast cancer incidence and risk factors, but also late-stage diagnosis, quality of cancer care and reduced access
to health care.
Susan G. Komen developed an evidence-based strategic plan for reducing breast cancer mortality focused on increasing
access to breast cancer screening and care. Komens approach resulted from an analysis of the breast cancer continuum of care
in Panama City, aimed at understanding the factors that determine whether women enter the breast cancer continuum of care,
and the barriers that prevent women from completing the continuum. Komen utilized the results of the study to drive programs
and develop evaluation plans.
We conducted four focus groups and individual interviews with 35 breast cancer survivors living in Panama City to assess the
factors that determine access to the breast cancer continuum of care, as well as personal perceptions of breast cancer treatment
and follow up. In addition, we held 69 in-depth interviews to capture the communitys perceived barriers in accessing cancer care
from three stakeholder groups: public and private medical professionals, non-profits, and private insurers. Interviews and focus
groups were audio taped, transcribed and the content analyzed for emergent themes and patterns.
Based on the data, we drafted three different paths of breast cancer care depending on the type of health insurance: Seguro
Social (Social security system), private insurance, or no insurance. In all scenarios we identified health care system deficiencies
within the primary, secondary and tertiary care levels, in particular low medical breast cancer training and lack of delineated
referral guidelines. Community level gaps identified include lack of information, awareness, and social support networks.
To address these gaps, Komen implemented strategic activities that included a program to fund interventions to conduct
awareness campaigns, train primary care medical professionals, and reduce barriers to access breast health and cancer care that
in Panama is mostly localized in the capital. Impact was measured using an impact chart based on evidence-based and best
practice interventions in the areas of public awareness, community and professional education, and continuum of care services.
With an investment of $500,000 over two years we were able to reach 2 million people with breast health and breast cancer
messaging through 92 television and 268 radio events, and 2,500 views through social media. We funded a total of 36 medical
trainings on standard breast health care guidelines by level of care, protocols and referral processes reaching a total of 1,328
medical providers and 65 community leaders, some from indigenous communities (caciques). Finally, we provided housing
assistance to 130 women from rural areas that had traveled to receive breast health and breast cancer services, including
treatment and follow-up if needed.
Our results show that evidence-based programs provide the opportunity to improve service in areas of most need.

Title: Breast cancer prevention awareness and breast examination attitudes among Hong Kong women who work in a medical
environment A pilot study
Janice Tsang1, Siu Ting Yung1, Alston Chiu1, Henry Sze1, Cindy Lai Shan Wong1, Wendy Chan1, Agnes Cheung1, Eve Lai1, Sin
Ting Wong1, To Wai Leung1 and Dora Kwong1. 1Queen Mary Hospital, University of Hong Kong, Hong Kong, China.
Body: Background: Breast Cancer is the commonest female cancer worldwide and the most common female cancer in Hong
Kong. Overseas studies have explored the knowledge and attitudes towards breast cancer, but most are targeted at general
public. Similar studies were done among female health care personnel but mainly overseas with paucity of data within the
Chinese community. This is a pilot study aiming at exploring the breast cancer prevention awareness and breast examination
attitudes among Hong Kong women who work in a medical environment. Methods: This is a cross-sectional, qualitative pilot
study targeting female nurses working at the Queen Mary Hospital, Hong Kong. We planned for 300 samples as pilot study to
validate the questionnaire. This involved self-administrated questionnaires during 1st October 2013 to 30th November 2013.
Altogether, 1800 questionnaires were distributed to nurses at Queen Mary Hospital through the Central Nursing Department.
Results: A total of 317 completed self-administered questionnaires were collected. The age of the respondents ranged between
20-60, with the majority in the age group of 40-49 (28.4%), followed by 20-29 (27.4%). The mean years in practice was 17.
Among the six diseases given for ranking (diabetes mellitus, breast cancer, hypertension, depression, cardiovascular disease,
cerebrovascular disease), breast cancer is ranked as the most important disease for their age (37.2%). However, many
respondents perceived the degree of risk of breast cancer during their life time to be average (34.1% with median score 5) or
even lower (35% under median score 5). The knowledge towards breast cancer risk factor remained limited. Most of them did not
identify late menopause (74.5%), advanced maternal age (70.6%) and young age at menarche (58.7%) as risk factors. For breast
examinations, most respondents had performed breast self-examination (BSE) (91.2%), less than half of them performed BSE on
a regular basis (only 39.4% performed BSE once every month). The need of my age (72.2%) and I am more aware of
breast cancer (54.6%) are claimed to be the major factors encouraging them to perform BSE. Relatively few respondents had
ever undergone clinical breast examination (CBE) (41.0%). Similar to the case of BSE, those who had undergone CBE are found
to practise it irregularly (23.3%), comparing to 10.1% practising CBE once every year. More respondents with secondary
educational level or above had mammography (p=0.000) or ultrasound (p=0.007) done than respondents with primary educational
level or below which is statistically significant. Conclusion: The general breast cancer risk awareness among nurses in this pilot
study is of concern. There is an unmet need to develop breast cancer preventive intervention program targeting at nurses. This
pilot study helps to validate the proposed survey, and serves as the basis for our future design of appropriate intervention to
enhance breast cancer awareness and improve early breast cancer prevention for female healthcare workers. Further study is
warranted to be extend to female doctors, healthcare assistants and ward clerks etc.

Title: CYP19A1 and ESR1 polymorphisms and selected early-onset side effects during combined endocrine therapy in the
IBCSG TEXT trial for premenopausal women with hormone receptor-positive (HR+) early breast cancer
Harriet Johansson1, Kathryn P Gray2, Olivia Pagani3, Meredith M Regan2, Giuseppe Viale1, Valentina Aristarco1, Debora Macis1,
Antonella Puccio1, Susanne Roux4, Rudolf Maibach4, Marco Colleoni1, Manuela Rabaglio-Poretti4, Alan S Coates5, Richard D
Gelber2, Aron Goldhirsch1, Roswitha Kammler4, Bernardo Bonanni1 and Barbara A Walley6. 1European Institute of Oncology,
Milan, Italy; 2IBCSG Statistical Center, Dana-Farber Cancer Institute, Boston, MA; 3Institute of Oncology of Southern Switzerland,
Lugano Viganello, Switzerland; 4IBCSG Coordinating Center, Bern, Switzerland; 5International Breast Cancer Study Group,
Sydney, Australia and 6NCIC Clinical Trials Group, Kingston, ON, Canada.
Body: Background: Single nucleotide polymorphisms (SNPs) of the aromatase enzyme (CYP19A1) and estrogen receptor alpha
(ESR1) may be associated with breast cancer susceptibility and endocrine-mediated side effects. The IBCSG Tamoxifen (Tam)
and Exemestane (Exe) Trial (TEXT) includes a translational research project to assess whether selected SNPs may influence
treatment efficacy and toxicity. We report on early-onset hot flashes and sweating (HF/S) and musculoskeletal symptoms (MS;
myalgia, arthralgia, stiffness) with respect to CYP19A1 and ESR1 SNPs under combined endocrine therapy.
Patients and Methods: 2,672 premenopausal women with HR+ early breast cancer were randomized to treatment with the
GnRH-agonist triptorelin (Trip)+Tam or Trip+Exe for 5 years. Randomization was stratified according to intended use of
chemotherapy (yes/no) and lymph node status (N- vs N+). Estrogen-depletion side effects (HF/S and MS) were recorded at
baseline, 3-monthly during the first year and 6-monthly thereafter using the NCI CTCAE v3.0. DNA was centrally extracted from
whole blood with Qiagen kits. SNPs of CYP19A1 (rs4646 and rs10046) and ESR1 (rs207764, rs2234693 and rs9340799) were
analyzed by a pyrosequencing method (Diatech Pharmacogenetics S.r.l., Jesi, Italy). Control genotypes (wt/wt; wt/v; v/v) for all
SNPs were processed in each run. Logistic regression was used to analyze two endpoints: presence or absence of grade (gr) 2-3
HF/S during first 6 months and gr 2-4 MS during first 12 months. Four genetic models were used to explore associations with side
effects: genetic (wt/wt; wt/v; v/v), additive, dominant and recessive.
Results: DNA was isolated and genotyped for 1970 (74%) consenting women. Clinical characteristics and outcomes of this cohort
were consistent with the overall trial. At baseline median age was 44, median BMI 24 kg/m2 and 86% had regular menses. During
follow-up, 43% reported gr 2-3 HF/S and 27% reported gr 2-4 MS. The 5 SNPs did not deviate from Hardy-Weinberg equilibrium
(p>0.30) and minor allele frequency ranged from 29%-48%. The CYP19A1 SNP rs10046 (C>T) was associated with gr 2-3 HF/S.
Specifically, women with variant homozygote genotype (T/T) had a reduced risk of HF/S (39% T/T vs 45%). The univariate odds
ratio (OR) was 0.77 (95%CI: 0.62-0.96; p=0.02) compared with other variant groups. The multivariate model showed consistent
results after adjusting for age, BMI, menstrual status, chemo use, treatment allocation (Tam vs Exe) and baseline HF/S. The
other 4 SNPs were not associated with the selected side effects.
Conclusions: Our analysis indicates association of HF/S with CYP19A1 rs10046 genetic variants. The lack of association
between early-onset of selected estrogen-depletion side effects and the other 4 SNPs may be masked by the concurrent Trip.
Ovarian reserve before treatment may also influence the impact of SNP genotypes on early-onset side effects. Further analysis
with details of concurrent medications and treatment adherence will contribute to the interpretation of these results.
Funded by Susan G. Komen for the Cure.

Title: Effect of using LHRH analog during chemotherapy (CT) on premature ovarian failure and prognosis in premenopausal
patients with early-stage, hormone receptor-positive breast cancer: The primary analysis of a randomized controlled phase III trial
Jian-Wei Li1, Guang-yu Liu1, Ke-Da Yu1, Ya-jie Ji1, Miao Mo2, Li Lei3, Jiong Wu1, Gen-hong Di1, Yi-feng Hou1, Zhen Hu1, Can-ming
Chen1, Zhen-Zhou Shen1 and Zhi-Ming Shao1. 1Cancer Center and Cancer Institute, Shanghai Medical College, Fudan University,
Shanghai, China; 2Cancer Center and Cancer Institute, Shanghai Medical College, Fudan University, Shanghai, China and 3The
First People's Hospital of Kunming, Kunming, Yunan, China.
Body: Background: Whether administration of LHRH analog during CT in premenopausal patients with early-stage, hormone
receptor-positive disease would reduce CT-induced premature ovarian failure (POF) is still controversial. Moreover, whether
LHRH analog would influence the prognosis of patients is unknown yet. This randomized study is to evaluate whether
administration of LHRH analog during CT would reduce POF and effect the prognosis of breast cancer.
Methods: This is arandomized, controlled phase III clinical trial. Premenopausal patients age <46 with stage I-IIIA,
ER/PR-positive BC to be treated with CT were randomized into two arms. Concurrent arm received CT with goserelin (GN) 3.6
mg SQ starting 0-7 days prior to the first CT dose, and after CT, the researchers determined whether continuous GN to 2-5years
or cessation. Sequential arm received CT without GN, and after CT, the researchers determined GN 2-5 years immediately or
after restoration of ovarian function or bilateral ovariectomy. Neoadjuvant CT was allowed. Five-year tamoxifen after CT was
administered to all patients. The primary endpoint is POF, defined as amenorrhea for the prior 12 months and post-menopausal
FSH or not assessed after last CT dose or last GN dose. Other endpoints include efficacy of neoadjuvant CT and relapse-free
survival (RFS), defined as time to the first of these events: loco-regional recurrence, contralateral breast cancer or distant
metastasis.
Results: Between 2/09 and 5/13, the trial has finished enrollment, 216 patients were enrolled. The median age were 37.5 in
combined arm (n=108) and 39 in sequential arm (n=108), respectively. The median follow-up time was 27.4 months and 25.7
months, respectively. 15 patients and 21 patients received neoadjuvant CT, respectively. There were no significant difference in
age, tumor stage and CT regimens (p>.05). The median cycles of GN were 25, respectively. 47% had complete primary endpoint
data. POF rate were 5/42 (11.9%) in the combined arm and 16/60 (26.7%) in the sequential arm. POF rate (and post-menopausal
FSH) rate were 1/42 (2.4%) in the combined arm and 8/60 (13.3%) in the sequential arm. In neoadjuvant CT subgroup, each has
1 patient achieved pathological complete remission, and there was no significant difference in objective clinical response. There
were 9 patients in the combined arm and 3 patients in the sequential arm had occured RFS events (including 2 and 0 deaths,
respectively, OR=3.18, 95%CI:0.84-12.09, P=.075).
Conclusions:LHRH analog administration with CT might be associated with less POF and did not affect the efficacy of
neoadjuvant CT, however, had no RFS benefit, it may need longer follow-up. We will conduct an interim analysis in November
2014.
Clinicaltrials.gov Registry Number: NCT01712893.


Title: Factors influencing on discontinuation of adjuvant anastrozole in postmenopausal Japanese breast cancer patients: Results
from a prospective multicenter cohort study of patient-reported outcomes
Chiyomi Egawa1, Shintaro Takao2, Kazuhiko Yamagami3, Masaru Miyashita4, Masashi Baba5, Shigetoshi Ichii6, Muneharu
Konishi7, Yuichiro Kikawa8, Junya Minohata9, Toshitaka Okuno10, Keisuke Miyauchi11, Kazuyuki Wakita12, Hirofumi Suwa13,
Takashi Hashimoto14, Masayuki Nishino15, Takashi Matsumoto16, Toshiharu Hidaka17, Yutaka Konishi18, Yoko Sakoda19, Akihiro
Miya20, Masahiro Kishimoto21, Hidefumi Nishikawa22, Seishi Kono23, Ikuo Kokufu24, Isao Sakita25, Koushiro Kitatsuji26, Koushi Oh27
and Yasuo Miyoshi28. 1Kansai Rosai Hospital, Amagasaki, Hyogo, Japan; 2Hyogo Cancer Center, Akashi, Hyogo, Japan; 3Shinko
Hospital, Kobe, Hyogo, Japan; 4Kohnan Hospital, Kobe, Hyogo, Japan; 5Itami City Hospital, Itami, Hyogo, Japan; 6Rokko Island
Hospital, Kobe, Hyogo, Japan; 7Hyogo Prefectural Nishinomiya Hospital, Nishinomiya, Hyogo, Japan; 8Kobe City Hospital
Organization Kobe City Medical Center West Hospital, Kobe, Hyogo, Japan; 9Kobe Kyodo Hospital, Kobe, Hyogo, Japan;
10
Nishi-Kobe Medical Center, Kobe, Hyogo, Japan; 11Miyauchi Clinic, Amagasaki, Hyogo, Japan; 12Chayamachi Breast Clinic,
Osaka, Osaka, Japan; 13Hyogo Prefectural Tsukaguchi Hospital, Amagasaki, Hyogo, Japan; 14Hashimoto Clinic, Kobe, Hyogo,
Japan; 15Takarazuka Municipal Hospital, Takarazuka, Hyogo, Japan; 16Kinki Central Hospital, Itami, Hyogo, Japan; 17Kobe
Century Memorial Hospital, Kobe, Hyogo, Japan; 18Kobe Urban Breast Clinic, Kobe, Hyogo, Japan; 19Hyogo Prefectural
Kakogawa Medical Center, Kakogawa, Hyogo, Japan; 20Kuma Hospital, Kobe, Hyogo, Japan; 21Meiwa Hospital, Nishinomiya,
Hyogo, Japan; 22Nishikawa Clinic, Amagasaki, Hyogo, Japan; 23Kobe University School of Medicine, Kobe, Hyogo, Japan;
24
Kokufu Breast Clinic, Takarazuka, Hyogo, Japan; 25Sakita Clinic, Nishinomiya, Hyogo, Japan; 26Kitatsuji Clinic, Amagasaki,
Hyogo, Japan; 27Kobe Adventist Hospital, Kobe, Hyogo, Japan and 28Hyogo College of Medicine, Nishinomiya, Hyogo, Japan.
Body: Background: Adjuvant five-year treatment with aromatase inhibitors is standard for postmenopausal women with estrogen
receptor positive breast cancer. However, aromatase inhibitor-related adverse events including joint symptoms and vasomotor
symptoms have a strong impact on patients' quality of life and sometimes result in treatment discontinuation. The aim of this
study is to determine risk factors for discontinuation of endocrine therapy in Japanese postmenopausal breast cancer patients
treated with adjuvant anastrozole in a prospective cohort study based on patient-reported outcomes (PROs).
Patients and Methods: A total of 391 postmenopausal Japanese women with estrogen receptor-positive breast cancer and
treated with adjuvant anastrozole were enrolled from 28 centers in this prospective cohort study (SAVS-JP, UMIN000002455).
PROs assessment was obtained at baseline, 3, 6, 9 and 12 months which included joint and vasomotor symptoms. Long-term
adherence of anastrozole was obtained form 364 out of 391 patients (median follow-up: 44 months, range: 5-105months). We
analyzed the relationship of discontinuation of anastrozole with joint and vasomotor symptoms induced by treatment, and
patients characteristics.
Results: Among 364 patients, 64 (17.6%) discontinued, 297 (81.6%) are ongoing and 3 (0.8%) have completed five-year
anastrozole treatment. The reasons for discontinuation were recurrence: 20 (31.3%), secondary malignancies: 5 (7.8%), death
from non-breast cancer: 1 (1.6%) and adverse events: 38 (59.4%). These 38 patients who stopped treatment caused by adverse
events were compared with other 323 patients. Joint and vasomotor symptoms were categorized into grade 0 (no symptom or no
change from baseline), grade 1+2 (mild+moderate) and grade 3 (severe). Grades of joint symptoms were significantly associated
with discontinuation of anastrozole (Grade 0: 9.7%, grade 1+2: 7.8%, grade 3: 25.0%, p=0.02). Patients with longer time after
menopause (16 years or longer) were significantly higher frequency of discontinuation as compared with shorter time after
menopause (0-15years) (14.9% vs 8.0%, p=0.04). Univariate analysis revealed that grade 3 joint symptoms (odds ratio: 3.67,
95% confidence interval: 1.34-10.04, p=0.01) and longer time after menopause (OR: 2.01, 95%CI: 1.01-4.00, p=0.04) were
significant risk factors for discontinuation. By multivariate analysis, both grade 3 joint symptoms and long time after menopause
were independently associated with discontinuation.
Conclusion: In the present study, we have identified that grade 3 joint symptoms and longer time after menopause were risk
factors for discontinuation of adjuvant anastrozole. These data might give us useful information for counseling in patients with
adjuvant aromatase inhibitors for postmenopausal Japanese women.

Title: Aromatase inhibitor induced musculoskeletal syndrome (AIMSS) in Australian women with early breast cancer: An Australia
and New Zealand Breast Cancer Trials Group (ANZBCTG) survey of members of the Breast Cancer Network Australia (BCNA)
Janine M Lombard1,2,3, Nicholas Zdenkowski1,2,3, Kathy Wells4, Nicca Grant4, Linda Reaby3, John F Forbes1,2,3 and Jacquie
Chirgwin2,3,5. 1Calvary Mater Newcastle, Newcastle, NSW, Australia; 2University of Newcastle, Newcastle, NSW, Australia;
3
Australia and New Zealand Breast Cancer Trials Group, Newcastle, NSW, Australia; 4Breast Cancer Network Australia, VIC,
Australia and 5Monash University, VIC, Australia.
Body: Background: AIMSS is experienced by approximately half of women taking an aromatase inhibitor (AI), impairing quality of
life and in some leading to AI discontinuation. There is a lack of evidence for effective AIMSS treatments.
Aim: To investigate the importance of AIMSS in Australian women with early breast cancer.
Method: A survey invitation was distributed to 2390 members of the BCNA Review and Survey Group in April 2014. The online
questionnaire consisted of 45 questions covering demographics, AI use, clinical manifestations and risk factors for AIMSS,
reasons for AI discontinuation and efficacy of interventions used for AIMSS. AIMSS was defined as joint pain or stiffness that
developed or worsened after commencing an AI.
Results: Of 594 respondents, 370 (62%) were eligible. Reasons for exclusion were: preinvasive disease, locally
advanced/metastatic breast cancer, or other reason.
Eligible respondents had a median age range of 50-59 years. Duration of AI use varied (26%1year, 64% 1-5years, 10% 5years).
57% had received adjuvant chemotherapy. 43% of these commenced AI within 3 months of chemotherapy and 30% within 3-6
months of chemotherapy. A vitamin D test was performed in 64% of women and 68% were currently using vitamin D
supplements. Joint pain during menopause was reported by 22% of respondents.
AIMSS occurred in 302/370 women (81%). Of those who developed AIMSS, sites affected were feet (68%), hands or wrists
(65%), knees (62%), hips (56%), shoulders or elbows (49%), back (46%), or neck (3%). 34% of women had considered stopping
an AI because of AIMSS.
99 (27%) of respondents had discontinued AI for any reason and of these 68% discontinued because of AIMSS. Non-AIMSS
symptoms identified as reasons for discontinuation included fatigue, vaginal/urinary symptoms and hot flushes. In respondents
who discontinued AI, 20% ceased use in the first 3 months, 30% during months 3-12 and 38%12 months. 42% of respondents
who discontinued an AI restarted the same or a different AI after a treatment break.
To manage AIMSS 23% of respondents used doctor prescribed medications (eg anti-inflammatories, codeine, morphine,), 55%
over the counter (OTC) or complementary medicines (eg low dose anti-inflammatories, paracetamol, chondroitin, fish or krill oil,
glucosamine, and vitamin D) and 29% alternative therapies (eg acupuncture, massage, Tai Chi and yoga). Respondents
identified the following in each of the above categories as most successful in relieving AIMSS symptoms: doctor prescribed
anti-inflammatories, paracetamol and yoga. Doctor prescribed medications and OTC/complementary medicine either completely
or significantly relieved AIMSS in 12% and 25% of cases respectively. 27% of respondents found that one or more of the
interventions that they had used to manage AIMSS helped prevent AI discontinuation.
Conclusion: AIMSS is a significant issue for Australian women and is an important reason for AI discontinuation. Women use a
number of interventions to manage AIMSS, however their efficacy appears limited. Effective AIMSS interventions are needed, to
improve quality of life and reduce AI discontinuation.

Title: Co-SOFT: The cognitive function substudy of the suppression of ovarian function trial (SOFT)
Kelly-Anne Phillips1, Yang Feng2, Karin Ribi3, Jrg Bernhard3, Fabio Puglisi4, Meritxell Bellet5, Simon Spazzapan6, Per Karlsson7,
Daniel R Budman8, Khalil Zaman9, Ehtesham A Abdi10, Susan M Domchek11, Meredith M Regan2, Alan S Coates12, Richard D
Gelber2, Paul Maruff13, Frances Boyle14, John F Forbes15, Gini F Fleming16 and Prudence A Francis1. 1Peter MacCallum Cancer
Centre, ANZBCTG & IBCSG, Melbourne, Australia; 2IBCSG Statistical Center, Dana-Farber Cancer Institute, Boston, MA;
3
IBCSG Coordinating Center, Bern, Switzerland; 4University Hospital of Udine & IBCSG, Udine, Italy; 5Hospital Vall d'Hebrn &
SOLTI, Barcelona, Spain; 6Centro di Riferimento Oncologico & IBCSG, Aviano, Italy; 7Sahlgrenska University Hospital & IBCSG,
Gothenburg, Sweden; 8Hofstra North Shore-LIJ School of Medicine & NRG Oncology, Lake Success, NY; 9Centre
Pluridisciplinaire d'Oncologie CHUV & IBCSG, Lausanne, Switzerland; 10Tweed Heads Hospital, Tweed Heads, Griffith University,
ANZBCTG & IBCSG, Gold Coast, NSW, Australia; 11Abramson Cancer Center & ECOG-ACRIN, Philadelphia, PA; 12International
Breast Cancer Study Group & ANZBCTG, Sydney, Australia; 13Cogstate Ltd, Melbourne, Australia; 14Mater Hospital, ANZBCTG &
IBCSG, Sydney, Australia; 15Calgary Mater Newcastle, ANZBCTG & IBCSG, Newcastle, Australia and 16University of Chicago
Medical Center & Alliance for Clinical Trials in Oncology, Chicago, IL.
Body: Background: Cognitive impairment is a potential side-effect of breast cancer (BC) treatment. Estrogen is an important
neuromodulator that affects cognition. Estrogen depletion by oophorectomy or GnRH agonists may adversely affect cognition in
non-oncological settings, but there are few data regarding the cognitive effects of ovarian function suppression (OFS) in women
with breast cancer.
Patients and Methods: Between November 2003 and January 2011, 3066 premenopausal women with hormone receptor-positive
BC were randomised on the SOFT trial to 5 years of adjuvant endocrine therapy with tamoxifen alone, tamoxifen+OFS or
exemestane+OFS. OFS was achieved by the GnRH agonist triptorelin, oophorectomy or ovarian irradiation. Prior chemotherapy
was allowed, provided women had premenopausal estradiol levels at enrolment. Women eligible for Co-SOFT must not have
received any prior adjuvant endocrine therapy. At study entry (t1), and approximately 1 year after SOFT randomisation (t2),
objective cognitive function was assessed with a brief computerized test battery comprising 7 tasks (CogState Ltd: cogstate.com).
Subjective cognitive function, psychological distress, fatigue, insomnia and quality of life were also assessed. Co-SOFT recruited
86 of a planned 321 patients from 27 of 426 SOFT centres between November 2007 and January 2011, when Co-SOFT was
closed as the SOFT trial completed accrual. The protocol-specified primary comparison was the change in the composite score of
the CogState tasks over 1 year for women randomised to tamoxifen versus tamoxifen+OFS. However, due to low accrual this
was modified, prior to any analysis, to compare the tamoxifen versus the pooled tamoxifen+OFS and exemestane+OFS groups.
Cognitive test scores were standardized according to age-specific norms, averaged to compute the composite score and then
change between t1 and t2 calculated; a negative change in composite score indicates deterioration in cognitive function. Change
in composite score was compared using Wilcoxon rank sum test.
Results: Of 86 Co-SOFT enrolled patients, 74 underwent both t1 and t2 CogState testing and were included in the primary
analysis (7 withdrew consent/declined assessment, 5 missed testing due to scheduling). Of these 74 women, 20 were
randomised to tamoxifen and 54 to OFS+tamoxifen (28) or OFS+exemestane (26). Baseline characteristics were well balanced
between the 2 groups. During the first year 49 women utilised GnRH alone for OFS, 4 had GnRH followed by oophorectomy and
1 had oophorectomy alone. There was no significant difference in the changes in the CogState composite scores from t1 and t2
for patients randomised to tamoxifen alone compared with OFS+oral endocrine therapy (median, -0.057 versus -0.146
respectively, p=0.51). There were no significant between-group differences in the changes from t1 and t2 for any of the 7
individual cognitive tasks comprising the composite score.
Conclusions: The results of this 1-year longitudinal substudy suggest that the addition of OFS to oral endocrine therapy does not
significantly affect cognitive function in the setting of adjuvant BC treatment. Co-SOFT was limited by small sample size, so
further investigation of the impact of OFS on cognitive function in BC patients is warranted.

Title: Characteristics of recurrence after completing adjuvant tamoxifen therapy for 5 years
Eunshin Lee1, Han-Byoel Lee1, Young Joon Kang1, Yun-Gyoung Kim1, Tae-kyung Yoo1, Jongjin Kim1, Min Kyoon Kim1,
Hyeong-Gon Moon1, Dong-Young Noh1 and Wonshik Han1. 1Seoul National University College of Medicine, Seoul, Korea.
Body: Background Treatment with tamoxifen (TMX) reduces the recurrence rate and increase overall survival in patients with
hormone receptor positive breast cancer. Up to now, 5-year TMX therapy is generally accepted, but it is demonstrated that the
rate of late recurrence after 5 years is considerably higher in hormone receptor positive type than in other subtype. Several
clinical trials such as ATLAS and aTTom showed the benefit of continuing tamoxifen up to 10 years instead of stopping at 5 years
without increasing mortality due to the effect of extended tamoxifen medication.
Method We collected data of 1633 hormone receptor positive breast cancer patients who received surgery at Seoul National
University Hospital from 1997 to 2007, and had completed 5-year TMX therapy with no recurrence within 5 years after diagnosis.
Mean age of the patients was 43.3, and the patients have estrogen receptor or progesterone receptor. We included from the
stage I to stage IV patients underwent curative surgery and received adequate adjuvant therapy such as chemotherapy or
radiation therapy after surgery. We excluded the cases treated aromatase inhibitor (AI) or switched to AI.
Result Among these patients, recurrences after 5 years of TMX therapy were found in 93 patients (late recurrence group). Local
recurrences and distant metastases were found in 43 and 50 patients, respectively. Electronic medical records were
retrospectively reviewed for clinicopathological factors. When comparing between patients with no recurrence and patients with
late recurrence, p53 and HER-2 expression were significantly related to late recurrence (p=0.01, p<0.001 respectively). Also
when subgroup analysis was done for distant metastasis of late recurrence group, distant metastasis was significantly associated
with HER-2 expression and high nuclear grade (p=0.005, p=0.006 respectively). There are no relation between late recurrence
and age, stage and ki-67.
Conclusion our data shows that p53 and HER-2 expression is associated to late recurrence and especially HER-2 expression is
related to distant metastasis after completing TMX for 5 years. On the basis of the result of large clinical trials, extending TMX
therapy significantly reduces recurrence rate and increase survival. Our result support continuing TMX in patients with HER-2
expression and high nuclear grade is considerable after 5 years of TMX medication.

Title: Bortezomib enhances the efficacy of fulvestrant by promoting the aggregation of the ER in the cytoplasm
Doris Germain1, Kerin Adelson2, George Raptis3 and Samuel Waxman1. 1Icahn School of Medicine at Mount Sinai; 2Yale
University School of Medicine, New Haven, CT and 3North Shore-LIJ Cancer Institute.
Body: Background: Aromatase inhibitors (AIs) are standard treatment for Estrogen Receptor (ER)+ breast cancers in
post-menopausal women where the main source of estradiol comes from adipose tissue when the aromatase enzyme converts
androgens to estrogens. However, in premenopausal women, functioning ovaries flood the body with estrogens, and aromatase
inhibitors used alone offer no therapeutic benefit. In addition to tamoxifen and aromatase inhibitors, estrogen receptor
down-regulators, are a third type of anti-estrogen. The first of this class to be FDA approved is Fulvestrant, which acts by
promoting the proteosomal degradation of the ER. Like tamoxifen, fulvestrant binds directly to the ER but while tamoxifen has
both antagonist and agonist effects on the ER, fulvestrant is a pure antagonist. Other important advantages of fulvestrant over
tamoxifen are that 1) fulvestrant prevents the ER from binding DNA, 2) fulvestrant is not linked to increased risk of endometrial
cancer, 3) fulvestrant promotes permanent degradation of the ER.
The molecular machinery leading to the degradation of the ER in the nucleus following fulvestrant treatment is well described and
correlates with its ubiquitination in the nucleus. A less well-recognized mechanism, is fulvestrants ability to promote the
aggregation of the newly synthesized ER in the cytoplasm. Understanding that protein aggregates are toxic when not eliminated
by the proteasome, we took advantage of this effect of fulvestrant to ask whether combining fulvestrant with the proteasome
inhibitor Bortezomib could enhance the efficacy of Fulvestrant.
Results: We found that bortezomib enhances the aggregation of the ER in the cytoplasm following treatment with fulvestrant.
Further, these aggregates were found to be insoluble and to activate the unfolded protein response (UPR), a stress response that
leads to cell death. Further, bortezomib is able to prevent the activation of cytoprotective responses linked to the acquisition of
fulvestrant resistance. Furthermore, in a breast cancer mouse model of tamoxifen resistance, the combination induced tumor
regression. We currently are testing new generation proteasome inhibitors and the results will be presented at the meeting.
Conclusion: We conclude that adding bortezomib to fulvestrant enhances its efficacy by taking advantage of a previously poorly
recognized mechanismfulvestrants induction of ER aggregation in the cytoplasm. Further, our data suggest that this strategy will
block the ability of cells to acquired resistance to fulvestrant. Our group has developed a clinical trial that has tested this
combination and the results of this trial presented at the meeting.

Title: Possibility of GnRH agonist plus tamoxifen as an alternative treatment to AC (adriamycin plus cyclophosphamide)
chemotherapy plus tamoxifen treatment in premenopausal hormone responsive HER2 negative breast cancer patients
Heeseung Park1, Guiyun Sohn1, Byung Ho Son1, Jong Won Lee1 and SeiHyun Ahn1. 1College of Medicine, University of Ulsan,
Seoul, Korea.
Body: Background: The purpose of this study is to compare the treatment outcomes using GnRH agonist plus tamoxifen, and
adriamycin and cyclophopsphamide(AC) containing chemotherapy plus tamoxifen in hormone responsive premenopausal node
negative breast cancer patients.
Methods: Total 1027 premenopausal women with node negative, hormone receptor positive, Her-2 negative breast cancer were
included in this retrospective cohort study. 595 patients (57.9%) were treated with GnRH agonist together with tamoxifen, and 432
patients (42.1%) were treated with adriamycin and cyclophsophamide containing chemotherapy with tamoxifen.
Results: A median follow up period was81 months. In pT1 disease, 7 year DFS (disease free survival)was 98.9% for GnRH
agonist plus tamoxifen group and 97.4% for chemotherapy plus tamoxifen group, and in pT2 disease, 7 year DSS was 99.1% for
GnRH agonist plus tamoxifen group, and 96.9% for chemotherapy plus tamoxifen groupwith nostatisticalsignificance (p=0.252 &
0.230).In pT1 disease, 7 year RFS was 95.4% for GnRH agonist plus tamoxifen group and 92.9% for chemotherapy plus
tamoxifen group, and in pT2 disease, 7 year RFS was 91.6% for GnRH agonist plus tamoxifen group, and 91.7% for
chemotherapy plus tamoxifen group with no statistical difference between two treatment groups (p=0.353 & 0.836). In subgroup
analysis, ER+/PR+ patients and patientsyounger than 40 years old showed slightly better survival inGnRHagnonist plus tamoxifen
group when compared to chemotherapy plus tamoxifen group with p-value of 0.046 & 0.018.
Conclusion: Adding GnRH agonist to tamoxifen is reasonable alternative to adding AC chemotherapy to tamoxifen in
premenopausal luminal type node negative breast cancer patients.

Title: Tamoxifen treated patients have a better survival than patients treated with aromatase inhibitors - A population based
registry study in Sweden
Hkan Olsson1,2, Rickard Einefors1 and Per Broberg1. 1Clinical Sciences, Lund University, Lund, Sweden and 2Lund University,
Lund, Sweden.
Body: Background. Randomised trials suggest that therapy with aromatase inhibitors improves survival in breast cancer
compared with tamoxifen therapy in postmenopausal cases with hormone receptor positive breast cancer. Whether these results
from randomized studies transform into the general population is unknown. We have therefore compared survival for all breast
cancer cases in Sweden diagnosed 2000-2008 (n=54406) who received adjuvant antihormonal therapy.
Material and methods. The study includes all women with BC diagnosed in Sweden between 2000 through 2008 (n=54406). The
women had no previous cancer diagnosis during the period of 1958-1999. Dates of birth, BC diagnosis and TNM-stage where
directly extracted from the cancer registry. The womens antihormonal therapy was gathered from the Swedish Prescription
Registry (22213 women were on antihormonal therapy). Information regarding the cause of death and date of death was obtained
from the Cause of Death Registry and tbe Swedish Population Register up until the 31st of December 2012 and 31st of
December 2013 respectively. The breast cancer death and overall death have been calculated and the survival was compared
between tamoxifen and aromatase inhibitor treated breast cancer patients. Analyses were adjusted for TNM-stage and age at
diagnosis and restricted to women aged 50 and above.
Results. Patients being treated with tamoxifen had a better breast cancer prognosis compared with aromatase inhibitor treated
patients (HR 0.54, 95%CI 0.48-0.61). Restricting the analysis to stage 1 disease confirmed a better prognosis for tamoxifen
treated women (HR 0.48, 95%CI 0.34-0.66). A better prognosis could be seen in all age strata studies, 50-60.61-70.71-90. The
findings for overall survival gave similar results.
Conclusion .This population based observational study show that women treated with aromatase inhibitors have a worse overall
and breast cancer specific survival compared with tamoxifen treated women regardless of age and tumor stage.

Title: Prospective study of aromatase inhibitor-induced bone loss and lipid levels in early hormone sensitive breast cancer treated
with AI during 8 years
Christel Fontaine1, Lore Decoster1, Denis Schallier1, Leen Vanacker1, Jacques De Grve1, Marian Vanhoeij2, Guy Verfaillie2 and
Jan Lamote2. 1Oncologisch Centrum, UZ Brussel, Jette, Brussels, Belgium and 2Breast Surgery, Oncologisch Centrum, UZ
Brussel, Jette, Brussels, Belgium.
Body: Introduction: AIs are the preferred adjuvant hormonal treatment for postmenopausal(PM) estrogen receptor positive
(ER+) disease because of the improved efficacy over tamoxifen in terms of disease free survival and reduction in distant
recurrence. AIs have been associated with increased bone loss by blocking peripheral tissue estrogen synthesis, arthralgia and a
negative effect on lipid metabolism. In this prospective study we wanted to assess the amount of bone loss and the changes in
lipid levels in early breast cancer(EBC) patients (pts) treated with 3yrs of extended letrozole as part of the SOLE study after 5 yrs
of adjuvant AI. Objectives: the primary objective of the study is to calculate the mean percentage change in bone mineral
density(BMD)(g/cm) after 8 yrs of letrozole and to compare between the continuous(C) and intermittent(I) intake of letrozole. The
secondary objectives are to correlate the mean change in BMD with the initial value, BMI and to describe the evolution of the lipid
levels in both groups. Patients and methods: BMD was measured at the lumbar spine (L2-4) and hip by dual energy X-ray
absorptiometry (DEXA) in PM pts. with ER+ EBC. We also measured the fasting lipid levels baseline and after 8 yrs of an AI.
Differences between the two groups were assessed by the independent samples T-test and the correlation by the linear
regression analysis. Results: Fifty four pts were included in the study with a mean age of 62.5 yrs(+/-8.6 yrs). Seventeen pts are
too early to be evaluable, 8 pts stopped letrozole due to adverse events and 2 pts had no baseline DEXA values. Two pts
received tamoxifen 2 to 5 yrs before inclusion in the SOLE study. The remaining 25 pts showed a mean change in BMD for the
lumbar spine of -1.1 (10.6SD) and for the hip -4.4 (7.04SD). Currently 13 pts in the C arm demonstrated a mean decrease in BMD
for the (LS) of -2.7(10.2SD), and for the hip of -2.7(5.5SD). Twelve pts included in the I arm experienced a mean increase in BMD
for the LS of 0.65(SD11.2) and a mean decrease in BMD for the hip of -4.9(9.1SD). The difference between the two treatment
groups was not significant for the LS measurements after 8yrs (p=0.4) nor for the hip (p=0.23) neither was it at baseline (p=0.9;
0.1). Only three pts had fractures due to trauma. The overall fracture rate was only 0.05%.There was no correlation with the BMI,
but there was a significant correlation with baseline BMD.(p=0.002)The mean fasting cholesterol levels at 8yrs in the C arm was
214mg/dl(41.3SD) and in the I arm was 218mg/dl(35.6SD)and was not significantly different(p=0.8), nor was it at baseline(p=0.5).
Conclusion: This first prospective long term BMD and lipid follow-up study in PM EBC pts taking adjuvant letrozole beyond 5 yrs
shows a decline in BMD. Until now no significant differences were observed between continuous and intermittent letrozole. An
updated and detailed follow-up on more patients will be presented.

Title: The insulin like growth factor axis and development of tamoxifen resistance in breast cancer
Yousef M Hawsawi1, James Beattie2, Reem El-Gendy3, Valerie Speirs4 and Christopher Twelves4. 1University of Leeds School of
Dentistry, Leeds, West Yorkshire, United Kingdom; 2Leeds Institute of Cancer and Pathology, Leeds, West Yorkshire, United
Kingdom; 3BioMedical Health Research Centre, Leeds, West Yorkshire, United Kingdom and 4King Faisal Specialist Hospital and
Research Centre, Jeddah, Western, Saudi Arabia.
Body: Background
Insulin-like growth factors (IGF). IGFs are potent mitogens for breast epithelial cells that modulate the action of AKT, which has
been reported to be associated with tamoxifen resistance. To date, however, anti-IGF strategies have proved disappointing in
clinical trials We have investigated whether the IGFBP family of proteins, which modulate the activity of IGF, may play a role in
tamoxifen resistance, opening up the route for alternative anti-IGF based therapies.
Methodology
We investigated the expression of IGF axis genes in parental and tamoxifen-resistant (TamR) MCF-7 cells using qRT-PCR. Gene
and protein expression was confirmed using ELISA, Western, and ligand blotting. shRNA transfection was used to silence
candidate IGF axis genes in both cell lines. Cell sensitivity 4-hydroxytamoxifen (4-HT) was investigated by WST-1 cell
proliferation assay.
Results
IGF-IR, IGF-2R, IGFBP-2, -4 and -5 genes had the highest expression levels. IGFBP-5 was down-regulated 7-fold while IGFBP-2
was up-regulated by 2-fold in TamR v wt cells. Changes in IGFBP-5 and IGFBP-2 gene expression were mirrored in protein levels
measured in conditioned media by ELISA, Western and Ligand blot. Importantly knock down of IGFBP-2 in TamR cells restored
sensitivity to 4-HT suggesting a causal role for IGFBP-2 in the acquisition of tamoxifen resistance.
Conclusion
IGFBP-5 and IGFPB-2 are reciprocally regulated on the acquisition of tamoxifen resistance by MCF-7 cells. Preliminary studies
suggest that IGFBP-2 may play a role in the development of tamoxifen resistance in vitro.

Title: Factors associated with adherence to adjuvant endocrine therapy in patients with hormone receptor positive breast cancer
Jongjin Kim1, Wonshik Han1, Hyeong-Gon Moon1, Min Kyoon Kim1, Eunshin Lee1, Tae-Kyung Yoo1, Han-Byoel Lee1, Young Joon
Kang1, Yun-Gyoung Kim1, Tae Ryung Kim2 and Dong Young Noh1. 1Seoul National University College of Medicine, Seoul, Korea
and 2Gachon University Gil Hospital, Incheon, Korea.
Body: Background/Purpose
Adjuvant endocrine therapy in patients with hormone receptor positive breast cancer reduces recurrence and mortality, but many
patients are non-adherent to anti-hormonal medication. In order to increase the adherence, it is important to know about factors
associated with adherence. So we investigated factors associated with adherence to anti-hormonal medication using variable
questionnaires.
Methods
We carried out a cross-sectional survey of a sample of women who underwent surgery due to breast cancer in the Seoul National
University Hospital Breast Care Center from 2007 to 2011 and treated with anti-hormonal medication. Questionnaires were sent
to 1,000 patients. The questionnaire booklet included the Medication Adherence Report Scale-5(MARS-5), Womens Health
Questionnaire(WHQ), Beliefs about Medicine Questionnaire(BMQ), Satisfaction with Information about Medicines Scale(SIMS).
And to identify patients clinical characteristics, we reviewed electronic medical records, retrospectively.
Result
The response rate of questionnaire was 40.8%(408/1000). Of the answered patients, 263 patients were treated with tamoxifen
and 145 patients were treated with aromatase inhibitors(AIs). 197 of 408 answered patients(48.3%) were classified as
non-adherence. The rate of non-adherence was 132/263(50.1%) and 65/145(44.8%) in patients treated with tamoxifen and AIs.
Of the all answered patients, non-adherent patients had more depressed mood (p<0.001). Non-adherent patients scored lower on
positive beliefs as measured on BMQ-necessity (OR = 0.65, 95% CI 0.51 to 0.82) and higher on negative beliefs as measured on
BMQ-overuse (OR=1.81, 95% CI 1.29 to 2.54). Non-adherent patients also scored lower on satisfaction with information about
action and usage of anti-hormonal treatment as measured on SIMS-action and usage (OR = 0.47, 95% CI 0.38 to 0.65). Of the
patients treated with tamoxifen, non-adherent patients had more depressed mood (p=0.003), scored higher on BMQ-overuse
(OR=1.97, 95% CI 1.22 to 3.20) and scored lower on SIMS-action and usage (OR = 0.33, 95% CI 0.22 to 0.50). Of the patients
treated with AIs, non-adherent patients had more depressed mood (p=0.014), scored lower on BMQ-necessity (OR=0.52, 95% CI
0.36 to 0.75).
Conclusion
This study showed associations between depressive mood of breast cancer patients treated with anti-hormonal therapy and
adherence. And beliefs and satisfaction with information about medication also associated with adherence. To improve
adherence, we should evaluate and correct patients mood. And we should provide proper information about medications.

Title: Retinoic acid sensitizes triple-negative breast cancer cells to tamoxifen treatment
Krysta M Coyle1, Cheryl A Dean1, Diana B Jo1, Margaret Thomas1, Mohammad Sultan1 and Paola Marcato1. 1Dalhousie
University, Halifax, NS, Canada.
Body: Tamoxifen, an estrogen receptor (ER) antagonist, is often used as an adjuvant endocrine therapy in the successful
treatment of ER+ breast tumors. Tumors that lack ER, progesterone receptor (PR) and HER2 expression (i.e. triple-negative
breast cancers) cannot be treated with adjuvant endocrine therapies, like tamoxifen, and are often more aggressive. Inducing ER
expression is a potential strategy for sensitization of triple-negative breast cancers to adjuvant endocrine therapies. Given recent
evidence suggesting cross-talk between the retinoic acid (RA) and estrogen signaling pathways, we investigated if RA induces
expression of ER in triple-negative breast cancer cells. We hypothesize that this would lead to sensitization of the cells to
tamoxifen treatment. Quantitative PCR of mRNA isolated from triple-negative MDA-MB-231 cells treated with RA and estradiol
had increased ER transcript levels. Furthermore, treatment with estradiol and RA synergistically induced increased expression of
RA-inducible genes. In cell proliferation studies, neither RA nor estradiol treatment alone significantly altered the growth of
MDA-MB-231 cells; however, when treated with both estradiol and RA together, the growth of the cells increased significantly.
This suggests that the RA-mediated increase in ER expression sensitizes MDA-MB-231 cells to estradiol-induced cell growth.
Next, we investigated whether the increased ER expression sensitized MDA-MB-231 cells to tamoxifen treatment. Tamoxifen did
not decrease the growth of MDA-MB-231 cells; however, when applied in combination with both estradiol and RA, tamoxifen
significantly reduced MDA-MB-231 proliferation. Furthermore, tamoxifen treatment reduced the synergistic effects of estradiol/RA
on RA-inducible gene expression. Together, these results suggest that the use of RA in combination with tamoxifen warrants
further investigation as a potential treatment for triple-negative breast cancers. The success of the combination treatment of
tamoxifen and RA in the reduction of triple-negative breast cancer cell tumor xenografts will provide further justification for this
strategy in the treatment of triple-negative breast cancers.

Title: Intermittent dosing of aromatase inhibitors (AI) to improve tolerance in post-menopausal women
Vincent Launay-Vacher1, Rmy Salmon2 and Jean-Baptiste Rey3. 1Service ICAR Piti-Salptrire University Hospital, Paris,
France; 2Hpital des Peupliers, Paris, France and 3Clinical Pharmacy Institut Jean Godinot, Reims, France.
Body: Clinical rationale: A significant proportion of post-menopausal, patients treated with AI reports side-effects, especially
bone pain. In such patients, the difficulties to treat pain and to clearly identify its causes may lead to treatment discontinuation.
Individual cases reported an improvement in AI tolerance when dosage is reduced. The aim of this work was thus to analyse
pharmacological data in order to validate this concept ie. ensuring that intermittent dosing would not result in a possibly
deleterious under-dosing of AI.
Pharmacological rationale: Ageing is associated with physiological modifications that may impair drug pharmacokinetics (PKs).
The elimination can be altered, with decreased drug clearance (CL), resulting in an increased exposure to the drug, reflected by
increased AUCs. Drugs benefit-risk balance can thus be modified. The major sources of PKs alterations in the elderly are hepatic
and/or renal impairments (HI/RI).
It has been shown that AI PKs are altered in the elderly and/or in case of HI or RI. Exemestane AUC is increased 3-fold in the
elderly and in case of RI [1,2]. Letrozole AUC can double in case of HI [3] and a 42%-increase has been reported in the elderly
[4]. The renal CL of anastrozole is reduced by 50% in RI, resulting in a 10% decrease in total body CL. In HI, total body CL is 30%
lower as compared to normal [5].
These data demonstrate that elderly patients may be overexposed to AI when treated with the usual dosage, resulting in safety
issues occurring in some patients. As a result, a dosage adjustment approach could help prevent over-exposure and reduce
side-effects incidence/severity.
Proof-of-Concept studies: The reported increases in AI exposure being around 50%, an intermittent dosing schedule of 1
administration every other day could result in a similar drug exposure as compared to the usual daily schedule. AI elimination
half-lives in the absence of any alteration also are consistent with this dosing schedule (24, 48, and 50 hours for exemestane,
letrozole, and anastrozole, respectively). In order not to impair plasma concentrations, such a schedule is preferably suggested
as compared to a half-dose daily schedule. Prospective studies are needed, in which the PKs, efficacy, and safety of this
intermittent dosing schedule should be conducted.
References:
[1] Aromasin FDA Labeling Information; 2013 labeling revision;
http://www.accessdata.fda.gov/drugsatfda_docs/label/2013/020753s014lbl.pdf; accessed 04/29/2014
[2] Jannuzzo MG, et al. Cancer Chemother Pharmacol. 2004; 53(6): 475-81.
[3] Femara FDA Labeling Information; 2014 labeling revision;
http://www.accessdata.fda.gov/drugsatfda_docs/label/2014/020726s027lbl.pdf; accessed 04/29/2014
[4] Pfister CU, et al. Effect of age and single versus multiple dose pharmacokinetics of
letrozole (Femara) in breast cancer patients. Biopharm Drug Dispos. 2001; 22(5): 191-7.
[5] Arimidex FDA Labeling Information; 2013 labeling revision;
http://www.accessdata.fda.gov/drugsatfda_docs/label/2013/020541s027lbl.pdf; accessed 04/29/2014.

Title: Efficacy of adjuvant tamoxifen in hormone receptor-positive premenopausal breast cancer patients according to the body
mass index
Kadri Altundag1, Mehmet AN Sendur2, Sercan Aksoy1, Taner Babacan1 and Yavuz Ozisik2. 1Hacettepe University Cancer Institute,
Ankara, Turkey and 2Yidirim University, Ankara, Turkey.
Body: Introduction: Obesity is an independent risk factor for the development of breast cancer and has been associated with
poor breast cancer outcomes. But, this association usually depend on hormone-receptor positivity and ovarian activity. Obesity
was confirmed as an adverse prognostic factor in patients treated with aromatase inhibitors, but the adverse effects in patients
treated with tamoxifen was not known exactly. Thus, we aimed to examine the efficacy of adjuvant tamoxifen in hormone
receptor-positive premenopausal breast cancer patients according to the body mass index (BMI).
Material-Methods: Newly diagnosed hormone receptor-positive breast cancer patients who were premenopausal and
non-metastatic were enrolled to the study. Patients with BMI ranging between 18.5 and 24.9 kg/m2 as normal weight patients
(Arm A, n = 408), and patients with a BMI ranging 25 kg/m2 were overweight and obese patients (Arm B, n = 418).
Results: The median follow-up time for this analysis was 36 (6-327) months. The median age was 39.5 (22-57) and 43 (20-56) in
Arm A and Arm B, respectively (P<0.0001). The mean BMI was 22.1 1.8 kg/m2 and 29.2 3.3 kg/m2 of Arm A and Arm B,
respectively ( P = <0.001). In both normal weight and overweight patients, the other baseline clinico-pathologic properties and the
treatment history with radiotherapy and chemotherapy were similar and not statistically significant. In overweight and obese
patients the history of diabetes mellitus and hypertension was significantly higher compared to normal weight patients. In patients
with normal weight patients DFS rate was 88.5% and 78.2% whereas in overweight and obese patients DFS rate was 87.2% and
70.9% in the third and fifth years respectively (Figure 1) ( P = 0.43). In patients with normal weight patients OS rate was 98.5%
and 93.2% whereas in overweight and obese patients OS rate was 94.6% and 87.4% in the third and fifth years respectively
(Figure 2) ( P = 0.02).
Conclusion: Our study showed that BMI have no worse effect on recurrence risk in patients treated with tamoxifen in
hormone-receptor positive premenopausal breast cancer. Poor survival outcome was observed in overweight and obese patients
can be due to dose limitations of chemotherapeutic agents and higher rate of comorbid diseases.

Title: Mature data (> 25 years) on 2 years adjuvant tamoxifen treatment in premenopausal women with breast cancer: Time to
re-emphasize the progesterone receptor as predictive factor?
Maria Ekholm1,2, Pr-Ola Bendahl1, Mrten Fern1, Bo Nordenskjld3, Olle Stl3 and Lisa Rydn4. 1Division of Oncology and
Pathology, Lund University, Lund, Sweden; 2Ryhov County Hospital, Jnkping, Sweden; 3Divison of Oncology, Linkping
University, Linkping, Sweden and 4Division of Surgery, Lund University, Lund, Sweden.
Body: Background
In 2011 The Early Breast Cancer Trialists' Collaborative Group (EBCTCG) reported that five years of adjuvant tamoxifen (TAM)
significantly reduces the 15-year risks of breast cancer recurrence and death. Moreover, the estrogen receptor (ER) was
considered the clinically most relevant predictive factor for efficacy of TAM treatment. In premenopausal patients participating in a
randomized trial of 2 years of adjuvant TAM, we have previously reported on increased recurrence free survival in patients with
ER- and PR-positive disease and a trend for increased overall survival in patients with PR-positive tumors. The median follow-up
time was 13.9 years for patients without events. Since the effect on mortality may increase beyond year 10, we investigated more
mature of data.
Aims
To investigate the long-term effect of 2 years adjuvant TAM in premenopausal patients choosing cumulative mortality (CM) and
cumulative breast cancer mortality (CBCM) as end-points.
Methods
Premenopausal patients (n=564) with stage II breast cancer were included in a Swedish randomized multicenter trial between
1986-1991 and allocated to 2 years of TAM (n=276) or no treatment (n=288). The hormone receptor status of the tumor was
known for 541 (96%) of the patients included, determined by immunohistochemistry or cytosol-based methods. Less than 2%
were treated with adjuvant chemotherapy and they were equally distributed between the two groups. Median follow-up time was
26.3 years (22.7-29.7). Mortality data was obtained from the Swedish Cause of Death Register.
Results
Death from any cause was recorded among 312 (55%) of the patients, and 265 (47%) of the deaths were due to breast cancer.
Two years of TAM decreased CM and CBCM in all patients irrespective of hormonal receptor status (n=564) (CM: HR 0.81; 95%
CI: 0.65-1.02, p=0.070; CBCM: HR 0.79; 95% CI: 0.61-1.01, p=0.058), as well as in patients with ER-positive disease (n=346)
(CM: HR 0.77; 95% CI: 0.57-1.03, p=0.077; CBCM: HR 0.72; 95% CI: 0.52-1.01, p=0.051), however not strictly significant.
Importantly, in patients with PR-positive tumors, TAM significantly reduced CM and CBCM (CM: HR 0.73; 95% CI: 0.55-0.98,
p=0.037; CBCM: HR 0.68; 95% CI: 0.49-0.94, p=0.020). Moreover, in patients with ER- and PR positive tumors TAM decreased
CM (HR 0.74; 95% CI: 0.55-1.01, p=0.057) and CBCM (HR 0.69; 95% CI: 0-49-0.97, p=0.033), whereas there was no effect of
TAM in patients with ER-positive and PR-negative tumors (p=0.97 and p=0.99, respectively).
Conclusions
The present study demonstrates that 2 years of adjuvant TAM as monotherapy significantly reduced CM and CBCM in
premenopausal women with PR-positive tumors as well as CBCM in ER- and PR-positive tumors, after > 25 years of follow-up.
Five years of adjuvant TAM have been proven to reduce mortality rates at 15 years, but we herein show that also 2 years of TAM
yields a survival benefit at 25 years.
In the latest meta-analysis from EBCTCG, PR did not add predictive information in patients with ER-positive tumors, but
according to our results the significance of PR, as a predictive factor for TAM efficacy, should be re-emphasized in
premenopausal women.

Title: Phase I study of ARN-810, a novel and potent oral selective estrogen receptor degrader, in postmenopausal women with
metastatic estrogen receptor positive (ER+), HER2- breast cancer
Aditya Bardia1, Maura N Dickler2, Ingrid A Mayer3, Eric P Winer4, Umar Mahmood1, Gary Ulaner2, H Charles Manning3, Peter Rix5,
Jeffrey H Hager5, Debasish Roychowdhury5, Edna Chow Maneval5, Carlos L Arteaga3 and Jose Baselga2. 1Massachusetts
General Hospital Cancer Center, Boston, MA; 2Memorial Sloan Kettering Cancer Center, New York, NY; 3Vanderbilt-Ingram
Cancer Center, Nashville, TN; 4Dana-Farber Cancer Institute, Boston, MA and 5Seragon Pharmaceuticals, San Diego, CA.
Body: Background: Evidence that ER can signal in both ligand-dependent and independent manner in endocrine resistant
breast cancer (BC) provides rationale for therapies that are not only functional antagonists of ER but also reduce ER levels, thus
targeting both modes of signaling. Furthermore, mutations in ESR1 affecting the ligand-binding domain (LBD) that drive
ER-dependent transcription and proliferation in the absence of estrogen suggest that LBD-mutant forms are involved in mediating
clinical resistance and next generation ER modulators with robust activity in both wild type and mutant ER tumors are needed.
ARN-810 is a novel, orally bioavailable, ER antagonist that induces proteasomal ER degradation in BC cell lines at picomolar
concentrations and tumor regression in tamoxifen-sensitive and resistant BC xenograft models.
Methods: ARN-810 was tested using standard 3+3 dose escalation to assess safety, PK, and Recommended Phase 2 Dose
(RP2D). Key eligibility criteria included ER+ (HER2-) metastatic BC progressing 6 months (m) on endocrine therapy and 2
prior chemotherapies. Pre- and on-study tumor biopsies were obtained when feasible. Pharmacodynamics was assessed by
functional imaging with [18F]-fluoroestradiol (FES)-PET, tumor-based ER/PR/Ki67 IHC, and ER target gene expression. Plasma
PK was assessed following a single dose and at steady-state. Anti-tumor activity was assessed by clinical benefit rate (CBR)
[complete response, partial response, or stable disease 6m] and progression-free survival (PFS).
Results: From April 2013 to June 2014, 32 patients (pts) (median age 61 (range 43 75); median number of prior therapies = 3
(range 1 7); visceral metastases 54%) were enrolled at 5 doses (100, 200, 400, 600, 800 mg) and 2 different regimens (once
[QD] and twice daily) given orally with and without fasting. Increases in ARN-810 exposure were dose-dependent with no
apparent food effect. At 4 weeks of treatment, complete reduction in FES uptake consistent with full receptor saturation and/or
degradation was seen in 95% pts (21/22 scanned to date), including 2 pts with ESR1 mutations, suggesting ARN-810 exhibits
greater ER occupancy than that recently reported for fulvestrant 500 mg (van Krutchen et al, ASCO 2014). Evidence of reduced
ER levels and Ki67 staining was observed on treatment. To date, 19 pts (59.4%) remain on study with a preliminary CBR of 41%.
RP2D, PFS and gene expression results will be provided at time of meeting. The most common adverse events were grades 1/2
nausea, diarrhea, fatigue, and abdominal pain. There was 1 dose limiting toxicity (grade 3 diarrhea) at 800 mg QD which led to
expansion of that cohort, while in parallel, evaluation of the other dose regimens continues. No patients have discontinued the
study due to toxicity.
Conclusions: ARN-810 appears to be safe and tolerable, with predictable PK, promising anti-tumor activity, and
pharmacodynamic evidence of target engagement, ER degradation and reduced tumor proliferation in heavily pre-treated
metastatic ER+ BC. In Phase II, ARN-810 will be studied in patients previously treated with aromatase inhibitors and fulvestrant,
including those with ESR1 mutations.

Title: Visceral metastases from hormone receptor positive breast cancer are as sensitive to endocrine therapy as non-visceral
metastases
John FR Robertson1, Robert Paridaens2, Jan Bogaerts3, Yuri Rukazenkov4, Christine Campbell5 and Ian Bradbury6. 1University of
Nottingham, Derby, Derbyshire, United Kingdom; 2Universitair Ziekenhuis Gasthuisberg, Leuven, Belgium; 3European
Organisation for Research and Treatment of Cancer, Brussels, Belgium; 4AstraZeneca Pharmaceuticals, Macclesfield, United
Kingdom; 5Frontier-Science, Grampian View, Kincraig, United Kingdom and 6Frontier-Science, Grampian View, Kincraig, United
Kingdom.
Body: Background:
There remains a perception among many clinicians that visceral metastases (VMs) from hormone receptor positive (HR) breast
cancer (BC) respond less well to endocrine therapy (ET) than non-VMs and so should receive chemotherapy as first line
treatment.
Patients & Methods:
Four phase 3 randomised controlled trials (RCTs) of first line ET, all with tamoxifen (TAM) as their control arm) have been
reviewed Exemestane (Exe), fulvestrant 250mg (F250) and two with anastrozole (Ana); the latter, were study 27 in the Rest of
the World (ROW) & study 30 in North America (NA). All reported objective response (OR), clinical benefit (CB), time to
progression (TTP) / progression free survival (PFS): all have been published (1-5). Only HR positive tumors were included in this
review. Data have been analysed both for Tam control arms alone and also for the four different endocrine agents combined.
Results:
CB & OR for TAM alone in each study individually and then combined are shown in the Table. The OR and CB rates were similar
for non-VMs versus VMs in all studies except study 30 (NA) where CB rates were 59% for non-VM and 33% for VM (Test for
heterogeneity of CB rates was p=0.047). For the four studies combined, the CB rates for non-VM versus VMs were 64% and 57%
respectively (p=0.06) while the OR rates were 34% versus 30% respectively (p= 0.28).
When all endocrine agents were combined the OR rate between non-VMs and VMs was significantly different (p = 0.038) as was
the CB rate (p = 0.0015). Rates of CB and OR in study 30 (NA) again appear different between non-VMs and VMs (data not
shown).
The Median Duration of CB on Tam appear similar between non-VMs versus VMs, both for each study individually and when
combined (see Table); when combined the Medians were 420 versus 418 days respectively with a Hazard Ratio (HR)=0.952
(0.748-1.211) (p=0.69). When all endocrine therapies for the combined four studies were assessed the HR for DoCB between
non-VMs and VMs was 0.922 (0.779-1.093) (p=0.35).
For the TTP the HR of non-VMs versus VMs on Tam alone was 0.851 (0.715-1.011) (p=0.07) and for all endocrine therapies the
HR was 0.821 (0.727-0.926) (p=0.001).
Conclusions:
HR+VMs which achieve clinical benefit on ET remain controlled for as long as non-VMs as shown by the DoCB results for both
Tam and all endocrine therapies combined.
There was no significant difference in OR rates between VMs and non-VMs with Tamoxifen. There was when all endocrine
agents were combined and the difference appears to be primarily due to one study (30 NA). There is no confirmed
explanation for these differences.
TTP differences appear to be due primarily to the difference in initial CB rates in study 30 (see Table).
HR+ VMs have hormone sensitivity similar to non-visceral mets they respond as well and for as long as non-VMs. In the
absence of visceral crisis (ie immediately life-threatening disease) ET would appear to be the treatment of choice for VMs in the
same way as it is for non-VMs.
Clinical Outcome by VM and Non VM (Tamoxifen only patients) N (%)
Studies
Exe
Ana (EUR)
Ana (NA)
F250
Combined
N = 168
N = 144
N = 162
N = 209
N=683
N = 88
N = 71
N = 82
N = 135
N=376
CB (%)
66 (75)
42 (59)
47 (59)
87 (64)
242 (64)
OR (%)
41 (47)
26 (37)
15 (19)
45 (33)
127 (34)
TTP (Median in days)
287
254
305
234
277
DoCB (Median in days)
343
360
445
451
420
N = 80
N = 73
N = 80
N = 74
N=307
CB (%)
64 (80)
38 (52)
27 (33)
46 (62)
175 (57)
OR (%)
36 (45)
21 (29)
13 (16)
21 (28)
91 (30)
TTP (Median in days)
340
230
152
281
238
DoCB (Median in days)
376
505
411
339
418
Without VM (55%)
With VM (45%)

Title: Association of PIK3CA mutation with clinical response to specific endocrine therapies in metastatic hormone receptor
positive (HR+) breast cancer
Douglas S Micalizzi1, Dejan Juric1, Andrzej Niemierko1, Kerry L Reynolds1, Darrell Borger1, Sadhna R Vora1, Steven J Isakoff1,
Beverly Moy1, Leif W Ellisen1 and Aditya Bardia1. 1Massachusetts General Hospital, Boston, MA.
Body: Background: Phosphatidylinositol 3-kinase (PI3K) gene (PIK3CA) mutations are the most common somatic mutations in
breast cancer. Preclinical models suggest that activating mutations in PIK3CA may mediate resistance to endocrine therapy in
breast cancer, and multiple clinical trials testing combinations of endocrine therapy with PI3K inhibitors are ongoing. However, the
role of PIK3CA in modulating the clinical response to endocrine therapies is less clear, with some studies suggesting that PIK3CA
mutations are associated with improved prognosis in HR+ breast cancer. The primary objective of this study was to evaluate the
association of PIK3CA mutation with clinical response to endocrine therapies in metastatic HR+ breast cancer.
Methods: We identified patients with metastatic HR+ /HER2 negative breast cancer, including ER+/PR+ (estrogen
receptor/progesterone receptor) and ER+/PR-, to determine the time to progression (TTP) on first-line endocrine therapy for
metastatic disease. PIK3CA mutations, including 8 common hotspot mutations, were assessed by a robust, high-throughput
tumor genotyping assay (Snapshot), developed at our institution, using DNA derived from formalin-fixed, paraffin-embedded
(FFPE) tissue. Actuarial analysis of TTP was performed using Cox proportional hazard method to compute Hazard Ratio (HR)
and 95% Confidence Intervals (CI).
Results: Between 2009 and 2012, we identified 188 patients with HR+ metastatic breast cancer who had tumor genotyping
performed. PIK3CA mutations were identified in 32.2% of tumors, including mutations in both helical (exon 9) and kinase (exon
20) domains (60% and 40%, respectively). The PIK3CA mutant and wild type patients had a similar median age at diagnosis of
metastatic disease (55.8 and 56.7 years; p=0.2), ER+/PR+ tumors (75.0% vs 76.4%; p=0.8), and median TTP (8.2 versus 11.4
months; p=0.6). After adjusting for age at diagnosis and ER+/PR+ versus ER+/PR-, the TTP on first-line endocrine therapy did
not vary among patients with PIK3CA mutations versus wild type (HR: 0.94; 95% CI:0.62-1.43; p=0.8), but did vary by type of
endocrine therapy. Patients with PIK3CA mutations, as compared to wild type patients, had shorter TTP with fulvestrant (HR: 3.6;
95% CI:1.2-11.0; p=0.03), but not with aromatase inhibitors (AIs) (HR: 0.70; 95% CI:0.44-1.1; p=0.1), suggesting that mutant
PIK3CA may specifically modulate the response to fulvestrant therapy. We did not observe any difference in TTP for exon 9
versus 20 PIK3CA mutations, though numbers were small resulting in limited statistical power.
Conclusion: Among patients with metastatic HR+ breast cancer in this study, PIK3CA mutations are associated with decreased
time to progression with fulvestrant, but not with AIs, suggesting that the these mutations may mediate resistance to specific
endocrine therapies. Further studies are needed to confirm these findings and provide mechanistic insights to help guide optimal
selection of endocrine therapy and combination with PI3K directed therapy in metastatic HR+ breast cancer.

Title: Enobosarm for the treatment of metastatic, estrogen and androgen receptor positive, breast cancer. Final results of the
primary endpoint and current progression free survival
Beth Overmoyer1, Pedro Sanz-Altimira2, Ann H Partridge1, Martine Extermann3, Jane Liu4, Eric Winer1, Nancy Lin1, Michael
Hassett1, Leroy Parker1, Ryan Taylor5, Michael Hancock5, Susan Small5 and Mary Ann Johnston5. 1Dana-Farber Cancer Institute,
Inflammatory Breast Cancer Program, Boston, MA; 2Commonwealth Hematology-Oncology, Quincy, MA; 3Moffitt Cancer Center,
Tampa, FL; 4Illinois Cancer Center, Peoria, IL and 5GTx, Inc, Memphis, TN.
Body: Background: Historically, androgens have been utilized for the treatment of breast cancer (BC) as the androgen receptor
(AR) is the most highly expressed steroid receptor in BC (75-95% of estrogen receptor positive (ER+) and 50% of ER negative).
Reports of the use of androgens in metastatic BC (MBC) indicate that women progressing on tamoxifen have the ability to
respond to synthetic androgens with overall response rates in the range of 20-60%; however, these steroidal androgens also
exhibit virilizing side effects, thus limiting clinical use. A non-steroidal, tissue-selective, AR modulator (SARM), such as
enobosarm, offers a targeted approach of AR activation without virilization or estrogenic effects.
Methods: This is a phase II proof of concept study examining the efficacy and safety of once daily enobosarm 9 mg in
post-menopausal women with ER+ MBC who had responded to adjuvant and/or salvage endocrine therapy. Therapy is continued
until patients display evidence of disease progression. The proportion of AR+ patients with clinical benefit response (CBR) at 6
months is the primary endpoint; defined as patients with a complete response (CR), partial response (PR), or stable disease (SD)
as detailed in modified RECIST 1.1. AR status of metastatic disease will be correlated with CBR. Serum prostate specific antigen
(PSA) will be assessed as a biomarker of AR activation by drug. Secondary endpoint is progression free survival (PFS).
Results: Patient demographics: mean age 63.7 years, mean time from diagnosis 11.0 years, 72.7% prior chemotherapy, 89%
(17/19) AR+. After a median follow-up of 81 days (range 7-304 days), preliminary results of 22 patients: 9 SD as best response,
median duration 212 days. Current disposition of patients: 15 PD after a median 80 days (range 15-304 days), 4 SD (1 on
treatment for < 6 months), and 3 early discontinuations (days 7, 28, 255). Five patients have died due to PD off study. Among the
17 evaluable patients, 6 reached the primary endpoint (35.3%; 95% CI=16.6% to 59.4%) with increased PSA, thereby exceeding
the pre-defined statistical threshold requiring that at least 3 of 14 patients with an AR+ metastatic lesion demonstrate clinical
benefit. CR or PR has not been observed. Current six month Kaplan-Meier estimate of PFS is 43.8% (95% confidence
interval=19.5% to 68.1%). Enobosarm is well-tolerated with common grade 1/2 toxicities of nausea (8%), menopausal symptoms
(13%), pain (25%), fatigue (10%), weight change (4%); 5 (4%) grade 3 toxicities (3 unrelated to drug, 1 bone pain, 1 fatigue), and
no grade 4 or higher toxicities reported. Final results of the study will be available in the fourth quarter of 2014.
Conclusions: Enobosarm demonstrates promise as a novel endocrine agent for AR+ MBC. The primary endpoint has been
achieved, with 6/17 AR+ patients meeting statistical threshold for success (35% CBR at 6 months). Serum PSA appears to be a
surrogate marker for AR activity associated with enobosarm administration. Based upon these favorable preliminary findings, a
larger phase II study is anticipated.

Title: Role of pharmacogenetics in metastatic breast cancer (MBC) patients treated with exemestane as first-line hormonal
therapy. An Italian multicentre study
Sara Gagno1, Mauro Mansutti2, Daniele Rossini3, Diana Crivellari4, Chiara Zanusso5, Silvana Saracchini6, Donata Sartori7 and
Giampietro Gasparini3. 1Doctoral School in Pharmacological Science, University of Padua, Italy; 2S. Maria della Misericordia;
3
A.C.O. San Filippo Neri, Rome; 4CRO National, Cancer Institute, Aviano (PN); 5Post Graduate School Clinical Pharmacology,
University of Milan, Milan, Italy; 6A.O. Santa Maria degli Angeli, SC Oncology, Pordenone (PN) and 7Azienda ULSS n.13, U.O.C.
Oncologia ed Ematologia Oncologica Mirano- Dolo- Noale (VE).
Body: Background: In a subset of MBC patients exemestane is not effective. Single Nucleotide Polymorphisms (SNPs) of genes
involved in estrogen availability, in pharmacokinetics (PK) and pharmacodynamics (PD) of exemestane could be responsible for
responsiveness. CYP19A1_Ex11+410A/C rs4646 was found to correlate with improved overall survival in patients treated with
letrozole(1) and anastrozole (2).
Methods: This multicentre study enrolled 423 patients with diagnosis of MBC or locally advanced-ER-positive breast cancer
treated with exemestane as first-line hormonal therapy (prior adjuvant therapy and chemotherapy for metastatic setting were
admitted). Objective of the study was to prospectively assess the predictive role of polymorphisms in aromatase gene, in
particular CYP19A1_Ex11+410A/C rs4646, and in genes involved in the PK and PD of exemestane on drugs activity
(Response Rate - RR, time to progression - TTP). The correlation between SNPs and RR was assessed by the Fisher's exact
test two-way. The CYP19A1_Ex11+410A/C rs4646, CYP1B1*3_4326G/C (rs1056836) and other not pre-planned
polymorphisms were analyzed. Moreover, in a subset of patients, PK and PD analysis were also performed.
Results: At present time we report the preliminary results on 196 patients. Among these patients, CYP19A1_Ex11+410A/C
rs4646 SNP shows no significant association with RR, whereas in patients carrying at least one variant allele of
CYP1B1*3_4326G/C (rs1056836) polymorphism, a statistically significant association with a better response to exemestane was
found (dominant model: OR =.436, 95% CI =.21 to .89, p = .029; Clinical benefit, dominant model: OR =.362, 95% CI =.17-0.75, p
= .007, Fisher's Exact Test in accordance with the two-way). The same SNP seems, in the pharmacodynamic analysis, to be
associated with the level of estrogen suppression (p =.0508 according to the Kruskal-Wallis ANOVA) and associated to better RR
(p =.0287; Clinical B benefit: p =.0209 according to the Wilcoxon-Mann-Whitney test), providing a basis for phenotypic
association. The overall results on the entire cohort of patients enrolled are under evaluation as well as the data concerning TTP,
toxicity and the correlation with PR, HER2 and Ki67 tumor tissue expression.
Conclusions: A predictive role of the CYP1B1*3_4326G/C (rs1056836) polymorphism for RR to exemestane as first- line therapy
has been found. Our study suggests that a simple genetic evaluation from peripheral blood, performed prior to therapy, may allow
the identification of the patients who are more likely to be responsive to exemestane.

Title: Impact of visceral metastases on treatment patterns for hormone-receptor-positive (HR+)/human epidermal growth factor
receptor-2-negative (HER2-) metastatic breast cancer (mBC)
James Signorovitch1, Jenny Wang2, Ruo-Ding Tan1, Darren Thomason1, Andrew Kageleiry1 and Elyse Swallow1. 1Analysis Group,
Inc, Boston, MA and 2Novartis Pharmaceuticals Corporation, East Hanover, NJ.
Body: Introduction:
Guidelines recommend reserving chemotherapy (CT) for HR+/HER2- mBC until either endocrine-based therapy (ET) is no longer
effective or for patients with visceral crisis or need for rapid tumor control. The extent to which visceral metastases (VM),
regardless of whether they are contributing to visceral crisis, impact real-world treatment patterns in mBC is not well understood.
This retrospective analysis characterizes treatment patterns with ET and CT among HR+/HER2- mBC patients stratified by
presence of VM.
Methods:
Postmenopausal women (age 50 years) diagnosed with HR+/HER2- mBC were identified from the MarketScan database
(2002Q3-2012Q2). Patients who initiated ET without CT for mBC were followed until transition to CT, discontinuation of ET (>90
days without evidence of ET), or end of data or insurance eligibility. Upon initiation of each line of therapy, patient characteristics
including age, number of metastatic sites, and presence of VM were compared between those receiving ET and those receiving
CT. The number of lines of ET received before CT, and the total duration of ET were described for patients with and without VM.
Results:
Of the 19,120 patients who initiated treatment for mBC, 5,418 received 2nd-line treatment and 1,471 received 3rd-line treatment.
In each of these 3 lines, the corresponding numbers of patients receiving CT were 7,575 (40%), 2,397 (44%) and 650 (44%).
Among the patients who received CT in each line, the majority did so without evidence of VM: 5,821 (77%), 1,511 (63%) and 435
(67%), in the 1st, 2nd and 3rd lines, respectively. Within each line, the average patient receiving CT was approximately 3 to 5
years younger and had 0.02-0.29 more metastatic sites, compared with the average patient receiving ET. Among the 11,545
patients who initiated 1st-line ET, 9,315 (81%) did not have evidence of VM. Patients with versus without VM upon initiation of
1st-line ET received similar average numbers of lines of ET (1.32 versus 1.37 lines). However, the median time prior to transition
to CT, discontinuation of ET, or loss to follow-up was shorter for patients with VM compared to those without VM (5 months
versus 10 months). In addition, patients with VM appeared to have shorter durations of ET at each line compared with patients
without VM (Table).
Conclusions:
Although the present analysis could not distinguish visceral crisis from the broader group of patients with visceral metastases, it is
notable that a large majority of patients receiving CT did so without evidence of VM. This indicates an unmet need during the
study period (2002-2012) for effective disease control among patients without VM for whom CT was not a preferred option.
Table. Durations (months) of Endocrine Therapy
Total number of lines of ET observed, with and without VM
One Line
Line of ET / patient group
VM
Two lines
Without VM
VM
Without VM
Three lines
VM
Without VM
All Patients, median (IQR)
3 (1-8)
6 (2-16)
11 (6-23)
15 (8-27)
23 (14-35)
26 (16-41)
First line
3 (1-8)
6 (2-16)
6 (3-15)
7 (3-16)
8 (3-17)
9 (3-17)
Second line
3 (1-6)
4 (2-10)
6 (3-12)
5 (3-11)
Third line
3 (1-5)
4 (2-9)
VM = visceral metastases, IQR = interquartile range

Title: A phase III study of fulvestrant 500 mg versus 250 mg in postmenopausal Chinese women with advanced breast cancer
and disease progression following failure on prior antiestrogen or aromatase inhibitor therapy: Supporting superior clinical benefit
for the 500 mg dose
Zefei Jiang1, Qingyuan Zhang2, Zhimin Shao3, Kunwei Shen4, Li Li5, Jifeng Feng6, Zhongseng Tong7, Kangsheng Gu8, Xiaojia
Wang9, Binghe Xu10, Guofang Sun11, Huifang Chen11 and Yuri Rukazenkov12. 1Beijing 307 Hospital, Beijing, China; 2Tumour
Hospital of Harbin Medical University, Harbin, Heilongjiang, China; 3Tumor Hospital of Fudan University, Shanghai, China;
4
Shanghai Ruijin Hospital, Shanghai, China; 5First Affiliated Hospital of Dalian Medical University, Dalian, Liaoning, China;
6
Jiangsu Cancer Hospital, Nanjing, Jiangsu, China; 7Tianjin Cancer Hospital, Tianjin, China; 8First Affiliated Hospital of Anhui
Medical University, Hefei, Anhui, China; 9Zhejiang Cancer Hospital, Hangzhou, Zhejiang, China; 10Cancer Hospital, Chinese
Academy of Medical Sciences, Beijing, China; 11AstraZeneca China, Shanghai, China and 12AstraZeneca Pharmaceuticals,
Macclesfield, United Kingdom.
Body: Background: In the international Phase III COmparisoN of Faslodex In Recurrent or Metastatic breast cancer (CONFIRM)
study, fulvestrant 500 mg was associated with significantly longer progression-free survival (PFS) over the 250 mg dose (hazard
ratio [HR] 0.80; 95% confidence interval [CI] 0.68, 0.94; p=0.006) in postmenopausal women with advanced breast cancer (ABC)
following failure on prior endocrine therapy. There were no clinically meaningful differences between the treatment groups in
terms of the incidence or severity of adverse events. The present study was designed to compare the efficacy and safety of
fulvestrant 500 mg versus 250 mg in a Chinese population for registration purposes.
Methods: This was a Phase III randomized, double-blind study in a Chinese population (ClinicalTrials.gov: NCT01300351).
Postmenopausal women with estrogen receptor positive (ER+) ABC following failure on prior endocrine (antiestrogen [AO] or
aromatase inhibitor [AI]) therapy were randomized 1:1 to fulvestrant 500 mg or 250 mg. Patients (pts) were stratified by
post-AO/post-AI status and enrollment of post-AI pts was capped at 45%. Primary study endpoint was PFS. Consistency with the
global CONFIRM study was to be concluded if the HR for the treatment comparison of PFS was <1 (full analysis set; stratified
log-rank test); the study was not powered to detect significant differences between treatment groups. Secondary endpoints
included pharmacokinetics, ORR, CBR, DoR, DoCB, safety and tolerability.
Results: 221 pts were randomized to fulvestrant 500 mg (n=111) or fulvestrant 250 mg (n=110). 121 pts were in the post-AO
subgroup and 100 pts were in the post-AI subgroup. Demographic and baseline characteristics were balanced between
fulvestrant 500 mg and fulvestrant 250 mg and comparable with those in the global CONFIRM study. 98% (119/121) in the
post-AO subgroup and 92% (92/100) in the post-AI subgroup had adjuvant endocrine therapy, while only 12% (14/121) in the
post-AO subgroup and 51% (51/100) in the post-AI subgroup used salvage endocrine therapy. At the time of the primary analysis,
152 progression events (69%) had occurred (post-AO 59% [71/121]; post-AI 81% [81/100]). Median PFS was 8.0 months (m) in
the fulvestrant 500 mg group vs 4.0 m in the 250 mg group (HR 0.75; 95% CI 0.54, 1.03; p=0.078); the predefined criterion for
consistency with the global CONFIRM study was met. In a predefined subgroup analysis of PFS, the HR for fulvestrant 500 mg vs
250 mg was <1 in both post-AO (median PFS 8.1 m vs 5.6 m; HR 0.86; 95% CI 0.54, 1.37) and post-AI (median PFS 5.8 m vs 2.9
m; HR 0.65; 95% CI 0.42, 1.03) subgroups. Secondary endpoints favored fulvestrant 500 mg over 250 mg, with the exception of
median DoR. Safety and tolerability profiles were consistent with the known safety profile of fulvestrant.
Conclusions: Data from the present study support the superior clinical benefit of fulvestrant 500 mg vs 250 mg demonstrated in
the global CONFIRM study, in postmenopausal Chinese women with ER+ ABC. Hazard ratios favoring fulvestrant 500 mg were
observed in both the post-AO and post-AI settings.

Title: Real-world treatment durations without chemotherapy (CT) for hormone-receptor-positive (HR+)/human epidermal growth
factor receptor-2-negative (HER2-) metastatic breast cancer (mBC)
Elyse Swallow1, Jenny Wang2, Darren Thomason1, Ruo-Ding Tan1, Andrew Kageleiry1 and James Signorovitch1. 1Analysis Group,
Inc, Boston, MA and 2Novartis Pharmaceuticals Corporation, East Hanover, NJ.
Body: Background:
Guidelines recommend maximizing the duration of effective treatment that can be achieved without chemotherapy for HR+/HER2mBC, with the goal of avoiding toxicity and preserving quality of life. A target of three lines of ET is suggested, with CT reserved
for patients requiring rapid control of symptomatic disease or who no longer benefit from ET. This study describes treatment
patterns for HR+/HER2- mBC over the past decade.
Methods:
Postmenopausal women (age 50) with HR+/HER2- mBC were identified from the MarketScan databases (2002Q3-2012Q2).
Patients whose first treatment after mBC diagnosis included ET without CT were followed until first use of CT, discontinuation of
ET (>90 days without evidence of ET), or end of data or insurance eligibility. The distribution of agents used prior to CT was
assessed for up to 3 lines of therapy. Median treatment durations were calculated by agent and by line among patients who
received at least 3 lines of treatment that included ET without CT.
Results:
Of 19,120 HR+/HER2- mBC patients, 11,545 (60%) initiated a 1st-line therapy for mBC that included ET without CT. The
remaining 40% initiated with CT. Of the 11,545 patients who initiated ET, the majority received the non-steroidal aromatase
inhibitors (NSAIs) anastrozole (38%) and letrozole (27%) in the 1st-line. Among these patients, 3,021 (26%) patients received
2nd-line ET without CT, with the majority receiving exemestane (20%), anastrozole (18%), letrozole (18%) or fulvestrant (18%). A
total of 821 patients, comprising 7% of the patients who received 1st-line ET, received at least 3 consecutive lines of ET without
CT. The majority of these patients received exemestane (20%), tamoxifen (19%) or fulvestrant (19%) in the 3rd line. Among these
821 patients, the median treatment durations were 9, 6, and 3 months in the 1st, 2nd, and 3rd lines, respectively. Median
durations for each ET agent were generally lower in later versus earlier treatment lines (Table). At the time of data cutoff for this
study, limited data were available for everolimus treatment (n=21).
Conclusions:
The large majority of patients did not receive the guideline-recommended target of 3 lines of ET. Additionally, the shorter
durations of therapy observed within each subsequent line of ET suggest that there may be diminishing returns for sequential use
of ET monotherapies for HR+/HER2- mBC. These findings indicate an unmet need for more effective alternatives to available ET
monotherapies after the initial ET discontinuation during the study period (2002-2012).
Table. ET treatment duration (months) among the 821 patients who received at 3 three consecutive lines of ET without CT
Line 1
Treatments/line of ET*
N (%)
Line 2
Median
N (%)
Line 3
Median
N (%)
Median
Anastrozole
339 (41.3)
135 (16.4)
113 (13.8)
Letrozole
243 (29.6)
10
165 (20.1)
109 (13.3)
Tamoxifen
114 (13.9)
140 (17.1)
159 (19.4)
Exemestane
42 (5.1)
186 (22.7)
161 (19.6)
Fulvestrant
37 (4.5)
98 (11.9)
158 (19.2)
Megestrol
23 (2.8)
80 (9.7)
104 (12.7)
*Treatment duration of other ETs (i.e., toremefene, raloxifene and combination therapy) with less than 2% of users are not
reported.

Title: First report of clinicopathological analysis in neoadjuvant treatment phase in NEOS: A randomized study of adjuvant
endocrine therapy with or without chemotherapy for postmenopausal breast cancer patients who responded to neoadjuvant
letrozole
Takehiko Sakai1, Hiroji Iwata2, Yoshie Hasegawa3, Rikiya Nakamura4, Hiromitsu Akabane5, Shoichiro Ohtani6, Masahiro
Kashiwaba7, Naruto Taira8, Tatsuya Toyama9, Yutaka Yamamoto10, Tomomi Fujisawa11, Norikazu Masuda12, Takuhiro
Yamaguchi13, Hirofumi Mukai14 and Yasuo Ohashi15. 1Cancer Institute Hospital, Japanese Foundation for Cancer Research,
Tokyo, Japan; 2Aichi Cancer Center Hospital, Nagoya, Japan; 3Hirosaki Municipal Hospital, Hirosaki, Japan; 4Chiba Cancer
Center, Chiba, Japan; 5Hokkaido P.W.F.A.C. Asahikawa-Kosei General Hospital, Asahikawa, Japan; 6Hiroshima City Hospital,
Hiroshima, Japan; 7Iwate Medical University School of Medicine, Morioka, Japan; 8Okayama University Hospital, Okayama,
Japan; 9Nagoya City University Hospital, Nagoya, Japan; 10Kumamoto University Hospital, Kumamoto, Japan; 11Gunma
Prefectural Cancer Center, Ota, Gunma, Japan; 12NHO Osaka National Hospital, Osaka, Japan; 13Tohoku University Graduate
School of Medicine, Sendai, Japan; 14National Cancer Center Hospital East, Kashiwa, Japan and 15Chuo University, Tokyo,
Japan.
Body: Background: Whether adjuvant chemotherapy is required for patients with intermediate-risk endocrine-responsive
postmenopausal breast cancer remains unknown. The New primary Endocrine-therapy Origination Study (NEOS: N-SAS BC06
study: UMIN 000001090 (http://www.umin.ac.jp/) was a randomized controlled trial to verify the necessity of adjuvant
chemotherapy in node-, ER+, and HER2- postmenopausal breast cancer patients who responded to neoadjuvant endocrine
therapy. The primary registration and primary treatment have finished. This report evaluated clinical responses and radiological
findings in the neoadjuvant LET treatment phase of 6 months.
Methods: Patients meeting eligibility criteria received LET preoperatively in weeks 24-28 after primary enrollment. Patients
evaluated as complete response (CR), partial response (PR) or stable disease (SD) by each investigators underwent secondary
enrollment and will be divided at random into two arms, an arm given LET for 4.5-5 years after chemotherapy and another arm
given only LET for 4.5-5 years. Patients evaluated as progressive disease during LET treatment will receive discretionary
treatment. The primary endpoint was disease-free survival and secondary endpoints were overall survival, clinical response rate
in neoadjuvant treatment phase, pathological response, breast-conserving surgery rate, DFS/OS in subgroups according to
clinical response, safety, HRQOL, and cost-effectiveness.
Results: Between May 2008 and June 2013, 905 patients entered primary registration. We excluded 42 patients without
confirmed data. The 863 included patients characteristics at baseline are: median age: 63 years old, median Body Mass Index
(BMI):23.90, T1c:37%, T2:63%, and PgR+:79%, and 74% of patients had planned breast-conserving surgery (BCS). The clinical
responses were evaluated with calipers, ultrasound and MRI (or CT) at the baseline and end of treatment before surgery. Clinical
response rates were 2, 48, 46 and 4% in CR, PR, SD and PD, respectively. Excluding those who could not enter secondary
registration according to the protocol, 83 (56.1%) of 148 patients with planned total mastectomy were converted to BCS according
to final radiological evaluations before surgery. The correlation between the tumor size on MRI (r=0.87) after neoadjuvant LET
and the pathological invasive tumor size was stronger than with other modalities (r=0.68, 0.57, 0.33 by CT, US and calipers,
respectively). On univariate analysis, a small tumor size (T1c rather than T2), PgR+, HER2:2+ and a high BMI at the baseline
were significantly correlated with the clinical response. On multivariate analysis, PgR + (HR: 1.13, 95%CI: 1.01-1.27, p=0.032)
and a small tumor size (HR: 1.11, 95%CI: 1.03-1.17, p=0.003) were significant independent predictors of the clinical response.
Conclusion: This is the first clinical report of neoadjuvant hormone therapy for early breast cancer. Neoadjuvant LET therapy
improved BCS rates. MRI was useful for predicting the residual pathological invasive tumor size. PgR + and a small tumor size at
the baseline were significant independent predictors of the clinical response.

Title: Neoadjuvant letrozole and lapatinib is feasible in Asian postmenopausal women with estrogen receptor (ER) and human
epidermal growth factor receptor-2 (HER-2) positive breast cancer [Neo-All-In]: First efficacy and safety report
Ji Hyun Park1, Myoung Joo Kang2, Jin-Hee Ahn1, Jeong Eun Kim1, Kyung Hae Jung1, Gyung-Yub Gong3, Hee Jin Lee3, Byung-Ho
Son4, Sei-Hyun Ahn4, Hak-Hee Kim5, Hee Jung Shin5, Dae-Hyuk Moon6 and Sung-Bae Kim1. 1Asan Medical Center, University of
Ulsan College of Medicine, Seoul, Korea; 2Inje University Haeundae Paik Hospital, Busan, Korea; 3Asan Medical Center,
University of Ulsan College of Medicine, Seoul, Korea; 4Asan Medical Center, University of Ulsan College of medicine, Seoul,
Korea; 5Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea and 6Asan Medical Center, University of
Ulsan College of Medicine, Seoul, Korea.
Body: Purpose Neo-ALL-In (NCT 01275859) is a single center, prospective study aimed to evaluate the recruitment feasibility,
efficacy and safety profiles as well as biologic features of neoadjuvant letrozole plus lapatinib in postmenopausal women with ER
and HER2 positive breast cancer.
Methods Postmenopausal women with stage IIA to IIIB ER and HER-2 positive breast cancer were eligible. Patients received
combination therapy of letrozole 2.5 mg orally daily plus lapatinib 1,500 mg orally daily for 18-21 weeks before surgery. Clinical
responses were assessed by clinical palpation, ultrasonography (US), mammography and/or MRI. Tissue and/or blood samples
were collected for analysis of biomarkers at three time points (baseline, day 15, and before surgery). Baseline Fluorine-18
Fluorodeoxyglucose (18F-FDG) and Fluorine-18 Fluoroestradiol (18F-FES) PET-CT imagings were obtained.
Results Among twenty-four patients enrolled, 22 patients underwent surgery while 1 patient is currently on neoadjuvant therapy
and the other patient is waiting for surgery. Among 22 patients completed surgery, 16 patients (72.7%) completed planned
neoadjuvant letrozole and lapatinib, whereas 3 patients (13.6%) prematurely terminated the treatment and proceeded to surgery
due to minimal clinical response or progression. Except grade 3 liver toxicities revealed in 3 patients (13.6%), which resulted in
sequential dose reduction and discontinuation, adverse events were mainly grades 1 to 2 (Skin, 83.3%; GI, 77.3%), and these
were generally tolerable with excellent compliance. Overall clinical response rates including complete and partial response was
72.7% (n=16), and pathologic complete response in breast (pCR; ypT0-is) was 4.5% (n=1).
Clinical and pathologic responses of 22 patients by assessment modalities
pathologic response in
pathologic response
breast and lymph nodes
in breast (ypT0-is)
(ypT0-is N0)
Total (N=22, %)
Clinical
palpation
US
CR
1 (4.5)
0 (0.0) 0 (0.0)
1 (4.5) 1 (4.5)
0 (0.0)
1 (4.5)
PR
17 (77.3)
12
(54.5)
9 (40.9)
9
11 (50.0)
(40.9)
15 (68.2
15 (68.2)
No change
3 (13.6)
9
(40.9)
13 (59.1)
5
9 (40.9)
(22.7)
7 (31.8)
4 (18.2)
PD
1(4.5)
1 (4.5) 0 (0.0)
1 (4.5) 1 (4.5)
0 (0.0)
2 (9.1)
Not
available/evaluable
0 (0.0)
0 (0.0) 0 (0.0)
6
0 (0.0)
(27.3)
0 (0.0)
0 (0.0)
Mammography MR
Overall
response
CR, complete response; PR, partial response; PD, progressive disease; US, ultrasonography
In analyses of biomarkers thus far, 81.8% of patients showed stationary expression of HER-2, 54.5% of patients showed
decrease in Ki-67 expression, and 27.3% of patients showed increase in ER expression from baseline to surgery by
immunothistochemistry (IHC) staining. Decreased expression of ER after surgery by IHC staining was significantly correlated with
poor clinical response (p=0.004). However, no significant differences in baseline SUVmax in FDG-PET were found between
responders and non-responders (8.8 VS 10.7, p=0.53).
Conclusion This chemo-free combination neoadjuvant therapy was feasible, with comparable efficacy outcomes and
manageable toxicities profiles. Updated data on 18F-FES PET-CT and biomarkers will be provided.

Title: Association between ischemic cardiac events and targeting of the internal mammary nodal region with adjuvant radiation for
breast cancer
Adam L Liss1, Kent A Griffith2, Reshma Jagsi1, Jean M Moran1, Robin B Marsh1, Todd M Koelling3 and Lori J Pierce1. 1University
of Michigan, Ann Arbor, MI; 2Biostatistics Unit, University of Michigan, Ann Arbor, MI and 3University of Michigan, Ann Arbor, MI.
Body: Background: Cardiac toxicity has been well documented following adjuvant breast cancer radiation therapy (RT) using
outdated treatment methods. Limited information exists, however, regarding cardiac outcomes following contemporary RT fields
and techniques. Inclusion of the internal mammary nodal (IMN) field for appropriately selected patients with breast cancer
coincided with our departments transition to computed-tomography (CT) based planning. The objectives of this study, therefore,
were to assess the risk of ischemic cardiac events following IMN irradiation and to assess the risks following CT-based versus
two-dimensional (2D) planning.
Methods: All patients treated with adjuvant RT for breast cancer and without history of additional chest RT from January 1, 1984
December 31, 2007 in our department were assessed. CT planning for breast cancer began for select patients in 1997 and was
used for all patients as of 2001. The inclusion of the IMN region was determined by review of our clinical database. Ischemic
cardiac endpoints were defined as myocardial infarction, coronary artery bypass grafting procedure, angioplasty/stent placement,
and/or diagnosis of coronary artery disease. A text-based and diagnosis code-based search was used to flag possible endpoints
which were then confirmed by manual chart review. Hypertension (HTN), diabetes (DM), hyperlipidemia (HLD), and anthracycline
use were identified by a diagnosis code-based search.
Results: We identified 2,126 patients who received adjuvant RT. Median follow-up was 9.6 years. 311 (14.6%) patients had IMNs
targeted and 1,813 (85.3%) did not (data not available for 2 patients). RT to the IMNs was not associated with a higher risk of
ischemic cardiac events (HR: 0.88, P = 0.731). 1,072 (50.4%) patients had CT planning, 1,003 (47.2%) had 2D planning, and no
information was available for 51 (2.4%) patients. Overall, there were 56 (5.6%) ischemic events in the 2D cohort and 27 (2.5%) in
the CT cohort. After truncating follow-up to 10 years to account for differential potential follow-up, there were 28 (2.8%) and 23
(2.2%) ischemic cardiac events, respectively. The table lists the association of patient and treatment characteristics and risk of
ischemic cardiac events. HTN, HLD, and DM were each associated with a significantly increased risk of ischemic cardiac events.
Conclusions: After reviewing all patients treated in our department for breast cancer from 1984 2007, we were unable to find an
association between IMN irradiation and ischemic cardiac events. CT-based planning has been shown to allow accurate targeting
of the IMNs. These data suggest that even with inclusion of the IMNs, CT-based planning minimizes dose to the heart and
coronary arteries thereby decreasing the risk of ischemic cardiac toxicity. These results will be followed with time.
Association between characteristics and ischemic cardiac events
Characteristic
Hazard Ratio
95% Confidence Interval
P-value
RT to the IMNs
0.88
0.42-1.83
0.731
2D vs. CT planning
1.14
0.68-1.91
0.613
Left vs. right
1.21
0.79-1.85
0.372
Anthracycline use
0.75
0.44-1.26
0.270
HTN
5.61
2.06-15.30
0.001
HLD
1.90
1.25-2.90
0.003
DM
3.11
1.99-4.86
<0.001

Title: Utilization of hypofractionated radiation therapy for early stage breast cancer in women over 50 years of age
Malolan S Rajagopalan1, Craig Lehocky1, John C Flickinger1, Dwight E Heron1, Paniti Sukumvanich2, Joseph L Kelley2, Gretchen
M Ahrendt3 and Sushil Beriwal1. 1University of Pittsburgh Cancer Institute, Pittsburgh, PA; 2University of Pittsburgh Cancer
Institute, Pittsburgh, PA and 3University of Pittsburgh Cancer Institute, Pittsburgh, PA.
Body: Purpose: Hypofractionated whole breast irradiation (HF-WBI) following breast conserving surgery for early stage breast
cancer is now recommended for women over 50 years of age by the American Society for Radiation Oncology (ASTRO). Herein,
we explored the rate of HF-WBI utilization and its associated patterns of care using the National Cancer Data Base (NCDB), a
comprehensive oncology outcomes database which captures 70% of all newly diagnosed cancer patients in the US.
Methods: We utilized NCDB to identify women aged 50 diagnosed with T1N0/T1Nx invasive breast cancer or Tis (DCIS) who
underwent breast conserving surgery and adjuvant whole breast external beam radiation therapy. HF-WBI was defined when a
definitive radiation dose (40 Gy) was delivered in 23 fractions. We evaluated factors associated with the use of HF-WBI through
exploratory univariate and multivariable analyses.
Results: A total of 311,071 patients from 1998-2011 met inclusion criteria. The commonest HF-WBI regimen was 42.56 Gy in 16
fractions. The rate of HF-WBI utilization increased over time from 0.4% in 1998 to 16.5% in 2011 (p<0.001). HF-WBI was
delivered more often for women of advancing age. In 2011, HF-WBI was delivered in 10.0%, 14.2%, 24.2% and 35.6% of women
in their 50s, 60s, 70s and 80+ (p<0.001). Multivariable analysis correlated significantly increased use of HF-WBI with (in order of
association): later year of diagnosis, advancing age (decade), treatment in academic center, regional location in US, lower grade
of disease, white race, residence in a higher income area (p<0.001), greater comorbidity score (p<0.03), presence of invasive
cancer (p<0.01), right-sided disease (p<0.01), and greater distance from reporting facility (p<0.001).
Factors Associated with Increased Use of HF-WBI
Association with HF-WBI
Odds Ratio of HF-WBI Use (95% confidence interval)
Year of Diagnosis (2011 vs. prior to 2009)
7.14 (6.85-7.52)
Age (Decade, 80+ years)
6.45 (6.06-6.90)
Facility Type: Academic
4.00 (3.69-4.33)
Location (Mountain Region)
1.82 (1.68-1.97)
Grade 1
1.40 (1.33-1.47)
Race: White
1.38 (1.25-1.53)
Income Quartile 4 (>$46K/year)
1.30 (1.21-1.40)
Comorbidity Index (High, 2)
1.15 (1.02-1.31)
Invasive Cancer
1.09 (1.04-1.14)
Laterality (Right Side)
1.05 (1.02-1.09)
Distance from Treatment Facility
1.002 per mile (1.002-1.002)
Conclusions: This comprehensive patterns of care study of HF-WBI in women 50 years old with early stage breast cancer from a
national database identified a substantial increase in the utilization of this technique with time. Advancing patient age and
treatment at academic facility were also strong predictors of delivery of HF-WBI. Disparities in utilization of HF-WBI and overall
employment of this technique, even in later years, remain suboptimal. Strategies to identify and break barriers to the adoption of
HF-WBI should be explored.

Title: Long term outcomes in patients with phyllodes tumor of the breast: The UT MD Anderson experience
Bobbi A Porche1, Pamela K Allen2, Simona Shaitelman2, B Ashleigh Guadagnolo2, Wendy A Woodward2, Constance Albarracin2,
Abenaa M Brewster2, Kelly K Hunt2 and Welela Tereffe2. 1UT Medical School at Houston, Houston, TX and 2MD Anderson Cancer
Center, Houston, TX.
Body: Purpose: Phyllodes tumor is a relatively rare disease of the breast, and the risk factors for local versus distant failure are
unclear. The goal of this study was to identify factors influencing local, distant, and cause-specific survival outcomes in patients
with phyllodes tumors, and to assess the impact of local radiotherapy (RT).
Methods: We retrospectively reviewed the records of 387 patients recorded in the MD Anderson tumor registry as having
"cystosarcoma phyllodes" or "phyllodes". After excluding patients with recurrent disease at presentation to the institution,
concurrent invasive ductal carcinoma, absence of confirmation of phyllodes tumor, or inadequate follow up after surgery, 229
patients diagnosed from 1964-2011 were evaluable. The median tumor size was 4 cm (range, 0.7-28 cm); histology was benign in
29%, indeterminate in 7%, and malignant in 41% (23% unknown). Stromal overgrowth was present in 18% (n=40). Local therapy
consisted of breast-conserving surgery (n=184) or mastectomy (n=43); 15% (n=34) also received local RT. Chemotherapy was
administered in 9% (n=21). We used Kaplan-Meier analyses and Cox proportional hazards models to estimate the associations
between patient/tumor characteristics and treatment on local control (LC), distant metastasis-free survival (DMFS), and cause
specific survival (CSS).
Results: At a median follow-up of 76 months (range, 1-485 months), the actuarial 5-yr LC for the entire cohort was 75%; 5-yr
DMFS was 78%; and 5-yr CSS was 86%. Factors influencing LC included receipt of RT (5-yr LC 97% versus 72%, p=.005) and
age over 50 (85% versus 72%, p=.044). Neither malignant histology nor stromal overgrowth increased the risk of local failure.
Factors influencing DMFS included malignant histology (5-yr DMFS 65% versus 91%, p=.001) and stromal overgrowth (50%
versus 84%, p<.001). Factors influencing CSS included malignant histology (5-yr CSS 77% versus 100%, p=.002) and stromal
overgrowth (58% versus 92%, p<.001). Patients with stromal overgrowth or malignant histology were much more likely to receive
RT (p<.001 for both factors). RT improved 5-yr LC for both breast conserved and mastectomy patients, but did not improve DMFS
or CSS. In multivariable models, the following associations were noted (HR=hazard ratio; CI= 95% confidence interval): LC RT
(HR 0.13, CI .03-.55); DMFS malignant histology (HR 2.99, CI 1.41-6.34); CSS malignant histology (HR 4.18, CI 1.43-12.24).
Conclusions: Adjuvant RT after surgery improves local control in phyllodes tumor; however, it does not improve DMFS or CSS.
Stromal overgrowth and malignant histology are associated with worse DMFS and CSS, but do not impact local control. Therefore
malignant histology and stromal overgrowth should not be deciding factors in the use of local RT, as escalated local therapy in
patients with these tumor characteristics does not improve DMFS or CSS. Other strategies should be considered for patients with
phyllodes tumor who are at high risk of distant failure, including systemic therapy.

Title: The adoption of hypofractionated whole breast irradiation for early-stage breast cancer: A national cancer data base
analysis
Elyn H Wang1, Sarah S Mougalian2, Pamela R Soulos3, Charles E Rutter4, Suzanne B Evans4, Bruce G Haffty5, Cary P Gross6
and James B Yu4. 1Yale University School of Medicine, New Haven, CT; 2New Haven, CT; 3Cancer Outcomes, Public Policy, and
Effectiveness Research Center at Yale, New Haven, CT; 4Department of Therapeutic Radiology, New Haven, CT; 5Rutgers
Cancer Institute of New Jersey and Robert Wood Johnson Medical School, New Brunswick, NJ and 6Department of Internal
Medicine, New Haven, CT.
Body: Purpose
Hypofractionated whole breast radiation therapy (hypofractionation) for early-stage breast cancer is a treatment innovation that is
both supported by high quality randomized trial evidence and clinical guidelines, and is more cost effective and convenient than
conventional fractionation. However, whether hypofractionation has been adopted nationally, and what factors are related to its
adoption, are unknown.
Methods
We performed a retrospective study of breast cancer patients in the National Cancer Data Base from 2004-2011 who were
treated with radiation therapy and met eligibility criteria for hypofractionation. We used logistic regression to identify factors
associated with receipt of hypofractionation (vs. conventional fractionation).
Results
We identified 13,271 (11.7%) and 99,996 (88.3%) women with early-stage breast cancer who were treated with hypofractionation
and conventional fractionation, respectively. The use of hypofractionation increased significantly, with 5.4% of patients receiving it
in 2004 compared with 22.8% in 2011 (P<0.001 for trend). Patients living >50 miles from the cancer reporting facility had
increased odds of receiving hypofractionation (OR 1.57 [95% CI 1.44-1.72], p<.001).
Adoption of hypofractionation was associated with treatment at an academic center (p<0.001) and living in an area with high
median income (p<0.001). Hypofractionation was less likely to be used in patients with high risk disease, such as increased tumor
size (p<0.001) or poorly differentiated histologic grade (p<0.001).
Adjusted odds ratios for receipt of conventional fractionation vs hypofractionation
Feature (Reference)
OR (95% CI)
p-value
Community
0.38 (0.35-0.42)
<0.001
Comprehensive Community
0.51 (0.48-0.53)
<0.001
Other
0.64 (0.52-0.78)
<0.001
Northeast
0.93 (0.88-0.99)
0.02
Midwest
1.04 (1.00-1.11)
0.07
West
1.69 (1.59-1.78)
<0.001
60-69
1.22 (1.16-1.29)
<0.001
70-79
1.77 (1.66-1.89)
<0.001
>=80
3.12 (2.88-3.38)
<0.001
Black
0.96 (0.89-1.05)
0.41
White Hispanic
1.23 (1.10-1.36)
<0.001
Facility Type (Academic)
Facility Location (South)
Age in yrs (50-59)
Race (White)
Other
1.23 (1.13-1.34)
<0.001
No insurance
1.16 (0.96-1.41)
0.11
Medicaid
0.97 (0.86-1.10)
0.71
Medicare
1.05 (1.00-1.11)
0.05
Other government
0.81 (0.63-1.05)
0.12
Unknown
1.22 (1.03-1.44)
0.02
30 000 - 35 000
0.95 (0.87-1.04)
0.30
35 000 - 45 999
0.97 (0.90-1.06)
0.54
>= 46 000
1.25 (1.16-1.35)
<0.001
Unknown
1.04 (0.86-1.25)
0.70
Urban
0.88 (0.83-0.95)
0.001
Rural
1.10 (0.99-1.22)
0.07
>=50 miles
1.57 (1.44-1.72)
<0.001
Unknown
1.13 (0.91-1.40)
0.27
2005
1.03 (0.92-1.14)
0.65
2006
1.11 (1.00-1.24)
0.05
2007
1.31 (1.18-1.45)
<0.001
2008
1.99 (1.80-2.19)
<0.001
2009
2,88 (2.62-3.15)
<0.001
2010
4.10 (3.62-4.64)
<0.001
2011
5.48 (4.83-6.21)
<0.001
Intermediate
0.94 (0.89-0.97)
<0.01
Poor
0.86 (0.80-0.91)
<0.001
Undifferentiated
1.03 (0.5-1.63)
0.89
Unknown
0.90 (0.82-0.99)
0.04
10-29 mm
0.90 (0.87-0.94)
<0.001
30-49 mm
0.71 (0.62-0.80)
<0.001
>=50 mm
0.94 (0.67-1.32)
0.76
Unknown
0.63 (0.42-0.95)
0.03
Primary Payor (Private insurance)
Median Income Quartile (<$30 000)
Urbanization (Metro)
Distance from reporting facility (<50 miles)
Year of Diagnosis (2004)
Grade (Well-differentiated)
Tumor Size (<10 mm)
Conclusions
The use of hypofractionation is rising and is associated with increased travel distance and treatment at an academic center.
Further adoption of hypofractionation may be tempered by both clinical and non-clinical concerns.

Title: Long-term outcome of electron intraoperative boost and hypofractionated external beam radiotherapy after
breast-conserving surgery in premenopausal women
Maria Cristina Leonardi1, Anna Morra1, Cristiana Fodor1, Luigi Santoro1, Samantha Dicuonzo1, Roberta Lazzari1, Veronica
Dell'acqua1, Elena Rondi1, Federica Cattani1, Floriana Pansini1, Alberto Luini1 and Roberto Orecchia1,2. 1European Institute of
Oncology, Milan, Italy and 2University of Milan, Milan, Italy.
Body: Purpose: To report long-term clinical results on breast cancer patients treated with a electron intraoperative boost followed
by hypofractionated external beam radiotherapy (HEBRT) to the whole breast inside a clinical protocol.
Materials and Methods: Between June 2004 and December 2008, 357 premenopausal women with a diagnosis of early-stage
breast cancer were given breast conservative surgery (quadrantectomy and sentinel node biopsy +\- axillary dissection). During
surgery, an electron intraoperative boost of 12 Gy was administered to the tumour bed. Within 4 weeks after having received the
IORT boost, the whole breast was irradiated with hypofractionated scheme. HEBRT consisted of 13 daily fractions of 2.85 Gy up
to a total dose of 37.05 Gy, with 3-D conformal technique. Median age was 42 years (24-48). Pathologic stages were distributed
as follow: stage 0 4 (1.1%), I 192(53.8%), stage IIA 108 (30.3%), stage IIB 28 (7.8%), stage IIIA 18 (5%), stage IIIC 6 (1.7%), CR
1(0.3 %).
The biomolecular classification included 33 triple negative patients.
All patients but one received adjuvant systemic treatment, according to tumor features. Chemotherapy +/- endocrinotherapy was
offered to 149 patients, while endocrinotherapy alone was given to 197 patients. In all cases, adjuvant chemotherapy commenced
after the completion of HEBRT. Follow-up was carried out on a regular basis by breast surgeons and radiation oncologists.
Results: Median follow-up was 83.1 months (range 9.1-118.9). 340 patients were alive and 317 were free of any disease (89%).
Three patients (0.8%) had nodal regional recurrence, 12 patients (3.4%) had local recurrence, 26 patients (7.3%) had distant
metastasis.13 patients (3.6%) developed contralateral breast cancer.12 patients (3.4%) had a new primary tumor in other site.
15 patients (4.2%) died due to progression of disease, while two died of other causes not related to the breast cancer.
Conclusions: This shortened scheme including intraoperative anticipated boost and hypofractionated whole breast radiotherapy
showed a good local control at 5 years follow-up. The patients compliance is very high and in patients who are in need of
chemotherapy, the time gap between surgery and radiotherapy treatment is avoided.
As follow-up length increases, a greater number of patients will achieve the minimum 5 -year period to receive the first long-term
evaluation and the strength of the results on efficacy is bound to increase.

Title: Lobular histology and partial breast irradiation: To what extent is it a cautionary parameter?
Maria Cristina Leonardi1, Patrick Maisonneuve1, Anna Morra1, Nicole Rotmensz1, Samantha Dicuonzo1, Roberta Lazzari1,
Veronica Dell'acqua1, Federica Cattani1, Mauro Mastropasqua1, Alberto Luini1 and Roberto Orecchia1,2. 1European Institute of
Oncology, Milan, Italy and 2University of Milan, Milan, Italy.
Body: Aim: The high likelihood of multicentricity and multifocality linked to lobular histology (ILC) has called into question the
partial breast irradiation. Patients with ILC are categorized in the cautionary group according to ASTRO guidelines. We looked
into the population treated with intraoperative radiotherapy with electrons (ELIOT) as full dose to investigate the impact of lobular
compared to ductal histology (IDC) in terms of local recurrence.
Matherials and methods: From 1999 to 2007, 2220 patients were treated with breast conserving surgery and intraoperatory
radiotherapy with electrons ( 21 Gy at 90% isodose) as sole treatment. The study population includes both patients treated
outside and inside the clinical randomized phase III trial ELIOT. Out of 2220 patients, 255 (11.5%) presented ILC. We compared
this group with patients having IDC, treated in the same period. The rate of local recurrences has been analyzed, with a minimum
follow-up of 5 years. The role of patients and tumor features on local control has been evaluated.
Results: Compared to IDC group, patients with ILC were older ( 70, 20% vs. 11%), with no lymph node involvement( pN0, 79%
vs. 70%), low grading( G1-2,86% vs.72%), absence of vascular invasion( 81% vs. 98%), higher hormonal receptor status (ER +,
97% vs. 88%), HER2 negativity (96% vs. 89%), low proliferative index( Ki-67, 23% vs. 44%). No differences were noticed with
regard to the technical parameters of ELIOT. On the whole, the incidence of local relapse was 4.5% at 5 years (0.9/100 per year).
Within the first 5 years of follow-up, no differences were observed between lobular and ductal histology, but after 5 years the
difference became significant (p 0.05). Higher risk of local relapse was also observed in old patients ( over 60) with ILC.
Regarding the biomolecular classification, lobular carcinomas belonging to Luminal A subtype had a more aggressive local
behavior than Luminal A ductal carcinomas, showing a significantly higher rate of local relapse (p 0.0021).
Conclusion: In spite of a favorable tumor profile, a greater incidence of local relapse occurred in tumors with ILC compared to
ductal histology. When considering the anagraphic, histological and biomolecular variables, the old age (>60), the proliferative
index and the luminal A subtype turned out to be significant. Interestingly, the incidence of local relapse becomes statistically
significant 5 years after the treatment. This event points out the importance of an adequate follow-up length, especially in lobular
histology for which other reports on whole breast irradiation in literature describe similar biological behavior .

Title: Simultaneous integrated boost incorporated into a hypofractionated regimen using tomoDirect: Acute toxicity assessment
Maria Cristina Leonardi1, Anna Morra1, Federica Cattani1, Luigi Santoro1, Samantha Dicuonzo1, Raffaella Cambria1, Rosa
Luraschi1, Alessia Bazani1, Roberta Lazzari1, Veronica Dell'Acqua1 and Roberto Orecchia1,2. 1European Institute of Oncology,
Milan, Italy and 2University of Milan, Milan, Italy.
Body: Aim: We report on acute toxicity, which is the secondary endpoint of a phase II clinical trial specifically addressed to
assess the chronic toxicity of a hypofractionated scheme including a simultaneous integrated boost, with intensity modulated
radiotherapy.
Materials and methods: From 2/2012 to 12/2013 194 patients with early breast cancer entered a phase II study on
hypofractionated radiotherapy including a boost dose to the tumor bed. All patients were operated on conservatively. The whole
breast and the tumor bed are planned to receive a dose of 45 Gy and 50 Gy, respectively, in 20 fractions, over 4 weeks.
Treatment plans are generated using the TomoDirect modality,which is available on Tomotherapy Hi-Art System (Madison,WI).
Acute toxicity was evaluated according to the RTOG acute toxicity scale, up to 6 months after the treatment. Afterwards, chronic
toxicity is evaluated using LENT/SOMA scale.
Results: 95% of the volume of the breast and boost PTVs received 99% and 100% of the planned dose, as median values,
respectively and 0.1% (median value) of the entire breast volume received 100% of the boost dose.
The median maximum dose to the breast and to the boost PTVs was 113% and 103.3 %, respectively. At the end of treatment,
the acute toxicity, was distributed as follows.
As far as erythema was concerned, at the end of treatment, 58% of the patients experienced grade 1 erythema, which dropped to
23% one month later. Grade 2 erythema affected 37% of the cases, and after one month, it decreased to 2%.Only 1 patient
(0.5%) complained of Grade 3 erythema at the end of the treatment, which rapidly disappeared afterwards. With regard to breast
oedema, at the end of treatment grade 1 was observed in 16% of the cases, for whom it tended to remain stable after one month,
while grade 2 oedema was noted in 4% of the cases, decreasing to 1.5% on the first month- follow-up visit. Regarding
desquamation, dry desquamation ( grade 1) was observed in less than 10% of cases at the end of treatment, but it tended to
increase to 17% one month later. Patchy moist desquamation (grade 2) was present in 1.9% of the patients at the end of
radiotherapy, and in 1% of them, one month afterwards. Confluent desquamation (grade 3) was noted only in 1 patient (0.5%),
who was receiving concomitant chemotherapy with cyclophospamide, methotrexate and fluoruracil: it was still present 1 month
after the radiotherapy completion, as the patient continued to be on chemotherapy. No significantly different side effects were
observed between the whole breast and the boost area. No patients experienced any lung and cardiac symptoms.
Conclusion: The clinical results of this SIB hypofractionated scheme showed low acute toxicity. In spite of the high dose per
fraction, with the tumor bed receiving an even higher dose /per fraction, acute toxicity was within the limit acknowledged by
literature for conventional fractionation. This non-rotational treatment option allows us to deliver treatment with a traditional
tangent-like dose, without spreading low doses to the adjacent structures. Chronic toxicity will be assessed after 2 years.
Therefore, a longer follow-up is needed to assess the effective tolerance to the SIB schedule.

Title: Tangential fields (TgF) breast radiotherapy (RT): Prospective evaluation of the dose distribution in the sentinel lymph node
area (SLNa) as determined intra operatively by clip placement
Yazid Belkacemi1, Veronique Bigorie2, Quiong Pan1, Romain Bosc2, Ryan Bouaita1, Frederic Pigneur3, Elias Assaf4, Hakima
Badaoui5, Emmanuel Itti6 and Elie Calitchi1. 1APHP, Henri Mondor Breast Center, UPEC; 2APHP, Henri Mondor Breast Center,
UPEC; 3APHP, Henri Mondor Breast Center, UPEC; 4APHP, Henri Mondor Breast Center; 5APHP, Henri Mondor Breast Center
and 6APHP, Henri Mondor Breast Center, UPEC, Creteil, France.
Body: Purpose:
The coverage of axilla contents by radiation therapy (RT) is an important issue in the new era of minimal axillary surgery based
on sentinel lymph node biopsy (SLNB). Data from recent trials demonstrated equivalent survival in breast cancer (BC) patients
with 1-2 positive SLNs with or without axillary lymph node dissection (ALND). In a similar context, AMAROS trial showed that
complete RT coverage of the axilla is a better option than ALND regarding the risk of arm lymphedema. Our recent data showed
that axillary levels are underdosed when only tangential fields (TgFs) are used (Belkacemi et al, Ann Oncol 2013). We aimed to
evaluate the dose distribution in the SLNa defined intra operatively by clips placement. This could be a important for patients with
SLN involvement who have neither ALND nor RT to the axilla.
Materials/Methods:
Twenty-five patients have been prospectively included in this study. They had clips placement in the SLNa during the SLNB
procedure. Breast dose was ranged between 40 to 50Gy in 15 to 25 fractions. Additional boost to the tumor bed of 10 or 16Gy
was delivered in 21 patients. Level I-III and organs at risk were contoured using the RTOG contouring atlas. The SLNa was
defined as 1.5 cm in diameter around the clips. Dose-volume-histograms were analyzed regarding the volumes receiving 95% or
50% of the prescribed dose. Percentages overlap between TgFs and SLNa volume were analyzed to define 3 groups: 100%
overlap ("suitable group" with SNLa completely included in the TgF), > 50% overlap ("partially suitable group" with SLNa partially
included in the TgF) and 0-49% overlap or completely outside the TgFs ("unsuitable group").
Results:
The mean dose delivered to levels I, II, III and SLN area were 25, 5, 2 and 33Gy respectively. The volume covered by the
95%-isodose were respectively 2%, 0%, 0% and 4%. The average dose delivered to level I, II, III and SLN area were higher using
High TgFs vs STgFs (38 vs 22Gy, p=0.004; 11 vs 3Gy, p=0.019; 5 vs 2Gy, p=0.003; 43 vs 31Gy,p=0.02), respectively. HTgFs
covered better 50% of all axilla levels. Boost delivery and initial tumor site did not influence axilla coverage by the TgFs.
The SNLa was totally or partially covered in 48% and 28% of patients, respectively. The mean dose delivered to 95% of the SNLa
was only 22Gy using STgFs and 33Gy with the HTgFs. Using the STgFs, the SNLa was either totally (n=8/20) or partially
(n=6/20) covered by > 50% of dose. Average dose was 46, 34 and 8Gy, respectively in the 3 groups. HTgFs allowed a complete
coverage of the SLNa in all patients.
Conclusion:
In patients undergoing breast conservative therapy, TgFs provide a limited coverage of the SLNa. STgFs allowed total coverage
of this area in less than half of the patients. Thus, SLNa should be delineated in patients who have only SLNB procedure. Some
of these patients with nodal involvement without additional ALND could benefit from HTgFs irradiation or a better-personalized
nodal RT using a dedicated nodal RT technique such that reported in AMAROS trial. The last allow better coverage of the axilla
contents than TgFs.

Title: Patterns of practice of regional node irradiation in the sentinel node biopsy era: Results of the nodal radiotherapy (NORA)
survey. On behalf of the EORTC Breast Working Party of the Radiation Oncology Group (ROG)
Yazid Belkacemi1, Orit Kaidar-Person2, Philippe Poortmans3, Mahmut Ozsahin4, Maria-Clara Valli5, Nicola Russell6, Ian Kunkler7,
Julie Hermans8, Abraham Kuten9, Geertjan van Tienhoven10 and Helen Westenberg11. 1APHP, Henri Mondor Breast Center,
UPEC, Creteil, France; 2Rambam, Haifa, Israel; 3Institute Verbeeten, Tilburg, Netherlands; 4CHUV, Lausanne, Switzerland;
5
Oncology Institute, Southern Switzerland, Switzerland; 6The Dutch Cancer Institute, Amsterdam, Noord-Holland, Netherlands;
7
Edinburgh Cancer Centre, University of Edinburgh, Edinburgh, United Kingdom; 8EORTC Breast Working Party of the Radiation
Oncology Group (ROG), EORTC, Brussels, Belgium; 9Italian Hospital and Rambam Campus, Haifa, Israel; 10Academisch Medisch
Centrum, Amsterdam, Noord-Holland, Netherlands and 11Radiotherapeutisch Samenwerkingsverband Arnhem-Nijmegen
(Radian), Locatie Arnhems Radiotherapeutisch Instituut (ARTI), Amhem, Netherlands.
Body: Objective:
Predicting the outcome of breast cancer (BC) patients based on sentinel lymph node (SLN) status without axillary lymph node
dissection (ALND) is a matter of debate, especially when it comes to the definition of regional nodal irradiation (RNI). This is even
unclear in the framework of primary systemic therapy (PST). The aim of the NORA (NOdal RAdiotherapy) Survey was to examine
the patterns of RNI practiced by European Radiation Oncology centers.
Methods:
A web questionnaire was distributed to EORTC centers, and responses were received during the period between July
2013-January 2014.
Results:
A total of 81 European and 3 non-European answers were analyzed. While 3D planning is performed in 81 (96%) of the centers
for breast irradiation, nodal areas are delineated in only 51 (61%) of the centers. After breast conserving surgery (BCS), only 14
(17%) centers declared to treat systematically the internal mammary chain (IMC), when supraclavicular node RT (SCN-RT) is
indicated. Extra-capsular extension (ECE) is the main factor impacting decision-making regarding IMC and axillary nodal RT
(ALN-RT). Only half of the centers advocate SCN-RT for intermediate risk patients (1-3N+). For macro-metastatic SLN
involvement, there was a significant impact of ECE on decision-making independently from the number of positive LNs.
In the PST setting, only 49 (58%) centers take into account the histologic fibrotic changes of the axillary LNs in post-PST ypN0
patients with unknown pre-PST status. In ypN0 patients with inner and central BC, 32 (39%) and 23 (27%) centers centers
indicate SCN-RT and IMC-RT, respectively. SCN-RT is delivered by 30, 44, 58, 67 centers in patients with ypN(i+), ypN(mi),
1-2N+ and > 3-4N+ disease, respectively. ALN-RT is indicated by in 25% of the centers in patients with ypN(mi) or 1-2N+. Older
age > 70y was not considered as a limited factor for RNI.
Discussion:
The term of RNI is not uniformly defined in the literature. In the historic studies, RNI was dedicated to wide RT including whole
lymphatic pathways of the breast. ALN and IMC RT indications have declined with the time during the last decades because of
the risk of morbidity. However this practice is challenged by the recent RT trials that reported and increase of overall and distant
metastatic survival with RNI and IMC delivery in high-risk N0 and inner-central tumors. In addition, AMAROS trial showed
equivalence between axillary RT and ALND in SLN (mi+) patients. Taking these data altogether in the context of increased
avoidance of ALND, future challenges should concern the multidisciplinary patients selection for RNI and better personalization
of the lymphatic pathways RT according to the risk of locoregional and distant recurrences.
Conclusions:
The NORA Survey is unique by evaluating the impact of SLN status and PST on RNI decision-making. IMC RT is not frequently
coupled to SCN-RT. ECE is the main factor impacting decision-making for RNI. Only 58% of the centers take fibrotic changes in
the LNs into consideration in decision-making for RNI indications. Age is not considered as a limiting factor for prescribing RNI in
Europe.

Title: Low utilization of hypofractionated radiotherapy for the treatment of early-stage breast cancer in the US
Yvonne M Mowery1, Rachel A Greenup1, Kevin Houck1, Manisha Palta1, Janet K Horton1, Eun-Sil S Hwang1, Julie A Sosa1 and
Rachel C Blitzblau1. 1Duke University Medical Center, Durham, NC.
Body: Background: Large randomized controlled trials have shown that hypofractionated whole breast irradiation (HF-WBI) is
not inferior to or more toxic than conventionally fractionated whole breast irradiation (CF-WBI) for the treatment of early-stage
breast cancer. Royal Marsden data were published in 2006 (10-yr), the Ontario trial was reported in 2002 (5-yr) and 2010 (10-yr),
and the UK START trials were published in 2006 (5-yr) and 2013 (10-yr). We utilized the National Cancer Data Base (NCDB) to
evaluate patterns of radiotherapy fractionation for early-stage, node-negative breast cancer in the U.S. We hypothesized that
HF-WBI use would increase over time in response to emerging data supporting its use in this population.
Methods: We conducted a retrospective, population-based cohort study of women >18 years diagnosed with T1-2N0 invasive
breast carcinoma and treated with breast-conserving surgery between 2004 and 2011. Radiotherapy was categorized as
accelerated partial breast irradiation (APBI; 38-40 Gy/1-10 fractions), HF-WBI (40-56 Gy/15-24 fractions) or CF-WBI (50-66
Gy/25-40 fractions). Patients treated with alternate fractionation were excluded. Patterns of breast radiotherapy fractionation were
compared using the chi-square test. Multivariable logistic regression was performed for patients diagnosed in 2011, the year with
the highest levels of HF-WBI utilization.
Results: 217,789 patients in the NCDB met inclusion criteria. HF-WBI use increased over time, rising from 2.1% among eligible
patients in 2004 to 15.1% in 2011, while APBI use remained low at <2%. Utilization of HF-WBI was significantly higher in
academic centers than comprehensive community cancer centers and community-based cancer centers (p <0.0001). This pattern
persisted when controlling for T-stage, hormone receptor/HER2-status, diagnosis year, and patient age >30 years. Table 1 shows
frequency of HF-WBI use over time by center type. On multivariate analysis of patients diagnosed in 2011, the following factors
were associated with higher use of HF-WBI: treatment at an academic center, older patient age, hormone receptor positivity, pT1
tumor size, and rural residence (Table 2).
Conclusions: Utilization of HF-WBI in the US is rising, but remains low overall despite level I evidence showing its non-inferiority
to CF-WBI. Given the advantages of HF-WBI in terms of patient convenience and potential healthcare system costs, further
research is indicated to explore disparities in HF-WBI utilization in the US and to guide education of breast cancer providers.
Table 1: % HF-WBI use
Community Cancer Center
Comprehensive Community Cancer Center
Academic Center
2004
0.9%
1.7%
3.8%
2005
1.6%
2.1%
3.8%
2006
1.2%
2.0%
4.6%
2007
1.5%
2.8%
7.1%
2008
2.9%
5.3%
10.8%
2009
3.9%
8.5%
14.7%
2010
5.8%
10.7%
16.4%
2011
9.6%
13.9%
20.5%
Table 2. Multivariate analysis, year 2011 (n=5568)

OR
95% CI
Academic Center vs. Community Cancer Center
3.06
2.32
4.02
Academic Center vs. Comprehensive Community Cancer Center
1.78
1.53
2.08
Patient age, 50-90 vs. 18-49
2.37
1.86
3.01
T2 vs. T1
0.54
0.46
0.63
HER2+ vs. Hormone Receptor +/HER2-
0.75
0.59
0.97
ER-/PR-/HER2- vs. Hormone Receptor +/HER2-
0.66
0.52
0.84
Rural vs. urban
2.68
1.69
4.24

Title: No Show

Title: Development of a photonumeric scale for acute radiation dermatitis in breast cancer patients
Dean Shumway1, Eleanor M Walker2, Nirav Kapadia3, Thy Thy Do1, Kent Griffith1, Mary Feng1, Bonnie DePalma2, Reshma Jagsi1,
Yolanda Helfrich1, Erin Gillespie4, Alexandria Miller1, Adam Liss1 and Lori J Pierce1. 1University of Michigan, Ann Arbor, MI; 2Henry
Ford Hospital, Detroit, MI; 3Dartmouth-Hitchcock Medical Center, Lebanon, NH and 4University of California, San Diego, La Jolla,
CA.
Body: Purpose
Scales for rating acute radiation dermatitis (ARD) are inconsistent and have not been validated despite decades of clinical use,
making ARD difficult to report reliably. We sought to design a photonumeric scale to consistently describe ARD in breast cancer
patients undergoing radiation (RT).
Methods
Patients undergoing RT for breast cancer were enrolled on a prospective study that included photographs and reporting of
physician-rated erythema, hyperpigmentation, and CTCAE toxicity score at baseline and 2, 4, and 6 weeks after initiating RT.
Erythema and hyperpigmentation were also quantified using a hand-held colorimetric device. Photographs were taken using a
standardized protocol that included 3 views to fully assess the breast/chest wall, axilla, and inframammary fold. 209 photographs
from 35 patients with white skin (Fitzpatrick skin types I-IV) and 369 photographs from 50 patients with skin of color (Fitzpatrick
skin types V-VI) were clustered according to the apparent severity of ARD. Due to the prevalence of hyperpigmentation that
obscured erythema in patients with skin of color, separate images were used to illustrate ARD in this population. Two
photonumeric scales (for white skin and skin of color) were developed via an iterative process until group consensus was
achieved. Four raters with experience in the evaluation of ARD in breast cancer patients used the photonumeric scale to
independently score the entire collection of photographs, sequenced in random order. Intra- and inter-rater agreements were
assessed using weighted kappa scores.
Results
Of the 35 patients with white skin, 20% experienced severe erythema, and 40% experienced dry or moist desquamation. Of the
50 patients with skin of color, 34% experience severe hyperpigmentation, and 48% experienced dry or moist desquamation.
Using the photonumeric scales, we observed high intra-rater agreement for independent ratings of erythema or
hyperpigmentation (70 to 89% agreement fraction, kappa 0.55 to 0.81) and desquamation (79 to 87% agreement fraction, kappa
0.52 to 0.64). Similarly, we observed moderate to high inter-rater agreement for independent ratings of erythema or
hyperpigmentation (61 to 76% agreement fraction, kappa 0.40 to 0.62) and desquamation (69 to 84% agreement fraction, kappa
0.36 to 0.58). Quantitative measurements of erythema in white patients using colorimetry correlated strongly with photonumeric
grade (correlation coefficient 0.76, p<0.001), as did physician-rated erythema at the point-of-care (p<0.001). Fitzpatrick score was
not significantly associated with maximum photonumeric erythema grade (p = 0.14).
Conclusions
We report a new photonumeric scale with high intra- and inter-rater reliability for acute radiation dermatitis in breast cancer
patients. To our knowledge, this is the first rigorously evaluated scale that is applicable to patients across the spectrum of skin
pigmentation, including white skin and skin of color. The photonumeric scale will facilitate consistent reporting of acute radiation
dermatitis in research and clinical settings using a simple, standardized instrument. Future work will include prospective real-time
clinical validation with multiple raters and correlation with patient-reported outcomes.
Funded by a Munn Idea Grant (G011480).

Title: Short term toxicity of intra-operative radiotherapy for patients with breast cancer treated at a single center
Ayelet Shai1, Maoz Zur1, Michelle Leviov1, Arie Bitterman2, Shiloni Eitan3 and Mariana Steiner1. 1Oncology, Lin and Carmel
Medical Centers, Haifa, Israel; 2Carmel Medical Center, Haifa, Israel and 3Carmel Medical Center, Haifa, Israel.
Body: Introduction:
Intra-operative radiotherapy (IORT) is increasingly being used for the treatment of low risk breast cancer. In the randomized
TARGIT trial, late toxicity was not different between patients treated with IORT and external beam radiotherapy. Early toxicity of
IORT has not been fully characterized.
Methods:
IORT with one dose of 20 Gy using the Intrabeam device has been offered as an alternative to standard treatment as part of a
single center prospective single arm trial. Patients over 60 years with clinically node negative, < 2 cm invasive duct carcinomas
were eligible. Patients aged 50-60 years, patients with tumors 2-3.5 cm and patients with invasive lobular carcinomas were
treated with IORT if they were not candidates for standard treatment.
For the present study information regarding complications occurring within the first year after surgery was analyzed. Patients
treated between 2006 and 2012 were included in this analysis.
Results:
393 patients were treated from 2006 2012. 4 were lost to follow up. Median age was 70 years (55-90) and median clinical tumor
size was 1.2 cm (5-30).
102 patients (26.2%) had a complication. Infections were most frequent. Misdiagnosis of radiation dermatitis as infection in some
cases cannot be excluded. Clinically meaningful seromas occurred in 10.2% of patient and 8.2% had a wound dehiscence,
possibly related to rupture of pre-existing seromas.
Table 1
Complication type
N (%)
N with grade 3 (%)
Infection
42 (11)
3 (3.3)
Seroma
40 (10.2)
Wound dehiscence
33 (8.5)
2 (0.5)
Bleeding
11 (2.8)
3 (0.8)
Skin necrosis
2 (0.5)
N with grade 4 (%)
median time to recovery (range)

0.62 (0.1 - 3.8)
13.2 (0.3-50)
5.8 (0.36-15.3)
2 (0.5)
3.88 (3.45-4)
frequency, severity and time to recovery of acute complications of IORT.

18 patients (4.6%) had a grade 3 complication. 2 patients (0.5%) had grade 4 radiation induced skin necrosis of small size and
both resolved within 4 months. All other complications resolved completely (table 1). Risk factors for complications in this cohort
will also be presented.
Conclusions:
Intra-operative radiotherapy was associated with a relatively high rate of clinically meaningful seromas and wound dehiscence
and rare small size skin necrosis. Most complications were mild moderate. 5% of patients experienced grade 3 or 4
complications, and all resolved completely. Further study of the factors predisposing to severe complications is warranted.


Title: No Show

Title: The utilization of hypofractionated breast radiotherapy in the United States between 2003 and 2011
Dezheng Huo1, Yasmin Hasan1 and Katharine Yao2. 1University of Chicago, Chicago, IL and 2NorthShore University
HealthSystem, Evenston, IL.
Body: Background: Recently, several randomized trials have found that accelerated hypofractionated whole breast irradiation
(HF-WBI) had equivalent local recurrence rates and disease-free survival, compared to conventional fractionated whole breast
irradiation (CF-WBI). It is unknown if the pattern of care regarding HF-WBI has changed in the United States after the publication
of these trials, given its administrative advantage.
Methods: Using data from the National Cancer Database, this study included 559,362 non-metastasic female breast cancer
patients who received post-lumpectomy external beam radiotherapy to their breast or regional nodes. We examined the trend of
HF-WBI use in the U.S. and factors related to its use using logistic regression models.
Results: Of patients receiving external beam radiotherapy, about 5% underwent HF-WBI. The most common HF-WBI dose
schedule is 42.4-42.6 Gy in 16 fractions without or with boost radiation. There was a 20-fold increased trend in the use of
HF-WBI, increasing from about 0.6% in 2003 to 13% in 2011 (p<0.0001). Among patients with pT1-2N0-1 disease, a subgroup for
whom previous clinical trials of HF-WBI are mainly applicable, HF-WBI use increased from 0.6% in 2003 to 14% in 2011.
Although not included in previous clinical trials, patients with ductal carcinoma in situ also experienced an increased use of
HF-WBI, from 0.5% in 2003 to 12% in 2011. We found age was strongly correlated with HF-WBI use; the older a patient, the more
likely she received HF-WBI. African Americans were less likely to use HF-WBI. Patients with low grade were more likely to
receive HF-WBI than those with high grade and patients with negative nodes are more likely to received HF-WBI. Large volume,
academic cancer centers are more likely to adopt HF-WBI than community cancer programs. Interestingly, patients who live at
least 50 miles away from a cancer center are about 58% more likely to receive HF-WBI than patients living within 50 miles of a
cancer center.
Conclusions: The utilization of hypofractionated radiotherapy significantly increased from 2003 to 2011 in the United States, which
is presumably influenced by publication of several randomized trials. In 2011, HF-WBI accounted for more than 10% of all
external beam radiotherapy for early breast cancer patients. Patient and facility factors impacted the patterns of HF-WBI use.

Title: Sustained acceptable cosmetic outcomes and local control following accelerated partial breast irradiation using CT-guided
IMRT in the prone position: Results from a phase I/II feasibility study
Carmen Bergom1, Phillip Prior1, Kristofer Kainz1, Natalya V Morrow1, Ergun E Ahunbay1, Alonzo Walker1, X Allen Li1, Tracy Kelly1,
Adam D Currey1 and Julia White2. 1Medical College of Wisconsin, Milwaukee, WI and 2Ohio State University Comprehensive
Cancer Center, Columbus, OH.
Body: Objective/Purpose
External beam accelerated partial breast irradiation (EB-aPBI) can have potential challenges in daily reproducibility, although it
has broader potential use than aPBI using brachytherapy. Image-guide radiotherapy (IGRT) can improve daily reproducibility and
allow smaller treatment margins. Our institution utilized IG-IMRT to administer EB-aPBI in the prone position in a Phase I/II study
to increase dose homogeneity, conformality, normal tissue avoidance, and reliable targeting. Our preliminary results and toxicity
were promising. Here we report final physician- and patient-reported cosmetic outcomes from this prospective trial.
Women with node-negative invasive breast cancer or DCIS, tumors less than 3.0 cm, a negative sentinel lymph node biopsy, and
surgical clips demarcating the lumpectomy cavity underwent prone EB-aPBI using IG-IMRT on an IRB-approved phase I/II study.
The lumpectomy PTV represented a 2.0 cm lumpectomy cavity expansion. 38.5 Gy was delivered in 10 fractions over 5 days,
such that 95% of the prescribed dose covered greater than 99% of the PTV. Dose constraints for the whole breast, lungs and
heart were met.
Results
Twenty patients were enrolled, with a median patient age of 61.5 and a mean tumor size of 1.0 cm. 35% of patients had DCIS. At
a median follow-up of 18.9 months, 40% and 10% of patients had G1 and G2 fibrosis, respectively, and 95% of patients had good
to excellent physician-assessed cosmesis. At a median follow-up of 60.0 months (range 54-79 months), physician-assessed
cosmetic outcome was good to excellent in 80%, with 30% and 20% of patients experiencing G1 and G2 fibrosis. Patient-reported
outcomes at one year yielded 90% of patients with good to excellent cosmetic outcomes. At 3 years, 75% of patients reported
good to excellent cosmesis. Eighty-eight percent of patients were completely satisfied with the treatment and results, and 94% of
patients would choose aPBI again. With one local recurrence, the actuarial five year rate of local control was 95%.
Conclusions
These data demonstrate that EB-aPBI in the prone position using IG-IMRT continues to yield acceptable cosmetic outcomes at
longer term follow-up, and a very high percentage of patients would choose this treatment again.
(Supported by Komen Grant: BCTR0504070).

Title: Evaluation of separation techniques and dosimetry of the inframammary fold during whole breast radiation therapy
Guy Jones1, Peter Guion1, Jason Cheng1 and Aparna H Kesarwala1. 1National Cancer Institute, Bethesda, MD.
Body: Purpose/Objectives: The majority of patients receiving radiation therapy will experience some form of skin reaction,
ranging from localized erythema to moist desquamation. Radiation dermatitis of the inframammary fold (IMF), the most common
acute side effect of supine whole breast radiation therapy (WBRT), can be mitigated by separating the breast from the chest wall.
The purpose of this analysis was an objective evaluation and comparison of practical and inexpensive separation techniques.
Materials/Methods: A breast phantom was created on a human torso phantom using tissue-equivalent bolus material. Four
separation techniques were evaluated: 1) taping the inferior aspect of the breast to the shoulder or placing 2) 6" x 2" rolled bubble
wrap, 3) a 6" x 1" x 1" styrofoam block, or a 4) 50 mL 2" diameter plastic conical tube in the IMF. For the control setup, the breast
was not taped and nothing was placed in the IMF. Written and pictorial simulation instructions recorded by one radiation therapist
were independently used by additional therapists to reproduce each setup, with medial and inferior measurements taken for
comparison. Dosimetry of identical tangent 100 cGy WBRT fields was evaluated with thermoluminescent dosimeters (TLDs)
placed at each of four locations: the ipsilateral IMF, the ipsilateral chest wall 5 cm inferior to the IMF, and 5 cm superior to the
nipple of both the ipsilateral and contralateral breasts at the 12:00 position. Measurements and TLD data are reported as mean
standard error of the mean of 3 separate TLD readings. One-way ANOVA and posthoc analyses were used to determine
significance between groups with for significance set at p 0.05.
Results: Compared to simulation, the average differences measured medially and inferiorly among the conical tube (10 6 mm
medial; 0.6 0.2 mm inferior), styrofoam block (13 8 mm medial; 1.2 0.2 mm inferior), and bubble wrap (18 4 mm medial;
0.6 0.1 mm inferior) setups were not significant. The taping technique (0.2 0.1 mm) was more reproducible inferiorly than the
styrofoam block setup (p <0.01), but no different from the conical tube and bubble wrap setups. TLD readings demonstrated that
the IMF received less of the prescribed dose with the taping (108 cGy 1.4%), styrofoam block (109 cGy 0.9%), or bubble wrap
(110 cGy 0.9%) techniques compared to either the control (120 cGy 1.1%) or conical tube (121 cGy 0.4%) setups (each p
<0.01). TLD readings from the three other locations were not significantly different between any of the setups.
Conclusions: The taping, styrofoam block, and bubble wrap setups were most effective in sparing dose to the IMF. The only
difference in reproducibility was between the inferior aspects of the taping and styrofoam block setups, but taping had the
disadvantage of being more invasive than the other setups. While this unique analysis did not reveal a single superior solution,
the taping, styrofoam block, and bubble wrap separation techniques should be considered by radiation oncologists seeking
practical, inexpensive, and easily-implemented solutions to mitigate IMF dermatitis in WBRT patients.

Title: Effects of statin use on skin toxicity during breast radiation therapy
Scott M Lieberman1, Elizabeth Stone1, Christopher Chen1, Marlow Hernandez1 and Leah Roberts2. 1Cleveland Clinic Florida,
Weston, FL and 2Charles E Schmidt College of Medicine at FAU, Boca Raton, FL.
Body: Background: Recent San Antonio Breast Cancer Symposium abstracts have suggested that statin use confers a
significant benefit in breast cancer risk reduction and specifically in the subset of inflammatory breast cancer. Additional studies
evaluating the combination of statin use with various breast cancer treatments are currently on-going. The primary objective of
this study was to determine the effects of statin use on acute skin toxicity during breast radiation therapy (RT). Methods: We
performed a retrospective chart review of 325 patients diagnosed with breast cancer who underwent RT between 2004 and 2012.
Of these, 227 patients were considered evaluable. Patients were excluded if the course of RT was fewer than 16 treatments.
Patients were also excluded if follow-up at the end of RT was not obtainable. Of the evaluable patients, 46 were on statins at time
of RT, while 181 were not. Patients were evaluated at two time points - midway through RT and at completion of RT. Skin
reactions were measured using the RTOG grading system for acute radiation morbidity. Grade 1 reactions were erythema and
slight skin atrophy. Grade 2 indicated follicular, faint or dull erythema/ epilation/dry desquamation/ decreased sweating. Grade 3
was tender or bright erythema, patchy moist desquamation/moderate edema. Lastly, Grade 4 was ulceration, hemorrhage,
necrosis. Results: Our results showed a statistically significant increase in acute radiation skin toxicity in patients on statins at the
time of radiation therapy. By the midway point of RT, 67% of the statin group had developed Grade 1-3 skin toxicity vs. only 42%
of the control group. Odds ratio midway through RT was 2.75. By the completion of RT, 94% of the statin group compared with
67% of the control group had Grade 1-3 skin toxicity (p=0.005). Odds Ratio at completion of RT was 7.29. The statin group also
progressed to more advanced stages of toxicity more rapidly than our control group. Grade 1 skin reactions were seen in 62% of
our statin group vs. 41% of our control group midway through treatment (p=0.044). Grade 2-3 reactions were seen in almost 5%
of patients in the statin group vs. 1% in the non-statin group midway through treatment. At the end of treatment, 48% of patients
on statins had developed Grade 2-3 reactions when compared with 29% in the non-statin group (p=0.0199). These results still
remained significant when controlled for chemotherapy use. Conclusion: Our study revealed that patients on statins developed
acute skin toxicity at higher rates than our control group. These results were statistically significant at both the midway point as
well as at the completion of RT. Study limitations include the small sample size and the retrospective nature of the study. Our
data also raises the question of what effect statins may have on local recurrence rates as well as the potential effects of
radiosensitization in various breast cancer cell lines. These results are hypothesis-generating and require larger retrospective
analyses or prospective trials to verify this effect.

Title: Prospective study of proton radiotherapy for treatment of regional lymphatics in breast cancer
Julie A Bradley1, Roi Dagan1, Meng Wei Ho1, Christopher G Morris1, Zuofeng Li1 and Nancy P Mendenhall1. 1University of Florida
Proton Therapy Institute, Jacksonville, FL.
Body: Purpose: To compare dosimetric endpoints between proton therapy (PT) and conventional radiation and to determine the
feasibility of PT for regional nodal irradiation (RNI) in women with breast cancer.
Methods: From May 2012 to February 2014, 18 women (stage IIA-IIIB) prospectively enrolled on a pilot study. Median age was
51.8 years (range, 42-73), with equal division between breast-conserving therapy (BCT) and mastectomy and right and left-sided
cancers. Treatment targets (CTVs for breast/chest wall, supraclavicular, axillary, internal mammary nodes [IMNs]) and organs at
risk were delineated on CT scans, and PT and conventional plans were developed. PT alone was used for 10 patients (9
post-mastectomy, 1 after BCT) and combined proton-photon in 8 (all BCT). Acute toxicity was prospectively recorded using
CTCAE v4.0. A Wilcoxon signed-rank sum test compared the dose-volume parameters.
Results: Median followup was 10.6 months (range, 1.1-19.1). For all patients, the PT plan better met the dosimetric goals and
was used for treatment. Breast/chest wall coverage was adequate (V47.5=96.6% for both plans). PT improved coverage of level
II axilla (median D95, 48.9Gy [range, 44.8-51.4Gy] with PT vs 45.4Gy [range, 39.8-51Gy] with conventional; p=0.0005). Adequate
coverage of IMNs was consistently achieved with PT (median D95, 48.9Gy with a range of 44.8-51.4Gy) but not with conventional
(median D95, 45.4Gy with range of 39.8-51Gy; p=0.0005); PT reduced heart dose in all patients. Median ipsilateral lung V5
measured 35.3% for PT compared to 60.5% for conventional (p<0.0001). Ipsilateral lung V20 improved with PT (median, 21.6%
vs 35.5%; p<0.0001).
Table 1: Cardiac dose in left-sided patients (comparison of medians of the cohort; range in parentheses)
Heart V5
Mean Heart
Dose
Proton Plan
2.7%
(0-12%)
0.6Gy (0-3.2Gy) 0.5% (0-15.6%)
Conventional
Plan
34%
(7-60%)
5.9Gy (2-9.1Gy)
p-value
0.004
0.004
Mean Ventricle
Dose
Ventricle V5
31.4%
(8.7-79.8%)
Maximum LAD
dose
Mean LAD dose
0.1Gy (0-4Gy)
30.5 (13.4-42.6Gy)
1.7Gy
(0.5-28.5Gy)
5.4Gy (2.4-11Gy)
44.6Gy
(35.1-54.7Gy)
27Gy
(16.8-41.9Gy)
0.004
0.004
0.04
0.02
Table 2. Cardiac dose in right-sided patients (comparison of medians of the cohort; range in parentheses)
Heart V5
Mean Heart Dose
Ventricle V5
Mean Ventricle Dose
Proton Plan
0.3% (0-2.9%)
0.5Gy (0-0.8Gy)
3.3% (0-1%)
0.2Gy (0-0.5Gy)
Conventional Plan
13.2% (0.4-24.7%)
2.9Gy (1-5.1Gy)
3.3% (0-7.1%)
1.2Gy (0.8-1.7Gy)
p-value
0.004
0.004
0.008
0.004
Grade 3 dermatitis developed in 4 patients (22%), which was the only grade 3 toxicity. All patients developed grade 2 dermatitis;
other acute grade 2 toxicities included fatigue (n= 6) and esophagitis (n=5). No grade 4+ toxicities developed. Dermatitis resolved
by 1 month after PT for all but 1 patient who developed cellulitis (grade 2).
Conclusion: PT for RNI after mastectomy or BCT significantly improves cardiac dose especially for left-sided patients and lung
V5 and V20 in all patients without excessive acute toxicity. PT simultaneously improves target coverage for the IMN and level II
axilla, which may positively impact long-term survival in breast cancer patients.

Title: Safety of eribulin mesylate and concomitant palliative radiotherapy for metastatic breast cancer: A single-center experience
Icro Meattini1, Vieri Scotti1, Carla De Luca Cardillo1, Beatrice Detti1, Vanessa Di Cataldo1 and Lorenzo Livi1.
1
Radiotherapy-Oncology Unit, Florence University Hospital, Florence, Italy.
Body: Background. Eribulin mesylate is a recently approved new therapeutic option for patients with metastatic breast cancer
(BC). It is a structurally simplified synthetic analogue of halichondrin B that inhibits the growth phase of microtubule dynamics and
sequesters tubulin into non-productive aggregates, inhibiting microtubule polymerization, and inducing irreversible mitotic block at
G2-M phases and apoptosis.
Three phase II trials of eribulin in chemotherapy pretreated advanced BC patients were completed. In all these studies, eribulin
showed a manageable tolerability profile, with most of the common drug-related adverse events being neutropenia, fatigue,
alopecia, nausea, and anemia, according to the phase I trials findings. Eribulin was also associated with an overall low incidence
of peripheral neuropathy.
When new chemotherapy options become available, one of the most important questions regarding the sequence of treatments is
to address the safety of concomitant radiation treatments.
Recently has been reported a clinical case treated with eribulin mesylate and whole-brain radiotherapy (RT) in a heavily
pretreated patient with multiple visceral and bone metastases from triple negative breast cancer.
In our knowledge this is the first analysis focusing on the safety of eribulin in patients concomitantly treated with palliative RT.
Methods. All patients where heavily pretreated for metastatic breast cancer (4 previous chemotherapy lines), including
anthracyclines and taxanes-based regimens. Patients received eribulin mesylate (1.4 mg/m2 i. v., on days 1 and 8, of a 21-day
cycle).
Tumor response was assessed using the Response Evaluation Criteria in Solid Tumors (RECIST), every 8 weeks, or sooner if
disease progression was suspected.
Safety was assessed according to National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE), version
4.0. Measures of bone-pain scores and analgesic consumption were evaluated at each visit, using a patient-rated scoring system.
Results. Concerning 25 consecutive treated patients, median performance status at baseline was 2 (range 1-3). A median of 4
cycles of eribulin (range 2-10) were administered. All patients received palliative bone RT during eribulin treatment. No
suspension or delay in chemotherapy administration was necessary. 20/25 cases underwent spinal bone irradiation; 8 Gy in
single fraction was given in 13/25 patients, 12 cases received 20 Gy in 5 fractions.
The overall response rate was 24% and the clinical benefit rate was 40%. Toxicity was manageable and in line with main
published series; the most frequent grade 3 hematologic adverse events were neutropenia (56%), and anemia (20%).
Bone pain scores fell within a few weeks and remained below baseline value throughout the analysis. The mean pain score was 2
at baseline and 0.7 at the end of observation period. The mean analgesic score was 1.8 at baseline and 1.6 at the end of
follow-up.
Conclusions. Our clinical practice experience supports the evidence that eribulin has clinical activity and is characterized by a
manageable safety profile, also when combined to palliative RT. Concomitant treatment is feasible also in patients heavily
pretreated, and with poor performance status.

Title: Comparison of cardiac dose between accelerated partial-breast irradiation and whole-breast irradiation in breast cancer
patients
Kazuhiko Sato1, Yoshio Mizuno1, Hiromi Fuchikami1, Naoko Takeda1, Takahiro Shimo2, Jun Kubota2, Yuko Inoue3, Hiroshi Seto4
and Masahiro Kato2. 1Tokyo-West Tokushukai Hospital, Akishima, Tokyo, Japan; 2Tokyo-West Tokushukai Hospital, Akishima,
Tokyo, Japan; 3Inoue Ladies Clinic, Tachikawa, Tokyo, Japan and 4Seto Hospital, Tokorozawa, Saitama, Japan.
Body: [Background] Breast-conserving surgery (BCS) followed by whole-breast irradiation (WBI) has now become the standard
treatment for early-stage breast cancer. However, WBI is associated with an increased risk of coronary events, especially in
patients with preexisting cardiac risk factors. In radiotherapy (RT), the highest dose is likely delivered to the left anterior
descending artery (LAD), which is the typical site for ischemic heart disease. We initiated a prospective, observational study on
accelerated partial-breast irradiation (APBI) using multicatheter brachytherapy after BCS. In this study, we compared the radiation
dose to LAD between patients receiving APBI and those receiving WBI.
[Methods] The study participants included a cohort selected from consecutive patients who underwent BCS followed by RT since
November 2007. In the WBI group, patients received 50 Gy in fractions of 2 Gy to the entire breast. APBI was initiated on the day
of primary surgery in the form of multicatheter brachytherapy, at a dose of 32 Gy in 8 fractions. The planned target volume was
defined as the estimated tumor volume plus a 20-mm margin. Dose distribution analysis was performed on the basis of
postoperative CT using dosevolume histograms. LAD was outlined from its origin to each visible end using the planning CT
images. First, the mean and maximal total doses to LAD were calculated. Second, the radiotherapeutic biologically effective dose
of APBI was adjusted to that of WBI for comparisons between the two different RT schedules.
[Results] Of the 359 consecutive patients who underwent BCS followed by RT, we retrospectively reviewed 182 patients for
radiation dose to LAD. The 82 patients receiving WBI were randomly selected; 42 patients had right breast cancer and 40 had left
breast cancer. We selected 100 consecutive APBI patients with left breast cancer treated between September 2009 and
December 2013 because the LAD dose is considered to be virtually zero in right breast cancer patients. In the WBI patients, the
mean and maximal total LAD dose were significantly higher in left breast cancer patients (2.1 and 8.2 1.2 Gy, respectively)
than in right breast cancer patients (0.4 0.02 and 0.6 0.03 Gy, respectively; p < 0.0001). Among the APBI patients, the total
LAD doses were influenced by tumor location. The mean and maximal total LAD doses were significantly higher in patients with
inner quadrants or central tumors (2.5 0.2 and 4.4 2.5 Gy, respectively) than in those with outer quadrant tumors (1.0 0.1
and 2.1 0.3 Gy, respectively; p < 0.0001). After adjustment for the total LAD dose, the mean and maximal total LAD doses were
significantly decreased in APBI patients with outer quadrant tumors (1.1 0.2 and 2.4 0.4 Gy, respectively; p < 0.0001), but not
in those with central and inner quadrant tumors (2.9 0.3 and 5.4 0.6 Gy, respectively).
[Conclusions] Our results show that APBI may decrease the risk of coronary artery disease, especially in patients with outer
quadrant tumors in the left breast. Although APBI should be carefully interpreted until mature phase-III data are available, the risk
of ipsilateral breast tumor recurrences and LAD dose must be considered together while administering RT after BCS.

Title: Clinical outcomes according to elective nodal irradiation and molecular subtypes in high risk N1 breast cancer patients
Jeong Il Yu1, Won Park1, Doo Ho Choi1, Seung Jae Huh1, Seok Jin Nam1, Seok Won Kim1, Jeong Eon Lee1, Won Ho Kil1,
Young-Hyuck Im1, Jin Seok Ahn1, Yeon Hee Park1 and Eun Yoon Cho1. 1Samsung Medical Center, Seoul, Korea.
Body: Objectives: To evaluate the clinical outcomes according to the application of an elective nodal irradiation (ENI) and
molecular subtypes in high-risk pathologic N1 (pN1) breast cancer.
Methods: We performed a retrospective comparison study with high-risk pN1 patients who received curative resection followed
by chemotherapy at Samsung Medical Center from January 2009 to June 2011. High risk was defined as having more than two of
the following risk factors: 2 axillary lymph nodes (ALNs) or level II ALN metastasis, lymphovascular invasion and extracapsular
extension. We compared clinical outcomes according to the application of ENI.
Results: Among 278 patients, 159 patients received ENI while 119 did not and their characteristics were displayed.
Table 1. Characteristics of 278 pathologic N1 breast cancer patients according to the usage of elective nodal irradiation (ENI).
Variables
Age (years)
Surgery (%)
Pathology (%)
T stage (%)
Histologic grade (%)
Nuclear grade (%)
Dissected ALN number
Estrogen receptor (%)
Progesterone receptor (%)
HER-2 amplification (%)
Hormonal therapy (%)
Chemotherapy (%)
Trastuzumab (%)
no ENI (n=119)
ENI (n=159)
P-value
Median
48
48
0.50
Range
27-80
27-71
MRM
90 (75.6)
2 (1.3)
BCS
29 (24.4)
157 (98.7)
IDC
116 (97.5)
153 (96.2)
Other
3 (2.5)
6 (3.8)
T1
56 (47.1)
81 (50.9)
T2
63 (53.9)
78 (49.1)
1-2
84 (70.6)
102 (64.2)
35 (29.4)
57 (35.8)
1-2
79 (66.4)
90 (56.6)
40 (33.6)
69 (43.4)
Median
19
17
Range
2-61
5-48
Positive
92 (77.3)
124 (78.0)
Negative
27 (22.7)
35 (22.0)
Positive
87 (73.1)
123 (77.4)
Negative
32 (26.9)
36 (22.6)
Positive
33 (27.7)
24 (15.1)
Negative
86 (72.3)
135 (84.9)
No
29 (24.4)
31 (21.9)
Yes
90 (75.6)
128 (78.1)
No
13 (10.9)
5 (3.1)
Yes
106 (89.1)
154 (96.9)
No
5 (15.2)
1 (4.2)
Yes
28 (84.8)
23 (95.8)
<0.001
0.74
0.55
0.30
0.11
0.17
1.00
0.48
0.01
0.38
0.01
0.19
During follow-up, 21 patients (6 in ENI and 15 in no ENI) had recurrence, and loco-regional recurrence developed in 8 patients, 6
of whom had not received ENI.
Table 2. Patterns of recurrence according to ENI and molecular subtype
No ENI
ENI
Site
Recurrence (%)
Recurrence (%)
P-value
Loco-regional-n
119
6 (5.0)
159
2 (1.3)
0.08
Luminal A
36
0 (0.0)
41
0 (0.0)
Luminal B
57
4 (7.0)
86
0 (0.0)
HER-2 enriched
12
0 (0.0)
1 (11.1)
Triple negative
14
2 (14.3)
23
1 (4.3)
119
10 (6.5)
159
5 (3.1)
Luminal A
36
2 (5.6)
41
0 (0.0)
Luminal B
57
6 (10.5)
86
2 (2.3)
HER-2 enriched
12
0 (0.0)
0 (0.0)
Triple negative
14
2 (14.3)
23
3 (13.0)
119
15 (12.6)
159
6 (3.8)
Luminal A
36
2 (5.6)
41
0 (0.0)
Luminal B
57
9 (15.8)
86
2 (2.3)
HER-2 enriched
12
0 (0.0)
1 (11.1)
Triple negative
14
4 (28.6)
23
3 (13.0)
Distnat-n
Total-n
0.06
0.01
In both univariate and multivariate analysis, ENI, adjuvant chemotherapy and endocrine therapy were the significant prognostic
factors in recurrence-free survival (RFS). ENI showed higher RFS than no ENI when they were analyzed according to molecular
subtypes, except HER-2 enriched. RFS was 96.9% at 3-years in luminal A patients with no ENI.
Conclusions: The application of ENI may improve RFS in patients with high-risk pN1 breast cancer, so ENI should be
considered in those patients. Close observation without ENI might be an option in patients with luminal A who had received
optimal systemic managements.

Title: Skin toxicities of obese African American breast cancer patients treated with hypofractionated radiation therapy
Arun G Paul1, Amy Collins2, Gregory Dyson2 and Neha Amin3. 1Detroit Medical Center, Detroit, MI; 2Barbara Ann Karmanos
Cancer Center, Detroit, MI and 3Wayne State University, Detroit, MI.
Body: Objective: Hypofractionated radiation therapy (HFRT) for early stage breast cancer is an established treatment option with
equivalent cancer outcomes and better cosmetic results than standard fractionation. There is limited published information about
expected skin changes of African-American breast cancer patients (pts) undergoing HFRT. While HFRT is not recommended for
women with large breasts, the use of prone position may allow for homogeneous HFRT plans. We prospectively monitored and
reported on skin changes in African-American pts who received HFRT for their breast cancer.
Methods: A retrospective analysis at a single institution from 12/2012 to 08/2013 was performed to identify early stage breast
cancer pts who underwent breast conservation surgery and received adjuvant whole-breast HFRT. An assessment form had
been created to prospectively document weekly changes in radiation dermatitis (CTCAE V.4 Grade 0-4) and hyperpigmentation
(none, faint, moderate, severe). Photographs of the treated area were collected before and at the end of treatment. Treatment
planning guidelines were per RTOG 1005.
Results: There were 15 African-American pts with Tis-T2N0M0 breast cancer who were treated with HFRT to a dose of 4256 cGy
in 16 fractions (266 cGy per fraction) followed by a lumpectomy cavity boost of 1000-1250 cGy (250cGy per fraction). There were
12 (80%) pts with right-breast cancer and 6 (40%) who were treated in the prone position. The median age was 61 (36-70). The
median body mass index (BMI) for the pts treated in the prone position (42.2 [36-54]) was greater than the median BMI for pts
treated in the supine position (29.7 [26-43]). The median breast volume of the prone pts (2335cc [2163-3369]) was more than
twice the median volume of supine pts (920cc [231-1459]). The median separation distance for prone and supine pts were
25.1cm (17.5-31.2) and 22.5cm (17.5- 31.8), respectively. Radiation dermatitis: None (6%), Grade 1 (60%), Grade 2 (34%),
Grade 3 (none). Only three pts had desquamation:1 pt had dry, and two pts had both dry and minimal moist desquamation in the
infra-mammary fold. Hyperpigmentation: none (6.5%), faint (40%), moderate (47%), severe (6.5%). The areas of severe
hyperpigmentation correlated to areas of desquamation. The median DMax, mean heart dose, and ipsilateral lung V20Gy were
107.2% (105.2 109.9), 60 cGy (29 544), and 8.25% (0.0 23), respectively.
Conclusion: African-American pts treated with HFRT experience minor skin toxicities with mostly Grade 1 dermatitis and
moderate hyperpigmentation changes. Obese African-American pts can safely be treated with HFRT if they can be placed in the
prone position. Some pts with a separation of > 25cm can be safely and effectively treated with HFRT.

Title: Phase II trial of hypofractionated post mastectomy radiation
Matthew M Poppe1, Bruce G Haffty2 and Atif J Khan2. 1Huntsman Cancer Hospital, University of Utah, Salt Lake City, UT and
2
Rutgers Cancer Institute of New Jersey & Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ.
Body: Purpose: The efficacy of irradiating the chest wall and draining lymph nodes after mastectomy in improving local regional
control has been firmly established by multiple trials comparing mastectomy alone to mastectomy with postoperative radiation.
The currently accepted standard of care involves 25 fractions of chest wall and nodal radiation plus an optional 5 fraction scar
boost. Given recent publications with intact breast radiation demonstrating equivalence with shorter course radiation treatment,
we sought to establish a short course of radiation for women post mastectomy that would be equivalent to conventional radiation
in terms of local control and complications.
Methods: Starting in 2011, Stage II and III, unilateral breast cancer patients were prospectively consented to participate in our
study at Rutgers Cancer Institute of New Jersey and the University of Utah. We designed a phase II non-inferiority trial, with plans
to enroll 65 patients, allowing for a 18% loco-regional recurrence rate if the true rate is 10%. Patients were allowed to receive
chemotherapy before or after adjuvant radiation with radiation starting 14-63 days after surgery or adjuvant chemotherapy.
Patients underwent 3D simulation and were treated with 6MV photon radiation using 3D conformal tangents or intensity
modulation. Regional nodal radiation was included when clinically appropriate with inclusion of internal mammary nodes at the
treating providers discretion. The prescription dose was prescribed at 333cGy per day to the chest wall and regional nodes for 11
daily fractions followed by a recommend scar boost of 333cGy per day for 4 fractions. Chest wall PTV was keep within 90-115%
of the prescribed dose, the maximum brachial plexus dose was keep below 3919cGy (107%) and no portion of the heart could
receive >2Gy/day. The maximum dose of the boost field was kept to 120% of the boost prescription. Any breast reconstruction
technique was allowed, per institutional standard.
Results: 48 patients have thus far been enrolled with a median follow up interval of 17 months. Looking at the 43 patients with > 4
months of follow-up, we have only seen a 4.7% grade 3 or greater acute or late toxicity rate, with one grade 3 wound
complications, and one grade 3 fibrosis. There has been one chest wall recurrence, for a 2.3% local recurrence rate.
Conclusion: Early results would appear to show that hypofractionated post mastectomy radiation appears to be safe and feasible.
Acute toxicity is less than anticipated, however it is too early make any conclusions on efficacy or late effects, with only 48 out of
65 patients enrolled. We hope to use the results of this phase II trial to design a randomized prospective trial comparing
hypofractionated post mastectomy radiation to standard fraction.

Title: Effect of margin width on local recurrence in invasive lobular carcinoma treated with multimodality therapy
Yasuaki Sagara1, William T Barry2, Ines Vaz-Luis2, Fatih Aydogan1, Jane E Brock1, Eric P Winer2, Mehra Golshan1 and Otto
Metzger-Filho2. 1Brigham and Women's Hospital, Dana-Farber Cancer Institute, Boston, MA and 2Dana-Farber Cancer Institute,
Boston, MA.
Body: Background: Recent consensus guideline on margins for breast conserving surgery recommends the use of no ink on
tumor as the standard for an adequate margin. Current recommendation extends to invasive lobular carcinoma (ILC), however
the data in this subset is limited by small numbers. In the present analysis we sought to evaluate the influence of margin status on
outcomes in ILC and mixed tumors.
Methods: We performed retrospective cohort study and reviewed 809 eligible patients diagnosed with ILC (337 with pure ILC; 472
with mixed ILC) with Stage I III treated at Dana Farber/Brigham and Womens Cancer Center (DFBWCC) between May 1997
and Dec 2007. Clinico-pathologic data was extracted following the Clinical Research Information Systems (CRIS) Database
procedures and manually reviewed to confirm inclusion and details of margin status. Margin status was defined using the last
ASCO/ASTRO/SSA consensus guidelines criteria. Analysis results were considered to be statistically significant when the
two-tailed p-value was <0.05.
Results: Breast conservation was performed in 399 patients (49%). Margin status at the initial attempt for breast conservation was
defined as follows: 180 (45%) negative, 64 (16%) positive, 71 (18%) 1mm margin, and 84 (21%) close margins (> 1 and < 3
mm). Following initial lumpectomy, 102 (25%) patients underwent additional surgery (96 re-excisions and 6 mastectomies) and
residual invasive disease was found in 40 patients. Whole-breast radiation therapy was performed in 376 patients (96%). In
multivariate models adjusted for classic clinico-pathologic factors, tumor size (HR= 1.8 95% CI 1.0 to 3.3, p=0.05), multifocality
(HR= 2.0 95% CI 1.1 to 3.6, p= 0.02) and ILC subtype (HR= 2.0 95% CI 1.0 to 3.7, p=0.04) were correlated with positive margins,
while year of diagnosis, age and pre-surgical MRI findings were not statistically significant.
With 72 months median follow-up, 12 ipsilateral breast cancers (3.1%), 5 other locoregional (1.2%) and 15 distant (3.8%)
recurrences were observed after definitive breast conserving therapy. The incidence of locoregional recurrence (LRR) was 4.3%
and similar for ILC and mixed ILC (p=0.76). In univariate analysis positive surgical margin was associated with LRR (HR=5.1, p=
0.03) and disease-free survival (DFS) (HR=8.9, p .001), but due to limited number of cases and events this could not be
adjusted for other clinico-pathologic prognostic factors in a mulitvariate model. Close surgical margins, margins within 1mm and
multifocality were not associated with increased LRR or worse DFS. Re-excision did not impact on DFS for patients with close
margin (p= 0.57) and within 1 mm margin (p= 0.85). By contrast, significant improvement of DFS following re-excision was
observed in patients with positive margin (p= 0.01).
Conclusions: Following lumpectomy, local recurrence rates for ILC patients with close surgical margin and 1mm margin are low
and equivalent to those in patients with negative margins. This study supports the validity of using no ink on tumor as the
standard for an adequate margin for patients diagnosed with pure or mixed ILC treated with multimodality therapy.

Title: Multidisciplinary breast cancer care is associated with a higher rate of breast conservation in comparison with
non-multidisciplinary care
Caitlin L Gomez1, Pin-Chieh Wang1, Nicole A Dawson1, Robyn L Dvorak1, Nova Foster1, Anne Hoyt1, Sara A Hurvitz1, Amy
Kusske1, Charles Y Tseng1 and Susan A McCloskey1. 1University of California, Los Angeles, CA.
Body: Background
Mastectomy rates, both therapeutic and prophylactic, are on the rise in the United States. After recent implementation of a
multidisciplinary breast clinic for newly diagnosed women at our institution, we sought to examine the impact of multidisciplinary
care on surgical decision making.
Materials/Methods
Women with newly diagnosed breast cancer at our institution are referred to a multidisciplinary breast clinic where they are seen
by a team of breast specialists (MDC) or to an individual practitioner (non-MDC) for initial consultation. We retrospectively
analyzed rates of breast conserving surgery (BCS) and mastectomy among women with newly diagnosed breast cancer seen in
either setting. For mastectomy cases, we designated the mastectomy as clinically indicated vs not clinically indicated based on
National Comprehensive Cancer Network (NCCN) guidelines for breast conservation. T-test and chi-square were used to
examine the comparability between MDC and non-MDC cohorts. Logistic regression was used to evaluate the overall prevalence
of BCS among MDC and non-MDC cohorts. Stratification analysis was further conducted to examine BCS rates among women in
each cohort receiving neoadjuvant chemotherapy vs up front surgery.
Results
A total of 341 consecutive patients were analyzed, including 202 MDC and 139 non-MDC patients seen in initial consultation
between June 2012 and April 2014. The MDC and non-MDC cohorts were statistically equivalent in terms of age, tumor and
nodal stage, histology, biomarker status, receipt of neoadjuvant chemotherapy, and proportion with genetic mutations. In the
MDC cohort, 66% underwent BCS vs 42% in the non-MDC cohort (p<0.0001). Of those receiving neoadjuvant chemotherapy,
37% in the MDC cohort underwent BCS vs 12% in the non-MDC cohort (p=0.08). Of those proceeding to surgery without
neoadjuvant therapy, 70% underwent BCS in the MDC cohort vs 46% in the non-MDC cohort (p<.0001). Among mastectomies
performed in the MDC vs non-MDC cohorts, 77% and 41% respectively were clinically indicated (P<.0001). Rates of unnecessary
contralateral prophylactic mastectomy were comparable in both groups, 39% (p=0.99).
Conclusions
Breast cancer patients seen in an MDC setting at the time of initial diagnosis are significantly more likely than women seen in a
non-MDC setting to undergo breast conservation. We hypothesize that the MDC model of breast cancer care, via facilitation of
more informed medical decision making, may be a viable strategy to curtail rising mastectomy rates in the United States.

Title: Residual disease after breast conservation surgery: To excise or not to excise?
Catherine R Campo1, Erika Reategui1, Sarah P Cate1, John Rescigno1, Priyanka Mittar1, Alyssa Gillego1 and Susan K Boolbol1.
1
Mount Sinai Beth Israel Medical Center, New York, NY.
Body: Background:
The definition of adequate margins after breast conservation surgery for invasive breast cancer has been a highly debated topic.
A recent consensus statement by the SSO/ASTRO recommends re-excision only for positive margins, defined as tumor at ink. In
light of this consensus statement, we studied the characteristics of patients undergoing breast conservation surgery with
subsequent re-excision at our institution in order to examine factors predictive for residual disease (RD+) in the re-excision
specimen, as well as a second re-excision.
Methods:
In this IRB approved retrospective chart review of our breast cancer cases from 1998-2013, we reviewed 828 patients who
underwent breast conservation surgery with re-excision for invasive breast carcinoma. A close margin was defined as less than
2mm from the inked margins, and a positive margin was defined as tumor at ink. We analyzed various clinicopathologic features.
RD+ was missing in 7 patients (0.8%), who were excluded from that analysis. Pearson chi-square was used to test significance in
univariate analysis. Binary logistic regression was applied in multivariate analysis of factors significant at p<0.05.
Results:
Overall, 230 patients (28%) had RD+, and 103 patients required a second re-excision (12.4%) due to persistently positive or close
margins (44% of RD+). Factors not significant for RD+ were: diagnosis era (before June 2007 vs. after), menstrual status, age <
40, race, excision volume, tumor size, nodal status, grade, histology, lymphovascular invasion, and hormone receptor status. For
patients with only close margins, there was no difference in RD+ by margin width, as defined by <0.5mm, 0.5 to <1mm, 1mm to
<2mm. Factors significant for RD+ were: disease that was mammographically occult or calcifications only (35%) vs. mass or
architectural distortion (25%), p=0.003; positive margin vs. close margin (33% vs. 24%, p=0.004); margin positive or close for
DCIS (41%) vs. both DCIS/invasive (30%) vs. invasive alone (17%), p<0.001; presence of DCIS (30% vs. 16%, p=0.001);
extensive intraductal component (EIC) (42% vs. 24%, p<0.001); the number of positive or close margins (1: 16%, 2: 26%, 3+:
48%, p<0.001).
In multivariate analysis, the number of positive or close margins was significant, (p<0.001), as was a margin positive or close for
DCIS (p<0.001). Factors significant in univariate analysis for a second re-excision were similar as those for RD+. In multivariate
analysis, independent factors for second re-excision were: EIC (p=0.007), number of positive or close margins (p<0.001), and a
margin positive or close for DCIS (p=0.001).
Conclusions:
Residual disease at re-excision for positive or close margins was present in approximately one-fourth of patients. The probability
of residual disease was unrelated to margin positivity or margin width within 2mm. Consideration of re-excision should take into
account the burden of intraductal disease in the specimen, its presence at or close to the margin, and the number of positive or
close margins.

Title: Intraoperative ultrasound-guided breast conserving surgery for palpable and nonpalpable breast cancer
Guldeniz Karadeniz Cakmak1, Ali U Emre1, Oge Tascilar1, Burak Bahadir2 and Selcuk Ozkan1. 1Bulent Ecevit University School of
Medicine, Zonguldak, Turkey and 2Bulent Ecevit University School of Medicine, Zonguldak, Turkey.
Body: Objective: Intraoperative ultrasound guided (IUG) breast conserving surgery (BCS) is being increasingly embraced by
breast surgeons worldwide. Real-time sonographic localization provides tumor free margins with low excision volumes and
decreases rate of reoperations. We aimed to compare the efficacy of IUG-BCS for palpable and nonpalpable breast cancer with
respect to margin status, re-excision rate, tumor free tissue sacrifice and cost-time analysis. The relationship between
intraoperative assessment of gross macroscopic and ultrasonographic margins and frozen section results, were also analyzed.
Methods: Between 2011 and 2014, IUG-BCS were performed to 208 patients with the diagnosis of in situ or invasive carcinoma.
Intraoperative localization protocol includes ultrasound visualization of the lesion, tumor margin determination, and image
confirmation of specimen and tumor bed. Sonographic and macroscopic assessment of the surgical margins by surgeon was
followed by frozen section analysis of each margin. Moreover, cavity shaved margins from tumor bed were also obtained for
permanent section analysis.
Results: Of the 208 patients, 89 (42.8%) had palpable and 119 (57.2 %) had nonpalpable tumors. The sensitivity of intraoperative
ultrasound localization was 100% (208/208 cases). Patients were on average 55 years old (range, 25-92). There was no
difference with respect to patient characteristics including age, menopausal status, personal-family history, oral contraceptive
usage, body mass index and tumor localization. Mean tumor size was 1.14 cm for nonpalpable and 2.87 cm. for palpable tumors.
Negative margins were achieved in 92.43 % of nonpalpable (110/119) and 91.01% (81/89) of palpable lesions at the initial
procedure verified by frozen section analysis. The involved margins were correctly identified by the surgeon via specimen
sonography in 95.4% of the cases (15/17). According to frozen section analysis of the 1248 ultrasonographically clear margins,
re-excisions were required for 16 margins of palpable and 14 margins of nonpalpable tumors. The overall positive margin rate
determined by frozen section analysis was 2.4% (30/1248), with the majority of these patients (12/17) proved to have significant
degrees of pure DCIS or mixed invasive ductal with DCIS at final histopathologic evaluation. Three patients (2.5%) with
nonpalpable and two patients (2.24%) with palpable tumors required a second operation for either determination of close margins
or multifocality at cavity shaved margins, without residual cancer on pathological examination of the reoperative specimens.
IUG-BCS resulted in smaller excision volumes in nonpalpable cancers due to their size, as expected. However, calculated
resection ratio was found to be similar for both groups. A cost and time analysis determined nothing significant between groups.
Conclusion: IUG-BCS is an invaluable and effective modality for both palpable and nonpalpable breast cancer in obtaining clear
surgical margins with optimum resection volumes and reducing re-operations. Furthermore, frozen section analysis of the
specimen margins together with shaving cavity margins of the tumor bed for permanent analysis could be a feasible method for
minimizing the requirement for reoperations.

Title: Intra-operative imaging of surgical margins in breast conserving surgery using localized tissue dielectric response
J Michael Dixon1, Lorna Renshaw1, Oliver Young1, Dhananjay Kulkarni1, Talha Saleem1, Moshe Sarfaty2, Ramaswamy
Sreenivasam2, Catherine Kusnick2 and Jeremy Thomas1. 1Edinburgh Breast Unit, Edinburgh, Midlothian, United Kingdom and 2L
S Biopath, Mountain View, CA.
Body: INTRODUCTION:
Breast conserving surgery (BCS) removes the tumor with a surrounding margin of normal breast tissue, and to reduce local
recurrence clear margins are needed. Involved margins identified by pathology several days later require a second operation.
Between 20-40% of BCS patients require re-excision. There is a clear unmet clinical need for effective and efficacious
intra-operative, real time assessment of breast tumor margins. The handheld, battery operated TOUCH Imaging device displays
spatially resolved images of margins based on tissue spectral dielectric response. This novel imaging device enables surgeons to
assess in real-time the status of the excised tissue margins.
AIM:
To determine the efficacy and safety of TOUCH Imaging device as used by surgeons during BCS and its health economics.
METHOD:
The 1st study of the TOUCH Imaging device was run in the USA with freshly excised breast tissue as part of BCS. The study was
run in the pathology lab to determine the efficacy and safety of using the device. The efficacy was determined by surgeons
interpretation of device generated margin images compared with permanent section pathology. A follow up study in the UK is
currently underway. Surgeons were trained to use the device in the operating room on freshly excised tissue. The imaging was
performed after the surgery was completed with no subsequent intervention. The margins were scored as clear or positive and
the localization of the abnormality was marked using special ink for specific pathology attention. In the learning phase 18 BCS
cases were performed. The main study is being run in two phases and is currently underway. Phase-1 aimed to determine the
accuracy of the device when used by individual surgeons in real time had been completed. Analysis is on the way, and phase-2
will start following the confirmation of device accuracy and will involve using the device to direct intra-operatively which margins
require further excision. Phase 2 will provide insight to the potential health economics of using the TOUCH Imaging device.
RESULTS:
Data from the first study in the USA on 50 BCS specimens, imaging 214 distinct margins are shown in the table. The sensitivity of
the device is high; the false negative rate low, the specificity and false positive rates are satisfactory.
TOUCH Compared to Permanent Section Pathology
Positive Pathology
Negative Pathology
Positive TOUCH
66 ( 30.8%)
50 (23.4%))
Negative TOUCH
5 ( 2.3%)
93 ( 43.5%)
% (95% CI)
% (90% CI)
Sensitivity
92.96 (84.33, 97.67)
92.96 ( 85.76, 97.18)
Specificity
65.03 (56.62, 72.81)
65.03 (57.93, 71.66)
False Positive Rate
34.97 ( 27.19, 43.38)
34.97 (28.34, 42.07)
False negative Rate
7.04 ( 2.33, 15.67)
7.04 ( 2.82, 14.24)
The training study in the UK on 18 BCS specimen results showed 50% True Positive, 39% True Negative, 11% False Positive,
6% False Negative rates. There were 5 cases (28%) of positive margins < 1 mm and the device images identified 4 with 1 case
(6%) being missed.
Results from the ongoing study will be presented.
CONCLUSIONS:
The TOUCH Imaging device is safe and efficacious.
It is easy for surgeons to use in the operating room.
This device has the potential to reduce BCS re-operation rate.

Title: Margin Probe device is able to reduce re-excision rate of breast conserving surgery in invasive and pre-invasive breast
cancer independent from any patient or tumor related factors
Jens Uwe Blohmer1,2, Julia Tanko1, Ragna Voelker1, Julia de Grahl1 and Jessica Gross1. 1Sankt Gertrauden-Krankenhaus, Berlin,
Germany and 2Charit, Berlin, Germany.
Body: Background: A positive margin status after breast conserving surgery (BCS) is one of the strongest predictors of local
recurrence of intraductal and invasive carcinoma. As much as 20-50% of all patients with BCS need to undergo a second
operation in order to receive free margins. In this study we tested the clinical performance of Margin Probe (Dune Medical
Devices), a novel device for intraoperative margin evaluation.
Methods: A prospective clinical trial was performed: The device was applied to 150 lumpectomy specimen from consecutive
patients with BCS treated during the first three months in 2013. The re-excision rate was compared to the re-excision rate of a
historical group of 156 patients treated with BCS during the first three months in 2012, without the application of the device. We
analyzed whether Margin Probe is affected by tumor morphology, grading, size of the tumor, breast density, age,
body-mass-index or the use of wire-marker.
Results: Due to the application of Margin Probe the re-excision rate decreased significantly by 51% from 39.7% to 14.6%. In the
subgroup of intraductal carcinoma (DCIS) the re-excision rate was reduced about two thirds from 66.7% to 23.1%. In the
subgroup of invasive lobular carcinomas the re-excision rate decreased from 37.0% to 19.0%. Margin Probe results are not
affected by grading, tumor size, breast-density, age, body-mass-index or wire-marker application.
Conclusion: Margin Probe is an effective tool for detection of positive margins during BCS and significantly decreases the
re-excision rate. It is not limited to invasive carcinoma but also detects involved margins in DCIS as well as in invasive lobular
carcinoma. It does not interfere with any of the factors we examined.

Title: Translating the concept of intrinsic subtypes into an oncoplastic cohort of more than 1000 patients-predictors of recurrence
and survival
Mahdi Rezai1, Stephanie Kellersmann2, Sarah Knispel2, Rainer Kimmig1 and Peter Kern1,2. 1Breast Center Dsseldorf, Dsseldorf,
North-Rhine Westfalia, Germany and 2University Hospital of Essen, Essen, North-Rhine Westfalia, Germany.
Body: Introduction:
The concept of breast cancer experienced a paradigm shift by Srlie T. et al. with intrinsic subtypes as prognostic classification of
breast cancer [1]. We validated this concept in a large cohort study of oncoplastic surgery.
Patients and methods:
We analysed 1035 patients with oncoplastic surgery (2004-2009) and survival parameters related to histopathological
approximated intrinsic subtypes.
Data were retrieved from customized questionnaires and patients charts. Survival data were determined from cancer registries.
Results:
A total of 944 patients with primary unilateral breast cancer, median age 58 years, were eligible for analysis. At a median
follow-up of 5.3 years, LRR was 4.1%, with 5-year-OS of 94.5% and DFS of 90.9%. Stage distribution was as follows: T1a 3%,
T1b 12 %, T1c 44,2 % and T2 was 22,1%. 70,4 % of patients were nodal-negative and nodal involvement was predominantly low.
Intrinsic subtypes, not T-size, nodal-status, resection margin width nor histopathology, governed the prognosis of this cohort.
Triple-negative and Her2 non luminal breast cancer had the highest recurrence and the lowest survival rates compared to
Luminal A: Recurrence TNBC 11,3 %, Her2pos non luminal 9,3 %, Luminal A 2,5 %; Overall survival: TNBC 91,3 %, Her2 non
luminal 93,7 %; Luminal A: 96,3 %. Our data confirmed the intrinsic subtype concept on a large basis in oncoplastic surgery.
(1) Srlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, Hastie T, Eisen MB, van de Rijn M, Jeffrey SS, Thorsen T,
Quist H, Matese JC, Brown PO, Botstein D, Lnning PE, Brresen-Dale AL: Gene expression patterns of breast carcinomas
distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA 2001; 98: 1086910874.

Title: Breast contour preserving surgery; a matter of patient age?
Annelotte C van Bommel1, Marie-Jeanne T Baas-Vrancken Peeters2, Marc A Mureau3, Margriet van der Heiden - van der Loo4,
Carolien H Smorenburg2 and Thijs van Dalen5. 1Leiden University Medical Center, Leiden, Netherlands; 2Netherlands Cancer
Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 3Erasmus MC Cancer Institute / University
Medical Centre Rotterdam, Rotterdam, Netherlands; 4Comprehensive Cancer Centre the Netherlands (IKNL), Utrecht,
Netherlands and 5Diakonessenhuis Utrecht, Utrecht, Netherlands.
Body: Background
Breast cancer treatment is a multimodality treatment with different options for the individual patient, aiming to preserve the breast
contour in the vast majority of the patients. Different treatment modalities can be used for this goal: primary breast conserving
surgery (BCS), BCS after neoadjuvant therapy and immediate breast reconstruction after ablative surgery. Which treatment
strategy is chosen may depend on patient preferences as well as patient age. The aim of the present analysis was to describe the
rate of breast contour preserving procedures (BCPP) as related to patient age.
Material and methods
All invasive M0 breast cancer patients diagnosed and operated in the Netherlands between 2011 September 2013 were
selected from the national NABON Breast Cancer Audit.
Primary BCS rates, BCS rates after neoadjuvant therapy, and rates of immediate reconstructions after ablative surgery were
calculated for various age groups: <40, 40 to 50, 50 to 70 and 70 years.
Results
A total of 34.577 patients were identified. Sixty percent (20.583 patients) underwent BCS. Primary BCS rates increased with age:
47%, 60%, 69% in patients aged <40 years, 40 to 50 years and 50 to 70 years, respectively. In patients aged 70 the BCS rate
was considerably lower (48%).
Immediate reconstructions were performed in 16% of all mastectomy patients. An inverse relationship was observed with patient
age; the rate of immediate reconstructions decreased from 38% in patients <40 years to 1% in patients 70 years of age. A
similar decrease in neoadjuvant therapy was seen over the various age groups in patients treated with BCS from 21% to 4% in
patients <40 and 70, respectively.
The table shows different treatment modalities. After combining these outcomes the rate of breast contour preserving surgery was
similar in patients aged <70 (62-73%).
Table 1. Various treatment modalities for invasive breast cancer patients
< 40 years
40 50 years
50 70 years
70 years
Primary breast conserving surgery
47%
60%
69%
48%
Breast conserving surgery and neoadjuvant therapy
21%
14%
6%
4%
Immediate reconstruction with mastectomy
38%
29%
16%
1%
Breast contour preserving surgery
62%
68%
73%
48%
* Proportion of patients with primary breast conserving surgery, breast conserving surgery after neoadjuvant treatment and
immediate reconstruction combined with mastectomy.
Conclusions
Patient age affected both the BCS with or without neoadjuvant therapy and the percentage of immediate reconstructions after
mastectomy. Overall, more similar percentages in preservation of the breast contour are achieved for various age groups
compared to BCS alone. Therefore, combining different treatment modalities into breast contour preserving surgery provides a
more complete overview of maintaining the breast contour related to various age groups than BCS alone.

Title: Magnetic sentinel node and occult lesion localization in breast cancer: Initial results of the MagSNOLL trial
Muneer Ahmed1,2, Ashutosh Kothari2, Hisham Hamed2, Sumit Goyal3, Tibor Kovacs2, Anand David Purushotham1,2, Bauke
Anninga1,2, Phillippa Young3, Julie Scudder2, Sarah McWililliams2, Sarah Pinder1,2, Quentin Pankhurst4, Ian Monnypenny3 and
Michael Douek1,2. 1King's College London, London, United Kingdom; 2Guy's and St Thomas' NHS Foundation Trust, London,
United Kingdom; 3University of Wales at Llandough, Cardiff, United Kingdom and 4University College, London, United Kingdom.
Body: Trial registration:
ISRCTN: 68689512
MREC No: 13/LO/0636
UKCRN ID: 14979
INTRODUCTION:
One-third of breast cancers diagnosed are non-palpable. These breast cancers require localization-guided surgery and axillary
staging using sentinel lymph node biopsy (SLNB). We present our experience of the first 20 patients undergoing a novel
technique of magnetic-guided lesion localization and concurrent SLNB, avoiding the need for wire-guided localization and
radioisotopes.
MATERIALS AND METHODS:
Co-localization of the primary tumour and sentinel lymph nodes (SLNs) using a novel magnetic technique was initially undertaken
on a protocol-driven predefined minimum of 10 patients with palpable breast cancers to assess the feasibility of the magnetic
tracer to safely localize at the point of injection and concurrently drain to the lymphatics. In order to confirm proof of principle, the
subsequent 20 patients with non-palpable breast cancers were analysed. An ultrasound-guided intra-tumoral injection of magnetic
tracer was performed (0.5 mL Sienna+, Endomagnetics Ltd, UK). Once successful lesion localization was confirmed (peak
magnetometer counts retained at the site of injection), the technique was undertaken in patients with non-palpable breast cancers
awaiting wide local excision and SLNB. All patients underwent SLNB with the magnetic and combined technique (radioisotope
and Patent Blue Dye).
RESULTS:
A total of 33 patients were recruited and 1 patient excluded due to breach of the trial protocol. This left a remainder of 32 patients
for consideration, of which 12 patients (1 bilateral) possessed palpable breast cancer and 20 patients non-palpable breast cancer.
Peak magnetometer counts were retained at the site of injection in all palpable (n=13) and non-palpable (n=20) breast cancers.
Re-excisions (second operations) for involved margins were performed in 2 patients with non-palpable breast cancers (10%). The
mean volume of excised specimens was 49 cm3 (SD 30.6). The SLN identification rate for the magnetic technique alone was
84% (28/33) overall and 85% (17/20) for non-palpable lesions. For the combined technique (radioisotope and blue dye) the SLN
identification rate was 97% (32/33) overall and 100% (20/20) for non-palpable lesions. When the magnetic technique is combined
with blue dye, the SLN identification rate overall was 97% (32/33) and for non-palpable breast cancers was 95% (19/20). The
mean number of sentinel nodes excised was 1.75 versus 2.05 for the magnetic and combined techniques respectively.
CONCLUSION: Magnetic lesion localization and concurrent SLNB is a feasible technique. Further optimisation and validation of
this technique, in a larger trial, is required.

Title: Pilot study of a passive non-radioactive electromagnetic wave technology to localize non-palpable breast lesions
Charles E Cox1, Melissa Themar-Geck2, Ronald Prati2, Michelle Jung1, Jeff King1 and Steven C Shivers1. 1USF Breast Health
Program, Tampa, FL and 2Breast Care Center at Florida Hospital Tampa, Tampa, FL.
Body: Background and Objectives: The standard preoperative technique for localizing non-palpable breast lesions is wire
localization (WL). Radioactive seed localization (RSL) has been described as an alternative approach to address the number of
clear disadvantages associated with WL. Yet, despite its proven advantages, the adoption of RSL has been impacted by
considerable regulatory requirements for the handling of radioactive materials. To advance the progress made with RSL and
eliminate the issues associated with radioactive components, the SAVI Scout surgical guidance system has been developed.
The SAVI Scout surgical guidance system is an FDA-cleared medical device that utilizes passive non-radioactive electromagnetic
wave technology to provide real-time guidance during excisional breast procedures. The purpose of this pilot study is to
determine the safety and efficacy of the SAVI Scout system in localizing and directing the removal of non-palpable breast lesions
during excisional biopsy and lumpectomy procedures. The preliminary results from two institutions are reported.
Materials and Methods: Following a feasibility study using the SAVI Scout system in resected breast tissues ex vivo, Institutional
Review Board approval was granted for both institutions for women with a non-palpable breast lesion requiring preoperative
localization for excision. Participating patients underwent localization and excision with the SAVI Scout system, which consists of
an electromagnetic wave reflective device (reflector), handpiece and console. Using mammographic or ultrasound guidance, the
reflector was placed percutaneously up to 7 days prior to the scheduled excisional procedure. At surgical excision, the surgeon
used the SAVI Scout handpiece to locate the reflector, which was removed along with the surrounding breast tissue. The console
provides audible feedback of reflector proximity to the handpiece. Successful reflector placement, localization and retrieval were
the primary endpoints.
Results: After the first training case (data not used), a total of 17 patients have participated in the study to date. The reflectors
were successfully placed with mammographic guidance in 9/9 patients and with ultrasound guidance in 8/8 patients. Reflectors
were placed an average of 1.7 days (range 0-6 days) before surgery. Five patients underwent excisional biopsy and 12 patients
had a lumpectomy. The intended lesion and reflector were successfully removed in 17/17 patients. Reflector migration did not
occur and no adverse events occurred. Of the 14 patients in which final pathology is currently available, 6 patients had no
invasive or in situ carcinoma identified, 4 had no tumor in the excision, and 3/4 patients with tumor had clear margins. One patient
had a focally positive margin and was recommended for re-excision.
Conclusions: The preliminary data show the SAVI Scout system to be a safe and effective tool for the localization of
non-palpable breast lesions. Ongoing accrual to this pilot study will validate these findings with planned enrollment of a total of
additional 50 patients in the next 60 days at up to 3 additional sites.

Title: Anticipating ACA's impact on breast cancer screening for medically underserved women reached through the Avon Breast
Health Outreach Program: Understanding the Massachusetts experience
Lindsay Senter1, Marvin R Aliaga1, Yixin Hu2, Kelly Morrison Opdyke1, Kathryn Gates-Ferris1 and Marc Hurlbert3. 1Cicatelli
Associates Inc (CAI), New York, NY; 2Mailman School of Public Health, Columbia University, New York, NY and 3Avon
Foundation for Women, New York, NY.
Body: Background
The Avon Breast Health Outreach Program (BHOP) supports community-based organizations to conduct education and outreach
to link low-income and uninsured women to routine breast cancer screening and care. Through the Affordable Care Act (ACA),
the number of uninsured individuals is expected to decrease as of 2014. While ACA implementation varies by state, BHOP
grantees may benefit from understanding the health reform experience in Massachusetts (MA) in 2006 and its impact on the
number and proportion of uninsured clients served over time.
Objective
Describe changes in client volume, health insurance status, and demographics following implementation of health reform in MA to
anticipate upcoming impact of ACA on BHOP grantees.
Methods
Confidential client intake records of continuously funded BHOP grantees from 2004-2012 were analyzed; dataset included
females aged 40-64, recruited for breast cancer screening. We compared records for 11,199 clients served by 4 MA BHOP
programs with 283,720 clients served by 52 control agencies funded in the US during the same time period. We examined
trends across years in agency-specific rates of health insurance coverage, and client volume adjusting for annual funding.
Changes in key demographic characteristics over time were also analyzed.
Results
The proportion of uninsured MA clients decreased dramatically from 46.6% to 6.2% from 2004-2012, with the biggest decrease
occurring in the 3 years following health reform (42.1% to 13.0% from 2006-2009); as compared to only a slight drop among
controls from 2004-2012 (69.3% to 67.4%). After adjusting for changes in annual BHOP funding, MA experienced a 74% increase
in client volume from 2006-2009, compared with a 15% increase among controls. From 2004-2012, the mean age of clients (53
years) remained stable and similar for MA and controls. The proportion of racial/ethnic minorities served increased considerably
in MA, from 62.0% to 77.7%, versus a smaller increase of 60.1% to 63.0% in controls. Likewise, the proportion of clients born
outside the US rose from 37.2% to 50.7% in MA, compared with 33.7% to 36.6% among controls. However, the proportion of
low-income clients (annual household income <$25,000) decreased overall from 90.4% to 83.1% in MA, and from 83.1% to
79.1% for controls.
Conclusion
In BHOP, Massachusetts observed a marked decrease (most noticeably in the 3 years following health reform) in the proportion
of uninsured clients served, while overall client volume and the proportion of racial/ethnic minorities and foreign-born women
increased.
Discussion
Historically, many BHOP grantees have relied on funding from the CDCs National Breast and Cervical Cancer Early Detection
Program or charity funds to pay for screening services for uninsured clients. As more clients nationally gain access to health
insurance, BHOP grantees need to be prepared for the changing landscape and potential increase in client volume, as
experienced by MA. Additionally, grantees should plan to update systems that aid in health insurance eligibility determination,
assist with enrollment, and strengthen provider partnerships and referral systems that accept multiple types of insurance.

Title: Adaptive and acquired mechanisms of resistance to PI3K inhibitors
Sarat Chandarlapaty1, Maurizio Scaltriti1, Marie Will1, Zhiqiang Li1, Ana Bosch-Campos1, Sarit Schwartz1, Vanessa
Rodrik-Outmezguine1, Michael Berger1, Baselga Jose1 and Neal Rosen1. 1Memorial Sloan Kettering Cancer Center, New York,
NY.
Body: Background: Activation of PI3K alpha is among the most frequent oncogenic events in breast transformation, commonly
occurring through mutations in the p110 alpha subunit of PI3K or amplification of the HER2 receptor tyrosine kinase. Therapeutic
targeting of activated PI3K signaling has been made feasible by the development of highly selective and potent ATP competitive
inhibitors of p110 alpha some of which are in late stage clinical development. Unfortunately, clinical response to these
compounds has not been seen for the majority of patients and when response occurs it is often short-lived. We have
hypothesized that relief of feedback regulation of PI3K signaling constitutes a major mechanism for tumor cell adaptation to
inhibitors of oncogenes such as PI3K.
Methods: To understand the basis for resistance to PI3K inhibitors we studied the consequences of acute and prolonged PI3K
inhibition in patient samples as well as in cell line and murine models of breast cancer. We examined the effects of selective
inhibition of PI3K alpha on upstream and parallel signaling pathway components utilizing several phosphoproteomic methods
including mass spectrometry, antibody microarrays, and immunoblotting. We further investigated the consequences of PI3K
inhibition upon the genome using next generation sequencing.
Results: In the context of tumors with high levels of RTK signaling such as HER2+ breast cancers, inhibitors of PI3K alpha led to
only transient inhibition of PI3K/AKT signaling with reactivation of signaling closely coinciding in time with loss of feedback
suppression of RTKs such as HER3 and IGF1R. This reactivation of AKT signaling could be blocked using inhibitors of the
induced RTKs and this was associated with a marked increase in tumor cell death. A second mechanism of reactivation of PI3K
signaling was observed in examining the genomes of tumors where inhibition of PI3K alpha was associated with acquired
inactivating mutations in PTEN. Loss of PTEN in such tumors was associated with failure of the PI3K alpha inhibitor to block
PI3K/AKT signaling but sensitivity to combined PI3K alpha plus PI3K beta inhibition in laboratory models. Finally, in the context of
hormone dependent tumor models, inhibition of PI3K alpha was associated with an increase in estrogen receptor activation. This
was observed as increases in ER protein expression, ER phosphorylation, and ER binding to established ER target promoters.
Combined inhibition of ER and PI3K alpha was demonstrated to be synergistic in these models in vivo.
Conclusions: Feedback suppression of upstream and parallel signaling pathways poses a major limitation to the antitumor effects
of single agent PI3K alpha inhibition. Combination approaches to potently inhibit PI3K alpha, PI3K beta, and either ER or
HER3/IGF1R may prove more effective and durable in the clinic.

Title: Impact of a culturally syntonic door-to-door breast cancer early detection intervention
Carmen J Calfa1, Julie G Wilkinson2, Mindy M Williams3, James M Pann4, Angela Yehl5, Stephanie E Hoogenbergen6 and Andrea
D Ivory6. 1Memorial Cancer Institute, Hollywood, FL; 2ImmunoSite Technologies, Miramar, FL; 3Memorial Division of Breast
Surgical Oncology, Hollywood, FL; 4Nova Southeastern University, North Miami Beach, FL; 5Nova Southeastern University, Fort
Lauderdale, FL and 6Women's Breast Health Initiative Florida Affiliate, Miami Lakes, FL.
Body: Introduction. Disparities in the detection of breast cancer persist despite efforts to reach underserved populations and
increase mammography utilization. Black and Hispanic women are more likely to be diagnosed at a later stage compared to
White women. Lack of health insurance and limited awareness of breast cancer are significant predictors of screening behaviors;
other reasons remain poorly understood and warrant further research. Our study presents highlights of the Womens Breast
Health Initiative Florida Affiliate (WBHI) intervention and key findings related to its impact.
Methods. The intervention targeted single-family home neighborhoods with high incidence of late-stage breast cancer, median
income at 200% of the poverty level, and access to affordable health clinics. Neighborhood mobile mammography or
transportation to local facilities was provided, followed by navigation to assure proper follow up. Eligible women were uninsured,
40 years of age and were qualified for an annual screening mammogram.
During 2011-2012
Intervention Days
45
Educational Materials
21,079
Volunteers*
2,025
Women Interviewed
5,441
Eligible Women
643
Appointments Scheduled
581
Screening Days
15
Screening Mammograms Completed
409
*culturally relevant & trained

Descriptive and inferential statistical methods were used to analyze participant data divided into 3 groups based on neighborhood
racial demographics: "Black" or "Hispanic" data groups had 75% Black or Hispanic residents respectively, while the "Other" did
not have a predominant race. Post-visit phone surveys of 1,871 participants were conducted within 30 days.
Results. The intervention significantly increased screening rates amongst eligible women from 22% to 64%, had a 0.24% breast
cancer diagnosis rate and motivated 59% of the women to propagate awareness. Health insurance coverage rates did not differ
significantly between the 3 groups. Notably, amongst the insured, those from "Hispanic" neighborhoods had the highest recent
mammogram rate while, amongst the uninsured, those from "Black" neighborhoods had the highest. Neighborhood culture and
prior mammogram history impacted the efficacy of the intervention. Among the "Hispanic" and "Other" groups, women with no
prior mammograms were less receptive to receiving one compared to those who had prior mammograms. This phenomenon was
not observed in the "Black" group. Womens responses and perceived benefit to the educational package differed according to
the neighborhood race and ethnicity:
"Black"
"Hispanic"
"Other"
Benefited
70%
85%
88%
Spread the Word
54%
57%
65%
Discussion. WBHI reached the goal of increasing screening rates in this underserved population, thus showing that this type of
intervention can be effective. Educational packages presented in a face-to-face format within a door-to-door context were found
useful by most women. Our findings highlight disparities beyond a womans access to breast healthcare and education by
showing that neighborhood culture impacts response to outreach intervention and breast healthcare. This generates the
hypothesis that given equal access to care, disparities in using screening mammography will remain. Educational methods further
tailored to racial and ethnic characteristics may play a significant role in closing disparity gaps.

Title: Association of BRCA1 mutations with impaired ovarian reserve: A plausible connection between infertility and
breast/ovarian cancer risks
Sara B Giordano1, Navdha Mittal2, Kristin Smith2 and Mary Ellen Pavone2. 1Medical University of South Carolina, Charleston, SC
and 2Northwestern University, Chicago, IL.
Body: Purpose: Mutations in either the BRCA1 or BRCA2 gene are associated with breast and ovarian cancer susceptibility.
Lifetime risk estimates for ovarian cancer in the general population indicate that 1.4 percent (14 out of 1,000) of women will be
diagnosed with ovarian cancer compared with 15 to 40 percent of women (150-400 out of 1,000) who have a BRCA1 or BRCA2
mutation. For decades, scientists have been attempting to establish a link between the risk of developing cancer and infertility.
One prospective study suggested an association with BRCA1 mutations and occult primary ovarian insufficiency. This study
showed a novel association between low response to ovarian stimulation and BRCA1 mutations, suggesting a link between
double-strand DNA break repair dysfunction, infertility, and breast/ovarian cancer risks.
If substantiated, this may place BRCA mutation positive women at higher risk for chemotherapy induced ovarian failure. We also
know that reproductive factors including low parity and infertility may place women at increased risk for breast cancer. Taking into
account that 1% of females suffer from primary ovarian insufficiency and that the underlying mechanism remains unknown most
of the time, discovery of susceptibility gene may have implications for understanding the link between infertility and breast/ovarian
cancer.
We hypothesize that mutations in the BRCA1 gene will adversely affect ovarian reserve, as measured by AMH levels.
Participants and Methods: Subset analysis of serum samples taken from Northwestern Ovarian Cancer Early Detection Program
(NOCEDPP) of reproductive aged women who had testing for the BRCA mutation. Using an IRB approved protocol, the
NOCEDPP database was searched for women 18-45 years old who had previous BRCA testing. These women must have also
provided consent for their stored serum to be used for research purposes. We excluded those with a BRCA2 mutation, a previous
history of cancer and/or cancer treatment, and those with a previous unilateral or bilateral oophorectomy or other ovarian surgery.
Statistical analysis was done using parametric and nonparametric testing.
Results: A total of 125 met the criteria for our study. 66 women were BRCA1 mutation-positive and 59 were BRCA mutationnegative. The median age for BRCA1 mutation-positive women was 33.5 years while BRCA mutation-negative patients was 37
years (p<0.05). Body mass index, gravidity, parity, and duration of birth control were similar between the groups. Overall median
AMH values were 2.6ng/mL in both groups. However, when broken into age groups, BRCA1 mutation-positive women aged
35-39 had a significantly lower AMH level when compared to BRCA mutation-negative women (median 3.6ng/mL vs 1.3ng/mL,
p<0.05).
Conclusion: AMH values were significantly lower in BRCA1 mutation-positive women aged 35-39. Women with the BRCA1
mutation should be counseled regarding this potential decrease in ovarian reserve.

Title: Formative evaluation of interventions addressing culturally relevant needs of young breast cancer survivors
Rochelle Shoretz1, Adina Fleischmann1, Elana Silber1, Mary Ann Hall2 and Ashani Johnson-Turbes2. 1Sharsheret, Teaneck, NJ
and 2ICF International, Atlanta, GA.
Body: Background. Young women diagnosed with breast cancer face challenges impacting their quality of life, psychosocial
functioning, and reproductive health outcomes. Ashkenazi Jewish women experience a disproportionately high prevalence of
genetic mutations in the BRCA1/2 genes linked to a higher incidence of hereditary breast and ovarian cancer, and have unique
needs based on genetic risk and cultural factors. Sharsheret, a national not-for-profit organization with an expertise in supporting
Jewish young breast cancer survivors (YBCS), developed two interventions, the Peer Support Network (PSN) and Genetics for
Life (GFL), to address the culturally specific needs of Jewish YBCS. In 2011, Sharsheret conducted a formative evaluation to
strengthen its existing PSN and GFL interventions.
Methods. Four focus groups were conducted with 27 YBCS having participated in the PSN or GFL. The groups investigated the
information and support needs of YBCS of Jewish descent and YBCS perceptions of PSN and GFL content, materials and
delivery channels. Two groups (online and in-person) investigated the needs of YBCS, PSN and GFL program content, and PSN
and GFL program delivery related to the psychosocial health of YBCS. Two groups (online and in-person) explored YBCS needs,
PSN and GFL program content, and PSN and GFL program delivery related to the reproductive health of YBCS. Participants
were recruited using in-person methods, email blasts, social media and via Sharsheret website postings. Qualitative data were
analyzed using a notes-based thematic analysis.
Results. Across groups, respondents identified the primary information needs for YBCS included resources addressing genetics,
side effects and consequences of treatment (e.g., effects on fertility, intimacy and premature menopause) rather than treatment
itself. Respondents identified that their support needs included culturally relevant peer support and genetic resources at the time
of diagnosis, treatment and after treatment, as well as support for partners, family members and friends. Participants preferred to
receive information and support through diverse modalities, including health care providers, the internet, peer supporters, support
groups, symposia and teleconferences and mass media. Respondents recommended outreach efforts for the PSN and GFL
interventions in locations that attract young people, including pediatricians offices, schools, salons, and college campuses, and to
clergy, school principals, local Jewish organizations and health care providers. Participants also suggested potential
technology-based modifications to PSN and GFL intervention delivery, including online intake forms, a LiveChat option, video
testimonials, interactive expert column, online pedigrees and family conference calls with a genetic counselor or clinical staff
member.
Conclusion. YBCS identified the PSN and GFL interventions as critical resources attentive to their culturally specific information
and support needs. Focus group participants provided strategies to enhance intervention content and reach. Healthcare
professionals serving YBCS of diverse backgrounds can utilize the data gleaned from this evaluation to shape culturally relevant
interventions and materials to address the unique needs of YBCS.

Title: Sentinel node surgery after neoadjuvant chemotherapy in patient with axillary node involvement: The French GANEA 2
prospective multi-institutional trial
Jean-Marc Classe1, Loic Campion1, Severine Alran2, Christine Tunon de Lara3, Pierre Francois Dupre4, Christelle Faure5, Nicolas
Paillocher1, Serge Lasry2, Marie Pierre Chauvet6, Gilles Houvenaeghel7, Marian Gutowski8, Pascaline De Blaye9, Charlotte Ngo10,
Emmanuel Barranger11, Jean Luc Veraeghe12, Celine Lefebvre13, Jean Francois Rodier14, Virginie Bordes1, Helene Charitansky15,
Gwenael Ferron15 and Pierre Gimbergues16. 1Institut de Cancerologie de l'Ouest, Saint Herblain, France; 2Institut Curie, Paris,
France; 3Institut Bergoni, Bordeaux, France; 4CHU Morvan, Brest, France; 5Centre Lon Brard, Lyon, France; 6Centre Oscar
Lambret, Lille, France; 7Institut Paoli-Calmettes, Marseille, France; 8Centre Valdorel, Montpellier, France; 9CH Les Oudairies, La
Roche sur Yon, France; 10Hopital Europeen Georges Pompidou, Paris, France; 11Centre Lacassagne, Nice, France; 12Centre
Alexis Vautrin, Nancy, France; 13CHU Angers, Angers, Pays de la Loire, France; 14Centre Paul Strauss, Strasbourg, France;
15
Institut Claudius Regaud, Toulouse, France and 16Centre Jean Perrin, Clermont-Ferrand, France.
Body: Background
Half of the patient treated with neoadjuvant chemotherapy (NAC) for a large operable breast cancer has no axillary lymph node
involvement at the time of surgery. Sentinel lymph node detection (SLND), performed after NAC, must select patient who should
be spared of an axillary lymph node dissection (ALND). The application of SLND for staging the axilla after NAC for patient who
initially had a proven axillary lymph node involvement remains controversial because of a low detection rate (DR) and a high false
negative rate (FNR).
Objective
The aim of GANEA 2 trial was to assess the DR and the FNR of SLND after NAC in the particular case of patients with a proven
axillary lymph node involvement.
Patients and Method
GANEA 2 was validated by scientific and ethical national boards.
Inclusion criteria: FIGO stage T2-T3 infiltrating breast carcinoma, indication of NAC, surgery (radical or conservative) after NAC
and signed consent form,
Exclusion criteria: inflammatory cancer, local relapse, previous surgical removal of the tumour, mental disorder, pregnancy or no
contraceptive method, contra-indication to NAC, NAC interrupted due to progressive disease.
Design: Diagnosis and indication to plan a NAC, control of inclusion and exclusion criteria, consent form signature, axillary
sonography with fine needle cytology before NAC to select patients with a proven lymph node involvement. After NAC patients
underwent both SLND, with the combined technique Blue dye and radiolabeled colloid, and complementary ALND.
Pathological procedure: Pathological analysis, of sentinel and non sentinel nodes, carried out according to standard methods and
classified according the last American Joint Committee staging system and Sataloff classification.
Studied parameters were detection rate, false negative rate and Sataloff grading on tumor and lymph nodes. We evaluated
particularly the likelihood that the FNR in patients with one or more SLN examined was greater than 10%.
Patients with no lymph node involvement before NAC underwent only a SLND with an ALND only in the case of SLN
macro-metastasis with a rigorous follow up. They are not part of this abstract.
Results
From July 2010, to February 2014, 242 patients from 19 institutions were enrolled, with a proven axillary lymph node involvement
before NAC. After NAC, 1/3 had metastasis free axillary lymph node (80/142).
Detection rate was 83.1% (201/242). Half of the patients with a detection failure had an involved ALND. The false negative rate
was 14.2% in the whole series but 24.5% in the case of only 1 SLN resected, and 8% in case of more than 1 SLN resected. In
case of involved SLN, half of the patients had involved ALND. Considering the node Sataloff scoring, 18 of the 20 false negative
cases were grade C or D (n=15 grade C, metastatic disease and therapeutic effect; n = 3 grade D, metastasis and no therapeutic
effect).
Conclusion
Among patients treated by NAC for a large operable breast cancer with proven involved lymph node before NAC, who had only 1
SLN examined, the false negative rate was 24.5%. SLND with the combined technique, provides a FNR of less than 10% only in
the case of 2 or more SLN resected.

Title: Methods impacting the false negative rate of sentinel lymph node surgery in patients presenting with node positive breast
cancer (T0-T4,N1-2) who receive neoadjuvant chemotherapy Results from a prospective trial ACOSOG Z1071 (Alliance)
Judy C Boughey1, Karla V Ballman2, William F Symmans3, Linda M McCall4, Elizabeth A Mittendorf3, Gretchen M Ahrendt5, Lee G
Wilke6, Bret Taback7 and Kelly K Hunt3. 1Mayo Clinic, Rochester, MN; 2Alliance Statistics and Data Center, Mayo Clinic,
Rochester, MN; 3University of Texas MD Anderson Cancer Center, Houston, TX; 4Alliance Statistics and Data Center, Duke
University, Durham, NC; 5Magee-Womens Surgical Associates, Pittsburgh, PA; 6University of Wisconsin Hospital and Clinics,
Madison, WI and 7Columbia University Medical Center, New York, NY.
Body: Background: The American College of Surgeons Oncology Group (ACOSOG) Z1071 trial (Alliance) reported a false
negative rate (FNR) of 12.6% with sentinel lymph node (SLN) surgery after neoadjuvant chemotherapy in women presenting with
node-positive breast cancer. Proposed methods to decrease the FNR include clip placement in the positive node at initial
diagnosis with confirmation of resection of the clipped node at surgery and inclusion of residual metastatic cells identified by
immunohistochemistry (IHC) for cytokeratins and disease measuring <0.2mm on H&E in the definition of node positivity after
chemotherapy. Herein we evaluate the impact of these methods on FNR.
Methods: Z1071 was a prospective multi-institutional trial in which women with clinical T0-4,N1-2,M0 breast cancer underwent
both SLN surgery and axillary dissection (ALND) after neoadjuvant chemotherapy. The primary endpoint defined nodal
metastasis as disease >0.2mm on H&E. Slides were submitted for central IHC staining when H&E-negative, unless performed
locally. For this analysis, we expanded the definition of node positivity to include metastases less than 0.2mm on H&E and
metastatic deposits identified by IHC.
In 171 cases, a clip was placed in the node at initial biopsy and in these cases the SLNs and ALND were evaluated by x-ray or
pathology to document clip location. FNR in the clip cohort is reported using the primary endpoint definition of node positive
disease (>0.2mm on H&E).
Results: In the 113 cases where the clip was found in the SLN the FNR was 6.4% (1.8-15.5%). In the 29 cases where the clip
was found in the ALND specimen, the FNR was 22.2% (6.4-47.6%).
IHC was available on 470 of 525 patients with cN1 disease and 2 SLNs resected. Using the definition of H&E metastasis
>0.2mm, the FNR in these patients was 11.3% (34/301, 8.0-15.4%) which decreased to 8.7% (27/311, 5.6-11.8%) when including
any disease <0.2mm. SLNs from 16 patients had disease <0.2mm in size. Seven patients previously classified as false negative
SLN changed to true positive with identification of disease in the SLN. Nine patients changed from node-negative to node-positive
with the only disease being <0.2mm disease found in the SLN. Nodal pathologic complete response rate changed from 36.0% to
33.8% with the inclusion of metastases <0.2mm.
Residual disease identified in SLNs or ALND
FNR (%)
95% CI
SLN metastases >0.2mm by H&E
470
301 (64.0%)
11.3
8.0-15.4
SLN metastasis any size on IHC or H&E
470
311 (66.2%)
8.7
5.6-11.8
Clip in SLN
114
62 (54.4%)
6.4
1.8-15.5
Clip in ALND
29
18 (62.1%)
22.2
6.4-47.6
Clip location unknown
29
20 (69.0%)
14.3
3.0-36.3
Clip not placed
354
207 (58.5%)
13.5
9.1-18.9
Node positive definition
Clip location
Conclusion: Placement of a clip at initial diagnosis of node positive disease with identification of the clip during the SLN surgery
reduces the FNR. Use of IHC with inclusion of metastases <0.2mm in the definition of residual nodal disease in women with node
positive breast cancer after chemotherapy also improves the accuracy of SLN surgery. Use of one or both of these methods
should be considered when performing SLN in this setting.

Title: Sentinel node procedure before or after neoadjuvant chemotherapy in clinically node negative or positive patients; results
from 3 phase III studies of the Dutch breast cancer trialists group (BOOG)
Vivianne C Tjan-Heijnen1, Birgitte E Vriens1, Maureen J Aarts1, Judith R Kroep2, Cock J van de Velde2, Gerrit-Jan Liefers2, Ayoub
Charehbili2, Petronella G Peer3 and Maaike de Boer1. 1Maastricht University Medical Centre, Maastricht, Netherlands; 2Leiden
University Medical Center, Leiden, Netherlands and 3Radboud University Medical Center, Nijmegen, Netherlands.
Body: Background
In breast cancer patients treated with neoadjuvant systemic therapy, the timing of the sentinel node (SN) procedure in patients
with clinically node-negative disease (cN0) at diagnosis and the role of the SN procedure after neoadjuvant therapy in patients
with initially clinical node-positive disease (cN+) still remains to be elucidated.
Methods
Between February 2006 and May 2012, 657 patients were enrolled in three clinical trials on neoadjuvant systemic therapy under
the auspices of the Dutch Breast Cancer Research Group (BOOG). In the INTENS and NEO-ZOTAC study, patients were treated
with TAC or AC-T neo-adjuvant chemotherapy (doxorubicin (A), cyclophosphamide (C) and docetaxel (T)). In the TEAM IIa trial,
patients were treated with neoadjuvant endocrine therapy (exemestane for 6 months). Timing of the SN procedure in cN0 disease
mainly depends on local policy and period of inclusion. We compared the pN0 rate after SN procedure performed before versus
after neoadjuvant systemic therapy, in patients with cN0 disease at initial diagnosis. Further, we assessed the SN negative and
the false-negative sentinel node rate in patients with initially cN+ breast cancer and conversion to cN0 when treated with
neoadjuvant systemic therapy. The false-negative rate was obtained by dividing the number of patients who were SN-negative
but non-SN positive by the number of patients who had a positive SN or a positive non-SN.
Results
In total, 271 patients (n=93 INTENS, n=107 NEOZOTAC, n= 71 TEAM IIa) underwent a SN procedure (n=233 cN0; n=38 cN+).
Of patients with cN0 breast cancer at diagnosis, 131 (56%) underwent the SN procedure before and 102 (44%) after neoadjuvant
systemic therapy, with ypN0(sn) or ypN0(i+)(sn) in 90/131 (69%) and 69/102 (68%) patients, respectively. Patients with initially
cN+ disease who converted to cN0 disease after systemic therapy had a negative SN in 29% (11/38) of cases. The false-negative
rate was 20% (6 of 30). More detailed analyses will be presented at the meeting.
Conclusion
To our knowledge, this is the first study comparing the impact of the timing of the sentinel procedure in patients with cN0 disease.
We showed, that irrespective of the timing of the SN procedure - before or after neoadjuvant systemic therapy - in patients with
cN0 breast cancer at diagnosis, two-third of patients had a negative sentinel node. In patients who converted after neoadjuvant
chemotherapy from cN+ to cN0 the false negative sentinel node rate was 20%.


Title: Correlation of percutaneously biopsied axillary lymph nodes marked with black tattoo ink prior to neoadjuvant
chemotherapy with sentinel lymph nodes in breast cancer patients
Nicole Choy1, Jafi Lipson1, Sunita Pal1, Debra Ikeda1, Long Trinh1, Kimberly Allison1, Michael Ozawa1, Amanda Wheeler1 and
Irene Wapnir1. 1Stanford University School of Medicine, Stanford, CA.
Body: Introduction:
Sonographic evaluation of the axilla and percutaneous biopsy of abnormal lymph nodes with fine needle aspiration (FNA) or core
needle biopsy (CNB) has become more common practice in patients with newly diagnosed breast cancer prior to neoadjuvant
chemotherapy (NAC). Sentinel lymph node (SLN) biopsy is considered the gold standard for axillary staging in clinically node
negative breast cancer patients. Currently, there is no clear correlation of sonographically detected abnormal lymph nodes and
open surgical assessment. We conducted an exploratory pilot study which marked suspicious axillary lymph nodes with black
tattoo ink at the time of percutaneous needle biopsy prior to NAC. Black nodes visualized during axillary surgery were evaluated
in comparison to SLNs.
Methods:
Breast cancer patients with clinical and/or sonographically suspicious axillary lymph nodes prior to NAC were included in the
study. Following FNA or CNB biopsy of node, 0.1 to 0.5 ml of a sterile, highly purified, biocompatible fine carbon suspension
(Spot) was injected into the cortex of the lymph node and adjacent soft tissue. A total of 12 patients were injected with black ink
prior to NAC. Intraoperative presence of black pigment was assessed and correlation between sentinel and tattooed nodes were
evaluated.
Results:
Nine patients had a positive percutaneous lymph node biopsy prior to NAC. The average number of days that elapsed between
injection and to surgery was 130 days. A successful SLN procedure was performed in all patients. A black tattooed node was
identified in all patients and correlated to a SLN. 7 patients were down-staged in the axilla and 6 patients did not go onto
completion axillary dissection. One patient with a negative SLN had a completion axillary dissection, but no additional positive
lymph nodes were found. Four patients with positive SLN had a completion axillary dissection (1 of whom was a false negative
percutaneous biopsy). In all four patients, the positive sentinel node contained visible black ink. There was one patient who had
an additional positive sentinel node, which was not black. Two axillary dissections contained additional positive nodes.
Conclusion:
Black ink tattooing with sterile black ink (Spot), successfully marked suspicious lymph nodes prior to NAC. These correlated to
a SLN. In node positive patients with a partial response in the axillary lymph nodes following neoadjuvant chemotherapy,
previously marked, black-inked node proved to be the persistent positive node. Tattooing of lymph nodes at the time of
percutaneous biopsy may improve the accuracy of surgical axillary staging by aiding in the intra-operative identification of
previously biopsied nodes.

Title: In patients with breast cancer, pre-operative sentinel node biopsy using intradermal microbubbles and contrast enhanced
ultrasound predicts volume of axillary metastases
Karina Cox1, Jennifer Weeks1, Ritchie Chalmers1, Pippa Mills1, David Fish1 and Ali Sever1. 1Maidstone and Tunbridge Wells NHS
Trust, Maidstone, Kent, United Kingdom.
Body: Introduction
In patients with early breast cancer, there is a trend towards conservative axillary surgery but there are concerns that patients
with high volume axillary metastases may receive sub-optimal axillary treatment leaving them vulnerable to local recurrence.
Determining the metastatic status of sentinel lymph nodes (SLN) in the pre-operative/ diagnostic period would allow focussed
surgical planning and facilitate decisions regarding adjuvant therapy early in the patient pathway.
Methods
654 patients with primary invasive breast cancer and a normal grey-scale axillary ultrasound were included in the analysis.
Pre-operatively, patients received periareolar intra-dermal injection of microbubble contrast agent, breast lymphatics were
visualised by ultrasound and followed to identify axillary SLN. Contrast-pulse sequencing and grey-scale ultrasound modalities
were used to image SLN. Sentinel lymph nodes were then subjected to core biopsy. Patients subsequently underwent tumour
excision and axillary node clearance (ANC) or surgical sentinel node biopsy (SNB) using blue-dye and isotope +/- completion
ANC.
Results
Sentinel lymph nodes were clearly visualised in 605 patients and successfully (B2-B5) core biopsied in 555. The test identified
53% of all SLN metastases with 100% specificity. The negative predictive value was 88%. The prevalence of axillary lymph node
metastases was 23%. Given a benign (B2) SLN biopsy result, the post-test probability that a patient had SLN metastases on
subsequent surgical excision was 12%. Following ANC, 55% of patients with a malignant (B4/5) biopsy result were found to have
high volume axillary disease (2 macro metastases or more) whereas 21% of patients with an initial benign (B2) biopsy result and
metastatic cells found at SNB had high volume axillary disease identified after completion ANC (P value less than 0.001). In total,
only 2% of patients with a benign (B2) SLN core biopsy result had high volume axillary metastases and half of these had either
multifocal cancer or invasive lobular carcinoma.
Conclusions
SLN can be readily identified and biopsied in the breast clinic using intradermal microbubbles and CEUS. In patients with primary
invasive breast cancer and a normal grey scale axillary ultrasound, a benign (B2) SLN core biopsy result may be highly predictive
of either no metastases or low volume metastatic disease in the ipsilateral axilla. This group of patients is therefore likely to
benefit from axillary conservation and in certain cases, it may be appropriate to completely omit a surgical SNB.

Title: Reoperative sentinel lymph node biopsy for ipsilateral breast tumor recurrence: Impact of axillary surgery and radiation
therapy on lymphatic drainage patterns
Ayaka Sato1,4, Takehiko Sakai1, Fumiko Moroo1, Kiyomi Kimura1,5, Masahiko Tanabe1,6, Rie Horii2, Futoshi Akiyama3 and Takuji
Iwase1. 1Cancer Institute Hospital, Japanese Foundation for Cancer Research, Tokyo, Japan; 2Cancer Institute Hospital,
Japanese Foundation for Cancer Research, Tokyo, Japan; 3The Cancer Institute of Japanese Foundation for Cancer Research,
Tokyo, Japan; 4University of Tokyo Hospital, Tokyo, Japan; 5International University of Health and Welfare Mita Hospital, Tokyo,
Japan and 6Juntendo University Hospital, Tokyo, Japan.
Body: Background: While sentinel lymph node biopsy (SLNB) is a well-established method for patients with clinically
node-negative primary breast cancer, reoperative sentinel lymph node biopsy (re-SLNB) for ipsilateral breast tumor recurrence
(IBTR) still remains controversial. The aim of this study was to evaluate the location of reoperative sentinel lymph nodes
(re-SLNs) and the factors related to aberrant drainage patterns after breast conserving surgery (BCS).
Methods: Between April 2005 and August 2013, we performed re-SLNB on 132 patients who developed IBTR with no metastatic
lymph nodes in preoperative examination. Both injection of radioisotope (Tc-99m phytate) and preoperative lymphoscintigraphy
were performed on 97 patients of them, who were eligible for this retrospective study. We divided these patients into two groups
according to their previous axillary surgery: 55 patients with just SLNB or no axillary surgery to the group S, and 42 patients with
previous axillary lymph node dissection (ALND) to the group D. The patients in the group S were divided into two subgroups: 19
with adjuvant radiation therapy (RT) and 36 without RT. We compared visualization rates of lymph drainage patterns in these
groups statistically by using Chi-square test. We evaluated the relationship between previous local treatments and lymphatic
drainage patterns.
Results: Number of cases in which lymphatic drainage was visualized were 52 (94.5%) in the group S and 33 (78.6%) in the
group D (P < 0.001). Re-SLNs were observed at ipsilateral axilla in 47 patients (90%) in the group S and 19 patients (58%) in the
group D (P < 0.001), at internal mammary (IM) in 9 (17%) and 16(49%) (P < 0.01), and at contralateral axilla in 8 (15%) and
5(15%), respectively.
Within the group S, lymphatic drainage was visualized in 34 (94.4%) of 36 patients without RT and 18 (94.7%) of 19 patients who
had received RT after BCS. Re-SLNs were visualized at ipsilateral axilla in 34 (100%) and 13 (72%) (P < 0.01), at IM in 5 (15%)
and 4 (22%), and at contralateral axilla in 1 (3%) and 7 (39%) (P < 0.001).
Re-SLNB was successfully performed in 62 (94%) of 66 patients whose lymphatic drainage was observed at ipsilateral axilla.
Metastases for carcinoma in re-SLNs were found in 8 (13%) of them. At IM, re-SLNB was successful in 15 (60%) of 25, and 2
(13%) of them were positive for cancer. At contralateral axilla, re-SLNB was successful in 8 (62%) of 13 while no metastases
were found among them.
Conclusion: Previous ALND affected the visualization rates of re-SLNs. In patients after previous ALND, SLN was more likely to
be in IM. Adjuvant RT effected a change in lymphatic drainage patterns, but did not affect the visualization rate of re-SLN. At
ipsilateral axilla, the rates of successful re-SLNB and metastases at re-SLNs were similar to SLNB for primary breast cancer.
Metastasis to SLN of recurrent cancer is one of the most important information for management of IBTR. However, it was very
difficult to predict localization of re-SLN because lymphatic drainage patterns could have been affected by previous treatments.
Therefore, we recommend re-SLNB using radioisotope with lymphoscintigraphy in surgery for IBTR.

Title: Does axillary lymph node ratio still have a prognostic value in neoadjuvant setting?
Seung Pil Jung1, Hye Yoon Lee1, Woo Young Kim1, Hee Yong Kwak1, Jae Bok Lee1, Hoon Yub Kim1 and Jeoung Won Bae1.
1
Korea University Hospital, Korea University College of Medicine, Seoul, Korea.
Body: Background : Axillary nodal status is one of the most important prognostic factor in breast cancer. The lymph node ratio
has been reported as an independent prognostic factor because the dissected and involved lymph node could be different across
the institutions along with surgical and pathologic procedure. Neoadjuvant chemotherapy has been preferred treatment method in
locally advanced breast cancer for the purpose of reducing tumor mass in breast and/or axillary area. Neoadjuvant chemotherapy
would downstage axillary nodal status and lower total number of excised lymph nodes compared with upfront surgery. Therefore,
there has been emerged question whether the axillary nodal status and lymph node ratio following neoadjuvant chemotherapy
could accurately predict the prognosis. To answer this question, we evaluated the axillary nodal status and lymph node ratio as a
prognostic factor after neoadjuvant chemotherapy.
Methods : One hundred thirty two patients were eligible in this study between 2005 and 2011. The patients underwent breast
surgery after 3 or 6 cycles of anthracycline with or without taxane based chemotherapy according to their axillary nodal status.
The cut-off range of lymph node ratio were divided to low (0.20, n=45), intermediate (2<, 0.65, n=25) and high (<0.65, n=6).
Clinical and pathologic factors including hormone status, tumor size, lymphovascular invasion and response to chemotherapy
were evaluated and survival results were also investigated with nodal status.
Results : Mean age of total patients was 48 years. Mean follow up period was 40 months. The pathologic complete responses
were observed 33% in breast, 25% in axillar, 17.4% in both areas, respectively. Tumor subtype was an independent factor for
pathologic response rate. With multivariate analysis, clinical tumor size, pathologic nodal status and the presence of
lymphovascular invasion were statistically significant to disease free survival in overall cohort. In node negative patients group
(n=57), the hormone receptor status and molecular subtype were associated with the disease free survival using multivariate
analysis. Whereas, pathologic tumor size, pathologic nodal status and lymph node ratio, especially lower cut-off limit 0.20, were
significant risk factor for disease free survival in node positive (n=75) patients.
Conclusion : Neoadjuvant chemotherapy significantly affect axillary nodal status. Traditional nodal staging has been accepted as
important prognostic factor. The response to neoadjuvant chemotherapy depends on tumor subtype. In this study, we proved that
nodal ratio could be a candidate as prognostic factor in neoadjuvant setting and 0.20 is acceptable lower cut-off value of lymph
node ratio.

Title: Application of the Z0011 criteria on Dutch breast cancer patients
Nicole C Verheuvel1, Vivianne CG Tjan-Heijnen2, Adri C Voogd2,3 and Rudi MH Roumen1,2. 1Mxima Medical Center, Veldhoven,
Netherlands; 2Maastricht University, School GROW, Maastricht, Netherlands and 3Dutch National Cancer Registry, Eindhoven,
Netherlands.
Body: Background
The American College of Surgeons Oncology Group reported a randomized controlled trial (Z0011-trial) among women with
T1-2N0M0 breast cancer treated with breast conserving therapy. It showed that axillary lymph node dissection may be redundant
in selected sentinel node positive patients. Though, it raises questions as to the general applicability of these results. Therefore,
this study aims to examine the practice changing effect and the clinical relevance of the criteria described in the Z0011 trial, when
applied to a Dutch population of patients diagnosed with invasive breast cancer.
Methods
This is a multicenter study including patients with T1-2N0-1M0 invasive breast cancer treated in either of the 10 different hospitals
affiliated with the Dutch National Cancer Registry region South in the period between January 2007 and December 2012. All
patients underwent a clinical examination, a mammography and sonographic examination of the breast and the ipsilateral axilla.
Chi-square analyses or a Fishers exact test were used to assess differences in patient and tumor characteristics the Z0011
population and the Dutch cohort. A p-value of 0.05 was considered statistically significant.
Results
A total of 11,031 patients had invasive breast cancer, of whom 5368 patients were treated with breast conserving therapy. In
6.8% axillary status was determined by the ultrasound guided lymph node biopsy and in 93.1% axillary metastases were found
trough the sentinel node procedure after a negative or inconclusive ultrasound. When applying the Z0011 criteria to our study
population of node positive patients, 5.7% of all breast cancer patients, equalling 11.8% of patients receiving breast conserving
therapy, met the Z0011 criteria.
Conclusion
These results suggest that when applying the Z0011 criteria to all node positive patients, including patients with a positive sentinel
node and patients with a positive ultrasound guided lymph node biopsy, only in 5.7% of all invasive breast cancer patients could
be spared an ALND and its morbidity. This is a small practice changing effect. Nevertheless, it is worthwhile to consider
implementing the findings of Z0011 trial in daily clinical practice.


Title: Prospective evaluation of the reliability of the combined use of two models to predict non-sentinel lymph node status in
breast cancer patients with metastatic sentinel lymph nodes: The MSKCC nomogram and the Tenon score PHRC-NOTEGS
study
Roman Rouzier1, Catherine Uzan2, Alexandra Rousseau3, Eugenie Guillot1, Sonia Zilberman3, Charles Meyer4, Pablo Estevez1,
Pierre-Franois Dupre5, David Kere6, Virginie Doridot7, Gauthier D'halluin8, Xavier Fritel9, Nicolas Pouget10, Chafika Mazouni2,
Tabassome Simon3 and Charles Coutant10. 1Institut Curie, Paris, France; 2Gustave Roussy, Villejuif, France; 3APHP, Paris,
France; 4Hpitaux Civils de Colmar, Colmar, France; 5CHU Brest, Brest, France; 6Institut Jean Godinot, Reims, France; 7Centre
Mdical Rpublique, Clermont-Ferrand, France; 8Centre Clinical Soyaux, Soyaux, France; 9CHU Poitiers, Poitiers, France and
10
Centre Georges Franois Leclerc, Dijon, France.
Body: Background: Several mathematical models have been developed to predict non-SN status in patients with breast cancer
with SN metastasis. The Memorial Sloan-Kettering Cancer Center nomogram and Tenon score outperform other methods in
academic studies but their exportability at multiple geographic locations and practice settings has never been reported. The
purpose of this study was to prospectively evaluate the combined use of the MSKCC nomogram (Memorial Sloan-Kettering
Cancer Center) and Tenon score to select, in patients with metastatic sentinel lymph node (SN), those at low risk of metastatic
non-SN in whom additional axillary lymph node dissection (ALND) could be avoided.
Material and methods: From January 2011 to July 2012, data on 3157 patients with breast cancer from 65 institutions (university
affiliated, general, regional hospital, nonprofit private hospital and private practice) were prospectively recorded (NCT01509963).
Selection criteria were patients aged over 18 years old with untreated invasive T1-2 breast cancer with an indication of SN
procedure. The primary outcome measure was the false negative rate in patients with both a 10% probability of metastatic
non-SN with the MSKCC nomogram and a Tenon score 3.5 (i.e. low risk): proportion of patients with metastatic non-SN at
additional ALND. The hypothesis was a 5%5% rate in this group of patients. Other patients were considered at high risk.
Because of the results of the Z011 and IBCSG 23-01 trials, additional ALND was not mandatory.in case of metastatic SN.
Results: Among the 2936 patients, at least one SN was metastatic (isolated tumor cells, micro- or macrometastasis) in 696
patients (23.7%). Among them, 178 did not have completion ALND. Among patients with completion ALND (n=518, 74.4%), 67
(13%), 437 (84%) and 14 (3%) patients were at low, high and undetermined combined risk while 47.5% were at low risk in
patients without completion ALND (p<.001). This study did not meet its primary objective as the false negative rate in patients
with low risk was 16.4% (11/67) [95% confidence interval: 8.5%-27.5%]. The false negative rate was significantly higher in
patients in the high risk group: 33.9% (148/437) [95% confidence interval: 29.6%-38.4%] had non-SN metastasis (p=.004).
Analyzed individually, Tenon score and MSKCC nomogram had complementary performances (number of patients at low risk:
35.3% and 18.5%, false negative rates: 21.9% and 17%, concordance index: 0.63 and 0.65, respectively).
Conclusion: In this controlled prospective trial, metastatic SN patients with both a 10% probability of metastatic non-SN with the
MSKCC nomogram and a Tenon score 3.5 had completion axillary dissection in 47% of cases: in these patients, the false
negative rate was statistically over 5% and did not reach the primary endpoint. Further evaluation is warranted to determine the
outcome of patients with and without axillary dissection.

Title: Impact of the ASOCOG Z0011 trial on a multi-institutional mixed practice setting
Swapnil D Kachare1, Nasreen A Vohra1, Kathyrn M Verbanac1, Timothy L Fitzgerald1, Emmanuel E Zervos1 and Jan H Wong1.
1
East Carolina University, Greenville, NC.
Body: Background
The ASOCOG Z0011 trial is widely considered to be practice changing, but reports to date on the adoption of Z0011 are largely
confined to tertiary academic institutions with breast cancer surgical specialty services. The aim of this study was to assess the
impact of Z0011 on the care of breast cancer patients in a mixed academic/ non-academic practice setting.
Methods
All patients who underwent surgical treatment for breast cancer in the Vidant Healthcare System, a multi-county network,
between the years January 1, 2009 and September 20, 2013 were identified. Patients with Stage IV disease and ductal
carcinoma in situ were excluded.
Results
One thousand six hundred four patients met initial inclusion criteria, 661 in the pre-Z0011 period and 973 in the post-Z0011
period. The majority of patients were female (99%) and Caucasian (65%). Infiltrating ductal cancer (67%) was the most common
tumor type. Tumors on average were 23mm in size with the majority being ER (74%), PR (63%) positive and Her2 (82%)
negative. Most patients had node negative disease (73%). Table 1 summarizes the characteristics of the pre- and post-Z0011
study populations.
Patient and Tumor Characteristics
Variable
Pre-Z0011
Post-Z0011
p value
Female (%)
656 (99.2)
968 (99.4)
0.86
Race (%)
0.49
White
448 (67.8)
642 (65.9)
Black
205 (31.0)
314 (32.2)
Mean size (mm)
19.9
22.3
Histology (%)
0.06
0.07
Infiltrating ductal cancer
434 (65.7)
663 (68.1)
Infiltrating lobular cancer
30 (4.5)
62 (6.4)
ER Positive (%)
484 (73.3)
722 (74.2)
0.92
PR Positive (%)
406 (61.5)
622 (63.9)
0.61
Her2 Negative (%)
304 (78.4)
897 (83.5)
0.01
Triple Negative Subtype (%)
83 (23.9)
168 (18.9)
0.07
Node Positive Status
175 (27.5)
926 (26.7)
0.71
A total of 142 patients met Z0011 criteria, with T1-2 disease, underwent partial mastectomy and had a positive SNB. There were
a similar number of patients who met Z00111 criteria in the pre- (n=63) and post-Z0011 (n=79) periods, p>0.05. In the post-Z0011
period, patients were more likely to undergo SNB alone (56 vs. 19%, p<0.001). Post-Z0011 the decrease in ALND was in those
who had immediate ALND (34% vs. 71%), with no substantive change in the delayed group (10.1 vs. 9.5%), p<0.001. Patient who
underwent SNB alone had smaller tumors (22 vs. 32mm, p<0.001), but similar histologic subtypes (p=0.60) and receptor status
(p0.15) as compared to patients who underwent SNB with ALND. Over the entire study period patients in an academic practice
were less likely to undergo SNB alone (31 vs. 43%). In the post-Z0011 period, patients cared for in an academic practice had an
increase in SNB alone from year 1 (2012) to year 2 (2013) (40% to 57%), while the non-academic practices had a decrease in
SNB alone from year 1 to year 2, (63% to 56%).
Conclusion
In this mixed practice setting there was an overall decrease in ALND in T1/T2 patients with low metastatic axillary burden
following publication of Z0011. Immediately post-Z0011 the non-academic practices more rapidly adopted SNB alone, however
these differences were no longer apparent two years following Z0011. Continued longitudinal studies will further assess the
variability in the incorporation of Z0011 among various surgical practices.


Title: Pre-operative axillary staging results in over-treatment in some breast cancer patients
Andrew Pieri1, Henry Cain2 and Sebastian Aspinall3. 1Wansbeck Hospital, Ashington, Northumbria, United Kingdom; 2Royal
Victoria Infirmary, Newcastle Upon Tyne, Tyne & Wear, United Kingdom and 3University Hospital of North Tyneside, North
Shields, Tyne & Wear, United Kingdom.
Body: Introduction
Following a diagnosis of breast cancer, pre-operative ultrasound staging of the axilla is recommended. Patients found to have
metastatic disease on biopsy or FNA can proceed directly to axillary clearance (ALNC) at the time of their breast tumour excision.
However, although ultrasound staging is a sensitive test to detect axillary disease, it does not differentiate between low and high
volume nodal metastasis. Recent evidence suggests that, in selected patients with low volume axillary disease following SLNB,
completion ALNC may be safely omitted. This has been reflected in the recently updated ASCO guidelines. In light of this current
trend towards axillary conservation, some patients undergoing ALNC after positive axillary staging may be over-treated.
The aims of this study is to establish the nodal burden of patients undergoing ALNC following positive axillary staging and to
compare the nodal burden of those patients with those undergoing completion ALNC after a positive SLNB.
Methods
Data was collected prospectively over 12 months from nine hospitals within the North of England Cancer Network. Age, tumour
characteristics, breast operation, axillary staging results, axillary operation(s) and nodal results were recorded.
Results
A total of 1010 patients with breast cancer underwent pre-operative axillary staging. 215 patients (21%) had an ALNC. Of these,
115 (53%) cases underwent a primary ALNC after positive staging. The remaining 100 (47%) patients had a completion ALNC
following a positive SLNB. The nodal burden for patients undergoing ALNC is shown in the table below:
Total number of positive nodes
ALNC after positive staging
ALNC after positive SLNB
n patients
% patients
n patients
% patients
32
28
47
47
13
11
24
24
12
10
11
11
11
10
39
34
14
14
Totals
115
100
Analysis of the data revealed that tumour size was a significant predictor of low volume axillary disease (i.e. <3 nodes positive)
[p=0.02, Mann-Whitney test].
Conclusion
In this study, 46% of patients undergoing ALNC after positive staging had low volume disease (2 or less positive nodes). In the
context of recent evidence if these patients had undergone a SLNB following the positive pre-operative staging they may have
avoided a completion ALNC.
This study demonstrates tumour size to be a significant predictor of low volume axillary disease and thus may be an important
factor to consider alongside pre-operative staging when selecting patients who may be better managed by a SLNB rather than
proceeding directly to ALNC.

Title: Noninvasive assessment of axillary lymph node metastases in breast cancer using EpCAM antibody
Shu Wang1, Jiajia Guo2, Liulu Zhang3 and Houpu Yang4. 1Peking University People's Hospital Breast Center, Beijing, China;
2
Peking University People's Hospital Breast Center, Beijing, China; 3Peking University People's Hospital Breast Center and
4
Peking University People's Hospital Breast Center.
Body: Background:Axillary lymph node metastasis is one of the most important prognostic determinants for patients with breast
cancer. Its a big progress from axillary lymph node dissection (ALND) to sentinel lymph node biopsy (SLNB), however, SLNB
also has some limitations, including the detection failure rate, false negative rate, tracer allergies et al. Since about 70% of early
breast cancer patients have negative SLNs, surgery maybe overtreatment for these patients and noninvasive assessment of
axillary staging seems to be a better choice. Its not easy to detect and mapping metastasis lymph nodes in vivo non-invasively
because we dont have specific cancer markers, but what we know is that most breast cancer cells come from epithelial origin
and should have epithelial markers, and this kind of markers should not show up in lymph system. EpCam is a surface epithelial
marker and expressed in 80-100% breast cancer patients. The aim of this study was to evaluate the value of EpCAM antibody in
detecting breast cancer lymph node metastases non-invasively.Methods:VX2 rabbit model which is an epithelial tumor model
with lymph node metastasis was used in this experiment. Anti-EpCAM was labeled with a near infra-red (NIR) fluorescence
imaging agent Cy7 and selected as metastasis tracer. New Zealand white rabbits were assigned to study group and control group
at random. Twelve rabbits in the study group received VX2 tumor injection to bilateral mammary glands and 12 rabbits in the
control group were tumor free. Twenty four cases of axillary lymph nodes metastasis were investigated in each group because of
bilateral experiments. Anti-EpCAM/Cy7 was subcutaneously injected around the tumors in study group or in mammary gland in
control group at a dosage of 0.1ml (10umol/L). After 30 hours of screening by a NIR fluorescence detection machine, all the
axillary lymph nodes in both groups were removed for pathological examination and the fluorescence status of each node was
recorded. Result:The fluorescence signal could be detected 2 hours after injection in 24 cases in the control group and 23 cases
in the study group. Thirty hours later, fluorescence disappeared in all cases in the control group and all the lymph nodes resected
were pathological negative. While in study group, fluorescence signal could be detected continuously in 14 cases and 13 were
proven to be pathological metastasis. The case failed to have fluorescence image from the beginning finally proved to be vascular
invasion and cancerous node formation. The sensitivity, specificity rate of anti-EpCAM/Cy7 to detect axillary lymph node
metastasis was 92.8% (13/14) and 90.0% (9/10). Conclusion:The NIR dye labeled EpCAM antibody (anti-EpCAM/Cy7) has a
high accuracy in evaluating the axillary lymph node status of breast cancer in the animal experiment. Our tiny preclinical
experiment showed the value of using EpCAM antibody as a new method for noninvasive lymph node evaluation for breast
cancer and potentially extended in a lot of other epithelial originated cancers.

Title: Role of 18 FDG PET/CT in axillary staging in early breast cancer
Cristina Noblia1, Eugenia Azar1, Dolores Mansilla1, Amilcar Osorio2, Eduardo Armanasco1, Diana Montoya1, Martin Ipia1, Gaston
Berman1, Eduardo Gonzalez1, Christian Gonzalez2, Gabriel Bruno2, Patricia Parma2, Carla Pulero1 and Ana Alvarez1. 1Angel H.
Roffo Institute, Buenos Aires, Caba, Argentina and 2Nuclear Diagnostic Center Foundation, Buenos Aires, Caba, Argentina.
Body: OBJETIVE To evaluate the 18F-FDG PET/CT capacity for the detection of axillary metastases and compare results with
sentinel lymph node biopsy (SLNB).MATERIAL AND METHODS: 99 female patients with clinical T1T2N0 breast cancer were
included. Patients with recent breast or axillary surgery, T3T4 disease, ductal carcinoma in situ, inflammatory carcinoma,
uncontrolled diabetes mellitus and pregnant or lactation patients were excluded. Pre-operative FDG PET-CT was performed 15
days before surgery, SLNB took place with the combined method (radioisotopes and patent blue). Pearson and Spearman
correlation test were used to evaluate association of main variables with a significance level (p) of 0.05.RESULTS: Breast
PET-CT results: 80 positive PET/CT. Negative PET/ CT in 19 patients (4invasive lobular carcinoma, 7 tumors <7mm) Sensitivity
81%, specificity 100%.97 patients were operated (2 were stage IV with no surgery criteria). We found significant correlation of the
FDG tumor uptake (SUV) with tumor size (p< 0.0001), histological grade (p< 0.009), nuclear grade (p<0.001), mitotic grade (p<
0.007) and Ki 67 (p< 0.0001).In all cases the correlation was positive.There was no correlation with hormonal receptors
rate.Axillary PET CT results: positive PET/CT in 16 patients (16%). Of this 6 SNLB were negative (FPR= 9%). Specificity 91%,
PET /CT were negative in 81 patients (84%), 17 had axillary metastases(5 micrometastases, 2 isolated tumor cells and 2 lobular
carcinoma).Sensitivity:37%.FNR=63%. Correlation of histopatological axillary metastases was positive with breast tumor SUV
(P<0.0002) and tumor size (p<0.01), but was not significant with histological, nuclear and mitotic grade, Ki 67,hormonal receptor
rate and molecular subtype CONCLUSION: 1- PET / CT does not provide benefits for axillary staging in initial stages due to its
low spatial resolution (6-8 mm) 2 A negative axillary by PET/CT does not replace the sentinel node technique. 3 We can
suspect the presence of axillary metastases when there is an intense FDG uptake in the breast tumor.

Title: Total tumoral load as a prediction tool of non-sentinel node metastases in patients with early breast cancer and positive
sentinel lymph node assesed by OSNA
Martin Espinosa-Bravo1, Francesc Prez-Ceresuela1, Sebastian Diaz-Botero1, Vicente Peg2 and Isabel T Rubio1. 1Breast Surgical
Unit, Breast Cancer Center, Hospital Universitari Vall d'Hebron, Universitat Autnoma de Barcelona, Barcelona, Spain and
2
Hospital Universitari Vall d'Hebron. Universitat Autnoma de Barcelona, Barcelona, Spain.
Body: Background: Total tumoral load (TTL) in the sentinel lymph nodes (SLN) assessed by the OSNA assay is a new variable
that is able to predict the likelihood of more axillary metastasis. Compared with the number of positive SLNs, the TTL is
independent of the number of metastatic SLNs and a better predictor of further nodal metastasis. Although establishing specific
cutoff-points of the TTL can be questionable because they may change in the future and increasingly become more patient
specific, we have seen that TTL <15.000 could be a good cut off by its high negative predictive value (NPV). Changes in practice
have been occurred after the ACOSOG Z0011 trial and nowadays, patients with 2 positive SLNs may spare an axillary lymph
node dissection (ALND) with no impact on the oncological outcome. In the Z0011, 27% of patients had additional non-SLN
metastasis in the cALND. The objective of this study was to compare two methods of SLN metastatic burden, the TTL and the
number of positive SLNs, for predicting non-SLN metastasis in patients with positive SLNs by OSNA.
Methods: This is a retrospective cohort study of 145 consecutive patients with cT1-T2 invasive breast cancer with
ultrasonographically node-negative treated between April 2010 to April 2013, where there is at least one positive SLN assesed by
OSNA. We design a prediction tool based on the TTL results determined by OSNA to calculate the likelihood of not finding more
positive non-SLN (http://www.vhebron.net/es/calculadora-risc-metastasi-2). Our group have demonstrated that TTL=15.000 could
be a good cut off by its high NPV and sensitivity, 85.5% and 76.7%, respectively.
Results: A total 325 SLNs were removed with a mean of 2.24. The type of SLN metastasis were macrometastasis in 85 patients
(58.6%) and micrometastasis in 60 patients (41.4%). When considering patients with number of positive SLNs, of 109 patients
with 2 positives SLNs and an cALND, 17 patients (22%) had non-SLN metastasis; and of 5 patients with 3 SLN, four (80%) had
non-SLN metastasis. Taking the TTL results, of 51 patients with TTL <15.000 copies/L with cALND (90.2% with one positive
SLN), 7 patients (14%) had non-SLN metastasis, were a media of 2.2 extra non-SLN metastasis.
Of 55 patients with TTL 15.000 copies/L (with 1 or 2 positives SLNs) with cALND, 20 patients (36%) had non-SLN metastasis
and there were a media of 2.5 non-SLN metastasis (range 1-8). Of 34 patients (62%) with one positive SLN with TTL 15.000
copies/L, 12 patients (35%) had non-SLN metastasis; and of 21 patients with two positive SLN with TTL 15.000 copies/L,
eight patients (38%) had non-SLN metastasis.
Conclusion: The total tumor load obtained by OSNA predicts non-SLN metastasis independent of the number of positive SLNs.
The assess of the TTL in our cohort of patients show that patients with one or two positives SLNs with high TTL had high
likelihood of non-SLN metastasis greater than 27% of Z0011 criteria. Prospective studies are needed to determine the clinical
impact of this variable in patients outcome in the management of patients with SLN assesed by OSNA.

Title: Preoperative axillary imaging with ultrasonography: Among the breast cancer patients with lymph node metastases, can we
identify the patients who may omit axillary dissection?
Naoto Kondo1, Takashi Fujita1, Masataka Sawaki1, Masaya Hattori1, Akiyo Yoshimura1, Mari Ichikawa1, Junko Ishiguro1, Yayoi
Adachi1, Haruru Kotani1, Tomoka Hisada1 and Hiroji Iwata1. 1Aichi Cancer Center Hospital, Nagoya, Aichi, Japan.
Body: (Introduction) ACOSOG Z11 and EORTC AMAROS showing little benefit to axillary dissection(ALND) for early stage
breast cancers with limited nodal disease have led us to questioning the value of preoperative axillary imaging. It may not result in
a benefit to the patients to perform ALND by diagnosing a few or small axillary lymph node(ALN) metastases preoperatively.
(Aim) A purpose of this study is to determining the association between diagnostic method and metastatic number of ALN
metastases, and to find the patients who can omit ALND safely even if with ALN metastases.
(Methods)A database of consecutive primary breast cancer patients who underwent comlete ALND at our institution in 2008-2011
was analyzed. After we excluded patients treated with neoadjuvant systemic therapy, a total of 390 patients were included. By
diagnostic methods of ALN metastases, we classified them in four groups as follows. Group A (n=41) : suspicious ALNs on
axillary ultrasound(AUS) and ultrasound-guided fine needle aspiration cytology (FNAC) positive, Group B (n=47) : only one
abnormal ALN on AUS +/- FNAC, Group C(n=53) : multiple abnormanl ALNs +/- FNAC, Group D (n=249) : negative ALNs on
AUS but SLNB positive .
(Results) The median number (range) of ALN metastases were 3(1-22) in GropuA, 2(0-12) in Group B, 7(1-37) in Group C,
1(1-17) in Group D. There were significant differences in number of metastases between Group A/B/C and Group D (p=.02,
p=.01, p=.002). Paitents with 3 or less positive ALNs were 24.5% (13/53) in Group C, whereas 61.0% (25/41) in Group A and
68.1% (32/47) in Group B (p=.04, p=.02). We next evaluated the influence of patient- and tumour-related variables on the number
of positive ALNs in Group A/B. Factors such as age, tumour size (<20mm vs >20mm), ER status(positive vs negative), HER2
status (positive vs negative), nuclear grade (1/2 vs 3), menstruation status (pre vs post menopausal) were examined. However
there were no significant differences in all factors between the patients with 4 or more ALN metastases and patients with 3 or less
ALN metastases. As a result of multivariable analysis, relative risk (95%CI, p-value) of age was 0.92 (0.546-1.347, 0.84), tumour
size : 0.82 (0.511-1.418, 0.76), ER status : 1.06 (0.873-1.821, 0.76), HER2 status : 1.17 (0.853-2.390, 0.28), nuclear grade : 0.43
(0.420-1.22, 0.34), menstruation status : 0.85 (0.538-1.693, 0.89). (Conclusion) In our contemporary series, patients diagnosed
as ALN metastases preoperatively have significantly more involved nodes compared to SLNB positive patients regardless of the
diagnostic method, suggesting that such patients should proceed to ALND. Preoperative axillary imaging are useful to identify
node-positive breast cancer patients requiring ALND .

Title: Long-term follow-up of the node-negative breast cancer patients before treatment evaluated with sentinel node biopsy
alone following neoadjuvant chemotherapy
Hiroko Nogi1, Ken Uchida1, Makiko Kamio1, Kumiko Kato1, Yasuo Toriumi1 and Hiroshi Takeyama1. 1Jikei University School of
Medicine, Tokyo, Minato-ku, Japan.
Body: BACKGROUND Sentinel node biopsy (SNB) for the node negative breast cancer is standard treatment as an accurate
assessment of axillary lymph node status. The use of neoadjuvant chemotherapy (NAC) has increased during the past several
years. At present, the use of SNB following NAC is controversial. Proponents of SNB after NAC prefer a single surgical procedure
with potential for fewer axillary dissections.
OBJECTIVE Our objective was to examine SNB evaluation alone following NAC in patients with clinically node-negative breast
cancer before treatment and to evaluate the axillary lymph node recurrence for patients undergoing NAC versus patients
undergoing surgery first.
METHODS A total of 1176 patients with clinically node negative breast cancer underwent SNB from 2007 to 2013.
Clinicopathologic and survival data were reviewed and comparisons made between patients receiving NAC and those undergoing
surgery first. Lymphatic mapping was performed with both radioactive colloid and blue dye. Patients with negative SN for
metastasis were followed without axillary lymph node dissection (ALND). Patients with metastases to a SN underwent ALND.
RESULTS Of the patients, 180 (15.3%) underwent SNB following NAC and 996 (84.7%) underwent surgery first. SN identification
rates were 98.3% in the NAC group and 98.9% in the surgery first group. 152 (84.4%) patients in the NAC group and 877 patients
(88.1%) had negative SNs. At median follow-up of 49.2 months, 1 patient (0.007%) in the NAC group and 3 patients (0.003%) in
surgery first group had axillary lymph node recurrences. There were not any significant differences between the NAC group and
the surgery first group.
CONCLUSION The SNB following NAC in patients with node-negative breast cancer has a low axillary recurrence rate and could
be acceptable.

Title: The ratio and size of positive sentinel lymph nodes predicts the involvement of non-sentinel lymph nodes following
completion axillary lymph node dissection
Rajiv V Dave1, Muhhamed N Chauhan1, Maria Ghaus1, Sana Ahmed1, Shiv Shapra1, Joshua Marriott1, Craig Sayers1, Zbigniew
Kryjak1 and Deedar Ali1. 1Pinderfields Hospital, Wakefield, West Yorkshire, United Kingdom.
Body: Background: The role of Completion Axillary Lymph Node Dissection (CALND) following positive Sentinel Lymph Node
Biopsy (SLNB) is being actively debated. The involvement of our unit in the POSNOC trial (which has a no-treatment arm), has
prompted a review of our units CALND results, in order to examine predictors of involvement of non-sentinel lymph nodes
(n-SLN), to better inform our patients during recruitment.
Methods: We retrospectively analysed our experience of SNLB between July 2008 to 2013. A total of 1152 breast cancer patients
underwent SLNB based on lymphoscintigraphy, intraoperative gamma probe detection, and blue dye mapping using
99mTc-nanocolloid and Patent Blue V injected peri-areola.
Results: Out of 1152 SLN biopsies performed, 224 were positive for metastatic disease. 203 patients were anaesthetically
capable of progressing to CALND. On univariate analysis, involved n-SLN on CALND could not be predicted by age (<50 years;
16/77 vs 50 years; 93/125, p=0.272), size of tumour (<50mm; 44/193 and 50mm; 4/10, p=0.188), procedure (mastectomy;
24/93, WLE; 24/106, p=0.444), lymphovascular invasion (22/97 vs 26/106, p=0.812), number of positive SLN (2; 45/196, >2; 3/7,
p=0.213), receptor status; ER (negative; 3/13, positive 45/187, p=0.619), PR (negative; 6/28, positive 41/169, p=0.478), Her2
(negative; 43/175, positive 5/25, p=0.414), triple negative (2/8 vs 46/192, p=0.612). There was a trend toward higher incidence of
positive n-SLN with increasing grade (G0-2; 28/139 vs G3; 20/64, p=0.062) and extracapsular spread (14/41 vs 32/149, p=0.073),
but these did not reach statistical significance. Positive n-SLN on CALND was however predicted by macrometastases in SLN
(macrometastases; 39/141 vs micrometastases; 9/62, p=0.029) and ratio of positive nodes (<0.5; 18/109 vs 0.5; 30/94,
p=0.008). There were 4/224 recurrences (3 distant metastases and 1 loco-regional), which were not predicted by any of the
clinicopathological variables investigated. 3 patients who recurred only had one positive node on SLNB.
Conclusion
In our series of more than 200 SLNB, a ratio of >0.5 positive SLN yield and presence of macrometastases in positive SLN, were
associated with positive n-SLN on CALND.

Title: Management of the axilla in breast cancer patients: Do we over treat patients with preoperative diagnosis of nodal
metastasis?
Ioannis Michalakis1, Jaroslaw Krupa1 and Louisa Dunk2. 1Glenfield Hospital, Leicester, United Kingdom and 2Glenfield Hospital,
UHL Trust, Leicester, United Kingdom.
Body: Introduction: Axillary lymph node status is the most significant single prognostic factor in breast cancer patients. Axillary
ultrasound scan (AUS) followed by fine needle aspiration (FNA) is the gold standard modality in preoperative staging of the axilla
and spares the patient a possible second operation. Sentinel lymph node biopsy (SLNB) accurately stages the axilla with low
axillary recurrence rates and reduced morbidity. Increasing evidence suggests that surgical removal of the axillary lymph nodes
(completion ALND) in early breast cancer with limited nodal disease yields no advantage in terms of either overall or disease-free
survival.The purpose of this audit is to identify possible changes in management of the axilla in patients having preoperative
diagnosis of nodal metastasis.
Methods: We reviewed the available data from breast cancer patients that had positive FNA after AUS (441) or positive nodes at
a SLNB (324) in our unit from 2007 to 2013.
Results:410 of the 441 (93%) patients with a preoperative diagnosis of nodal metastasis received completion ALND. 147 of the
410 (36%) patients had neoadjuvant chemotherapy and of these 34 (23%) had complete pathological response at the axilla. 101
of the 263 (38%) patients who did not receive neoadjuvant chemotherapy had 1 or 2 nodes involved and also had a T1 or T2
tumour.
Patients with positive axillary FNA and ALND
Numper of patients
ALND without neoadjuvant chemotherapy:263
ALND after neoadjuvant chemotherapy:147
Age (median)
64
48
Tumor size (T)
T1:56 T2:190 T3:16
Histological Grade
G1:11 G2:105 G3:144
G1:15 G2:84 G3:28
ER status positive
200 (76%)
96 (65%)
Her 2 status positive
49 (19%)
44 (30%)
Median nodes removed
20
17
Median positive nodes
Patients with <3 nodes positive
103 (39%)
72 (49%)
Total number of patients 410

161 of the 324 (47%) patients with positive SLNB had completion ALND. 67 of the 161 (42%) patients received chemotherapy
prior the completion ALND. All the patients who had a positive SLNB and who did not receive completion ALND received
radiotherapy to the axilla.
Patients with positive SLNB
Number of patients
Radiotherapy to the
axilla:163
With completion ALND after

chemotherapy:67
With completion ALND before

chemotherapy:94
Age (median)
64
50
59
Tumor size (T)
T1:83 T2:73 T3:5
T1:20 T2:39 T3:8
T1:37 T2:48 T3:7
Histological Grade
G1:27 G2:93 G3:40
G1:8 G2:31 G3:28
G1:13 G2:60 G3:21
ER status positive
144 (88%)
57 (85%)
90 (96%)
Her 2 status positive
10 (6%)
16 (24%)
14 (15%)
16
17
Patients with <3 nodes positive at the

150 (92%)
SLNB
50 (75%)
72 (77%)
Patients with <3 nodes positive in total

after completion ALND
40 (60%)
58 (62%)
Median nodes removed

Median positive nodes
Total number of patients 324

Conclusions: Breast cancer patients with a preoperative diagnosis of nodal metastasis are treated in a different way than the
patients with positive SLNB. Axillary treatment may be improved if we could identify patients with a preoperative diagnosis of
nodal metastasis that fulfil the criteria of the Z11 trial and so potentially save patients unnecessary axillary surgery.

Title: Internal mammary sentinel lymph node biopsy with modified injection technique: High visualization rate and accurate
staging
Peng-Fei Qiu1, Bin-bin Cong1, Rong-Rong Zhao1, Yan-Bing Liu1, Guo-Ren Yang1, Peng Chen1 and Yong-Sheng Wang1.
1
Shandong Cancer Hospital&Institute, Jinan, Shandong, China.
Body: Background: Although the 2009 AJCC incorporated the internal mammary sentinel lymph node biopsy (IM-SLNB) concept,
there has been little change in surgical practice patterns due to the low internal mammary sentinel lymph nodes (IM-SLNs)
visualization rate (VR) with the traditional radiotracer injection technique (average 13%, 0-37%). In this study, various injection
techniques were evaluated in term of the IM-SLNs VR in clinically axilla lymph nodes (ALNs) negative patients, and the impact of
IM-SLNB on the diagnostic and prognostic value were analyzed both in clinically ALNs negative (NCT01642511) and positive
patients (NCT01668914).
Methods: 340 patients with T1-2 invasive breast cancer were enrolled. Clinically ALNs negative patients (n=293) were divided into
3 groups according to the study period and radiotracer injection technique. Group A: traditional technique (peritumoral
intraparenchymal injection) for the initial 58 cases; Group B: periareolar intraparenchymal injection under the ultrasonography
guidance in the latter 235 cases, the injection sites were chosen at the 6 and 12 oclock positions 0.5-1.0 cm from areola (about
2.0-4.0 cm from the nipple). Group B was then separated into 2 groups according to the radiotracer injection volume: Group B1,
low volume (<0.5ml/point, n=41); Group B2, high volume (0.5ml/point, n=194). Clinically ALNs positive patients (n=47) were
managed as group B2. IM-SLNB was performed for patients with IM-SLNs visualized on preoperative lymphoscintigraphy and/or
detected by intraoperative gamma probe.
Results: Clinically ALNs-negative: The IM-SLNs VR was significantly higher by lymphoscintigraphy in Group B than that in Group
A (63.0% vs. 13.8%, P<0.001), while the axillary VR was similar in these two groups (84.3% vs. 77.6%, P=0.227). The VR of
sentinel lymph nodes was improved by the intraoperative gamma probe compared to lymphoscintigraphy: Group A (axilla:
77.6%98.3%, P=0.001; internal mammary: 13.8%15.5%, P=0.794) and Group B (axilla: 84.3%98.7%, P<0.001; internal
mammary: 63.0%70.6%, P=0.078). Group B2 was found to have the highest IM-SLNs VR (74.2% vs. 53.7% Group B1,
P=0.009). The successful rate of IM-SLNB was 95.8%, and arrived 100% after 20 cases learning curve. In the patients who
underwent IM-SLNB, the IM-SLNs metastases rate was 8.7%, systemic treatment was changed only in 4.3%; however,
radiotherapy treatment was changed in 8.7%. For the patients with upper inner quadrant tumor, systemic and radiotherapy
treatment was changed in 8.8% and 20.6%.
Clinically ALNs-positive: IM-SLNB was performed in all patients with drainage to IM-SLNs (VR: 70.2%, 33/47); the successful rate
was 100 % (33/33). The IM-SLNs was positive in 21.2% patients, and the internal mammary radiotherapy could be guided with
this IM-SLNB results.
Conclusions: The modified injection technique (periareolar intraparenchymal, high volume and ultrasonography guidance)
significantly improved the IM-SLNs VR, making the routine IM-SLNB possible in daily practice. IM-SLNB could provide individual
staging, prognosis, and decision-making of the internal mammary radiotherapy for breast cancer patients, especially in positive
ALNs and high-risk negative ALNs patients.

Title: Validation of diagnostic procedure for metastatic lymph nodes in breast cancer using a semi-dry dot-blot method and novel
anti-cytokeratin 18+19 antibodies
Ryota Otsubo1, Masahiro Oikawa1, Hiroshi Hirakawa2, Hiroshi Yano1 and Takeshi Nagayasu1. 1Nagasaki University Hospital,
Nagasaki, Japan and 2Chiba Aiyuuukai Memorial Hospital, Chiba, Japan.
Body: Introduction: Sentinel lymph node (SLN) biopsy is a common diagnostic procedure for breast cancer. However, because of
a shortage of pathology specialists in Japan and discordance between intra-operative and final pathological diagnoses of SLN
metastasis, new diagnostic modalities are desperately required. We previously reported a novel method of detecting metastasis in
SLNs by a semi-dry dot-blot (SDB) method with 93.3% sensitivity, 96.9% specificity and 96.6% accuracy. Here, we evaluated the
efficacy of the SDB method using novel anti-cytokeratin (CK) 18+19 antibodies to diagnose lymph node metastases in breast
cancer.
Materials and methods: We obtained 73 lymph nodes dissected from 43 patients with breast cancer from July 2013 to May 2014
at Nagasaki University Hospital and the Japanese Red Cross Nagasaki Genbaku Hospital, including 55 sentinel lymph nodes and
18 dissected axillary lymph nodes, which were sliced at 2-mm intervals and washed with phosphate-buffered saline. This lavage
fluid was used to diagnose lymph node metastasis by the SDB method; whereas the washed lymph nodes were blindly diagnosed
by pathologists using with hematoxylin and eosin (H&E) stain. Suspended cells in the lavage fluid were centrifuged; the cell pellet
was lysed with lysis buffer to extract protein, which was then challenged and visualized with anti-cytokeratin 18 and 19 antibodies,
each at 1g/ml and 0.1g/ml to distinguish between micrometastases or isolated tumor cells (ITC) and macrometastases; and
with chromogen on a dot-blot membrane. Diagnoses based on the SDB method were compared with their H&E counterparts.
When the SDB method and H&E-based examinations did not agree, we examined specimens immunohistochemically with
anti-cytokeratin18+19 antibodies.
Results: Of the 73 lymph nodes, 25 were assessed as positive and 48 as negative by the permanent pathological examination
with H&E. The SDB method made correct diagnoses in all positive cases and 42 of the 48 pathologically negative cases. We
found four micrometastases and one ITC in the positive cases, which were difficult to diagnose as positive at the lower antibody
concentration, but were clearly positive at the higher concentration. When the 6 discrepant cases were examined
immunohistochemically, we found two cases of ITC and one of micrometastasis. Sensitivity, specificity and accuracy of the SDB
method in detecting cancer cells were 100% (95% confidence interval [95% CI]: 100100%), 93.8% (95% CI: 86101%) and
95.9% (95% CI: 91100%), respectively.
Conclusions: The SDB method using anti-CK18+19 antibodies is simple and accurate for diagnosing lymph node metastases;
estimating metastatic amount may be possible with different antibody concentrations. We are producing an SDB kit that uses
these antibodies.

Title: Sentinel lymph node biopsy using intraoperative indocyanine green fluorescence imaging navigated with preoperative
computed tomography lymphography for early breast cancer
Hajime Abe1, Keiichi Yamazaki1, Masao Ogawa1, Masayasu Kawasaki1, Kohri Yoneda1 and Masao Kameyama1. 1Bell Land
Genaral Hospital, Sakai, Japan.
Body: Background: Sentinel lymph nodes (SLN) biopsy has been established as a standard of care in the treatment of early
breast cancer. The combination of the radioisotope and dye-staining methods is the most accurate way to identify sentinel lymph
nodes. We had reported a novel technique of SLN identification using fluorescence imaging of indocyanine green (ICG) injection.
However, if any lymphatic vessels are injured, further fluorescence navigation will be difficult because of ICG contamination in the
surgical field in overweight or obese patients. In this study, a new diagnostic approach for imaging lymphatic drainage and
identifying SLN using preoperative computed tomography-lymphography (CTLG) and an intraoperative ICG fluorescence method
was investigated.
Patients and method: Between January 2013 and April 2014,105 breast cancer patients without clinical evidence of lymph node
metastasis were treated. On the day before the operation, CTLG was performed using 64-row multidetector CT. We performed an
intradermal injection in the periareolar area, using 4 ml of contrast agent with 0.5 ml of local anesthetic. The contrasted lymph
flow and SLN were identified in reconstructed three-dimensional imaging. The SLN spot was indicated by CT laser light navigator
system. During the operation, fluorescence images were obtained using the fluorescence imaging system, Photpdynamic Eye
(pde-neo, Hamamatsu Photonics Co., Japan). After ICG was injected intradermally in to the periareolar skin, lymphatic drainage
was observed with fluorescence images. SLN biopsy was performed by referring to the marker preoperatively placed on the
CTLG. Moreover, we classified enhancement of SLN as "whole" and "partial" by CTLG with visual patterns, and examined the
relation with SLN metastasis.
Results: The median age of the 105 patients was 63 (range 33 87) years old. CTLG and fluorescence imaging was safely
performed in all patients. This method was visually easy to identify the location of SLN on the axillary skin even in obese patients.
CTLG could visualize lymphatic flow and accurately identify SLN in 99 (94%) of 105 patients, whereas fluorescence imaging
identified successfully lymphatic flow and SLN in all patients. Lymphatic flows of CTLG were completely consistent with those of
fluorescence imaging. The number of SLN identified by CTLG was significantly lower than that by fluorescence imaging (1.1 vs.
1.6, p<0.01). Fifteen patients (14%) were found to have lymph node metastases pathologically, and five of them had
micrometastases of lymph node. In 6 patients without detected SLN by CTLG, only one patient had lymph node metastasis. In
case of partial enhancement of SLN with CTLG, the rate of positive metastasis was significant higher compared to the cases of
whole enhancement (p<0.01).
Enhancement of SLNs by CTLG
Metastasis(+)
Metastasis(-)
Whole
59
Partial
10
26
Not identified
Conclusion: This new navigation method of CTLG and fluorescence imaging revealed more easy and effective to detect SLN
intraoperatively than fluorescence imaging alone. In addition, the information from CTLG may be helpful for the prediction of SLN
metastasis.

Title: Not all sentinel lymph nodes are equal A predictive model for axillary burden in early breast cancer
Yirong Sim1, Sue Zann Lim1, Shaun S Tan1, Alvona Z Loh1, Cindy Lim2, Preetha Madhukumar2, Gay H Ho2, Veronique KM Tan2
and Kong Wee Ong2. 1Singapore General Hospital, Singapore, Singapore; 2National Cancer Centre, Singapore, Singapore and
3
National University of Singapore, Singapore, Singapore.
Body: Introduction
The Z11 trial demonstrates that an axillary lymph node dissection (ALND) can be avoided in patients with low axillary burden who
undergo breast conserving surgery (BCS), combined with whole-breast radiotherapy and systemic therapy. Our group propose a
standardisation of sentinel lymph node biopsy (SLNB), incorporating 2 novel sentinel nodal stations (SNS) which represent
sequential echelons of SLN draining the breast the intercostalbrachial nerve (ICB) and the medial pectoral neurovascular
bundle (MP). By increasing the specificity and positive predictive value of SLNB, we aim to identify a subgroup of patients who
undergo mastectomy who can be spared from ALND and its associated complications.
313 female patients who underwent sentinel lymph node biopsy for breast cancer from 2 February 2012 to 19 December 2013
were prospectively studied. 12 patients had bilateral breast cancers, and each laterality was counted as distinct cases, giving a
total of 325 cases. 7 had breast surgery performed prior to the SLNB, and the rest had their oncological surgery performed at the
same setting of the SLNB. The surgeries were performed by three surgeons in the National Cancer Centre Singapore, using a
specific surgical technique to identify SLN at the 2 SNS. Relevant patient demographics, status of SNS and the rates of
metastatic non-sentinel lymph nodes were collected and analysed.
Results
A total of 325 SLNBs using the ICB and MP SNS were identified from 169 simple mastectomies, 35 skin sparing mastectomies,
and 129 breast conserving surgeries. The median age was 56 (range 27-89). The ICB SLN and MP SLN were identified in 313
(96.3%) and 258 (79.4%) cases respectively. In 249 (76.6%) cases were both ICB and MP nodes identified, of which 55 (16.9%)
had metastatic involvement of the SLN.
An axillary clearance was performed if at least one ICB or MP node was positive, and only 27 (49.1%) had further axillary
involvement. More than 2 positive SLNs had 100% positive predictive value for further axillary LN metastases. There was a low
sensitivity (29.6%) and high false negative rate (70.4%) for positive axillary nodes in patients with 2 positive SLNs.
MP nodal status, however, was 85.7% specific (p<0.001) and 48.1% sensitive (p=0.649) and a positive predictive value of 76.5%
for axillary nodal involvement. Logistic regression also shows that MP node status is significant for predicting axillary nodal status
(OR 5.57, p=0.006).
Conclusion
Our study shows that MP node status is specific, and has a positive predictive value for further axillary LN metastases. Therefore,
we propose that in all patients who undergo SLNB with their BCS or mastectomy, an axillary clearance should be performed if the
MP node is positive, regardless of the number of positive SLNs.

Title: Role of age, menopausal status and biological tumor characteristics on sentinel lymph node metastasis in early breast
cancer patients with favorable prognostic features: A retrospective, mono institutional study on 345 cases
Nicla La Verde1, Chiara Casiraghi2, Chiara Dazzani1, Irene Floriani3, Elena Biagioli3, Andrea Cordovana2, Chiara Gerardi3,
Gabriella Farina1, Anna Maria Croce4, Filippo Bianchi4, Claudio Lunghi2, Mauro Lamera2, Marcello Bonavita5, Elena Bernardin2
and Giorgio Gherardi4. 1A.O. Fatebenefratelli e Oftalmico, Milan, Italy; 2A.O. Fatebenefratelli e Oftalmico, Milan, Italy; 3Laboratory
of Clinical Research, IRCCS Istituto di Ricerche Farmacologiche Mario Negri , Milan, Italy; 4A.O. Fatebenefratelli e
Oftalmico, Milan, Italy and 5A.O. Fatebenefratelli e Oftalmico, Milan, Italy.
Body: Background: Sentinel lymph node biopsy (SLNB) is recognized as a standard procedure for women with early breast
cancer. Recently some studies reported the futility of axillary lymph node dissection in patients (pts) who are SLN positive but
bear favorable clinical and primary tumor biological characteristics. Since we believe that the choice of a proper axillary treatment
in early breast cancer could be personalized and axillary SLNB could be avoided in specific subgroups of pts, we designed this
study to identify key primary tumor characteristics and pts' clinical features that could influence the indication to perform SLN
biopsy.
Patients and methods: Retrospective analysis was carried out in women who had undergone surgery and SLNB for early breast
cancer (pT1-2) from 2005 to 2013 at the Fatebenefratelli e Oftalmico Hospital in Milan, Italy. All SLNs were examined
histologically in toto on seriated permanent sections. The association between SLN positivity and the following variables as age,
menopausal status, tumor size, histological grading, presence of extensive "in situ" components and lymphovascular invasion
(LVI), quantitative evaluation of Ki-67, HER2 expression, oestrogen and progesterone receptors, was assessed by means of
univariate and multivariate logistic models.
Results: The records of 345 pts with early breast cancer who underwent surgery were retrieved. Mean age was 61 years and
79% pts were postmenopausal. 85% were treated with quadrantectomy and 66% had only one SLN removed. SLN metastasis
was detected in 24% of cases. Tumor size was <2 cm in 76% of pts, and 86% were of luminal subtypes. Peritumoral LVI was
detected in 32% of cases. At univariate analysis a statistically significant association was found between tumor size [odds ratio
(OR) 1.05, confidence interval at 95% (95%CI) 1.01-1.08; p=0.005], histological grade (OR 1.50, 95%CI 1.04-2.16; p=0.029),
presence of LVI (OR 3.81, 95%CI 2.27-6.37; p < .0001). At multivariate analysis only LVI confirmed to increase the risk of SLN
positivity in the whole series (OR 3.26, 95%CI 1.89-5.64; p < .0001) as well as in subgroup of pts with luminal A and B subtype
(OR 3.51, 95%CI 1.92-6.44; p < .0001). Negative and positive predictive values of LVI were 83.9% and 42.2%, respectively.
Interestingly, in a pre-planned subgroup analysis according to menopausal status, an association between tumor dimension and
SLN positivity was found in premenopausal women (OR 1.1, 95%CI 1.01-1.21; p < .034), while in postmenopausal pts LVI was
associated with SLN positivity (OR 3.05, 95%CI 1.65-5.64; p < .0004).
Conclusions: Our data suggest that in a population with favorable early breast cancer (luminal subtypes, postmenopausal status
and small tumour size) LVI increases the risk of SLN metastasis. These results are confirmed in luminal subgroup. In addition the
absence of LVI has a significant negative predictive value. As far as menopausal status is concerned we found that the positivity
of SLN is associated with tumor dimension in premenopausal and with LVI in postmenopausal women.

Title: In patients with micrometastatic in sentinel lymph node biopsies, involvement of the non-sentinel lymph nodes cannot be
predicted by clinicopathological variables
Muhammad N Chauhan1, Rajiv V Dave1, Mauria Ghaus1, Sana Ahmad1, Craig Sayers1, Zbigniew Kryjak1 and Deedar Ali1.
1
Pindefields Hospital, Aberford Road Wakefield, West Yorkshire, United Kingdom; 2Pindefields Hospital, Aberford Road
Wakefield, West Yorkshire, United Kingdom; 3Pindefields Hospital, Aberford Road Wakefield, West Yorkshire, United Kingdom
and 4Pindefields Hospital, Aberford Road Wakefield, West Yorkshire, United Kingdom.
Body: Background: The Sentinel Lymph Node Biopsy (SLNB) procedure is recognised to be an accurate method of staging the
axilla in patients with early stage breast cancer. There remains a debate as to whether patients with micrometastases should
undergo completion axillary lymph node dissection (CALND). We aimed to assess the indicators for positive non-sentinel lymph
nodes (n-SLN) following CALND.
Methods: We retrospectively analysed our experience of SNLB between July 2008 to July 2013. A total of 1152 breast cancer
patients underwent SLNB based on lymphoscintigraphy, intra-operative gamma probe detection, and blue dye mapping using
99mTc-nanocolloid and Patent Blue V injected peri-areola. Statistical analysis was performed using Fishers exact and c2 for
categorical data.
Results: Out of 1152 SLNB biopsies performed, 224 (19.5%) were positive for metastatic disease; macrometastases in 150
(67%), micrometastases in 72 (32%) and ITC in 2. CALND was not performed in 20 cases (9 macrometastases, 10
micrometastases, and 1 ITC), largely due to concerns regarding fitness for anesthesia. Macrometastases on SNLB were more
likely to predict positive n-SLN on ANC {macrometastases; 39/141(27.7%) vs micrometastases; 9/62 (14.5%), p=0.029}. On
univariate analysis, positive n-SLN in CALND for patients with micrometastases on SLNB was not predicted by grade (G0-G2;
6/43, and G3; 3/19, p=0.565), size of primary breast tumour (<40mm; 8/58, 40mm; 1/4, p=0.475), lymphovascular invasion (5/30
vs 4/31, p=0.503), age (<50 years; 3/24 vs 50 years; 6/38, p=0.496), or number of positive SLNB (all patients had <2 positive
nodes on SLNB). Recurrences were detected in 4 patients, of which 1 was in a patient with micrometastases on SLNB. Out of the
4 recurrences, 3 were distant (liver and bone) and one was locoregional, with new disease in the contralateral breast).
Conclusion
In our series, 14.5% (9/62) of patients with micrometastases had positive n-SLB on CALND, which was not predicted by any
clinicopathological characteristics. We have recently changed our practice toward not routinely offering CALND in patients with
micrometastases, in keeping with current vogue. However, it is important to inform our patients that 14.5% of patients with
micrometastases on SLNB may have positive n-SLN.

Title: Prognosis of metastatic internal mammary sentinel nodes (IMSN) in breast cancer
Eli Avisar1, Shai Libson2 and Eduardo Perez1. 1University of Miami Miller School of Medicine, Miami, FL and 2Ben Gurion
University, Beer Sheva, Israel.
Body: Introduction:
The presence of positive internal mammary nodes has historically been associated with a significant worse prognosis. Recent
studies have also demonstrated a worse prognosis associated with drainage to the internal mammary nodes. Intense search for
those nodes, biopsy, appropriate staging and modern treatment for positive IMSN might improve outcome. We sought to study
the prognosis of pathologically positive IMSN at our institution.
Methods:
A retrospective analysis of a prospectively collected database including all breast cancer patients identified with an IMSN that was
biopsied between 2005 and 2012 was performed. Demographics, histologic markers, patterns of recurrence as well as survival
data were collected. Univariate and multivariate analysis were performed.
Results:
Thirteen of 82 patients with a biopsied IMSN were metastatic (16%). In 8 of those (62%), the biopsy resulted in a change in
staging. In all cases of positive IMSN additional radiation to the internal mammary chain was added. Eighteen percent (18%) of
positive nodes did not show drainage to the internal mammary basin during Lymphoscintigraphy. There was no statistical
difference in regional and distant recurrences between the patients with positive IMSN and those with negative IMSN.
Furthermore there was no difference in disease specific survival.
Conclusions:
Intense search for IMSN presence and biopsy of those nodes is associated with changes in staging and treatment for metastatic
IMSN. In our study, pathologically positive IMSN were associated with a non- inferior prognosis than negative IMSN.

Title: Axillary involvement in lobular breast cancer
Pilar Zamora1, Covadonga Marti1, Ana Roman1, Jose M Oliver1, Javier de Santiago1 and Jose I Sanchez-Mendez1. 1Hospital
Universitario La Paz, Madrid, Spain.
Body: Purpose:
Since the beginning of the use of the technique of selective sentinel node biopsy (SLNB), multiple models have been proposed to
predict both the probability of involvement of the sentinel lymoh node (SLN), and, in case it is positive, the likelihood of
non-sentinel lymph node (NSLN) involvement.
Some of the factors that increase the likelihood of SN involvement (multifocallity, hormonal receptors, larger size)are frequently
found in invasive lobular carcinoma.
Invasive lobular carcinoma (ILC) is the second most common histologic type of invasive mammary carcinoma, comprising 5%
15% of all invasive breast carcinomas. However, in recent decades an increase in the relative incidence of this variant has been
reported in the literature. This has been linked to the employment of hormone replacement therapy, the use of assisted
reproduction techniques (IVF) or greater availability of diagnostic tools such as breast ultrasound or MRI that allow their better
detection.
The aim of this study is to determine the extent of axillary involvement in both SLN and NSLN in patients with ILC
Patients and Methods
369 cases of infiltrating carcinoma candidates for SLNB technique between April 2010 and April 2013 were reviewed
retrospectively.
Patients must have a diagnosis of breast cancer of any histological type, with a screening ultrasound without clinical suspicion of
axillary involvement.
Cases of intraductal carcinoma that underwent SLNB were excluded and even those that were performed prior to neoadjuvant
treatment
Data on age, histological type, positivity / negativity of the SLN, implementation or not of axillary lymphadenectomy and outcome
of involvement of NSLN were collected.
Statistical calculations were performed with SPSS.
Results
Of the 369 selected patients, 291 (79.9%) had invasive ductal carcinoma (IDC), 55 (15.1%) ILC and 18 other histological types
(mucinous, colloid, tubular, papillary or mixed). The median age was 58.6 years.
SLN could not be found in 5 patientes. 225 (61.8%) had a negative SLN. In 139 (38.2%) patients, SLN was positive, 64 of these
cases showed micrometastasis while 75 presented macrometastasis.
Among IDC, 108 (37.1%) of the cases had a positive SLN, 53 (49.1%) of them with macrometastases.
Among ILC, 27 (49.1%) of the cases presented positive SLN, in 20 of them (74.1%) with macrometastasis.
Differences in both the SLN involvement and the presence of macrometastases observed between IDC and ILC were statistically
significant (p< 0.001). In those cases were a lymph node dissection was performed, the probability of finding NSLN affection was
18.8% in the case of IDC compared to 55% in patients with ILC (p< 0.001).
Conclusions
Axillary lymph node involvement in the case of candidates for SLNB is more frequent and more extensive in patients with ILC
than in patients with IDC.

Title: A new nomogram to predict axillary metastasis in breast cancer patients without axillary surgery
Valery Rodionov1,2, Vlada Cometova1,2, Sergey Panchenko1,2, Sereda Idrisova1,2, Yurij Savinov1 and Maria Rodionova3. 1Ulyanovsk
State University, Ulyanovsk, Russian Federation; 2Ulyanovsk Regional Clinical Cancer Center, Ulyanovsk, Russian Federation
and 3N.N. Blokhin Russian Cancer Research Center, Moscow, Russian Federation.
Body: Background: Several studies have concentrated on finding a combination of predictive parameters to establish a
mathematical model that can identify patients with no axillary metastasis for whom routine lymph node dissection could be safely
avoided. We developed a new model of nomogram (the Ulyanovsk Cancer Center axillary lymph node metastasis nomogram,
UCC-ALNM nomogram); it employs clinically and pathologically relevant variables and offers possible advantages over the others
nomograms.
The purpose of the study: To assess the predictive power of UCC-ALNM nomogram.
Methods: A total of 530 breast cancer patients treated between 2008 and 2010 were used as the modeling group for validating
the UCC-ALNM nomogram. Clinical and pathologic features of patients were assessed by multivariable logistic regression to
predict the presence of axillary metastasis in breast cancer patients. The predictive accuracy of our nomogram was measured by
calculating the area under the receiver-operating characteristic (ROC) curve (AUC). Clinical factors included into analysis were:
patients age and localization of the primary tumor. Pathological factors evaluated were: traditional pathological criteria (primary
tumor size, histological type, tumor grade, HR- and Her-2 status) and new total pathological index (Ulyanovsk prognostic index UPI), introduced by pathologists of the Ulyanovsk Regional Cancer Center. UPI is total score of six main pathological criteria that
characterize the malignancy of epithelial tumors: degree of cellular differentiation, cellular polymorphism, mitotic activity, growth
pattern, lymphovascular invasion, stromal reaction.
Results: By the multivariate analysis, patients age (p=0.04), tumor size (p<0.001), UPI (p<0.001), PR (p<0.001) and Her2 status
(p=0.02) were identified as independent predictors of axillary metastasis. The nomogram was then developed using the six
variables associated with axillary metastasis: age, tumor size, PR, Her2, UPI.
The new model was accurate and discriminating with an AUC of 0.7510 when applied to the modeling group.
Conclusions: UPI is a new predictive factor of axillary metastasis in breast cancer patients. UCC-ALNM nomogram.

Title: Sentinel node mapping with fluorescein and comparison with methylene blue and technitium sulphur colloid in early breast
cancer
Anurag Srivastava1, Rajendra A Badwe2, Amar Prem1, Vani Parmar2, Vuthaluru Seenu1, Anita Dhar1, Nita S Nair2, Rohini
Hawaldar2 and Vaibhav Sanmali2. 1AIIMS, New Delhi, Delhi, India and 2Tata Memorial Hospital, Mumbai, Maharashtra, India.
Body: Background: Sentinel lymph node biopsy is currently standard of care in node negative early breast cancer. There are
various tracers for detecting sentinel lymph nodes in breast cancer. Sentinel lymph node biopsy (SLNB) using Technetium tagged
Sulphur colloid is a convenient and safe method to assess lymph node status. However, sulphur colloid is radioactive and its use
needs gamma camera which is very costly and available in very few centers in India. Methylene blue dye is an economical
alternative for sulphur colloid having identification rate similar to the isotope. However, it may react with hemoglobin forming
meth-hemoglobin resulting in difficulty in pulse oximetry during operation and causes skin eruptions and necrosis. Hence, there is
a need to identify a sensitive, inexpensive and safe dye for sentinel node biopsy. In this study we have investigated the
effectiveness and safety of fluorescein in sentinel node biopsy in a cross-sectional analytical study compared to Methylene Blue
and Technitium Sulphar colloid.
Methods: This trial was conducted at two centers : Tata Memorial Centre, Mumbai and All India Institute of Medical Sciences New
Delhi in India.We examined 86 patients of early breast cancer with no palpable axillary nodes undergoing SLNB with three tracers
(Sulphur colloid /Methylene blue / Sodium Fluorescein). Patients underwent complete axillary dissection after identification of
sentinel node for validation of sentinel node biopsy. Hot nodes were identified using hand held gamma probe and fluorescent
nodes were identified using Ultra Violet lamp. All nodes were examined with Haematoxylin and Eosin staining. Since this study is
first with use of fluorescein for sentinel node biopsy in breast, we injected fluorescein at decreasing time interval from 12 hours to
5 minutes to ascertain the most appropriate time for injection of fluorescein in first 15 patients. The sentinel nodes were visualized
only when fluorescein was injected 5 minutes before incision.
Results: Sentinel nodes could be identified in eighty out of eighty six patients by combined tracers. In 72 out of 86 patients hot
nodes could be identified (83.7%) while in 54 out of 86 patients fluorescent nodes could be identified (62.8%). Blue nodes were
identified in 64 out of 86 patients(74.4%). Fluorescein delineated the lymphatic pathway going from breast to axilla in most
patients. No side effects of Fluoroscein were observed. Of 86 cases, histologically positive sentinel lymph node were found in 15
patients. False negative with combined three tracers was in 3/18 (16.6%) patients. False negative rate with individual tracers was
sulphur colloid- 3/16 (18.8%), Fluorescein 2/15(13.3%) and with Methylene blue 2/17(11.8%).
Conclusions: Fluorescein can be used as a low cost and effective alternative in sentinel lymph node biopsy for carcinoma breast
without the risk of radiation exposure. Moreover, excellent visualisation of lymphatic pathway during procedure can act as a guide
to trace the sentinel node.

Title: Comparison between positive and false-negative sentinel lymph nodes in breast cancer patients
Eriko Abe12, Naoki Hayashi3, Toshinao Onoda23, Yang Yang1, Mieko Uno1, Hideko Yamauchi3, Sachiko Ohde4 and Koyu Suzuki1.
1
St Luke's International Hospital, Tokyo, Japan; 2Yokohama Asahi Central General Hospital, Yokohama, Japan; 3St Luke's
International Hospital, Tokyo, Japan and 4St Luke's Life Science Institute Center for Clinical Epidemiology, Tokyo, Japan.
Body: Background:
Intra-operative sentinel node biopsy (SNB) is performed for clinically node negative invasive carcinoma (IC). Despite a negative
result for metastasis from rapid frozen section (FS) diagnosis of intra-operative SNB, some cases may be diagnosed positive
when permanent sections (PS) are analyzed (i.e., false negative). In order to reduce the false-negative rate, it is necessary to
determine why macrometastases measuring greater than 2 mm cannot be accurately diagnosed using frozen specimens. The
aim of this study was to compare the pathological characteristics of false-negative and positive cases of macrometastasis using
FSs. We also reviewed whether there were differences in postoperative prognoses between positive and false-negative cases.
Methods:
We retrospectively reviewed 1632 intra-operative SNBs collected from 2008 to 2011 at St. Lukes International Hospital, Tokyo,
JAPAN. Of these, 980 patients had undergone surgery for the treatment of IC without neoadjuvant chemotherapy. Lymph nodes
were sectioned every 2 mm and examined. For FSs, we performed hematoxylin and eosin (HE) staining, and for PSs, we used
HE and cytokeratin (AE1/AE3) staining. Image J (NIH Image, Bethesda, MD, USA) was used for assessment of the metastatic
area of lymph nodes. Micro- and macrometastases were defined as metastatic lesions measuring between 0.2 and 2 mm or
measuring 2 mm or more, according to TMN classification.
Results:
Using FSs, we identified 104 patients (10.6%) who were positive for macrometastasis, 16 patients (1.6%) who were positive for
micrometastasis, and 860 patients (87.8%) who were negative for metastasis. Of the negative cases, the result was changed to
micrometastasis in 37 cases (4.3%) and to macrometastasis (false-negative cases) in 14 cases (1.6%). Ten of the 14 patients did
not have additional axillary lymph nodes dissection. The mean age of the patients was 55.1 3.1 years for false-negative cases
and 52.3 1.2 years for positive cases. In terms of histological features, there were statistically significant differences between
false-negative and positive cases for invasive ductal carcinoma (IDC), invasive lobular carcinoma (ILC), and IDC plus ILC (p <
0.05). There were no statistically significant differences between false-negative and positive cases in terms of the total number of
slices for lymph nodes (8.4 1.6 vs. 9.7 0.6, respectively; p = 0.41) but tendency in the number of slices for metastases (2.4
1.2 vs. 4.8 0.5, respectively; p = 0.07). The maximum area ratio of metastasis was significantly lower in false-negative cases
than in positive cases (3.0% 0.1% vs. 20.0% 0.2%, respectively; p < 0.05). There was no significant difference in disease-free
survival between false-negative and positive cases (p = 0.43).
Conclusions:
Although it is necessary to reduce false-negative cases, no difference in prognosis was detected in our study. False-negative
cases had lower metastasis area ratios than positive cases diagnosed using rapid diagnosis of intra-operative SNBs, and ILC is
known to be high in histological features. Therefore, using AE1/AE3 stain together with rapid diagnosis of intra-operative SNBs
may lower the false-negative rate when a diagnosis includes ILC in pre-operation analysis.


Title: No Show

Title: Randomised controlled trial of stereotactic 11g vacuum-assisted or core biopsy for diagnosis and management of malignant
microcalcification
Sara M Bundred1, Jin Zhou1, Anthony Maxwell1, Sigrid Whiteside1, Julie Morris1, Janicke Harake2 and Nigel J Bundred1.
1
University Hospital of South Manchester NHS Foundation Trust, Manchester, United Kingdom and 2Royal Bolton Hospital,
Bolton, Lancashire, United Kingdom.
Body: Background
Vaccum-assisted biopsy (VAB) is claimed to improve diagnosis of malignant microcalcification (MM) and prevent underestimation
of invasive component compared to stereotactic core biopsy (SCNB) 14g biopsies. VAB is also reported to reduce requirement for
further surgical procedures versus SCNB.
Methods
In a randomised controlled trial, we compared both techniques in the firstline diagnosis of MM and subsequent surgical
management in two UK Breast Screening Units. Patients with MM were identified and gave written informed consent to the study
before randomisation 1:1 to SCNB or VAB. Exclusions included bleeding diathesis or significant comorbidity preventing surgery.
SCNB was performed by advanced radiography practitioners and VAB by experienced radiologists under local anaesthesia.
Quality of life was assessed using EORTC QLQ-BR23 questionnaire: Pre-biopsy, 2, 6, and 12 months post-randomisation.
The primary objective compared the final pathological diagnosis with the initial biopsy (VAB or SCNB) and required 110 patients
to be randomised to show a 25% improvement in diagnostic accuracy with VAB compared to SCNB with a 90% power.
Results
Three quarters of MM investigated was less than 15mm in size. Overall 787 women were investigated for eligibility. 476 did not
meet the inclusion criteria, 82 declined to participate and 100 met exclusion criteria. In total, 129 women were randomised. Both
groups were evenly matched for age 57.2 years (range 47-75) and lesion size but VAB took longer (p<0.001) and produced on
average, 4 more cores (12 versus 8) (p<0.01). Three (4.7%) VAB developed bleeding causing procedure abandonment. The PPV
for biopsy of MM was 30.2% and 1 patient withdrew before the procedure.
There was an 84% diagnostic accuracy of SCNB which precluded any effect on surgical outcome.
Analysis
VAB
Core SCNB
63
64
Accurate diagnosis
54 (86%)
54 (84%)
1.0
Subsequent therapeutic porcedures - 1
9 (56%)
17 (71%)
0.50
Subsequent therapeutic procedures - 2
7 (44%)
7 (29%)
0.50
DCIS upgrade to invasive
12.5%
8.7%
0.86
Both groups had significant Quality of Life (QOL) falls in Global Health (p<0.001) and Social Functioning over time (p<0.04) but
no overall difference between groups was seen. At 12 months, body image scores fell and fatigue increased in patients
undergoing surgery.
Conclusion
SCNB has equal accuracy as VAB in the firstline diagnosis of malignant microcalcification. Devicor provided the equipment to Dr
S.B. and Prof. N.B., to do this study.

Title: A review of the accuracy of wire localisation using digital breast tomosynthesis (DBT) in a prospective series of 81 patients
Simon DH Holt1, Khaldoun MY Nadi1, Amrita Gurung1, Helen R Williams1, Anita M Huws1 and Yousef M Sharaiha1. 1Prince Philip
Hospital, Llanelli, Carmrthenshire, United Kingdom.
Body: Introduction: Digital breast tomosynthesis allows identification of impalpable lesions in three-dimensions. We have devised
a technique for wire placement using a gridded plate that allows localisation of such lesions for subsequent surgical excision.
Method: 81 consecutive cases were studied prospectively between Jan 2012 and May 2014. The localisation is usually performed
by the operating surgeon under local anaesthetic shortly before the patient is transferred to the operating theatres. Our surgeons
consider the optimal position for the tip of the wire to be 1 cm beyond and 1cm to the side of the lesion placed using only the CC
or lateral projection. The patient is positioned with the aperture of a gridded plate over the lesion needing to be surgically
removed. A single DBT is acquired and the volumes reviewed. The lesion is identified and the depth in the breast recorded (Z
coordinate). The ideal position for the wire is then marked in that plane and the volume scrolled through to bring the calibrated
plate into focus. This identifies the X and Y coordinates, which are then recorded. The wire is then placed freehand in the line of
the beam and a further DBT performed to check the position of the wire without needing to reposition the patient. A specimen
x-ray is performed immediately following excision to confirm the target lesion is included. The results are verified at the
subsequent multidisciplinary team (MDT) meeting.
Results: Once the target is identified, the average time for the localisation procedure and check DBT was 4 minutes. This series
included both diagnostic (N=24, 30%) and therapeutic excisions (N=57, 70%). Patients presented both from screening (N=64,
79%) and symptomatically (N=17, 21%). The age range of the patients was 42-79y (mean 59y). 54 (67%) cases were undergoing
excision for calcifications, 15 (19%) for impalpable masses, 11 (13%) for radial scars or distortions and 1 for glandular
asymmetry. The final diagnoses following excision were IDC 34 (42%), DCIS 29 (36%), benign 13 (16%) and 1 case each of ILC,
spindle cell tumour, mucinous carcinoma, phyllodes and pleomorphic LCIS.
In all cases the target lesion was shown to be present in the specimen x-ray, confirmed on histology and ratified by the MDT.
Conclusions: Wire localisation using DBT is a quick and accurate method of localising impalpable lesions for surgical excision.

Title: No Show

Title: Are all small tumors low risk? Characterization of small invasive node negative breast cancers (BC) enrolled in the EORTC
10041/BIG 3-04 (MINDACT) trial
Konstantinos Tryfonidis1, Leen Slaets2, Giuseppe Viale3, Femke Snoo4, Laura Van't Veer5, Emiel Rutgers6, Martine Piccart7, Jan
Bogaerts2 and Fatima Cardoso8. 1European Organization for Research and Treatment of Cancer (EORTC), Brussels, Belgium;
2
European Organization for Research and Treatment of Cancer, Brussels, Belgium; 3European Institute of Oncology, University of
Milan, Milan, Italy; 4Medical Affairs, Agendia, Amsterdam, Noord-Holland, Netherlands; 5University California, San Francisco, CA;
6
Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands; 7Institute Jules Bordet, Brussels, Belgium and 8Breast
Unit, Champalimaud Cancer Center, Lisbon, Portugal.
Body: Background: Adjuvant systemic therapy for subcentimetric N0 BC is controversial. These tumors have good prognosis, but
a subgroup has high risk of relapse and requires adjuvant therapy. The best tool to identify this subgroup is unknown. We
clinically and biologically characterize pT1abN0M0 tumors from patients enrolled in MINDACT.Patients and Methods: All
MINDACT pts with pT1abN0 tumors were included. Descriptive statistical analysis included age, menopausal status, centrally
assessed tumor grade, Ki67, ER, PR and HER-2 status. Cut-off value for high Ki67 was 20%. IHC subtypes are per 2013 St.
Gallen consensus/ ESMO guidelines. Genomic tests (Mammaprint, Targetprint, BluePrint) were performed and their concordance
rate with central pathology evaluated. We report clinical (modified version of adjuvant online!) and genomic risk assessment
(MammaPrint) for these patients. Survival data will be reported after MINDACT primary analysis.Results:826 patients with
T1abN0 tumors were enrolled in MINDACT, [12.3% of all patients (6694) and 16% of all N0 tumors];310 (35%) patients were 60
years and 537 (65%) were postmenopausal. Most (710;86%) samples were tested by central pathology. Tumors were mainly
ductal (579;81.5%), ER + (650;91.5%), and PR+ (593;83.5%);44(6.2%)were HER-2 +; 520 (73.2%) had a low Ki67&174 (24.5%)
had high Ki-67 labeling index; 80 (11.3%) were grade 3 & 328 (46.2%) grade 2 tumors. According to Targetprint, 761/826 (92.1%)
tumors were ER+; 624/826 (75.5%) PR+; 45/826 (5.4%) HER2+. Concordance between Targetprint and central pathology was
higher for ER (kappa= 0.90) and lower for PR and HER2 (kappa=0.60 and 0.73 resp).Molecular subtype classification by IHC
surrogates showed 420/710 (59.2%) Luminal A tumors; 189/710 (26.6%) Luminal B; 36/710 (5.1%) Luminal B/ HER2+; 8/710
tumors (1.1%) HER-2 + and 37/710 (5.2%) TNBC. Using BluePrint 739/826 (89.5%) were Luminal; 29/826 (3.5%) HER-2 + and
58/826 (7.0%) basal. Combining BluePrint and MammaPrint 609/826 (73.7%) tumors were Luminal A&129/826 (15.6%) Luminal
B. Agreement between Mammaprint and Ki67 for distinction of luminal A vs B/HER-2 neg. tumors was low (kappa=0.31), similar
to the overall MINDACT results (kappa=0.35).Using clinical risk assessment 820/826 (99.3%) were low-risk (CL). Using genomic
risk assessment 624/826 (75.5%) were low risk (GL) and 201/826 (24.3%) were high-risk (GH). Overall, 624/826 (75.5%) were
both CL/GL and 196/826 (23.7%) were CL/GH. The percentage of discordant CL/GH cases among T1abN0 tumors (196/826,
23.7%) is higher than in all MINDACT cohort (592/6694, 8.9%) and among T1c-2-3N0 tumors (381/4462, 8.5%). T1ab tumors
were mainly 720/826 (87.2%) treated with BCS and radiotherapy; adjuvant treatment and compliance rate will be available at the
meeting.
Conclusions:1) Results show biological characteristics are relevant for T1abN0 BC& that size/nodal status could be insufficient
determinants of adjuvant systemic therapy. 2) A significant portion (23.7%) of such tumors, clinically classified as low risk, was
genomically high risk by MammaPrint.3) IHC surrogates don't always accurately reflect genomic molecular subtyping.


Title: Incidence of actionable genomic alterations in metastatic breast cancer: A prospective study
Raquel E Reinbolt1, Katlyn Tolliver1, Mahmoud Abdel-Rasoul3, Cynthia Timmers2, Bhuvaneswari Ramaswamy1, Rachel M
Layman1, Robert Wesolowski1, Maryam B Lustberg1, Ewa Mrozek1, Susan Ottman1, James Chen2, Charles Shapiro1, Michael C
Ostrowski2, Gustavo W Leone2 and Erin Macrae1. 1Ohio State University Comprehensive Cancer Center & The Stefanie Spielman
Comprehensive Breast Center, Columbus, OH; 2Ohio State University Comprehensive Cancer Center, Columbus, OH and 3Ohio
State University, Columbus, OH.
Body: Introduction: Foundation Medicine Incorporated (FMI) uses next generation sequencing (NGS) to simultaneously identify
oncogenic molecular events and match targeted therapies to these alterations. We conducted a prospective study of the FMI test
to describe the incidence of actionable genomic alterations found in advanced breast cancer patients.
Methods: This is a prospective, single center, single-arm trial in advanced breast cancer patients designed to assess FMI
testings feasibility and impact. Adult metastatic breast cancer patients with an estimated survival of 3 months were included,
had tumor sample available for testing, and were within 10 weeks of starting their current therapy line. A massively-parallel
sequencing platform (FoundationOne) used the patients tumor (FFPE tissue) to sequence for 236 cancer-related genes plus 47
introns of 19 genes often rearranged in cancer to high depth (>500x). Genomic alterations were categorized as actionable if
linked to an approved therapy in the solid tumor of study or another malignancy, a known or suspected contraindication to a given
therapy, or a clinical trial linked to the alteration. The number of genomic alterations was quantified; genes with > 1 alteration
were counted once.
Results: Forty-nine patients were consented to FMI testing. Forty patients (82%) successfully completed testing; analysis was
insufficient in 6 tissue samples, 2 failed DNA extraction, and 1 patients tissue was not received. Of the 8 unsuccessful tests,
sampling from the bone (50%) and liver (38%) were the most common sites of failure. Thirty-seven reports have been received;
27 (73%) samples were taken from metastatic tissues, 10 (27%) from the primary breast tumor. A total of 192 actionable
alterations were detected in tumors tested, with a median of 5 (range 1-11) alterations found per patient. Amplifications comprised
87 alterations, 105 were a base substitution, insertion/deletion or other copy number alteration. Of the 192 alterations, the TP53
gene was the most often altered [n=22 (11%)] and most frequent in ER/PR/HER2-neu negative breast tumors. Alterations in
PIK3CA [n=16 (8%)] and ZNF217 [n=8 (4%)] were the 2nd and 3rd most commonly altered genes. The NGS report provided 24 of
the 37 patients (65%) with a recommended FDA approved breast therapy; an additional 24 patients (65%) were suggested a
non-breast FDA approved therapy. At least 1 potential treatment or clinical trial was proposed to 36 patients (97%) with
everolimus (n=30) and temsirolimus (n=30) the most commonly recommended FDA approved interventions, and trametinib (n=6),
regorafenib (n=6), and pazopanib (n=6) the next most commonly suggested interventions. Patients had a median of 3 (range 0-7)
genomic alterations paired with a recommended clinical trial.
Conclusions: This prospective study shows NGS testing in advanced breast cancer patients is successful in over 80% of patients
screened. Two-thirds of patients were recommended a FDA approved therapy and nearly all patients were suggested at least one
potential clinical trial. Further studies to assess barriers to patients access to these recommended targeted therapies are
currently underway.

Title: Comparative analysis of somatic mutations in matched primary vs metastatic triple-negative breast cancers
Irene Cherni1, Maren K Levin2, Joyce A O"Shaughnessy2,3,4 and John D Carpten1. 1Translational Genomics Research Institute,
Phoenix, AZ; 2Baylor Sammons Cancer Center, Dallas, TX; 3Texas Oncology, Austin, TX and 4US Oncology.
Body: Background. Treatment of triple-negative breast cancer (TNBC) is clinically challenging due to disease heterogeneity and
aggressiveness which often result in high recurrence and mortality rates. We have previously reported on genomic alterations
found in 14 metastatic TNBC tumors (Craig et al 2012) uncovered through whole-genome and transcriptome sequencing and
found that majority of tumor changes in our cohort converged on RAS/RAF/MEK/ERK and PI3K/AKT/mTOR pathways revealing
potential therapeutic targets. In this study, we interrogated patients matched primary and metastatic cancers using exome
sequencing. Differentiating between primary and metastasis-specific genomic alterations will further our understanding of TNBC
recurrence.
Methods. Tumor tissue was microdissected from 10uM FFPE sections from primary tumors of nine TNBC patients. DNA was
extracted using Qiagen AllPrep kit. Whole-genome libraries were prepared from >250ng of DNA using Kapa-onBead method.
Exomes were captured using All Exon V5 capture kit (Agilent Technologies) and sequenced on Illumina HiSeq. List of somatic
variants (SNVs) in primary tissue was generated using Seurat variant caller and manually compared with the SNVs uncovered
previously in metastatic tumors. Presence (or absence) of all reported somatic SNVs was visually confirmed in IGV genome
viewer.
Results. Our analysis revealed the majority of somatic mutations found in metastatic lesions were also present in the primary
tissue (64.3% on average). Patient TNBC03 had an unusually high number of new mutations in the lung metastatic lesion
(72.7%). All TP53 mutations in our cohort were an early event present in both primary and metastatic tissues. In patient TNBC01,
acquired mutation in NF1 - an upstream negative regulator of RAS pathway - in addition to pre-existing PTEN deletion, CTNNA1
underexpression and RB1 mutation - suggests RAS pathway addiction which could have been responsible for disease recurrence
in this patient. Interestingly, NEDD4 mutation was found in a lung metastasis of another patient whose disease was refractory to
BEZ235 (PI3K/mTOR inhibitor) treatment. NEDD4 has recently been shown to be an upstream regulator of the PI3K pathway and
a possible therapeutic target in cancers with downregulated PTEN. Additional select acquired mutations in metastatic lesions of
our cohort included FGF14, ABCB10, MDM2, KIF1C, ATAD2B, PTPLAD2, MDM20, TDRD1, and USP6. Mutated CDH5 was
found in metastases in 2 of 9 patients.
Conclusion. Our findings demonstrate the complexity and variability in somatic events underlying TNBC recurrence, reinforcing
the need to re-biopsy metastatic TNBC lesions and to employ combination targeted therapies wherever possible.

Title: The heterogeneous clinical behavior of luminal breast cancers is associated with different mutational landscapes
Giampaolo Bianchini1, Eugenia Galeota1, Tingting Jiang2, Maurizio Callari3, Milvia Zambetti1, Takayuki Iwamoto4, Thomas Karn5,
Christos Hatzis2, Lajos Pusztai2 and Luca Gianni1. 1IRCCS Ospedale San Raffaele, Milan, Italy; 2Yale School of Medicine, New
Haven, CT; 3IRCCS Istituto nazionale dei Tumori, Milan, Italy; 4Okayama University Hospital, Okayama, Japan and
5
Goethe-University, Frankfurt, Germany.
Body: Background: Genomic studies in estrogen-receptor (ER) positive breast cancer showed few commonly mutated genes and
suggested that tumors with higher background of mutation rate were enriched in intrinsic aromatase-inhibitor-resistant tumors
(Ellis MJ Nature 2012). Different combinations of proliferation (PROL) and estrogen-receptor related genes (ERS) allow the
prediction of the risk of recurrence and endocrine treatment sensitivity (Bianchini G Breast Cancer Res 2013). The
highPROL/lowERS is enriched in intrinsic endocrine resistant tumors. We explored whether different genomic alterations could
contribute to the different clinical behavior of groups defined by PROL and ERS.
Method: We defined luminal breast cancers by PAM50 and retrieved gene expression and whole-exome sequencing data of 324
luminal breast cancer cases from the Cancer Genome Atlas (TCGA). PROL and ERS were calculated as previously reported.
Low and high expression groups were defined by median cut-points (Bianchini G Breast Cancer Res 2013). As a metric of the
background mutation rate of each sample, we used the overall mutation frequency in whole exome (Kandoth C Nature 2013). The
most frequently mutated genes (TP53, PIK3CA, MAP2K4, MAP3K1, GATA3, CDH1) were also assessed for their association with
PROL and ERS defined groups.
Results: The average number of mutations (Nmut) was 44.4 (5-366). High and low proliferation groups had different mutation
frequency (average Nmut 53.6 and 35.3, respectively; p=6.1E-06). LowERS group also had a higher Nmut (average Nmut 54.4
vs. 34.5 in high ERS; p=1.2E-06). Notably, neither high or low expression of the ER gene (p=0.89), nor positive and negative
progesterone receptor by immunohistochemistry (p=0.69) were associated with Nmut. LowERS group was associated with higher
Nmut in luminal A (p=0.005) and B (p=0.006) subtypes. The group with highPROL and lowERS had the highest Nmut (average
62.1), which was not significantly different from basal-like tumors. HighPROL/lowERS was significantly associated with higher
frequency of mutTP53 (32.3%) and lower rates of mutPIK3CA (28.1%), mutMAP2K4 and/or mutMAP3K1 (7.2%) and mutCHD1
(3.1%). The lowPROL/highERS had the lowest Nmut (average 29.8), with the higher rate of mutPIK3CA (61.4%) and
mutMAP2K4 and/or mutMAP3K1 (29.2%), and lower rate of mutTP53 (6.7%).
Discussion: In ER-positive breast cancers, the highPROL/lowERS group, which is characterized by the highest risk of early
recurrences despite endocrine treatment and chemotherapy, poor down-regulation of Ki67 after neoadjuvant aromatase-inhibitor
(AI), but the highest rate of pCR after neoadjuvant chemotherapy, was also associated with the highest frequency of mutations, of
the same order of magnitude of basal-like tumors. This data is in line with the observation that the number of mutations is linked
to intrinsic resistance to neoadjuvant AIs (Ellis MJ Nature 2012). These data could be also informative for interpreting results of
the combination of everolimus and endocrine therapy (exemestane or tamoxifen) that suggest higher efficacy in tumors with
acquired endocrine resistance (lower mutation rate) compared to intrinsic or primary endocrine resistance (higher mutation rate).

Title: FoundationOne profiling of TSC1 and TSC2-mutated advanced breast cancers
Norma A Palma1, Juliann Chmielecki1, Garrett Frampton1, Siraj Ali1, Maren Levin3, Jeffrey S Ross1,2, Deborah Morosini1, Gary
Palmer1, Vincent A Miller1, Phil Stephens1 and Joyce O'Shaughnessy3. 1Foundation Medicine, Cambridge, MA; 2Albany Medical
College, Albany, NY and 3Baylor Charles A. Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, TX.
Body: Background: The TSC1/TSC2 complex inhibits the mTOR pathway. Loss of function of this complex leads to
hyperperactivation of the mTOR pathway which promotes growth and survival. Few studies have examined the role of TSC1 and
TSC2 mutations in breast cancer and as a clinically relevant target for molecularly targeted therapy, such as everolimus, a
rapamycin analog.
Methods: Hybridization capture of 3769 exons of 236 cancer-related genes and 47 introns from 19 genes that are frequently
rearranged in cancer were fully sequenced to high, uniform coverage from a commercial CLIA-certified laboratory (Foundation
Medicine).
Results: 30 of 2,208 breast cancer pt samples (1.3%) harbored either TSC1 (13/30, 43%) or TSC2 (17/30, 56%) genomic
alterations (GAs). The samples with TSC1/2 consisted of breast carcinoma n.o.s. (17/30) and invasive ductal carcinoma (12/30);
one invasive lobular carcinoma (ILC) harbored a TSC2 alteration. Sequencing was performed on metastatic (18/30, 60%) and
primary (12/30, 40%) tumors. All TSC1 alterations resulted in truncation of the protein product; TSC2 GA consisted of truncations
(14/17) and point mutations (3/17). TSC1/2 GAs frequently co-occurred with GAs in TP53 (19/30; 63%) and amplifications of MYC
(9/30; 30%) and FGFR1 (7/30; 23%). Of the TSC1/2-mutant cases where hormone receptor and HER2/neu status was known
(23/30), 65% were ER-/PR-/HER2- (TNBC), 26% were ER+/HER2- and 9% were ER-/HER2+.
Patient case: We describe a postmenopausal node positive ILC (ER+/PR+/HER2-) pt who received adjuvant chemotherapy
followed by 9 yrs adjuvant anastrazole then with disease recurrence. On FoundationOne testing a TSC2 A1141T missense
mutation, along with GAs in CDH1, ETV6, and CSF1R, were found in her metastatic disease in the soft tissues of her left orbit
associated with substantial swelling resulting in loss of vision. While TSC2 A1141T has not been functionally characterized, in
vitro and in vivo data suggest sensitivity to mTOR inhibitors. Her disease stabilized but did not decrease in size with vinorelbine +
capecitabine. She was treated with letrozole + everolimus and had a near clinical complete response with restoration of vision for
5 + months. Collection of clinical data on additional pts is currently ongoing.
Conclusion: 1.3% of breast cancer cases harbored GAs in either TSC1 orTSC2. The majority of TSC1/2 GAs cause inactivation
of these negative regulatory proteins leading to activation of the mTOR signaling pathway. TSC1/2 GAs were enriched for TNBC
(65%) versus non-TNBC (35%) (p= 0.077; Fishers exact test). Truncation of TSC1/2 occurred at multiple codons, underscoring
the importance of examining the entire coding region of tumor suppressor genes for these inactivating events. Comprehensive
genomic profiling has the potential to identify the broad spectrum of these inactivating events, a subset of which may be clinically
relevant.

Title: Deep clonal profiling identifies distinct mechanisms of heterogeneity and evolution in breast cancer
Princy Francis1, Lucretia M Alvarez1, Elizabeth Lenkiewicz1, Mia Champion1, Lisa Evers1, Karen L Anderson1, Ann E McCullough1,
Michael T Barrett1 and Barbara Pockaj1. 1Mayo Clinic, Scottsdale, AZ.
Body: Background: Breast tumors exhibit intratumor heterogeneity resulting in targeted therapy resistance and other challenges
in disease management. To address the sources of heterogeneity, we performed a unique, in-depth analysis of clonal
architecture in primary chemoradiation-nave breast cancers. We combined DNA content-based flow cytometry and ploidy
analysis with aCGH and next-generation sequencing (NGS) in multiple biopsies from the tumors and involved lymph nodes (LNs).
Material and methods: We used DNA content-based flow sorting to isolate nuclei from distinct populations of diploid and
aneuploid tumor cells in surgical tumor samples from two chemoradiation-nave patients. Each sorted tumor cell population was
interrogated with aCGH and exome NGS. In Patient #1, we used 12 fresh frozen sections morphologically mapped from within a
HER2+, ER+, PR- primary invasive ductal carcinoma (IDC) of histological grade 3 with LN involvement and 2-3 sections from 2
out of 5 LNs. In Patient #2, 11 morphologically mapped fresh frozen sections were analyzed from a grade 2, ER+, PR+, HER2+,
BRCA2 mutant LN- IDC. In parallel, matching samples were processed for IHC assays.
Results: We identified multiple co-existing aneuploid populations within the biopsies. The 17 primary and LN biopsies from Patient
#1 fell into 6 distinct ploidy groups albeit with aberrant but homogenous aCGH profiles, characterized by SARC amplification and
homozygous deletions in ROBO1 and ROBO2. In contrast a dominant ploidy was identified throughout Patient #2 but with
heterogeneous aCGH profiles. Mutation profiles obtained through exome sequencing further confirmed that ploidy was the main
driver in Patient #1 whereas copy number aberrations played the key role in Patient #2 with the BRCA2 mutation (R3129X). A
dendrogram based on exome variant calls of the aneuploid populations in Patient #1 strongly correlated with ploidy group and
further revealed the specific clonal population characterized by a 5N ploidy and homozygous mutations in TP53 and PIK3CA as
the progenitor to the ploidies present in the distant LNs. Strikngly, both patients had no HER2 amplification or mutation across
their clonal populations, contradicting the initial IHC staining in a single core biopsy.
Conclusions Rather than inferring the presence of distinct tumor cell populations, our novel flow-sorting based approach of first
identifying the clonal populations and then interrogating their genomes, provides an objective method of exploring the sources
and clinical significance of tumor heterogeneity. Our approach of clonal analysis has broad implications in the study of tumor
heterogeneity and the identification of drivers in breast and other solid tumors that can advance more effective treatment and
clinical management of patients with this disease.

Title: Mutational landscape of metaplastic breast carcinomas
Charlotte KY Ng1, Britta Weigelt1, Salvatore Piscuoglio1, Y Hannah Wen1, Maria R De Filippo1, Luciano G Martelotto1, Rachael
Natrajan2, Raymond Lim1, Edi Brogi1, Larry Norton3, Anne Vincent-Salomon4 and Jorge S Reis-Filho1. 1Memorial Sloan Kettering
Cancer Center, New York, NY; 2Breakthrough Breast Cancer Research Centre, The Institute of Cancer Research, London, United
Kingdom; 3Memorial Sloan Kettering Cancer Center, New York, NY and 4Institut Curie, Paris, France.
Body: Introduction: Metaplastic breast carcinoma (MBC) is an aggressive histologic type of breast cancer, which preferentially
displays a triple-negative phenotype. These tumors are characterized by the presence of malignant cells exhibiting differentiation
towards squamous epithelium or mesenchymal elements, including spindle, chondroid, osseous and rhabdoid differentiation.
Unlike other rare histologic types of breast cancer such as adenoid cystic and secretory carcinomas, which are underpinned by
the MYB-NFIB and ETV6-NTRK3 fusion genes respectively, pathognomonic genetic alterations have not been identified in MBC.
It has been suggested, however, that the frequency of PIK3CA somatic mutations would be significantly higher in MBCs than in
other forms of triple-negative disease. Here we sought to characterize the mutational landscape of MBCs by means of high-depth
whole exome sequencing analysis.
Material and Methods: Twenty-one triple-negative MBCs were retrieved from the authors institutions. Representative sections
from frozen blocks were microdissected to ensure tumor cell content greater than 50%. DNA samples extracted from
microdissected tumor and matched peripheral blood leukocytes were subjected to high-depth (250x) whole exome sequencing on
an Illumina GAIIx or HiSeq2000. Somatic point mutations were called using MuTect and somatic insertions and deletions (indels)
were called using Strelka, Varscan2 and Haplotype Caller. Potentially pathogenic mutations were predicted using computational
algorithms including PolyPhen-2, Mutation Taster, Mutation Assessor, CHASM and FATHMM. Significantly mutated genes were
identified using MutSigCV. Pathway and network enrichment analysis of mutations was performed with Ingenuity Pathway
Analysis and HOTNET. The genomic landscape of MBCs was compared with that of triple-negative breast cancers (TNBCs)
analyzed as part of The Cancer Genome Atlas project.
Results: A mean of 135 somatic non-synonymous point mutations and indels were identified per MBC. The most frequently
mutated gene was TP53, found in 12/21 cases (57%), and the only significantly mutated gene as defined by MutSigCV (q<0.01).
The repertoire of somatic mutations found in MBCs was qualitatively similar to that of TNBCs of no special type, and recurrently
mutated genes were altered at similar frequencies in MBCs and TNBCs of no special type. When somatic mutations were
annotated in pathways and networks, MBCs were found to have potentially pathogenic mutations affecting genes directly related
to the PI3K pathway, including pathogenic non-synonymous mutations affecting PIK3CA, PIK3R1, PIK3R2, PIK3C2B, PIK3C2G
and PTEN, significantly more frequently than TNBCs of no special type (10 out of 21 MBCs vs. 11 out of 62 TNBCs; Fisher's
exact test p-value=0.0099).
Conclusion: The majority (57%) of MBCs harbored non-synonymous mutations affecting TP53. While the frequencies of
mutations affecting recurrently mutated genes in MBCs are similar to those found in other forms of TNBCs, MBCs significantly
more frequently harbor mutations affecting PI3K pathway-related genes than TNBCs of no special type.

Title: Benchmarking mutation function prediction algorithms using validated cancer driver and passenger mutations
Maria R De Filippo1, Charlotte KY Ng1, Jorge S Reis-Filho1 and Britta Weigelt1. 1Memorial Sloan Kettering Cancer Center, New
York, NY.
Body: Background: Massively parallel sequencing studies have identified large numbers of mutations of unknown biologic
significance. There is a pressing need for computational methods to predict and distinguish neutral from potentially pathogenic
mutations accurately, to help identify those mutations worth exploring experimentally and clinically. Although various bioinformatic
algorithms are available, they are based on different methodologies and assumptions, and their predictions for the same
mutations are not always concordant. In this study, we sought to benchmark the performance of 17 prediction algorithms using
functionally validated and pathognomonic mutations.
Methods: We curated the literature for functionally validated and pathognomonic mutations as our positive dataset (i.e.
pathogenic mutations). For the negative dataset (i.e. neutral mutations), we retrieved variants from the dbSNP database,
including only those with minor allele frequency >25%. We compiled a total of 7975 mutations (875 pathogenic and 7100 neutral).
The performance of each prediction algorithm, namely accuracy, specificity, sensitivity, positive predictive value (PPV) and
negative predictive value (NPV), were defined using the positive and negative datasets described above. Confidence intervals
were calculated by sub-sampling 2/3 of the functionally pathogenic mutations and equal number of neutral mutations 500 times.
To reduce the bias introduced by mutations included in the COSMIC database, we excluded those found in COSMIC v67,
resulting in 6048 mutations (212 pathogenic and 5835 neutral), and re-evaluated the performance of each prediction algorithm.
Results: Our analysis revealed that the overall accuracy varied considerably, with a median of 87% (range 78%-97%). In terms of
accuracy, FATHMM (cancer) statistically outperformed all other prediction algorithms (97%, 95% confidence interval (CI)
96%-98%), followed by MutationTaster 2 (94%, 95% CI 93%-95%). Sensitivity and specificity also varied (median 85%, range
77%-96% and median 89%, range 71%-100%, respectively). The most sensitive prediction algorithm, FATHMM (cancer) (96%,
95% CI 95%-97%) statistically outperformed all others. The most specific prediction algorithm was CHASM (breast) (100%, 95%
CI 94%-100%). While CHASM (breast) had the highest PPV (100%, 95% CI 99%-100%), FATHMM (cancer) had statistically
better NPV than all other prediction algorithms (96%, 95% CI 95%-97%). When COSMIC mutations were removed, FATHMM
(cancer) remained the most accurate (93%, 95% CI 91%-95%) though the difference was not statistically significant. In this
context, CanDrA (breast) was the most sensitive prediction algorithm (95%, 95% CI 93%-97%) and had the highest NPV (93%,
95% CI 90%-96%), while CHASM (breast) was the most specific prediction algorithm (100%, 95% CI 99%-100%) and had the
best PPV (99%, 95% CI 97%-100%).
Conclusions: Our results demonstrate that functional prediction algorithms varied in performance. Using this dataset of
mutations, FATHMM (cancer) outperformed all other prediction algorithms in terms of accuracy, sensitivity and NPV, and
remained the most accurate even when mutations catalogued in the COSMIC database were excluded.

Title: Comprehensive somatic SNV and CNV profiling for triple-negative breast cancer patients by targeted next-generation
sequencing
Taegyun Yun1, Byung-Chul Lee1, Jungsun Park1, Dongyoon Park1, Tae-Kyung Yoo2, Min Kyoon Kim2 and Wonshik Han2.
1
Bioinformatics Tech. Lab, Healthcare Group, Future Technology R&D Division, SK telecom, Seongnam-si, Gyeonggi-do, Korea
and 2Seoul National University College of Medicine, Seoul, Korea.
Body: Introduction: Triple-negative breast cancer is a subtype of breast cancer lacking expression of estrogen receptor,
progesterone receptor and HER2/neu protein markers. Despite having the worse prognosis, currently there is no molecular target
for this subtype. To discover the clues for therapeutic targets of TNBC, we examined somatic Single Nucleotide Variation (SNV)
and Copy Number Variation (CNV) profiling of TNBC patients using targeted next-generation sequencing (NGS).
Methods: A total of 414 breast cancer and normal breast samples were collected at Seoul National University Hospital (SNUH)
from 1995 to 2013. By sample condition, 94 samples were frozen tissues (47 tumor-normal matched pair), and 320 samples were
formalin-fixed and paraffin-embedded (FFPE) tissues (155 tumor-normal matched pair; Tumor 164, Normal 156). Genomic DNA
was extracted from samples and target enrichment was done by using Agilent SureSelect Human Kinome Panel (612 genes
including over 500 kinases). The paired-end libraries were constructed and sequenced on Illumina HiSeq 2000 instrument with
average 250x depth coverage. Generated sequence reads were aligned to human genome hg19 with bwa algorithm, and somatic
SNV and CNV were identified using VarScan2 algorithm. To confirm the existence of identified somatic SNV and CNV, we carried
out SNP microarray experiment (Illumina Human Omni5 Exome microarray) for 46 pairs of frozen tissue samples. There were
18,720 SNP probes in SNP microarray covering target region of kinome panel. If the same SNV and CNV were detected in both
NGS and microarray data, we considered them validated.
Results: In 46 paired frozen samples, 610 somatic SNVs were detected. The most frequent and non-synonymous mutations
were in PIK3CA (8.6%), BMPR1A (4.3%), FES (4.3%), TP53 (4.3%), ROCK1 (4.3%), PAK2 (4.3%), CDK18 (4.3%), OBSCN
(4.3%), LRRK1 (4.3%), CDC6 (4.3%), PIKFYVE (4.3%), and FGFR3 (4.3%). Chromosomal aberration was detected in
chromosome 1 (10 samples) and chromosome 8 (11 samples). 242 somatic CNV were found in 157 genes in 46 patients, but 99
gene amplifications showed only single occurrence. In other words, TNBC sample displayed highly heterogeneous CNV profile.
However, we detected two frequently amplified genes; PKHD1L1 (13%) and NRBP2 (10.9%). Further analysis on NGS data of
155 pairs of FFPE samples is in process to validate the results.
Conclusion: We investigated comprehensive somatic mutation profile in matched pair TNBC samples. Although these samples
showed highly heterogeneous mutation profile, its possible to detect some interesting SNV and CNV with over 4% frequency.
Further analysis may illuminate clinical meaning of those alterations.

Title: Interaction between smoking history and gene expression levels impacts survival of breast carcinoma patients
James L Wittliff1, Sarah A Andres1, Mohammad A Alatoum1, Katie E Bickett1, Theodore S Kalbfleisch1 and Guy N Brock1.
1
University of Louisville, Louisville, KY.
Body: Our investigations explore the association of cigarette smoking on breast cancer risk of recurrence and progression, in
contrast to studies that focused on tobacco use and risk of breast cancer occurrence. The goal was to decipher the interaction
between smoking history and expression levels of 22 gene candidates selected from microarray data obtained from laser capture
microdissected carcinoma cells from 247 de-identified patient tissue biopsies on disease recurrence and overall patient survival of
breast cancer patients. qRT-PCR was used to determine expression levels for NAT1, NAT2, COMT, SOD1, SOD2, BRCA1,
BRCA2, APOC1, ARID1B, CTNNBL1, MSX1, UBE2F, IRF2, NCOA1, LECT2, THAP4, RIPK1, AGPAT1, C7orf23, CENPN,
CETN1 and YTHDC2 selected from a previous study for 50 breast cancer patients with a history of cigarette smoking and 51
patients who had never smoked. For smokers and non-smokers separately, L1-penalized multivariable Cox regression models
were fit to predict patient disease-free and overall survival, with 1000 splits of the data into training (70%) and test (30%) sets to
determine predictive accuracy based on the C-index. The LASSO penalty was used to perform variable selection in each of the
training sets, and a permutation procedure was used to determine a significance threshold for the number of times a variable was
kept in the model. Multivariable analyses using the LASSO revealed CENPN, CETN1, CYP1A1, IRF2, LECT2, and NCOA1 to be
significant predictors for both disease recurrence and mortality among smokers. Additionally, COMT was highly associated with
recurrence, and NAT1 and RIPK1 were associated with mortality. In contrast, only IRF2, CETN1, and CYP1A1 were significant
for disease recurrence and mortality among non-smokers, with NAT2 additionally significant for survival. Median, 25th percentile,
and 75th percentile for the C-indexes based on the gene expression models are given in Table 1.
Table 1. C-indexes for the Four Gene Expression Models Based on 1000 Test Data Sets
Patient Group
Outcome
25th Percentile
Median
75th Percentile
Current Smokers
Overall Survival
0.73
0.80
0.86
Current Smokers
Recurrence-free Survival
0.67
0.73
0.78
Never Smoked
Overall Survival
0.53
0.59
0.66
Never Smoked
Recurrence-free Survival
0.51
0.59
0.66
Analysis of interaction between smoking status and gene expression values using the combined samples revealed significant
interactions between smoking status and CYP1A1, LECT2, CETN1. Molecular signatures consisting of 7-8 genes were highly
predictive for breast cancer recurrence and overall survival among smokers, with median C-index values of 0.8 and 0.73 for
overall survival and recurrence, respectively. In contrast, the median C-index values for non-smokers was only 0.59. Hence,
significant interactions between expression of crucial genes and cigarette smoking status appear to play a key role in predicting
clinical outcomes of breast carcinoma patients. Supported in part by a grant from the Phi Beta Psi Charity Trust (TSK & JLW) and
a Research of Women (ROW) grant to JLW from the EVP for Research and Innovation, University of Louisville.

Title: An academic cancer centers experience in establishing a breast cancer-focused multidisciplinary genomic tumor board
Michelle McGowan1, Shaveta Vinayak2, Steven Maximuk2, Roselle Ponsaran1, Patricia Marshall1, Lyndsay Harris2 and Paula
Silverman2. 1Case Western Reserve University, Cleveland, OH and 2University Hospitals Case Medical Center, Cleveland, OH.
Body: Background. Clinically, it is becoming increasingly important to assess genomic alterations in breast cancers using Next
Generation Sequencing to find potential targets for personalized treatment. However, care paths for using such results to guide
treatment have not yet been established. Previously reported studies suggest that developing an interdisciplinary review board to
evaluate genomic findings can help to guide patient care. We report on a prospective qualitative study of a multidisciplinary
genomic tumor board (MGTB) developed to discuss breast cancer patients genomic test results in the context of the clinical
history and current practice guidelines. Our goal was to observe and describe the establishment of the MGTB and its members
perceptions of a tumor board approach to interpreting genomic test results.
Materials and Methods. After IRB approval, we used qualitative case study methodology to study the MGTB and its participants,
which include medical, surgical, and radiation oncologists, pathologists, geneticists, epidemiologists, bioinformaticians,
pharmacologists, nurses, and bioethicists. MGTB meets on a monthly basis. Cases are anonymously presented by the treating
physician. Genomic tumor findings from commercial vendors and in-house studies are discussed. Treatment recommendations
are made by consensus, with level of evidence assigned based on Simon-Paik-Hayes biomarker guidelines. After obtaining
consent from MGTB members, MGTB proceedings and interviews were recorded and transcribed for thematic analysis. Social
dynamics within the MGTB, which includes interaction among MGTB members regarding interpretation and communication of
genomic findings, were analyzed.
Results. MGTB began meeting monthly in 10/2013. As of 6/2014, 24 cases have been presented. All were female with metastatic
breast cancer; 16/24 (67%) hormone receptor positive (ER or PR >1%), 7/24 (29%) triple negative, and 2/24 (9%) her2 positive.
Analysis of meeting and interview transcripts revealed that though inherently multidisciplinary, the MGTBs recommendations
seem to be driven by medical oncologists with input from geneticists and pharmacologists when present. MGTB
recommendations have become increasingly consistent over time. Based on genomic findings, the board often recommends
referring patients to clinical trials (18/24, 75%) and to Medical Genetics (9/24, 38%) for evaluation of suspected germline
abnormalities over other options such as drug therapies approved by the FDA for use in other tumor types.
Discussion and Conclusion. MGTB is a unique forum to discuss genomic test results and departs from typical tumor boards by
involving non-clinicians. This case study suggests the feasibility of standardizing interdisciplinary evaluation and level of evidence
recommendations for genomic results. The findings of this qualitative case study reveal MGTB members discomfort with
recommending off-label use of FDA-approved therapies. Hence, on an institutional level, team dynamics will determine how
testing and recommendations are established. Further study is required to assess institutional and professional variability in
offering genomic testing and making treatment recommendations on the basis of genomic findings.

Title: Multiple breast cancer risk variants are associated with differential transcript isoform expression in tumors
Jennifer L Caswell1, Scott Huntsman1, Donglei Hu1 and Elad Ziv1. 1University of California, San Francisco, San Francisco, CA.
Body: Introduction: Genome-wide association studies have identified over seventy single-nucleotide polymorphisms (SNPs)
associated with breast cancer. A subset of these SNPs regulate expression of nearby genes, but the functional effects of the
majority remain unknown. We hypothesized that some breast cancer risk SNPs may regulate alternative splicing of nearby genes.
Methods: We used RNA sequencing data and matched germline genotypes from The Cancer Genome Atlas for 358 estrogen
receptor (ER)-positive and 109 ER-negative breast tumors. For each risk SNP, we tested the association between genotype and
exon-, junction-, and transcript-specific expression of genes within 500 kilobases, adjusting for overall gene expression. We
excluded associations potentially related to error in mapping to pseudogenes or bias in mapping to the reference genome.
Results: Six breast cancer risk SNPs were associated with differential isoform expression of seven nearby genes at FDR < 0.05
in ER-positive tumors; five of these associations replicated (P-value < 0.05) in ER-negative tumors (Table 1). At five loci, the
pattern of association with alternative splicing mirrored the pattern of association with breast cancer. Of these five, two SNPs
were associated with alternative splice site usage in nearby genes (STXBP4, BABAM1), one with exon skipping (DCLRE1B), and
two with more complex exon usage patterns (PEX14, RAD51L1).
Associations between breast cancer risk SNPs and transcripts identified in ER-positive tumors at FDR < 0.05
SNP rsID
SNP Location
SNP Distance from

Gene
Gene
(ER+) P-value (ER+) FDR (ER+) P-value (ER-)
Component
rs6504950
chr17:53,056,571 STXBP4
In gene: intron 1
Exon 5:6 junction 1
0.59
1.5E-23
2.6E-19
3.1E-07
Exon 5:6 junction 2
-0.72
2.5E-23
2.6E-19
2.3E-10
Transcript uc010dcc
-0.42
1.5E-11
2.6E-08
-0.63
5.7E-13
1.4E-09
0.26
8.0E-10
1.2E-06
-0.48
1.3E-07
1.3E-04
4.7E-03
0.24
1.2E-06
9.1E-04
1.0E-04
Transcript uc002nfv
0.47
6.1E-08
6.5E-05
5.3E-05
Transcript uc002nfu
-0.27
5.3E-06
3.3E-03
6.3E-03
-0.45
1.4E-07
7.3E-06
1.5E-02
Exon 1:2 junction
0.50
1.6E-07
1.5E-04
Exon 2
0.34
3.5E-06
2.3E-03
Exon 7
-0.39
1.1E-06
8.8E-04
Transcript
uc001arm
-0.39
1.1E-06
8.1E-04
rs11552449* chr1:114,448,389 DCLRE1B
In gene: exon 1
Transcript uc001eei
Transcript
uc001eeg
Exon 1:3 junction
Exon 2
rs8170
chr19:17,389,704 BABAM1
In gene: exon 8
rs11249433 chr1:121,280,613 SRGAP2D
151 kB
Exon 2:3 junction

rs11552449* chr1:114,448,389 PHTF1
rs616488
146 kB
chr1:10,566,215 PEX14
In gene: intron 2
2.9E-02
rs999737
chr14:69,034,682 RAD51L1
In gene: intron 13
Exon 15
-0.33
6.2E-05
3.4E-02
Location is for hg19 build. is for effect of the breast cancer risk allele. FDR is when all exon, junction, and transcript tests are
considered together. P-values for ER- shown when < 0.05. * SNP is associated with transcript expression of two different genes.
Pattern of association with transcript does not correspond to pattern of association with breast cancer.
Conclusion: Regulation of differential transcript expression appears to be an important functional mechanism of breast cancer risk
SNPs. These associations allow us to identify likely candidate causal genes, and also specific exons and transcripts that may
mediate breast cancer risk, opening a new window into the link between risk variants, gene function, and outcome.

Title: Long range expression patterns detected by RNASeq in breast cancer reveal cluster enriched with triple negative breast
cancer
Alex Mankovich1, Vartika Agrawal1, Nilanjana Banerjee1 and Nevenka Dimitrova1. 1Philips Research, Briarcliff Manor, NY.
Body: Background: Breast cancer subtyping using gene expression is well established in breast cancer research. While it is
known that there are large chromosomal regions affected by copy number polymorphisms in breast cancer, it is not clear whether
expression patterns reflect large genomic events affecting longer portion of the chromosome. We present a method to quantify
long range gene expression patterns of larger genomic regions at 23kb, 100kb and 1mb.
Methods: We used TCGA-level 2 breast cancer gene expression data (RNA-Seq) generated at the Carolina Center for Genome
Sciences, UNC at Chapel Hill. We evaluated the long range expression patterns of 649 patients for which we had ER, PR and
Her2 status data by IHC. For 221 samples we had ER, PR, Her2, age, menopausal status, p53 and PIK3CA mutation status. Our
method defines long range expression within a window of a particular length (e.g. 23kb, 100kb). We take the mean expression
scores for all genes that fall within each window and concatenate these windows to obtain larger chromosome-wide patterns. The
final chromosome-wide vectors are concatenated to represent long range expression patterns across the entire genome. We then
evaluate the variation of these window scores across all samples and keep top 2% varying windows. Then, we apply hierarchical
clustering, and evaluate enrichment of clinically meaningful subtypes using hypergeometric test.
Results: Simple hierarchical clustering revealed clear separation of triple negative breast cancer samples at any level of
resolution: 23kb, 100kb in the available data set. While the 100kb resolution showed two distinct clusters, the 23kb resolution
showed three distinct clusters. Interestingly, the hierarchical clustering of samples (n=649) using the top 214 (2%) highly varying
long 100kb regions revealed a cluster which contained 99 samples and was enriched with 75 TNBC samples (out of 102 TNBC in
the entire set) (p=1.7E-53).
The long range expression of 23Kb regions on 221 samples revealed 319 (2%) such regions which further segregate the samples
into three different clusters: Cluster1: Luminal enriched cluster with 147 samples out of which 104 are ER+PR+Her2- samples
(p-value <<0.01), Cluster2: Her2+ enriched cluster with 31 samples out of which 23 are Her2+ (p-value<<0. 01), and Cluster3:
TNBC enriched cluster containing 33 samples out of which 26 are TNBC (p=1.5E-18). There are a total number of 38 TNBC
samples. Furthermore, in the TNBC cluster with 33 samples, 28 had p53 mutations and 24 of those were both TNBC and had p53
mutations (p-value << 0.05). In this TNBC cluster, there is enrichment of samples lacking PIK3CA mutations (n=31)(p-value
<<0.01). We found no clear association with age and menopausal status.
Conclusions: Hierarchical clustering relying on long range expression regions produces clusters that are enriched with well known
clinically relevant subtypes. The associations with known tumor biology are strong for different region sizes (23kb and 100kb).
This is the first study to report long range gene expression patterns that reveal data-driven close association with tumor biology.

Title: Re-assessment of 21-gene breast cancer assay utilization as a quality indicator by the Florida Initiative for Quality Cancer
Care (FIQCC) Consortium
Christine Laronga1, Weihong Sun1, Erin Siegel1, Ji-Hyun Lee1, William J Fulp1, Paul B Jacobsen1 and Jhanelle E Gray1. 1H. Lee
Moffitt Cancer & Research Institute, Tampa, FL.
Body: Background: In 2006, the FIQCC conducted a comprehensive review of quality of care indicators specific to breast cancer
(BC) based on QOPI/NCCN/ACOS and panel consensus, with feedback provided on indicator performance to encourage quality
improvement efforts. In 2009, at re-assessment of adherence to the previous indicators, we introduced a pilot quality indicator
from the NCCN treatment guidelines: discussion/recommendation of the 21-gene BC assay for treatment decisions. The current
study assesses quality improvement efforts targeting specifically the 21-gene BC assay.
Methods: After disclosure of results in 2012, 5 FIQCC sites agreed to participate in a re-assessment sub-study focused on the
21-gene BC assay. Following IRB approval, each site developed and implemented a site-specific quality improvement plan
focused on the discussion/utilization of this assay. Following this intervention, chart reviews were conducted for ER+ BC patients
first seen by a medical oncologist in 2013 and eligible for the 21-gene BC assay based on NCCN criteria (n=130). Data collected
included: 1) documentation of a discussion with the patient of the 21-gene BC assay; 2) documentation of
ordering/performance/results of the 21-gene BC assay; and 3) treatment decisions based on results and/or documentation of
reasons for non-compliance. Statistical comparisons of the previously collected 2009 data (n=262) for these 5 FIQCC sites and
new 2013 data were performed using the Pearson Chi-Square exact test based on Monte Carlo estimation.
Results: Overall, the mean age of patients was 62 years (range 30-89) for the combined 2009 and 2013 years. More women had
her2 negative BC in 2013 compared to 2009 (p=0.003). Since the 2009 dataset was not specific to the 21-gene BC assay
eligibility, a smaller proportion of patients had tumor sizes staged as T1c/T2/T3 [80.3% vs. 92.2% for 2009 as compared to 2013,
p=0.005] and less were node negative in 2009 [56.8% vs. 96.6%; p<0.001]. Based on NCCN criteria, only 60.9% of eligible
patients had a documented discussion or recommendation for the 21-gene assay in 2009 with significant variability among the 5
sites (p=0.011). A significant improvement was seen in 2013 (p<0.001) with 91.4% of eligible patients having documentation of a
discussion or recommendation for the 21-gene assay without significant variability amongst the practice sites. Significant
improvement in the percentage of patients having the assay ordered from 2009 to 2013 was seen (p<0.001) but variability
amongst the sites persisted (2009: p=0.002; 2013: p=0.018). Recurrence Score (RS) results were completely s recorded in both
2009 and 2013. All patients with high-risk RS were referred for chemotherapy. A small fraction of patients with low-risk RS was
referred for chemotherapy both years (2009: 5.4%, 2013: 3.8%; p=1.00).
Conclusions: Previously, the FIQCC identified quality improvement needs in multiple aspects of BC care, including
recommendation and ordering of the 21-gene BC assay. A dedicated quality improvement effort can significantly increase both
the discussion/recommendation and ordering of the 21-gene BC assay among eligible patients. These improvements have the
potential to alter treatment recommendations.

Title: Examining 3-dimensional genome architecture and transcriptional profiles using an optimized HiC-seq protocol in breast
cancer cells
Jennifer D Davis1, Christian Ross2, Hu Li2 and Amy Brock1. 1University of Texas, Austin, TX and 2Mayo Clinic, Rochester, MN.
Body: The post-genomic era ushered in a cadre of techniques to study epigenetic mechanisms involved in disease progression
and cancer. One understudied mechanism of gene expression/transcriptional control is 3-dimensional architecture of the genome.
Several high-throughput technologies have been adapted to examine 3D genome architecture; in particular those derived from
chromatin conformation capture (3C) sequencing technologies. Of 3C-derived techniques, HiC-seq, is the most extensive, and
examines all chromatin-chromatin contact sites within a genome, elucidating the entire 3D genome architecture. Although,
HiC-seq technique was published 2009, it has been under-utilized in research laboratories due to its technical complexity,
expense, and requirement for large volumes of starting material (50 million cells/500 g of DNA). Other barriers to use of HiC-seq
include the requirement for specialized equipment and laborious extraction techniques. However, the benefits of HiC-seq are high
for cancer researchers because a genome-wide "snap-shot" of the entire 3D genome of the cancer is displayed. This allows for
analysis of higher-order epigenetic control of gene expression. Furthermore, understanding the chromatin conformation of the
genome provides new insights into the cellular disease state and potentially novel therapeutic approaches to drug-resistant
cancers. Our study presents significant technical innovations, permitting HiC-seq technique to be adopted in laboratories without
extensive experience in 3C-seq technologies, and with the use of standard equipment and reagents, as well as smaller volumes
of starting materials. Our modified technique utilizes up to 100 fold lower amounts of starting material (500,000 cells/5-10 g of
DNA) per reaction. This adaptation of HiC-seq preparation offers the potential for simultaneous extraction of total RNA and Hi-C
processed chromatin for sequencing. To optimize our technique we utilized the estrogen-responsive breast cancer cell line,
MCF7. This cell line is typically utilized to study potential drug therapies in the context of estrogen-sensitive cancers. We stream
lined HiC-seq by optimizing nuclei extraction and allowing for collection of a cytoplasmic fraction for mRNA extraction and
sequencing analysis. We also optimized enzymatic digestion, ligation and DNA extraction methods by utilizing newer enzymatic
chemistries, commercially available silica column extraction methods and reagents in place of multiple phenol-based DNA
extraction methods. Our quality control results suggest that sample processing for reduced amounts of material is similar if not
superior to conventional HiC-seq. We leverage this powerful genomic high-throughput technique in order to simultaneously
examine the 3D architecture and transcriptional profiles of MCF7 breast cancer cells. Our computational workflow includes further
quality controls using HiC-User Pipeline software from the Bahraham Institute and the R Bioconductor package HiTC package.
Overall our optimizations in MCF7 breast cancer cells, and post-acquisition work-flow, make HiC-seq an accessible technology.
These innovations permit simultaneous analysis of architectural and transcriptional profiles in cancer cells.

Title: Assessing reproducibility of copy number arrays to assist breast cancer biomarker discovery
Cindy Q Yao1, Cheryl Crozier1, Mary Anne Quintayo1, Jane Bayani1, Melanie Spears1, Julie Livingstone1, Esther Jung1, Clement
Fung1, Victoria Sabine1, Paul C Boutros1 and John MS Bartlett1. 1Ontario Institute for Cancer Research (OICR), Toronto, ON,
Canada.
Body: Introduction:
Large-scale interrogation of the genome has emerged as an attractive method for identifying useful characteristics of cancer
biology; in particular, the study of copy number aberrations (CNA) has recently received tremendous attention. A number of
different technologies have been developed to assess the copy-number landscape, allowing us to better understand the role of
CNA in cancer cells. The OncoScan CNA platform (Affymetrix Inc.) has been particularly appealing for oncology due of its ability
to work well with formalin-fixed, paraffin-embedded (FFPE) materials, which is the primary form for storage of clinical samples. In
addition, its high resolution, rapid analysis time and ability to interrogate different genomic characteristics (CNA, loss of
heterozygosity or mutation) make the OncoScan platform highly popular: it has been widely cited in the literature for use in
biomarker discovery, clonal evolution and sub-clonal detection, as well as population-based analyses. While CNAs identified by
the OncoScan platform have shown good concordance with fluorescence in-situ hybridization (FISH) results, to date, no studies
have been conducted to thoroughly assess the reproducibility of the assay. In this study, we have assessed the reproducibility of
the OncoScan platform using identical samples performed in replicates across multiple chip batches. Moreover, we have
assessed the effect on reproducibility of DNA treatment, including elution in water or TE buffer, as well as in the use of varying
amounts of DNA.
Methods:
Affymetrix OncoScan FFPE Express 3.0 SNP Arrays were performed using the optimal input DNA as recommended by the
manufacturer as well as fewer input amounts for comparison. CNAs were called using BioDiscovery Nexus Copy Number
software (http://www.biodiscovery.com/software/nexus-copy-number/) using the SNP-FASST2 algorithm with modified
parameters (significance threshold of 1 x 10-9 and minimum number of probes per segment of 10).
Results:
Initial reproducibility analysis involving 12 samples repeated either 2, 4 or 6 times both within a single batch and across different
batches has revealed that CNA calls were concordant between replicates for the majority of the genome (ranges between 81% to
100%), suggesting high precision of the assay. In addition, we are in the process of assessing and comparing mutation calls
across replicates to gain a more in-depth understanding of the platform.
Conclusion:
This is the first study examining the reproducibility of OncoScan FFPE assays; initial results have suggested that the assay is
precise and has the potential for robust biomarker discovery. Additional characterizations would be interesting for evaluating its
use as a clinical tool in the long term.

Title: Discovery of novel amplified genes in primary breast cancer with copy number and gene expression analysis of whole
exome and transcriptome sequencing data
Eunshin Lee1, Woosung Lim1, Kyung-Min Lee1, Tae-kyung Yoo1, Jongjin Kim1, Han-Byoel Lee1, Yun-Gyoung Kim1, YoungJoon
Kang1, Min Kyoon Kim1, Hyeong-Gon Moon1, Dong-Young Noh1 and Wonshik Han Han1. 1Seoul National University College of
Body: Introduction: Copy number alteration of genome is common in breast cancer and tend to have more driver role than single
point mutations. Traditionally, genome-wide analysis of DNA copy number changes were done by array CGH or SNP array
method. Here, we did DNA whole exome sequencing (WES) and RNA-seq using Next Generation Sequencing (NGS) technology
to find common genes or chromosomal regions of which DNA copy was highly amplified and at the same time RNA expression
was also upregulated.
Materials and Method: RNA and DNA were extracted fromfresh frozen tissues of 93 breast cancer patients. WES and RNA-seq
were done using NGS technology (Illumina HiSeq 2000). As a control, normal DNA from all matched patients were also
sequenced. GATK was used to gain mean depth and coverage data for targeted regions.CNVs were calculated with ExomeCNV,
a statistical method to detect somatic CNVs using depth-of-coverage information from mapped short sequence reads.To estimate
expression levels, the relative transcript abundances were measured in FPKM using Cufflinks.
Results and Discussion: DNA of 1,737 genes were highly amplified (log R>1.0) in two or more samples. The two most commonly
amplified chromosomes were chromosome 8 and 17. We applied a cut-off for higher gene expression as relative FPKM >1.5.
ERBB2 amplifications and high expression were most common (21.5%) of all genes and it was in agreement withHER-2 IHC and
FISH result. Among previously reported amplified genes, FGFR1 (5.4%) and PVT1 (8.6%) in chromosome 8, CCND1, PAK1
(3.2%) and EMSY (4.3%) in chromosome 11, CCNE1 (4.3%) in chromosome 19 were also identified in this study. IGF1R high
amplification and expression was found in two samples, and ESR1, MDM2, KIT was found in only one sample each. We found
uncommon but novel and recurrent highly amplified and expressed genes: CLK4 in 5q (3.2%), AHI/MYB in 6q (3.2%),
MMP7(2.2%) andMALAT1 in 11q (1.1%), and NEK8 in 17q (4.3%) We designed FISH probe for this 5 new genes and confirmed
the high amplifications in each sample with FISH. Functional study of these genes will be followed for the driver role of these
genes in carcinogenesis and progression of breast cancer cells.

Title: Discovery of the genes that underpin the transition to malignant phenotype of breast tumors in highly consanguineous
region
Ishita Gupta1, Allal Ouhtit1, Somya Shanmuganathan1 and Hamad Al-Riyami1. 1Sultan Qaboos University, Muscat, Seeb, Oman.
Body: Breast cancer (BC), a multifactorial and heterogeneous disease is a predominant women form of cancer worldwide
affecting 22.9%. The rationale of this study is based on the following observations: 1) in Oman, a significantly increasing number
of younger females (25-40 years) present to the clinic with advanced stage of BC; 2) in Oman, the rate of consanguinity is
significantly high (52%); and 3) the transition from normal/begnin to malignant phenotype of breast tumor requires the
involvement of a subset of specific genes. The long-term objective of this study is to identify and validate the subset of genes that
are responsible for this malignant transformation using functional genomic studies, focus on this young age group of patients
attending BC clinic (sporadic and familial). RNA samples were isolated from 40 Breast Tumors and 40 Normal tissues and
analyzed by Microarray Gene Expression Profiling. Among a number of genes that were up and down regulated, BRIP1, HOXB3
and MAGED1 were identified as potential genes that might underpin the transition to the malignant phenotype; these genes were
validated by RT-PCR using the same RNA samples that were examined by microarray. Pathway analysis was carried out to
identify the major functional pathways connecting these genes. Ongoing sequencing of these genes using DNA extracted from
the same samples will ultimately identify any genetic alteration that can affect the normal function of these genes. Functional
validation assays aim to validate further the physiological relevance of these genes in tumor malignancy, and perhaps other novel
genes specific to BC in the Omani population. Identification and validation of these genes will potentially pave the way towards
the design of anti-cancer therapeutic strategies.

Title: No Show

Title: Identifying molecular targets and mechanisms of treatment resistance in inflammatory breast cancer (IBC) using
reverse-phase protein microarrays (RPMA)
Laura Austin1, Kimberly Limentani1, Juan Palazzo1, Tiffany Avery1, Rebecca Jaslow1, Ron Hencin3, Emanuel F Petricoin2,3 and
Massimo Cristofanilli1,4. 1Thomas Jefferson University Hospital, Philadelphia, PA; 2George Mason Universtiy, Fairfax, VA;
3
Theranostic Health, Inc, Rockville, MD and 4Inflammatory Breast Cancer International Consortium.
Body: Background
Inflammatory Breast Cancer (IBC) is a clinicopathologic diagnosis characterized by rapid progression and poor prognosis. Even
with the advent of targeted therapies and a multimodal approach, IBC is often treatment refractory and a therapeutic challenge for
all subtypes, including ER+ and HER-2 amplified (HER2+) disease (Masuda et al). Therefore identifying mechanisms of
resistance to molecularly targeted therapy could provide clues to improve management and outcome. Recent studies comparing
the gene expression profiles of IBC tumors with non-IBC demonstrated that HER2+ IBC have increased mTOR signaling
compared to their non-IBC counterparts (Iwamoto et al). mTOR activation is a mechanism for Trastuzumab resistance and may
contribute to treatment resistance in HER2+ IBC. The availability of molecular diagnostics evaluating phosphoproteins is an
appealing approach to predict treatment-sensitivity and select more effective combinations.
Methods
This is an observational analysis of 12 IBC patients who had tissue biopsy after progression on standard therapies including
HER-2 targeted therapies. Tissue analysis for expression of cancer-related phosphoproteins was performed using TheraLink.
The TheraLink assay uses reverse-phase protein microarrays (RPMA) to quantify HER1, HER2, and HER3 receptor
overexpression; it also evaluates for phosphorylation of the receptor which indicates activation. Phosphorylation of HER
downstream signaling pathways such as JAK2, AKT/mTOR and MEK1/2 are also detected. Additionally, next generation
sequencing (NGS) using FoundationOne was performed if sufficient tissue was available.
Results
All patients had IBC and most had metastatic disease (83%). According to subtype 25% of patients were ER+/HER2-, 42%
ER+/HER2+, 25% ER-/HER2+, 8% ER-/HER2-. 58% of tumors demonstrated HER1 activation, 75% had HER2 activation and
58% had HER3 activation. Interestingly, 83% had mTOR activation, and most of these patients also had accumulation of its
downstream proteins, S6 ribosomal protein and 4E-BP-1. 78% of patients with HER2 activation also had mTOR activation. Two of
the 4 patients who were HER2- by IHC/FISH had HER2 activation by RPMA. Six patients also had NGS on tissue; 75% had
concordance between HER2 activation on TheraLink and ERBB2 amplification on NGS, 67% had concordance with mTOR
activation on TheraLink and mutation in the mTOR pathway (PIK3CA mutation or PTEN loss) on NGS. One patient with triple
negative, chemo-refractory IBC who underwent 3 lines of neoadjuvant therapy prior to bilateral mastectomy was found to have
HER1, HER2, HER3 and mTOR activation; she was started on lapatinib and capecitabine and remains with no recurrent disease
and on treatment.
Conclusions
Patients with IBC often have activation of members of the HER family and mTOR pathway indicating molecular targets and
potential mechanisms of resistance in IBC. The concomitant use of NGS and RPMA is an intriguing approach to molecular
diagnostics in this aggressive and treatment refractory disease providing additional information on pathway activation leading to
expanded therapeutic options. Future prospective studies should clarify the potential impact in treatment selection and outcome.

Title: Circulating tumor DNA (ctDNA) provides molecular monitoring for inflammatory breast cancer (IBC)
Laura Austin1, Paolo Fortina1, Dragan Sebisanovic2, LaiMun Siew2, Aubrey Zapanta2, Benjamin J Schiller2, Gangwu Mei2, Helmy
Eltoukhy2, AmirAli Talasaz2 and Massimo Cristofanilli1,3. 1Thomas Jefferson University Hospital, Philadelphia, PA; 2Guardant
Health, Inc, Redwood City, CA and 3Inflammatory Breast Cancer International Consortium.
Body: Background
Inflammatory Breast Cancer (IBC) is a clinicopathologic diagnosis characterized by rapid progression, resistance to treatment and
poor prognosis. It is often an incurable disease with complex molecular features including somatic mutations that evolve in
relation to genomic instability and selective treatment pressure. Monitoring disease by performing multiple biopsies may not be
feasible and puts the patient at risk with each invasive procedure. Circulating DNA fragments carrying tumor-specific sequence
alterations (ctDNA) are found in blood and offer the possibility of longitudinal non-invasive molecular monitoring of the disease by
detecting actionable mutations.
Methods
This is an observational analysis of 35 IBC patients who failed standard therapies and had plasma analyzed for ctDNA detection.
Selection criteria included progression of disease after standard therapies, need to detect novel molecular abnormalities for
possible therapeutic targeting, or confirmation of genomic abnormalities already demonstrated in tissue analysis. Guardant Health
performed the plasma analysis (Guardant360); first ctDNA was isolated from plasma, then a panel of 54 gene mutations
associated with solid tumors as reported in the COSMIC database were sequenced to concurrently analyze somatic mutations
and gene amplification using single-molecule digital sequencing technology.
Results
All patients had IBC and 80% had metastatic disease; 37% of patients were ER+/HER2-, 14% ER+/HER2+, 23% ER-/HER2+,
and 26% ER-/HER2-. 94% of patients with stage III or IV tumors had ctDNA alterations detected. The most common mutations
were TP53 (49%), PIK3CA (20%), ERBB2 (17%), NOTCH1 (17%), and ALK (11%). Twelve patients also had next generation
sequencing (NGS) analysis of tissue biopsy and 75% of these patients demonstrated at least one concordant mutation. The
genomic information obtained from ctDNA, NGS or both was used to select treatments in 11 cases (31%). HER2 targeted therapy
was continued in four patients with HER2+ disease after ctDNA confirmed ERBB2 alteration or amplification. A patient with
ER+/HER2+ disease who progressed on HER2 targeted therapies, demonstrated ERBB2 and PIK3CA mutations on ctDNA; she
was changed to exemestane/everolimus with objective response for several months. Repeat ctDNA at time of progression
showed ERBB2 mutation only; she was changed to Trastuzumab/everolimus/vinorelbine. Moreover, a combination of lapatinib
and capecitabine was initiated on a patient with a triple negative, chemo-refractory tumor that on ctDNA revealed EGFR and
ERBB2 mutations; a repeated ctDNA after 5 months of therapy EGFR and ERBB2 mutations were not detected and she remains
without evidence of progression.
Conclusions
Evaluation and longitudinal monitoring of IBC patients using ctDNA allows for identification of genomic abnormalities in all
patients with advanced disease and to perform real-time molecular monitoring. The discovery of actionable genomic
abnormalities is driving the management of this aggressive and treatment refractory form of breast cancer with potential future
impact on outcome.

Title: Association of mammographic density and molecular breast cancer subtype
Brandy L Edwards1, Kristen A Atkins1, George J Stukenborg1, Wendy M Novicoff1, Krista N Larson1, Wendy F Cohn1, Jennifer A
Harvey1 and Anneke T Schroen1. 1University of Virginia, Charlottesville, VA.
Body: Background: While mammographic density is linked to increased breast cancer risk, limited yet conflicting data exists on
an association between density and developing specific molecular subtypes of breast cancer.
Methods: Eligible patients were enrolled in a larger study on breast density, diagnosed with cancer between 2003-2013, and had
pathology and films available for review. Density was classified qualitatively from existing radiology reports according to Breast
Imaging Reporting and Data System (BIRADS) classification and quantitatively by volumetric breast density measurements using
Volpara SolutionsTM software. Subtype was assigned by hormone receptor status, tumor grade and mitotic score (MS). Subtype
categories included: Luminal A (ER/PR + & grade 1; ER/PR + & grade 2 & MS=1; ER+/PR- & grade 1); Luminal B (ER+ & grade
3 or MS=3; ER+/PR- & grade 2; ER/PR + & grade 2 & MS=2); Her-2 + (ER+ or ER - & Her-2 +); Triple Negative (ER/PR-, Her-2
-). Relevant pre-cancer factors including patient age, race, BMI, family history of breast cancer, and biopsy showing LCIS were
included in analysis.
Results: Of 604 patients with invasive cancer, 457 had sufficient information for analysis. Among these, 233 (51%) had Luminal
A, 79 (17%) Luminal B, 59 (13%) Her-2 +, and 86 (19%) Triple Negative tumors. Younger women and those with denser breasts
based on quantitative measurements were more likely to have Her-2+ tumors (Table 1); this association was not seen using the
standard BIRADS classification. Triple Negative tumors were less common in patients with LCIS and more common in African
Americans. A multinomial logistic regression model controlling for pre-cancer patient factors demonstrated that while quantitative
breast density does not significantly differentiate between all molecular subtypes (p=0.140), the association between Her-2+
tumors and denser breasts using continuous quantitative measurements is significant (p=0.035).
Conclusion: Women with denser breasts by continuous-scaled quantitative measurements are at higher risk for Her-2+ tumors;
an association not delineated using standard BIRADS density classification. Delineating risk factors specific to molecular breast
cancer subtype may promote individualized risk prediction models and prevention strategies.
Table 1. Association between patient factors and molecular breast cancer subtype
Molecular Subtype
Variable
Luminal A
Luminal B
Her-2+
Triple Negative
P value
(n=233)
(n=79)
(n=59)
(n=86)
Age (median,IQR)
61 (54,70)
58 (50,67)
54 (46,70)
59 (48,67)
0.006
BMI (median,IQR)
27.1 (22.9,30.5)
25.7 (23.0,30.0)
26.0 (22.7,32.6)
28.1 (24.3,32.6)
0.173
Volpara density
(median,IQR)
7.18 (4.74,11.25)
8.68 (5.68,14.34)
10.25 (5.96,16.51)
7.00 (4.97,11.89)
0.002
BIRADS density (n,%)
0.183
Fatty
42 (18.0%)
15 (19.0%)
5 (8.5%)
16 (18.6%)
Scattered
103 (44.2%)
27 (34.2%)
20 (33.9%)
36 (41.9%)
Heterogeneous
74 (31.8%)
30 (38.0%)
25 (42.4%)
26 *30.2%)
Extreme
14 (6.0%)
6 (7.6%)
9 (15.2%)
7 (8.1%)
Unspecified
1 (1.2%)
1 (1.2%)
Race (n,%)
0.002
Caucasian
202 (86.7%)
65 (82.3%)
51 (86.4%)
61 (70.9%)
African American
23 (9.9%)
13 (16.5%)
6 (10.2%)
23 (26.7%)
Other
2 (0.9%)
2 (3.4%)
Unspecified
LCIS (n,%)
6 (2.6%)
1 (1.2%)
2 (2.3%)
48 (20.6%)
8 (10.1%)
8 (13.6%)
5 (5.8%)
0.004
34 (43.0%)
21 (35.6%)
39 (45.35%)
0.625
Family history of breast

107 (45.9%)
cancer (n,%)

Title: Modeling luminal breast cancer heterogeneity: Combination therapies to suppress hormone receptor-negative
subpopulations among receptor-positive ones
Allison L Scaling1, Aaron J Knox1, Maurcio P Pinto1, Brian S Bliesner1, James M Haughian1, Hany A Abdel-Hafiz1 and Kathryn B
Horwitz1. 1University of Colorado Anschutz Medical Campus, Aurora, CO.
Body: All Luminal breast cancers are heterogeneous, containing substantial numbers of estrogen (ER) and progesterone (PR)
receptor-negative cells among the ER+PR+ ones. We have identified two such Luminal-derived ERPR cells: The first, we call
"Luminobasal" (LB), are ERPR and cytokeratin 5 (CK5)-positive. The second, we call "Double Negative" (DN), are ERPR
and CK5. Currently, neither is targeted for treatment. Luminobasal cells: To address the relationships between true Luminal
ER+PR+CK5 and Luminobasal ERPRCK5+ cells in Luminal cancers and tightly control their ratios we generated isogenic
pure Luminal (pLUM) and pure Luminobasal (pLB) cells from solid tumor xenografts starting from the same parental T47Dco
Luminal human breast cancer cell line. Cells were gene profiled and unique immunohistochemical (IHC) biomarkers for each
were confirmed. Interaction dynamics studies show that in mixed-cell 3D colonies, pLUM or Luminal MCF-7 cells suppress growth
of pLB cells. Similarly, in mixed-cell solid tumor xenografts, pLUM cells suppress pLB proliferation. Alarmingly, in mixed-cell
models, monotherapy of a pLUM or MCF-7 subpopulation with the antiestrogen Fulvestrant inadvertently expands the number of
pLB cells. This suggested that pLB cells also need to be treated. An 89 drug FDA-approved oncology library and high throughput
screening methods were therefore used to show that pLB cells are specifically targeted by the EGFR inhibitors Gefitinib and
Erlotinib. We then showed, in both mixed-cell 3D colonies and mixed-cell solid mouse tumors, that combination therapy using
Fulvestrant plus Gefitinib constitutes a robust treatment strategy that targets both cell populations simultaneously. Double
Negative cells: DN cells are EGFR and therefore unaffected by EGFRi. However they express the Luminal marker Claudin-3
(CLD3). Among other conditions, DN cells arise as a discrete CLD3+ subpopulation within CLD3 pLB xenografts. This allowed
us to use CLD3 for FACS isolation and purification of DN cells from xenografts. Purified cells were gene expression profiled,
which showed that DN cells have a unique genomic signature that distinguishes them from LUM and LB cells. They express
diagnostic markers suitable for IHC and FACS-sorting, some of which also present potential therapeutic targets. We propose that
heterogeneous Luminal breast cancers may be best treated with combination therapies that include endocrine, EGFRi and/or DN
inhibitors. Importantly, optimizing combination regimens that eradicate or suppress not only ER+PR+ cells but also diverse
ERPR cells in Luminal disease requires that primary tumors be pre-screened for appropriate biomarkers, including but not
limited to ER, PR, CK5 and EGFR.

Title: Demonstration of immune cell and pathway heterogeneity in Singapore DCIS samples using novel hyperplexing method
(MultiOmyx)
Nicole E LaPlante1, Yunxia Sui1, Michael J Gerdes1, Sean Dinn1, Rong Zhang1, Sireesha Kaanumalle1, Elizabeth McDonough1,
Christina Lowes1, Craig Allred2, Fiona Ginty1, Thomas Foo1 and Puay-Hoon Tan3. 1GE Global Research, Niskayuna, NY; 2GE
Healthcare, Aliso Viejo, CA and 3Singapore General Hospital, Singapore, Singapore.
Body: Breast cancer is the most common malignancy in Singapore women with rising incidence across all ethnic groups
(Chinese, Malays and Indians). Ductal carcinoma in situ (DCIS) is the earliest non-invasive stage of disease and has been shown
to account for approximately 26% of diagnoses in all women participating in the Breast-Screen Singapore program. Despite
availability of breast screening, there are still Singapore women presenting with locally advanced breast cancer. The goal of the
current study was to investigate pathway and immune cell heterogeneity in low, intermediate and high nuclear grade DCIS using
a newly developed method for in situ hyperplexed analysis of multiple proteins in a single FFPE tissue (MultiOmyx). FFPE
samples from patients (n= 15) diagnosed with DCIS were provided by Singapore General Hospital. Patients were of Chinese
origin, ranged from 50-59 years, were all post-menopausal, and included low (n=5), intermediate (n=5) and high grade (n=5)
samples. All histological diagnoses were reviewed by a single pathologist. Following a single antigen retrieval step, DAPI and
cytokeratin staining was conducted and imaged at 10X. Based on DAPI, cytokeratin and autofluorescence, an H&E-like image
was generated for each sample. Using this image, approximately 30 regions of interest (ROI) were selected per slide/patient,
including normal and DCIS regions with and without immune cells. In total, 15 biomarkers were imaged following an iterative
process of IF staining and dye inactivation. These included CD4 (T-cells), CD8 (T-cells), CD20 (B-cells), CD68 (macrophages),
CD10 (myoepithelial cells), CD44v6 (cancer stem cell), Her2, Her4, EGFR (ErbB family proteins), pmTOR (cell growth), SLC7A5
(Mammostrat) and epithelium and cell segmentation markers (pan-cadherin, pan-cytokeratin, S6 and Na+K+ATPase). Images
were evaluated for quality and processed to generate single cell quantification data for each marker in the epithelium and stroma.
Single cell data was compared with clinical and histological features (grade, DCIS/ low immune cells, DCIS/ high immune cells).
Cell-based k-means clustering was then performed for biomarkers in both the epithelial cells (Her4, Her2, panCK, SLC7A5,
EGFR, NaKATPase, pmTOR, and CD10) and stromal cells (CD4, CD8, CD20, CD68). Using consensus clustering, the optimum
number of clusters was found to be six in both the epithelium and stroma. No distinct clusters were associated with histological
grade. Normal ROIs tended to be associated with higher CD10 and EGFR, moderate CD20 and low or negative for other
markers. DCIS with low immune cells tended to have higher Her4, pmTOR and moderate CD68 while DCIS with high immune
cells tended to have high Her2 and moderate to high CD4, CD20 and CD68. A larger study is needed to associate the findings
with outcome. This technology provides a way to elucidate mechanisms of early disease and illustrates the complex inter-play
between grade, pathway activation and immune response.

Title: Gene-expression molecular subtyping of immunohistochemistry-typed triple-negative breast cancer tumours
Mario Campone1, Loc Campion1, Delphine Loussouarn2, Isabelle Valo1, Catherine Gurin-Charbonnel1, Antoine Vanier1, Wilfried
Gouraud1, Catherine Guette1, Vronique Verrile1, Agns Chassevent1 and Pascal Jzquel1. 1Institut de Cancrologie de l'Ouest,
Saint Herblain, France and 2Hopital G&R Lannec, Saint Herblain, France.
Body: BACKGROUND
Recently, it has been shown that annotation of triple-negative breast tumours (TNBC) by means of immunohistochemistry (IHC)
or gene-expression signatures (GES) gave different results: 20 to 30% of these tumours were not basal-like, but luminal A, B or
HER2-E. In this study, we aimed at identifying gene-expression molecular subclassification of IHC-TNBC.
MATERIALS AND METHODS
Patients
Studied bi-centric cohort retrospectively included 107 randomly selected women whose primary breast tumours lacked IHC
expression of estrogen receptor, progesterone receptor and HER2. Patients were diagnosed and treated primarily between 1998
and 2007 at the ICO-Gauducheau (n=65) and the ICO-Papin (n=42).
Gene expression profiling
Gene expression analysis was performed using Affymetrix Human Genome U133 Plus 2.0 Arrays that correspond to
approximately 20,000 genes.
Statistical analysis
Unsupervised analysis was performed by means of fuzzy clustering. Independent IHC-TNBC cohort (GSE21653; n=87) was used
for external validation.
"Fuzzy cluster" functional annotation
To annotate "fuzzy-clusters", we used clinicopathologic characteristics, gene-expression signatures (GES) (PAM50, Proliferation
score, TNBCtype, Immune response, Claudin-low), Gene Ontology enrichment and IHC (CK5, CK5/6, HER1, AR, Ki-67, FOXA1).
RESULTS
Fuzzy clustering partition individualized 3 clusters in our cohort: C1, C2 and C3. "Fuzzy cluster" functional annotation results are
displayed in Table 1. C1 was composed of a mixture of non-basal-like subtypes. C2 and C3 were basal-like and IHC-TNBC.
Except an adenoid cystic case, C2 was exclusively composed of basal-like subtypes. C3 was essentially characterized by
immune response and included 26% of claudin-low subtypes. For this reason, C3 should rather be named "immune response
cluster". External validation confirmed our results.
Table1. "Fuzzy cluster" functional annotation results.
N "fuzzy cluster"
C1 (n=24)
C2 (n=48)
C3 (n=35)
P-value
Age (mean years)
64.6
56.8
51.9
<0.05
SBR
Low
High
<0.05
NPI
Low
High
<0.05
Clinicopathologic characteristics
Gene-expression signatures
PAM50
Luminal B (61%), A (22%)
Basal-like (98%)
Basal-like (91%)
Proliferation score
Low
High ++
High +
TNBCtype
Luminal androgen receptor
Basal-like 1, mesenchymal
Immune response
Immune response
Steroid hormone response
Immune response
Immune response
Claudin-low
0%
Low frequency (2%)
Higher frequency
(26%)
Gene Ontology enrichment
<0.05
Biological processes
Epithelial cell differentiation, hormone

metabolic process
Cell adhesion, locomotion,

chemotaxis
Immune response
Immunohistochemistry
CK5 and/or HER1
positive
50%
91%
<0.05
AR positive
73%
5%
<0.05
Ki-67 positive
29%
87%
<0.05
FOXA1 positive
83%
9.6%
<0.05
CONCLUSION
Heterogeneity within basal-like and IHC-TNBC is still controversial and require further research to understand the complexity of
the disease, and to identify molecular drivers that can be therapeutically targeted. Annotation discordances between IHC and
GES clearly indicate that a robust molecular subtyping method must be found. Correct molecular assignment would permit to
orientate 20 to 30% of IHC-TNBC patients towards targeted therapy (hormonotherapy or anti-HER2).

Title: Accuracy of estrogen receptor, progesterone receptor, HER2 Status and Ki67 labeling index between core needle and
surgical excisional tumour in 910 patients with breast cancer
Rikiya Nakamura1, Naohito Yamamoto1, Yoshiyuki Itookubo1, Ayako Nakagawa1 and Makiko Itami1. 1Chiba Cancer Center
Hospital, Chiba, Japan.
Body: Purpose
Accurate determination of ER, PgR, HER2 status and Ki67 labeling index was very important in making decision for adjuvant and
neoadjuvant treatment for patients with breast cancer. However discordance of ER, PgR, HER2 status and Ki67 labeling index
(Ki67-LI) between core needle biopsy (CNB) and surgical specimens (SSp) varied among reported studies.
The aim of the present study was to compare the accuracy of CNB with that of SSp for ER, PgR,HER-2 status, Ki67-LI detection
in breast cancer.
Patients and methods
All patients diagnosed with an early breast cancer in our Cancer Center Hospital between January 2005 and May 2014 was
included but exclusion criteria of patients with large tumour requiring neoadjuvant chemotherapy. All CNB were performed under
ultrasound guidance, with at least four 14-gauce core biopsies being obtained for pathological examination. ER, PgR, HER-2
status and Ki67-LI were assessed in CNB and in SSp. ER and PgR were determined by
Immunohistochemistry (IHC). The cut points for ER and PgR positive was10%. The cut point for Ki67-LI high expression was
20%. HER2 was determined by IHC and scored from 0 to 3+. FISH analysis was carried out in HER2 2+ cases and in discordant
cases.
Results
A total of 910 patients were assessed. CNB can be used with confidence for ER and HER2 determination. Rates for sensitivity,
specificity, negative predictive value and positive predictive value for CNB compared with SSp are 99.3%,95.6%,97.2%,98.5% for
ER, 94.5%,81.8%,88.2%,90.7% for PgR ,93.0%,99.1%,99.6%,93.9% for HER2, and 87.7%,69.8%,81.1%,79.4% for Ki67-LI,
respectively. Specially, the most impact factor of discordance for Ki67-LI and HER2 is Tumour size and that for ER and PgR is
lobular carcinoma.
Conclusion
CNB can be used with confidence for ER and HER2 determination. For PgR and Ki67-LI due to substantial discordance results
from CNB should be used with caution.


Title: Utility of simultaneous HER2 protein and gene assessment for the evaluation of discrepancy and intratumoral heterogeneity
of HER2 status and the prediction of prognosis in invasive breast cancer using the gene-protein assay (GPA)
Sasagu Kurozumi1, Mary Padilla2, Masafumi Kurosumi3, Hiroshi Matsumoto1, Yuji Hayashi1, Katsunori Tozuka1, Shigenori E
Nagai4, Kenichi Inoue4, Hanako Oba3, Jun Horiguchi5, Izumi Takeyoshi5, Jim Ranger-Moore2, Eslie Dennis2 and Hiroaki Nitta2.
1
Saitama Cancer Center, Saitama, Japan; 2Ventana Medical Systems, Inc, Tucson, AZ; 3Saitama Cancer Center, Saitama,
Japan; 4Saitama Cancer Center, Saitama, Japan and 5Gunma University Graduate School of Medicine, Gunma, Japan.
Body: Background:
The eligibility of patients for HER2-targeted therapies is determined by the evaluation of HER2 gene amplification and HER2
protein overexpression. The gene-protein assay (GPA, Ventana Medical Systems, Inc., USA) is a new method for simultaneous
evaluation of HER2 immunohistochemistry (IHC) and dual in situ hybridization (DISH) using a single tissue section. In this study,
we investigated the relationship between HER2 IHC and DISH results evaluated by GPA. In addition, we analyzed the correlation
between HER2 status and prognosis of invasive breast cancer patients.
In this study, invasive carcinoma tissues of 280 consecutive patients treated in Saitama Cancer Center in 2000-2001 (median
follow-up: 130 months) were examined. Among HER2 positive initial samplings, no patients originally received adjuvant
trastuzumab therapy. However, 76% of HER2 positive cases of recurrence received trastuzumab therapy. GPA was performed on
a section of routinely processed primary tumor and the status of HER gene and protein were separately evaluated in whole areas
of tumor sections using the following FDA criteria: DISH (negative: HER2/CEN17 < 2, positive: HER2/CEN17 2.0) and IHC
(score 0 to 3+). In IHC score 2+ patients group, final HER2 positivity was decided according to DISH results using criteria of FDA
criteria. Recurrence-free survival (RFS) and cancer-specific survival (CSS) stratified by IHC and DISH results were analyzed. In
addition, patterns of heterogeneity were grouped according to the following 4 phenotypic and genotypic types: A) IHC 2+/DISH+;
B) IHC 2+/DISH-; C) IHC 1+ & 0/DISH+; and D) IHC 1+ & 0/DISH-. The presence of heterogeneity in relation to prognosis was
analyzed in the IHC 0 & 1+/DISH- group.
Results:
The HER2 IHC 3+ group (27.5%), both with or without trastuzumab therapy, had significantly worse survival than HER2 IHC 1+ &
0 group (RFS: P=0.0039; CSS: P=0.0362) and HER2 DISH+ group (27.5%) had significantly worse survival than HER2 DISHgroup (RFS: P=0.0056; CSS: P=0.0497). HER2 positive group defined by FDA criteria had significantly worse RFS than HER2
negative group (P=0.0211). HER2 IHC 1+ & 0/DISH+ group had significantly worse RFS than IHC 1+ & 0/DISH- group
(P=0.0208). In the HER2 IHC 1+ & 0/ DISH- group, patients with heterogeneity (33 cases) had significantly worse survival than
those without heterogeneity (RFS: P=0.0176; CSS: P=0.0199).
Conclusions:
HER2 GPA technology might be useful for evaluating the discrepancy and heterogeneity of HER2 IHC and DISH results at single
cell levels simultaneously and the presence of HER2 tumor cell heterogeneity might be a potent prognostic factor in HER2
negative breast cancer patients. Further clinical research must be conducted for clarification of the relationship between the
presence of HER2 intratumoral heterogeneity and the effectiveness of HER2-targeted therapies.

Title: No Show

Title: The non-coding transcriptome of hypoxic breast cancer: Novel insights of clinical relevant long non-coding RNA in hypoxia
signalling
Hani Choudhry1,3, Johannes Schodel2, Ashwag Albukhari3, Syed Haider3, Francesca Buffa3, Peter J Ratcliffe4, David R Mole4,
Ioannis Ragousis5 and Adrian L Harris3. 1King Abdulaziz University, Jeddah, Saudi Arabia; 2Friedrich-Alexander-University,
Erlangen-Nuremberg, Germany; 3University of Oxford, Oxford, United Kingdom; 4Henry Wellcome Building for Molecular
Physiology, University of Oxford, Oxford, United Kingdom and 5McGill University and Genome Quebec Innovation Centre,
Montreal, Canada.
Body: Hypoxia is associated with aggressive and poor prognosis of breast cancer. Generally, pan-genome analyses of hypoxia
have focussed on protein-coding genes, however, the role of non-coding RNAs, in particular long non-coding RNAs (lncRNA) in
hypoxia is not well characterised. We undertook an integrated genomic analysis of the hypoxic transcriptome in MCF7 breast
cancer cells, employing total RNA-seq together with ChIP-seq for the hypoxia-inducible transcription factor (HIF) and for
epigenetic marks of transcriptional activation (RNApol2 and histone H3K4me3).
Analyses revealed that all classes of RNA are significantly regulated by hypoxia including piwiRNA, miRNA, tRNA, and
sn/snoRNA. Significant numbers of lncRNAs were upregulated in hypoxia and were associated with increased RNApol2 and
H3K4me3 markers and with HIF binding, indicating direct transcriptional activation of lncRNAs by HIF. The most hypoxiclly
upregulated lncRNA was NEAT1, which is a direct transcriptional target of HIF-2a but not HIF-1a. The role of NEAT1 in cancer
has not been previously studied.
We demonstrated that hypoxic NEAT1 induction is common in breast cancer cell lines and xenografts models treated with
bevacizumab. NEAT1 directly induces the formation of nuclear paraspeckle bodies in hypoxia. Moreover, it contributes to
tumourigenicity by increasing cell proliferation, colony formation, and reducing apoptosis. In addition, we report that NEAT1 is
required to retain hypoxia induced hyper edited Junctional Adhesion Molecule A (JAM-A) mRNA in the nucleus, thus preventing
export into the cytoplasm for translation. Finally, in a large cohort of 2000 breast cancers, high levels of NEAT1 were associated
with poor clinical outcomes and clinicopathological features.
Our results extend knowledge of the hypoxic transcriptional response into the spectrum of non-coding transcripts. These findings
provide novel mechanisms of transcriptional regulation in hypoxia and open new avenues to find novel pathways and targets to
develop therapies for breast cancer.

Title: Preliminary translational results from PROMIX, a phase II trial of preoperative chemotherapy with the addition of
bevacizumab in large operable and locally advanced HER2-negative breast cancer
Niklas Loman1,5, Ida Johansson5, Judith Bjhle3, Jan Frisell2, Tobias Lekberg3, Lisa Rydn4, 5, Anna von Wachenfeldt3, Jonas
Bergh3, Thomas Hatschek3 and Ingrid Hedenfalk5. 1Lund University Hospital, Lund/Malm, Sweden; 2Karolinska University
Hospital, Stockholm, Sweden; 3Karolinska University Hospital, Stockholm, Sweden; 4Lund University Hospital, Lund/Malm,
Sweden and 5Department of Clinical Sciences, Lund, Sweden.
Body: Background: Preoperative chemotherapy in breast cancer (bc) provides unique possibilities to evaluate effects of therapy
by studying response and changes in the tumor during the course of treatment. A pathologic complete response (pCR) correlates
positively with long term prognosis in high-proliferating bc. In triple negative bc (TNBC) the prognosis was still relatively serious in
cases with pCR in one large meta-analysis (Cortazar Lancet 2014).
Methods: 150 cases were included in this multicenter study, and treated with six cycles of epirubicin and docetaxel, adding
bevacizumab from cycle 3, before surgery. Core needle biopsies were collected at free hand or with ultrasound (US) guidance
and snap frozen at base-line and after cycle 2, tissue was also collected at surgery. Subtyping was performed using
immunohistochemistry (IHC) of ER, Ki67 and HER2 according to modified St Gallen criteria; and using bead array gene
expression profiling (GEX) according to PAM50.
Results: Biopsies were successfully retrieved from 145/150 pts at baseline, 138 after cycle 2 and 139 at surgery. The mRNA yield
was adequate for GEX from 123/145 (85%) at baseline, 82/138 (59%) at cycle 2 and 71/139 (51%) at surgery, the decrease being
a result of tumor shrinkage during treatment.
Initial PAM50 subtypes were as follows: luminal A (LA) 20%, luminal B (LB) 45%, HER2 5 %, basal like (BL) 22% and normal like
(NL) 8%. PAM50 at baseline differed compared to IHC subtypes. Among IHC defined LA-like cases 15/33 (45%) were classified
as LB by PAM50. Similarly, among IHC LB-like 22/57 (39%) were classified as non-LB (6 basal, 8 LA, 3 HER2 and 5 NL), while
among IHC TNBC 7/28 (25%) were classified as non-BL subtypes (1 LA, 3 HER2 and 3 NL).
Of the pts with a baseline GEX analysis, 17 (14%) achieved a pCR. The observed pCR rates among PAM50 subtypes were: LA
8%, LB 5%, HER2 17%, BL 53% and NL 20%. For non-pCR cases, 39/52 (75%) of the tumors changed PAM50 subtype between
baseline and surgery. The majority changed to the NL subtype. 33% of the LB tumors changed to the LA subtype.
Currently, after 2.2 years of follow-up, 16 pts are deceased due to bc. Among BL cases, 6/9 pts with a pCR at surgery remain
alive; while 3/9 have died from bc. Exploratory analyses using functional gene modules (Desmedt Clin Cancer Res 2008) suggest
that patients with BL tumors who have died have higher scores for PLAU/invasion and lower scores for STAT1/immune response
compared with those who are still alive. Tumor size at baseline did not obviously correlate with outcome.
Conclusion: We show that biological material can be retrieved from a substantial fraction of cases treated within a multicenter
study of preoperative chemotherapy. The success rate may be ameliorated by routine use of US guidance. The distribution of
subtypes differs between modified IHC St Gallen criteria and PAM50, especially within the luminal subtypes. The pCR rate is
highest among cases with a BL tumor at baseline. Shift of the gene signature between different subtypes during the course of
treatment is frequent. In this set of relatively large tumors, the prognosis among BL bc appears to be adverse in spite of a pCR.

Title: Anastrozole and everolimus in hormone receptor-positive metastatic breast cancer: Safety profile, activity and associations
of molecular alterations in the PI3K/AKT/mTOR pathway
Jennifer J Wheler1, Filip Janku1, Stacy L Moulder1, Aung Naing1, Sarina A Piha-Paul1, Gerald S Falchook1, Ralph G Zinner1,
Apostolia M Tsimberidou1, Siqing Fu1, David S Hong1, Johnique T Atkins1, Roman Yelensky2, Vince Miller2, Philip J Stephens2,
Vincente Valero1, Funda Meric-Bernstam1 and Razelle Kurzrock3. 1UT MD Anderson Cancer Center, Houston, TX; 2Foundation
Medicine, Cambridge, MA and 3UC San Diego, Moores Cancer Center, La Jolla, CA.
Body: Background
Combining aromatase inhibitors with PI3K/AKT/mTOR inhibitors in patients with hormone receptor (HR)-positive metastatic breast
cancer has demonstrated clinical efficacy. There is limited data on associations between molecular signatures and activity.
We evaluated the combination of anastrozole and everolimus in 56 patients with HR-positive, metastatic breast cancer. The
primary objective was to establish safety and maximum tolerated dose (MTD). Dose limiting toxicities (DLTs) were defined as
serious grade 3 or 4 toxicities related to treatment that occurred during cycle 1. Dose level 1 was anastrozole 1mg PO QD and
everolimus 5 mg PO QD and dose level 2 was anastrozole 1 mg PO QD and everolimus 10 mg PO QD (a dose level -1 included
everolimus 2.5 mg PO QD). Secondary endpoints included evaluation of antitumor activity and molecular associations with
response. When tissue was available, Next Generation Sequencing (NGS) was performed using genomic libraries selected for all
exons of 236 (or 182) cancer-related genes sequenced to average depth of >500 in a CLIA laboratory (Foundation Medicine,
Cambridge, MA, USA). An analysis was then performed for all classes of genomic alterations.
Results
The median age was 59 (range, 37-82) and the median number of prior therapies in the metastatic setting was 3 (range, 0-13).
The initial oral daily dose of anastrozole 1 mg oral and everolimus 10 mg PO daily was well tolerated. Five dose-limiting toxicities
(DLTs) were seen at full doses, including grade 3 thrombocytopenia (1 patient), grade 3 neutropenia (1 patient), grade 3
increased liver enzymes (1 patient), grade 3 hyperglycemia (1 patient) and, grade 3 mucositis (1 patient). The most common
grade 3 or 4 treatment-related toxicities were neutropenia (5%), increased liver enzymes (5%), and hyperbilirubinemia (3%). Of
the 56 patients on study, 36 were tested for at least one molecular alteration in the PI3K/AKT/mTOR pathway. Twelve of these 36
patients had NGS analysis of their tumor tissue. Eighteen of 36 patients (50%) tested had at least one alteration in the pathway,
including mutations in PIK3CA (n=16), PIK3R1 (n=1), and AKT1 (n=2); PTEN protein loss (n=1); and, AKT3 amplification (n=1).
Sixteen of 56 evaluable patients (29%) achieved stable disease (SD) /partial response (PR)/complete response (CR) 6 months
(n = 3 (5%) with PR/CR). Thirteen of the 16 patients who achieved SD/PR/CR 6 months were tested for a genetic alteration in
PI3K/AKT/mTOR pathway and 7 of these patients (54%) had at least one alteration in the pathway, including mutations in
PIK3CA (n=6), PIK3R1 (n=1), and AKT1 (n=1); PTEN loss (n=1); and AKT3 amplification (n=1).
Conclusions
Combination anastrozole 1 mg and everolimus 10 mg is well tolerated and is active in heavily-pretreated patients with HR-positive
breast cancer. The presence of a molecular alteration in the PI3K/AKT/mTOR pathway did not predict for clinical activity of this
combination.

Title: Gene expression associated with poor prognosis of young TNBC patients
Thomas Karn1, Lajos Pusztai2, Cornelia Liedtke3, Giampaolo Bianchini4, Balazs Gyrffy5, Christos Hatzis2, Achim Rody3, Volkmar
Mller6, Marcus Schmidt7, Uwe Holtrich1 and Sven Becker1. 1Goethe-University, Frankfurt, Germany; 2Yale Cancer Center, New
Haven, CT; 3University Hospital Lbeck, Germany; 4Ospedale San Raffaele, Milan, Italy; 5Hungarian Academy of Sciences,
Budapest, Hungary; 6University Hospital, Hamburt, Germany and 7Johannes Gutenberg University Mainz, Mainz, Germany.
Body: Background:
Among TNBC patients those of very young age (<40 years) display a significantly worse prognosis (Liedtke et al. 2013 Breast
Cancer Res Treat). We verified this result in 1161 TNBC samples with full Affymetrix gene expression data of which 845 patients
had both detailed age and follow up information.
We split the full sample set into a finding cohort of 394 TNBC and a validation cohort of 767 TNBC encompassing 309 and 536
samples, respectively, with both age and follow up data. We then used significance analysis of microarrays (SAM) in the finding
cohort to look for genes whose expression is associated with young age (<40 years). Identified genes were analyzed for their
correlations to known metagenes characteristic for different TNBC subtypes. Subsequently the whole analysis was repeated in
the validation cohort.
Results:
We identified 98 and 222 probesets by SAM in the finding and validation cohort, respectively. Only a subset of identical probesets
(19.4%) was re-identified in the validation. However, the gene lists were similarily enriched for correlations to specific metagenes
and the respective TNBC subtypes. We found that young age among TNBC is positively correlated with increased proliferation
and a basal-like subtype, but negatively correlated with the molecular apocrine phenotype and the claudin-low subtype. We also
observed a negative correlation of young age with stromal enrichment and EMT but no difference in lymphocyte infiltration as
judged by specific metagenes. However, despite that TNBC of patients <40 years have a poor prognosis (P=0.006) and are
clearly enriched in basal-like subtype (70.7% vs 55.3%; P=0.001), still basal-like TNBC do not differ from non-basal-like TNBC in
prognosis. Moreover, even within the group of basal-like TNBC the prognostic effect of age<40 was still observed (P=0.008).
Interestingly, the changes in expression of single genes and metagenes were observed over a wide age range of 35-85 years. In
contrast the difference in prognosis was markedly pronounced at the 40 year age limit. Thus, potential critical differences in gene
expression associated with poor prognosis of TNBC <40 years seem to be concealed by many age-associated changes without
prognostic value.

Title: A four gene signature predicts anthracycline benefit: Evidence from the BR9601 and MA5 breast cancer trials
Melanie Spears1, Nicola S Lyttle1, Fouad Yousif1, Alison F Munro2, Christopher Twelves3, Kathleen I Pritchard4,5, Mark N Levine6,
Lois Shepherd7 and John MS Bartlett1,2,5. 1Ontario Institute for Cancer Research, Toronto, ON, Canada; 2Edinburgh Cancer
Research Centre, Edinburgh, United Kingdom; 3University of Leeds and Cancer Resarch UK Centre, Leeds, United Kingdom;
4
Sunnybrook Odette Cancer Centre, Toronto, ON, Canada; 5University of Toronto, Toronto, ON, Canada; 6McMaster University
and Hamilton Health Sciences, Hamilton, ON, Canada and 7NCIC Clinical Trials Group, Queen's University, Kingston, ON,
Canada.
Body: Background: Chromosome instability (CIN) in solid tumours is associated with poor prognosis and results in numerical and
structural chromosomal aberrations. Recent evidence from both the BR9601 and MA.5 trials has demonstrated CEP17
duplication as a predictive marker of anthracycline benefit. CIN25 and CIN70 gene expression profiles have previously been
published and predict survival response. An analysis of the BR9601 and MA5 clinical trials was performed to test the role of CIN
gene expression signatures as a marker of anthracycline sensitivity.
Methods: RNA was extracted from patients in both the BR9601 and MA5 studies and analysed through Nanostring technology.
Log-rank analyses explored the prognostic values of the signatures on distant relapse-free survival (DRFS). Cox-regression
models tested independent prognostic value on DRFS in the presence of treatment, age, tumour size, nodal status, ER status
and grade, and treatment by marker interactions.
Results: Of the 761 samples available from the BR9601 and MA5 cohorts we successfully analysed 703 (92.4%). High CIN25 and
CIN70 scores were associated with age (p<0.0001), grade (p0.0001), PgR negativity (p<0.0001) and ER negativity (p<0.0001). In
univariate analysis, high CIN25 score was associated with decreased DRFS (HR: 0.74, 95% CI 0.54-0.99, p=0.046). In a
multivariate analysis with adjustment for size, nodal status, ER, pathological grade, HER2, CIN25, treatment and treatment by
marker only pathological grade, nodal status and tumour size were significant predictors of outcome.
A more limited set of genes that reflected CIN was established by examining the expression profile of the genes and clustering
them. The combined cohort was split into a 60% training and 40% validation set. The area under the curve (AUC) was calculated
and the gene signature with the greatest AUC was selected and termed CIN4. Patients with low CIN4 score benefited from
anthracycline treatment compared to those that had high CIN4 score (HR 2.72, 95% CI 1.48-5.02, p=0.001). No significant benefit
with CMF treatment was observed in (HR: 1.02, 95% CI 0.58-1.82, p=0.92). After multivariate analysis the treatment by marker
interaction for CIN4 had a hazard ratio of 2.10 (95% CI 2.18-30.38, p= 0.001).
Conclusion: High CIN70 and CIN25 scores were associated with an aggressive phenotype and showed a potential increased
sensitivity to anthracycline therapy compared to those with low CIN scores. CIN4 was an independent predictor of anthracycline
benefit for DRFS. However, further work in larger patient cohorts such as NEAT is warranted.

Title: Identification of specific gene signatures and alternative splice variants using exon array in Inflammatory breast Cancer
Florence Lerebours1, Sophie Vacher2, Marie Le Cann2, Sophie Rondeau2, David Gentien2, Steven Van Laere3, Franois Bertucci4,
Pierre de la Grange5 and Ivan Bieche2. 1Institut Curie, Hopital Rene Huguenin, Saint-Cloud, France; 2Institut Curie, Paris, France;
3
General Hospital Sint Augustinus, Wilrijk, Belgium; 4Institut Paoli-Calmettes, Marseille, France and 5Genosplice, ICM, Hopital
Pitie Salpetriere, Paris, France.
Body: Inflammatory Breast Cancer (IBC) is a rare form of breast cancer with a particular phenotype and aggressiveness.
Although significant progress has been made in the management of IBC, the prognosis remains poor. Intensive research on IBC
has allowed identification of several candidate genes and pathways, but molecular findings unique to IBC have yet to be identified
to improve treatment and survival. Transcriptomic studies have revealed a marked heterogeneity of IBC with limited gene overlap
between studies, preventing the identification of a sensitive and specific signature in IBC.
In order to detect both gene expression levels and alternate RNA splice isoforms, we chose to perform splice-sensitive array
profiling using Affymetrix Exon Array in a well-defined series of 33 IBC (core biopsies from previously untreated T4d carcinomas)
compared with 28 stage I to non-inflammatory stage III breast cancers.
Gene expression analysis allowed the identification of classical molecular subtypes in both IBC and non-IBC, with
overrepresentation of basal-like (17/33 vs 9/28) but no luminal A in IBC (vs 11/28). Based on Fold-Change (FC) > 1.5 and p-value
< 0.05, 495 genes were significantly dysregulated between IBC and non-IBC, including in particular up-regulated hemoglobin
genes and down-regulated ER-related genes in IBC compared to non-IBC. Most activated pathways were hematopoeitic cell
lineage, cytokine-cytokine receptor interaction, chemokine signalling pathway and complement and coagulation cascade. We
defined a 21-gene signature discriminating IBC from non-IBC with 8% error rate. To get rid of genes associated with molecular
subtypes like ER-related genes, an analysis restricted to basal-like BC (17 IBC compared to 9 non-IBC) was performed allowing
the definition of a 29-gene signature discriminating the whole groups of IBC from non-IBC with 3.3% error rate. Validation of this
signature in the gene expression datasets from the World IBC Consortium will be presented.
Specific exon expression analysis revealed that when based on FC splicing-index and p-value, 266 exons representing 177
distinct genes were differentially regulated between IBC and non-IBC. After manually curation of results, 13 splice events
representing 12 distinct genes were retained as good candidates for alternative splicing in IBC (EVL, RPL10, MYH10, HSPA8,
DOCK7, DIDO1, RPL4, TRAK1, RGS1, LMO4, SMARCA4, ZNF337).
To confirm gene-signatures specific to IBC, altered pathways and major splice variants, we are performing a validation study
using quantitative RT-PCR in the screening set (n=33) and in an independent series of 140 IBC compared to 200
non-stage-matched non-IBC. Finally the most dysregulated genes will also be studied at the protein level using
immunohistochemistry.

Title: Feasibility and sensitivity of fine-needle aspiration biopsies for the detection of somatic mutations using next-generation
sequencing in breast cancer
Han-Byoel Lee1, Jisun Kim2, Kyung-Min Lee3, Je-Gun Joung4, Hae-ock Lee4, Min Kyoon Kim1, Eunshin Lee1, Jongjin Kim1,
Tae-Kyung Yoo1, Yun-Gyoung Kim1, Young Joon Kang1, Han Suk Ryu5, In-Ae Park5, Hyeong-Gon Moon1, Dong-Young Noh1,
Woong-Yang Park6 and Wonshik Han1. 1Seoul National University College of Medicine, Seoul, Korea; 2Asan Medical Center,
Seoul, Korea; 3Cancer Research Institute, Seoul National University College of Medicine, Seoul, Korea; 4Samsung Genome
Institute, Samsung Medical Center, Seoul, Korea; 5Seoul National University College of Medicine, Seoul, Korea and 6Samsung
Genome Institute, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea.
Body: Background/Purpose: Next-generation sequencing (NGS) is being incorporated rapidly into clinical practice. Fine-needle
aspiration biopsy (FNAB) specimens have been used feasibly in molecular analysis including direct sequencing and microarrays.
They are readily available and enriched in malignant cells, thus providing opportunities for genomic analysis for more clinical
samples. In this study, we assessed the feasibility and sensitivity of FNAB for the detection of somatic mutations by NGS
compared to bulk tissue.
Methods: Bulk tissue and FNAB was sampled via skin superficial to the palpable tumor from surgically resected breast cancer
specimen. DNA was extracted from the bulk tissues and FNAB samples obtained from twelve patients. Somatic mutations
detected from whole exome sequencing (WES) by next-generation sequencing (NGS) (HiSeq 2500, Illumina) were analyzed for
corresponding pairs of bulk tissue and FNAB. Verification of somatic mutations detected exclusively from FNAB and known to be
clinically relevant to breast cancer was carried out by Sanger sequencing. Invasive tumor percentages of bulk tissues were
evaluated using hematoxylin and eosin (H&E)-stained sections.
Results: Average depth of coverage were 158.8x and 158.3x for bulk tissue and FNAB, respectively. Number of detected somatic
mutations ranged from 2 to 153 (median 18.5) and 19 to 210 (median 39.5) for bulk tissue and FNAB, respectively. Ten
specimens had more mutations detected exclusively from FNAB than from bulk tissue. Allele fractions plotting of corresponding
pairs of bulk tissue and FNAB showed good, intermediate, and poor correlation in five, two, and five specimens, respectively.
H&E-stained sections of bulk tissue from the five specimens with good correlation contained an invasive tumor percentage of 45
to 98%, whereas those from five specimens with poor correlation contained 0 to 25%. Three of the poorly correlated bulk tissues
were judged to have 0% of invasive tumor. Among mutations detected exclusively from FNAB, eighteen different genes of interest
in 22 foci were evaluated for both FNAB and corresponding bulk tissue by Sanger sequencing. In the results, three mutations
(PIK3CA, TP53 x2) were verified in FNAB samples but not in the bulk tissue.
Conclusion: WES was successfully carried out in all pairs of bulk tissue and FNAB from twelve breast cancer patients. In samples
with high tumor content somatic mutation profiles showed high correlation between the two samples whereas samples with low
tumor content failed to show correlation. The failure was mostly due to the scarcity of tumor portions in the bulk tissues, indicating
that FNAB more reliably retained malignant tumor portion. This study suggests that FNAB is an easy and feasible method, and
furthermore, provides a more reliable specimen for NGS analysis where somatic mutations could be identified for potential
prognostic or therapeutic benefits.


Title: Functional mapping of the oncogenic kinome activity of triple-negative breast cancer cells
Bo Pan1,2, Miki Mori1,3, Pei Rong Evelyn Lee1, Marij Hartog1, Qiang Sun2, Laura van 't Veer1 and Jean-Philippe Coppe1.
1
Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA; 2Peking Union Medical College
Hospital, Beijing, China and 3Showa University, 5-8 Hatanodai 1-chome, Shinagawa-ku, Tokyo, Japan.
Body: Background: Triple-negative breast cancer (TNBC) accounts for approximately 10 to 15% of all breast cancer cases. The
management of TNBC cases will significantly improve once molecular mechanisms specific to TNBC cells will be identified and
treated accordingly. Evidence suggests that TNBC cells display deregulated kinase-dependent signaling cascades that differ from
non-triple-negative breast cancer cells. We hypothesized that uniquely divergent phospho-circuits could be distinguished between
TNBC and non-TNBC cell lines. By revealing such unique, dysfunctional phospho-signaling network, our long-term objective is to
identify kinases that underpin triple-negative breast cancer development, and can be inhibited using targeted therapy.
Methods: The kinome activity of TNBC and non-TNBC cell lines was identified (HCC70, MDA-MB-231, MDA-MB-436 compared
to AU565, MCF-7, T47D). The functional phospho-signature of each breast cancer cell was analyzed using a high throughput
experimental platform that monitors the level of activity of myriad kinases at once. This technique uses 242 phospho-sensing
probes and 78 controls in an aqueous-based assay to simultaneously and directly measure the phospho-catalytic activity of
phosphorylating enzymes in cell lysates. We mapped the most significantly deranged phospho-signaling cascades and the
related kinases.
Results: Using 6 cell lines tested under various conditions, we generated 72 phospho-signatures, out of a total of 23,040 data
points. After validating the repeatability and robustness of the assay, the kinase activity signature of each breast cancer cell line
was analyzed and compared to each other using unsupervised hierarchical clustering. The phospho-sensing assay revealed the
heterogeneity of kinase activity networks among breast cancer cells. These data also established that phospho-signaling
cascades related to AKT, ERK, and SRC kinases were differentially altered in TNBC and non-TNBC cell lines.
Conclusions: We successfully identified unique phospho-circuits of TNBC and non-TNBC cell lines. Our goal is now to test
whether specific kinase inhibitors can efficiently kill or prevent the growth of TNBC cell lines in culture and animal models. We will
expand our approach into a high-content, functional kinome-screening platform to characterize the phospho-fingerprint of breast
cancer cells and tissues, and explore the druggable, kinase-dependent mechanisms critical to triple-negative breast tumors.

Title: Explore the differences and relationships between normal and ADH, DCIS and IDC tissues in breast based on Raman
spectroscopy
Bing Han1, Haipeng Zhang1, Chao Zheng1, Ye Du1, Lijia Liang2, Weiqing Xu2 and Zhimin Fan1. 1First Hospital of Jilin University,
Changchun, Jilin, China and 2State Key Laboratory for Supramolecular Structure and Materials, Jilin University, Changchun, Jilin,
China.
Body: Background: Atypical ductal hyperplasia (ADH) is a premalignant breast lesion associated with an increased risk of breast
cancer which shares some but not all the features of ductal carcinoma in situ (DCIS). The mechanism that ADH changes to DCIS
and even invasive ductal carcinoma (IDC) is still not clear now. There is a pressing need to develop screening techniques that are
accurate, minimally invasive and, more sensitive to identify early neoplastic changes and hence improve early breast cancer
diagnosis. Our aim was to explore the differences and relationships between normal, ADH, DCIS and IDC lesions of the breast
based on biochemical characteristics determined by Raman spectroscopy
Methods: Frozen sections were collected from 39 patients (all ladies) who underwent surgical resection or mammotome biopsy at
the Department of Breast Surgery, the First Hospital of Jilin University. After operation the samples were immediately frozen at
-20-25 and two contiguous sections (6 m thickness) were cut from a sample by freezing microtome. One was stained with
haematoxylin and eosin for routine histopathological analysis; the same position of the other section was detected by Microscopic
confocal Raman spectrometer (HORIBA JY Lab800, 633nm) with its mirror image (the H&E section). After Raman measurement,
the sections were routinely processed, stained with H&E and histologically examined. Support vector machine (SVM) was used to
differentiate different breast lesions.
Results: A total of 475 Raman spectra were obtained from 9 normal breast tissues, 7 ADH, 8 DCIS, and 15 IDC breast tissues.
Pronounced mean Raman spectra differences were observed between normal tissues, ADH, DCIS and IDC tissues. The
significant features of normal tissues are 1301, 1438, 1652, and 1743 cm-1, these peak positions are attributed to lipids, and the
spectra profiles of normal tissues have no strong protein peaks. Most noticeable was the increased protein and reduced lipid
levels of ADH tissues compared to normal tissues. The peak relative intensity of 1158 cm-1 which is attributed to the vibrational
modes of C-C stretch of proteins has significantly increased from normal tissues to DCIS and IDC in breast. The major spectra
differences in ADH, DCIS and IDC spectra were evidenced by a red shift with a broad peak of CH2, the intensity of the stretching
vibration peak of carotenoids, a relatively strong band of amide-I, and the nuclear acid peak. ADH tissues had the largest
constituent variations between subjects. During the disease progression, IDC tissues have smaller inter-subject constituent
variations than DCIS and ADH tissues. The accuracies of SVM are 93% and 95% in discriminating normal and IDC, while only
50% and 51% in discriminating ADH and DCIS.
Conclusion: Malignant transformations in tissue are associated with complex biochemical changes and may provide an effective
way to evaluate the malignancy by Raman spectroscopy as it reveals differences between normal tissues, ADH, DCIS and IDC
tissues in the breast. Further study to explain the biochemical relationships between these differences will shed more light into a
better understanding of the mechanism by which ADH and DCIS convert to IDC.

Title: Genetic profiling of breast cancer confirms a pivotal role of EGFR pathway in the development of acquired resistance to
tamoxifen
Rabab M Gaafar1, Abeer A Bahnasy1, Osman M Mansour1, Nasr M Allahloubi1, Ibrahim A Malash1 and Sabry M Shaarawy1.
1
National Cancer Institute, Cairo University, Cairo, Egypt; 2National Cancer Institute, Cairo University, Cairo, Egypt; 3National
Cancer Institute, Cairo University, Cairo, Egypt; 4National Cancer Institute, Cairo University, Cairo, Egypt; 5National Cancer
Institute, Cairo University, Cairo, Egypt and 6National Cancer Institute, Cairo University, Cairo, Egypt.
Body: Background: Acquisition of resistance to tamoxifen remains a major drawback in the treatment of oestrogen receptor
(ER)-positive breast cancers. Aberrant expressions of some genes in the EGFR pathway were associated to the acquisition of
resistance in some studies.
Patients and methods: This prospective study was carried out on 157 female patients with hormone receptor positive, locally
recurrent inoperable and/or metastatic breast cancer who presented to the National Cancer Institute, Cairo University during the
period from October 2010 to October 2012. All patients received tamoxifen. Patients were divided into tamoxifen responsive and
refractory groups according to their response to therapy.
In an attempt to understand the contribution of EGFR pathway to the development of resistance to Tamoxifen, we assessed the
genetic profile of the EGFR pathway genes in the 2 groups. RNA was extracted from tumor and normal tissue samples obtained
from all patients and the expression level of 92 genes was evaluated using the SABioscience array (Qiagen) with four
house-keeping genes.
Results: Age ranged between 29 and 79 years but age 50 or > 50 did not correlate to resistance. Evaluation of Hormone
receptor status showed that 58.59% were positive for both ER and PR, 32.48% were ER positive, PR negative and 8.91% ER
negative PR positive. Neither hormone receptor status nor nuclear grade was correlated to drug resistance. There was a strong
correlation between response to tamoxifen and disease site as patients with bone only disease demonstrated noticeable good
and maintained response compared to those who had visceral involvement (p- 0.005). The expression levels of all genes were
assessed in both studied groups: the responders (the control) and the refractory (tested) groups. Fifty six genes were differentially
over-expressed in the refractory group compared to the responding group, among which only JAK1, COL1A1, GAB1, FN1 and
MKNK1 showed a significant difference between responders and refractory groups. Thirty four genes were differentially
expressed (reduced expression) in the refractory compared to the responders. Moreover, CYP2D6 *3, *4 were significantly
prevalent in the refractory group (86.6%), whereas variants *10/*10 and *10/*3 were more common in the in the responding group
(85.5%) (p = 0.027)
Conclusion: A panel of 5 genes (M JAK1, COL1A1, GAB1, FN1 and MKNK1) in EGFR pathway together with CYP2D6
polymorphisms could be used to predict patients response to tamoxifen though this has to be verified in an extended study
including larger sample.

Title: Signatures of endocrine resistance in the tamoxifen and exemestane adjuvant multinational trial (TEAM)-UK cohort
Jane Bayani1, Mary Anne Quintayo1, Cindy Q Yao1, Syed Haider1, Cassandra Brookes2, Paul C Boutros1,3, John MS Bartlett1 and
Daniel W Rea2. 1Ontario Institute for Cancer Research, Toronto, ON, Canada; 2Cancer Research UK Clinical Trials Unit,
University of Birmingham, Birmingham, United Kingdom and 3University of Toronto, Toronto, ON, Canada.
Body: Introduction: There are a number of commercially-available tests to stratify risk for women diagnosed with early breast
cancer. While such "Generation I" tests are increasingly being used, a consensus is growing that these tests are moderately
accurate in assessing risk. Moreover, Generation I tests fail to direct more personalized treatment. There exists, therefore, a clear
need for more informative "Generation II" tests that have theranostic targets. To this end, we have performed an mRNA
abundance-based analysis using the UK cohort of the TEAM trial to identify signatures of endocrine resistance, from which
pathways for putative therapeutic intervention may be identified.
Methods: RNA extracted from 790 patients in the UK-TEAM cohort were profiled using a 165-gene NanoString codeset. The
gene list was compiled from targets that comprise many of the existing risk assessment tests, in addition to genes known to be of
importance for breast cancer pathogenesis. Signal intensities were normalized using the R statistical environment; 336 different
combinations of preprocessing methods were assessed and the most optimal method selected using unbiased criteria.
Results: Univariate survival analysis revealed a number of significantly prognostic candidates. Using inter-gene correlation and
consensus clustering, we identified five gene clusters. Not surprisingly, these clusters included a strong proliferation, hormone
signalling and cell migration component. Derivation of risk scores using Cox proportional hazards model, with the inclusion of age
and nodal status, generated a signature identifying patients with differences in distant relapse-free survival (DRFS). Moreover, the
composition of the gene-list made it possible to characterize the patients into their intrinsic subtypes and to determine their
relative risk for recurrence relative to assessment tools available today. The added value of subtyping and the gene clusters
identified in this discovery cohort will be discussed.
Conclusions: The impact of test-guided therapy using multi-parametric tests is increasingly being felt in the clinic, and is
reshaping modern health-care economics. A successful Generation II multi-parametric test will better discriminate those that are
truly at high risk for recurrence following endocrine therapy and offer potential therapeutic options for intervention for those who
would not benefit from current modalities.

Title: Transcriptomic profiling of patient sequential tumours provides cutting edge view of global metastatic expression changes
following endocrine therapy
Jean McBryan1, Ailis Fagan1, Damian McCartan1, Jarlath C Bolger1, Fiona T Bane1, Christopher Byrne1, Marie McIlroy1, Arnold D
Hill1 and Leonie S Young1. 1Royal College of Surgeons in Ireland, Dublin, Ireland.
Body: Disease recurrence is a common problem in breast cancer and yet the mechanisms enabling tumour cells to evade
therapy and colonise distant organs remain unclear. Here, for the first time, RNAsequencing has been performed on matched
primary, nodal and liver metastatic tumours from three tamoxifen-treated patients following metastatic disease progression.
Despite all primary tumours being of a luminal subtype and all cancers metastasising to the liver, the extent of patient
heterogeneity at the gene level was striking. Less than 3% of the genes differentially expressed between sequential tumours were
common to all patients. Larger divergence was observed between primary and liver tumours than between primary and nodal
tumours, reflecting both the latency time to disease progression and the genetic impact of endocrine therapy. Furthermore, a
xenograft model demonstrated the ability of tamoxifen to drive disease progression and establish distant metastatic disease in the
endocrine resistant setting. Common functional pathways altered during metastatic, endocrine-resistant progression included
ECM receptor interactions and focal adhesions. This novel global analysis highlights the influence of primary tumour biology in
determining the transcriptomic profile of metastatic tumours, as well as the need for adaptations in cell-cell communications in
order for tumour cells to successfully colonise distant host organs.

Title: New treatment options for metastatic breast cancer revealed by reverse-phase protein microarray and genomic profiling
Casey B Williams1, Pradip De1, Nandini Dey1, Kirstin A Williams1, Jessica Klein1, Michelle King1, Misty Small1, Brigitte Cyr1,
Scooter Willis1, Yuliang Sun1, Jennifer Carlson1, Brandon M Young1, Amy Krie1 and Brian Leyland-Jones1. 1Avera Cancer
Institute, Sioux Falls, SD.
Body: Background: The optimal treatment strategy for patients with metastatic breast cancer (MBC) is currently unknown.
Resistance to standard therapies like anthracyclines and taxanes limit the number of treatment options in many patients to a
small number of non-cross-resistant regimens. We hypothesized that genomic and proteomic profiling of clinical MBC samples
would identify genomic alterations that are linked to targeted therapies, and that this could facilitate a personalized approach to
therapy for our patients.
Methods: We retrospectively analyzed 31 consecutive metastatic breast cancer patients that had genomic and/or proteomic
studies sent over a 4 month period from February to May 2014. All patients were seen in our Genomic Oncology Clinic and
subsequently underwent a rebiopsy of a metastatic site. Formalin-fixed, paraffin-embedded (FFPE) tissue was sent to either
Foundation Medicine for genomic profiling, Theranostics for reverse-phase protein microarray, or both. Standard
immunohistochemistry for ER, PR, and HER2 was also performed on all patients.
Results: Genomic or proteomic alterations were identified in all 31 patients. All patients harbored at least one actionable target
and a treatment recommendation for a currently available FDA approved drug or drug combination was able to be suggested in
all but one patient. The most commonly observed genomic alterations were within PIK3CA (26%), TP53 (23%), CCND1 (19%),
and MYC (16%). Over 30 distinct genomic alterations were identified. Proteomic results were available from 16 patients.
Activation of the AKT/mTOR pathway was evident in a majority of patients. A change in HER2 status was also found in 26% of
patients. 16% of cases underwent a negative to positive conversion in HER2 status while 10% of cases underwent a positive to
negative conversion. It is notable that all 5 patients that underwent a negative to positive conversion in HER2 status had biopsies
taken from metastatic disease in the liver.
Conclusions: All patients in this retrospective study harbored genomic or proteomic alterations that are associated with targeted
therapies. Treatment recommendations were suggested in all but one patient and a majority of patients are receiving the
suggested therapy. Our data demonstrate that routine genomic and proteomic analysis in a clinical setting makes a significant
positive impact for patients.

Title: Exploring type II microcalcifications in benign, premalignant and malignant breast lesions by shell-isolated
nanoparticle-enhanced Raman spectroscopy (SHINERS)
Haipeng Zhang1, Bing Han2, Chao Zheng2, Ye Du2, Lijia Liang3, Weiqing Xu3 and Zhimin Fan2. 1The First Hospital of Jilin
University, Changchun, Jilin, China; 2First Hospital of Jilin University, Changchun, Jilin, China and 3State Key Laboratory of
Supramolecular Structure and Materials, Jilin University, Changchun, Jilin, China.
Body: Background: Type II microcalcifications, most often seen in proliferative lesions, including both benign and malignant
pathologies. But whether the emergence of type II microcalcifications associated with cell canceration is still not clear now.
Raman spectroscopy is a powerful, non-invasive analytical tool which can provide detailed and meaningful information about
biochemical composition of tissues at molecular level. Our aim was to find the differences and relationships of type II
microcalcifications between fibroadenoma, ADH tissues, and DCIS, IDC in breast based on their various biochemical
characteristics by Raman spectroscopy.
Methods: The frozen sections were collected from 15 patients (all female; ages 25-57) who underwent surgical resection or
mammotome biopsy at the Department of Breast Surgery, the First Hospital of Jilin University. After operation the samples were
immediately frozen at -20-25 and two contiguous sections (6 m thickness) were cut from a sample by freezing microtome.
One was stained with haematoxylin and eosin for routine histopathological analysis and found the microcalcification locations by
three expert breast pathologists; the same position of the other section was detected by Microscopic confocal Raman
spectrometer (HORIBA JY Lab800, 633nm) with its mirror image (the H&E section). After the spectra we needed had been
obtained, the Au@SiO2 shell-isolated nanoparticles (SHINs) were added to the surface of frozen sections immediately and then
the spectra with SHINs were collected.
Results: A total of 122 Raman spectra and 119 SHINERS spectra were obtained from the microcalcifications in 3 fibroadenoma
tissues, 3 ADH, and 5 DCIS, 4 IDC breast tissues. The Raman signals were significantly enhanced by SHINs. Except the major
peak at 958-960 cm-1 which attributed to the vibrational modes of calcium hydroxyapatite, the results show no calcium oxalate
dihydrate peaks but several other chemical species. The peaks of these species appear at 1002, 1072, 1126, 1446, 1556, and
1657 cm-1, these are attributed to amino acid residue (phenylalanine), nucleic acids, lipids, carotenoids, and Amide I,
respectively. The fibroadenoma and ADH microcalcifications mean spectral have the same peak at 1072cm-1, which belongs to
the O-P-O stretch of nucleic acids, but in DCIS and IDC tissues the peak changes to 1078cm-1. The ADH have more obvious
peak at 1657 cm-1 which assigned to different vibrational modes of the backbone and Amide I, but in DCIS and IDC tissues, the
Amide I bands were disappeared. In the DCIS microcalcifications mean spectral, the peaks attributed to amino acid residue
(phenylalanine) at 1003, 1031 cm-1 and lipids (CH2 and CH3 bending) at 1301, 1441 cm-1 show more stronger. Meanwhile,
compare with DCIS, the IDC spectral shows more stronger nucleic acids peaks and weaker lipids peaks.
Conclusion: We have demonstrated the potential of SHINERS to differentiate Type II microcalcifications found in fibroadenoma,
ADH, and DCIS, IDC breast tissues. The results presented in this paper provide the biochemical characteristics of the Type II
microcalcifications among these tissues, and may represent a key factor responsible for mechanisms of carcinogenesis.

Title: Gene expression in synchronous primary, axillary nodal and disseminated breast cancer cells
Reuben J Broom1, Cherie Blenkiron3, Richard Harman2, Erica Whineray Kelly2, Stephen Allpress2, Katherine Gale2, Sam Rice2,
Peter Shin2, Anna Brown2, Annette Lasham3 and Cristin Print3. 1Auckland City Hospital, Auckland, New Zealand; 2North Shore
Hospital, Auckland, New Zealand and 3University of Auckland, Auckland, New Zealand.
Body: Disseminated Tumour Cells (DTC) have been found in the bone marrow (BM) of up to 70% of patients with breast cancer
metastatic to axillary lymph nodes. They are a biologically and therapeutically interesting tumour cell population. Although there is
uncertainty about their relationship to cancer prognosis, DTC provide a valuable window into the processes by which primary
tumour (PT) cells disseminate. DTC can be enriched from BM by immunoaffinity using antibodies directed against epithelial cell
markers such as EpCAM. To better understand the changing gene expression patterns that may accompany breast cancer
metastasis, we have compared the whole genome RNA expression profiles of matched PT, axillary lymph node metastases
(LNM), and EpCAM-enriched cells from the BM of seven patients. Compared to PT, axillary LNM had consistently altered
expression of RNAs encoding matrix metalloproteinases (MMP), growth factors, transcription factors, as well as downstream
targets of Catenin-a, Tumour Necrosis Factor (TNF)-a and miR-22. Once the gene expression patterns of potentially
contaminating BM cells were subtracted, compared to PT and LNM EpCAM-enriched MB cells had consistently elevated
expression of RNAs encoding metabolic enzymes, ribosomal proteins, and DNA-binding factors such as YB-1. The gene
expression changes we identify are candidate mediators of breast cancer metastases and represent attractive targets for further
study on a cell by cell basis.

Title: Development of a fluorescent reporter system to delineate self-renewing cells in triple negative breast cancer
Praveena S Thiagarajan1, Masahiro Hitomi1, James S Hale1, Alvaro G Alvarado1, Baltin Otvos1, Kevin Stoltz1, Maksim Sinyuk1,
Awad Jarrar1, Qiao Zheng1, Dustin Thomas1, Thomas Egelhoff1,2,3, Jeremy N Rich2,3,4, Justin D Lathia1,2,3 and Ofer Reizes1,2,3.
1
Cleveland Clinic Lerner Research Institute, Cleveland, OH; 2Cleveland Clinic Lerner College of Medicine of Case Western
Reserve University, Cleveland, OH; 3Case Comprehensive Cancer Center, Cleveland, OH and 4Cleveland Clinic Lerner Research
Institute, Cleveland, OH.
Body: Background : Advanced cancers including triple-negative breast cancer (TNBC), the most aggressive breast cancer
subtype, contain a self-renewing, tumorigenic cancer stem cell (CSC) population. CSCs contribute to tumor progression and
therapeutic resistance. Despite an effective early response to chemotherapy, TNBC relapses with a highly heterogeneous and
refractory metastatic disease enriched in CSCs. Apart from being chemo- and radio-resistant, CSCs display a high degree of
multipotency, heterogeneity and plasticity. Due to the complex nature of CSCs, elucidating the characteristics of CSCs in TNBC
progression continues to be a challenge. Thus, developing anti-CSC therapies to be integrated into clinical paradigms represents
an immediate priority.
Rationale : A major obstacle to the identification of CSC regulatory mechanisms is the lack of experimental systems that enable
the reliable enrichment of CSCs from non-CSCs for comparative analysis. While CSCs have been isolated from TNBC using
CD44+/CD24- and/or ALDH+ phenotype, this enrichment paradigm requires refinement as it is not universally applicable and
lacks the ability to study CSCs in real time. The limitations of these systems have excluded their application in studying the
molecular heterogeneity among breast cancer tumors. CSCs are molecularly characterized based on the expression of cell
surface receptors and the embryonic stem cell transcription factors NANOG, OCT4 and SOX2. NANOG, the master regulator of
stem cell self-renewal, has emerged as a pro-carcinogenic factor in cancer cell lines with CSC behavior. Our previous studies
have shown that silencing NANOG in cancer cells leads to reduced proliferation and self-renewal based on in vitro and in vivo
experiments.
Hypothesis : We hypothesized that a Nanog promoter could be used to effectively enrich for TNBC CSCs.
Results : We generated two TNBC cell lines (MDA-MB-231 and HCC70) harboring NanogGFP reporter gene by lentiviral
transductions. Increased NANOG mRNA and protein expression was observed in flow cytometry-sorted GFP+ cells compared
with GFP- cells. GFP+ cells were enriched for the CSC markers CD49f and CD44+/CD24-. GFP+ cells also demonstrated an
increased protein expression of the stem cell transcription factors NANOG, SOX2 and OCT4. Limiting dilution analyses revealed
increased self-renewal and significantly higher stem cell frequencies in GFP+ cells. GFP+ cells demonstrated a mesenchymal
phenotype with increased invasive capacity. Subcutaneous injections of GFP+ cells showed significantly higher in vivo tumor
initiation and progression than GFP- cells. Flow cytometry-sorted GFP+ and GFP- cells enriched for CSCs and non-CSCs,
respectively. Using this system, we performed a high-throughput flow cytometry screen and identified an additional novel CSC
marker for TNBC.
Conclusion : We have developed a novel TNBC CSC reporter system using a GFP reporter driven by the Nanog promoter and
identified a novel CSC marker. We have defined and validated this robust system wherein we have characterized and monitored
the role of CSCs and non-CSCs in TNBC tumor initiation and progression. Using this approach, identifying CSC targets in TNBC
could unravel the potential for development of innovative therapeutic strategies.

Title: Breast cancer stem-like cell activity correlates with tumour progression to metastasis but not with clinical or tumour
characteristics
Sacha J Howell1, Denis Alferez2, Katherine Spence2, Rachel Eyre2, Fran Shaw1, Bruno Simoes2, Angelica Santiago-Gomez2,
Maria Bramley3, Mohamed Absar3, Zahida Saad4, Sumohan Chatterjee4, Cliona Kirwan1, Ashu Gandhi5, Anne C Armstrong6,
Andrew M Wardley6, Gillian Farnie1 and Robert B Clarke1. 1University of Manchester, Institute of Cancer Studies; 2CRUK
Manchester Institute; 3Penine Acute Hospitals NHS Trust; 4Salford Royal NHS Foundation Trust; 5University Hospitals of South
Manchester NHS Foundation Trust and 6Christie NHS Foundation Trust, Manchester, United Kingdom.
Body: Introduction: Breast cancers exhibit cellular heterogeneity, containing both stem-like and more differentiated cells. The
activity of cancer stem cells (CSC) is likely to be dependent on the microenvironment or niche. Using 158 patient tumour samples,
correlations between niche-independent breast CSC activity and clinical and tumour characteristics were tested.
Methods: 104 early breast cancer surgical samples and 54 unrelated metastatic samples from pleural or ascitic fluid were
harvested. To test CSC activity, isolated cells were grown in both primary (formation) and secondary (self-renewal)
mammosphere (MS) culture. Tumour initiating activity was also tested by transplanting breast cancer fragments or cells into the
sub-cutaneous flanks of NSG mice (n=84 early and n=10 metastatic).
Results: No correlation was found between MS growth, MS formation (%), MS self-renewal (%) or in vivo tumour initiation and
breast cancer sub-type, grade, node status or Nottingham prognostic index. 33% of the samples that formed MS in vitro initiated
tumours in vivo while only 9% that failed to form MS initiated tumour growth. Metastatic compared to early BC samples grew MS
more frequently (53/54 compared to 81/104), and had a higher primary MS formation efficiency (1% vs 0.6%; P<0.001) although
rates of MS self-renewal were similar. Tumour initiation in vivo was also more frequent in metastatic than early breast cancer
samples (7/10 versus 25/84; P<0.02).
Conclusions: In summary, niche-independent breast CSC activity measured in vitro by MS assay and in vivo by xenograft growth
is not directly correlated with standard clinical parameters. However, both in vitro and in vivo CSC activity are increased in
metastatic samples. These results suggest that breast CSC activity is independent of other prognostic indicators but may predict
for poor outcome tumours. Relapse free survival data are maturing and will be presented with analysis of primary tumour ALDH1
expression.

Title: Prostaglandin E receptor EP4 is a therapeutic target in breast cancer stem cells
Amy Fulton1,2, Namita Kundu1, Tyler Kochel1, Xinrong Ma1 and Jocelyn Reader1. 1University of Maryland Greenebaum Cancer
Center, Baltimore, MD and 2Baltimore VA Medical Center, Baltimore, MD.
Body: There is an urgent need to identify mechanisms underlying the survival of cells expressing cancer stem cell or
tumor-initiating properties. The cyclooxygenase-2 (COX-2) pathway is highly expressed in many breast tumors and contributes to
poor outcomes. The COX-2 product prostaglandin E2 (PGE2) promotes tumor growth and metastasis by acting on the
G-protein-coupled receptor EP4. We compared the expression and function of COX-2 and EP4 in mammosphere-forming and
bulk populations derived from a panel of human and murine luminal and basal type tumor cells with different metastatic
capacities. Expression of both EP4 and COX-2 were markedly increased in mammosphere-forming cells derived from basal
and/or metastatic cells relative to the bulk population, but neither COX-2 or EP4 levels were elevated in mammospheres derived
from luminal or non-metastatic cells. Breast cancer stem cells, expressing elevated EP4 are correspondingly more sensitive to
inhibition with EP4 antagonists both in vitro and in limiting-dilution assays in vivo. Somewhat to our surprise, cancer stem cells
remain relatively sensitive to lysis by Natural Killer cells. We have also shown that EP4 blockade protects NK cells from
tumor-induced immune suppression. These studies identify EP4 as a potential therapeutic target in the general tumor cell
population and show that EP4 targeting may selectively inhibit cells with tumor-initiating or stem cell properties. EP4 antagonists
can directly inhibit cancer stem cells and tumor metastasis and indirectly support NK-mediated mechanisms of tumor control. We
are delineating the mechanisms by which EP4 and COX-2 are upregulated in cells with stem-like properties. In addition to EP4
and COX-2, STAT3 is upregulated in stem cells. Inhibition of STAT3 reduces mammosphere-forming capacity. Our studies
support a mechanism whereby COX-2/EP4 signaling induces STAT3 to support breast cancer stem cell survival by a
feed-forward mechanism.

Title: Proteosome inhibitor bortezomib inhibits NFB and effectively overcomes cancer stem cell escape triggered by Wnt
inhibitor therapy in FOXC1+ basal-Like/claudin-low breast cancer
Partha S Ray1,2, Connie Y Tsai2, Tania Ray3, Bomy Kim2 and Tor W Jensen2. 1University of Illinois College of Medicine, Urbana
Champaign, IL; 2Carle BioMedical Research Center, Urbana, IL and 3Oracle Biosciences, Champaign, IL.
Body: Cancer stem cells (CSCs) are considered to be an important contributing factor towards treatment failure, cancer
recurrence, and mortality. CSCs are known to be more enriched in the basal-like and claudin-low subtypes of breast cancer. The
Wnt/ catenin signaling pathway is well known as a regulator of embryonic development and stem cell biology, and is prominently
active in basal-like and claudin-low breast cancers. We have previously reported that transcription factor FOXC1 plays a critical
role in mediating aggressive cell traits in basal-like/claudin-low breast cancer. In this study, we sought to investigate the link
between Wnt signaling and FOXC1 and its potential in regulating CSC biology in basal-Like/claudin-low breast cancer. We
observed that exposure of the MDA-MB-231 basal-like/claudin-low cell line (low constitutive FOXC1 expressor) to Wnt3a (a
canonical Wnt signaling ligand), resulted in increased expression of FOXC1. Reciprocally, overexpression of FOXC1 in MCF10A
human mammary epithelial cells led to a pronounced increase in Wnt signaling activity, strongly suggestive of a direct or indirect
positive feedback loop between Wnt signaling and FOXC1. More importantly, BT549 and HS578t basal-like/claudin-low cells
(high constitutive FOXC1 expressors) proved to be more sensitive to treatment with the Wnt inhibitor ICRT3 as evidenced by
decreased cell viability when compared to MCF7 (luminal) or SKBR3 (HER2) breast cancer cell lines. Furthermore the decrease
in cell viability appeared to be proportionate to the level of FOXC1 expression. Upon pharmacological inhibition with ICRT3 and
biological inhibition with siRNA knockdown of LRP6, (a canonical Wnt signaling cell surface receptor) a decrease in FOXC1
expression level was observed in a dose and time dependent manner. This effect was particularly pronounced in mammosphere
cultures enriched for BT549 cancer stem-like cells. Inhibition of Wnt signaling reduced mammosphere formation efficiency of
BT549 cells, suggesting that Wnt inhibition targets cancer stem cells (CSCs) in the basal-like/claudin-low breast cancer subtype.
More importantly, however, after an initial 4 day incubation period, some cells are observed to persist and later display renewed
enhancement of mammosphere formation ability. Profiling of such cells interestingly revealed depletion of FOXC1+ve cells but
persistence of cells displaying pronounced up regulation of stereotypical embryonic stem cell Transcription Factors OCT4, SOX2
and NANOG, strongly suggestive of a potential primitive stem cell/quiescent cell state escape mechanism. qRT-PCR based
pathway activation analysis revealed marked activation of NFB signaling in the residual cells that withstood Wnt inhibition.
Simultaneous pharmacologic inhibition with Wnt inhibitor ICRT3 and the proteasome inhibitor Bortezomib (known to inhibit NFB
signaling) effectively targeted BT549 cancer stem cells in mammosphere culture and prevented the persistence/emergence of
any residual cells. Taken together, our findings suggest that combination therapy approaches are likely required to effectively
target breast cancer stem cells.

Title: Expression of the pluripotency transcription factor SOX2 in primary breast cancers (BCs): Correlation with
clinicopathological features (CPfs) and recurrence
Barbara Pistilli1, Giovanni Benedetti1, Mauro Finicelli2, Tiziana Squillaro2, Andrea Marcellusi3, Tommasina Biscotti4, Alfredo
Santinelli4, Paola Mariani5, Paolo Decembrini6, Giancarlo Ciccioli7, Luciano Latini1, Antonio Giordano8 and Umberto Galderisi9.
1
Macerata Hospital, Macerata, Italy; 2Human Health Foundation, Spoleto, Italy; 3University of Rome "Sapienza", Rome, Italy;
4
Politecnica University of Marche, Ancona, Italy; 5Macerata Hospital, Macerata, Italy; 6Civitanova Hospital, Civitanova Marche,
Italy; 7Macerata Hospital, Macerata, Italy; 8Sbarro Institute for Cancer Research and Molecular Medicine, Temple University,
Philadelphia and 9Second University of Naples, Naples, Italy.
Body: BACKGROUND
SOX2 is one of the pluripotency transcription factors expressed by stem cells, which plays a central role in controlling the
expression of genes implicated in embryonic development and stemness manteinance. Key regulators of embryonic stem cell
(ESC) identity, such as NANOG, SOX2, OCT4 and GDF3 resulted overexpressed in a variety of solid tumors with a possible role
in cancer progression and prognosis. SOX2 expression has mainly been reported in basal-like BC subtype, suggesting a role in
conferring a less differentiated phenotype. In our analysis we evaluated a heterogenous group of BC tissues to determine
whether the expression of ESC-regulating genes correlated with CPfs and recurrence.
METHODS
140 primary invasive BC specimens were collected from 137 female patients who underwent surgery. mRNA expression for
SOX2, OCT4, NANOG, GDF3 genes was assessed by RT-PCR. Immunoistochemistry (IHC) was performed for SOX2 with
mouse monoclonal antibody (1:50, Y17, Santa Cruz Biotechnology, USA). Correlations with molecular subtypes, menopausal
status, grading, ER, PR, ki67 ( 20% and > 20%), HER2, T-size and node status were evaluated by Fisher's exact test and 2
test. Association of ESC-genes, CPfs and DFS was estimated by univariate and multivariate Cox-regression analysis (p 0.05).
Survival analysis (DFS and OS) were calculated by Kaplan-Meier curves and compared by log-rank test.
RESULTS
In 117 samples assessable by RT-PCR the genes resulted expressed as follows: NANOG=52 (44.5%), SOX2=11 (9.4%),
GDF3=9 (7.2%), OCT-4=0. Correlation of mRNA gene expression with CPfs was statistically significant between NANOG and
grade 2, GDF3 and node-negative status, SOX2 and higher ki67 (p=0.019, p=0.029, p= 0.035, respectively). Six out of 11 SOX2+
tumors were HER2+ (data not statistically significant); in the remaining 5 samples the fluorescence in situ hybridization was
performed but no HER2 amplification was detected. At univariate analysis of DFS, SOX2 expression (HR=2,36; p=0.002), ki67
(HR= 2,19; p=0,028), T-size >1 (HR=2,06; p=2.011), node-status (HR=2,21; p= 0.014); ER/PR(HR=0,58/HR=0,59,
p=0.065/p=0.068) resulted statistically significant. At multivariate analysis, SOX2 (HR=2,99; 95% CI 1,41-6,30; p=0.004),
node-status (HR=2,44; 95%CI 1,25-4,76; p=0.009) and T-size >1 (HR=1,77; 95% CI=0,99-3,13; p=0.051) were independently
associated with increased risk of recurrence. An earlier recurrence was observed in SOX2+ patients (median 34.9 months; 95%
CI: 7.5-62.2) than SOX2- patients (median: 60.3 months; 95% CI: 32.6-88.1) (p=0.017); OS resulted shorter in SOX2+ (68.2
months 95%CI: 63.7-151.4 vs 145.3 months 95% CI: 80.5-210.2) albeit not statistically significant (p=0.104). IHC analysis
showed a positive score for SOX2 protein expression in all of 11 samples with SOX2 mRNA amplification; SOX2+ protein was not
detected in 20 samples randomly selected among the tissues not expressing SOX2 mRNA.
CONCLUSIONS
Our analysis confirm that ESC-regulating genes correlate with specific CPfs (grading, node-status and ki67). Notably, SOX2
resulted to be an independent prognostic factor, as it was associated with a risk of recurrence increased by 3 times, irrespective
of other CPfs.

Title: Mammospheres derived from circulating epithelial tumor cells are an indicator for presence of metastasis in patient with
breast cancer
Monika Pizon1, Dorothea Zimon1, Ulrich Pachmann1 and Katharina Pachmann1. 1Transfusion Center, Bayreuth, Germany.
Body: Background:
A major obstacle in the successful treatment of cancer is the occurrence of metastasis. The presence of CETCs is closely related
to tumor recurrence, but the mechanisms through which CETCs promote metastasis are still unclear. The aim of this study was to
determine the proliferative capacity of CETCs by analyzing the frequency of mammosphere formation with subsequent
phenotypic characterization of the spheres arising in breast cancer patients.
Methods:
CETCs were cultured under condition favoring growth of mammospheres from 38 patients with breast cancer, including a
subpopulation of 13 patients with metastatic disease. Cell viability, stem cell marker expression and ALDH 1 activity was
evaluated by fluorescence scanning microscope (Olympus ScanR).
Results:
Sphere formation was observed in 74 % of patients with breast cancer. Patients with distant metastasis had higher numbers of
mammospheres (median 13,0 vs 1,0; p0,001) compared to patients without metastasis. In multivariate analysis, a high number
of mammospheres was associated with the presence of metastasis. The mammospheres area under the ROC curve was 0,99.
Six or more spheres classified metastatic disease with a sensitivity and specificity of 96 % and 100 %, respectively. These results
suggest that above cut-off six in the number of mammospheres is statistically highly indicator for disease progression. Analysis of
surface marker expression profile of mammospheres showed that spheres cultured from CETCs had typical phenotype of cancer
stem cells with a high enzymatic activity for ALDH 1 with the ALDEFLUOR assay. There was no sphere formation in a control
group with 50 healthy donors.
Conclusion
This study demonstrates that a small fraction of CETCs has proliferative activity.
Identifying the CETC subset with cancer stem cells properties may provide more clinically useful prognostic information and may
be a new indicator for the presence of metastases.

Title: StemScreen, an innovative platform technology for the identification of novel cancer stem cell-directed compounds
Janice Chen1, Hai Li2, Marcie A Glicksman3, Charles Karan2, Christopher Brooks1 and Eric K Rowinsky1. 1Stemline Therapeutics,
Inc, New York, NY; 2Columbia University Medical Center, New York, NY and 3Brigham and Women's Hospital, Cambridge, MA.
Body: Cancer stem cells (CSCs) are thought to play significant roles in breast cancer initiation and progression. Conventional
therapeutic agents target tumor bulk but spare CSCs, leading to tumor recurrence and relapse. Therefore, drugs that eliminate
both tumor bulk and CSCs may represent the most effective treatment strategy for breast cancer. We have developed a
proprietary cell-based screen called StemScreen that permits the rapid testing and identification of novel CSC-directed
compounds in a high-throughput manner. StemScreen takes advantage of a landmark discovery that many cancer cell lines
harbor stable populations of CSCs. Our initial studies utilized a breast cancer cell line transfected with a unique expression vector
that is only active in the 1-5% putative CSC population. The StemScreen platform is unique because it allows for screening of
compounds within the context of the CSCs natural microniche environment and has advantages over traditional drug discovery
methods that have been designed to identify compounds that only target tumor bulk, but not CSCs.
In an effort to increase the throughput and make the assay compatible with existing compound libraries, we miniaturized the
platform to a 384-well format and optimized the line for high-content screening. A series of compound hits identified from
screening these cells against smaller libraries of known active compounds as well as larger and more diverse compound libraries
will be presented. We are also currently expanding this technology into other tumor types. We believe that this approach
represents a major technological advance in oncology drug discovery and that this platform will be instrumental in the discovery of
unique new therapies that have a dual effect on CSCs and non-CSC tumor bulk.

Title: SIRT1 inhibitors significantly reduce cancer stem cells and block epithelial mesenchymal transformation in breast cancer
cells
Songlin Zhang1, Min Li1, Baoxiang Guan1 and Robert Brown1. 1UTHSC at Houston, Houston, TX.
Body: Breast cancer stem cells are contribute to distance metastasis, breast cancer recurrence and drug resistance. SIRT1, a
nicotinamide adenine dinucleotide (NAD+)-dependent histone deacetylase, play significant role in DNA repair, cell survival, stem
cells and epigenetic modification. SIRT1 has been related to cancer stem cells in chronic myelogenous leukemia. In this study,
we investigate the role of SIRT1 in breast cancer stem cells.
SIRT1 inhibitors cambinol and Ex527 are used for treatment in several breast cancer cell lines, including T47D, TB549,
MDA-MB-231 and MDA-MB-468. Stem cell markers, SOX-2 and Nanog, are significantly decreased in SIRT1 inhibitor treated
cancer cells by qRT-PCR and western blot in T47D cell line. The ALDH positive cells in MDA-MB-231 cell line are significant
reduced from 29% to 3.2% with Adeflour assay, and the CD44 expression is significantly reduced in CD44/CD24 flow cytometry
analysis. Using qRT-PCR, SIRT1 inhibitors significantly down regulate vimentin (-3.7 folds), N-cadherin and smooth muscle actin
(-2.7 folds) and up regulate the gene of E-cadherin, indicating SIRT1 inhibitors can block the epithelial mesenchymal
transformation (EMT) of breast cancer. SIRT1 inhibitors can significantly (40-50%) block the cancer cell invasion and migration in
several triple negative breast cancer cell lines. SIRT1 inhibitors can inhibit cell proliferation on all tested breast cancer cell lines,
and can induce significant cell apoptosis in T47D cells. In xenograft tumor study, SIRT1 inhibitor cambinol can significantly reduce
tumor growth and inhibit tumor metastasis.
The molecular mechanism of SIRT1 inhibitors in blocking EMT and reducing cancer stem cells is likely associated with blocking
the Wnt pathway. Several down stream target genes of Wnt pathway, such as cyclin D1, c-Myc and c-Jun, are significantly down
regulated after using SIRT1 inhibitor cambinol, and the changes are more significant in TGF-beta stimulated cancer cells.
Beta-catenin is significantly reduced including the active beta-catenin, and the decreasing beta-catenin protein may be related to
the decreased Dvl proteins.
In summary, our study demonstrated that SIRT1 inhibitors can reduce breast cancer stem cell population in several cancer cell
lines, block epithelial mesenchymal transformation, and inhibit breast cancer cell invasion and metastasis. SIRT1 inhibitors
appear to inhibit Wnt pathway in cancer cells to block EMT and affect cancer stem cells.

Title: Inflammation promotes stem-like characteristics and increases stemness of breast epithelial cells
Jennifer Sims-Mourtada1,2, Kimberly M Arnold1,2, Lynn Opdenaker1,3 and Daniel Flynn1,2. 1Center for Translational Cancer
Research, Helen F. Graham Cancer Center, Christiana Care Health Services, Inc, Newark, DE; 2University of Delaware, Newark,
DE and 3University of Delaware, Newark, DE.
Body: Triple negative breast cancer (TNBC) accounts for approximately 15% of all breast cancers and is more likely to affect
younger women, women of African American descent, and BRCA1 mutation carriers. TNBC is typically an aggressive cancer and
is defined as the absence of positive staining for estrogen (ER) and progesterone (PR) receptor, and a lack of amplification of
HER2. Due to its receptor status, TNBC is insensitive to the conventional targeted treatment utilized in ER/PR/HER2 positive
breast cancer leading to poor prognosis and early visceral metastasis with survival rates for women who relapse within 5 years of
treatment being significantly lower than hormone receptor positive breast cancer. Therefore, understanding the mechanisms that
drive the formation of TNBC can aid in determining targets to develop better treatment options. We investigated differences in the
tumor microenvironment between TNBC and ER positive breast cancer patient tissue samples. Hematoxylin & eosin stain of
invasive ductal carcinomas revealed that TNBC patients had higher amounts of tumor infiltrating leukocytes compared to ER
positive breast cancers. In premalignant breast cancer patients, ER negative ductal carcinoma in situ (DCIS) patients had
significantly higher infiltrating leukocytes compared to ER positive DCIS indicating that inflammation may be a contributing factor
in TNBC. Research on molecular profiling of TNBC has revealed that a subset of gene expression patterns are associated with
the basal-myoepithelial cells rather than the luminal cells, however, previous evidence indicates that TNBC arises from luminal
progenitors which undergo an abnormal conversion to basal-like progenitors with enhanced growth and survival promoting
characteristics. We propose this conversion occurs due to the presence of inflammatory signals early in tumor development which
trigger a cascade of events that leads to acquisition of a basal-like phenotype in luminal progenitor cells and upregulation of
stem-like properties, resulting in the formation of basal-like breast cancer/TNBC. Treatment of non-malignant breast epithelial
cells with IL-6 increased sphere-forming efficiency under stem cell growth promoting conditions and an inhibitor of the IL-6/STAT3
pathway decreased ALDH enzymatic activity as measured by Aldefluor assay. Also, treatment with IL-6 downregulated the
luminal marker, epithelial specific antigen (ESA), and increased the expression of the myoepithelial marker, CD49f, on the cell
surface of non-malignant breast epithelial cells, indicating an increase in the basal-like cell population and a loss of the luminal
cell population. Therefore, inflammation appears to be a common feature in TNBC compared to ER positive breast cancer and
the presence of an inflammatory stimulus in the breast may promote the conversion of luminal to basal-like cells, promoting the
upregulation of stem-like properties and development of TNBC. These data suggest that anti-inflammatory drugs could potentially
be used to inhibit these events from occurring in the development of TNBC, leading to decreased tumor growth and better
treatment options.



Title: The expression of aldehyde dehydrogenase1 in cytoplasm or stroma of breast cancer is associated with clinical prognosis
Xin Guan1, Yi Dong2, Aiping Shi1 and Zhimin Fan1. 1Breast Surgery, Bethune First Hospital of Jilin University, Changchun, Jilin,
China and 2Breast Surgery, Jilin Tumor Hospital, Changchun, Jilin, China.
Body: Background: Previous studies have shown that ALDH1 expression in breast tumor cells was associated with poor
prognosis.Whether tumor microenvironment affect prognosis of breast cancer patients is unknown by far.
Aim: We performed a research to study whether ALDH1 expression in cytoplasm or stroma of breast tumors is associated with
prognosis of breast cancer patients.
Methods: As many as 160 cases with invasive carcinoma were retrieved.ALDH1 staining was respectively assessed on
cytoplasmic and stroma of tumour cells.Chi-square test was used to evaluate the associations between ALDH1 expression and
clinical parameters.Cox proportional hazards analysis was used for relapse-free survival(RFS).
Results: There was significant association between ALDH1 expression in stroma and Age,ER.
Association between ALDH1 expression status and clinicopathological parameters
ALDH1
expression in
stroma
ALDH1 expression in
cytoplasm
N
ALL CASES
Positive (%)
Negative (%) P
Positive (%) Negative (%) P
56 (35)
104 (65)
64(40)
Age (years)
96 (60)
0.867
0.022
50
94
32 (34.0)
62 (66.0)
30 (31.9)
64 (68.1)
>50
66
24 (36.4)
42 (63.6)
34 (51.5)
32 (48.5)
Tumor size
0.430
0.188
54
17 (31.5)
37 (68.5)
18 (33.3)
36 (66.7)
2-5
100
38 (38.0)
62 (62.0)
46 (46.0)
54 (54.0)
>5
1 (16.7)
5 (83.3)
1 (16.7)
5 (83.3)
Number of metastatic lymph

nodes
0.452
0.364
80
27 (33.8)
53 (66.2)
34 (42.5)
46 (57.5)
4-9
57
18 (31.6)
39 (68.4)
19 (33.3)
38 (66.7)
>9
23
11 (47.8)
12 (52.2)
11 (47.8)
12 (52.2)
ER in primary breast tumor
0.156
0.025
Positive
109
34 (31.2)
75(68.8)
37 (33.9)
72 (66.1)
Negative
51
22 (43.1)
29 (56.9)
27 (52.9)
24 (47.1)
PR in primary breast tumor
0.591
0.598
Positive
112
41 (36.6)
71 (63.4)
43 (38.4)
69 (61.6)
Negative
48
15 (31.3)
33 (68.7)
21 (43.8)
27 (56.3)
HER-2 in primary breast tumor
0.460
0.585
Positive
43
17 (39.5)
26 (60.5)
19 (44.2)
24 (55.8)
Negative
117
39 (33.3)
78(66.7)
45 (38.5)
72 (61.5)
Significance level of p<0.05 was considered to be statistically significant.
In multivariate analyses of RFS,ALDH1 expression in cytoplasm is associated with a worse prognosis.However,ALDH1

expression in stroma is not related with prognosis.
Univariate and multivariate analyses of RFS
Univariate Analysis
Multivariate Analysis
Characters
OR
95% CI
OR
95% CI
Age 50/>50
0.812
0.436-1.515
0.513
Tumor size 2/>2
2.473
1.144-5.343
0.021
2.035
0.938-4.417
0.072
Number of metastatic lymph nodes 3/>3
6.346
2.815-14.306
<0.001
5.655
2.477-12.910
<0.001
ER in primary breast tumors -/+
0.660
0.354-1.231
0.191
PgR in primary breast tumors -/+
0.533
0.288-0.988
0.046
0.835
0.358-1.948
0.676
HER-2 in primary breast tumors -/+
0.969
0.487-1.928
0.929
ALDH1 in primary breast tumors -/+
2.578
1.403-4.737
0.002
2.774
1.502-5.124
0.001
ALDH1 in primary breast stroma -/+
0.712
0.356-1.473
0.378
Significance level of p<0.05 was considered to be statistically significant.

Conclusions: We conclude that ALDH1 expression in stroma might suggest the presence of Cancer stem cells in tumor
microenvironment.The ALDH1 expression in cytoplasmic of breast tumors rather than in stroma was an independent factor for
prognosis.
Key words:Breast cancer stem cell;ALDH1;Tumor microenvironment.


Title: Clinical relevance and biological properties of oligometastatic breast cancer in lung; prognostic impact of CD44+/CD24/low
cells
Rei Mimoto1, Tadashi Kobayashi1, Yoshimi Imawari1, Makiko Kamio1, Kumiko Kato1, Hiroko Nogi1, Yasuo Toriumi1, Ken Uchida1
and Hiroshi Takeyama1. 1Jikei University School of Medicine, Tokyo, Japan.
Body: Background: Metastatic breast cancer is a systemic disease. Our aim was to evaluate the clinical outcomes of pulmonary
metastasectomy of recurrent breast tumors and to identify possible prognostic factors.
Methods: We reviewed data from a registry of patients with lung metastases from breast tumors who received pulmonary
metastasectomy in Jikei University Hospital between 2004 and 2011. We analyzed prognostic factors for overall survival (OS) and
progression free survival (PFS) after metastasectomy. We also investigated lung metastases for the prevalence of
CD44+/CD24/low tumor cells and evaluated their prognostic significance.
Results: Among 17 patients with lung metastasis of breast tumors, 5-year OS and PFS were 72% and 36%, respectively. Better
OS was observed among patients with oligometastatic breast cancer (OMBC). Patients with OMBC, estrogen receptor (ER)
positive cells, and disease free intervals (DFI) of >8 years had better PFS. The average prevalence of CD44+/CD24/low tumor
cells in lung metastases of breast cancer was 21%, ranging from 0 to 90%. The presence of CD44+/CD24/low tumor cells
influenced the progression after lung metastasectomy, with median PFS times of only 6 months in patients with high-prevalence
of cancer-initiating cells. CD44+/CD24/low cells with cancer-initiating properties were present in only 9% 12 of patients with
OMBC but were found in 73% 21 of patients with non-OMBC.
Conclusion: Pulmonary metastasectomy may be a treatment option for OMBC patients with lung metastases. Better prognosis of
OMBC may be related to low levels of cancer-initiating cells.

Title: No Show

Title: The impact of ALDH1 on chemo-resistance and prognosis according to intrinsic subtype in breast cancers
Kumiko Kida1, Takashi Ishikawa2, Akimitsu Yamada1, Kazutaka Narui2, Sadataka Sugae1, Mikiko Tanabe2, Yasushi Ichikawa1 and
Itaru Endo1. 1Yokohama City University Graduate School of Medicine, Yokohama, Kanagawa, Japan and 2Yokohama City
University Medical Center, Yokohama, Kanagawa, Japan.
Body: [Background]Aldehyde dehydrogenase 1 (ALDH1) has been identified as a breast cancer stem cell marker. The clinical
significance of ALDH1 as a chemo-resistant and prognostic indicator has been reported recently. However, the analysis
according to each intrinsic subtype was not reported.
[Aims] To investigate the impact of ALDH1 on chemo-resistance and prognosis according to intrinsic subtypes in invasive breast
cancers.
[Methods]
1) Patients and tumor specimens; A total of 653 primary breast cancer patients were enrolled in this study from 2004 to 2013 at
the Yokohama City University Medical Center in Japan. We performed immunohistochemical analyses using paraffin-embedded
core needle biopsy sections prior to the treatment.
2) Correlation of ALDH1 with clinicopathological factors;
Analyses were performed to investigate association of ALDH1 expression with other biomarkers and clinicopathological factors in
breast cancers. Age, histologic type, tumor size, nodal status, ER/PgR/HER2 status, nuclear grade, Ki67, Topo2, p53, CK5/6 and
EGFR were observed.
3) Neoadjuvant patient cohort study
234 breast cancer patients receiving neoadjuvant chemotherapy were enrolled. The correlation between ALDH1 and pathological
complete response (pCR) rate was investigated in each intrinsic subtype.
4) Prognostic cohort study
We performed a Cox analysis of disease free survival and overall survival of all 653 cases according to each subtype, taking
account of clinicopathological factors.
[Results]
ALDH1 expression in tumor cells was seen in 139 of 653 cases (21.3%). The ALDH1 expression correlated significantly with
tumor size, clinical node metastasis, clinical staging, nuclear grade and HER2 status positively, ER and PgR status negatively.
ALDH1 expression was significantly seen in HER2-positive cancers and triple negative type.
In neoadjuvant study, we analyzed 234 patients treated with neoadjuvant chemotherapy including 63 luminal type, 20
luminal-HER2 type, 45 HER2-enriched type and 106 triple negative type. The pCR rate was significantly lower in patients with
ALDH1-positive cases (13.5%vs.30.3%,p=0.003). In multivariate analysis, ALDH1 and ER are correlated with pCR rate
significantly. According to the intrinsic subtypes, the correlation between pCR and ALDH1 expression was extremely significant in
triple negative type (p=0.003). In HER2 positive type, ALDH1 expression had tendency with low pCR, but with no significance. In
luminal type, two patients achieved pCR and both had no ALDH1 expression.
In prognostic analysis, patients with ALDH1 expression had significantly poor disease free survival (DFS; p<0.001) and overall
survival (OS; p=0.044). In the multivariate COX regression model, ALDH1 expression was an independent prognostic indicator of
DFS (p=0.033), but not significant predictor of OS (p=0.124). According to each intrinsic subtypes, ALDH1 had a higher impact on
prognosis of luminal type ,though not significant on triple negative type.
[Conclusions]
Breast cancers with ALDH1 expression posse biologically aggressive phenotypes that tend to have a poor prognosis.
Chemoresistance was significantly seen in ALDH1-positive triple negative type, on the other hand, impact on prognosis was seen
in luminal type more highly than triple negative type.

Title: EMT activates PERK-eIF2 signaling and sensitizes cells to perturbations in endoplasmic reticulum function
Yuxiong Feng1, Ethan S Sokol1,2, Catherine A Del Vecchio1, Sandhya Sanduja1, Jasper HL Claessen1, Theresa A Proia1, Dexter X
Jin1,2, Ferenc Reinhardt1, Hidde L Ploegh1,2, Qiu Wang3 and Piyush B Gupta1,2,4,5,6. 1Whitehead Institute for BioMedical Research,
Cambridge, MA; 2Massachusetts Institute of Technology; 3Duke University; 4Koch Institute for Integrative Cancer Research;
5
Harvard Stem Cell Institute and 6Broad Institute.
Body: Epithelial-to-mesenchymal transition (EMT) plays an important role in cancer progression. By undergoing an EMT, cancer
cells acquire a spectrum of malignant properties, including invasiveness, multi-drug resistance and stem-like traits. Although they
play an important role in tumor progression and resistance, few vulnerabilities of EMT cancer cells have been reported to date. To
identify specific vulnerabilities of EMT cells, Using small molecule and RNAi screens, we have discovered that induction of EMT
greatly sensitizes cells to agents that perturb endoplasmic reticulum (ER) function. This unexpected sensitivity to ER stress is
mainly due to the expression and secretion of large amount of extracellular matrix (ECM) proteins by cells that have undergone
an EMT. In line with their increased secretory load, EMT cells display a branched ER morphology and constitutively activate the
PERK-eIF2 branch of the unfolded protein response (UPR). Using a PERK-specific inhibitor, we found that PERK activation is
also required for EMT cells to invade and metastasize. In human tumor tissues, EMT gene expression correlates strongly with
both ECM and PERK-eIF2 genes. In summary, our findings identify a novel vulnerability of cells that have undergone an EMT,
and demonstrate that the PERK branch of the UPR is required for their malignancy.

Title: IGF1R/FAK crosstalk is essential for epithelial-to-mesenchymal transition (EMT), migration, and invasion of TNBC cells
LaTonia D Taliaferro-Smith1, Elaine Oberlick2, Tongrui Liu1, Tanisha Z McGlothen1 and Ruth O'Regan1. 1Winship Cancer Institute
of Emory University, Atlanta, GA and 2Harvard Medical School, Boston, MA.
Body: Triple negative breast cancers (TNBCs) are highly metastatic and deadly tumors that currently lack effective prevention
and treatment options. Insulin-like growth factor 1 receptor (IGF1R) and focal adhesion kinase (FAK) both play a role in several
developmental processes, and both IGF1R and FAK signaling pathways have been linked to a number of common pathological
diseases, including breast cancers. In particular, overexpression of IGF1R and FAK are closely associated with metastatic breast
tumors. However, the relationship between the IGF1R and FAK signaling cascades in metastatic TNBCs remain largely unknown.
The present study aimed to investigate the interrelationship between IGF1R and FAK crosstalk in regulating the malignant
properties of TNBC cells. Using stable small hairpin RNA (shRNA)-mediated IGF1R silencing and stable full-length IGF1R
plasmid over-expression methods, we demonstrated that IGF1R was essential for maintaining mesenchymal morphologies of
TNBC and regulating the expression of EMT markers, including vimentin, ZEB-1, Snail-1, E-cadherin, ZO-1, and claudin-1. Using
colony formation, spheroid migration, and Matrigel invasion assays, we further showed that IGF1R promoted migratory and
invasive TNBC phenotypes, including increased colony formation, cell migration, and invasion. Most importantly, IGF1R-mediated
migration and invasion required FAK activation and could be augmented using pharmacological inhibitors of FAK. Our findings in
TNBC cells demonstrate a novel role of the IGF1R/FAK signaling pathway in regulating critical processes involved in the
metastatic cascade. These results may improve the current understanding of the basic molecular mechanisms of TNBC
metastasis and provide a strong rationale for co-targeting of IGF1R and FAK as therapy for mesenchymal TNBCs.

Title: Pax-5 regulates EMT and MET in breast cancer through FAK1 regulation
Sami Benzina1, Pierre O'Brien1, Annie-Pier Beauregard1, Roxann Gurrette1, Stphanie Jean1 and Gilles Robichaud1. 1Universit
de Moncton, Moncton, NB, Canada.
Body: Metastasis accounts for 90% of deaths in breast cancers patients. Therefore, the study of genetic factors regulating cancer
malignancy is a top priority to mitigate the morbidity and mortality associated to this disease. One of these factors, Pax-5,
normally regulates key biological functions such as cell viability, growth, and differentiation. However, an aberrant expression of
this factor results in the development and progression of cancer.
In this study, we developed breast cancer cell models with conditioned expression of the Pax-5 to evaluate signaling pathways
relevant to breast metastasis and cancer progression. We found that Pax-5 extinguishes several aspects of cancer aggressivety
such as: proliferation, spheroid formation, migration and invasion. At the molecular level, we found that Pax-5 modulates cancer
malignancy through the regulation of various components of the epithelial to mesenchymal transitioning (EMT) process in addition
to key signaling targets such as: NFB and the Focal Adhesion Kinase 1 (FAK1). We also demonstrate that Pax-5 decreases
FAK1 level trough up-regulation of miR-135b, a direct repressor of FAK1 expression. Altogether, our findings suggest that the
presence of the Pax-5 lead to less aggressive breast cancers by promoting mesenchymal to Epithelial transitioning (MET). These
findings bring light to molecular mechanisms driving breast cancer malignancy and benefit our quest in the development of
diagnostic and therapeutic strategies against breast cancer progression.

Title: Treatment of metastatic breast cancer using two nanoparticles combined with siRNA targeting Twist1 to inhibit EMT
James B Finlay1, Cai M Roberts1, Gina Lowe1, Ling Peng3, Jeff I Zink2, Fuyuhiko Tamanoi2 and Carlotta A Glackin1. 1City of Hope
Beckman Research Institute, Duarte, CA; 2University of California, Los Angeles, CA and 3Aix-Marseille University Centre
Interdisciplinaire de Nanoscience de Marseille, Marseille, Provence-Alpes-Cte d'Azur, France.
Body: Breast cancer is the 2nd leading cause of cancer related deaths among women in the US with over 240,000 diagnoses
and 40,000 deaths expected in 2014. Among the more serious and deadly forms of breast cancer are the Triple Negative Breast
Cancers (TNBC) (ER-, PR-, HER2-). Mortality rates among patients rise dramatically when these cancers spread beyond the
primary tumor site. Therefore reduction of tumor cell dispersion is a key component to minimizing mortality rates.
Epithelial-Mesenchymal Transition (EMT) is the process by which cancer cells downregulate proteins associated with cell to cell
adhesion (e.g. E-cadherin) resulting in cells that are able to detach from neighboring cancer cells, invade adjacent tissue, and
eventually enter the circulatory system or lymphatics. The process of EMT is tightly regulated by the transcription factor Twist1,
which is often overexpressed in breast cancer. Therefore, Twist1 serves as an excellent therapeutic target whose downregulation
would result in fewer metastatic cancer cells and correspondingly reduce cancer mortality.
Our lab has designed and tested (in vitro) two siRNA based therapeutics that target Twist1 in a TNBC cell line (SUM 1315).
These siRNA molecules have been strategically designed and modified to make them resistant to degradation, enhance silencing
effects and diminish their immunogenicity. To overcome the inherent problems with delivery of siRNA (both in vitro and in vivo) we
have been testing two nanoparticle delivery systems.
Recent advances in nanotechnology have led to the development of a variety of nanoparticles that provide valuable tools for
cancer therapy. We are testing two different types of nanoparticles in this study: The first is a Polyamidoamine (PAMAM)
dendrimer (YTZ3-15) which is a truncated 3rd generation dendrimer which has been modified with a lipophilic tail to enhance
cellular uptake. The second is a mesoporous silica nanoparticle (MSNs) which is able to not only deliver siRNA, but is also
contains pores which allow simultaneously delivering chemotherapies such as doxorubicin. This novel siRNA gene silencing and
nanotechnology-based therapy should allow us to exert more precise temporal control during breast cancer treatment.
When complexed with either nanoparticle we have found significant Twist1 knock down in vitro as well as reduction of Twist1
target genes that are associated with EMT. We have also shown that Twist1 silencing reduces migration and invasion of SUM
1315 breast cancer cells in vitro Recently we began testing these siRNA+Nanoparticle complexes in an orthotopic murine model
using Firefly Luciferase expressing SUM 1315 cells in immunocompromised mice (NSG). The results of this in vivo research have
shown that there is significant uptake of the siRNA+Nanoparticle complexes in the tumors when compared to other organs.
Results of these studies are significant because TNBC is particularly drug resistant and metastatic; and superior therapies are
urgently needed to effectively treat patients with these breast cancers. This approach paves the way for TNBC treatment that
incorporates Twist1 knock down (resulting in renewed chemosensitivity) and simultaneous delivery of a chemotherapy reagent.

Title: A potential role for Janus protein tyrosine kinases in the regulation of epithelial-mesenchymal transition in a model of
epidermal growth factor induced breast cancer epithelial-mesenchymal transition
Teneale A Stewart1, Iman Azimi1, Felicity M Davis1, Erik W Thompson2,3,4, Andrew J Brooks5, Sarah J Roberts-Thomson1 and
Gregory R Monteith1. 1School of Pharmacy, University of Queensland, Woolloongabba, Queensland, Australia; 2St Vincent's
Institute, Fitzroy, Victoria, Australia; 3University of Melbourne, St Vincent's Hospital, Fitzroy, Victoria, Australia; 4Institute of Health
and BioMedical Innovation and School of BioMedical Sciences, Queensland University of Technology, Kelvin Grove, Queensland,
Australia and 5Institute for Molecular Bioscience, University of Queensland, St Lucia, Queensland, Australia.
Body: Background. Epithelial-mesenchymal transition (EMT), a process whereby tumorigenic epithelial cells acquire an invasive
and migratory phenotype, is an important event in the invasion-metastasis cascade. As such, intracellular signaling pathways
involved in the regulation of EMT represent potential therapeutic targets in the treatment and prevention of invasive cancer
subtypes. The calcium ion, a highly versatile intracellular messenger, plays an important role in processes important in
tumorigenesis including invasion and metastasis, and altered calcium signaling has been identified in various cancers. We
recently identified that activation of signal transducer and activator of transcription 3 (STAT3) and expression of specific EMT
markers in the MDA-MB-468 cell line model of epidermal growth factor (EGF) induced EMT display some calcium dependence.
While the calcium permeable ion channel TRPM7 was shown to partially regulate this STAT3 activation and vimentin expression,
the precise mechanisms of their regulation are not yet fully understood. The aim of this research was to investigate the upstream
intracellular signaling pathway involved in EGF stimulated STAT3 activation and the subsequent induction of EMT in this model.
Methods. MDA-MB-468 basal-like breast cancer cells were pre-treated for 1 hr with the Janus protein tyrosine kinase (JAK)
inhibitor, JAK inhibitor I (1 and 10 M), or the Src family tyrosine kinase inhibitor, PP2 (0.1, 1 and 10 M), followed by stimulation
with EGF (50 ng/mL) for 10 or 20 min, and 24 h to assess effects on STAT3 activation and/or EMT marker expression,
respectively. Total cellular protein was isolated following inhibitor treatment EGF stimulation, and the level of phosphorylated
STAT3 (10 or 20 min) or vimentin protein expression (24 h), was analyzed using Western blotting. Cellular RNA was isolated
following inhibitor treatment EGF stimulation and levels of vimentin mRNA (24 h) were assessed using real time RT-PCR.
Results. Treatment of MDA-MB-468 breast cancer cells with JAK inhibitor I resulted in a significant decrease in EGF stimulated
STAT3 phosphorylation, while inhibition of Src family tyrosine kinases with PP2 also significantly decreased EGF stimulated
STAT3 phosphorylation. In addition to its effects on STAT3 phosphorylation, pre-treatment of MDA-MB-468 cells with JAK
inhibitor I also appeared to decrease EGF-induced vimentin protein and mRNA expression, indicating a potential role for Janus
protein tyrosine kinases in the induction of EMT in this model.
Conclusions. Janus protein tyrosine kinase signaling appears to play a role in the regulation of STAT3 activation, and the
induction of the EMT marker vimentin in the MDA-MB-468 cell line model of EGF-induced EMT. Future studies will focus on
investigating the specific JAK family member(s) involved in the EGF-STAT3 signaling pathway, as well as the nexus between
calcium and identified regulators of EGF stimulated STAT3 activation, and EMT marker expression, in this model of breast cancer
EMT.

Title: Niclosamid overcomes epithelial-mesenchymal transition of lapatinib resistance through inhibiting Src activation in HER2
positive breast cancer
Junjun Liu1, Xiaosong Chen1, Yan Mao1 and Kunwei Shen1. 1Comprehensive Breast Health Center, Ruijin Hospital, Shanghai
Jiaotong University School of Medicine, Shanghai, China.
Body: Background: The Src pathway is known to regulate tumor matastasis and plays a role in epithelial-mesenchymal
transition(EMT). Lapatinib, although, has improved clinical outcome for HER2 positive breast cancer patients, acquired resistance
to lapatinib remains an important reason influencing its clinical efficacy. Our previous research shows lapatinib acquired
resistance HER2 positive breast cancer cell lines, SKBR3-R and BT474-R have EMT phenomenon and multiple pathway
activated. Niclosamide is an FDA-approved antihelminthic agent, which is found has cytotoxicity effect on tumor stem cells
recently. We wished to determine the effect of niclosamide on EMT and lapatinib resistance cells and investigate niclosamide as
a potential therapeutic agent for HER2 positive breast cancer.
Methods: SKBR3 and BT474, two HER2 positive breast cancer cell lines, were continuously exposed to increasing doses of
lapatinib to establish two stable cell lines resistant to lapatinib, SKBR3-R and BT474-R. Cell proliferation was determined by
CCK8 assay. Protein expression was determined by western-blotting. Invasion ability was analyzed by transwell assay. FITC
staining flow cytometry (FCM) was conducted to observe the percentage of apoptosis.
Results: Both two HER2 positive lapatinib resistant cell lines, SKBR3-R and BT474-R had EMT phenomenon. Niclosamide had a
stronger inhibition effect on lapatinib resistant cell lines than parental cell lines, and induced more apoptosis by FCM.
Western-blot showed niclosamide could reverse the EMT phenomenon of SKBR3-R and BT474-R with E-cadherin up-regulated
and snail,vimentin down-regulated at the concentration of 0.5-1M. A significant reduction of Src signaling was also confirmed, as
well as the downstream Akt and Erk pathway. After adding niclosimide for 48 hours, SKBR3-R and BT474-Rs capability of
invasion were inhibited.
Conclusion: Our results suggested that niclosamide had a strong cytotoxic effect on HER2 positive lapatinib resistant breast
cancer cell lines. The EMT induced by lapatinib resistance could be reversed by niclosamide, associated with the inhibition of Src
pathway activation, as well as the downstream pathways. Niclosamide treatment produced reduced levels of invasion, serving as
a novel therapeutic way for lapatinib-resistant breast cancer patients.

Title: Prostaglandin E receptor 2 (EP2) regulates breast cancer stem-cell like property and promotes epithelial-mesenchymal
transition
Vivian Y Shin1, Man T Siu1, John C Ho1, Isabella Cheuk1, Jiawei Chen1 and Ava Kwong1,2. 1University of Hong Kong and 2Hong
Kong Hereditary Breast Cancer Family Registry, Hong Kong.
Body: Background: Presence of cancer stem-like cells (CSCs) is the main obstacle for poor treatment response and mortality in
breast cancer patients. Prostaglandin E receptors have been reported to play a role in epithelial-mesenchymal transition (EMT)
and metastasis, however, the contribution on cancer stem cell compartment remains unstudied.
Methods: Human xenograft breast cancer model was used to study the expression of EP receptors during cancer development.
Construction of stable EP2-expression cells was used to study tumorigenesis and characterization of EP2 receptor. Functional
role of EP2 receptor on cell proliferation, flow cyometry, invasion and EMT gene expression array were performed in transfected
cells. Expression of EP2 receptor was compared in primary tumor tissues by immunostaining and real-time PCR.
Results: EP2 receptor was predominantly expressed in animal tissues during cancer development, as well as in human primary
tumor tissues. In mouse xenograft model, MB-231-EP2 clone developed a more aggressive tumor with a larger tumor size and
showed a significant increase in cancer stem cell marker aldehyde hydrogenase (ALDH1) expression. In vitro study, MB-231-EP2
clone increased colony formation capacity and S-phase entry by the regulation of E-cadherin, TWIST1 and ALDH1. Importantly,
we found that Twist1 expression level was higher in breast cancer patients than healthy controls and was associated with ALDH1
expression.
Conclusions: These findings implicated that EP2 receptor was crucial to nurture CSC phenotype and promote tumorigenesis in
breast cancer. Blocking of EP2 might be a potential therapeutic strategy to improve treatment response for breast cancer
patients.

Title: Interaction of mesenchymal stem cell with breast cancer lineage cells and evaluation of their biological behavior
Fernanda Marques Rey1, Willian Abraham da Silveira1, Heriton Marcelo Ribeiro Antonio1, Renata D Sicchieri1 and Daniel
Guimaraes Tiezzi1. 1Ribeiro Preto School of Medicine, Ribeiro Preto, So Paulo, Brazil.
Body: Breast carcinoma is a highly prevalent and incident disease. The interaction between tumor stroma and malignant
epithelial cells has been reported as a mechanism of resistance to cytotoxic treatment. Mesenchymal cells originating from the
tumor stroma create a favorable microenvironment for cancer stem cells (CTTs) maintenance. CTTs are able to repopulate the
host with tumor cells of the same origin on distant sites. It is postulated that this interaction modulates the ability of tumor to
invade and disseminate. The aim of this study is to analyze the association MCF-7 cell line with human mesenchymal stem cells
(MSC), and evaluate the biological behavior and phenotypic changes. MSC cells derived from Wharton's jelly were co-cultured
with MCF-7. Cell migration assay: MSC cells were seeded into 12 well plates and co-cultured with MCF-7 cells in the proportions
1% MSC + 99% MCF-7, 10% MSC + 90% MCF-7, 30% MSC + 70% MCF-7 and 100% MCF-7 in 5x104 total cells. After 5 days of
culture, two lines cross were traced on the bottom of the well. An inverted microscope with 10X objective was used to photograph
at time 0 hours, 24 hours, 48 hours and 120 hours until confluence. We analyzed the phenotypic changes and mean migration
time of cell culture and co-culture. The confluence of the cell layer of the area of migration occurred after 120 hours only in
co-culture of MSC 30% and 70% MCF-7(163,66m to 0 m). Flow cytometry: MSC cells were seeded into 6 well plates and
co-cultured with MCF-7 cells in the same proportions as described above in 1x105 total cells, and maintained at 37 C with 5%
CO for 5 days. Were quantified cells with CD44+/CD24- profile with anti-CD44 antibody (APC) and anti-CD24 (PE). The ANOVA
with Tukey's post-test were used for statistical analysis. Data were presented as mean standard deviation, p< 0.05. There was
no significant difference in CD44+/CD24- in the comparison groups: MSC vs. cell co-cultures; cell co-cultures vs. cell co-cultures;
MCF-7 vs. 1% MSC+ 99% MCF-7; MCF-7 vs. 10% MSC + 90% MCF-7. We observed a significant increase in CD44+/CD24population comparing MCF-7 vs. co-culture of 30% MSC + 70% MCF-7 (95% CI of diff, -55.56 to -27.71). Conclusions: The
co-culture with MSCs an MCF-7 is associated with an increase in cell migration and, the is a increase in CD44+/CD24- cells.
These findings suggest that the interaction of mesenchymal stem cells with MCF-7 may be is able to acquire potential metastatic
profile.

Title: Autoimmunity in breast carcinogenesis. A new paradigm
Flix Fernndez Madrid1,2, Marie-Claire Maroun1, Leonard Lipovich1, Larry Tait2, Azadeh Stark3, Richard Zarbo3, Lisa Polin1 and
Dhananjay Chitale3. 1Wayne State University, Detroit, MI; 2Karmanos Cancer Institute, Detroit, MI and 3Henry Ford Health
System, Detroit, MI.
Body: The B cell response in breast cancer [BC] may have a tumor-promoting effect. The objective of this work was to
demonstrate that TAAs and cytokines inflict autoimmune damage to the breast creating a chronic inflammatory milieu that
promotes BC progression
Methods: The DCIS.com nude mouse xenograft model of BC was treated with rituximab and dexamethasone. Immunoscreening
of a cDNA library, PCR, and sequence determination. Immunohistochemistry [IHC] for inflammatory cytokines, immunoglobulins
and complement and the TUNEL assay.
Results: Deposition of immunoglobulins and complement was shown on breast epithelial and stromal cells and the involvement
of IL-4, IL-6, IL-17, TNF-alpha and NFkB was demonstrated by IHC staining. We identified autoantibodies recognizing multiple
human breast epithelial and stromal antigens and the Fc and V regions of immunoglobulins. Rituximab and dexamethasone
treatment markedly modified the macroscopic and microscopic appearance of BC and the pattern of cytokine deposition, induced
increase in vascularity, marked stromal hypertrophy and massive debulking of the tumor by drastically reducing the tumor mass
through apoptosis.
Conclusions: These results suggest that autoantibody- and cytokine-mediated autoimmune damage, triggered by TAAs creates
a chronic inflammatory milieu with generation of pro-inflammatory and tumor promoting signals supporting BC progression. The
dramatic results of the treatment of the xenograft model with rituximab and dexamethasone illustrate the potential of treating BC
as a chronic rheumatic autoimmune disease. The demonstration of a major role of autoimmunity in BC progression may have a
transformative impact on prevention, diagnosis and treatment of BC.

Title: Autoimmunity to centrosomal and proteasome proteins in breast cancer. A link to chromosome instability?
Marie-Claire Maroun1, Ofelia Olivero2, Judith Abrams3, Wei Chen3, Azadeh Stark4, Carol Peebles5, Larry Tait3, Dhanajay Chitale4
and Flix Fernndez Madrid1,3. 1Wayne State University, Detroit, MI; 2National Cancer Institute, NIH, Bethesda, MD; 3Karmanos
Cancer Institute, Detroit, MI; 4Henry Ford Health System, Detroit, MI and 5INOVA Diagnostics Inc, San Diego, CA.
Body: BACKGROUND. Perturbation of centrosomal or centrosome-associated proteins has been observed in nearly all human
solid tumors and has been implicated in the origin of chromosomal instability. Ubiquitin-proteasome degradation is highly
dependent on the organizing capabilities of the centrosomes. Here we report that autoantibodies in breast cancer [BC] sera target
centrosomal proteins as well as important proteins involved in proteasome protein degradation.
METHODS. We immunoscreened a T7 cDNA library of BC proteins and the association of the cloned autoantigens with BC was
studied by autoantigen microarray analysis. We used immunohistochemistry [IHC] to investigate the expression of the
centrosome and centrosome-associated autoantigens identified.
RESULTS. Immunoscreening with BC sera led to the identification of autoantibodies recognizing epitopes developing in a family
of proteins located on the centrosomes such as NIMA-related kinase 7, dynein heavy chain domain 3, peri-centriolar material-1,
isomorph CRA, and stathmin-1, the ubiquitin-conjugating enzyme E2, the proteasome 26S subunit and the SUMO/sentrin
peptidase. Antibody reactivity to these proteins which are associated with centrosome assembly, microtubule function and protein
degradation were highly associated with the diagnosis of BC. IHC staining of paraffin-embedded BC sections with specific
antibodies showed that aurora and stathmin-1 and other centrosome antigens are expressed in BC.
CONCLUSIONS. The discovery of autoantibodies targeting important centrosome and proteasome proteins associated with BC
suggests that this immune reactivity could be related to autoimmunity developing in BC. Our findings indicate that these
autoantibodies might be biomarkers of early BC and suggest the possibility of a link with chromosomal instability in BC.

Title: The association between primary breast cancer and thyroid cancer: A meta-analysis
Kyle R Joseph1, Senarath Edirimanne1 and Guy Eslick1. 1University of Sydney, Sydney, NSW, Australia.
Body: Background: Previous studies have suggested an aetiological link between breast cancer and thyroid cancer, however,
there has never been a formal meta-analysis which collates the existing evidence supporting the hypothesis that breast cancer or
thyroid cancer predispose an individual to developing the other.
Method: A systematic search was carried out using the electronic databases Pubmed and Medline. We searched for articles
containing epidemiological evidence of breast cancer following thyroid cancer and vice versa. Additionally, we searched for
articles that included epidemiological data involving the incidence of all second primary malignancies following both breast cancer
and thyroid cancer, and compared the data sets.
Results: The meta-analysis performed in a total of 18 studies, showed that there is a significantly increased risk of developing
thyroid cancer as a second primary malignancy of breast cancer (RR=1.59, 95% CI: 1.28-1.99; I2=79.99, p<0.001). Additionally,
there was marginally increased risk of developing breast cancer as a second primary malignancy of thyroid cancer (RR=1.24,
95% CI: 1.18-1.30; I2=24.67, p=0.18), compared to the general risk of developing a second primary malignancy following thyroid
cancer.
Conclusion: The findings of this meta-analysis suggest that there is less likely to be a common aetiological factor of the two
cancers, but rather a specific aspect of the pathophysiology of breast cancer which pre-disposes individuals to develop thyroid
cancer. Elucidation of the common mechanisms between breast cancer and thyroid cancer will have important implications in
both diagnostic and therapeutic management of these cancers. Benefit of thyroid ultrasound screening after breast cancer
surgery needs to be assessed.

Title: MBC alliance: Coordinating metastatic research from lab bench to clinical trials
Marc S Hurlbert1, Musa Mayer2, Stephanie Reffey3, Elly Cohen4, Susan Colen4, Samantha Finstad5, Katherine McKenzie6, Alison
Butt7, Ginny Mason8 and Lynne Davies9. 1Avon Foundation for Women, New York, NY; 2AdvancedBC.org, New York, NY; 3Susan
G. Komen, Dallas, TX; 4BreastCancerTrials.org, San Francisco, CA; 5National Cancer Institute, Bethesda, MD; 6California Breast
Cancer Research Program, Oakland, CA; 7National Breast Cancer Foundation Australia, Sydney, New South Wales, Australia;
8
Inflammatory Breast Cancer Research Foundation, West Lafayette, IN and 9International Cancer Research Partnership, London,
England, United Kingdom.
Body: The Metastatic Breast Cancer (MBC) Alliance[1] consists of non-profit advocacy, funding organizations and industry
partners who seek to transform and improve the lives of women and men living with MBC. There is no cure for MBC and
metastasis is the cause of virtually all breast cancer deaths. One objective of the Alliance is to understand the MBC research
landscape from basic lab research through translational and clinical trials, to epidemiology, quality of life and patient reported
outcomes. The MBC Alliance partnered with the International Cancer Research Partnership (ICRP[2]), an alliance of
governmental and charitable organizations from the USA, Canada, Europe, Australia and Japan that fund regional, national and
international cancer research grants and awards. The ICRP database of members funded grants contains >60,000 grants since
2000, from >80 member organizations, totalling over $14 billion USD. Each project is coded to a Common Scientific Outline
(CSO), a classification system of broad areas of cancer research. Objective: To review the last ten years of breast cancer
research grant funding related to metastasis and clinical trials to identify the most promising molecular targets, pathways and
therapeutics in development for MBC. Methods: ICRP partner-funded grants related to MBC were queried from the ICRP
database and clinical trials were queried from Clinicaltrials.gov and breastcancertrials.org. We also conducted interviews with 75
experts including advocates, scientists, clinicians, and leaders of professional societies and cooperative groups. The database
searches resulted in a pool of 150 open clinical trials for MBC and 17,985 breast cancer-relevant grant awards. Relevant grant
awards and clinical trials were then coded with the assistance of keywords and manual review, to one or more, of 6 categories
relevant to the hallmarks of cancer (Hanahan & Weinberg[4]) and metastasis (Steeg[5]). Results: The ICRP database contained
2,250 unique awards related to MBC. The awards are predominantly basic research (70%) and translational (24%) and that
profile has not changed significantly over a 5-year period 2008-2012. Awards active in 2012/2013 show that the main areas of
focus are: i) early steps to invasion (10%) and, ii) metastatic colonization (15%). Data from 130 trials have been assigned to the
Hanahan/Weinberg framework: sustaining proliferative signal, resisting cell death, enabling replicative mortality, genome
instability and mutation, tumor promoting inflammation, activating invasion and metastasis, avoiding immune destruction, and
other categories like cancer stem cells. Within these categories, the molecular targets of the drugs being used in the trial were
further subdivided (e.g., PI3, RAF, CDK). Of the trials analyzed, 71 were Ph I or PhI/II, 47 Ph II and 11 Ph III. Conclusion: Using
publicly available databases we were able to develop a comprehensive list of molecular targets, pathways and therapeutics for
MBC that will enable improved coordination of MBC research. [1] https://www.mbcalliance.org/; [2]
https://www.icrpartnership.org/index.cfm; [3] CSO 1.4 (Biology of progression) and keywords associated with metastasis.; [4]
Hanahan D and Weinberg RA, Cell 2011;144:646-674.; [5] Steeg PS. Nature Medicine 2006;12(8)895-904.

Title: ICRP analysis: Environmental influences in breast cancer
Marc S Hurlbert1, Senaida Poole2, Kari Wojtanik3, Samantha Finstad4 and Lynne Davies5. 1Avon Foundation for Women, New
York, NY; 2California Breast Cancer Research Program, Oakland, CA; 3Susan G. Komen, Dallas, TX; 4National Cancer Institute,
Bethesda, MD and 5International Cancer Research Partnership, London, England, United Kingdom.
Body: The International Cancer Research Partnership (ICRP[1]) is an alliance of governmental and charitable organizations from
the USA, Canada, Europe, Australia and Japan funding regional, national and international cancer research grants and awards.
One key activity of the partnership is a database of information about members funded grant projects (N>60,000 grants, from 80
members, totalling over $14 billion USD). Each project is coded to a Common Scientific Outline (CSO), a classification system of
broad areas of cancer research. Breast cancer is the most common cancer in women worldwide, however, only about 5-10% of
breast cancer is attributable to genetic predisposition,[2] and about one third of cases are attributable to known genetic or other
risk factors. In 2013, the Interagency Breast Cancer and Environmental Research Coordinating Committee (IBCERCC)[3]
recommended that funding organizations plan strategically to accelerate the pace of scientific research on breast cancer and the
environment. Thus, the ICRP has developed a mechanism to track activity and trends in research into environmental influences
on breast cancer, to provide a baseline for future assessment of progress. Methods: ICRP-funded grants related to
environmental influences on breast cancer were queried from the ICRP database. We focused on three time points: awards that
were active in 2006, 2008 or 2010. The search resulted in a pool of 11983 breast cancer-relevant awards that was narrowed
further to 1107 awards of relevance using a combination of keyword searches and specific Common Scientific Outline (CSO)
codes.[4] Relevant awards were then coded with the assistance of keywords and manual review, to one or more, of 6 categories
of environmental research: Behavior-Lifestyle, Behavior-Tobacco exposure, Chemicals-Chemical pollutants,
Chemicals-Exogenous hormones, General Infection-Microorganisms and Radiation. Results: Between 2006 and 2010, the
numbers of active awards declined and funding levels also fell. Most of the funded research is focused on behavioral/lifestyle
factors in breast cancer (e.g., diet, alcohol intake, and shift work patterns). Further analysis of the Behavioral/Lifestyle category
reveals that the major area of activity is in the role of nutrition/alcohol in cancer, closely followed by the contribution of obesity and
reproductive factors (age of menarche, parity etc.). Conclusion: We were able to utilize a CSO filter to identify trends in funded
grant projects related to the environment and breast cancer. The decline in numbers and research funding between 2006 and
2010 is concerning. As breast cancer incidence continues to increase, research efforts to understand the causes of increased
incidence are essential. Further research investment in these areas may be required.
[1] https://www.icrpartnership.org/index.cfm
[2] http://www.niehs.nih.gov/health/assets/docs_a_e/environmental_factors_and_breast_cancer_risk_508.pdf
[3] https://www.niehs.nih.gov/about/assets/docs/summary_of_recs_508.pdf (Accessed 28/2/14)
[4] CSO areas 2.1, 2.3, 1.2, 2.4, 6.2 (Etiology, Basic Biology of cancer initiation, and surveillance).

Title: Got patient advocates? The value of patient advocate participation in a large research study to develop personalized
risk-based breast cancer screening strategies
Vernal Branch1, Carolyn Achenbach2, Kathleen G Ross3, Wendy F Cohn2, Martin D Yaffe3, William A Knaus4 and Jennifer A
Harvey2. 1Virginia Breast Cancer Foundation, Richmond, VA; 2University of Virginia, Charlottesville, VA; 3University of Toronto,
Toronto, ON, Canada and 4Northshore University Research Institute, Evanston, IL.
Body: Background: Over the last several years there has been confusion among women about breast cancer screening and
patient advocates are increasingly used to help women understand the changes. In 2009, the U.S. Preventive Task Force
(USPSTF) recommended that women under the age of 50 do not need routine screening. New state laws require breast density
results to be given to women and their providers for making their screening choices. Women are not sure what to do with this
information and are offered little guidance to personalize their screening recommendations. The goal of this study is to develop a
risk model for improved personalized breast cancer screening recommendations. Patient advocates were incorporated throughout
the design, recruitment, analysis, and dissemination phases of the study.
Methods: The research team included three patient advocates who participated as full members in the bi-weekly and quarterly
team meetings throughout the duration of the study. All advocates were breast cancer survivors. The primary components of the
study included focus groups to understand womens knowledge and views on breast density as well as personalized screening, a
telephone survey to gain a broader view on these topics, and recruitment to a case:control study to build a breast cancer risk
model that incorporates an automated measure of breast density.
Results: Enrollment was completed over one year with 3,445 women; 839 cases and 2,606 controls. Study design and resulting
recruitment strategies were reviewed early with regular feedback by the patient advocates. At the advice of the advocates,
Facebook was chosen as primary social media, resulting in nearly 200 posts (stories) and 1583 likes for the project. Many of the
posts were generated by or featured advocates. Regarding the focus groups, the advocates developed the questions. Women
were informed about the study by the advocates and educated about breast density. The advocates were key in using the focus
groups to find the right language for enrollment materials, obtain their perception of the importance of the study, and understand
their views regarding a new model for personalized screening for women. The advocates were likewise key in developing
questions for and analyzing results of the telephone survey. In the analysis phase, the advocates assisted the team in
understanding the results of the risk questionnaires. For example, most women did not know the type of breast cancer that they
had been diagnosed with or even if it was invasive. The advocates confirmed how and why even highly educated women would
not necessarily retain this information. Finally, the advocates will have a strong role in the eventual dissemination of the study
findings to women.
Conclusions: The investigators have developed a breast cancer risk model that includes an automated measurement of breast
density, with the goal of personalizing screening for women. The inclusion of patient advocates throughout all phases of the study
improved knowledge and insight of the investigating team. Their role extended beyond community engagement and development
of study materials. The advocates became integral members of the study team.

Title: Measurements of advocate participation in and contributions to the Mayo Clinic breast cancer SPORE
Cynthia Chauhan1, Wayland Eppard1, Lori Denison1, Debra Gearhart1, Lawanna Holmes1, Ruth Kraft1, Linda Miller1, Mary A Sitta1,
Mary L Smith1, James Ingle1 and Matthew Goetz1. 1Mayo Clinic, Rochester, MN.
Body: [Background] Patient advocates bring a unique, valuable perspective and presence to SPOREs. However, although
patient advocates are recommended members of SPORE grants, guidelines are not given on the role of advocates or on
measurement of the performance of advocates.
[Methods] The Mayo Clinic Breast Cancer SPORE Patient Advocate Advisory Committee has developed a participation
measurement tool. This tool will underpin and augment evaluations of the individual advocate contributions by the program PI and
the SPORE project leaders. The Advocate Advisory Committee Chair will review the evaluations with the individual advocates.
Annually, the advocate committee will review the performance of the group by the standard of the metrics.
[Results] Advocates kept a log of their SPORE activities for four months then worked with a statistician to discern patterns of
measurable, evaluable behaviors. The determined evaluable areas are training, scientific activity, meeting attendance,
communication. Currently, those metrics are being tracked. Advocates developed the metrics together and report activities at the
monthly advocate meetings. Currently, activities are simply tracked and categorized. No minimum standards are set. This will be
the third step in the development of the metrics.
[Conclusions] This approach is the first step in the development of meaningful evaluation of advocate SPORE activities. It
quantifies the participation of advocates in the SPORE and provides the basis for developing a qualitative performance
evaluation. The SPORE PI and the Committee Chair will use the evaluations to interpret the involvement of advocates in SPORE
research and activities and to plan interventions to strengthen the advocate contributions to the SPORE.

Title: ICRP analysis: Obesity research in breast cancer
Kari Wojtanik1, Rhonda Aizenberg2, Susan Higginbotham3 and Lynne Davies4. 1Susan G. Komen, Dallas, TX; 2Pancreatic Action
Cancer Network; 3American Institute for Cancer Research and 4International Cancer Research Partnership.
Body: The International Cancer Research Partnership (ICRP) is an alliance of governmental and charitable organizations from
the USA, Canada, Europe, Australia and Japan, funding regional, national and international cancer research grants and awards.
One key activity of the partnership is the ICRP database of information about members funded grant projects (N>60,000 grants,
from 80 members, totaling $13.6 billion USD). Each project is coded to a Common Scientific Outline (CSO) classification, a
classification system of broad areas of cancer research. Obesity has been associated with an increased risk of developing
several cancer types, including breast cancer. Worldwide, obesity rates have nearly doubled since 1980 (WHO), and there is
significant concern that rising rates of obesity will result in additional obesity-related cancer incidence. Breast cancer is the most
common cancer in women worldwide and its incidence has risen in most countries in the last 30 years.1 In addition, there is
convincing research evidence that body fatness is linked to breast cancer incidence (postmenopause).2 With this in mind, the
ICRP has analyzed obesity-related breast cancer research in its portfolio over three time periods: 2006, 2008 and 2010.
Methods: Using a combination of keyword searches and manual review, a total of over 1040 awards over the period 2006-2010
were found in the ICRP portfolio that were related to obesity and cancer. Of these, 353 awards were considered to be relevant to
breast cancer (relevance 25%). These were assessed by Common Scientific Outline (CSO) areas.3 Results: The numbers of
obesity-relevant awards and research investment were higher for breast cancer than for any other cancer type in the ICRP
portfolio from 2006 to 2010. Research was being conducted across all CSO areas, from basic biology, etiology, prevention, early
diagnosis/prognosis to treatment and cancer survivorship. Between 2006 and 2010, there was a slight decrease in etiology
research (CSO2), and an increase in research into cancer survivorship (CSO6). It is notable that training awards are increasing,
indicating that the research organizations contributing data to this analysis consider workforce training to be a priority area.
Conclusion: We were able to use the ICRP database to identify trends in funded grant projects related to obesity research and
breast cancer. Despite increased numbers of awards, the overall stasis in research funding over this period, and the decline in
investment in etiology is concerning. As breast cancer incidence continues to increase, research efforts to understand the causes
of increased incidence are essential. Further research investment in these areas may be required.
1 http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx (accessed 7th January 2014)
2 World Cancer Research Fund / American Institute for Cancer Research. Continuous Update Project Report. Food, Nutrition,
Physical Activity and the Prevention of Breast Cancer, 2010
3 https://www.icrpartnership.org/CSO.cfm


Title: onlineDeCISion.org: An interactive web-based clinical decision aid for DCIS treatment
Elissa M Ozanne1, Natasha K Stout2, Katharine Schneider3, Djra Soeteman4, Deborah Schrag2, Michael Fordis3 and Rinaa S
Punglia2. 1Dartmouth Institute for Health Policy, Hanover, NH; 2Harvard Medical School; 3Baylor College of Medicine, Houston, TX
and 4Tufts Medical Center.
Body: Purpose: Treatment decisions regarding Ductal Carcinoma in Situ (DCIS) are complex, and patients often have inaccurate
and incomplete understanding of the risks and benefits they face. Our objective was to create a web-based decision aid
(onlineDeCISion.org) that can be used in clinical practice to guide both clinicians and their patients with these decisions.
Methods: We developed a web-based clinical decision aid to provide tailored information about DCIS treatment choices including
an individual patients risk of recurrence, likelihood of long-term breast preservation and survival outcomes following up to 6
different treatment strategies for DCIS (lumpectomy, lumpectomy with radiation, lumpectomy with tamoxifen, lumpectomy with
radiation and tamoxifen, and mastectomy with or without breast reconstruction). The decision aid is populated by our previously
developed simulation model of DCIS outcomes. A theoretical framework and best-practices for web-based decision tools guided
the development of the decision aid including semi-structured interviews and usability testing with a diverse group of
multidisciplinary clinicians and patient advocates.
Results: The decision aid was designed to include these key features: 1) descriptions of treatment options; 2) ability to input
patient health-adjusted age; 3) tailored likelihood of time-specific (10-year and lifetime) recurrence and survival outcomes; and 4)
projections of downstream effects of each treatment. The decision aid provides default recurrence risks based on clinical trial data
but allows clinicians to customize 10-year DCIS and invasive recurrence risks to retain flexibility to display expected outcomes for
individual patients. These estimates can be based on the patients actual age, or age adjusted for health status, allowing for a
more realistic expectation of the benefits each treatment holds.
Conclusion: Our web-based decision aid displays tailored outcomes following different treatment strategies for DCIS, allowing
patients to be better informed about the tradeoffs of treatments available to them and select treatments consonant with their
personal preferences, improving the quality of decision making for DCIS. The interactive design features allow users of the
decision aid the ability to address uncertainty around risks of recurrence and comorbidity risks and facilitate the use of the
decision aid across diverse populations. While the decision aid warrants further evaluation, the results of our study promise to
improve decision making in patients with DCIS.

Title: Treatment adverse effect registration. A PC based electronic patient - hospital communication portal that allows registration
of adverse effects of chemotherapy and returns advice and recommandations for action
Jakob Due1, Helle Gert Christensen1 and Peter Michael Vestlev1. 1Roskilde Hospital, Roskilde, Region Zealand, Denmark.
Body: Background
App. 200 new breast cancer patients are treated each year with chemotherapy at the outpatient ward at Roskilde Hospital in the
Region of Zealand. The adverse effect of the treatment determines whether the patients are able to receive the planned dose.
Studies indicate that the easier access patients have to accurate and relevant information about their disease, the better they
manage. Other studies indicate that the better the patients handle the adverse reactions to chemotherapy, the more they comply
to treatment as planned.
The patients are the best to report the experience of their side effects, so it was an obvious choice to get the patients to
self-report adverse effects.
Aim
The primary aim of the study was to create an online portal that could provide patients with better access to relevant guidance
and accurate information about self assessed side effects
A secondary aim was
- to provide the patients with relevant information at the right time in order to increase their empowerment to act and comply with
the treatment.
- the development of a database based on all the registrations done by the patients for further research and developmental
projects.
Methods
We created, as part of a national project, an online portal, where the patients can make real time records of their side effects. The
adverse effects were graded according to CTCAE standards. According to the grading the patients get practical online advice on
how to act. The portal enabled the patients to get a visual overview of the time-line of their own adverse effects, and a possibility
to compare the timelines of adverse effects with other similar patients. The patients were interviewed by telephone as to their
experience with the use of the portal
Results
55 pt were included, 35 pt gave feed-back to the portal. Median age 61 r. Average IT kompetence level was 3,9 (out of 5):
"experienced user and adequate in surfing on the internet"
Results
28 out of 35 patienter experienced benefit from the advices and instructions given on the portal. It was after the first series of
treatment and in connection with newly experienced side effects that the advices were experienced as most valuable It was
appreciated that the portal could give "round the hour" advices It was appreciated that the advices were practical instructed the
patients as to what they could do themselves.
There were (amoung patients, family and health professionels) a need for a more direct access to the advices and instructions
without login. For a small minority of the patients the portal was not regarded as positive. 15 off the 55 patients did not succeed in
using the portal either because of their general condition during their treatment or because they found it too complicated.
Conclusion
There is a patient demand for an around the clock realtime support to the experienced adverse events.
- It is of great importance that the portal is easy to access and intuitive in use.
- It is essential to find the right balance between secure login and easy access for the patients.
- Mobile access to the portal would bring the help closer to the patients and generate more data for the database.

Title: Decision-making surrounding adjuvant chemotherapy in young women with early stage breast cancer
Shoshana M Rosenberg1, Karen Sepucha2, Kathryn J Ruddy3, Lidia Schapira2, Steven Come4, Virginia Borges5, Evan Morgan1,
Nancy U Lin1, Shari Gelber1, Rulla M Tamimi6 and Ann H Partridge1. 1Dana-Farber Cancer Institute, Boston, MA; 2Massachusetts
General Hospital, Boston, MA; 3Mayo Clinic, Rochester, MN; 4Beth Israel Deaconess Medical Center, Boston, MA; 5University of
Colorado, Aurora, CO and 6Brigham and Women's Hospital, Boston, MA.
Body: Background: There is an increasing recognition that many young women with breast cancer will have favorable outcomes
without chemotherapy. We sought to characterize decision-making surrounding adjuvant chemotherapy treatment (CT) in this
population for whom chemotherapy has historically been a standard of care.
Methods: As part of an ongoing, multi-center, prospective cohort of young women diagnosed with breast cancer at age 40 and
younger, we identified 657 women with Stage I-III breast cancer. Participants were asked to complete surveys by mail that
included questions about socio-demographics, decision-making, and treatment history within the first year following diagnosis.
Tumor characteristics were ascertained via pathology and medical record review. We used Chi-square tests to compare:
decisional involvement (patient-driven vs. shared vs. physician-driven), degree of confidence, and feeling informed about the CT
decision (the latter two measured on a 0-10 scale, categorized as follows: 0-5=low; moderate=6-8; 9-10=high) between women
who did and did not receive CT. To explore clinical appropriateness of the CT decision, we used logistic regression to assess the
relationship between tumor characteristics and non-receipt of CT among women with Stage I/II disease.
Results: Among women with Stage I (n=250), II (n=312), and III (n=95), disease, 66%, 95%, and 100%, received CT,
respectively. A greater proportion of women who had CT were highly confident with their decision compared with women who did
not have CT (80% vs. 60%, p<0.0001); women who did not have CT were more likely to report a low level of feeling informed
about the CT decision compared to women who received CT (20% vs. 5%, p<0.0001). Women who did not have CT were also
more likely to report the final CT decision as made by their doctor (49% vs. 28%) and less likely to report a shared decision (33%
vs. 59%, p<0.0001). Non-receipt of CT in women with Stage I/II disease (n=546) was associated within having node negative
disease, T1 (vs. T2 or larger), Her2- negative, and hormone receptor positive tumors.
Conclusion: Although non-receipt of CT would be expected to be viewed favorably by patients and doctors, we found that women
who received CT felt more confident and better informed than those who received no CT. Given that women who did not have CT
were also less likely to perceive the CT decision as shared, improved communication together with better decisional support may
be beneficial, especially for women who do not receive adjuvant chemotherapy.

Title: Characterizing the metastatic breast cancer patient experience around preparing for a treatment decision
Joanne Buzaglo1, Melissa Miller1, Anne Morris1, Allison Harvey1 and Mitch Golant1. 1Cancer Support Community, Philadelphia,
PA.
Body: Background: An estimated 155,000 people are living with metastatic breast cancer (MBC) in the US. With new
developments in treatment, people are living longer with MBC and are confronted with more complex treatment decisions.
Patient-provider communication is typically inadequate and patients are not fully prepared for communicating effectively with their
doctor.
Methods: Since March 2013, the Cancer Support Community has registered 909 people living with MBC to the Cancer
Experience Registry, an online initiative designed to learn and raise awareness about the psychosocial impact of cancer. 572
registrants responded to questions about their experience with making treatment decisions. This sample was 99% female, 91%
Caucasian, and 69% with a college degree and median age 56. Median time since MBC diagnosis was 3 years.
Results: Before making a treatment decision, nearly all (91%) reported receiving information about their cancer type; 76%
received information about their treatment choices. Only 41% indicated they received information about clinical trials prior to
making a treatment decision. Just over half reported they had quite a bit of knowledge about their treatment options. However,
22% had little or no knowledge about their treatment options. Thirty-eight percent received treatment decision support prior to
making a treatment decision; 45% would have liked more support. Twelve percent had little or no involvement in their treatment
decision-making process. Nearly one-third (29%) did not feel they had a treatment choice and 28% reported they did not have
enough time to make a treatment decision. Those who wrote down a list of questions prior to their first visit to discuss treatment
options with their health care provider felt significantly more prepared to discuss their treatment options (p<0.001). About
two-thirds of MBC registrants were satisfied with various aspects of the treatment decision-making process: outcome of the
treatment(s) received (70%); doctors explanation of the benefits of each option (67%); how they arrived at a decision (66%); how
much they participated in making the decision (64%); and their doctors explanation of the risks and side effects (64%). Sixty-nine
percent thought it would be important to get help with gathering information, and 68% with developing a written list of questions
before their meetings with cancer specialists; only 47% thought it important to obtain audio-recordings of appointments.
Conclusion: Although over two thirds of these women were satisfied with various aspects of treatment decision making including
their communication and interaction with their doctor around the decision, nearly 30% of women thought that they had no choice
or felt rushed in making a decision. Those women who prepared a list of questions prior to a consultation with the doctor were
significantly more prepared in making an appropriate decision. While a small majority of patients report being knowledgeable
about treatment options, a significant proportion report not having enough knowledge or support to fully engage in a treatment
decision. Further efforts are needed to address gaps in the delivery of decision support to MBC patients.

Title: eHealth in modern breast cancer treatment: New possibilities in communication between patients, doctors and nursing staff
Rachel Wuerstlein1, Thomas Kirkovits1, Caroline Drewes1, Daniel Schiltz1, Ingo Bauerfeind2, Renate Haidinger3, Kerstin Paradies4,
Ursula Goldmann-Posch5, Timo Schinkoethe1 and Nadia Harbeck1. 1Brustzentrum am Klinikum der Universitaet Muenchen, CCC
of LMU, Munich, Bavaria, Germany; 2Gynaekologie und Frauenheilkunde am Klinikum Landshut, Landshut, Bavaria, Germany;
3
Brustkrebs Deutschland e.V., Hohenbrunn/Munich, Bavaria, Germany; 4KOK Konferenz onkologischer Kranken- und
Kinderkrankenpflege, Hamburg, Germany and 5Mamazone Frauen und Forschung gegen Brustkrebs e.V., Augsburg, Bavaria,
Germany.
Body: Introduction:
Lack of compliance and adherence in oral and s.c. treatment of breast cancer (BC) are huge problems leading to significant
impacts in morbidity and mortality. During long term treatment, constant patient contact cant be secured and possible side effects
not be treated adequately. Where conventional mailing systems failed, as reported in the PACT program, eHealth could be a
possible solution to increase adherence among patients and to ameliorate the communication between patients, oncologist and
nurses.
The objective of this study is to investigate the actual internet usage habits and property of new media among BC patients, their
oncologists and the nursing staff to find new possible ways to improve compliance and adherence in long term treatment.
Methods:
By using 3 different questionnaires (33 items), the actual usage of internet and modern media among BC patients and their
healthcare professionals (oncologists and nursing staff) is surveyed. Also, the equipment of media (computer, smartphone, etc.)
in private as well as in business use is investigated. Huge care and attention is given to possible future eHealth systems for
additional patient support. The collected data also includes age, sex, workplace and, in case of medical professionals, their
emphasis.
Patients completed the questionnaire at two patient conventions and before consultations, oncologists and nurses were asked to
answer the questionnaire at several local BC meetings.
Results:
631 patients, 120 oncologists and 96 nurses completed the questionnaire in 2013.
The internet usage in general and for health related issues is very high among all three subgroups (patients: 93% and 77%,
respectively; oncologists: 100% and 98%; nurses: 92% and 93%). Among patients, even above age 60, 51% report to use the
internet every day. Taking a look at participants equipment of new media, the property of a personal computer is very high (78%;
99%; 95%).
Medical professionals as well as the majority of patients can imagine getting additional support during long term therapy using
eHealth technologies (e.g. for monitoring of and interventions concerning side effects) (see table).
Discussion:
This survey, which is the first BC specific study representing internet usage habits among BC patients and their medical
professionals, shows high acceptance of new interactive ways of communication between patients, doctors and nurses who are
all taking part in treatment of BC. Introducing eHealth may help increase compliance and improve and individualize the
doctor-patient-relationship which will possibly lead to decreased mortality and higher patient and staff satisfaction.
Table 1
Patients
Doctors
Nursing Staff
Number of participants
631
120
96
Internet usage in general
92.7% (495/534)
100% (119/119)
91.7% (88/96)
PC
77.7% (490/631)
99.2% (119/120)
94.8% (91/96)
Smartphone
25.5% (161/631)
70.8% (85/120)
45.8% (44/96)
Property of media:
Wish for additional patient support:

via Internet
53.8% (276/513)
66.4% (79/119)
56.8% (54/95)
via Smartphone
24.2% (119/492
51.3% (60/117)
31.1% (28/90)
Registration of side effects via electronic devices
41.1% (204/497)
71.2% (84/118)
56.1% (51/91)

Title: Group visits to provide gynecologic care for women affected by breast cancer
Sally R Greenwald1, Sarah Watson1, Tami S Rowen1 and Mindy Goldman1. 1University of California, San Francisco, CA.
Body: Background
Advancements in treatment have resulted in a growing number of women now living with or affected by breast cancer. This
patient population has complex and unique gynecologic needs. Two thirds of breast cancers are hormone positive and these
patients will often be given Tamoxifen or aromatase inhibitors which have specific gynecologic effects. Additionally, oophorectomy
may be done as part of breast cancer treatment. Chemotherapy may induce menopause with the need for safe and effective
treatments for menopausal symptoms. Gynecologists have a very important role in providing care to women affected by breast
cancer. Group visits are increasingly used to provide education and care in the setting of chronic disease, including breast
cancer. There is limited data regarding the use of this type of healthcare delivery in providing gynecologic focused care to this
complex and growing patient population.
Methods
This is a cross-sectional study measuring patient expectations and satisfaction with the practice of group visits as a model for
providing gynecological health information to women affected by breast cancer. From July 2010 through February 2014 we
surveyed women who participated in a two hour new patient visit that involved a one hour information session followed by an
approximately 30 minute individual exam with a gynecologist. The goals were to measure the role of a small group format and an
individuals sense of satisfaction with their gynecology clinic experience. Demographic data was collected to determine if there
were any correlations with satisfaction and womens age, stage of disease, hormone receptor and menopausal status.
Results
A total of 104 women participated during the study period. Over 80% of participants in group visits felt that their expectations were
met during the visit. Eight five percent felt they had a better knowledge of treatments for gynecologic side effects from cancer
treatments. When assessing disease and demographic factors that might be associated with satisfaction, we found that women
with hormone receptor positive breast cancer and those with treatment induced menopause were more likely to prefer group visits
than women with hormone receptor negative breast cancer or natural menopause. Women with stage 3-4 disease preferred the
group education visit experience more than women with stage 1-2 disease.
Conclusion
There is a need to provide comprehensive gynecologic care and education to the expanding population of women impacted by
breast cancer. Our data suggests that group visits are an effective and well-received strategy for providing gynecologic care to
this niche patient practice. Incorporating a group visit model into a practice is a cost-effective educational modality to providing
sources of information and support. Based on our observed differences in satisfaction and preference for group visits by stage of
disease, menopause status and hormone receptivity, we recommend investigations grouping similar patients to further improve
unique patient education and satisfaction.

Title: No Show

Title: Understanding potential gaps in treatment discussions between caregivers/patients with metastatic breast cancer and
oncologists
Musa Mayer1, Helen L Coons2, Stephen Jones3 and Deana Percassi4. 1AdvancedBC.org, New York, NY; 2Women's Mental Health
Associates, Denver, CO; 3Molecular Health, The Woodlands, TX and 4Nielsen Consumer Insights (formerly Harris Interactive,
Inc.), Rochester, NY.
Body: About 5% of newly diagnosed cases of breast cancer in the US involve cancer that has metastasized to distant parts of the
body, and about 20%-30% of patients diagnosed with early-stage breast cancer later experience a distant metastasis. The 2013
Global Count Us, Know Us, Join Us survey found that 53% of US women with metastatic breast cancer (MBC) surveyed wished
they had more time to discuss their needs during healthcare visits, and 60% believed their cancer treatment options are limited.
To better understand unmet patient emotional, informational, and care needs and potential communication gaps in discourse with
physicians, the Make Your Dialogue Count survey was developed for (1) women 21 years of age with MBC; (2) caregivers of
women with MBC who are 21 years of age; or (3) licensed US medical oncologists who treat 5 women with MBC per month.
Surveys were conducted online, by paper, and by telephone in the US from June through August 2014. Respondents to one
survey were not necessarily related to or associated with respondents to another survey. Patient and caregiver data were
unweighted. Oncologist data were weighted by geographic region and years in practice by sex to align with actual proportions in
the population.
The survey was completed by 359 patients, 234 caregivers, and 252 oncologists. Patients/caregivers were mainly white
(81%/73%) or hispanic (8%/18%). Most oncologists practiced general oncology (74%), for a mean of 18 years.
Patients/caregivers lacked some basic disease knowledge: 20%/29% did not know the HER2 status, 16%/20% did not know the
hormone receptor status, and 28%/32% reported being told by oncologists that MBC was possibly curable. At initial diagnosis of
MBC, all groups felt it was important to discuss expectations for treatment efficacy and side effects and long-term plans and
goals, but patients/caregivers reported that these topics were not discussed often enough. When patients were asked what
emotions they felt, and oncologists what emotions they observed in their patients at the time of initial diagnosis of MBC, surveyed
patients were less scared, anxious, distressed, and worried, but also less committed, determined, hopeful, and confident than
oncologists observed their patients to be. Most (71%) patients had 1treatment change (median of 3 changes among those with
treatment changes). Patients at the time of treatment change were less scared and distressed and more hopeful than they had
been at the time of initial diagnosis of MBC, suggesting adaptation to and/or acceptance of disease status over time. Patients said
they were committed to proactively managing side effects (96%) and wanted more information on side effect prevention and
minimization (73%), yet only 57% openly discussed side effects with their oncologists. Still, 46% of patients wished their
healthcare teams did more to help them manage side effects.
The Make Your Dialogue Count survey identified gaps in the discourse between patients/caregivers and oncologists that directly
affect disease management. Closing these gaps and understanding the needs of patients throughout their journey are necessary
to establish effective patient/caregiver-oncologist relationships and provide patients with the care they need.

Title: Treatment of metastatic breast cancer patients by community health practitioners: Practice pattern and competence
assessments
Sara R Fagerlie1, Alison Heintz1 and Maureen Haas1. 1Educational Concepts Group, LLC, Atlanta, GA.
Body: Background: The complexity of current treatment and management for patients with metastatic breast cancer continues
to rise. The volume and pace of scientific advances make it challenging for the community practitioner to stay abreast of optimal
patient care. Education is key in disseminating critical information to practitioners and allows professional reflection of appropriate
therapeutic decision making and peer discussion. Understanding the base knowledge, competence, and current practice patterns
of the community practitioner is critical to identification community needs and implementation of education that impacts patient
care.
Methods: During 2013 and 2014, educational outcomes assessments were gathered during 2 education programs consisting of
47 live independent continuing medical education (CME) activities held within community practices across the USA. Participants
were asked a series of case-based questions via an audience response system to assess baseline knowledge, competence, and
identify practice patterns Assessments were repeated following the 1-hour CME certified activity. Long-term assessment was
conducted electronically 6-weeks following the educational initiative.
Results: The programs educated 958 practitioners, including 515 physicians. Practice patterns for frontline treatment varied at
baseline. Education resulted in treatment more highly aligned with practice guidelines and recent advances. At baseline 75% of
practitioners and 30% of the physician target audience were unable to identify appropriate next steps in a borderline HER2+
patient. The education reduced the gap by 25%. 6-weeks following the activity 100% of the assessed physician target audience
agreed or strongly agreed that the education influenced their interpretation of HER2 testing. Practice gaps in practitioners ability
to identify, recall, and interpret practice changing clinical trials were also identified. Newness of treatment data, lack of
reimbursement, and treatment side effects were the most common barriers identified in applying the education into clinical
practice.
Conclusions: The results highlight the diversity of current clinical practice in the community practice setting for patients with
metastatic breast cancer and the positive impact of focused education. Knowledge and competence practice gaps were identified
to pinpoint specific areas of benefit for the community practice setting in the treatment of patients with metastatic breast cancer.

Title: Managing breast and ovarian cancer risk: A novel approach to teaching residents comprehensive risk reduction and
management strategies
Deborah S Lindner1. 1Northwestern University Feinberg School of Medicine, Chicago, IL.
Body: Introduction: Awareness about genetic predisposition to breast and ovarian cancer and access to genetic testing has
increased dramatically over the last decade, particularly with the increase in direct to consumer marketing of genetic testing.
While professional organizations have published bulletins addressing this topic, a working knowledge of how to identify and
manage high risk patients is lacking among many physicians. The goal of this pilot study was to assess knowledge deficits in
breast and ovarian cancer risk assessment and management among primary care residents and to determine whether a novel
approach to teaching this subject will help bridge the perceived knowledge gap.
Methods: A novel case based learning module was developed by a multidisciplinary team of Breast Surgeons, Gynecologic
Oncologists, Obstetrician/Gynecologists, Psychologists and Genetic Counselors as a practical approach to assessing risk level
based on personal and family history and applying standard recommendations for screening and risk reduction. This approach is
unique, as lectures in an academic setting often focus on management of high risk patients, but not on risk stratification and risk
reduction in all comers. Surveys were completed by residents after the workshop to assess perceived knowledge before and after
the presentation.
Results: 106 residents from Internal Medicine, Obstetrics and Gynecology and Family Medicine residency programs were
surveyed after participating in the lecture and case based learning module. 87 (82%) responded that the information presented
was new to them. 101 (95%) responded that the training increased their knowledge about the options available to young women
for risk reduction and early detection of breast and ovarian cancer. 105 (99%) responded that after attending the training, they
understood how to identify patients that are at an increased risk for developing breast and ovarian cancer and should be referred
for genetic counseling and testing. 100 (94%) indicated they would incorporate material in the training into their everyday practice.
Discussion: Resident surveys confirmed there is a knowledge gap among trainees in breast and ovarian cancer risk assessment
and management. While professional organizations have created guidelines for patient management, most trainees are either
unaware of recommendations, or feel the information is not being presented in a way that allows them to use it practically in
patient management. The novel educational workshop presented here increased perceived knowledge among residents, allowed
them to correctly identify high risk patients, and increased their working knowledge of risk reduction strategies among high and
low risk patients. Recent studies have demonstrated poor coordination of care and lack of followup for the majority of high risk
patients who do not opt for risk reducing surgery within the first few months after diagnosis of a gene mutation. The need for more
comprehensive education among emerging medical professionals in this area is clear. Using a practical, risk stratification
approach to identify and manage high risk patients is successful in bridging the knowledge gap among residents in primary care
specialties.

Title: Multidisciplinary breast cancer care registry and quality control system in the Netherlands: The NABON breast cancer audit
Vivianne C Tjan-Heijnen1, Annelotte C van Bommel2, Margriet van der Heiden-van der Loo3, Pieter Westenend4, Bart de Vries1,
Carolien H Smorenburg5, Agnes Jager6, Marc B Lobbes1, Ruud M Pijnappel7, John H Maduro8, Henk Struikmans9, Marc A
Mureau6, Marga Schrieks10, Carol Richel10, Maike Schepens11, Thijs van Dalen12, Michel W Wouters5, Marie-Jeanne T Vrancken
Peeters5, Emiel J Rutgers5 and Sabine Siesling3. 1Maastricht University Medical Centre, Maastricht, Netherlands; 2Leiden
University Medical Center, Leiden, Netherlands; 3Comprehensive Cancer Centre Netherlands, Utrecht, Netherlands; 4Albert
Schweitzer Hospital, Dordrecht, Netherlands; 5Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands; 6Erasmus
Medical Centre Cancer Institute UMC, Rotterdam, Netherlands; 7University Medical Centre, Utrecht, Netherlands; 8University
Medical Centre, Groningen, Netherlands; 9Medical Centre Haaglanden, Den Haag, Netherlands; 10Breast Cancer Patient
Organisation, Netherlands; 11Health Insurer Netherlands, Netherlands and 12Diakonessenhuis, Utrecht, Netherlands.
Body: Background
Previous quality assessments in oncology focused on surgical issues such as number of annual operations per hospital.
However, high-quality of care depends on an excellent interplay between all disciplines involved in cancer care. The NABON
(National Breast Cancer Organisation of the Netherlands) has, therefore, developed a multidisciplinary set of 30 quality indicators
to check and improve breast cancer care. Health insurers have recently decided to use this information for purchasing.
Methods
The NABON Breast Cancer Audit (NBCA) started in 2011. Data on all newly diagnosed patients with invasive breast cancer in the
Netherlands are collected by the Netherlands Cancer Registry (n=61 hospitals) or by the physicians themselves (n=31 hospitals).
Data capture is facilitated using a web-based portal and feedback to participating hospitals on their own data is being done every
week. Since 2012 all Dutch hospitals participate. A set of quality indicators on process and outcome was selected following
established clinical guidelines and is being supervised by a multidisciplinary steering committee.
Results
Data of all 41,958 breast cancer patients treated between 2011 and 2013 were collected. In 2013, 94% of patients were
discussed in the multidisciplinary team prior to first treatment and 98% after surgery. BI-RADS score was used in 98% of
radiological reports. After neo-adjuvant chemotherapy, 7.6% of patients had positive specimen margins following first breast
conserving surgery compared to 5.0% and 20% of patients following primary breast conserving surgery for invasive cancer and
ductal carcinoma in situ, respectively. Eighteen percent and 36% of patients underwent immediate breast reconstruction after
mastectomy for invasive cancer and ductal carcinoma in situ, respectively. Pathological analysis showed 12% of patients had
HER2 positive and 85% had ER positive disease. Neo-adjuvant or adjuvant systemic therapy was given to 63% of patients and
neo-adjuvant or adjuvant chemotherapy to 39% of patients. Time between diagnosis and first treatment was generally short: 49%
to 85% underwent primary surgery with and without immediate reconstruction, respectively, and 76% underwent neo-adjuvant
chemotherapy within 5 weeks from first biopsy. At the conference we will show that variation between hospitals was not related to
annual surgical volume per hospital.
Conclusion
The NBCA is a unique national system to provide and confirm quality assessment in breast cancer and to drive improvements in
quality of multidisciplinary breast cancer care. Present results show an overall high quality of care in the Netherlands and provide
insight in items of improvement.

Title: Education and information preferences for women with triple-negative breast cancer: Should personal or medical
demographic variables impact program tailoring?
Kathleen D Swiger1, Jocelyn A Sendecki3, Janine E Guglielmino2, Hope S Rugo, Carey K Anders4, Susan M Domchek5, Arin
Ahlum Hanson2, Hayley Dinerman6 and Catherine Creme Henry2. 1Consultant, Greensboro, NC; 2Living Beyond Breast Cancer,
Haverford, PA; 3Thomas Jefferson University, Philadelphia, PA; 4University of North Carolina, Chapel Hill, NC; 5University of
Pennsylvania, Philadelphia, PA and 6Triple Negative Breast Cancer Foundation, Norwood, NJ.
Body: Purpose: To determine interest in tailored breast cancer education and information among women diagnosed with breast
cancer based on personal and medical demographics, including subtype of breast cancer, cancer stage, age, living situation
(single, married, children, student, retired), race/ethnicity, or sexual orientation.
Respondents and Methods: An 80-question online survey was designed to identify education, information, and support needs
among women diagnosed with breast cancer after 2006. Respondents self-identified their breast cancer subtype and selected up
to 3 options from a list of preferences about ways in which education and information needs should be tailored. Options were
"breast cancer type," "age," "stage," "living situation," "race/ethnicity," "sexual orientation," "no preference," and "other." The
responses of 656 women diagnosed with triple-negative breast cancer (TNBC) (25.1%) were compared to those of 1,954 women
with other subtypes of breast cancer (74.9%). Logistic regression was used to assess differences in tailoring preferences between
TNBC and non-TNBC women.
Results: Compared to non-TNBC women, TNBC women had a significantly stronger preference for information tailored to breast
cancer subtype (71% vs. 49%, p<0.001) and race/ethnicity (5% vs. 2%, p=0.002) and significantly lesser preference for tailoring
based on cancer stage (43% vs. 47%, p=0.004) and living situation (15% vs. 25%, p<0.001). The difference in preference for
racial/ethnic tailoring between TNBC and non-TNBC participants may be due to the overall low proportion of non-white
respondents to the survey (18.4%) and the higher proportion of non-whites in the TNBC group (18% vs. 11%, p <0.001). There
was a significantly different pattern of preference by age group between TNBC and non-TNBC women. TNBC women under 40
had a stronger preference for tailoring to their age group than younger non-TNBC women (age <30, 89% vs. 77%, age 30-39,
74% vs. 62% p=0.02). Across women of all breast cancer subtypes, cancer stage played a role in tailoring preference. Women
with higher stage (stage >=2) breast cancer had a stronger preference for tailoring materials based on cancer stage (p<0.001);
this pattern did not differ by TNBC status.
Conclusion: Education and information tailoring preferences show marked differences by breast cancer subtype. Women with
TNBC strongly prefer education and information be tailored to their breast cancer subtype and their race/ethnicity, but are less
interested in cancer stage or living situation-specific tailoring. Additionally, younger age influenced preferences in women with
TNBC. Across all women, advancing cancer stage influenced preference toward stage-tailored materials. Healthcare providers,
cancer centers, and breast cancer organizations should consider developing education and information tailored to the needs of
TNBC patients and survivors.

Title: Improving completeness of client-level data collection in the Avon Breast Health Outreach Program through electronic tablet
technology
Lindsay Senter1, Marvin R Aliaga1, Kelly M Opdyke1, Kathryn Gates-Ferris1 and Marc Hurlbert2. 1Cicatelli Associates Inc (CAI),
New York, NY and 2Avon Foundation for Women, New York, NY.
Body: Background: The Avon Breast Health Outreach Program (BHOP) supports community-based organizations to provide
breast cancer education and outreach, and navigate low income and uninsured women to breast cancer screening and treatment.
BHOP organizations collect demographic and health information on clients through a standardized interview using the Clinical
Intake Form (CIF). Research demonstrates that transitioning away from paper completion to audio-computer assisted
self-interview allows for increased rates of survey completion. Given the advancement of tablet technology, there is significant
opportunity for organizations to adopt electronic data collection of clinical information to improve data quality.
Objective: This presentation compares rates of CIF completion collected among a low income, diverse client population via two
modalities; self-administered paper-based interviews (SAPI) versus electronic collection using iPads.
Methods: The CIF is organized into three sections: demographics [12 items], breast health [9 items], and access/use of health
services [5 items]. In 2013, 26 BHOP grantees adopted and utilized iPads to administer the CIF with the iSurvey application, after
receiving training and technical assistance on tablet usage. Data were restricted to women >=18 years, who self-administered the
CIF, English-only, collected between March and December 2013, and among only those grantees who employed both modalities.
The BHOP dataset was analyzed to compare CIF completeness across the two modalities. Completeness is defined as having a
response to every question in the CIF; missing is defined as any question that was not responded to or was left blank. Percent of
completeness and mean number of missing responses in the CIF overall and by section were analyzed by modality. Odds Ratios
(OR) on completeness by modality were also calculated controlling for race, income, education, place of birth and residence.
Results: There were 8,004 CIFs analyzed; 2,144 via iPad (27%) and 5,860 via SAPI (73%). Among the iPad mode, 84% of the
CIFs had complete data, compared to 47% in the SAPI. The mean number of missing values across the entire CIF was 0.2 via
iPad, compared to 1.3 with the SAPI. SAPI variables had ten or more times as much missing data as the iPad variables (e.g.,
"self-reported breast symptoms" item had 7% missing data in SAPI vs. 1% in iPad). Adjusting for key demographic
characteristics, clients were seven times more likely to complete the entire CIF via the iPad as compared to the SAPI [Overall
OR= 7.3(6.0-9.0)].
Conclusions: Given the significantly higher rates of CIF completion using iPads, there are important implications for health care
organizations to adopt and utilize electronic collection of their health interview data. Future research should study iPads ability to
handle high patient volume in less time and with more accuracy compared to SAPI.
Discussion: Tablets are a relatively inexpensive option for organizations dedicated to assuring complete data collection. While
administrative barriers arise with new technology, the significantly positive aspects of tablet data collection of increased quality
and completeness overshadow such burdens.

Title: Perceptions about cancer clinical trials among metastatic breast cancer patients: Findings from a patient powered registry
Anne Morris1, Melissa Miller1, Allison Harvey1, Mitch Golant1 and Joanne Buzaglo1. 1Cancer Support Community, Philadelphia,
PA.
developments in treatment, people are living longer with MBC. Advances in treatment are handicapped by patients limited (3-5%)
participation in cancer clinical trials (CCTs). While much effort has been placed on better training physicians and health care
providers in talking to their patients about the appropriateness of such participation, barriers remain.
Experience Registry, an online initiative designed to learn and raise awareness about the psychosocial impact of cancer. 557
registrants responded to questions about their beliefs, attitudes and experience with CCTs. This sample was 99% female, 91%
Caucasian, and 69% with a college degree and median age 56. Median time since MBC diagnosis was 3 years.
Results: Only half of registrants reported that a member of their health care team spoke to them about participating in CCTs.
Nearly one-quarter (23%) report that they took part in a CCT. 48% considered a CCT as a treatment option for their MBC and
34% did not know if a CCT was available to them. 67% are uncomfortable with being randomly assigned to a treatment. 18% do
not trust the medical establishment and fear they will be used as a "guinea pig" for research. 63% fear receiving a placebo in a
CCT; however, only 6% report that they dont understand what clinical trials are. The table below shows differences in beliefs
between those who report participating in a CCT and those who have not participated in one.
Statement
percent agree or strongly

percent agree or
agree if did NON take part in
strongly agree
CCT n=419
percent agree or strongly

agree if DID take part in
CCT n=126
I am uncomfortable with being randomly

assigned to a treatment
67
68
60
0.001<
I do not trust the medical establishment

and fear I will be used as a guinea pig for
research
18
20
0.01
I fear receiving a placebo in a clinical trial
63
66
52
0.001<
I do not understand what clinical trials are
0.001<
I fear side effects that might come with

treatment on a clinical trial
54
58
43
0.001<
I would be unable to fulfill trial requirements

due to logistical barriers such as
22
transportation
24
13
0.001<
There are no clinical trials available in my

community
22
25
12
0.001<
My health insurance would not cover it
27
30
18
0.001<
Conclusion: This population of MBC patients reports a range of experiences and often conflicting beliefs about CCTs. While this
survey is limited to self-report data, there is persistent suspicion among these women about participating in CCTs. Although these
women report that they understand what a CCT is, they fear receiving a placebo and being randomized to different treatment
arms without at least the standard of care. These misconceptions persist even among people who have participated in a CCT.
These findings highlight the need for treatment decision counseling that educates patients about all their treatment options
including CCTs and for training of health care providers to better inform patients about CCTs.

Title: The financial costs of metastatic breast cancer and the decisions patients make to cope with costs: Findings from the
Cancer Experience Registry
Joanne Buzaglo1, Anne Morris1, Melissa Miller1, Allison Harvey1 and Mitch Golant1. 1Cancer Support Community, Philadelphia,
PA.
developments in treatment, people are living longer with MBC and have to manage greater financial burden related to care,
including copays and out of pocket costs.
Experience Registry, an online initiative designed to raise awareness about the psychosocial impact of cancer. 496 registrants
responded to questions about the financial cost of MBC. This sample was 99% female, 91% Caucasian, and 69% with a college
degree and median age 56. Median time since MBC diagnosis was 3 years.
Results: 38% of registrants report being seriously or very seriously concerned about health insurance or money worries and 46%
reported currently experiencing intrusive ideation about the financial cost of care. Registrants reported experiencing a significant
burden from MBC related expenses, as shown in the table below.
Please tell us how much of a burden the following additional expenses caused you. (0=not at all to 4=very much)
3 (%)
copays for medical treatments (e.g., surgery, chemotherapy, radiation, etc.)
492 46.1
Prescription drugs and over the counter medication (including co-pays)
492 40.9
Copays for medical appointments/visits
492 39.3
Diagnostics or treatment not covered by insurance
492 39.7
Complementary medicine or alternative therapy (e.g., vitamins, homeopathy)
491 37.9
Travel (parking, gas)
489 20.6
Medical supplies
489 21.3
Late fees on bills
492 19.6
Counseling or therapy
488 19.9
Because of these expenses, registrants have: foregone vacations, celebrations, and social events (53%); sold property (12%);
refinanced their house (13%); filed for bankruptcy (5%); downsized their living accommodations (16%); liquidated their assets
(19%); depleted their savings (40%); borrowed against or used money from a retirement plan (27%); cut their grocery expenses
(42%); applied for or used public assistance (14%); chosen a treatment that is not as effective but costs less (9%); asked their
doctor if there was a less expensive treatment (19%); tried to negotiate payments with credit companies (24%); negotiated with
service providers to reduce costs (22%); used pharmaceutical assistance programs (26%); accepted money from friends or family
(39%); and cashed in a life insurance policy early (8%), among others. In order to reduce the cost of treating MBC, registrants
often or always postpone seeking psychological counseling or support (20%) and delay follow up on recommendations for
complementary treatment such as physical or occupational therapy and nutrition counseling (13%).
Conclusion: MBC places a significant financial burden on patients, which can result in patients taking measures that can
significantly impact their quality of life. Future implications for research include the development and evaluation of interventions
designed to enhance doctor-patient communication and support (e.g., financial counseling) to ensure that the financial cost of
MBC does not negatively impact the patients quality of life, course of cancer care, and health outcomes.

Title: "The curriculum for recovery," a five week educational and experiential support group for breast cancer survivorship
recovery
Robert E Fisher1, Marianne Stenhouse1 and Anh Lai-O'Connell1. 1Rocky Mountain Cancer Centers- Longmont, Longmont, CO;
2
Rocky Mountain Cancer Centers- Longmont, Longmont, CO and 3Rocky Mountain Cancer Centers- Aurora, Aurora, CO.
Body: Introduction:
Breast cancer survivors benefit from supportive, professionally facilitated group discussions to assist their individual recovery from
treatment and place their cancer in a life perspective. We have developed a multi-session group discussion to aid survivors in this
path of recovery.
The Pink Ribbon Survivors Network (www.PinkRibbonSurvivorsNetwork.org) has developed an online library for breast cancer
survivors titled, "The Curriculum for Recovery." This library contains professionally vetted online articles in 19 categories related
to survivorship issues facing breast cancer survivors. The library houses over 400 articles, in total. We have developed a 5
session program of discussion for breast cancer survivors integrating this library resource as a basis for discussion material.
The 5 sessions are titled: 1.) Physical Recovery and New Adjustments, 2.) Nurturing Yourself: The Importance of Self-Care 3.)
Fostering Positive Relationships, 4.) Understanding your Emotions: Doubt and Hope in Survivorship, and 5.) Prioritizing, Leaving
Your Legacy, Moving Forward.
Each session draws from specific categories of the Curriculum for Recovery Library. Participants are asked to read specific
articles from the online library in preparation for group discussion. The 5 sessions lead the participants from the issue of
immediate physical recovery to gaining a philosophic life perspective on the impact of breast cancer in their lives.
Results:
Website usage: Since its inception in July, 2012, the website's resources have been accessed from 2300 cities in 128 countries.
It's libraries have been viewed 9,000 times, tallying over 30,000 page views. The Survivors' online library hosts information in 19
different categories. Separate online libraries specialize in clinical literature for cancer care professionals and primary care
professionals.
During registration for this group, participants will be asked about coping issues, and their goals in this group participation.
Feedback after each session will be obtained. We intend to develop a manual for professional facilitators to use in order to be
able to facilitate similar groups in their center. This manual will be available online within our website's libraries.
Discussion:
Studies have shown that breast cancer survivors benefit from group participation for support, education and discussion during
their breast cancer recovery. We have developed a multi-session group discussion program that can be replicated free of charge
at any location, utilizing the online resource noted here, along with the comprehensive facilitator's guidelines. This program is
designed to bring survivors from initial issues of early recovery to gaining a broad perspective of how breast cancer has affected
the individual's life and her future. This program is well represented in a brief live presentation at the San Antonio Breast Cancer
Symposium that will be illustrative of its value to breast cancer survivors, and the ease of replication at other locations.

Title: Psychological status of early breast cancer (EBC) patients (pts) after surgery and before the starting of aromatase Inhibitors
(AIs). Results of a single-institution, prospective study
Marina E Cazzaniga1, Annalisa Valentini1, Francesca Gallina1, Francesca Riva1, Alessandra Crippa1, Marco Tagliabue1, Federica
Cicchiello1, Diego Cortinovis1 and Paolo Bidoli1. 1AO S Gerardo, Monza, MB, Italy.
Body: BACKGROUND
AIs are the milestone adjuvant treatment for postmenopausal early HR+ BC pts. Non-steroidal AIs induce a significant oestrogen
deprivation, responsible in approximately 40% of the pts for muscle-skeletal adverse events (MSKs) and for discontinuation in
25%. The present study aims to evaluate the impact of the diagnosis communication and of the need for AI therapy on the future
development of MSKs events in a prospective cohort of EBC pts.
PATIENTS AND METHODS
The analysis of patients psychological attitude was conducted by using 3 different tests: SCL 90-r for psychological symptoms,
SF-36 for the self-perception of the health status and COPE-NVI for evaluation of coping. After a full explanation of the AIs
potential adverse events by the oncologist, the questionnaires were self-administered with the support of a dedicated psychologist
at the moment of the first visit (T0). For the future correlation between baseline psychological attitude and the development of
AI-related events, the presence and the intensity of pain were measured by NRS 11-points scale and by Tender Points (TPs)
evaluation. Pearson correlation analysis was used to calculate correlations between Pain Scale, Physical Synthetic Index of Sf-36
and NRS/TRPs scores. Wilcoxons Test was used to compare baseline scores with subsequent ones.
RESULTS
From October 2012 to April 2013, 70 EBC pts entered the study. Exclusion criteria included severe osteoporosis or other co
morbidities that contraindicate therapy with AIs and documented diagnosis of psychosis or cognitive deterioration. Median age
was 64.8 years (43-88). The vast majority of the pts (584.3%) presented stage IA-IIA disease. Fifteen pts aged less than 55 years
were menopausal, 13 have previously received chemotherapy as part of the planned adjuvant programme. Thirty-four pts (48.6%)
had a baseline NRS>0 and 18 NRS>5; 26 pts presented at least 1 positive TP. Regarding the SCL 90-r analysis, all the pts have
reported standardized scores within the normal ranges for somatization, anxiety and phobic anxiety. The analysis of SF-36
questionnaires indicated that pts showed a significant lower score in comparison to the reference population in the ISF Index
(44.17 vs 48.7) and in the Physical Health Scale (47.5 vs 74.13). These differences could suggest a greater difficulty to do daily
activities and a worse global index of self physical perception
At baseline, there was a statistically significant correlation between Physical Pain Scale of SF-36 and NRS Pain score (p=0.001)
and between Physical Synthetic Index of SF-36 and NRS Pain Score (p<0.001). No differences have been observed in the COPE
NVI questionnaires between the studied population and the reference one.
CONCLUSION
Early BC pts who are candidate for endocrine adjuvant therapy with AIs often present an impairment in terms of physical activity
(daily and working) and perception of physical pain. The analysis of these parameters along the 1st year of AI therapy and their
correlation with MSK adverse events onset is ongoing.

Title: Sources of information and influence on decisions regarding contralateral prophylactic mastectomy: A prospective study
Swati A Kulkarni1, Martha Van Haitsma1, Kristen Wroblewski1, Sarah Rabbit2 and Kathy Yao2. 1University of Chicago, Chicago, IL
and 2NorthShore University HealthSystem, Evanston, IL.
Body: Background: The number of women with a newly diagnosed ipsilateral breast cancer requesting contralateral prophylactic
mastectomy (CPM) has increased significantly in recent years. Little is known about the sources of information women utilize as
part of their decision making process regarding the choice of CPM.
Methods: A 55-item survey validated with breast cancer survivors was administered to 136 women with newly diagnosed breast
cancer before they underwent surgical therapy at two institutions. Women were asked about the influence of medical personnel,
individuals outside of medicine, websites and social media on their decision to keep or remove their healthy breast.
Results: The median age was 58 years (range 30-85); 69.2% were White, 26.7% were African American and 4.2% were Asian;
Thirty-eight (28.6%) of patients had a first-degree relative with breast cancer. Seventy-five (58.6%) had a lumpectomy, 41 (32%)
had a unilateral mastectomy and 12 (9.4%) had a CPM. About 80% of women felt the information they were given about breast
cancer treatment was consistent. Fifty-five patients (41.7%) sought out extra information beyond what the doctor gave them.
Sources of information that had strong or some influence were websites and books recommended by their doctor (44%),
websites, followed books not recommended by their doctor (30.4%), and magazines (15.4%). Only 6.7% said that social media
had some or strong influence on their decision making; 61.7% stated that they completely or mostly relied on their doctors advice
to make treatment decisions and ranked their doctors spoken advice (74.3%) as the most important source of information in
making their decision to keep or remove their healthy breast. However, Eighty-two women (62.6%) stated that information
provided by staff did not include any information about removing the healthy breast. Forty-one (36%) stated that no one discussed
contralateral breast cancer risk and 67 women (59%) stated that no one talked with them about the possibility that cancer could
turn up somewhere else even if both breasts were removed. Of medical personnel, women stated that their surgeon (81%) had
the most influence on their decision, followed by their oncologist (59.1%) and reconstructive surgeon (29.8%). When asked who
influenced their decision outside of medicine, 58.7% of women stated that other breast cancer survivors influenced them.
Reconstructive surgeons and the spouse/partner had more influence on those <50 years old compared to >50 years old (p=0.03).
Oncologists (p=0.004), nurses and trained counselors (p=0.06-0.08), had more of an influence for African Americans compared to
Whites.
Conclusions: Many women with newly diagnosed breast cancer do not receive information about CPM, contralateral breast
cancer risk or how CPM affects recurrence from their physician. However, women still highly value their doctors advice and
information provided by their doctor but one third of women state that websites identified on their own strongly influence their
decision making for CPM. These findings highlight an opportunity for physicians to educate women about the utility of CPM as
part of their surgical treatment.

Title: The association between exercise behavior and patient-reported outcomes in women with early breast cancer receiving
locoregional radiation therapy
Ritu Arya1, Lee W Jones2, Rachel C Blitzblau1, Manisha Palta1, Lisa Massa1, Gloria Broadwater1 and Janet K Horton1. 1Duke
Cancer Institute, Durham, NC and 2Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Purpose/Objectives: RT is associated with acute treatment-related complications that can lead to poor quality of life (QOL)
and fatigue. Exercise has been shown in other cancer treatment settings to improve negative outcomes. We conducted a
prospective pilot study to explore the association between exercise, patient-reported outcomes (PROs), and radiation therapy
(RT) toxicities.
Materials/Methods: Patients with surgically excised ductal carcinoma in situ or invasive breast carcinoma receiving curative
radiation were enrolled. Each patient completed an exercise behavior/QOL survey at two time points: before the patients 5th
fraction of radiation and during the last week of treatment. Limb girth measurements to evaluate lymphedema and assessment of
shoulder range of motion (ROM) were completed on all patients at the same two time points. Skin toxicity was assessed weekly
throughout radiotherapy. Exercise behavior was quantified with the Godin Leisure Time Exercise Questionnaire (metabolic
equivalent [MET] hours per week). Patients with >7 METs/week were designated the exercise cohort (n=10) and those with
<7METs/week the non-exercise cohort (n=10). PROs were measured using standardized questionnaires. Continuous variables
were compared using the Wilcoxon rank sum test and percentages were compared using Fishers exact test.
Results: Median patient age was 56 (range 28-70) years. Median MET in the exercise cohort was 12.9/week (range 7.5-35.0,
n=10); 0.0/week in the non-exercise cohort (range 0-5.8, n=10). Women in the exercise cohort experienced significant
improvement in depression scores over the course of treatment as compared to those who did not exercise (p=0.013). Those in
the exercise cohort also reported less fatigue on the FACT-Fatigue subscale at treatment completion (exercise: 133.0;
non-exercise: 121.0) (p=0.6). In addition, only 30% of exercisers suffered from grade 2 dermatitis compared to 70% of
non-exercisers (p=0.2), despite a similar body mass index (26.4 exercise cohort versus 28.1 non-exercise). Exercisers also had
greater ROM in the affected (91.7 vs. 85.2%, p=0.1) and contralateral shoulder (95 vs. 90%, p=0.048) at treatment completion.
No differences in pain or sleep scores were noted and lymphedema was mild (<3cm) in the entire patient cohort.
Conclusion: The vast majority of current exercise oncology literature indicates that physical activity is an independent predictor of
quality of life metrics in cancer patients. Our study notes a trend towards improved outcomes with increased exercise during
radiation therapy, suggesting that accrual of additional patients to our pilot study is worthwhile. Ultimately, a concurrent exercise
intervention may improve quality of life and reduce acute toxicity in patients undergoing breast radiation treatment.

Title: Chemotherapy-induced fatigue and mitochondrial function in early stage breast cancer
Namrata I Peswani1, Kanchana Herath1, Brian P Dranka1, George M Lessmann1, Donna McAllister1, Raymond G Hoffmann1,
Balaraman Kalyanaraman1 and Christopher R Chitambar1. 1Medical College of Wisconsin, Milwaukee, WI.
Body: Introduction
Patients with early stage breast cancer receiving chemotherapy (CT) may experience persistent fatigue. The mechanisms for
fatigue are not well understood. We hypothesized that CT-induced fatigue may involve perturbations in mitochondrial function. We
therefore examined the development of fatigue in patients during adjuvant or neoadjuvant CT and measured mitochondrial
function in their peripheral blood mononuclear cells (PBMCs).
Methods
Females with Stage I-III breast cancer patients, ages 35-75 years, were enrolled in an IRB-approved study. Patients with
coexisting illnesses associated with chronic fatigue were excluded. Patients self-reported fatigue severity using the 14-question
Fatigue Symptom Inventory (FSI) that rates fatigue intensity on a 10-point scale. Data [FSI and PBMCs] were collected prior to
CT, mid-point in the course of CT, 2 - 3 weeks after completion of CT, and 3 and 6 months later. Mitochondrial function in PBMCs
from patients was measured using a Seahorse Bioscience XF24 analyzer at corresponding times.
Results
Results on CT-induced fatigue are available for 67 patients. The overall fatigue score for each patient was measured as the sum
of the scores for all 14 questions in the FSI. The average fatigue score for the 67 patients prior to chemotherapy was 20. Baseline
fatigue scores for patients who had adjuvant CT were much greater than for patients receiving neoadjuvant CT (R2=1). Patient
fatigue scores doubled after starting chemotherapy (average score 40, p < 0.001). Even though fatigue scores improved after
treatment completion (average score 26 at 6 months), the scores did not return to baseline 6 months later (p=0.02).
In our population, CT-induced fatigue did not correlate with patient age regardless of the CT regimen [r=0.3 and r= -.2, for
doxorubicin plus cyclophosphamide (AC) vs docetaxel plus cyclophosphamide (TC) groups, respectively)] or a decrease in
hemoglobin (r= -.3 and r=0.03, for AC and TC, respectively). None of these correlations could explain more than 9% of the
change in treatment-induced fatigue (p=0.2).
Preliminary analysis of mitochondrial function in 12 patients shows that 9 patients had a decrease in mitochondrial reserve
capacity with an increase in CT-induced fatigue following completion of 4 cycles of CT. 3 patients had a decrease in mitochondrial
reserve capacity but reported no significant change in their fatigue scores. Mitochondrial function data on all patients have been
collected and are presently being analyzed.
Conclusion
Our study shows that early stage breast cancer patients treated with adjuvant or neoadjuvant CT may experience fatigue that can
persist for months after completion of treatment and cannot be explained by the CT regimen, age, or anemia. Our initial results
suggest a link between CT-induced fatigue and changes in mitochondrial function in some patients. Mitochondrial function
analysis and its correlation with fatigue for all 67 patients will be completed and reported at SABCS. Mitochondrial function may
prove to be a biomarker for CT-induced fatigue in certain patients and may help identify patients for whom interventions that
stimulate mitochondria biogenesis, including pharmacologic agents and exercise, may be beneficial in ameliorating this
side-effect.

Title: Psycho-oncological intervention in breast cancer patients - A quantitative analysis of tumor associated fatigue treatment
Christian Eichler1,2, Pia Multhaupt2, Sibylle Multhaupt2, Friedrich Wolff1 and Mathias Warm2,4. 1Holweide Hospital, Cologne,
Germany; 2Brustzentrum, Krankenhaus Holweide, Cologne, Germany and 3University of Witten/Herdecke, Witten, Germany.
Body: Abstract
Background: Although tumor associated fatigue (TAF) or cancer related fatigue (CRF) is not a new concept, no real headway has
been made in the quantitative analysis of its successful treatment. Since 20 to 30% of all breast cancer patients suffer from
anxiety and/or depression within the first year of their diagnosis, this issue needs to be addressed and a standard treatment
protocol has to be developed. Multimodal approaches are currently being used including increased physical activity,
pharmaceutical therapy, and psycho-oncological intervention. This study focused on developing a simple, reproducible protocol
for the psycho-oncological support of tumor associated fatigue patients.
Methods: Between the year 2011 and 2012, 23 breast cancer patients fulfilled the diagnosis TAF requirements and were
introduced into this study. All patients had received surgery and were currently being monitored in an adjuvant setting. Our
method focused on a psycho-oncological support group using a predetermined, highly structured and reproducible treatment
manual. Eight weekly, 90 minute sessions were conducted and patients were evaluated before and after this eight session block.
Tumor fatigue specific questionnaires such as the multidimensional fatigue inventory (MFI) as well as the hospital anxiety and
depression scale (HADS) were used in order to quantitatively evaluate patient TAF.
Results: Of the 23 patients enrolled in the study, only 7 patients fulfilled the TAF diagnostic criteria after the psycho-oncological
group treatment. This represents a 70% reduction in diagnosable tumor associated fatigue. The HADS analysis showed a 33%
reduction in patient anxiety as well as a 57% reduction in patient depression levels. The MFI scores showed a significant
reduction in 4 of the 5 evaluate categories. With the exception of the "mental fatigue" MFI category all results were statistically
significant.
Conclusion: This study showed that a highly structured, psycho-oncological group intervention will produce significant
improvements in breast cancer patient tumor associated fatigue levels after only 8 sessions. This type of group therapy should be
recommended as a standard of care for all TAF breast cancer patients.

Title: Examining diagnostic and therapeutic delays using patient reported outcomes
Odicie Fielder-Kimbrough1, Mayra Serrano1 and Kimlin T Ashing1. 1City of Hope, Duarte, CA.
Body: Background: Diagnostic and treatment delays in cancer patients lead to poorer outcomes. African American and Latina
women with breast cancer (BC) are more likely to experience delays. This study explores delays in African American and Latina
breast cancer survivors (BCS) and association with physician-patient communication variables.
Methods: A total of 320 BCS (88 African-American and 232 Latina-American)diagnosed with stage 0-3 BC were recruited from
the California Cancer Registry and Hospitals. BCS completed a questionnaire assessing patient reported time to diagnosis and
initiating treatment, and quality of care variables pertaining to physician-patient interaction.
Results: The mean number of days waited after finding a lump or abnormality and seeking diagnostic care to obtaining a cancer
diagnosis was 57.9 days in this cohort. Bivariate analysis revealed that Latina BCS were significantly more likely to wait longer to
obtain diagnosis. ( p = .033). Once diagnosed, African-American BCS were more likely to report greater days delay to initiating
treatment (p = .007). The mean number of days delayed for initiating treatment between African American and Latinas, however,
was not significantly different, 35.3 v. 34.5 days (p=.107). We investigated the surgeon-patient dyad in influencing delays. The
better patients rated the relationship, the fewer days in delayed treatment reported (p=0.038). Additionally, 11% of BCS (n=34)
did not believe that their doctors gave them enough information regarding their treatment. The regression analysis showed that
the patients evaluation of the adequacy of doctors information was a significant predictor of days delayed treatment (p<0.0001).
If the patient believed that enough information was provided, then the treatment was initiated within 6 days compared to the 40
days average reported by patients who endorse inadequate information.
However, the majority of patients (87%) reported a "good" or "very good" relationship with her surgeon. Although the majority of
patients reported a "good" or "very good" relationship with her Oncologist (65%), the relationship was rated as "poor" or "very
poor" by 23% of patients. There was no statistical difference in rating of relationship with providers between the ethnic groups
(p=.865).
Conclusions: Our results suggest that poor patient-physician interactions are associated with BC diagnostic and therapeutic care
delays in both African American and Latina BCS. As healthcare shifts towards improving variables of quality of care and patient
satisfaction, attention to improving these interactions and further exploration will be warranted. Further, improving timely access to
diagnostic and therapeutic care may reduce BC disparity and improve cancer equity.

Title: Socioeconomic factors and the use of complimentary and alternative therapies in patients with early stage breast cancer
Swetha Panati1, Kamran Shahid1, Kaylin S Watson1, Lauren Adair1, Sanjay Juneja1, Kimberly Nguyen1, Runhua Shi1 and Gary V
Burton1. 1LSU Health Sciences Centre- FWCC, Shreveport, LA.
Body: BACKGROUND: Treatments not considered a part of conventional cancer care are known as Complementary and
Alternative Medicine (CAM) and are becoming increasingly popular. These CAM therapies, divided into Alternative Medical
Systems (AMS), Mind-Body Interventions (MBI), Biological Based Therapies (BBT), Manipulative Therapies (MT) and Energy
Therapies (ET), may or may not benefit patients (pts). Socioeconomic factors which may be associated with CAM use have not
been well defined.
AIM: To further define socioeconomic factors associated with the use of specific CAMs in pts with early stage breast cancer.
METHODS: 513 early stage breast cancer pts were interviewed between 4/2012 and 6/2014 using an IRB approved survey. The
pts were interviewed after completion of all adjuvant chemotherapy, radiation and at least 6 months of adjuvant hormone therapy.
Data collected included pt demographics, age, race, employment, insurance, marital status, income, education, religion and
residence. All pts had literacy testing using a validated reading test. Pts were questioned on the use of 36 specific CAMs within
the 5 CAM divisions, why they chose to use CAMs, and their opinion relative to benefits. Chi square test was used to evaluate the
data.
RESULTS: CAM use was common with 100% of pts using or participating in at least 1 CAM class. CAMS within the MBI division
were used by most patients with statistically increased use seen in non-protestant religion and employed pts (p=0.02 and 0.04).
BBT division use was associated with age (p=0.001) and marginal for being insured (p=0.06). AMS use appeared to be more
common in unemployed pts but was not statistically significant (p= 0.07).
The most common specific CAMs used were prayer (95%), exercise (65%), deep breathing (39%), and music therapy (38%).
Statistically significant socioeconomic associations with specific CAM subclass utilizations included: Advancing age was
associated with herbal supplement use. White pts used PET, music and art therapy as compared to black pts who used dance
therapy and progressive relaxation. Insured pts used herbal, yoga and pilates. Pts with high income participated in yoga while low
income used deep breathing, progressive relaxation, music and dance therapy. Pts with higher reading level (education) used
deep breathing, music, dance and pet therapy.
CONCLUSION: All pts, with early stage breast cancer, utilize CAMs and see their use as an important part of their cancer
therapy. Specific CAM subgroups use was associated with advancing age, employment, income, race and reading levels.
Although all pts use CAM therapy, the vast majority of CAMs pose no risk and could benefit the individual pt. Utilization of various
herbal supplements, which could pose a risk, is seen in all socioeconomic groups, however, a statistically significant increased
use was seen with advancing age and having insurance. This study, however, did not obtain information from the 2% of pts seen
at our institution who abandoned all conventional therapy and were lost to follow-up.

Title: Breast cancer in young women: Fertility preservation as a component of treatment planning and discussion
Hugo Nunes1, Ftima Vaz1, Margarida Brito1, Cludia Melo2, Ana Teresa Almeida Santos2 and Antnio Moreira1. 1Servio de
Oncologia Mdica do Instituto Portugus de Oncologia de Lisboa Francisco Gentil, Lisbon, Portugal and 2Servio de Reproduo
Humana do Centro Hospitalar da Universidade de Coimbra.
Body: Background: as part of education and informed consent before cancer therapy, health care providers should address the
possibility of infertility with patients (pts) treated during their reproductive years, and be prepared to discuss fertility preservation
options and/or to refer all potential patients to Reproductive Specialists. I here we describe the recently established referral
protocol of our institution with a fertility preservation centre. Both institutions belong to the Portuguese National Health System.
Methods: the referral protocol includes female pts, 39 years old, with a diagnosis of breast or gynecological cancer or sarcoma
with indication to curative intention chemotherapy. The possibility of fertility preservation is primarily discussed at the cancer
centre and only interested pts are referred to the reproductive clinic, where specialized counseling is provided. Fertility
preservation procedures (FPPs) (oocyte, embryo and/or ovarian tissue cryopreservation) are conditional to a previous evaluation
of the ovarian pool (OP). OP markers are: antral follicle count, follicle stimulating hormone (FSH) and anti-mullerian hormone
(AMH). Only good prognosis cases are eligible for FPPs.
Results: After primary discussion with eligible breast cancer (BC) pts, 17 were referred to the reproductive clinic: median age at
BC diagnosis was 34 yrs (29-38), median previous pregnancies number of 0 (0-2) and median number of children of 0 (0-2).
Marital status: most (64%) pts had a companion but 36% did not. All pts except 1 were candidates for immediate chemotherapy,
either in the adjuvant or neoadjuvant setting. In the adjuvant setting (9 pts), median time from surgery until starting systemic
treatment was 61.5 (43-146) days (aprox. 8,8 weeks); the case with a 146 day interval before chemotherapy was due to surgical
complications and not to FPP. In the neoadjuvant setting (5 pts), median time between multidisciplinary decision and systemic
treatment beginning was 17 (11-24) days (2,4 weeks). One patient was proposed to hormone therapy only after BC surgery and
underwent FPP before systemic treatment. The timing for primary discussion of FPP was in multidisciplinary decision
consultation. After counselling at the reproductive clinic, 2 pts declined FPP (%). For the other 13 pts oocyte cryopreservation was
the FPP mostly used, although in one case embryo cryopreservation was performed. One case of ovarian hyperstimulation
syndrome was observed but this FPP complication resolved without sequela and this event did not cause delay in chemotherapy
treatment.
Conclusions: our collaboration protocol allows for an efficacious referral of BC pts seeking fertility preservation counselling.
Although BC patients may be focused initially on their cancer diagnosis, health care providers are encouraged to advise them
regarding potential threats to fertility. Formal collaboration between cancer and reproductive centres, like the one described here,
are crucial so as to allow for the widest array of options for fertility preservation and to prevent delay in cancer treatment. We
intend to follow up these patients in order to realistically understand the impact of this practice in fertility and quality of life of
cancer patients.

Title: Knowledge and Information seeking about personalized breast cancer therapy
Deevakar Rogith1, Rafeek A Yusuf1, Shelley R Hovick2, Bryan M Fellman3, Susan K Peterson3, Allison Burton-Chase4, Yisheng
Li3, Elmer V Bernstam1 and Funda Meric-Bernstam3. 1University of Texas School of BioMedical Informatics, Houston, TX; 2Ohio
State University, Columbus, OH; 3University of Texas MD Anderson Cancer Center, Houston, TX and 4Albany College of
Pharmacy, Albany, NY.
Body: INTRODUCTION: Breast cancer patients and providers are increasingly interested in personalized cancer therapy.
Information-seeking behaviors and knowledge about personalized cancer therapy, cancer genetics, and molecular testing may
influence patients participation in clinical trials and decision making regarding their care. We evaluated breast cancer patients
knowledge and information seeking behaviors regarding personalized cancer therapy (PCT).
METHODS: The study population included newly registered female breast cancer patients at The University of Texas MD
Anderson Cancer Center prior to their first clinical visit. Of 308 consecutive patients who were invited to participate, 100 (32%)
completed a self-administered questionnaire assessing their knowledge and information seeking preferences regarding PCT.
Knowledge regarding cancer genetics and PCT research was assessed using 16 true/false questions (Cronbachs =0.88). A
knowledge score was computed from the total number of correct responses.
RESULTS: Respondents were predominantly white (70%), older (median age 55 years; SD=12.9; range 26-84), educated (78%
with college degree or higher) and higher incomes (54% >$50,000/year); 71% had been diagnosed with breast cancer for at most
one year at time of participation. Knowledge regarding cancer genetics and PCT was moderate (M=8.68, SD=3.8). Although most
participants (85%) could correctly identify the definition of PCT, many (59%) did not know that somatic mutations are not
hereditary. Many (75%) knew that molecular testing can reveal risk for other hereditary cancers. Less than half (46.5%) knew
about the availability of PCT in clinical trials. A minority (27%) indicated that they had sought information regarding PCT. They
sought for information related to specific treatment options. Higher education (p<0.01) and income levels (p<0.05) were
associated with higher knowledge scores and with seeking PCT information (p<0.01). Those who had previously undergone any
genetic testing also were more likely to seek information about PCT (29.6% vs 9.9%, p<0.05). Other demographic and clinical
variables like age, race, duration of illness, cancer stage did not correlate with the knowledge score or information seeking
behavior.
CONCLUSION: Study participants could define PCT, but had limited knowledge of its availability and underlying treatment
principles. This may be due, in part, to the fact that few participants had sought information about PCT. Understanding patients
knowledge and prior information seeking regarding PCT may inform clinicians, who are likely to be patients initial source of
information about PCT.

Title: The effect of low intensity exercise for mental status and quality of life of Japanese early breast cancer patients
Makiko Kamio1, Kazumi Kawase1, Hiroko Nogi1, Ken Uchida1 and Hiroshi Takeyama1. 1Jikei University School of medicine, Tokyo,
Japan.
Body: Objective: Breast cancer is one of the most common cancers, and it accounts for 20% of malignant tumors of women.
Psychological damage of the breast cancer patient is serious. In our institution, approximately 30% of the early breast cancer
patients diagnosed as depressed mental status before operation. Breast cancer patients have to receive hormone therapy,
chemotherapy, and radiation therapy for a long term after operation. Depressed mental status affects execution of the treatment,
and the interruption of the treatment brings the risk of cancer recurrence. As a treatment for depressed mental status of cancer
patients, medication and psychotherapy are common, but some previous reports showed exercise is effective. Exercises provided
in these reports were various kinds and strength, and the standard prescription is not established. We investigate the effect of
walking program as a mild aerobic exercise, whether it improves the depressed mental status and quality of life (QOL) of breast
cancer patients.
Methods: 25 early breast cancer patients were recruited. Depressed mental status and QOL were assessed by Center for
Epidemiologic Studies Depression Scale (CES-D) and Medical Outcomes Study 36-item Short Form Health Survey (SF-36).
Physical activity was measured by accelerometer (Lifecorder PLUS, Suzuken), and estimated using metabolic equivalent foe task
(MET). Participants mounted accelerometer after discharge, and baseline physical activity was recorded till 1 month after
operation. After recording baseline physical activity, we instructed to perform walking or mild aerobic exercise. Scores of CES-D
and SF-36, and physical activity data were measured again after two months.
Results: Physical activity after intervention was significantly increased as compared to baseline (8.55.5 vs 12.36.9
MET.h/week, p=0.02 ). At pre-operation phase, nine patients (36%) regarded as depressed mental status by CES-D, and
depressed patients decreased to five at the time of 3 months after operation. A subscale of SF-36 ( mental health ) reduced
significantly compared with pre-operation and 3 months after operation ( p<0.01 ). Physical activity after intervention was
correlated with the scores of CES-D and all subscales of SF-36.
Conclusion: This study demonstrated the possibility that walking program as a mild exercise improves mental status and QOL of
perioperative breast cancer patients. It is necessary to examine in more detail for clinical application.

Title: Perception of women after undergoing microbiopsy following a positive breast cander screening
Birgit Carly1, Mireille Aimont1, Nicolas Beauloye1 and Fabienne Liebens1. 1CHU St Pierre, Brussels, Belgium.
Body: Background: Some authors have challenged the benefit of breast cancer mammographic screening due to high prevalence
of overdiagnosis and overtreatment.
Purpose: To assess the anxiety and stress of women undergoing a microbiopsy following a false positive breast cancer screening
and to evaluate whether this experience refrains them from continuing breast cancer screening.
Methods: 296 patients underwent breast microbiopsy for positive breast cancer screening B1, B2 and B3 lesions in our Breast
Center between 2011 and 2013. Patients undergoing microbiopsy for B4 and B5 lesions were excluded from the study. The
patients were thereafter interviewed by our two psychologists by phone within a time range of 6 to 18 months following the biopsy.
Results: 201 women answered the questionnaire (68% response). 5.5% had a breast lesion at risk and 75% had a benign lesion.
Half of the women rated their anxiety as high (7-8 on a scalebetween 0 10) and 15% as very high (9-10/10) before the biopsy.
While waiting for the result, 42% rated their anxiety as high (7-8/10) and nearly a third as very high (9-10/10). But 80% accepted
to continue breast cancer screening and 90% would accept to undergo again a microbiopsy if necessary.
Conclusion: microbiopsy provides high stress in most patients undergoing it. Nevertheless, 90% of those who had a false positive
results accept it and are willing to undergo a biopsy again if necessary.

Title: Exploratory post hoc analyses of patient-reported outcomes (PROs) in the IMELDA randomized phase III trial: Maintenance
bevacizumab (BEV) capecitabine (CAP) after initial first-line BEV plus docetaxel (DOC) for HER2-negative metastatic breast
cancer (mBC)
Dinesh Doval1, Saverio Cinieri2, Hakan Bozcuk3, Jean-Yves Pierga4, Kadri Altundag5, Xiaojia Wang6, Sudeep Gupta7, Guillermo
Lopez Vivanco8, Vineet Gupta9, Ewa Chmielowska10, Jose Bines11, Philippe Montcuquet12, Alfred Namour13, Emilio Alba14, Giorgio
Mustacchi15, Paulo Cortes16, Sabine de Ducla17, Ulrich Freudensprung17, Lesley Fallowfield18 and Joseph Gligorov19. 1Rajiv Gandhi
Cancer Institute & Research Center, Delhi, India; 2Medical Oncology Department and Breast Unit, Brindisi, Italy; 3Akdeniz
University Medical Faculty, Antalya, Turkey; 4Institut Curie, Universit Paris Descartes, Paris, France; 5Hacettepe University
Cancer Institute, Ankara, Turkey; 6Zhejiang Cancer Hospital, Hangzhou City, China; 7Tata Memorial Hospital/Center, Mumbai,
India; 8Hospital Universitario Cruces, Vizcaya, Spain; 9Bangalore Institute of Oncology, Bangalore, India; 10Centrum Onkologii
Prof. F Lukaszczyka, Bydgoszcz, Poland; 11Instituto Nacional de Cancer, Rio de Janeiro, Brazil; 12Clinique Saint Vincent,
Besanon, France; 13National Cancer Institute, Cairo University, Cairo, Egypt; 14Hospital University Clinic Virgen de la Victoria,
Mlaga, Spain; 15Medical Oncology, University of Trieste, Trieste, Italy; 16University Hospital of Santa Maria, Lisbon, Portugal; 17F.
Hoffmann-La Roche Ltd, Basel, Switzerland; 18Brighton & Sussex Medical School, University of Sussex, Falmer, United Kingdom
and 19APHP Tenon, IUC-UPMC, Paris, France.
Body: BACKGROUND The addition of CAP to maintenance BEV demonstrated statistically significant and clinically relevant
improvements in progression-free survival (PFS [primary endpoint]; HR 0.38 [95% CI 0.270.55]; log-rank p<0.001) and overall
survival (OS [secondary endpoint]; HR 0.43 [95% CI 0.260.69]; log-rank p<0.001) in patients (pts) without disease progression
(PD) on initial first-line BEVDOC for HER2-negative mBC in the IMELDA trial. This benefit was achieved despite the smaller
than planned sample size due to premature recruitment discontinuation because of regulatory withdrawal of BEVDOC.
METHODS Pts with HER2-negative measurable mBC, ECOG PS <2, and no prior chemotherapy for mBC were eligible. After 36
cycles of BEVDOC, pts without PD were randomized to either BEV alone or BEVCAP (BEV 15 mg/kg q3w; CAP 1000 mg/m2
bid d114 q3w) until PD. PROs (secondary endpoint) were assessed using the EORTC QLQ-C30 completed at screening (before
BEVDOC), at randomization to CAP vs no CAP, then every 3 cycles until PD, and at (but not beyond) PD. Analyses of mean
change from randomization were prespecified. A 28-day window around the scheduled timepoints from randomization was
applied to maximize the number of questionnaires available for analysis. Exploratory post hoc analyses included mixed-model
repeated measures (MMRM; modeling weighted treatment effect from randomization across all available timepoints) and
responder analyses using the global health status/QoL subscale. Pts were categorized as having improved (10-point increase),
stable (change of <10 points), or worsened (10-point decrease) scores from randomization [Osoba, 2005].
RESULTS Adherence with questionnaire completion was 6585% for all assessment timepoints during the first year of
maintenance therapy. MMRM analysis of the global health status/QoL subscale showed no difference between the treatment
arms in change from randomization (least squares mean estimate 0.40 [95% CI 6.07 to 6.87]). Similar results were observed for
other subscales, including the diarrhea symptom subscale.
No. of pts (%)
BEV (N=94)
BEVCAP (N=91)
N=51
N=59
Improved
15 (29.4)
17 (28.8)
Stable
26 (51.0)
34 (57.6)
N=29
N=57
Improved
11 (37.9)
12 (21.1)
Stable
12 (41.4)
30 (52.6)
N=23
N=43
7 (30.4)
16 (37.2)
Week
Week
Week
9a
18a
27a
Improved
Stable
12 (52.2)
20 (46.5)
N=15
N=35
Improved
4 (26.7)
14 (40.0)
Stable
9 (60.0)
17 (48.6)
Week
36a
aNo.
of patients with completed questionnaires at both randomization and the respective week. Only weeks with 10 pts in both
arms shown.
CONCLUSIONS The IMELDA sample size was smaller than planned but protocol adherence with PRO completion was relatively
high. Prespecified change from randomization and exploratory post hoc MMRM analyses of PROs suggest that the clinically
meaningful PFS and OS benefit from adding CAP to BEV is achieved while maintaining QoL, with no difference between BEV
and BEVCAP treatments. Responder analyses over time showed improved or stable global health status/QoL scores in the
majority of pts at each timepoint in both treatment arms.

Title: Stage-related risk categorization and influence of free margins on survival in triple negative early breast cancer - a
population-based study of 2037 TNBC patients with adjuvant chemotherapy
Peter Kern1,2, Gunter von Minckwitz1, Carolin Ptter1, Annika Flach1, Sofia Pavlidou1, Rainer Kimmig1 and Mahdi Rezai2. 1German
Breast Group, Neu-Isenburg, Hessen, Germany.
Body: Introduction: Triple negative breast cancer (TNBC) represents 10-20% of all breast cancer entities [1][2] and has a known
aggressive behavior and poor outcome. Patients treated in the setting of randomized clinical trials often do not represent actual
treatment characteristics in real-life scenarios. To determine the stage-related survival and effect of surgical performance in
TNBC with current multimodal treatment, we set out to analyze data of a large population-based registry of primary breast
cancers which covering >50% of all breast cancer cases in Germany.
Patients and methods: We analyzed data from a prospectively collected cancer registry of >200 certified breast units of the
West-German Breast Center (WBC) in Germany from 2009-2011. From a cohort of 39570 primary breast cancer patients treated
in this period, 12759 underwent adjuvant systemic therapy, out of which 2037 were TNBC cases with adjuvant chemotherapy.
Inclusion criteria were triple negative breast cancers (Her2-new1+/2+ (Fish negative) and estrogen receptor (ER) and
progesterone receptor (PR) <10%) and adjuvant chemotherapy, unilateral and non-metastasized breast cancer. Only those
patients were included who have been followed-up within the first 3 years. Exclusion criteria were neoadjuvant chemotherapy,
bilateral breast cancer and metastatic disease.
The use of first, second and third generation chemotherapy was analyzed as well as the effect of clear/unclear resection margins
and its impact on survival data.
Results:
2037 patients were eligible for this study. Overall survival rates were as follows: T1 a and T1b 100 %, T1c 90,7 %, T2 90,9 %, T3
68,1 % and T4 64,3 %. No statistical differences were detected in between stages T1 and T2, and also not in between T3 and T4.
Combining T1/T2 and T3/T4 and performing group-wise comparisons, differences for combined stages were highly statistically
significant (3,9 x E-09). Inflammatory TNBC was prognostically worst with a survival-rate of 33,3 % at 24-months. (p<0,001)
Unclear resection-margins versus clear margins in TNBC exerted a negative impact on DFS (87 vs. 73 %; p=0,00002) and DDFS
(p=0,0004). Age was an independent risk factor for survival with a cut-off at 35 years.(p=0,044)
Third-generation chemotherapies (anthracycline+taxanes) were associated with a significant improved overall-survival at
24-months compared to first generation chemotherapies (non-anthracycline, non-taxane) (95 % vs. 87 %; p=0,0029)
Conclusion: Standard 3rd generation (anthracycline- and taxane-containing) chemotherapy and optimal surgical performance with
clear margins is vital for patients with early, triple-negative breast cancer (TNBC). Within T1 and T2 stages, no stage-related
deterioration of prognosis was detected, however these stages were markedly different from stages T3/T4, declining from
90-100% to 64-68 %.
This analysis of a large database of a population-based study demonstrates that tumor size, margins and guideline-adapted
chemotherapy matter in triple-negative, early breast cancer.
[1] Schwentner et al. 2013
[2] Elsawaf et al. 2013.

Title: Nipple sparing mastectomy in patients with BRCA1 and BRCA2 mutations
Aidan T Manning1, Andrea Pusic1, Caitlin Wood1, Anne Eaton1, Michelle Stempel1, Deborah Capko1 and Virgilio Sacchini1.
1
Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Introduction: Nipple sparing mastectomy (NSM) is now performed with increasing frequency in both therapeutic and
prophylactic breast surgery. The role of NSM in BRCA1 and BRCA2 mutation carriers has not been well described. The aim of
this study was to review our experience with NSM in this high-risk population.
Methods: A review of the breast database was performed to identify all patients with documented BRCA mutations who
underwent NSM at Memorial Sloan Kettering Cancer Center. Data extracted from the database included patient demographics,
type of mutation, indication for surgery, type of reconstruction, and complications. For patients undergoing therapeutic
mastectomy, data on disease stage, axillary procedures, and follow-up were also extracted.
Results: 177 NSMs (88 bilateral, 1 unilateral) were performed in 89 female patients with a documented BRCA mutation between
September 2005 and December 2013. There were 56 patients with BRCA1 mutation, 26 with BRCA2 mutation, and 7 with genetic
variants of uncertain significance. 26 patients had a therapeutic NSM for breast cancer (stage 0: n=6; stage 1: n=15; stage 2:
n=5) and concurrent contralateral prophylactic mastectomy (CPM). The mean tumor size was 1.46cm (range, 0.1-3.5cm), and all
were node negative following sentinel lymph node biopsy. 63 patients had NSM for prophylaxis. The mean age of patients
undergoing therapeutic NSM was 41 years (range, 26-59) and prophylactic NSM was 39 years (25-59). There was an incidental
diagnosis of ductal carcinoma in situ (DCIS) in 4 women undergoing CPM and 4 patients undergoing prophylactic NSM, including
1 patient diagnosed with bilateral DCIS, and an incidental diagnosis of atypia in 8 patients undergoing prophylactic NSM. In 26
patients undergoing therapeutic NSM, at median follow-up of 2.34 years (range, 0.45-6.06) there were no local or regional
recurrences. One patient developed distant metastases and subsequently died from her disease, and 1 other patient died from
metastatic ovarian cancer. In 63 patients undergoing prophylactic NSM, at median follow-up of 2.15 years (range, 0.11-8.14)
there were no newly diagnosed breast cancers or deaths. Following NSM, 5 patients (5.6%) required subsequent excision of the
nipple-areolar complex (3 cases for close or positive DCIS on final histology, 1 case for infection with necrosis, and 1 case for
ongoing nipple discharge). All 89 patients had immediate breast reconstruction (tissue expander: n=80; permanent implant: n=8;
autologous (DIEP) flap: n=1). Postoperative complications are shown in Table 1.
Postoperative complications following 177 nipple sparing mastectomies performed in 89 patients with BRCA mutations
No. of Breasts; n (%)
No. of Patients; n (%)
Skin desquamation
68 (38.4)
40 (44.9)
Necrosis requiring debridement
18 (10.2)
13 (14.6)
Infection
7 (4.0)
7 (7.9)
Hematoma
3 (1.69)
3 (3.4)
Complication requiring implant or tissue expander removal
6 (3.4)
6 (6.7)
Conclusion: NSM is an acceptable choice for patients with BRCA gene mutations undergoing therapeutic or prophylactic
mastectomy with no evidence of compromise to oncological safety. This report shows an acceptable complication rate, and
patients rarely required subsequent excision of the nipple-areolar complex.

Title: Variation in the use of mastectomy (MAST) in women with small node negative breast cancer (BC) treated at US academic
institutions
Ines Vaz Luis1,2, Melissa E Hughes1, Angel Cronin1, Hope S Rugo3, Stephen B Edge4, Beverly Moy5, Richard Theriault6, Michael J
Hassett1, Eric P Winer1 and Nancy U Lin1. 1Dana-Farber Cancer Institute, Boston, MA; 2Intituto de Medicina Molecular, Lisboa,
NA, Portugal; 3University of California, San Francisco, CA; 4Baptist Cancer Center, Memphis, TN; 5Massachusetts General
Hospital, Boston, MA and 6University of Texas MD Anderson Cancer Center, Houston, TX.
Body: Background: More than two decades ago several trials have shown equivalent survival between breast conserving
surgery (BCS) and MAST. Among a contemporary cohort of patients (pts) with Stage I BC who would be expected to be
candidates for BCS; we examined the initial choice of surgery and factors associated with it.
Pts and methods: Prospective cohort study including pts with clinical Stage I BC treated at a National Comprehensive Cancer
Network center that participated in the BC outcomes database from 2000-09. Descriptive analyses were performed examining the
proportion of pts who initially underwent MAST vs BCS. Factors associated with initial surgery were analyzed using multivariable
logistic regression.
Results: Of 10,249 pts with clinical Stage I BC, 2,361(23%) underwent MAST as the initial surgery and 7,888 (77%) BCS. Of
those, 8% were ultimately converted to MAST. The median time from diagnosis to initial surgery was longer among the MAST
group (4 vs. 6 weeks).
Patient, tumor, care and institutional factors were associated with higher rates of initial MAST: 30% of pts with <50 years of age
had a MAST vs. 17% of those 70; 41% of pts with body mass index (BMI) < 18.5 kg/m2 (underweight) had a MAST vs. 20% of
those with a BMI >30 kg/m2 (obese). There was significant institutional variation, with rates of initial MAST ranging from 14-30%.
The use of preoperative magnetic resonance imaging (MRI) was associated with a higher rate of initial MAST (32% vs.22%).
Differences by tumor subtype were observed, 38% of pts with HER2+/HR- tumors had initial MAST vs. 22-28% among other
subtypes. In the multivariate model, age, BMI, comorbidity, income, center, stage, tumor subtype, grade, histology and
preoperative MRI were associated with the choice of initial surgery.
Multivariate logistic model to investigate factors associated with initial MAST
MAST vs BCS
OR
95%CI
Age
<0.01
<50
50-59
0.7
0.6 - 0.8
60-69
0.7
0.6 - 0.8
70+
0.6
0.5 - 0.7
BMI (kg/m2)
<0.01
<18.5
1.8
18.5-<25
25-<30
0.8
0.7 - 0.9
30
0.7
0.6 - 0.8
Unknown
0.8
0.6 - 1.0
1.3 - 2.6
Comorbidity
0.05
1+
1.1
Race
1.0 - 1.3
0.8
Non Hispanic white
Non Hispanic black
0.8-1.2
Hispanic
0.8-1.3
Other
1.1
0.9-1.4
Insurance
0.16
Manged care/ Indemnity
Medicare
0.9-1.2
Self-pay/Medicaid
1.3
1-1.6
Other
0.6-1.6
Median Household income
0.02
1 (low)
0.9
0.8-1.1
0.9
0.8-1.1
0.9
0.8-1.0
5 (high)
0.8
0.6-0.9
1-1
Year of diagnosis
Center
0.35
<0.01
0.9
0.7-1.1
0.7
0.6-0.9
1.4
1.1-1.7
0.4
0.3-0.6
0.8
0.6-1.1
1.3
1-1.6
0.6
0.4-0.8
Clinical Stage
<0.01
T1a
T1b
0.8
0.6-0.9
T1c
0.8-1.2
T1NOS
1.1
0.9-1.4
Tumor Subtype
<0.01
HER2+HR+
HER2+HR-
1.5
1.1-1.9
HER2-HR+
0.9
0.7-1.0
HER-HR-
0.7
0.6-0.9
Grade
<0.01
Low Intermediate
High
1.2
1.1-1.4
Histology
Ductal
<0.01
Lobular
1.4
1.2-1.7
Mixed
1.3
1.1-1.5
Other
0.7
0.5-0.8
Preoperative MRI
No
Yes
1.8
<0.01
1.6-2.1
Conclusions: Among a cohort of pts with small node negative BC, 23% elected to have MAST with significant variation
associated with choice of treatment, while some of this variation is likely appropriate and clinically indicated, further studies to
assess pt understanding of the tradeoffs between BCS and MAST is warranted. These findings need to be considered in light of
the increasing number of pts who are choosing MAST/bilateral MAST.

Title: The role of preoperative MRI in negative margins after breast conserving surgery in patients with invasive breast cancer or
purely DCIS
Elvira L Vos1, Adri C Voogd2, Cornelis Verhoef1, Inge-Marie Obdeijn1 and Linetta B Koppert1. 1Erasmus MC Cancer Institute,
Rotterdam, Netherlands and 2Eindhoven Cancer Registry, Eindhoven, Netherlands.
Body: Introduction Clinical benefit of preoperative MRI in invasive breast cancer (IBC) patients and patients with purely ductal
carcinoma-in-situ (DCIS) remains controversial. We aimed to study the role of preoperative MRI in negative margins after breast
conserving surgery (BCS) and reexcision rate in a large population-based cohort.
Methods Retrospective analyses were performed in women diagnosed with IBC and purely DCIS and surgically treated between
2011-2013 extracted from the Eindhoven Cancer Registry. Patients were excluded in case of: neo-adjuvant systemic therapy,
(clinical or pathological) tumor stadium T4, distant metastasis, and unknown resection margin status. According to preoperative
MRI use, the study population was divided into a non-MRI and MRI group. All multivariable analyses were adjusted for baseline
differences between non-MRI and MRI group with P<0.1. No information was available on the exact reason to perform MRI
preoperatively in these patients.
Results A total of 3,116 patients were eligible of which 2,238 (71.8%) patients were treated by breast conserving surgery (BCS)
and included for analyses. Preoperative MRI was performed in 592 (30.6%) IBC patients and 55 (18.3%) DCIS patients. In IBC
patients, differences in non-MRI and MRI group were: median age (62 vs 58), histology (lobular type in 5.7% vs 24.5%), median
tumor size (13 mm vs 15 mm), her2neu receptor status and specific tumor locations. Median (interquartile range) time between
diagnosis and surgery in the non-MRI and MRI group were 21 (16-28) and 31 (22-40) days respectively (univariable and
multivariable P<0.001). Negative margin was attained in 1,135 (84.4%) and 489 (82.6%) patients respectively (OR 0.88 95%CI
0.68-1.14 P=0.326). After adjustment for baseline differences and for factors associated with negative margin with P<0.1 (i.e.
histology, tumor size, differentiation grade, progesterone receptor %, her2neu receptor status, regional lymph node stadium, and
tumor location) MRI use was not associated with negative margin (OR 1.09 95%CI 0.81-1.45 P=0.587). Reexcision was
performed in 96 (7.1%) and 62 (10.5%) patients (P=0.013). In case reexcision was needed, conversion to mastectomy occurred
in 32 (2.4%) vs 20 (3.4%) patients (P=0.210). In patients with purely DCIS, only median age differed in non-MRI and MRI group
(61 vs 57). Median (interquartile range) time between diagnosis and surgery in the non-MRI and MRI group were 22 (16-31) and
35 (23-49) days respectively (univariable P<0.001 and multivariable P=0.024). Negative margin was attained in 197 (80.1%) and
42 (76.4%) patients respectively (OR 0.80 95%CI 0.40-1.61 P=0.538). After adjustment for baseline differences and for factors
associated with negative margin with P<0.1 (i.e. differentiation grade and tumor location) MRI use was not associated with
negative margin (OR 1.51 95%CI 0.72-3.16 P=0.280). Reexcision was performed in 39 (15.9) and 10 (18.2) (P=0.672). In case
reexcision was needed, conversion to mastectomy occurred in 12 (4.9%) vs 3 (5.5%) patients (P=0.859).
Conclusion In both IBC and DCIS patients, preoperative MRI delayed time between diagnosis and surgery, but was not
associated with a higher percentage of negative margins after BCS.

Title: Radioguided occult lesion localization (ROLL). A 12-year single institution experience of 942 ROLL excisions
Simon C Hawkins1, Iain M Brown1, Kirsten Stepp1, Sue Widdison1, Donna Christensen1, Mohsen El-Gammal1, Alistair Paterson1
and Philip Drew1. 1Royal Cornwall Hospitals NHS Trust, Truro, Cornwall, United Kingdom.
Body: Introduction
Despite widespread adoption for the localization of impalpable breast lesions the wire guided localization (WGL) technique has a
range of disadvantages including high re-operation rates. Radio-guided occult lesion localization (ROLL) has been proposed as
an alternative technique, which may have superior clinical outcomes and be more flexible in its clinical application. A retrospective
study of outcomes following ROLL therapeutic wide local excisions (WLE) and diagnostic excision biopsies (DEB) was performed.
This documented 12 years experience of the ROLL technique at a single institution to provide a comprehensive analysis of the
largest UK based series of ROLL excisions and sentinel node occult lesion localisations (SNOLL).
Methods
942 patients were identified who had been referred with non-palpable breast lesions from 2000-2012. 576 patients underwent
WLE following a biopsy proven diagnosis of breast cancer. 366 patients underwent DEB. Prospective data collection was
performed using bespoke proformas by the operating clinician. These data were supplemented with a retrospective analysis of
prospectively collected data held on patient electronic records including evidence of radiological excision on specimen
radiographs and histopathological characteristics including excision margin status. Sub-group analysis was performed to examine
the outcomes from "same day" and "next day" protocols for ROLL/SNOLL procedures and to examine the effect of residual
radioactivity levels in the excision bed on margin status.
Results
99.0% of ROLL WLE returned histological diagnoses of invasive cancer or DCIS and 98.4% of radiological abnormalities were
identified on post-excision specimen radiographs. 97.5% of radiological abnormalities were identified on post-excision
radiographs following DEB. Surgical localisation was rated as easy in 92.3% of excisions and 29 adverse events were recorded
as a result of the localisation procedure (3.1%). Complete histological excision was recorded in 77.8% of the WLE sub-group
based on defined margin criteria with a median specimen weight of 50g. 31.7% of DEB were pathologically upgraded to a
diagnosis of breast cancer. No significant difference was seen in complete excisions rates between "next day" (76.4%, n=250)
and "same day" (78.8%, n=326) ROLL protocols (p= NS). No significant difference was seen when levels of residual excision bed
radioactivity and complete excision rates were compared (p=NS). Sentinel Lymph Node (SLN) localisation was successful in
97.6% of cases (n=205) with an SLN positivity rate of 12.7%. No significant difference was identified between SLN localisation
rates for "next day" vs. "same day" SNOLL protocols (p=NS). The presence of microcalcification as the radiological abnormality
(p=0.0005), underestimation of lesion size on pre-operative imaging (p=0.0005) and symptomatic referral (p=0.001) were factors
that predisposed to incomplete excision margin status.
Conclusions
ROLL/SNOLL can be safely and effectively used to accurately localize impalpable breast lesions in agreement with current
evidence. In addition to highly accurate localization, ROLL also has technical and logistical advantages that may make it more
acceptable than WGL for clinicians and patients.

Title: Flat epithelial atypia on core biopsy and upgrade to cancer: A systematic review and meta-analysis
Anatoliy V Rudin1, Hoskin L Tanya1, Ann M Farrell1 and Amy C Degnim1. 1Mayo Clinic.
Body: BACKGROUND: Flat epithelial atypia (FEA) is a recently described breast lesion that may coexist with cancer or atypical
ductal hyperplasia (ADH). Currently, there is no consensus on whether to surgically excise FEA diagnosed by core needle biopsy.
Our aim was to perform a systematic review and meta-analysis of pertinent studies to determine the frequency of upgrade to
cancer or ADH at surgical excision of "pure" FEA (no other high risk lesion on core biopsy).
METHODS: A retrospective search was performed using MEDLINE, Embase, Scopus and Web of Science databases from
1/2003- 4/2014 to capture studies on core biopsy diagnosed FEA followed by surgical excision. Search terms were: FEA, flat
epithelial atypia, DIN1A, columnar cell, breast diseases, core needle biopsy. Inclusion criteria were: 1) manuscript with original
data on FEA diagnosed on core needle biopsy, 2) data included outcome of cancer at surgical excision, 3) English Language.
RESULTS: Of 224 articles, 30 met inclusion criteria.
Study
Total Pure FEA cases
N excised
N upgraded to cancer
% upgraded
Lim, 2006
62.5
20
Kunju 2007
14
85.7
12
25
Martel 2007
63
38.1
24
38
Piubello, 2009
33
60.6
20
Chivukula, 2009
39
89.7
35
14
Senneta,2009
41
87.8
36
Hayes, 2009
100
13
Davashian, 2009
12
100
12
17
Tamasino, 2009
54
44.4
24
13
Noske, 2010
43
69.8
30
Lee, 2010
15
46.7
14
Ingegnoli, 2010
18
83.3
15
20
Noel, 2010
62
32.3
20
Flegg, 2010
NA
NA
40
Sohn, 2011
36
66.7
24
Lavoue, 2011
60
100
60
13
Rakha, 2011
24
100
24
21
Soloranzo,2011
33
84.8
28
14
Verschuur-Maes, 2011
69
34.8
24
38
Peres, 2012
128
79.7
102
10
10
Uzoaru, 2012
145
65.6
95
Bianchi, 2012
190
100
190
18
Biggar, 2012
51
100
51
Yamaguchi, 2012
17
47.1
13
Polom, 2012
20
100
20
10
Khoumais, 2013
104
90.4
94
10
11
Villa, 2013
142
85.2
121
Becker, 2013
303
78.9
239
10
Ceugnart, 2013
63
82.5
52
Uzan, 2013
35
100
35
Across 1420 patients with pure FEA in 30 studies, observed proportions with upgrade to cancer varied from 0-40% with a pooled
estimate of 11% (95% CI: 8-15%) using a random effects model. Test for heterogeneity was statistically significant (p < 0.0001, I2
statistic=58%). After excluding 7 studies with <50% (or unreported %) of all FEA cases excised, heterogeneity decreased
(p=0.07, I2=32%) and the pooled estimate of cancer upgrade was 9% (95% CI: 7-12%). For upgrade to ADH, 693 subjects with
pure FEA were analyzed from 21 studies. The percent upgraded to ADH ranged from 0-60% with significant heterogeneity (p <
0.0001, I2 = 64%). Excluding 4 studies with <50% of FEA cases excised did not improve heterogeneity (p < 0.0001 and I2 =
66%). The random effects model pooled estimate of upgrade to ADH was 16% (95% CI: 11-23%) for the 21 studies and 17%
(95% CI: 12-25%) for the subset of 17 studies.
CONCLUSION: The pooled upgrade rate of FEA to cancer was 9-11% and upgrade to ADH was 16-17%. For patients with FEA
on core needle biopsy, surgical excision should be strongly considered.

Title: Prophylactic mastectomy in young women with breast cancer
Cristiane Metran Nascente1, Frances Wright1, Christel Helwig1, Nim Li1, Rodica Mandel1, Alex Kiss2 and Ellen Warner1.
1
Sunnybrook Health Sciences Centre, Toronto, ON, Canada and 2ICES, Toronto, ON, Canada.
Body: Background: In recent years women with unilateral breast cancer (BC) have been increasingly requesting contralateral or
bilateral prophylactic mastectomy (PM) despite only requiring unilateral mastectomy (M) or lumpectomy (L) respectively. This
trend has been observed even in women at low risk for contralateral BC, particularly among younger women.
Objective: Determine the rates of unexplained PM (UPM) and factors associated with this choice, among a prospective cohort of
newly diagnosed young women with BC at our centre.
Methods: We reviewed the records of all patients diagnosed with BC at age 40 years from Feb 08 to Sept 13 at Sunnybrook
who consented to have their clinical data entered in a prospective research database. Psychosocial data - HADS [Hospital
Anxiety and Depression Scale] and IES [Impact of Event Scale] - were prospectively collected from all consenting women at
baseline and at completion of active treatment. Among the 149 patients treated with curative intent, 43 received unilateral M, 59 L
and 47 contralateral or bilateral PM. These groups were compared with respect to demographics, risk factors, use of
pre-operative MRI, pathology, treatment and surgeon gender. Differences in variables between the groups were analyzed using
analysis of variance for continuous variables and chi-square test for categorical variables.
Results: The mean age of all patients was 35 years old (21-40 range). Among the 47 patients who had PM, the first surgery was L
for 15 (32%), unilateral M for 27 (57%) and bilateral M for 5 (11%). Twenty-four (51%) fulfilled high risk criteria (19 BRCA
mutation, 4 strong family history, 1 chest radiation) and 2 (4%) had suspicious findings in the contralateral breast. The other
twenty-one patients (45%) underwent UPM for "peace of mind" alone. Thirty-three (33%) of UPM patients had T3-4 tumors. There
were no statistically significant differences between the UPM group and the combined M plus L groups, except for more
multicentric/ multifocal disease in the UPM groups (43% vs 9% p0.0001).
Discussion: Fourteen percent of young women at our centre, a 1/3 with a relatively poor prognosis had UPM. Data correlating
surgical choice with baseline and follow-up psychological variables will be presented. There appears to be a need for an
educational/psychological intervention at the time of BC diagnosis for non-high-risk women contemplating preventive breast
surgery to ensure that such a choice is truly informed.

Title: Initial experience with an ambulatory extended recovery program for patients undergoing mastectomy
Aidan T Manning1, Danielle Cassella1, Stacy Ugras1, Beverly Tseng-Reyes1 and Lisa Sclafani1. 1Memorial Sloan Kettering Cancer
Body: Introduction: An ambulatory surgical program has been introduced at Memorial Sloan Kettering Cancer Center for patients
undergoing select procedures that require a single overnight stay. The aim of this study was to review the initial experience with
this program for patients undergoing mastectomy and to determine the rate and cause of unanticipated hospital admission.
Methods: All patients undergoing mastectomy with or without implant based reconstruction from March 2013 to February 2014
inclusive were entered into the Ambulatory Extended Recovery (AXR) program and data were recorded in a prospectively
maintained AXR database. Data on patient demographics, type of procedure performed, and whether the patient remained on the
AXR program were extracted. Electronic Medical Records were reviewed for all patients who required hospital admission in order
to determine the reasons for this.
Results: 926 consecutive patients (905 female, 21 male) requiring mastectomy with or without implant based reconstruction were
entered into the AXR program during this one-year period (mean age was 51 years, range 21-90). The procedures performed
were as follows: bilateral mastectomy with reconstruction (n=433, 46.8%); bilateral mastectomy without reconstruction (n=48,
5.2%); unilateral mastectomy with reconstruction (n=255, 27.5%); and unilateral mastectomy without reconstruction
(n=190,20.5%), with or without axillary procedures. Reconstructive procedures deemed suitable for the AXR program included
tissue expander or permanent implant insertion only. 861 of 926 patients (93%) remained on the AXR program and were
discharged following overnight stay. 62 patients (6.7%) (61 female, 1 male) did not complete the AXR program and required
hospital admission (mean age, 52 years; range, 22-81). 3 additional patients (0.3%) required hospital admission on occasions
that the AXR unit was at maximum capacity. Reasons for admission are shown in Table 1. Of the 26 patients with postoperative
hematoma, 17 were brought back to the Operating Room for definitive management and 9 patients were treated conservatively.
Following admission, most patients (52 of 62, 83.9%) were fit for discharge after 1 day. Of the remaining 10 patients, 9 were
discharged after 2 days and 1 after 5 days.
Reason for hospital admission for 62 of 926 patients scheduled for breast surgery on the Ambulatory Extended Recovery (AXR)
program
Reason for Admission
Hematoma
26
Pain control
17
Nausea
Urinary Retention
Cardiac Issues
Other (respiratory issue, poor mobility, social admission, fall on ward, pyrexia)
Total
62
Conclusion: Unilateral and bilateral mastectomy, with or without implant based reconstruction, is safely performed in the setting of
an AXR program. Only a small minority of patients will subsequently require hospital admission, most commonly for management
of postoperative hematoma or inadequate pain control.

Title: Implication of breast-conserving therapy in the subtype era of breast cancer
Sanghwa Kim1, Hyung Seok Park1, Jee Ye Kim1, Jegyu Ryu1, Seho Park1 and Seung Il Kim1. 1Yonsei University College of
Body: Background
Molecular subtypes of breast cancer are one of the key factors to predict clinical outcomes in women with breast cancer. Some
specific subtypes like triple-negative breast cancer (TNBC) show more aggressive behaviors and worse clinical outcomes than
luminal A subtype, thus there are conflicting data regarding optimum local surgical strategy for TNBC. We evaluate the clinical
outcomes of women who underwent breast-conserving therapy (BCT) compared to those underwent mastectomy regarding to
breast cancer subtypes using a large single center cohort.
Methods
A total of 5353 women who underwent BCT or mastectomy due to primary breast cancer from 1990 to 2010 were retrospectively
reviewed. Cases with initial distant metastases or those with neoadjuvant chemotherapy were excluded. Clinicopathological
characteristics, overall survival (OS), and recurrence-free survival (RFS) were analyzed using the Chi-square test, Kaplan-Meier
survival analysis, and log-rank test. Cox proportional hazard models were used for multivariate analyses. In order to explore the
role of BCT in TNBC, we performed sub-group analysis using patients with TNBC in the cohort.
Results
BCT was performed in 1866 cases and mastectomy in 3487 cases. The mastectomy group had higher T stage (T2-3: 53.5% vs.
26.5%, p<0.001), higher N stage (N2-3 17.1% vs. 5.1%, p<0.001), and more HER2 over-expression (11.6% vs. 8.3% p<0.001)
than the BCT group had. The BCT group consisted of more Luminal A and TNBC subtypes than the mastectomy group. (Luminal
A 56.2% vs 41.6%, TNBC 19.9% vs 12.7%, all p<0.001)
The 5-year RFS rates of the BCT group in luminal A, B, and TNBC were better than those in the mastectomy group (Luminal A
93.7% vs 89.4%, p<0.001, Luminal B 94.9% vs. 86.3%, p=0.01, and TNBC 92.9% vs 79.8%, p<0.001). However, the 5-year RFS
of HER2-enriched subtype was not significantly different according to the operation type (p=0.85). The BCT group in luminal A
and TNBC showed better OS than the mastectomy group (Luminal A, p=0.002, TNBC, p<0.001), however, the difference of OS
between the BCT and mastectomy groups in luminal B and HER2-enriched subtypes was not significant (p>0.05).
In multivariate analyses, T and N stage, breast cancer subtypes, histologic grade, and the status of adjuvant chemotherapy were
independent prognostic factors for RFS and OS (all p<0.05). However, The statistical significance of RFS and OS of local
therapy, BCT vs. Mastectomy, according to breast cancer subtypes in the univariate analyses disappeared in multivariate
analyses (HR for RFS=0.90, 95%CI=0.70-1.16, HR for OS=0.83, 95%CI=0.614-1.122). In the sub-group analysis, BCT showed
comparable outcomes compared to mastectomy in patients with TNBC in multivariate analyses. (HR for RFS=0.89,
95%CI=0.38-2.06, HR for OS=1.007, 95%CI=0.987-1.028)
Conclusions
Clinical outcomes were not affected by surgical approaches regarding to breast cancer subtypes. BCT is an acceptable surgical
approach regardless of breast cancer subtype, and even in selective patients with TNBC.

Title: Assessment of diagnostic and therapeutic value of ductoscopy biopsy in single intraductal papillary lesion compared with
open surgery
Zheli Xu1, Wanying Xing1, Qiang Li2, Deli Xing3 and Yiqi Gu1. 1Breast Surgery of the 3rd Clinical Medical College of Norman
Bethune Health Science Center of Jilin University, Changchun, Jilin, China; 2China-Japan Union Hospital of Jilin University,
Changchun, Jilin, China and 3China-Japan Union Hospital of Jilin University, Changchun, Jilin, China.
Body: Purpose
Papillary lesions (PL) account for 12 % of all breast neoplasms and fiberoptic ductoscopy is a practical and direct method
compared with open biopsy in diagnosis and treatment of intraductal papillary lesions. This study aimed to assess efficacy of
ductoscopy in the diagnosis and management of single intraductal papillary lesion.
Methods
A total of 232 patients at China-Japan Union Hospital of Jilin University who were diagnosed with single intraductal papillary
lesion by fiberoptic ductoscopy were enrolled from March 2011 to November 2013. All patients underwent ductoscopic
papillomectomy before open surgery. The final pathologic diagnoses were made by using specimen from both ductoscopic
papillomectomy and surgeries. Any intraductal papillomatous lesion or surface abnormalities were considered as positive findings
during ductoscopy.
Results
Histopathologic investigations of surgically excised or ductoscopically removed lesions revealed that 217 out of 232 cases were
positive. In 187 cases pathological changes were only found in ductoscopic papillomectomy specimen, while the number of cases
which was only found in open biopsy was 24. In 6 cases ductoscopic papillomectomy and surgeries specimen both showed
positive.
Conclusions
The accuracy of ductoscopy biopsy in diagnosis was 88.9% (193/217) and specificity was 100%. In 187 out of 217 cases (86.2%)
ductoscopic papillomectomy alone could completely remove the lesion. Ductoscopy is an effective tool for diagnosis and
treatment of single intraductal papillary lesion.

Title: Multimodality perioperative analgesia with paravertebral nerve block and gabapentin reduces narcotic use and hospital
length of stay in mastectomy patients
Zandra H Cheng1, Vlad Frenk1, Jennifer D Bishop1, Theresa Bowling1 and Helen A Pass1. 1Stamford Hospital, Stamford, CT.
Body: Postoperative pain control is the major determinant in hospital length of stay (LOS) in patients undergoing mastectomy.
Using a multimodality approach for peri-operative analgesia (PA) with paravertebral nerve block (PVB)(regional anesthesia) as
well as including pre- and postoperative oral gabapentin, we significantly reduced both LOS and narcotic usage (NU) when
compared to PVB alone or conventional postoperative management (CPM) with on demand postoperative pain medications. A
single institution, retrospective chart review of patients undergoing mastectomy from 2009 to 2014 was performed (n = 129; 84
bilateral, 45 unilateral) with a subset analysis performed on patients undergoing tissue expander (TE) reconstruction (n=86) or
bilateral mastectomies (n=84). Patients were grouped by PA type (CPM, PVB by catheter infusion, and PVB with gabapentin
(PVB+G)). Data were analyzed via using Student t-test and significance was defined as p<0.05. As seen in the table below, LOS
and NU decreased with increasing multimodal PA approach. LOS was significantly decreased by PVB+G compared to CPM and
PVB for all mastectomies, bilateral mastectomies, and mastectomies with tissue expander (TE) reconstruction. NU was
significantly decreased by PVB+G compared to CPM and PVB for all mastectomies and bilateral mastectomies, and trended
toward decreasing NU in TE reconstruction.
All (n = 129)
(1) CPM (n=51)
(2) PVB (n=35)
(3) PVB+G (n=53)
p 1 vs 3; p 2 vs 3
LOS (days)
2.3 (0.84)
2.1 (0.71)
1.61 (0.54)
<0.0001; <0.0005
NU (mg)
73 (42.9)
52 (32.3)
39 (25.3)
<0.001; 0.04
(1) CPM (n=25)
(2) PVB (n=21)
(3) PVB+G (n=40)
LOS (days)
2.2 (0.50)
2.3 (0.63)
1.61 (0.44)
<0.001;<0.001
NU (mg)
86 (32.6)
55 (34.4)
43 (36.3)
<0.001; 0.10
(1) CPM (n=20)
(2) PVB (n=24)
(3) PVB+G (n=40)
LOS (days)
2.4 (0.7)
2.2 (0.72)
1.65 (0.52)
<0.009;<0.001
NU (mg)
84 (37.4)
58 (32.4)
38 (24.5)
<0.001;0.0066
TE reconstruction (n=86)
Bilateral (n=84)
NU = narcotic use normalized to morphine sulfate

We have found employing a multimodality approach to PA with the addition of the GABA analogue, gabapentin, to regional
anesthesia via PVB catheter infusion significantly improves the postoperative course of patients undergoing mastectomy
procedures by decreasing LOS and NU.

Title: Does the implementation of an enhanced recovery programme impact on postoperative outcomes in populations with
significant comorbidity and social deprivation?
Ee Von Woon1, Adrian Wong1, Juliette Murray1 and Alison Lannigan1. 1Wishaw General Hospital, Wishaw, Lanarkshire, United
Kingdom.
Body: Background:
Enhanced Recovery Programmes (ERP) are a well established evidence-based model of care which are intended to reduce the
impact of surgery and safely reduce the length of inpatient stay for surgical patients. In Lanarkshire, Scotland, the ERP was
introduced in Wishaw General Hospital in 2012 but has yet to be extended to neighbouring Monklands Hospital. We audited the
impact of the ERP by comparing post-operative outcomes between these district general hospitals in the same health board
servicing adjacent catchment areas which are both in areas of significant social deprivation.
All patients who underwent breast surgery from August 2012 to August 2013 inclusive were identified from a prospectively
collected electronic database. Parameters analysed included ASA grades, length of postoperative stay, rate of post-operative
complications and re-admissions. The relative social deprivation of patients was calculated by cross referencing their postcodes
with the Scottish Index of Multiple Deprivation (SIMD) 2012.
Results:
294 and 152 patients underwent 336 and 161 breast operations in Wishaw and Monklands respectively. The mean age of these
patients was 57 in Wishaw and 54 in Monklands. 30% of patients in Wishaw and 19% of patients in Monklands had ASA grade 3
(range = 1 to 3, p=0.08). In both hospitals, the most common diagnosis was breast cancer (Wishaw: n=257, 76%; Monklands:
n=110, 68%; p=0.17) and the most common procedure performed was wide local excision (Wishaw: n=235, 70%; Monklands:
n=114, 71%; p=0.73). The mean postoperative stay was 0.9 days in Wishaw compared to 2.0 days in Monklands (p<0.001).
Postoperative complications were higher in Monklands (n=37, 23%) compared to Wishaw (n=52, 15%, p=0.04). There was no
significant difference in the ASA grades of patients who developed complications, rates of readmissions, A&E visits or reoperation
(p>0.05). Although the Wishaw patient cohort is living in significantly more deprived areas (mean SIMD rank=2543) compared to
the Monklands patient cohort (mean=2915, p=0.02), there was no significant difference in deprivation status between patients
with complications in these two catchment areas (p=0.65).
Conclusion:
The ERP is a safe and effective protocol for breast surgery patients with low complication rates and its implementation halved the
inpatient admission time in our cohort. The savings derived from this would outweigh the running costs of the ERP.

Title: Predictors of bilateral mastectomy in breast cancer patients
Yun Fu1,2, Zhigang Zhuang2, Michelle Dewing1,4, Apple Sophia3 and Chang R Helena1. 1Revlon/UCLA Breast Center, , David
Geffen School of Medicine at UCLA, Los Angeles, CA; 2Shanghai First Maternity and Infant Hospital, Tongji University School of
Medicine, Shanghai, China; 3David Geffen School of Medicine at UCLA, Los Angeles, CA and 4University of Colorado Health
Breast Surgery, Colorado Springs, CO.
Body: Background: Breast cancer surgery has experienced an evolutional change from radical mastectomy to conservative
techniques due to the migration of stage of disease at diagnosis and adjuvant use of multimodality treatment. While breast
conservation is preferred by most, the trend of bilateral mastectomy has been on the rise in the United States. The aim of this
study is to determine factors that may affect patients choice of bilateral mastectomy.
Methods: This is a retrospective study of 376 patients diagnosed with primary invasive breast cancer who were treated by
bilateral or unilateral mastectomy (BM or UM) at the Revlon/UCLA Breast Center between Jan. 2002 and Dec. 2010. Patients in
the bilateral mastectomy (BM) group were further divided into groups of bilateral mastectomy for bilateral breast cancer and for
unilateral breast cancer and contralateral prophylactic mastectomy.
Results: When compared with the UM group, the following factors were found to be associated with the BM: younger age
(p<0.001), pre-menopause (p<0.001), having a family history of breast cancer (p<0.001) or ovarian cancer (p=0.017), BRCA 1
and 2 mutations (p<0.001, p=0.011, respectively), more breast biopsies (p=0.007), history of breast augmentation (p=0.014),
more MRI study within 6 months before the surgery (p=0.029), more likely to have reconstruction surgery (p<0.001) and sentinel
lymph node biopsy (SLNB) (p<0.001). Multivariate analysis indicated that patients with smaller tumor size (p<0.001, OR 0.087),
negative nodes (p<0.001), sentinel lymph node biopsy as nodal surgery (p<0.001, OR 0.259), BRCA mutation (p=0.020, OR
6.537) and positive family history (p=0.001, OR 2.732) were more likely to choose bilateral mastectomy with reconstruction using
tissue expanders or implants (p<0.001, OR 4.546).
Conclusion: Bilateral mastectomy is associated with lower TN stage, requiring only SNLB, presence of BRCA mutation and/or
high risk family history. Tissue expanders or implants based reconstructions were more frequently chosen by patients with
bilateral mastectomy.

Title: Does therapeutic mammoplasty reduce mastectomy rates?
Jennifer Pollard1, Pang Wong1, James Mansell2, Juliette Murray1, Alison Lannigan1, Julie Doughty2, Laszlo Romics3, Sheila
Stallard2 and Christopher Wilson2. 1NHS Lanarkshire, Wishaw; 2Western Infirmary, Glasgow and 3Victoria Royal Infirmary,
Glasgow.
Body: Introduction: Therapeutic mammoplasty (TM) is increasing in popularity as a method for enhancing breast conserving
surgery. Studies have shown its oncologically safe, whilst improving cosmetic outcome1,2,3. Therapeutic mammoplasty has
been gaining popularity in West of Scotland over the past five years. Initially it was thought this may reduce requirement for
mastectomy and immediate reconstruction. We have recorded type of surgery patients would have required had they not been
suitable for TM, looking at changing demographic of surgical workload.
Methods: Prospective data collected about patients undergoing TM in West of Scotland since 2011 in Victoria Infirmary, Western
Infirmary and in NHS Lanarkshire. We reviewed clinical indications for TM, surgical alternative, Body Mass Index (BMI) and
smoker status.
Results: Seventy-nine patients were identified. In 67 cases, alternative surgical option of mastectomy or standard conservation
was recorded. Mean BMI was 29. 41% of patients had contralateral surgery for symmetry at the same time. In 28 cases (35%)
TM avoided need for mastectomy. In 39 cases (49%) it was felt that cosmetic result would be improved by TM compared with
standard conservation. During the study period, rates of mastectomy with immediate reconstruction as a proportion of total
number of treated cancers have remained similar.
Conclusions: Whilst introduction of therapeutic mammoplasty in our region has improved options offered to patients and likely
cosmetic outcomes, it has not had a major impact in reducing mastectomy rates or demand for immediate reconstruction. It has
probably increased surgical workload of plastic surgeons as these cases are often performed as joint procedures.
References:
1. Iwuchukwu OC, Harvey JR, Dordea M, Critchley AC, Drew PJ. The role of oncoplastic therapeutic mammoplasty in breast
cancer surgery A review. Surgical Oncology. 2012;21:133-141
2. McCulley SJ, Macmillan RD. Planning and use of therapeutic mammoplasty Nottingham approach. British Journal of Plastic
Surgery. 2005;58:889-901
3. Clough K, Lewis J, Couturand B, Fitoussi A, Nos C, Falcou M. Oncoplastic techniques allow extensive resections for
breast-conserving therapy of breast carcinomas. Annals of Surgery. 2003;237(1):26-34.


Title: Intraoperative ultrasound guided wire bracketing of nonpalpable breast lesions for excision
David T Rock1, Aimee L Stewart1 and Samith Sandadi1. 1Regional Breast Care, 21st Century Oncology, Fort Myers, FL.
Body: Background. Mammographic wire localization is the primary means of localizing nonpalpable breast lesions for excision.
Unfortunately, that procedure is associated with patient discomfort and vasovagal responses, scheduling difficulties, and
problems with inaccurate wire placement or wire displacement. In addition, the wires often enter far from the target and take an
indirect course resulting in high positive margin rates. We reviewed our experience with intraoperative ultrasound guided
bracketing of nonpalpable lesions to determine if the accuracy of the excision and patient factors could be improved.
Methods. Patients that underwent intraoperative ultrasound guided wire bracketing were identified by electronic chart review.
After the patient was anesthetized, high frequency linear array ultrasound was used to identify the target lesion. Under direct
ultrasound visualization, the lesions were bracketed with 4 hook-wires placed perpendicular to the chest wall at the radial and
antiradial margins of the lesion. Excision of the lesion was then performed with circumferential dissection around all 4 localization
wires.
Results. Intraoperative localization was planned on 119 patients. The lesion was identified by preoperative ultrasound in 110 of
those patients (92%). The remaining patients required mammographic localization. In 43 patients (85%) the lesion was identified
for bracketing, in the remaining 17 patients the biopsy clip was seen by ultrasound and bracketed. Only 3 patients had positive
margins (no tumor touching ink) requiring re-excision (2.7%).
Conclusions. The results of our experience show that intraoperative ultrasound guided wire bracketing is a reliable alternative to
mammographic localization and is associated with a lower positive margin rate than reported for mammographic wire localization
with a single wire. In addition, patient comfort is improved, cost is reduced, and there are no scheduling issues. The technique is
especially useful for oncoplastic excisions where the incision is remote from the target lesion.

Title: Impact on survival of primary tumor resection in women with de novo metastatic breast cancer. The GEICAM Alamo I-III
breast cancer registry (1990-2001)
Sara Lpez-Tarruella1, Mara Jos Escudero17, Miguel Martn1, Carlos Jara2, ngel Guerrero3, Ana Lluch4, Ana Santaballa5,
Purificacin Martnez del Prado6, Juan Lao7, Emilio Alba8, Antonio Fernndez9, Raquel Andrs10, Antonio Llombart11, Norberto
Batista12, Ignacio Porras13, Jos Manuel Lpez-Vega14, Encarna Adrover9, Lourdes Calvo15, Marina Polln16 and Eva Carrasco17.
1
IIS Gregorio Maran, Madrid, Spain; 2F H de Alcorcn, Alcorcn, Spain; 3IVO, Valencia, Spain; 4H C Valencia, Valencia, Spain;
5
H U La Fe, Valencia, Spain; 6H U Basurto, Bilbao, Spain; 7H G U Miguel Servet, Zaragoza, Spain; 8C H Virgen de la Victoria,
Mlaga, Spain; 9CHUA, Albacete, Spain; 10H C U Lozano Blesa, Zaragoza, Spain; 11H U Arnau de Vilanova, Lleida, Spain; 12H U
de Canarias, Sta Cruz de Tenerife, Spain; 13C H Reina Sofa, Crdoba, Spain; 14H U Marqus de Valdecilla, Santander, Spain;
15
C O Juan Canalejo, A Corua, Spain; 16ISCIII, Madrid, Spain and 17GEICAM, San Sebastin de los Reyes, Madrid, Spain.
Body: Introduction: Retrospective data from institutional series and population-based databases have suggested a potential
benefit of the primary tumor (PT) surgery in de novo metastatic breast cancer (MBC) patients (pts). Recently reported prospective
data from 2 randomized trials and a multicenter registry questioned the real role of the local approach in the modern
individualized systemic treatment era. Methods: The ALAMO (A) is a retrospective analysis of pts diagnosed with BC between
1990 and 2001 across 56 GEICAM hospitals in Spain. Patterns of BC presentation (tumor and host characteristics), treatment
and survival were recorded in 3 cohorts, AI (1990-93, 4529 pts, closed by 2000), AII (1994-97, 10453 pts, closed by 2003) and
AIII (1998-2001, 10675 pts, closed by 2007). MBC pts at first diagnosis excluding those without complete information about their
PT surgery were included. Descriptive, Kaplan-Meier and Cox regression analyses were carried out. Results: 5.5% (N=1415) of
the ALAMO database pts were initially diagnosed with MBC, 1331 fulfilled the present analysis criteria (327 from AI, 619 from AII
and 385 from AIII). Median age was 63.1 years (range: 21.6-96.0), 51.8% had single-organ metastasis, and their distribution
according to the predominant site of disease was skin/soft tissue (16.2%), bone (33.7%), and visceral (48.4%). Surgery of the PT
was done in 44.5% (N=592) of pts (512 with radical procedures, 46 with palliative procedures and 34 unknown); besides, 427 pts
underwent axillary dissection. Initial local treatment was the choice for 380 pts (358 surgery and 22 radiotherapy), 722 received
initial systemic therapy (480 chemotherapy, 214 endocrine treatment and 28 both), 29 received best supportive care and for 200
pts the treatment sequence could not be established. Pts in the surgery (S) group were younger (19.5% vs 11.8% were <44
years-old in the S vs non-S group respectively), with oligometastatic disease (61.9% vs 43.9% with single-organ involvement in
the S vs non-S group respectively) and with different sites of disease (40.2% vs 54.3% with visceral and 39% vs 29.8% with bone
metastasis in the S vs non-S group respectively). With a median follow-up of 1.9 years, the 5-yr overall survival (OS) was 25.4%
in the entire de novo MBC population, with a median OS of 3.3 yrs in the S-group vs 1.9 yrs in the non-S-group (HR 1.69,
p<0.0001). Subgroup analyses showed a benefit of PT surgery in OS regardless the number of metastasis and site of disease,
but didn't show this benefit analyzing pts according to BC subtypes. The multi-adjusted HR for surgery was 1.38 (p=0.037). The
multivariate Cox regression analysis model included the site of disease (p=0.028), the histopathologic grade (p=0.019) and the
hormone receptor status (p=0.007). Discussion: The Alamo data line up with previously reported population-based registries,
which highlight the better survival outcome of de novo MBC pts undergoing PT surgery. However, the consideration of the
biological heterogeneity of BC has changed the landscape of systemic treatment. Only well designed randomized controlled trials
will have the power to discriminate between a consistent bias and a real biologic effect of the PT surgery.

Title: Oncologic outcome of 2,217 patients with breast cancer after negative sentinel lymph node biopsy without axillary lymph
node dissection
Jong Tae Park1, Hak Min Lee1, Sung Gwe Ahn1, Seung Ah Lee2 and Joon Jeong1. 1Gangnam Severance Hospital, Yonsei
University College of Medicine, Seoul, Korea and 2Eulji general Hospital, Eulji University College of Medicine, Seoul, Korea.
Body: Background
Sentinel lymph node biopsy (SLNB) is a standard method for nodal staging and enables to omit axillary lymph node dissection
after negative sentinel lymph node (SLN). False-negative rate of SLNB has been reported as 5-7%. To dissolve the concerns
regarding the axillary recurrence due to false-negative cases after negative SLNB, we investigated oncologic outcomes of a large
number of breast cancer patients with negative SLN after SLNB.
Methods
A total of 2,475 patients with clinically node-negative breast cancer underwent SLNB at Severance hospitals between December
1998 and December 2013. 99-m Tc radiocolloid or blue dye was used in SLNB with periareolar intradermal injection technique.
Intraoperative frozen examination of SLN was performed. If frozen test for SLN showed negative result, further axillary node
dissection was omitted. Survival rates were estimated by the Kaplan-Meier method. Multivariate survival analysis was performed
using the Cox-regression hazard model.
Results
Among 2,475 patients, 2,217 patients had no metastatic focus on SLN. A mean of 2.4 sentinel lymph nodes was dissected in all
of the patients. At a median follow-up of 36 months, only 13 patients (0.6%) had axillary recurrence, and a median time to the
recurrence was 20 months (range 7-45 months). During the follow-up period, 22 local, 16 regional, and 44 distant recurrences
were observed. In multivariate analysis for axillary recurrence, tumor size and refusal of systemic therapy were demonstrated to
be independent risk factors for axillary recurrence (hazard ratio (HR) 1.77, 95% confidence interval (CI), 1.08-2.89, P=0.024; HR
0.15, 95% CI, 0.03-0.75, P=0.021). Disease-free survival and axillary recurrence-free survival rates at 3 years were 96.5% (95%
CI, 0.960-0.970) and 99.2% (95% CI, 0.990-0.994), respectively.
Conclusions
The axillary recurrence rate was very low in patients with negative SLN. Our findings supported that SLNB is a reliable procedure
and its oncologic safety is not affected by the chance of axillary recurrence after negative SLNB.

Title: Papillary breast lesions: Association with malignancy and upgrade rates on surgical excision
Asma Mursleen1, Hanh Tam Tran, Omar M Ghanem and Maen Farha2. 1Medstar Union Memorial Hospital, Baltimore, MD.
Body: Introduction: The management of benign intraductal papillary (IDP) lesions is controversial. Radiological and clinical
findings alone cannot distinguish benign or malignant IDPs. Surgical excision of IDPs diagnosed by core or vacuum assisted
biopsies was the standard of practice. There is an increasing body of literature suggesting that observation may suffice.
Design: This study investigated upgrade rates of benign IDPs on surgical excision. A listing of all cases where the pathology
report included Papillary or papilloma was obtained after IRB approval. The upgrade rate on excised lesions and the association
of IDPs with atypia and malignancy were assessed. Clinical and radiological presentations of papillary lesions was obtained from
the medical records. We also reviewed recently published data on the subject for comparison.
Results: We reviewed 65 pathology reports of IDPs diagnosed at our institution between 2011-2014. Papillary lesions were
stratified in to three categories: benign, associated with atypia and associated with malignancy. Upgrade rates on surgical
excision of benign IDPs diagnosed with core biopsy was calculated. Of the 65 papillary lesions 67.7% were benign, 10.8% were
associated with atypia, and 21.5% were associated with malignancy. The upgrade rate on 28 excised benign IDPs was 7.14%.
A review of 15 studies published between 2011-14 on IDPs shows an average upgrade rate of 17% in the routine excision group
versus 2% in the observation group.
Intraductal Papillomas and upgrade rates on excision
Upgrade rate(%) in the routine excision studies
Upgrade rates(%) in the observation group
19
8.9
39
19
1.6
23
3.1
11.5
3.8
0
Numb of studies
Average rate
17.23
1.79
Clinical presentation included: Palpable mass, nipple discharge, itching, mammographic mass/asymmetry or calcifications. This
study found 69.3% of papillary lesions presented as mammographic findings including nodules/asymmetry (28%), calcifications
(24%), masses (16.7%). The remaining 29.3% of the papillary lesions presented as palpable masses(14.7%), discharge (4%),
itching (2.67%), and pain/tenderness(1.3%).
Conclusion: IDPs present as a radiologic finding in 2/3 of the cases and a clinical complaint in the remaining patients. A review
of the upgrade rates in the literature shows a range of 0-39%. Although larger cores from vacuum assisted biopsies may lower
the upgrade rates, a good number of studies show rates that do not justify observation only. As concerning is the association of
papillary lesions with atypia and malignancy in 32.3% of our patients indicating a possible causal relation. Our data and review of
the literature indicate that short of a prospective randomized trial to settle this question, IDPs should be routinely excised.

Title: Randomized clinical trial comparing 2-octylcyanoacrylate versus intradermic suture with nylon: Similar cosmetic results with
different safety profile
Ruffo Freitas-Junior1, Thiago S Becker1, Rosemar MS Rahal1 and Regis R Paulinelli1. 1Federal University of Goias (UFG),
Goiania, Goias, Brazil.
Body: Introduction: There are multiple options available for closing surgical incisions. This study compared the cosmetic results
between the use of 2-octylcyanoacrylate and nylon sutures in elective breast surgery. Objectives: To compare the efficacy and
cosmetic outcome between the adhesive and conventional suturing with nylon on the cutaneous synthesis mammary surgeries.
Methods: we performed a prospective, randomized, controlled trial with 79 patients, 37 in group 2-octylcyanoacrilate and 42 in the
group with nylon suture. We evaluated the surgical aspect of the scar in 40 and 180 days, the occurrence of complications (such
as dehiscence, hematoma, infection and allergic reactions), the size of the wound and breast lesions, surgical time and skin
closure time, hemoglobin and preoperative WBC, age, height and weight of patients. Statistical analysis was performed by t test
and chi-square. Results and discussion: the study was stopped before the end of the recruitment of patients for the presence of a
greater number of dehiscence in the adhesive group (OR: 11.42, 95% CI 1.36 - 96.02, p=0.007). There were no significant
differences between the groups regarding other complications, not in relation to surgical time and the aesthetic appearance of
scars after 40 and 180 days. The average size of the wound was greater in the adhesive group than in the suture, being
respectively 32.97 (+ 10.54) mm and 27.64 (+ 9.56) mm, no correlation of size with the largest number of dehiscence.
Conclusion: The results showed that the cosmetic scar appearance of 2-octylcyanoacrilate is equivalent to those obtained with
the intradermal nylon suture, but the risk of dehiscence is higher.

Title: No Show

Title: No Show

Title: The oncologic safety of nipple-areolar complex skin sparing mastectomy compared with skin sparing mastectomy: 8 years
follow up results
Kyung Jun Yeu1, Jeong Yeong Park1, Jung Eun Choi1, Su Hwan Kang1, Young Kyung Bae2 and Soo Jung Lee1. 1Yeungnam
University College of Medicine, Daegu, Korea and 2Yeungnam University College of Medicine, Daegu, Korea.
Body: Background: Skin sparing mastectomy (SSM) has been conducted in breast cancer patient because of both oncologic
safety and cosmetic satisfaction. Preservation of nipple-areolar complex (NAC) is helpful to keep more natural breast shape, but,
can cause anxiety about local recurrence. This study reviewed long term follow-up result of SSM and nipple-areolar skin sparing
mastectomy (NASSM), retrospectively.
Patients and methods: This study included 272 primary breast cancer patients who received SSM (94 patients) or NASSM (178
patients) except bilateral breast cancer from September 1996 to December 2008. Frozen section was conducted for analysis of
NAC resection margin status. In case of positive resection margin, NAC was sacrificed. Local recurrence and overall survival of
SSM and NASSM group was analyzed.
Results: The mean follow-up was 94.9 months. 81 NAC resection margins (29.8%) were invaded by tumor cells. The positive
resection margin of NAC was associated with presence of ductal carcinoma in situ (p=0.005), especially extensive intraductal
component (p=0.005) and invasive carcinoma with multiplicity (p=0.048). The patients in NASSM group tended to have more
worse disease free survival than those in SSM group (75.3% vs 86.2%, P=0.087). But, in analysis of only local recurrence
including NAC, there were 25 cases (14.0%, 7 in skin flap and 18 in NAC) of local recurrence in NASSM group and 8 (8.5%) in
SSM group. Local recurrence free survival of the NASSM group was 86.0% and that of the SSM group was 91.5% (P=0.278).
Distant recurrence after surgical treatment for local relapse occurred in only one SSM case. There was no significant difference
for the overall survival between NASSM and SSM group (97.8% vs 96.8%, p=0.556).
Conclusion: In this study, result of long term follow up showed that patients in NASSM group tend to have more local recurrence
than patients in SSM group, even if there is no statistically significance. However, surgically well-controlled local recurrence of
skin flap and/or NAC did not affect on overall survival. NASSM is alternative method for SSM with oncological safety and better
cosmetic outcome.

Title: DCIS in low resource settings: The black swan of breast health care?
Miriam Mutebi1, Lydia Cairncross1, Eugenio Panieri1 and Hannah Simonds1. 1Groote Schuur Hospital, University of Cape Town,
Western Cape, South Africa.
Body: In situ carcinomas of the breast may constitute between 25-30% of all screen detected tumors in countries that practice
routine screening. Of the in situ tumors, at least 80% are ductal carcinomas in situ (DCIS). In the United States, DCIS accounts
for at least 25% of all newly diagnosed breast cancers. Presentations may vary in non screened populations.
Aim: To determine the clinical presentation and surgical management of patients presenting with isolated ductal carcinoma in situ
at a single tertiary centre in the Western Cape, South Africa over a 5 year period.
( Jan 2008-Dec 2012)
Secondary: To review the diagnostic techniques most commonly used.
Results:
42 patients with isolated DCIS were identified (41 females and 1 male) with an average age of 58 years. DCIS comprised less
than 1% (42/3768) of all breast malignancies managed in this period. Most patients presented with a breast lump (23/42). 6
patients presented with nipple ulceration and 5 patients with a
nipple discharge. 8 patients had their lesions picked on mammography.
The diagnosis was made on core biopsy in 14 patients while 8 patients required excision of the palpable lump to make the
diagnosis. 6 patients had a punch biopsy, 4 patients underwent a micro-ductectomy and 8 patients required a stereotactic biopsy.
2 patients had a diagnostic ROLL performed.
In terms of primary surgical management, 23 patients underwent a primary mastectomy, 6 patients had a WLE and 6 patients had
a ROLL with therapeutic intent. 3 patients declined surgery and 1 was transferred to a different unit. 3 patients were poor surgical
candidates and were placed on tamoxifen only. In terms of axillary management, 9 patients had a concurrent axillary clearance
and 9 patients had a sentinel lymph node biopsy.
Of the 12 patients who had initial BCT (either ROLL or WLE), half required mastectomy for close or involved margins and 2
patients required re-excision of margins. 2 patients had an immediate reconstruction and 2 underwent a delayed reconstruction.
All the axillary node clearances and sentinel lymph nodes were negative for metastatic disease.
Considerations: The spectrum of presentation of breast malignancies differs markedly in resource limited and unscreened
populations. The commonest clinical presentation for DCIS was a breast lump and our incidence of isolated DCIS (1%) was much
lower than that reported in international series.
A number of patients with failed core biopsies required excision biopsies for diagnosis. ROLLs and stereotactic biopsies were also
used to make a diagnosis, but the skills and equipment for this are frequently lacking in most regional centers. The inappropriately
high axillary clearance rate for our patients could be explained by diagnostic concerns over concurrent invasive disease and the
previously limited availability of sentinel lymph node biopsy facilities.
Conclusion: Though rare, the management of DCIS in this set up serves to highlight the challenges of diagnosing and managing
breast malignancies in LMICs. Practical interventions like increasing health worker training in core biopsy methods, and in the use
of supportive aids like ultrasound, alongside increasing human capacity in cancer diagnostics, could help to improve the
management scope of cancers in LMICs.

Title: Cytology and histopathology evaluation of sub-nipple tissue during intraoperative and postoperative time to predict
neoplastic involvement of the nipple in patients with breast carcinoma
Giuliano M Duarte1, Maria Virginia Thomazin1, Andr Oliveira1, Fernado Tocchet1, Luciana Moreira1, Adriana Worschech1 and
Renato Z Torresan1. 1UNICAMP State University of Campinas, Campinas, SP, Brazil.
Body: Introduction: The analyses of sub nipple tissue (SNT) have been used by some surgeons to preserve or not the nipple in
nipple sparing mastectomy for breast carcinoma. Therefore, the intraoperative study of SNT becomes an important tool. However,
it is uncertain if the SNT evaluation can safely predict the nipple involvement for carcinoma. The aim of this study was to evaluate
the accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) of intraoperative imprint
cytology, intraoperative frozen section and postoperative paraffin histopathology of SNT to predict involvement of the nipple in
women with breast carcinoma.
Methods: It was realized a prospective study with 38 consecutives breast carcinoma women (stage 0, I, II and III) underwent to
mastectomy. It was excluded inflammatory carcinoma and clinical evident nipple involvement. After mastectomy, the nipple
areolar complex were dissected in an approximate thickness of 5mm simulating nipple-sparing flap dissection. Then, the SNT
were dissected and submitted to imprint citology and frozen section during intraoperative time. Subsequently, it were submitted to
routine paraffin histopathology analysis. The nipple was examined separately by paraffin histopathology (considered the gold
standard). We considered any atypical cells like positive findings in all exams (cytology, frozen and paraffin).
Results: The mean of patients age was 59 years, the mean of tumor size was 34 mm in clinical exam and 31mm in pathological
exam. The clinical and mammographic means of distance from tumor to nipple were 23 and 34 mm, respectively. The imprint
citology frozen section and paraffin histopathology of SNT showed: sensitivity 42.9%, 42.9% and 57.1%; specificity 80.6%, 96.8%
and 100%; accuracy 73.7%, 86.8% and 92.1%; PPV 33.3%, 75% and 100%; NPV 86.2%, 88.2% and 91.2%, respectively. When
we associated both intraoperative exams (imprint cytology and frozen section), the specificity (80.6% x 100%, p=0.01) and
accuracy (76.3% x 92.1%, p=0.02) were worse than postoperative exam. The false negative of postoperative analyses was 8.8%.
Conclusion: These preliminaries outcomes showed a moderate sensitivity and good specificity of three exams, the low false
negative rate of postoperative paraffin exam. Our dates suggest that SNT evaluation is a good method to predict nipple
involvement and possibly the postoperative evaluation (histopathology) is better than intraoperative evaluation (imprint cytology
and frozen section).

Title: Trend in primary tumor resection and disease specific survival in patients with metastatic breast cancer: A SEER database
analysis (1988-2011)
Mahvish Muzaffar1, Swapnil D Kachare1, Timothy L Fitzgerald1, Jan H Wong1, Kathryn Verbanac1 and Nasreen A Vohra1. 1East
Carolina University, Brody School of Medicine, Greenville, NC.
Body: INTRODUCTION: The value of primary tumor resection in patients with metastatic breast cancer is a topic of ongoing
debate. We aimed to analyze the trend and impact of primary tumor resection on survival over the last two decades using a
national database. We hypothesized that there would be a decreasing utilization of primary tumor resection over time with an
increase in disease specific survival.
METHODS: All patients with stage IV breast cancer at diagnosis between the years 1988 and 2011 were identified in the SEER
database. Univariate and multivariate descriptive and survival analyses were performed.
RESULTS: A total of 41,601 patients with stage IV breast cancer were included in the study, 98.9 % (41,162) were females and
1.1% were males. Forty percent underwent surgery. Table 1 summarizes the other significant differences in demographic and
tumor-related characteristics of patients who did and did not receive PTR. Over the 23- year study period there was a statistically
significant temporal trend of decreased primary tumor resection (62% of patients underwent PTR in 1988, 42.4% in 2000 and
27.7% in 2011). On univariate analysis, patients who underwent PTR had a greater median disease-specific survival (DSS), (34
vs. 18 months, p<0.0001). Younger age (p<0.0001), non-African American race (p<0.0001), lower T and N-stage (p<0.0001),
positive hormone receptor status (p<0.0001), lower grade (p<0.0001), mucinous histology (p<0.0001), radiation therapy
(p<0.0001), and surgery performed in the latter years (p<0.0001) were also associated with improved DSS. On multivariate
analysis increasing age (p<0.0001), AA race (p=0.0001), higher T and N stage (p<0.0001), negative hormone receptor status
(p<0.0001), higher grade (p<0.0001), no history of radiation therapy (p=0.002), and surgery in earlier years were associated with
increased mortality (p<0.0001). Not undergoing PTR was independently associated with increased mortality, (p<0.0001).
CONCLUSIONS: Results from this retrospective study suggest a survival advantage in patients with stage IV breast cancer who
undergo primary tumor resection. However, there has been a marked reduction in the number of patients undergoing surgery,
most likely reflecting more focused patient selection. Ongoing randomized controlled trials will help address the impact of primary
tumor surgery on survival of patients diagnosed with metastatic disease.
Table 1: Comparison of patient and tumor characteristics in the surgery vs no surgery group (all p<0.0001)
Variable
Surgery
No surgery
Female (%)
16,328 (98.7)
24,750 (99.1)
White Race (%)
13,198 (79.7)
19,328 (77.4)
T1 stage (%)
3,067 (18.5)
2,096 (8.4)
N1 stage (%)
4,173 (25.2)
5,834 (23.4)
ER Positive (%)
9,642 (58.3)
12,300 (49.3)
PR Positive (%)
7,532 (45.5)
9,346 (37.4)
Her2 Positive (%)
414 (4.5)
745 (3.0)
Grade II (%)
4,844 (29.3)
5,400 (21.6)
Infiltrating ductal cancer (%)
11,539 (69.7)
12,461 (50.0)
Radiation Therapy (%)
6,338 (38.3)
6,937 (27.8)

Title: Nipple sparing mastectomy: Risks of wound complication in the setting of neo-adjuvant or adjuvant chemotherapy and/or
radiation therapy
Selyne Samuel1, Rebecca Viscusi1, Amy Waer1, Victor Gonzalez1, Pavani Chalasani1, Craig Hurst1, Ethan Larson1, Robert
Livingston1 and Michele Ley1. 1University of Arizona, Tucson, AZ.
Body: Background: Surgical care of breast cancer has evolved significantly over the past 40 years. Nipple sparing mastectomy
(NSM) and skin sparing mastectomy (SSM) have become an increasingly used surgical management for women with malignant
breast disease. To date, there are limited recommendations about the role of NSM in patients receiving aggressive adjuvant
therapy. The purpose of this investigation is to determine whether NSM in the setting of neo-adjuvant or adjuvant chemotherapy
and/or radiation therapy increased the risks for wound complications.
Methods: A retrospective chart review of nipple sparing mastectomies at a single institution was performed from 2007 to 2014.
Multiple data points including neo-adjuvant or adjuvant chemotherapy and/or radiation therapy, obesity, smoking history, and type
of reconstructive surgery were examined in detail.
Results: We counted the procedures by breasts affected and identified 76 NSM in the time period which met criteria. Of the 76
NSMs, 27 breasts received neo-adjuvant(20) or adjuvant chemotherapy and/or radiation(7) therapy. 21 breasts in the NSM group
developed wound complications including skin flap necrosis (5), total nipple necrosis and loss (11) hematoma (2) infection(2) and
seroma(1). The complications were seen in 11 in the non adjuvant treated setting (14%) and 9 (33%) in the adjuvant therapy
setting.
Conclusions: Nipple sparing mastectomy are emerging as safe and adequate options for the management of malignant breast
disease. Our results show there is a significant risk of wound complication associated with neo-adjuvant and adjuvant
chemotherapy and radiation therapy in the setting of nipple sparing mastectomy. We are pursuing the development of new
surgical techniques and guidelines to reduce these risks in these high risk patients.

Title: Flap fixation reduces seroma in patients undergoing mastectomy: A significant implication for clinical practice
James Van Bastelaar1,2, Arianne Beckers1, Maarten Snoeijs1 and Yvonne Vissers1,2. 1Atrium Medical Center, Heerlen, Limburg,
Netherlands and 2Orbis Medical Center, Sittard, Limburg, Netherlands.
Body: Background
Seroma formation is a common complication following mastectomy for invasive breast cancer. Seroma formation can cause
problems in wound healing and infection, thus leading to seroma aspirations and repeated visits to the out-patient clinic. The key
to reducing seroma formations seems to partly lie in the obliteration of dead space. However, the techniques used to achieve this
goal are subjects of much controversy and debate. Mastectomy flap fixation (FF) is achieved by reducing dead space volume
using interrupted subcutaneous sutures. We hypothesized that obliteration of the dead space following mastectomy would
significantly reduce seroma formation, resulting in fewer outpatient visits and seroma aspirations. This study, aiming to reduce
seroma formation after mastectomy, is a retrospective study that compared conventional mastectomy with mastectomy and flap
fixation.
Methods
All patients undergoing mastectomy due to invasive breast cancer were eligible for inclusion. From May 2012 March 2013 all
patients undergoing mastectomy in 2 hospitals were treated using flap fixation. The skin flaps were sutured on to the pectoral
muscle using polyfilament absorbable sutures. The data was retrospectively analyzed and compared to a historical control group
(HC) that was not treated using flap fixation (May 2011 March 2012). In the HC group, only the skin was sutured. Closed
suction drainage was applied to all patients in both groups.
Results
180 patients were included; 92 in the FF group and 88 in the HC group. There were no significant differences in patient
demographics. 36% of patients developed seroma in the group that underwent flap fixation; 59% of patients developed seroma in
the historical control group (P=0.002). Seroma aspiration was performed in 15% of patients in the flap fixation group as opposed
to 43% of patients in the control group (P<0.001). Multivariate analysis showed that the effect of flap fixation varied with the extent
of surgery: whereas flap fixation reduced seroma formation in patients undergoing simple mastectomy or mastectomy with
sentinel node recovery, FF did not reduce seroma formation in patients undergoing mastectomy with axillary lymph node
dissection (i.e. modified radical mastectomy; P=0.04 for interaction.
Effects of flap fixation on seroma formation stratified by operation type
Historical Control
Flap fixation
Mastectomy
8/14 (57%)
1/6 (17%)
0.16
Mastectomy and SN
26/42 (62%)
13/52 (25%)
<0.001
Modified Radical Mastectomy
18/32 (56%)
19/34 (56%)
0.98
In contrast, flap fixation was associated with fewer seroma aspirations in all types of surgery (P=0.80 for interaction).
Effects of flap fixation on seroma aspiration stratified by operation type
Historical Control
Flap fixation
Mastectomy
6/14 (43%)
1/6 (17%)
0.35
Mastectomy and SN
19/42 (45%)
7/52 (14%)
0.001
Modified Radical Mastectomy
13/32 (41%)
6/34 (18%)
0.04
Conclusion
Patients undergoing mastectomy flap fixation displayed a significant reduction in seroma formation and fewer patients were
subjected to seroma aspiration. Patients undergoing flap fixation required fewer seroma aspirations. Flap fixation is an effective
surgical technique in reducing dead space and therefore seroma formation in patients undergoing mastectomy for IBC.

Title: External dosimetry and in vivo measurements improve surgical comprehension of intraoperative radiotherapy using
Intrabeam
Pierre-Francois Dupr1, Dounia Bouzid2, Petra Miglierini1, Olivier Pradier1, Dimitris Visvikis2, Julien Bert2 and Nicolas Boussion2.
1
Cancer Institut CHU, Brest, France and 2LaTIM INSERM UMR 1101 CHU Brest, Brest, France.
Body: Background
IORT is an option for patients with limited access to radiotherapy or patients wishing to avoid the prolonged course of External
Beam radiotherapy (EBRT). The recent 5-years results of the non inferiority trial TARGIT-A suggest that with a short follow-up the
5-years risk for local recurrence is greater for the IORT arm. Several options can be suggested to improve those results.
Objectives
We choose to enhance surgical comprehension of IORT with Intrabeam by performing in vivo measurements comparing to
external dosimetry taking into account tissues heterogeneities using a preoperative CT scan.
Material and Methods
Nine patients enrolled in french randomised trial RIOP-InCA underwent Intrabeam procedure for breast cancer treatment. Each
patient had a preoperative CT scan to simulate the treatment on a Monte Carlo platform using an x-ray source model; in order to
calculate the dose actually delivered by the Intrabeam system which takes into account the tissues heterogeneities. In vivo
measurements using thermoluminescent dosimeters are also performed to evaluate the dose to the skin. Comparisons are done
between simulated and measured data. Relative depth dose curves are also compared.
Results
The median age was 68.3 years [58-87]. The mean maximal diameter of lumpectomy was 71.3 mm [50-125]. The median size
was 10.8 mm [5-19] and the median margin status was 12.8 mm [2-32]. 2 patients had micro metastatic involvement of sentinel
lymph node without axillary clearance. SBR score was 1 for 5 patients and 2 for 2 patients. All histological subtype were ductal,
one patient presented associated DCIS. 6 patients presented luminal A phenotype, 2 luminal B and 1 luminal B with HER-2 neu
over-expression. 2 patients received supplemental EBRT (HER-2 neu over expression and SLN micro metastatic involvement). 2
patients had adjuvant sequential chemotherapy and all had adjuvant ani-aromatase hormono-therapy.
In vivo measurements on the skin using TLDs gives a mean dose of 1.3 Gy 1.1 Gy [0.1-4.9 Gy] in comparison with external
dosimetry which had a mean dose of 1.5 Gy 0.8 Gy [0.3-6.2 Gy] at the same positions. No patient received a dose superior to
the prescription of 6 Gy. Relative depth dose curves give a mean deviation of +36.7% 24.8% and +50.6% 16.8% in the
tangential and perpendicular axes respectively from the manufacturer. Using the shielding allow a reduction of the dose by 10
concerning the ribs.
Results
Relative deviation
SD
Mean deviation dose to the skin
In vivo/external
13.3%
2.2%
Mean deviation dose
Shielding/ no shielding
-60.2%
42.6%
Depht Dose curve
Tangential axe
36.7%
24.8%
Perpendicular axe
50.6%
16.8%
Conclusions
Results about the dose as a function of depth show clearly that we cannot consider breast as equivalent to water at this energy
and may taking in account breast density .It seems to be important to spread the skin correctly from the incision, to recover
carefully the applicator and use the shielding in order to avoid secondary effects as skin necrosis and ribs failures.
Our X-ray source model allows to have realistic dose distribution wich hepls in better surgical comprehension of IORT particularly
in the set up of the applicator.

Title: Immediate implant-based breast reconstruction following total skin-sparing mastectomy: Defining the risk of preoperative
and postoperative radiation therapy on surgical outcomes
Anne Warren Peled1, Frederick Wang1, Robert D Foster1, Rachel Lentz1, Michael Alvarado1, Cheryl A Ewing1, Laura J Esserman1,
Barbara Fowble1 and Hani Sbitany1. 1University of California, San Francisco, CA.
Body: Background
Radiation therapy is an increasingly common adjuvant treatment in breast cancer therapy. As total skin-sparing mastectomy
(TSSM) and immediate reconstruction becomes more widely performed, further defining the risks of various treatment regimens
on surgical and reconstructive outcomes after TSSM is important. In this study, we assess the effects of premastectomy and
postmastectomy radiation therapy on outcomes following TSSM and immediate prosthetic reconstruction.
Methods
All patients undergoing TSSM and immediate tissue expander/implant reconstruction at our institution between 2006 and 2012
were identified from a prospectively maintained database. Cohort 1 included patients undergoing TSSM and reconstruction
without any radiation. Cohort 2 included patients with a prior history of radiation before TSSM and reconstruction. Cohort 3
included patients undergoing radiation after TSSM and reconstruction. Complication rates were compared between cohorts.
Results
A total of 580 patients were identified undergoing 903 breast reconstructions following TSSM. Cohort 1 included 727 breasts,
cohort 2 included 63 breasts, and cohort 3 included 113 breasts. Compared to patients without radiation, patients with prior
radiation were more likely to develop severe infection (7.3% vs. 20.6%, p = 0.001), incisional breakdown (3.1% vs. 9.5%, p =
0.01), and expander/implant loss (5.1% vs. 20.6%, p < 0.0001). Similarly, patients with postmastectomy radiation had higher rates
of severe infection (22.1%, p < 0.0001) and expander/implant loss (17.7%, p < 0.0001) compared to patients without radiation,
though equivalent rates of incisional breakdown (6.1%, p = 0.57). All three cohorts showed similar low rates of partial or complete
nipple-areolar complex necrosis (1.2% vs. 3.2% vs. 0%, respectively).
Conclusions
Both preoperative and postoperative radiation following TSSM and immediate prosthetic reconstruction result in higher, but
acceptable, complication risks. Risks of infection and expander/implant loss in irradiated patients who have undergone TSSM are
similar to those reported after skin-sparing mastectomy. Further, the low rates of complications related to nipple-areolar complex
skin preservation in irradiated patients are equivalent to those seen in non-irradiated patients, supporting the safety of performing
TSSM in patients who require radiation therapy.

Title: Complications of breast reconstruction in a provincial population of breast cancer survivors
Elaine S Wai1,3, Ling Hong Lu2, Cheryl Alexander1, Mary L Lesperance2, Mary L McBride1,3, Scott Tyldesley1,3 and Chris Taylor3.
1
BC Cancer Agency Vancouver Island Centre, Victoria, BC, Canada; 2University of Victoria, Victoria, BC, Canada and
3
University of British Columbia, Vancouver, BC, Canada.
Body: Background: Breast cancer is the most prevalent cancer in women in Canada. There are varying reports of the use of
breast reconstruction in various populations and very few prospective trials or population reports of complications related to
breast reconstruction in breast cancer patients. The goal of this study is to describe the use of breast reconstruction in breast
cancer survivors in British Columbia, the risk of complications after breast reconstruction, and factors associated with a higher risk
of complications.
Methods: Electronic Records from the BC Cancer Agency were used to identify women diagnosed with invasive breast cancer
between 2001-2008 in British Columbia. Administrative hospital records used with permission of the British Columbia Ministry of
Health, obtained through PopulationData BC, were used to obtain details about all surgeries,hospital admissions and associated
complications for the cohort. Analyses of reconstruction for those treated with breast-conserving surgery (BCS) versus
mastectomy were done separately. For those who underwent mastectomy following BCS, complications after BCS before
mastectomy were attributed to the BCS, and those that occurred after mastectomy were attributed to the mastectomy. These
analyses included a description of the type of reconstruction procedures used in this era, and the frequency and types of
complications after the reconstruction procedures. Logistic regression was used to identify risk factors for complications post
reconstruction.
Results: The cohort consisted of 18,642 women; 4868 had mastectomy alone, 8633 had BCS alone, 2894 had both, and 2247
had neither surgery.
Of those undergoing BCS (11527), 746 had reconstructive procedures. Most of these were separate procedures; 55% were 50
years of age. The most common complication was "miscellaneous complication of internal prosthetic devices, implants or grafts".
One hundred eight (14%) of those with reconstruction after BCS had complications within 30 days of the reconstruction
procedure, 134 (18%) had complications within one year of the procedure.
Of those undergoing mastectomy (7762), 7497 had their mastectomies within 1 year of diagnosis. Ninety-six had bilateral
mastectomies. 1766 underwent reconstructive procedures; 74% were immediate procedures, 57% were 50 years of age. The
most common complication was "miscellaneous complication of internal prosthetic devices, implants or grafts". One hundred
sixty-seven (9%) of those with reconstruction after (or with) mastectomy had complications within 30 days of the reconstruction
procedure, 227 (13%) had complications within one year of the procedure. Regression analysis showed that no radiotherapy
(OR=0.65) and premenopausal status (OR=0.70) were significantly associated with a lower risk of complication post
reconstruction.
Conclusion: Complications related to reconstructive procedures for breast cancer happened in the minority of breast cancer
survivors. They were more common in those with BCS, and were more likely for those who underwent adjuvant radiotherapy or
were postmenopausal.

Title: Mastectomy without drains: Flap adhesion and closure of dead space using TissuGlu vs standard wound closure (SWC) a parallel cohort comparison
Stefan Paepke1. 1Klinikum rechts der Isar der Technischen Universitt Mnchen, Frauenklinik, Munich, Bavaria, Germany.
Body: Background: Mastectomy(ME) remains associated with high rates of minor complications and occasional revision surgery
for wound healing/seroma. Closed suction drain use in an effort to reduce seroma related complications is the current standard
procedure but drain efficacy has been contested in multiple reports. Mechanical closure of dead space with suturing techniques
has been shown to be effective but this is often not practical in these patients and increases operating time. Fibrin sealants may
reduce early drainage volume but have not been shown to allow elimination of drains. Recent reports have shown that closure of
dead space can be achieved through the use of a new resorbable surgical adhesive (TissuGlu, Cohera Medical, Pittsburgh)
allowing for the elimination of drain use in that procedure. Our work with this adhesive began in 2013, with use initially limited to
revisions and patients at high risk of postoperative complications; later extended use to ME in the general patient population
without drain placement. Here we present the results of our prospective observation of postoperative outcomes in this early
series.
Materials & Methods: A parallel cohort comparison was performed for non-inferiority of TG/No Drains vs. SWC with respect to
postoperative fluid management and seroma related complications. All primary ME +/- SLNB and/or ALND were prospectively
documented for operative procedure variables and outcomes, including male patients and bilateral cases. 56 MEs patients (66
MEs) from Feb 2013 through Feb 2014 were documented and analyzed. Total drained volume, aspirations, minor wound healing
complications and need for revision surgery were compared.
Results: Patient demographics and procedure variables were comparable in the two groups.
Patient demographics & procedure variables
Pt demographics & procedure variables
SWC
TG/ND
Patients
35
21
Mean Age
61.5
62.7
Mean BMI
25.7
25.7
Mean total drained + aspirated fluid per ME was 78 ml in the TG/No Drain group vs. 512 ml in the SWC group, an 85% reduction
(p=0.00). Four MEs required aspiration in the test group (15%) vs. 10 (25%) in the SWC group (ns). There were 3 revisions in the
SWC group and none in the test group (ns).
Outcomes
Outcomes
SWC
TG/ND
Number (mastectomies)
40
26
Mean Days with drains
6.4
-100%
Mean Total Drained Volume/mast. (ml)
463
-100%
Mean Total Drained + Aspirated (ml)/mast
512
78
Number (%) of mast. aspirated
10
25% 4
15% 0.539
-38% Fisher's Exact due to Small # of incidence
Number (%) of mast. w minor complications 13
33% 7
27% 0.625
-17% two proportion
Number (%) of mast. Requiring revision
8% -
0% 0.273
-100% Fisher's Exact Due to Small # of incidence
p value
0.000
-85% Mann-Whitney
Discussion/Conclusions: Our experience suggests a clear advantage in the TG/no-drain approach in mastectomy. In addition to
the improved patient comfort and earlier discharge with TG, patients with drains had significantly higher total fluid collection and
still required more aspirations on average than the test group. Non-inferiority for minor complications was not only achieved, but
there seems to be a trend toward reduced complications in the no drain cohort. A prospective, randomized study is planned to
verify our initial impressions.

Title: Is the combination of fat grafts and platelet rich plasma effective and safe? An experimental study in rats
Ruffo Freitas-Junior1, Alexandre R Blumenschein1, Marise AR Moreira1, Maria-Auxiliadora PC Cysneiros1 and Roseana N
Pereira1. 1Federal University of Goias (UFG), Goiania, Goias, Brazil.
Body: Autologous fat grafts and lipofilling can be used for breast reconstruction after breast conserving therapy. Fat is considered
an ideal filler because of its low cost, ease of harvest, abundance in the human body and low immunogenic and allergic reaction
due to its autologous nature. Platelet rich plasma (PRP) is a plasma fraction, with platelet count above baseline, generally
obtained via centrifugation of blood. PRP theoretically promotes tissue regeneration due to fact that it concentrates a greater
amount of growth factors essential in the process of tissue regeneration and neovascularization. This paper aims to examine if
the association of fat grafts and PRP improves graft viability in female rats, through an experimental, randomized and blinded
study, which involved 47 animals. These animals underwent fat graft harvest from their inguinal fat deposits and fat grafting
subcutaneously to their cranial region. In 22 animals the fat graft was mixed with PRP and in 25 the fat was grafted by itself. After
a 100 day period, the animals were sacrificed and the fat grafts were analyzed using scores from 0 (absent) to 4 (abundant), in
optical microscopy by two independent and blinded pathologists, by means of the following variables: fat graft cell viability, fat
necrosis, tissue inflammation and fibrosis. Regarding fat graft cell viability, the PRP group scored moderate/abundant in 63% of
the cases and the fat graft only group scored absent/slight in 72% of the cases (p<0.05). The PRP group presented lower fat
necrosis scores and lower tissue inflammation scores when compared to the fat graft only group (p<0.05). The presence of tissue
fibrosis was rarely observed in both groups. Tumors (dermoid cysts) within the fat grafts were observed in 3 animals in which the
grafts were mixed with PRP. It is concluded that PRP improves the viability and integration of fat grafts in rats, but more studies
are needed to fully understand the exact mechanisms that lead to this improvement and assess the safety of the method for use
in humans.


Title: Prediction of local recurrence following mastectomy
Sarah Al-Himdani1, Simon Timbrell1, Kian Tan1, Julie Morris1 and Nigel Bundred1. 1University Hospital of South Manchester NHS
Foundation Trust, Manchester, United Kingdom.
Body: Background: Rates of local recurrence following mastectomy vary considerably in the literature, ranging from 1.7% to
11.5%. Previous studies have attempted to identify risk factors for recurrence and whether these relate to aggressive disease or
to inadequate local treatment. We aimed to identify local recurrence rates at our institution to determine risk factors associated
with local recurrence.
Methods and materials: A retrospective study of all women undergoing mastectomies for unilateral T1-T3 breast cancer between
2000-2005 was undertaken. Kaplan-Meier curves were created and the log-rank test calculated to assess for differences between
groups. The incidence of local, regional and systemic recurrence was identified at 5 and 8 years post-mastectomy. Incomplete
excision/margin involvement was defined as margins 1mm or less. Demographic, biological and treatment-related factors
associated were used to identify predictors of local recurrence.
Results: In total, 581 women were studied (mean age: 59; range: 22-96); 462 simple mastectomies (SM) and 106 skin-sparing
mastectomies (SSM) which included 199 (34%) screen-detected and 382 (66%) symptomatic cancers. Mean follow up was 6.7
years (range: 5-12.4) and longest was 12.4 years. SSM patients had a lower mean age (p=<0.001), with a smaller tumour size
(p=0.022), fewer positive lymph nodes (p=<0.001) and less co-existing ductal carcinoma in situ (DCIS) (p=<0.001) than SM
patients. Recurrence rates for SM were 5.6% and 6.9% and for SSM, recurrence rates were 5.7% and 9.0% at 5 and 8 years
respectively. Recurrence in node negative patients was 2.2% at 5 years. Predictors of local recurrence were node positivity (HR
8.03; 95% CI 3.2-20) and margin involvement/close margins (HR 3.34; 1.64-6.8).
Conclusion: No significant difference between recurrence rates in SSM and SSM groups was identified. The presence of over four
positive lymph nodes and incomplete excision margins indicates an increased risk of recurrence and in these patients, immediate
SSM reconstruction should be considered prudently.

Title: A systematic review of the clinical evidence regarding PIP breast implants
Umar Wazir1, Abdul Kasem1 and Kefah Mokbel1. 1London Breast Institute, Princess Grace Hospital, London, United Kingdom.
Body: Background: Mammary implants marketed by Poly Implant Prothse (PIP) were found to contain industrial grade silicone
and this caused heightened anxiety and extensive publicity regarding their safety in humans. These implants were used in a large
number of patients worldwide for augmentation or breast reconstruction.
Methods: Articles were identified by searches of Medline, PubMed, Embase and Google Scholar databases up to March 2014
using the terms: "PIP", "Poly Implant Prothse","breast implants" and "augmentation mammoplasty" or "silicone" In addition the
websites of regulating bodies in Europe, USA and Australia were searched for reports related to PIP mammary implants.
Results: PIP mammary implants are more likely to rupture than other implants and can cause adverse effects in the short to the
medium term related to the symptoms of rupture such as pain, lumps in the breast and axilla and anxiety. Based on
peer-reviewed published studies we have calculated an overall rupture rate of 14.5% (383/2635) for PIP implants.
Conclusions: There is no evidence that PIP implant rupture causes long-term adverse health effects in humans so far. The
long-term adverse effects usually arise from inappropriate extensive surgery, such as axillary lymph node dissection or extensive
resection of breast tissue due to silicone leakage.

Title: Immediate breast reconstruction with anatomical implants following mastectomy: The Radiation perspective
Merav Akiva Ben-David1,2, Hila Granot1, Ilana Gelernter3 and Michael Scheflan4. 1Chaim Sheba Medical Center, Ramat-Gan,
Israel; 2Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel; 3School of Mathematics, Tel Aviv University, Tel Aviv,
Israel and 4Assuta and Herzliya Medical Centers, Tel Aviv, Israel.
Body: Purpose: Immediate implant-based breast reconstruction followed by postmastectomy radiation therapy (PMRT) is
controversial because of the risk of compromised treatment plans, particularly left-sided plans including internal mammary nodes
(IMNs) as target. Using direct-to-implant reconstruction with anatomical implants may improve plans due to their unique shape
over the chest wall. This single-institution study evaluated the effects of immediate breast reconstruction with anatomical implants
on the quality of PMRT delivered by 3-dimensional conformal radiation therapy (3D-CRT).
Methods and Materials: Patients undergoing immediate direct-to-implant reconstruction with anatomic implant, performed by a
single surgeon, were treated between 2008 and 2013. For each patient, 2 plans were created and calculated, including or
excluding IMNs. No electron fields were used. The primary endpoint was the dose distribution among reconstructed breast (RB),
heart, lungs and IMNs and between right and left breasts. Six patients were treated with the Varian RPM system for left sided BC
due to anterior heart position.
Results: Of 29 consecutive patients, 11 received right-sided and 18 received left-sided PMRT. For plans excluding IMNs, mean
Dmean was 49.09 Gy (98.2% of the prescribed dose) for right and 48.51 Gy (97.0%) for left RBs. For plans including IMNs, mean
Dmean was 49.15 Gy (98.3%) for right and 48.46 Gy (96.9%) for left RBs. Mean RB D95 ranged from 42.05 Gy (left side with
IMNs) to 45.15 Gy (right side with IMNs). Mean IMN Dmean was almost identical for left- and right-sided treatment (47.89 Gy and
47.27 Gy, respectively; P=.340). Heart doses were very low, with mean Dmean values of 1.25 Gy (range, 0.83-1.46) and 1.56 Gy
(range, 1.23-2.10) for left-sided plans excluding and including IMNs, respectively (P<.001). Mean lung V20 values ranged from
13.80% for left-sided treatment excluding IMNs to 19.47% for right-sided treatment including IMNs.
Conclusion:
PMRT can be delivered effectively and safely by 3D-CRT after direct-to-implant breast reconstruction with anatomical implants.
The reconstructed breast and the IM chain coverage were excellent and the heart dose was very low, (probably due to the use of
the RPM system). Lungs V20 was comparable with other publication. Our study support the hypothesis that immediate breast
reconstruction per se is not an impediment to the delivery of high-quality PMRT by modern 3D-CRT technology. Based on our
findings, reconstructive surgeons should consider the use of anatomical implants for immediate breast reconstruction in patients
requiring subsequent PMRT.

Title: Variation in UK reconstructive practice in the face of post-mastectomy radiotherapy
James R Harvey1, Nigel J Bundred1, Cliona C Kirwan1, Ashu Gandhi1 and Paula J Duxbury2. 1Manchester Academic Health
Sciences Centre, University of Manchester, Manchester, United Kingdom and 2Nightingale Centre, UHSM, Manchester, United
Kingdom.
Body: Approximately 30-40% of women are not offered immediate breast reconstruction because the possibility of
post-mastectomy radiotherapy (PMRT) is unknown at the time of mastectomy. Breast reconstruction may be delayed until final
pathology is available and need for radiotherapy established. Surgical literature is replete with studies of varying quality, reporting
complication rates for a range of reconstructive procedures, highlighting the need for surgical trials of reconstructive techniques in
women at risk of PMRT. Decisions for these patients are complex, involving multiple clinicians including surgeons and
oncologists. To inform a surgical trial design, we aimed to determine current UK surgical practice and gain an understanding of
the drivers behind decision-making.
Methods: A questionnaire, validated in a pilot population, was posted to Consultant members of the Association of Breast
Surgery (UK). We collected data on current practice in conducting Delayed, Immediate and Delayed-immediate reconstructive
surgery. We collated data on type and volume of procedure performed and factors affecting decision-making including delay to
adjuvant treatment, risk of complications, perception of patients quality of life (QoL) and aesthetic satisfaction.
Results: Of 355 surgeons, 130(37%) responded. Of these, 77% felt the current evidence base was not adequate to guide
surgical decisions and 80% felt a need for further trials to guide best treatment. Despite a lack of scientific evidence
demonstrating a difference in cosmesis or QoL between Immediate and Delayed reconstruction, 85% felt there is not equivalent
cosmesis and 71% felt there is not equivalent QoL between the two groups.
There is considerable heterogeneity in reconstructive approach to patients at risk of PMRT (Table 1). Delayed reconstruction
remains the most popular option, being regularly used by 94% of surgeons despite only 34% of surgeons believing the majority of
patients are satisfied with the approach. Significantly fewer surgeons perform Immediate implant based reconstruction (with or
without ADM) than Delayed (p<0.01). Implant reconstruction is performed by 71% of surgeons in patients at risk of PMRT, but
only 44% of surgeons felt patients were happy with the final results.
The three most important drivers in making a reconstructive choice were 1. Effect of PMRT on the cosmetic result 2. Minimising
risk of complications and avoiding delay to adjuvant treatment 3. Pre-operative uncertainty over the need for PMRT.
Table 1
No. of responses
Never
Sometimes
Often
Always
Delayed
50
51
19
Immediate implant
58
41
Immediate ADM+Implant
45
49
14
2-stage (delayed-immediate) Expander to permanent reconstruction
28
55
33
Immediate Autologous
29
59
22
Conclusions: Surgeons employ a variety of approaches to reconstruction in the face of PMRT, the most common approach
being delayed reconstruction. Decision-making is based upon individual surgeons perception of risks including likely delay to
adjuvant therapy and effect of PMRT on the reconstruction. Drivers appeared to be more surgeon-centred rather than
patient-based. There is awareness of a lack of evidence to support decision-making and the need for high quality studies.
Randomised clinical trials are needed to provide an evidence base for outcomes.

Title: Technical results and complication rates after nipple-sparing mastectomy and direct-to-implant breast reconstruction using
porcine acellular dermal matrix for implant coverage
Roland Reitsamer1 and Florentia Peintinger1. 1University Hosptial Salzburg, Breast Center Salzburg, Paracelsus Medical
University, Salzburg, Austria.
Body: Objective:
To evaluate the feasibility and complication rates of nipple-sparing mastectomy (NSM) and direct-to-implant breast reconstruction
using a porcine acellular dermal matrix (ADM) for implant coverage.
Methods:
NSM and direct-to-implant breast reconstruction using a porcine ADM for implant coverage was performed in a series of 91
breasts in 63 patients. Technical results, complications, and cosmetic results were collected from patient records.
Results:
Short-term complications within one month comprised minor complications as minimal nipple necrosis in 13.2%, and infection in
1.1%. Major complications, as hemorrhage with surgical evacuation in 4.4%, and implant loss due to skin breakdown in 4.4%,
totaled in 8.8%. Within three months after primary surgery 3 further severe complications with reoperation ocurred. In 87.9% no
second surgical intervention was necessary and cosmetic results were good or excellent. After a mean follow-up of 12 months
(range 5 - 35 months) no further complications could be observed. Patient satisfaction was high for 92.3% of the patients and
subjective cosmetic result was excellent in 90.1% and satisfactory in 2.2%.
Conclusion:
Direct-to-implant breast reconstruction using a porcine ADM for implant coverage after NSM is an innovative approach resulting in
a high patient satisfaction. The technique is challenging but feasible and complication rates are acceptably low.

Title: Antibiotic prophylaxis in prosthesis-based mammoplasty: A systematic review
Naisi Huang1,2, Mengying Liu1,2 and Jiong Wu1,2. 1Shanghai Cancer Center, Fudan University, Shanghai, China and 2Shanghai
Medical College, Fudan University, Shanghai, China.
Body: Background Implants are used in over two-thirds of breast reconstructions and all aesthetic augmentations. Although
considered as an aseptic surgery, infection rate after prosthesis-based breast surgery is reported to be 2% to 2.5% of patients
and represents the leading cause of morbidity after reconstructive and aesthetic surgery. Therefore, prophylactic antibiotics is
supported by several studies to prevent surgical site infection (SSI) and capsular contracture (CC). However, there is no high
quality evidence on antibiotic prophylaxis in prosthesis-based mammoplasty.
Methods An electronic search was conducted in Medline, Embase, and Cochrane. Studies of prosthesis-based breast surgery
with control group and antibiotic prophylaxis were included. Two authors independently screened, assessed and then extracted
information from the included studies. Mantel-Haenszel method was used to conduct meta-analysis. All analysis was performed
by STATA 12.0. Studies that were not qualified to be included into meta-analysis were summarized and described.
Results 15 studies were included into analysis. Average Jaded score for RCT studies was 2.5 (1-4). Average Newcastle score for
non RCT studies was 7.55 (4-9). 3 studies compared systematic antibiotic prophylaxis with no antibiotic use, while no significant
difference was observed. Compared with antibiotic prophylaxis within 24 hours, extended systematic antibiotic prophylaxis more
than 24 hours postoperatively could significantly reduce infection rate (RR=0.638, CI 0.453-0.898). The average SSI rate was
4.6% in extended antibiotic group versus 11.1% in control group. 15.38 patients needed to be treated to prevent 1 case of SSI.
However, in subgroup analysis, extended antibiotic prophylaxis could significantly decrease SSI rate in implant reconstruction
surgery (RR=0.508, CI 0.349-0.739), but not in aesthetic breast surgery (RR=1.458, CI 0.602-3.528). Topical antibiotic irrigation
could reduce CC rate (RR=0.472, CI 0.316-0.707), while might not be able to reduce infection rate. The average CC rate was
4.86% in topical antibiotic prophylaxis group, versus 6.81% in control group. 51.28 patients needed to be treated to prevent 1
case of Baker grade III or IV CC. Cephalosporins were the most commonly preferred antibiotic regimen in included studies.
Nevertheless, there was no consensus of antibiotic prophylaxis combination or timing.
Conclusions Extended systematic antibiotic prophylaxis will significantly reduce SSI rate, especially in implant breast
reconstruction. Topical antibiotic irrigation would decrease CC rate, while might not be able to reduce infection rate.
Cephalosporins are generally recommended as antibiotic prophylactic regimen which cover the most commonly identified
implantassociated bacteria. Risk factors such as chest irradiation and diabetes should be take into consideration when
prescribing antibiotic prophylaxis. More better-designed RCTs are awaited to demonstrate the proper antibiotic regimen in
prosthesis-based breast surgery to reduce complications.

Title: Integrated immunologic assessment of tumor infiltrating lymphocytes (TILs) and peripheral blood to assess synergy of
cryoablation (cryo) plus ipilimumab (ipi) in early stage breast cancer (ESBC) patients (pts)
David Page1, Jianda Yuan2, Adi Diab6, Zhiwan Dong2, Arielle Ginsberg2, Phillip Wong2, Ryan Emerson4, David Redmond5, Brian
Blum1, Zhenyu Mu2, Chunjun Zhao2, Christopher Comstock1, Elizabeth Morris1, Elizabeth Comen1, Alan Kotin1, Janice Sung1, Edi
Brogi1, Monica Morrow1, Stephen Solomon1, Virgilio Sacchini1, Majid Maybody1, Deirdre Neville1, Harlan Robins3, Sujata Patil1,
Jedd Wolchok1, Clifford Hudis1, Larry Norton1, James Allison6, Padmanee Sharma6 and Heather McArthur1. 1Memorial Sloan
Kettering Cancer Center, New York, NY; 2Ludwig Institute for Cancer Immunotherapy at Memorial Sloan Kettering Cancer Center,
New York, NY; 3Fred Hutchinson Cancer Research Center, Seattle, WA; 4Adaptive Biotechnologies, Seattle, WA; 5Cornell
University, New York, NY and 6MD Anderson Cancer Center, Houston, TX.
Body: Background: In pts with ESBC, cryo combined with cytotoxic T-lymphocyte antigen 4 blockade was well tolerated and did
not delay standard-of-care mastectomy. As observed in mice, cryo+ipi may liberate tumor-associated antigens, synergistically
activate tumor-reactive T-cells, and confer long-term anti-tumor immunity. Because singular biomarkers of response to
immunotherapy have not been well defined, we conducted an integrated immunologic assessment to explore potential predictors
of immune activation and response.
Methods: Serial blood and pre-/post-treatment tumor tissue were collected from 18 pts treated with cryo (6 pts), single-dose ipi at
10mg/kg (6 pts), or cryo + ipi (6 pts). A Meso Scale Discovery platform was used to measure plasma cytokine interferon gamma
(IFN). Multiparameter flow cytometry was used to evaluate peripheral and intratumoral T-cell and myeloid cell quantity, T-cell
phenotype (effector versus regulatory), proliferation state (Ki67), and activation state (inducible costimulator [ICOS] expression).
Finally, a DNA deep sequencing platform was used to conduct T-cell repertoire analysis of peripheral T-cells and TILs.
Results: Sustained >2-fold elevations (1 month post-treatment) in plasma IFN were observed in the majority (4/6) of pts
receiving cryo/ipi (median 6-fold increase), but in the minority of pts receiving cryo (0/6, median 0-fold) or ipi (2/6, median 0-fold).
Similarly, sustained >2 fold elevations in ICOS expression in peripheral CD3+CD4+ T-cells, a known pharmacodynamic marker of
ipi, were observed in the majority (5/6) of pts receiving cryo/ipi (median 4-fold increase), but in the minority of pts receiving cryo
(0/6, median 0-fold) or ipi (2/6 ipi; median 1-fold). No trends were observed in peripheral myeloid derived suppressor cells.
Analysis of TILs by flow cytometry identified increased numbers of proliferating CD8+ T-cells (CD8+Ki67+) in ipi and cryo/ipi
groups relative to cryo alone; furthermore, the ratio of proliferating (CD8+Ki67+) to regulatory (CD4+CD25+FoxP3+) cells was
enhanced in the cryo/ipi group. Finally, analysis of T-cell repertoire in TILs demonstrated that cryo/ipi generated an influx of novel
T-cell clones, with select clones surging dramatically in predominance and circulating within the periphery.
Conclusions: Utilizing an integrated assessment, we identified evidence of immunologic synergy with combination cryo/ipi
versus either therapy alone. Of the tested parameters, peripheral CD4+ ICOS expression, plasma IFN, Ki67-gated TIL
effector/regulatory ratios, and clonal repertoire analysis were identified as promising biomarkers of immune activation. These
findings will inform a prospective assessment of potential immunologic biomarkers of immune response and clinical benefit in a
phase 2 study of cryo-immunotherapy in ESBC.

Title: HER2 discordant results in local vs. central testing in the phase 3 nelipepimut-S trial and implementation of Leica Bond
Oracle HER2 Immunohistochemistry (IHC) System for low and intermediate levels (1+, 2+) of HER2 protein expression as a
companion diagnostic
Michelle Melisko1, Elizabeth A Mittendorf2, Sufia Safina3, Michael Schenker4, Murray A Brunt5, Maria Litwiniuk6, John Mackey7,
Katarina Petrakova8, Svitiana Alieva9, Lacey Chance10, Gavin S Choy10, Mark Ahn10, Adamm Hamm11, Sonia Kumar12 and Hope S
Rugo1. 1University of California, San Francisco, CA; 2U.T. MD Anderson Cancer Center, Houston, TX; 3Republican Clinical
Oncology Center, Kazan, Republic of Tatarstan, Russian Federation; 4Sf Nectarie Oncology Center, Craiova, Romania;
5
University Hospital of North Staffordshire, Stoke on Trent, United Kingdom; 6Greater Poland Cancer Center, Poznan, Poland;
7
Cross Cancer Institute, Edmonton, AB, Canada; 8Masaryk Memorial Cancer Institute, Cancer Care Clinic, Brno, Czech Republic;
9
Donetsk Regional Antitumor Center, Donetsk, Ukraine; 10Galena Biopharma, Inc, Portland, OR; 11Aptiv Solutions, Durham, NC
and 12Leica Biosystems, Danvers, MA.
Body: Background : The distinction between HER2-positive (IHC 3+ or 2+ with FISH ratio >/= 2) and not overexpressing HER2
has been the focus of many diagnostic tests over the past years in association with development of HER2-targeted therapies. The
paucity of therapies developed for the low to intermediate HER2 protein expression populations has resulted in limited attention to
their diagnostic precision and accuracy. The development of NeuVaxTM (nelipepimut-S; Galena Biopharma, Inc.) in the defined
population requires a HER2 IHC 1+/2+ diagnostic that precisely and accurately ensures identification of targeted patients. We
describe discordance rates between local and central testing performed to identify tumors with HER2 IHC 1+/2+ expression that
supports the development of a method to validate HER2 1+ and 2+ (FISH < 2.2) patients who receive nelipepimut-S adjuvant
therapy.
Methods : The Prevention of Recurrence in Early Stage, Node-Positive Breast Cancer with Low to Intermediate HER2
Expression with NeuVaxTM Treatment (PRESENT) study, is a multicenter, multinational, prospective, randomized, double-blind,
controlled Phase 3 study assessing efficacy and safety of the peptide vaccine nelipepimut-S, in HLA A2 or A3 positive patients
with early stage, node positive breast cancer expressing low and intermediate levels (IHC 1+/2+) of HER2 protein. PRESENT
2-step screening includes HLA testing and central lab confirmation of HER2 1+ or 2+ expression using the DAKO HercepTest.
Results : As of 2 June 2014, 1454 patients underwent central IHC testing for HER2 and had a quantifiable result of 0, 1+, 2+, or
3+ for both local and central test. Per local testing, 61% (HER2 1+, n=612; HER2 2+, n=275) were eligible and 39% (HER2 0,
n=468; HER2 3+, n=99) were ineligible. Of those eligible by local testing, 67.5% (n=599) were confirmed as eligible per central
testing for a discordance rate of 32.5% (n=288). Of the 288 discordant samples tested centrally, 73.6% (n=212) and 26.4%
(n=76) were reported as HER2 0 and 3+, respectively. 8.7% (76/877) of patients found to be HER2 1+ or 2+ by local testing were
determined to be HER2+ (IHC3+) by central testing.
Conclusions : Current tests for HER2 expression are defined by their ability to determine 3+ positivity, yet significant
discordance still occurs with nearly 9% false negative rate in this trial. Similarly, marked discordance exists between local and
central laboratory test results for HER2 by IHC at 1+/2+ levels of expression. The relatively high discordance rate observed may
be due, in part, to the lack of a validated IHC assay for low-to-intermediate expression of HER2 (0, 1+, and 2+). In order to
improve accuracy of testing and to develop a companion diagnostic for nelipepimut-S, the Leica Bond Oracle HER2 IHC System
has been validated to determine samples across the IHC spectrum (0, 1+, 2+ and 3+) and is now incorporated into HER2
screening for the PRESENT trial as a companion diagnostic to increase accuracy, precision, and specificity in discerning HER2
1+ and 2+ patients.

Title: A Phase Ib Study of an adjuvant GM-CSF-Secreting Breast Cancer Vaccine
Karen S Anderson1, Beth Overmoyer2, Christine Canning2, Jennifer Savoie2, Garrick Wallstrom1, Eric P Winer2 and Glenn
Dranoff2. 1Biodesign Institute, Tempe, AZ and 2Dana-Farber Cancer Institute, Boston, MA.
Body: Background: Vaccination with tumor cells engineered to secrete granulocyte-macrophage colony stimulating factor
(GM-CSF) generates potent, specific, and long-lasting anti-tumor immunity in multiple tumor models. Here, we present analysis
on the safety, feasibility, and biologic activity of an autologous GM-CSF-secreting breast cancer vaccine in the high-risk adjuvant
setting.
Methods: Following IRB approval and patient consent, 18 patients with stage II-III breast cancer underwent tumor procurement for
vaccine development at the time of breast surgery. Patients were required to have at least 4 cm of primary tumor, or have
received neoadjuvant chemotherapy with at least 2 cm of residual tumor. 11 patients had insufficient tumor following neoadjuvant
chemotherapy. 7 patients had sufficient tumor harvested to produce and subsequently receive vaccine therapy. Breast cancer
cells were transduced with a replication defective adenoviral vector encoding GM-CSF and irradiated. Vaccinations were started
4-12 weeks after completion of all chemotherapy, immunotherapy, and radiation therapy. Vaccines were delivered
subcutaneously (sc) and intradermally (id) weekly for three weeks, then every other week for a total of 6 doses. Immune
monitoring included skin biopsies of vaccine sites, measurement of leukocyte populations, and proteomic-based assessment of
antibody responses.
Results: Tumor cell yields ranged from 0.07-31.6 x 106 cells. Dose levels were based on cellular yield, ranging from 105- 4 x106
cells/dose for the 7 patients with sufficient cell numbers. Vaccinated patients were 32-65 years of age, and all received 6 vaccines
total. Three patients developed relapse within one year of the start of vaccinations, one of whom died at 14 months. The
remaining four patients remained disease-free 23-34 months from start of vaccine. Toxicities related to treatment were mild and
included Grade I/II local injection-site reactions, as well as grade I/II fatigue, fever, upper respiratory symptoms, cough, and joint
pain. One episode of grade 3 fatigue was observed. Increases in antibody responses (p < .05) were observed for 17 antigens in
at least 4 out of 5 patients evaluated.
Conclusion: For larger tumors, breast cancer cells can be harvested from a subset of patients in sufficient number for autologous
vaccine production at the time of breast surgery. Autologous vaccination can induce immune responses with limited toxicity. The
proteomic-based identification of antigen-specific immune responses following vaccination will be presented.

Title: A phase 1/2 study of Ad.p53 DC vaccine with indoximod immunotherapy in metastatic breast cancer
Hatem H Soliman1, Susan E Minton1, Roohi Ismail-Khan1, Hyo S Han1, Nicholas N Vahanian2, Charles J Link2, Gene Kennedy2,
Howard Streicher3, Daniel Sullivan1 and Scott J Antonia1. 1Moffitt Cancer Center, Tampa, FL; 2NewLink Genetics Inc, Ames, IA
and 3National Cancer Institute, Bethesda, MA.
Body: Background: Indoleamine 2,3 dioxygenase (IDO) is a tryptophan-catabolizing enzyme that causes immunosuppression in
the tumor microenvironment. Indoximod is an IDO pathway inhibitor. Preclinical data suggests indoximod enhancs the activity of
dendritic cell (DC) vaccines. Ad.p53 is an adenovirus used to generate autologous dendritic cell (DC) vaccines against p53
epitopes. We initiated a phase 1/2a trial of indoximod + Ad.p53DC to explore the safety and efficacy of the combination along with
response to subsequent chemo. The phase 1 safety data were previously presented and the treatment was well tolerated with no
DLTs. (Soliman, ASCO 2013) This abstract includes new phase 2a safety/efficacy data and updated outcomes on all phase 1/2
metastatic breast cancer patients who received Adp53DC+indoximod and any subsequent response to salvage chemo.
Methods: The phase 2a study combined indoximod 1600mg PO BID with up to 6 Ad.p53 DC vaccinations q2wks. The trial used
a single arm, Simon two stage design (n=12 in 1st stage, 25 in 2nd stage) with objective response as the primary endpoint. One
response out of 12 was required for progression into second stage. The study had 90% power to detect 20% response rate with a
p=.09. Patients with measurable, metastatic breast cancer, <3 lines of chemo in metastatic setting, p53 IHC >5%, ECOG 0-2, no
autoimmune disease were eligible. Study treatment continued until disease progression or unacceptably toxicity.
Results: Twelve phase 2 patients were accrued, 9 (7 TNBC, 2 ER+/HER2-) received 1 dose of Ad.p53DC+indoximod (3 did not
due to rapid disease progression during vaccine preparation). Six patients had 1 prior line of chemo. Seven (58%) subjects
experienced any grade AE, there were no treatment related AEs G3. All treatment attributable AEs were G1-2, <10% frequency,
and included anemia, nausea, lymphopenia, photophobia, and headache. All discontinuations were due to disease progression.
Best response to immunotherapy in phase 2=1 SD, 8 PD. Phase 2 median TTP = 6.85 wks (3.8-18.1), OS = 18.1 wks (3.8-52)
with 3 patients alive as of June 2014. Five patients who received 2 cycles of chemotherapy after immunotherapy demonstrated
1 PR, 3 PD, 1 pending scan. For the entire phase 1/2 breast cohort (21 in phase 1, 9 in phase 2) the median TTP = 9.85 weeks
(3.8-22.1), OS=38.7 (3.8-122.1) weeks, with 1 patient from the phase 1 cohort alive as of June 2014. Ten out of 21 evaluable
patients (1 CR, 7 PR, 2 SD, 11 PD) had clinical benefit from chemotherapy after immunotherapy (9 patients had rapid decline and
did not get one full cycle of therapy). All but one of the responders had seen prior chemotherapy in the metastatic setting
(including a CR after 4 prior lines of chemo). Median OS in the chemo responders was 69.4 wks (30.1-122.1).
Conclusions: Indoximod+Ad.p53DC was well tolerated. Across phase 1/2 the best response to immunotherapy alone was SD in
4 pts while 10 of 21 (47%) (including 1 CR) responded to subsequent chemotherapy in this largely pretreated cohort. There may
be a chemosensitization effect of indoximod+Ad.p53DC. Future trials should combine this treatment with chemotherapy in
appropriately selected patients.

Title: Discovery of fully human monoclonal antibodies as therapeutic candidates for the treatment of HER2-negative breast
cancer
Christy Boozer1, Paul Algate1, Aurelio Bonavia1, Mark Branum1, Po-Ying Chan-Hui1, Alison Fitch1, Brad Greenfield1, Claire
Sutherland1 and Kristine Swiderek1. 1Theraclone Sciences, Seattle, WA.
Body: Unlike HER2-positive breast cancer, there are limited treatment options for patients diagnosed with triple negative and
endocrine resistant HER2-negative breast cancer, and as such HER2-negative breast cancer represents a significant unmet
medical need. The goal of this work is to use Theraclones proprietary I-STAR platform to mine the memory B cell immune
repertoire of breast cancer patients for the discovery of therapeutically relevant monoclonal antibodies (mAbs) and targets that
may be exploited as candidates for treatment of HER2-negative breast cancer.
Matched serum and PBMC samples were collected from breast cancer patients at multiple clinical sites. The selected patient
populations included, but were not limited to, patients who were treatment nave, those who had received immunotherapy or
adjuvant chemotherapy, and patients who had an exceptional clinical response to treatment and/or disease. Serum antibody
binding to a diverse panel of 5 well characterized breast cancer-derived cell lines of luminal and basal sub-types was determined.
Utilizing this approach, patients with a robust serological profile across multiple breast cancer cell lines were identified and
prioritized for memory B cell repertoire analysis via the I-STAR platform. Using a high throughput and miniaturized, multiplex flow
cytometry assay, the secreted IgG antibodies from over 85,000 individually enriched and expanded B cell clones were screened
for binding to the tumor cell line panel and a large number of positive B cell clones were identified. The screening hits can be
binned into several unique binding profiles, many of which were confirmed to be shared across multiple patient samples. Deep
sequence analysis of the variable regions of the antibodies produced by the B cell clones demonstrated that several of the
screening hits were derived from clonally related B cells; the majority of the screening hits represented antibodies derived from
unique B cell clones. A representative set of these antibodies was expressed recombinantly for further in vitro characterization.

Title: Bi-specific antibodies targeting signaling pathway crosstalk are a new breast cancer immunotherapeutic strategy
Yanliang Zhang1, Edwige Gros, Sarabjit Chagar, Jian Cao, Heyue Zhou1, Kouros Motamed, Gunnar F Kaufmann1 and Yanwen
Fu1. 1Sorrento Therapeutics, Inc, San Diego, CA.
Body: All of the currently approved therapeutic anti-cancer antibodies are monospecific and therefore only capable of interfering
with the biological function of a single molecular target. However, breast cancers mostly involve crosstalk of often synergistic
signal transduction pathways, and thus, isolated blockade of a single signal transduction pathway is frequently met by escape
mechanisms, such as upregulation of redundant pathways, rendering the monospecific immunotherapy less effective.
Using both chemical and molecular biology techniques, Sorrento has developed new approaches to generate IgG-like bi-specific
antibodies (BsAbs) targeting either two compensating signal transduction pathways, such as HER family members, or a breast
cancer specific antigen and an immuno-regulatory molecule such as PDL1 or PD1. The chemical biology method, which involves
specific hetero-dimerization of two half antibody molecules using bio-orthogonal chemistry, was used to generate an anti-c-Met
and anti-PDL1 chemical bi-specific antibody (CBA). Lastly, employing a molecular biology approach, an anti-EGFR and ErbB3
scFv-Fc bi-specific antibody was produced. Progress will be presented on in vitro characterization and cell-based functional
assays of these BsAbs.

Title: c-MET is a potential therapeutic target for antibody-drug conjugates in breast cancer
Yanwen Fu1, Edwige Gros1, Alice Lee1, Kimberly Johnson1, Hong Zhang1, Silpa Yalamanchili1, Kouros Motamed1, Gary Chen1,
Bryan Jones1, David Miao1 and Gunnar F Kaufmann1. 1Sorrento Therapeutics, Inc, San Diego, CA.
Body: The transmembrane receptor tyrosine kinase c-MET plays a key role in malignant transformation of epithelial cells by
activating signal transduction pathways essential for cellular proliferation, survival, migration and invasion. Over expression of
c-MET, with or without gene amplification, has been reported in primary breast cancers and correlates with poor prognosis.
Inhibition of c-MET signaling, via tyrosine kinase inhibitors (TKIs) or antagonistic antibodies, is usually not sufficient for sustained
treatment efficacy, thus, we believe that antibody drug conjugates (ADCs) offer the promise and potential of delivering more
potent anti-tumor activity. We have generated ADCs containing a proprietary human anti-c-MET antibody (STI-D0606) with either
a tubulin inhibitor or a DNA damaging agent. STI-D0606, a fully human antibody (IgG1) selected from Sorrento's G-MAB antibody
library, was conjugated with a cytotoxin via site-specific bioconjugation. The conjugates retained binding affinity and showed
potent cell killing in a variety of c-MET-positive cell lines. Progress will also be reported on overall efficacy of c-MET ADCs in a
preclinical xenograft model.

Title: Overall survival (OS) in the IMELDA randomized phase III trial of maintenance bevacizumab (BEV) with or without
capecitabine (CAP) for HER2-negative metastatic breast cancer (mBC)
Joseph Gligorov1, Jose Bines2, Emilio Alba3, Giorgio Mustacchi4, Saverio Cinieri5, Vineet Gupta6, Jean-Yves Pierga7, Hakan
Bozcuk8, Rabab Gaafar9, Sudeep Gupta10, Guillermo Lopez Vivanco11, Xiaojia Wang12, Romulo Costa13, Kadri Altundag14, Ewa
Chmielowska15, Sabine de Ducla16, Ulrich Freudensprung16, Paulo Cortes17 and Dinesh Doval18. 1APHP Tenon, IUC-UPMC, Paris,
France; 2Instituto Nacional de Cancer, Rio de Janeiro, Brazil; 3Hospital University Clinic Virgen de la Victoria, Mlaga, Spain;
4
Medical Oncology, University of Trieste, Trieste, Italy; 5Medical Oncology Department and Breast Unit, Brindisi, Italy; 6Bangalore
Institute of Oncology, Bangalore, India; 7Institut Curie, Universit Paris Descartes, Paris, France; 8Akdeniz University Medical
Faculty, Antalya, Turkey; 9National Cancer Institute, Cairo University, Cairo, Egypt; 10Tata Memorial Hospital/Center, Mumbai,
India; 11Hospital Universitario Cruces, Vizcaya, Spain; 12Zhejiang Cancer Hospital, Hangzhou City, China; 13Instituto do Cncer do
Estado de So Paulo, So Paulo, Brazil; 14Hacettepe University Cancer Institute, Ankara, Turkey; 15Centrum Onkologii Prof. F
Lukaszczyka, Bydgoszcz, Poland; 16F. Hoffmann-La Roche Ltd, Basel, Switzerland; 17University Hospital of Santa Maria, Lisbon,
Portugal and 18Rajiv Gandhi Cancer Institute & Research Center, Delhi, India.
Body: BACKGROUND The open-label randomized phase III IMELDA trial demonstrated that adding CAP to maintenance BEV
until disease progression (PD) after initial BEVdocetaxel (DOC) provides statistically significant and clinically meaningful
improvements in both progression-free survival (PFS [primary endpoint]; hazard ratio [HR] 0.38 [95% CI 0.270.55]; log-rank
p<0.001) and OS. We present OS in subgroups representing stratification factors and clinically important populations.
METHODS Patients (pts) with HER2-negative measurable mBC, ECOG PS <2, and no prior chemotherapy for mBC were
eligible. After 36 cycles of BEVDOC, pts without PD were randomized to BEV alone or BEVCAP (BEV 15 mg/kg q3w; CAP
1000 mg/m2 bid d114 q3w) until PD. Stratification factors were estrogen receptor (ER) status, visceral metastases, response
status, and lactate dehydrogenase (LDH) concentration. OS from randomization was a secondary endpoint. The planned sample
size of 360 enrolled pts (290 randomized) was calculated assuming a PFS HR of 0.70 (median PFS 5.88.3 months) with 80%
power at 2-sided =0.05 after 244 PFS events. Recruitment was stopped prematurely after regulatory withdrawal of the
BEVDOC combination but pts who had already been enrolled and randomized were followed as originally planned.
RESULTS Between Jun 2009 and Mar 2011, 284 pts were enrolled and treated. Of these, 99 were not eligible for randomization
(most commonly due to PD [41%] or AEs/toxicity [31%]) and 185 (65%) were randomized. At the time of the primary PFS
analysis, representing study closure, median follow-up (from randomization) was 31.6 months. Median OS from randomization
was 23.7 months in the BEV arm and 39.0 months in the BEV CAP arm (events in 36% of pts). The HR for OS in the two
randomized arms showed consistency between subgroups, favoring the BEV CAP arm in all subgroups analyzed.
Subgroup
No. of events/No. of pts (%)

BEV
Unstratified HR (95% CI)
BEVCAP
(0.26-0.69)a
1-y OS rate (%)

BEV
BEVCAP
72
90
All
53/94 (56)
33/91 (36)
0.43
<65 y
46/81 (57)
27/77 (35)
0.51 (0.32-0.82)
72
93
65 y
7/13 (54)
6/14 (43)
0.50 (0.16-1.60)
68
79
Triple negative
16/21 (76)
10/25 (40)
0.44 (0.19-0.99)
62
90
Hormone receptor positive
37/73 (51)
23/66 (35)
0.53 (0.31-0.89)
75
91
ER
positiveb
36/69 (52)
23/64 (364)
0.53 (0.32-0.90)
75
90
ER
negativeb
17/25 (68)
10/27 (37)
0.44 (0.20-0.99)
64
91
<3 metastatic organ sites
17/40 (43)
17/48 (35)
0.75 (0.38-1.49)
81
93
3 metastatic organ sites
36/54 (67)
16/43 (37)
0.39 (0.22-0.71)
65
88
Visceral metastasesb
38/65 (58)
23/62 (37)
0.43 (0.26-0.73)
70
92
No visceral metastasesb
Complete or partial
Stable
responseb
diseaseb
15/29 (52)
10/29 (34)
0.76 (0.34-1.70)
76
88
36/68 (53)
24/68 (35)
0.61 (0.37-1.03)
73
89
14/22 (64)
6/20 (30)
0.22 (0.08-0.63)
68
100
Non-measurableb
3/4 (75)
3/3 (100)
0.30 (0.03-2.98)
67
67
LDH 1.5ULNb
50/89 (56)
30/85 (35)
0.49 (0.31-0.76)
72
94
3/5 (60)
3/6 (50)
1.01 (0.20-5.00)
60
44
LDH >1.5ULNb
aStratified
analysis.
bStratification
factor.
CONCLUSIONS. Combining maintenance BEV with CAP until PD after initial BEVDOC for mBC provides a statistically
significant and clinically meaningful improvement in OS (secondary endpoint), seen consistently irrespective of baseline
characteristics.

Title: A phase II study of bevacizumab in combination with modified FOLFOX6 in heavily pretreated patients with HER2-negative
metastatic breast cancer
Ting Li1,2, Biyun Wang1,2, Xichun Hu1,2, Zhonghua Wang1,2, Jian Zhang1,2, Si Sun1,2, Jun Cao1,2, Fangfang Lv1,2, Leiping Wang1,2,
Sheng Zhang1,2, Chen Ni1,2, Zhenhua Wu1,2 and Jie Xie1,2. 1Fudan University Shanghai Cancer Center, Shanghai, China and
2
Shanghai Medical College, Fudan University, Shanghai, China.
Body: Backgroud
Bevacizumab combined with modified FOLFOX6 (mFOLFOX6) is a standard regimen for colorectal cancer with good tolerability.
Our previously published study by Sun et al showed a moderate efficacy and excellent safety of mFOLFOX6 for metastatic breast
cancer (MBC). This study was to determine the efficacy and safety of adding bevacizumab to mFOLFOX6 in heavily pretreated
patients with human epidermal growth factor receptor 2 (HER2)-negative MBC.
In this open label, single-arm phase II study, bevacizumab, 5 mg/kg every two weeks or 7.5 mg/kg every three weeks, in
combination with mFOLFOX6, which were oxaliplatin 85mg/m2, leucovorin 400mg/m2 and 5-Fu 400mg/m2 intravenously on day
1 following 5-Fu 2400 mg/m2 continuous intravenously 46 hours every 2 weeks, was administered to patients who failed at least 2
prior chemotherapy regimens in metastatic setting. The prior exposure to taxane, anthracycline, vinorelbine, capecitabine and
gemcitabine was 97.9%, 93.8%, 64.6%, 64.6% and 81.3%, respectively. The primary objective was progression free survival
(PFS), and secondary objectives included objective response rate (ORR), overall survival (OS) and safety.
Results
48 patients were enrolled with a median of 49.5 years old (range, 34 to 73 years old) and a median of 3 prior chemotherapy
regimens (range, 2 to 6). A median of 4.0 cycles (range, 0.5 to 8.0 cycles) were delivered with 45 patients undergoing treatment
discontinuation, including 7 (14.6%) due to completion of 6 cycles, 1 (2.1%) due to completion of 8 cycles, 18 (37.5%) due to
disease progression, 5 (10.4%) due to adverse events, 4 (8.3%) due to withdrawal of informed consent, 2 (4.2%) due to
physicians decision, 2 (4.2%) due to economic reasons and 6 (12.5%) with unknown causes. The median PFS was 6.0 months
(95% confidence interval [CI], 3.6 to 8.5 months), ORR was 48.8% and median OS was 10.2 months (95% CI, 8.5 to 11.9
months). Most adverse events were grade 1 or 2 and grade 3 or 4 toxicities occurring in more than one patient were neutropenia
(75.0%), leukopenia (50.0%), thrombocytopenia (20.8%) and anemia (6.3%).
Conclusion
Adding bevacizumab to mFOLFOX6 has significant anti-tumor activity and excellent safety in heavily pretreated HER2-negative
MBC patients, which warrants conduct of further confirmative trials.
Adverse events compared with our historical mFOLFOX6 trial
Sun et al
Li et al
Toxicity
Grade 3-4, n (%)
Grade 3-4, n (%)
Anemia
0 (0)
3 (6.3)
Neutropenia
14 (22.6)
36 (75.0)
Thrombocytopenia
9 (14.5)
10 (20.8)
Nausea/vomiting
1 (1.6)
0 (0)
Diarrhea
1 (1.6)
0 (0)
Neuropathy
3 (4.8)
0 (0)
<0.05
Chemotherapy as the third line or beyond treatment for metastatic breast cancer
Number of
Patients
ORR
(%)
median PFS
(months)
median OS
(months)
Bevacizumab+mFOLFOX6 4, 3-7
48
48.8
6.0
10.2
Aogi et al
Eribulin
21
14.3
1.9
7.3
Cortes et al
Eribulin
5, 3-6
269
14.1
2.6
10.4
Rha et al
Gemcitabine
12
17
Smorenburg et
al
Gemcitabine
23
1.9 (median TTP)
7.8
Udom et al
Vinorelbine
3, 3-5
20
35
2.75 (median TTP) NA
Rivera et al
Paclitaxel
35
20
NA
NA
Rivera et al
Paclitaxel
33
18
NA
NA
Abrams et al
Paclitaxel
172
23
NA
NA
Study
Regimens
Li et al
Median Line,
range
NA=Not Available, TTP=Time To Progression


Title: Characteristics and outcome of breast cancer patients enrolled in cancer therapy evaluation program (CTEP) sponsored
phase I clinical trials
Filipa Lynce1, Larry Rubinstein2 and Pamela Harris2. 1Washington Cancer Institute, MedStar Washington Hospital Center,
Washington, DC and 2Cancer Therapeutic Evaluation Program, National Cancer Institute, Rockville, MD.
Body: Background: Breast cancer is the most commonly diagnosed cancer and the second leading cause of cancer-related
death among women. Given the availability of approved therapies and large clinical trials, historically few breast cancer patients
are referred for consideration of a phase I trial. Although efficacy is not the endpoint of phase I studies, we were interested in
determining whether clinical benefit rates differed in patients with breast cancer from patients with other cancers enrolled in phase
I solid tumor trials. In addition a better knowledge of the characteristics and outcomes of patients with breast cancer enrolled in
phase I clinical trials may contribute to better clinical trial design and patient selection.
Methods: We performed a retrospective analysis of all Cancer Therapy Evaluation Program (CTEP) sponsored phase I trials for
patients with solid tumors, from 1992 to 2012. CTEP phase I database which is maintained by Theradex Inc was queried. Patients
treated on phase I/II trials or including hematological malignancies were excluded. We conducted an analysis of demographic
variables, variables related to disease characteristics and outcomes of patients with breast cancer and compared them to patients
with other oncological diagnoses enrolled in the same trials.
Results: A total of 8119 patients who participated in 225 trials were identified. Of these 1367 (16.8%) patients had breast cancer.
Compared with patients with other oncological diagnoses, breast cancer patients were older (51.9 vs. 45.4 years), less likely to be
white (73% vs. 81.6%, p<0.001) and more likely to have received a high number of previous lines of therapy (34.2% of patients
with breast cancer had received 6 or more prior regimens vs. 11.9% of patients with other cancers, p<0.001). Performance status
(PS) was similar between the two groups of patients with the majority of patients presenting with PS of 1 (57.3% in the breast
cancer group and 60.9% in the non-breast cancer group). The clinical benefit rate was higher for patients with breast cancer
(56.1% vs. 42%, p<0.001). Breast cancer patients remained on study longer than non-breast cancer patients (mean 136.3 vs.
95.8 days, p<0.001). The toxicity related deaths were less than 1% in both groups. Discontinuation of treatment due to toxicity
was observed in 6.6% of patients with breast cancer and 5.6% of patients with other cancers.
Conclusions: An analysis of the CTEP-sponsored phase I trials revealed that breast cancer patients, enrolled in these trials, were
older and more heavily pre-treated than patients with other types of cancer. However they seem to have better outcomes and
similar toxicities when compared to the other patients. These data suggest that patients with breast cancer who have likely
exhausted treatment with approved agents should be considered for phase I trials.

Title: Potential benefits of hypnosis sedation on different modalities of breast cancer treatment
Martine Berliere1, Sarah Lamerant1, Philippe Piette2, Aurore Lafosse1, Laurence Delle Vigne1, Fabienne Roelants1, Christine
Watremez1, Marie-Agnes Docquier1, Lafita Fellah1, Isabelle Leconte1 and Franois Duhoux1. 1Cliniques Universitaires St Luc
(UCL), Brussels, Belgium and 2Grand Hopital de Charleroi (GHdC), Charleroi, Belgium.
Body: Background: In oncology, hypnosis has been used for pain relief in metastatic patients but rarely for induction of
anaesthesia.
Material and methods: Between January 2010 and February 2014, 220 patients from our breast clinic (Cancer Center Cliniques universitaires Saint-Luc - Universite catholique de Louvain) were included in an observational, non randomized study
approved by our local ethics committee. 110 consecutive patients underwent breast surgery (lumpectomy or mastectomy +/axillary lymph node dissection or sentinel lymph node biopsy) while on general anaesthesia (group I) and 110 consecutive
patients underwent the same surgical procedures while on hypnosis sedation (group II).
The stages and the tumor characteristics were well balanced between the two groups. After surgery, 28 patients received
chemotherapy in group I and 27 patients in group II. Radiotherapy was administered to 96 patients of group I and 95 patients of
group II. Currently, 83 patients of group I and 82 patients of group II are receiving endocrine therapy.
Different parameters were studied for each treatment modality.
Results: Duration of hospitalization was statistically significantly reduced in group II vs. group I (3.3 days vs. 4.4 days) (CI 95%
range: -1.48 -0,72, p=0,0000000578) for all surgical procedures. The same results were observed for mastectomies alone (3.1 vs.
5.3 days) (CI 95 % range: -3,19 -1,31, p=0,0002 ) and for lumpectomies (3.1 vs. 4.3 days) (CI 95 % range: -1,024 -0,364,
p=0,00065). The number of post-mastectomy lymph punctures was reduced in group II (1 to 3 (median value n=1.6) vs. group I (2
to 5 (median value n=3.1, p=0.01), as was the quantity of lymph removed (103 ml versus 462.7 ml) (p=0,0297) in the group of
mastectomies.
Concerning chemotherapy, the incidence of asthenia was statistically decreased (p=0.015) in group II. There was a statistically
non significant trend towards a decrease in the incidence of nausea/vomiting and muscle pain in group II (respectively p=0.1 and
p=0.2).
The frequencey of severe radiodermitis (p=0.01) and post-radiotherapy asthenia (p=0.01) were significantly reduced in group II.
Finally, compliance to endocrine therapy was improved in group II (p=0.05), while incidence of hot flashes (p=0,00029), joint or
muscle pain (p=0,000139) and asthenia (p=0,00002) were statistically significantly decreased in group II.
Discussion: Hypnosis sedation exerts beneficial effects on nearly all modalities of breast cancer treatment. The absence of a
significant benefit for chemotherapy-induced nausea/vomiting and muscle pain observed is probably due to the small number of
patients receiving chemotherapy in our study.
Conclusion: Benefits of hypnosis sedation on breast cancer treatment are very encouraging and further promote the concept of
integrative oncology.

Title: Differential prognosis of pulmonary lumpectomy in patients with a history of curative breast cancer
Zhebin Liu1, Yihua Sun2, Fangjing Ma3, Hong Ling1, Keda Yu1, Guangyu Liu1, Genhong Di1, Zhen Hu1, Canmin Chen1, Yifeng
Hou1, Jiong Wu1, Zhenzhou Shen1 and Zhimin Shao1. 1Breast Cancer Institute, Shanghai Cancer Center Fudan University;
Shanghai Medical College, Fudan University, Shanghai, China; 2Fudan University Shanghai Cancer Center, Shanghai, China and
3
First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang Uygur Autonomous Region, China.
Body: Objective: Treatment of patients with a history of breast cancer who are diagnosed as newly found solid pulmonary
nodule is still controversial. Our study is to find out whether the pulmonary nodule represents a primary cancer, metastasis, or
benign lesion secondary to breast cancer can be differentially prognostic.
Methods: A total of 165 consecutive patients who underwent surgery or biopsy for pulmonary nodules between 2007 and 2013
after curative operation for breast cancer were reviewed. The postoperative survival rate was analyzed and the difference in
survival rate was assessed between groups of pulmonary nodules representing a primary cancer, metastasis, and benign lesion.
Results: Among these 165 patients, median age was 59 and mean follow-up was 51 months (range from 5 to 398). The
pathologic diagnoses of pulmonary nodules were pulmonary metastases of breast cancer in 71 patients (43%), primary lung
cancer in 59 (35.8%), and other diagnoses in 35 (21.2%) (hyperplasia in 11; pneumonia in 9; granuloma in 6; sclerosing
hemangioma and pulmonary fibrosis in 4 each). In those who were diagnosed as malignant pulmonary tumor, about one half
(56.4%) were secondary malignant and the rest were metastatic. There was no statistically difference in median age or disease
free interval between operation date of breast surgery and pulmonary surgery or biopsy among these groups. However, those
who were diagnosed as primary malignant or benign lesion had significantly higher incidence of isolated pulmonary nodule shown
on their CT scan results (P=0.036). The incidence of positive lymph node involvement and negative ER/PR status was
significantly higher in metastatic breast cancer patients (63.8% and 70.6%, respectively). Among patients with metastatic
pulmonary nodules, rates of discordance between primary breast tumor and metastasis foci were 20.3%, 23.7% and 14.7% for
ER, PR and HER2, respectively, which changed a lot the regimen used for treatment of metastatic breast cancer. Both the 5-year
and 10-year disease-free survival from the initial mastectomy were significantly shorter in metastatic breast cancer patients
(72.9%, 35.3%) when compared with patients with primary pulmonary cancer (82.9%, 53.8%) or pulmonary benign lesions
(91.6%, 67.8%). Prognostic factors in metastatic breast cancer patients were a disease-free interval of >60 months with 10-year
survival of 48.3% (P<0.05), solitary lung metastasis is associated with a survival rate of 78.9% after 5 years and of 48.7% after 5
and 10 years, and this is statistically significant compared to multiple metastases.
Conclusions: We think that pulmonary lumpectomy is the best option in selected cases of solitary or multiple pulmonary nodules
from breast cancer, which can be useful for differential diagnosis, predicting prognosis and deciding the drug treatment strategy in
some cases.

Title: Isolated loco-regional recurrence in the breast: Re-quadrantectomy and systemic treatment
Victoria Costanzo1, Veronica Fabiano1, Mercedes Maino1, Federico Colo1, Reinaldo Chacon1, Adrian Nervo1, Jorge Nadal1, Martin
Loza1, Jose Loza1, Daniel Mysler1 and Mora Amat1. 1Instituto Alexander Fleming, Buenos Aires, Argentina.
Body: Background: Local or regional recurrence in breast cancer is associated with poor prognosis. Mastectomy is the classic
indication in these situations. However, second attempts of breast conservation in previously radiated patients with small,
histologically favorable local relapse have been reported. Also, prospective data (The CALOR Trial) suggest that adjuvant
chemotherapy should be recommended in patients with isolated loco-regional recurrence, especially if the recurrence is HR
negative. The purpose of our study was to determine the frequency of use of re-quadrantectomy and systemic treatment in
isolated loco-regional recurrence (ILR) in breast cancer.
Methods: Retrospective review of the database of our institution. Inclusion criteria: ILR in surgical bed, breast or lymph nodes with
complete resection (mastectomy, re-quadrantectomy or axillary resection) with or without systemic treatment and radiotherapy.
Findings: 4695 patients were analyzed from april 2000 to april 2014. 78 patients had a ILR, 66 were analyzed. Median age 49,5
years (27-86). 54,5% of the patients were postmenopausal and 16,6% had bilateral involvement at diagnosis. Characteristics of
the primary tumor: Mean tumor size 2 cm, 66,6% were HR positive, 6% HER2 positive, 13,6% TN and 14% unknown. Initial
treatment: breast conserving surgery 62 p (sentinel node 25, axillary node disection 36); mastectomy 4 patients. 42 patients
underwent adjuvant chemotherapy, 57 patients radiotherapy and 62 patients hormonotherapy. Time to recurrence: 94,3 months.
Characteristics of the ILR: Mean tumor size 1.8 cm. 65% were HR positive, 12% triple negative, 6% HER positive and 17%
unknown. Treatment at relapse was as follows: 27 re-quadrantectomy, 34 mastectomy, 1 lymph node dissection; all these
patients had previous quadrantectomy. The other patients underwent to mastectomy (1) and complete resection of the lesion (3).
29 patients received chemotherapy (43.9%), : 70% anthracyclines and taxanes, 6% trastuzumab, 30% other and 65% received
hormonotherapy. Local progression free survival 30,7 months; distance progression free survival 100 months. Local relapse was
evidenced in 30% of patients who had a re-quadrantectomy procedure.
Conclusions: Second attemp of breast conservation in previously radiated patients with small, histologically favorable ILR and
long recurrence free interval disease is feasible as described. The target population is of good prognosis as evidenced the long
free disease interval at recurrence (94,3 m), the mean size of tumor at recurrence (1.8 cm) and the high proportion of HR positive
disease. The percentage of local recurrence after re quadrantectomy is as described in some reports, 30% (20-35%). The
indication of chemotherapy was mostly in HR negative or HER 2 positive tumors, but the low number of patients does not permit a
valid comparaison.

Title: Low dose capecitabine is effective and relatively nontoxic in breast cancer treatment
John Carpenter1. 1University of Alabama, Birmingham, AL.
Body: Capecitabine has been widely used in treatment of both early and advanced breast cancer in the United States since
1998. The dose limiting toxicity is diarrhea, which is dose limiting and can be severe. Lower doses are widely used without any
apparent compromise in efficacy and are consistently less toxic (see Naughton, Clinical Breast Cancer 2010). Here a continuous
dose regimen of capecitabine 1000 mg. daily was used in 24 patients with breast cancer, both early and advanced, from March
2012-September 2013. Four received the drug as adjuvant treatment, 8 as preoperative therapy, and 12 for advanced disease.
Duration of treatment ranged from 0.5 to 15 months. Toxicity was modest: 1 had moderate nausea, none experienced vomiting, 2
mild anemia (one with marrow replacement by tumor), 2 mild palmar irritation, and 1 grade 1 thrombopenia. No diarrhea was
seen. One patient died of hepatic necrosis and portal vein thrombosis; the relationship to treatment was uncertain. 12/20 with
measurable disease experienced a response to treatment (1 complete and 11 partial responses by RECIST criteria), 1 had stable
metastatic disease for 9 months, and 7 had progression of disease. The median duration of response was 5 months (range 2-15
months). Use of capecitabine in this dose and schedule for treatment of breast cancer appears comparably effective to higher
conventional doses and dramatically less toxic.

Title: Can synthetic 2D mammography be used to select patients in whom there is no need to review the digital breast
tomosynthesis from which they were constructed? A review of 1871 consecutive mammogram sets of patients presenting
symptomatically or for follow up
Simon DH Holt1, Ali Moalla1, Helen R Williams1, Khaldoun MY Nadi1, Anita M Huws1, Amrita Gurung1, Daniel Thomas1 and Yousef
M Sharaiha1. 1Peony Breast Care Unit, Prince Philip Hospital, Llanelli, Carmarthenshire, United Kingdom.
Body: Introduction: A synthetic 2D mammogram can be created by combining the individual optimally enhanced 1mm slices of a
digital breast tomosynthesis (DBT). Theoretically this could help overcome the problems associated with standard 2D
mammograms (superimposition of structures hiding small cancers) and the significantly increased reading time and decreased
conspicuity of microcalcifications associated with DBT. It may avoid the need for the double x-ray exposure required for
combination 2D and DBT currently being suggested to maximise the specificity and sensitivity of mammography. The hypothesis
we wished to test is, "if the synthetic 2D is normal, is there any advantage to looking at the DBTs as well?"
Method: We have reviewed 1000 consecutive cases presenting symptomatically or to our follow up clinic all of which underwent
DBT on a Hologic Dimensions machine. From the 3D data sets synthetic 2D mammograms were constructed (Hologic C-view
technology). One breast radiologist with 12 years experience of interpreting mammograms and more recently 4 years experience
in interpreting DBTs was asked to first review the 2D synthetic mammograms (each breast CC and MLO) and report them before
then reviewing the DBTs and issuing a final report. The mammograms were reported M1 to M5 using the standard BIRADs
criteria. The BIRADs scores for each breast were recorded prospectively and entered into a database.
Results: 1000 consecutive patients were studied between October 2013 and March 2014. The average age of the women was
58.1 years (range 29 to 92). Of these some were under follow up after mastectomy so in total there were 1871 individual
mammogram sets reported. Table 1 summarises the correlation between C-view and DBT reporting. The correlation between the
two modalities is very close, but importantly there was only one patient in whom the C-view was reported normal or benign (M1 or
M2) but the DBT reported a possible abnormality (M3). However, 31 cases reported as suspicious or malignant M3, 4 or 5 by
C-view were subsequently downgraded to benign after review of the DBT.
BIRADS Classification - Synthetic 2D v DBT
Synthetic 2D
M1
DBT
M2
M3
M4
M5
M1
610
M2
1096
10
21
M3
34
M4
21
M5
47
Table 1
Conclusion: In a symptomatic and follow up clinic, our study suggests that much radiologists time and x-ray exposure to the
patient could be saved by using synthetic 2D mammograms derived from the DBT data rather than using separate 2D studies.
Only if the C-view is reported M3, 4 or 5 is it necessary to review the DBT but for all these patients the DBT is already available
without further x-ray exposure or recall.

Title: Clinical, anatomical and histological characteristics of breast lesions visualized by photoacoustic mammography; first
clinical study in CK project
Elham Fakhrejahani1, Masae Torii1, Yasufumi Asao1, 3, Iku Yamaga1, Toshiyuki Kitai2, Masako Kataoka1, Shotaro Kanao1,
Masahiro Takada1, Tsuyoshi Shiina1 and Masakazu Toi1. 1Kyoto University, Kyoto, Japan; 2Kishiwada City Hospital, Kishiwada,
Osaka, Japan and 3Canon Inc, Tokyo, Japan.
Body: Background: Photoacoustic mammography (PAM) is an optical imaging technique potentially capable of imaging breast
vasculature as well as measuring hemoglobin oxygen saturation(SO2) in focal breast lesions. We presented the initial observation
on first 20 cases using the first generation PAM (PAM-01) prototype made by Canon Inc.(Tokyo, Japan) (spatial resolution:2mm)
in SABCS 2013. Here we report the clinico-pathological characteristics of all cases recruited in the first clinical study in CK project
(Kyoto University/Canon joint research project) between Aug 2010 and March 2012.
Methods: 57 patients were recruited in IRB approved study at Kyoto University Hospital, Japan. Forty-two breast harboring
lesions and when possible contralateral breasts were evaluated by PAM-01. Axial maximum intensity projection (MIP)s were
obtained and signals from consecutive MIPs confirmed to be associated with the tumor location in MRI by an expert breast
radiologist were considered to be the region of interest (ROI).The same depth was used for ROI in the normal breast as control.
Histological sections from the widest area of the lesions were evaluated post-excisional by immunohistochemistry using
anti-CD31 as endothelial marker and anti-carbonic anhydrase IX (CA IX) as hypoxia marker. Histological slides were scanned
and divided into 1680 squares (0.84x0.84 mm2)(Hamamatsu Inc. Japan) for image analysis. Total vascular perimeter (TVP)-in
mm- was calculated for all the histological section by using Image Pro-Plus 7.0 software (Media Cybernetics, USA). Tumor area
was measured in mm2. TVP index was calculated as TVP/area.
Results: Photoacoustic signal was detected in 30 lesions out of 42 at the depth of 26.8 12.8 mm from which 80% were located
superior to nipple. CA IX positive cases in comparison with CA IX negative cases significantly showed higher TVP index (p-value
=0.028 Mann-Whitney Test ) suggesting more angiogenic profile of hypoxic tumors. However, lesions without any detectable
signal were reported to have only a bigger mass size histologically (26.6 vs. 14.8 mm, p-value 0.28, Mann-Whitney Test)
regardless of their TVP index or CA IX expression level. Moreover,SO2 was calculated 70.9% for signals located inside tumors
and 85.5% for signals associated with subcutaneous vessels in the same breast (p-value <0.0001 Wilcoxon Test) compatible with
the SO2 data from normal counterpart at tumor depth in the other breast (81.9%) (p-value 0.0001 Wilcoxon Test). Tumors bigger
than 2 cm also tended to have lower SO2 (65.6%) compared with tumors smaller than 2 cm (70.1%) in accordance with higher
TVP index in bigger tumors (p-value <0.001 Student T-test).
Conclusion: This is the largest clinical study of PAM till today and the correlation between histological profile of hypoxia and tumor
microvasculature with PAM signal visibility as well as tissue SO2 seems promising. However, the improvement of techniques and
resolution is necessary to develop a more clinically applicable non-invasive functional breast imaging modality for analyzing
breast tumor vasculature and hypoxia. Some of these improvements have taken place for the second generation PAM (PAM-02)
which is now under a clinical evaluation study.

Title: Are more frequent early follow up mammogram protocols necessary after breast conserving surgery and radiation therapy?
Chavi Kaufman1, Tamara Fulop1, Susan K Boolbol1, David Lucido1, Suzan Naam1, Alyssa Gillego1 and Manjeet Chadha1. 1Beth
Isreal Medical Center and Roosevelt Hospital, NYC.
Body: Background: In an era of choosing wisely in healthcare there are many initiatives in radiology and oncology that are being
evaluated for appropriateness of practice based on clinical evidence-base. More frequent follow up mammogram protocols in the
first 3 years after BCT are a widely accepted practice. However, such breast imaging schedules have no strong clinical
evidence-base to support the added testing every 6-months, nor a rationale to justify patient anxiety and added unnecessary
biopsy procedures. The goal of our study was to evaluate the frequency of BIRADS score 4 on short follow up mammograms in a
population of patients treated with breast conserving surgery and RT (BCT).
Materials and Methods: This is an IRB approved study. From 2001- 2007, we identified 681 patients who underwent BCT and
who also underwent follow up mammograms at our cancer center. We reviewed short follow up mammograms defined as those
obtained within the first 3-years after BCT. The BIRADS score was tabulated in all cases. Further, it was determined to study the
frequency of BIRADS 4 score only, because this was deemed a clinically significant finding that routinely warranted additional
evaluation.
Results: Median age of the study group is 51 years (31- 80 years). Among the 681 patients a total of 3648 follow up mammogram
sessions were obtained. The median number of follow up mammogram sessions per patient during the observation period of
3-years was 6 (range 2-6 mammogram sessions). In 85% of patients followed the mammogram scores were BIRADS 1 to 3. In
15% of the patients the BIRADS score 4 was reported at least once. Among the BIRAD 4 category of patients, 56% had this
score reported in the ipsilateral breast following BCT for breast cancer, and 44 % had BIRADS 4 reported in the contralateral
breast. Specifically, in the ipsilateral breast the frequency of BIRADS 4 was 8.2%. We also observed a significant trend with a
higher frequency BIRADS 4 reported within the first year of follow up in the ipsilateral breast as compared with the contralateral
breast. In the 2nd and 3rd year the frequency of BIRADS 4 was < 10% and comparable between the treated ipsilateral breast and
the normal contralateral breast. The pathology correlations showed a significantly lower yield of cancer in the BIRADS 4 subset of
patients.
Conclusions: Similar to the observations of the ACRIN 6666 study 1 on BIRADS 3 we observed a very low yield of cancer from
frequent follow up protocol for the ipsilateral breast using the BIRADS 4 category. In promoting responsible medical care it is
important to establish appropriate follow up guidelines and selection of schedules for groups of patients individualized by risk.
Annual follow up mammograms might be adequate frequency for the ipsilateral breast. This study warrants further evaluation and
standardized protocols.
1 ACRIN 6666 study, Radiology 2013
This study is funded by the Ellen Blair Grant.

Title: Mammographic changes after oncoplastic reduction mammoplasty
Anne Warren Peled1, Merisa Piper1, Laura J Esserman1, Robert D Foster1, Hani Sbitany1 and Elissa R Price1. 1University of
California, San Francisco, CA.
Body: BACKGROUND:
Reconstruction of partial lumpectomy defects with reduction mammaplasty techniques can improve aesthetic outcomes.
However, the impact of the significant tissue rearrangement on post-operative mammographic findings and subsequent
recommendations for biopsy has not been well-studied.
METHODS:
A retrospective review of 50 patients who underwent partial mastectomy with immediate oncoplastic reduction mammaplasty
reconstruction from 2001 to 2008 was performed. Mammography reports at 6 months, 1 year, 2 years, and 3 years
post-operatively were reviewed for Breast Imaging Reporting and Data System (BI-RADS) scores, predominant findings, and
recommendations for subsequent imaging or biopsy.
RESULTS:
At six months post-operatively, 49 patients (98%) had benign findings of post-surgical changes, while one patient had
microcalcifications and underwent subsequent surgical re-excision with residual DCIS on pathologic analysis. At one year, 94% of
patients continued to have benign mammograms; of these mammograms, 94% reported only post-surgical scarring, while 6%
described benign-appearing scattered or dystrophic calcifications. Of the three patients with suspicious mammograms at one
year, all underwent core biopsies with benign results. At two years, rates of fat necrosis (2%) and scattered or dystrophic
calcifications (10%) increased, though all mammographic findings were considered benign and none required additional imaging
or biopsies. By three years post-operatively, an additional two patients (4%) developed suspicious findings and underwent
biopsies confirming local recurrence (one invasive, one in situ). The remaining 96% continued to have benign mammographic
findings and were subsequently followed with routine biannual mammographic surveillance.
Overall, 88% of patients required no additional intervention in the three-year period following oncoplastic reduction mammaplasty.
Of patients recommended for biopsy, malignancy was discovered in 50%, including two local recurrences (4% of total patients).
This data is similar to a previously published large study (1841 women) of mammographic surveillance after partial mastectomy
without oncoplastic reconstruction at our institution, which found a 6% malignancy rate at 5 years.
CONCLUSIONS:
Although substantial tissue rearrangement is performed at the time of oncoplastic reduction mammaplasty, our results
demonstrate low rates of abnormal post-operative mammograms and subsequent biopsies over the first three years following the
procedure. These findings support the use of oncoplastic reduction mammaplasty as a strategy for improving reconstructive
outcomes in patients undergoing partial mastectomy.

Title: The aggregate number of false-positive recalls and biopsies performed under different breast cancer screening strategies in
the US
Carlie K Thompson1, Cristina O'Donoghue2, Elissa M Ozanne3, Martin Eklund1 and Laura J Esserman1. 1University of California,
San Francisco, CA; 2University of Illinois at Chicago, Chicago, IL and 3Dartmouth Institute for Health Policy and Clinical Practice,
Lebanon, NH.
Body: Introduction
The ideal screening program optimizes benefits while reducing harms. The balance of benefits and harms with breast cancer
screening is debated. This study aims to inform breast cancer screening policies by calculating the aggregate number of
false-positive recalls and false-positive breast biopsies under different screening strategies.
Methods
We created a model to estimate the number of screening mammograms, false-positive recalls and false-positive biopsies
performed per year in the US based on current practice. The percent of women that participate in screening under current
practice was estimated from the 2010 CDC Behavioral Risk Factor Surveillance System. The model also enabled the comparison
of 3 screening strategies: annual, biennial, and United States Preventive Services Task Force (USPSTF) guidelines, using a
target participation rate of 85%. The number of women at risk in each age group was taken from the US Census and excludes
women who have had breast cancer in the past five years. False-positive recall rates and biopsy rates were obtained from
Hubbard et al. (Ann Intern Med, 2011). Analyses were performed using R statistical software. Outcomes for this analysis were the
total number of false-positive recalls and biopsies. Monte Carlo methods were used to compute 95% confidence intervals.
Results
Screening Strategies
Frequency
Modeled participation
40-49 years
50-69 years
70-85 years
61-71%
74-75%
72-74%
Current practice
Variable
Annual strategy
Every year
85%
85%
85%
Biennial strategy
Every 2 years
0%
85%
0%
USPSTF
Every 2 years
20% (high risk)
85%
25-37.2%
40-49 years
50-69 years
70-85 years
21,994,479
36,820,954
12,610,766
Annual
9.2
8.6
9.6
Biennial
10.4
9.7
10.8
Annual
0.85
1.04
1.5
Biennial
1.01
1.24
1.8
Model Inputs
Women at risk
False-positive recall
rate (%)
False-positive biopsy
rate (%)
Formulas
Number of screening
mammograms
Women at risk x percent women

screened
Number of
false-positive recalls
Number of screening mammograms

x false-positive recall rate
Number of
Number of screening mammograms
false-positive biopsies
x false-positive biopsy rate
Outcomes
Current practice
Annual strategy
Biennial strategy
USPSTF
6.1x107
1.6x107
(5.6x107-6.5x107)
(1.4x107-1.7x107)
2.1x107
(1.9x107-2.2x107)
Number of screening
mammograms
4.6x107 (4.3x107-5.0x107)
Number of
false-positive recalls
4.2x106 (3.4x106-5.3x106)
5.4x106
(4.1x106-6.9x106)
1.5x106
(1.1x106-1.9x106)
2.1x106
(1.5x106-2.6x106)
Number of
false-positive biopsies
5.3x105 (4.1x105-6.4x105)
6.5x105
(4.8x105-8.2x105)
2.0x105
(1.5x105-2.5x105)
2.7x105
(2.1x105-3.5x105)
Conclusion
Compared to annual screening, we estimate that following the USPSTF guidelines would result in 62% fewer false-positive recalls
and 58% fewer false-positive biopsies. The absolute magnitude is in the range of 374,000 biopsies annually. Given that these
screening strategies have been projected to be of equal benefit, more widespread adoption of the USPSTF guidelines could
decrease the patient risk and anxiety associated with recall and biopsy without impacting the benefits. Developing tools for
personalized risk assessment would facilitate adoption of the USPSTF guidelines. These tools can be tested by comparing the
personalized and annual screening strategies prospectively and can inform screening policies.

Title: Impact of breast density notification laws on radiology practices: A survey of 110 radiology facilities
Lina Nayak1, Kanae K Miyake1, Yueyi Irene Liu1, William R Thomas1, Edward A Sickles2, Bonnie N Joe2, Karen Lindfors3, R J
Brenner4, Stephen Feig5, Lawrence W Bassett6, Jessica W Leung7, Haydee Ojeda-Fournier8, Jonathan Hargreaves3, Elissa
Price2, Jafi A Lipson1, Allison W Kurian1, Elyse Love3, Donna D Walgenbach3, Lauren Ryan2, Meg Durbin9, Bruce L Daniel1, Linda
Garcia6 and Debra M Ikeda1. 1Stanford University, Stanford, CA; 2University of California, San Francisco, CA; 3University of
California, Davis, CA; 4Alta Bates Summit Medical Center, Berkeley, CA; 5University of California, Irvine, CA; 6University of
California, Los Angeles, CA; 7California Pacific Medical Center, San Francisco, CA; 8University of California, San Diego, La Jolla,
CA and 9Palo Alto Medical Foundation, Palo Alto, CA.
Body: Purpose: Breast Density Notification laws, passed in 15 states as of April 2014, mandate that breast density information be
given to patients often without guidance on modalities, patient selection or funding for supplemental screening. The purpose of
this study is to assess the impact of breast density notification laws on radiology practices, specifically regarding breast cancer
risk assessment and supplemental screening studies.
Methods:
We performed an anonymous 20-question web-based survey to Society of Breast Imaging radiologists using a Qualtrics Survey
Tool between 8/2013-3/2014, with questions on radiology practices, breast cancer risk assessment, breast density measurement,
supplemental screening tests, and support for referring physicians and patients. We compared survey results between groups
using Fishers exact test.
Results:
121 radiologists from 110 facilities (48 academic, 43 large private hospital, 15 small private hospital and 4 other) representing 34
USA states and 1 Canadian site responded. 49% of facilities (54/110) were in states with an enacted breast density notification
law. 37% of facilities (40/109) performed risk assessment, 26% (28/109) did not perform risk assessment, and 38% (41/109) did
not but reported family history/other risk factors, with no significant difference in performing risk assessment between facilities
with or without an enacted law (p-value 0.71). Of the 37 facilities performing risk assessment, 60% used the Gail model, 22%
used the Tyrer-Cuzick model and 11% used the modified Gail model (multiple answers allowed [m.a.a.]). Of the 15 facilities
performing risk assessment, 40% answered "yes" when asked whether performing risk assessment is a new task because of the
density law. Breast density was estimated by only visual assessment in 98% of facilities (103/105), and by computer-based
determination with or without visual assessment in 2% (2/105). Supplemental screening studies offered included magnetic
resonance imaging (MRI) (88%, 92/105), handheld whole breast ultrasound (HHWBUS) (48%, 50/105), tomosynthesis (39%,
41/105), and automated WBUS (8%, 8/105) (m.a.a.). There was no significant difference in supplemental screening studies
offered between facilities with or without an enacted law (p-value 0.26). In anticipation of the law, facilities implemented HHWBUS
(33%, 16/48), tomosynthesis (6%, 3/48), automated WBUS (6%, 3/48) or none (60%, 29/48) (m.a.a.). Facilities with the enacted
law prepared for the law with referring physician discussions (69%, 34/49), website (49%, 24/49), educational talks for referring
physicians (43%, 21/49) or patients (31%, 15/49) (m.a.a.).
Conclusion:
Our survey showed variations in available supplemental screening modalities and policy implementation at each facility. There
was no significant difference in performing risk assessment and supplemental screening studies between facilities with or without
an enacted breast density notification law.

Title: Breast cancer screening by women from BRCA1 & BRCA2 mutation positive families
Melanie Wuttke1, Roger Milne2, Prue Weideman1, Sandra Picken1, Charmaine Smith1, Michael Friedlander3, kConFab
Investigators1, Sue-Anne McLachlan4, John L Hopper2 and Kelly-Anne Phillips1. 1Peter MacCallum Cancer Centre, Melbourne,
Victoria, Australia; 2University of Melbourne, Melbourne, Victoria, Australia; 3Prince of Wales Hospital, Sydney, New South Wales,
Australia and 4St Vincent's Hospital, Melbourne, Victoria, Australia.
Body: Purpose: Breast screening may detect breast cancer (BC) at an early stage but optimal screening depends on a womans
level of risk and age. This study estimates the contemporary prevalence of BC screening by Australian women from families with
a BRCA1/2 mutation identified by Family Cancer Clinics.
Methods: Subjects were carriers and true non-carriers from families with a BRCA1/2 mutation enrolled in the nationwide
kConFab cohort. They are followed up every 3 years with a questionnaire which includes use of breast ultrasound,
mammography (MMG), magnetic resonance imaging (MRI) and clinical breast examination (CBE) over the previous 3 year
period. Data from each womans most recent questionnaire (completed Oct 2009- March 2014) were used in the analysis. All
knew their mutation result. Those who had risk-reducing mastectomy, previous cancer, were evaluated for benign breast disease,
pregnant or breastfeeding or received their mutation result in the most recent follow-up round were excluded. Screening
behaviour was categorized based on current national guidelines (see table). Associations with underscreening and overscreening
were assessed using unconditional logistic regression to estimate odds ratios and 95% confidence intervals.
Definitions of overscreening and underscreening for carriers & true non-carriers
Groups
Overscreening*
Underscreening*
Breast imaging less often
than annually if age 30
Carriers
Breast imaging (MMG +/- MRI) more often than annually
True non-carriers
Age <40: any screening MMG; age 40: MMG more often than every 2 years OR Age 50-70: MMG less
any MRI or ultrasound screening
often than every 2 years
*Definitions derived from Cancer Australia Guidelines which do not recommend for or against CBE in any group and differ from
US guidelines.
Results: Of 372 eligible participants, there were 92 mutation carriers (42 BRCA1, 50 BRCA2) and 280 true non-carriers. 1% of
carriers were overscreening, 86% were screening annually while 13% were underscreening. MRI, which is funded for mutation
carriers aged 30-50 years, was used by only 52% (13/27) of carriers aged <50 years and by 7 carriers aged >50. Underscreening
carriers were more likely to be single (OR=2.78, 95%CI=1.28-5.88, P=0.009) while those with a first degree cancer affected
relative were less likely to underscreen (OR= 0.40, 0.19-0.85, P=0.017). CBE was undertaken by 79% of carriers at least yearly;
15% had no or irregular CBE.
115/280 (41%) non-carriers were overscreening, 55% were screening appropriately and 5% were underscreening.
Underscreening non-carriers were more likely to be single (OR 3.85, 95%CI=1.00-14.9, P=0.05). Predictive of overscreening in
non-carriers was having a cancer affected first degree relative (OR=2.92, 1.55-5.51, P=0.001). Non-carriers were less likely to
utilise CBE (54%).
Conclusion: Most mutation carriers in kConFab are having regular screening MMG, but MRI is underutilised even when funded.
The reasons for low usage of MRI requires further research and is concerning given its increased sensitivity over MMG.
Overscreening is common in true non-carriers of BRCA mutations. Family cancer history and marital status may predict
inappropriate screening behaviour.

Title: The contribution of the national health system to mammographic screening in Brazil
Ruffo Freitas-Junior1, Danielle CN Rodrigues1, Rosangela S Correa2, Joo-Emilio Peixoto3 and Rosemar MS Rahal1. 1Program of
Mastology, Federal University of Goias, Goiania, Goias, Brazil; 2CNEN, Comissao Nacional de Energia Nuclear, Goiania, Goias,
Brazil and 3National Cancer Institute, Rio de Janeiro, Brazil.
Body: INTRODUCTION: In Brazil, access to mammography is provided through the National Health System (SUS), the
Supplemental Health System, or is paid for directly by the patient. SUS is the official government system and was established to
conform to the constitutional requirement that health is a right of all Brazilian citizens and a duty of the state. In recent years SUS
breast cancer control policies have advanced and government strategies for early detection have strengthened. However,
evaluations are needed to monitor the effectiveness of these actions. OBJECTIVE: To describe the coverage of mammography in
breast cancer screening conducted by the National Health System in Brazilian macro-regions and states in 2012. METHODS: An
ecological study, where the estimate of coverage was the number of exams performed expressed as a percentage of the number
of exams expected in the target population of women 50-69 years old. The exams performed refer to target population
mammography production data from the Outpatient Information System (CIS) of DATASUS. To calculate the expected number of
exams for this population, the biennial screening recommendations of the National Cancer Institute (NCI) and an estimate of the
female population based on the census of the Brazilian Institute of Geography and Statistics (IBGE) were used. RESULTS: The
coverage estimate for SUS-performed mammography in Brazil in 2012 was 26.6%. Stratified by macro region, the lowest
coverage was in the northern region (12.1%) and highest in the South (34.6%).
National coverage of screening mammography in Brazil, performed by the National Health System (SUS), according to the
geographical regions.
Macro-regions
Coverage
Minimum
Maximum
North Region
12.1
0.2
27.2
Northeast Region
23.9
10.3
37.2
Southeast Region
28.5
15.5
32.5
South Region
34.6
30.3
41.0
Midwest Region
14.8
10.3
19.7
p*
0.001
* Chi-square test
Units of the Federation an estimated coverage ranged from 0.2% to 41%, the lowest in the state of Amapa and the largest in
Santa Catarina. 48% of states were in the range of 10.1% to 20% coverage. CONCLUSION: The results imply that the
contribution of SUS to mammography screening in Brazil is higher in both macro-regions and states with higher income and
better organised health system.

Title: Qatar experience of Standard risk breast cancer screening program (BCSP)
Salha Bujassoum1, Reena Alassam1, Hekmat Bugrein1 and Mufid Elmistiri. 1Hamad Medical Corporation, Doha, Qatar.
Body: Background
Qatar has one of the highest age-adjusted breast cancer incidences in the Arab world. Although this is much lower than the
incidence in the West. Breast cancer incidence in Qatar was 45 per 100,000 in 2003-200753 per 100,000 in 2008-2011.These
higher incidence rates in Qatar are mainly due to the growing population. The prevalent age group, for Qatari and non-Qatari
patients, was 40-50 years old. This suggests that the age-specific incidence of breast cancer in Qatari women is unlike the
pattern usually seen in Western nations where median age at diagnosis is 61 years, moreover the diagnosis is often at advanced
stages of breast cancer. These factors led to establishment the first hospital based (BCSP) in Qatar. It uses a distributed model of
mammography service. The program launched 2008, accepts eligible asymptomatic women at ages 40 - 69 years.
Methods
A retrospective study was done during the period from April 2008-December 2013.
Our aim is to describe our experience of (BCSP) in Qatar and to monitor performance indicators. Our (BCSP) includes an office
call and recall as well as triple assessment. We also discuss positive cases in multi-disciplinary meeting.
Supplement women satisfaction survey conducted along with screening for 100 women showed highly over all satisfaction
reached 90%.
Results
Total number of screened women was 4264 with an increasing participation, year by year. Out of these, Qatari patient's accounts
for 1145, and non Qatari for 3119. The age group of cases was (43-51). Total breast biopsies were 82, of which 45 were positive
of breast carcinomas, (37) invasive ductal carcinoma, (8) noninvasive ductal carcinoma. The Invasive cancer detection rate was
8.2 %. The positive predictive value (PPV) was 46%. Sensitivity value has improved from 51% in 2008 to 70% in 2012 as well as
specificity value that has increased from 77% in 2008 to 83% in 2012.
Conclusion
Public acceptance of the breast cancer screening program in Qatar gradually increased and women highly satisfied about the
services, incidence rate higher in age group (43-52) years. invasive detection rates were higher compare to the western counters
, this indicate either aggressive behavior of the disease or women come forward to screening without knowing that they are
symptomatic, in this part of the world. We have a unique population of multinationals that merits tailored screening tools .and
intensify the public awareness and early defection screening program.

Title: Diagnostic performance of PET/CT and bone scintigraphy in breast cancer patients with suspected bone metastasis
Naoki Niikura1, Jun Hashimoto2, Toshiki Kazama2, Jun Koizumi2, Rin Ogiya1, Mayako Terao1, Risa Oshitanai1, Toru Morioka1,
Banri Tuda1, Takuho Okamura1, Yuki Saito1, Yutaka Imai1 and Yutaka Tokuda1. 1Tokai University School of Medicine, Japan and
2
Tokai University School of Medicine, Japan.
Body: Introduction
Previous retrospective studies suggest that 18FDG PET/CT (PET/CT) has superior sensitivity and specificity to bone scintigraphy
(BS) in detecting breast cancer bone metastases, but the difference in efficacy between these techniques has not been
confirmed. Potentially, PET/CT may detect bone metastases more accurately than BS does. To test this hypothesis, this
prospective study compared the diagnostic efficacy to detect bone metastases between PET/CT and BS in breast cancer
patients. We also compared the response of bone metastases assessed by the PET/CT or BS with the bone metastases.
Method
This single-institution prospective study included consecutive patients with breast cancer diagnosed by biopsy and suspected
bone metastases at the Breast Diseases Unit at Tokai University Hospital, Kanagawa, Japan between September 2011 and
March 2014. Inclusion criteria were as follows: bone pain, elevated alkaline phosphatase, elevated tumor marker, and suspected
bone metastases on BS. Two nuclear medicine physicians interpreted the PET/CT and BS images. Bone involvement was
confirmed by biopsy, especially in the case of oligometastasis. If biopsy proved difficult to perform, conventional imaging and
additional directed radiological studies and follow up were helpful. This study was approved by the Institutional Review Board of
the Tokai University School of Medicine and is registered with UMIN, number 000006003. All patients provided informed consent.
Result
Thirty patients were initially enrolled, but two patients were excluded from analysis because they declined further follow-up
imaging. The median patient age at diagnosis was 59 years (range, 3174 years). Among the 28 patients, bone pain was
observed in 6 patients, elevated alkaline phosphatase in 4, elevated tumor marker in 17, and suspected bone metastases were
detected on BS in 7 patients. Among 10 patients were diagnosed bone metastases, PET/ CT detected 10 of 10 bone metastases,
however BS detected 7 of 10 bone metastases. PET/CT and BS were not highly concordant in detecting osseous metastases;
among 19/28 paired studies (68%), 2 (10%) were positive for metastasis, and 17 (90%) were negative. Nine occurrences (32%)
were discordant; of these, 2 of 9 were PET/CT positive and BS negative; 5 of 9 were PET/CT positive and BS equivocal; one
case was PET/CT negative and BS equivocal; and one was PET/CT equivocal and BS negative.
Conclusion
This study supports the use of PET/CT for detecting suspected osseous metastases. A large prospective study is needed to
determine whether PET/CT could replace bone scintigraphy in detecting suspected bone metastases.

Title: Whole transcriptome analysis of AR+ ER/PR- metastatic breast cancers treated with bicalutamide on TBCRC011
Tiffany A Traina1, Ayca Gucalp1, William Polkinghorn1, Steven Isakoff1, Sara Tolaney1, Lisa Carey1, James Ingle1, Lisle Nabell1,
Andres Forero1, Hope Rugo1, Kimberly Blackwell1, Minetta Liu1, Matthew Soloway1, Lisle Mose1, Dilip Giri1, Agnes Viale1, Clifford
Hudis1, Charles Sawyers1 and Joel Parker1. 1Translational Breast Cancer Research Consortium.
Body: Background: The heterogeneity of TNBC includes a subtype that is androgen dependent and whose growth is inhibited
by antiandrogen therapy. We conducted a multicenter phase II trial which established the activity of bicalutamide for the treatment
of patients (pts) with metastatic ER/PR-negative breast cancer TBCRC011 (Gucalp CCR 2013). We now report results of whole
genome, next generation sequencing of the 26 evaluable pts to molecularly classify the study population and to identify potential
biomarkers of response to bicalutamide. We hypothesized that those pts whose tumors express increased AR output are most
likely to benefit from anti-androgen therapy.
Methods: Archival formalin-fixed, paraffin-embedded (FFPE) samples were collected from either primary or metastatic site and
RNA was isolated using standard techniques. Whole transcriptome sequencing was performed using the Illumina HiSeq 2000
platform. This data set was compared to that of TCGA, the PAM50 and the Lehmann TN subtypes for molecular classification.
Bioinformatic analysis defined a model of AR transcriptional output based on androgen stimulated AR+ ER/PR- breast cancer cell
lines (HCC202, SUM185, MDA453) that was applied to the TBCRC011 data set as a predictor of response. Gene set analysis
was performed to test secondary hypotheses.
Results: 21/26 pts provided adequate gene expression estimates for further analysis. Principal component analysis in
comparison to whole transcriptomes from breast cancers in the TCGA show that TBCRC011 tumors associate with the basal like
(BL) intrinsic subtype, and this was confirmed by PAM50 classification. Relative AR expression of TBCRC011 cases was similar
to that of BLBC in the TCGA. There is a weak but statistically significant positive correlation between AR expression by IHC and
transcriptome (r=0.395, p=0.034). An expression-based model of AR activity was derived from AR+ ER/PR- cell lines exposed to
androgen. TBCRC011 samples segregate with the TCGA samples predicted to be AR responsive by this model. However, the AR
output signature did not predict clinical benefit (CR+PR+SD>24 weeks) as defined in TBCRC011. Unexpectedly, by the Lehmann
TNBC subtype criteria TBCRC011 tumors were not consistently molecularly classified as luminal AR (BL1-1, BL2-1,
Immunomodulatory-5, Luminal AR-4, Mesenchymal-2, Mesenchymal Stem Like-2, Unclassified-6). None of the 4 pts with clinical
benefit were classified as BL by the Lehmann model. Gene set analysis revealed that estrogen signal pathways were associated
with the number weeks on therapy, and that samples with high scoring AR by IHC exhibited expression patterns of medullary
breast cancer.
Conclusions: The majority of samples demonstrated expression variation consistent with TCGA basal like breast cancer.
Although cell line models predict enrichment of AR activity in both the TCGA basal like and TBCRC011 cohorts, this AR output
signature did not predict clinical benefit of bicalutamide. Molecular classification found clinical benefit beyond the luminal AR
subtype but exclusive of basal like 1 and 2. Whether IHC or molecular subtype is the optimal biomarker for pt selection for
antiandrogen therapy remains uncertain.

Title: Multiple subtypes of triple negative breast cancer are dependent on androgen receptor
Valerie N Barton1, Nicholas C D'Amato1, Michael A Gordon1, Britta M Jacobsen1 and Jennifer K Richer1. 1University of Colorado
Anschutz Medical Campus, Aurora, CO.
Body: Triple negative breast cancer (TNBC) constitutes 10-20% of invasive breast carcinomas and has the lowest five-year
survival rate. Currently, there are no targeted therapies for TNBC. Recent studies demonstrate that the androgen receptor (AR) is
expressed in up to a third of TNBC. AR is highly expressed in the "luminal AR (LAR)" molecular TNBC subtype, but is also
present in other TNBC subtypes and may present an opportunity for targeted therapy. We hypothesized that AR+ TNBC critically
depend on AR and that AR inhibition will decrease tumor burden in preclinical models of breast cancer. To determine the extent
to which AR+ TNBC depend on AR, we inhibited AR activity with the AR antagonist enzalutamide (ENZ) and shRNAs against AR
in multiple TNBC subtypes. Treatment with ENZ prevented AR nuclear localization in response to DHT in multiple TNBC cell
lines, reduced baseline proliferation in 2D culture (p<0.05), and decreased anchorage independent growth in soft agar (p<0.01).
AR knockdown significantly reduced proliferation (p<0.001) and increased apoptosis (p<0.001) compared to cells transduced with
a non-targeting control. In addition to reduced proliferation, AR knockdown or treatment with ENZ altered cellular morphology
from stellate to round in Matrigel and significantly decreased migration (p<0.05) and invasion (p<0.001) of cell lines spanning
multiple TNBC subtypes. Microarray profiling and ELISA of TNBC lines treated with DHT and ENZ suggested that AR regulation
of the EGFR ligand amphiregulin (AREG, p<0.05) is a mechanism by which AR influences proliferation, migration and invasion in
TNBC. Indeed, treatment with exogenous AREG rescued decreased proliferation, migration and invasion of AR knockdown cell
lines (p<0.05). In vivo, ENZ significantly decreased tumor viability (p=0.008) and increased necrosis (p=0.009) in SUM159PT
xenografts. Our findings suggest that AR influences both proliferation and invasion of AR+ TNBC cells representing multiple
TNBC subtypes and provide promising preclinical data on the efficacy of ENZ in AR+ TNBC. Thus, inhibition of AR by
anti-androgens such as ENZ may represent an effective targeted therapy for TNBC.

Title: PIK3CA mutations in androgen receptor-positive triple negative breast cancer confer sensitivity to the combination of PI3K
and androgen receptor inhibitors
Brian D Lehmann1, Joshua A Bauer1, Johanna M Schafer1, Christopher S Pendleton1, Luojia Tang1, Kimberly C Johnson1, Xi
Chen1, Justin M Balko1, Henry Gomez2, Carlos L Arteaga1, Gordon B Mills3, Melinda E Sanders1 and Jennifer A Pietenpol1.
1
Vanderbilt-Ingram Cancer Center, Nashville, TN; 2Instituto Nacional de Enfermedades Neoplsicas, Lima, Peru and 3University
of Texas MD Anderson Cancer Center, Houston, TX.
Body: Triple negative breast cancer (TNBC) is a heterogeneous collection of biologically diverse cancers, which contributes to
variable clinical outcomes. Previously, we identified a TNBC subtype that has a luminal phenotype and expresses androgen
receptor (AR+). TNBC cells derived from these luminal AR+ tumors have high frequency PIK3CA mutations. The purpose of this
study was to determine if targeting PI3K alone or in combination with an AR antagonist is effective in AR+ TNBC.
Methods: We determined the frequency of activating PIK3CA mutations in AR+ and AR- TNBC clinical cases. Using AR+ TNBC
cell lines and xenograft models we evaluated the effectiveness of PI3K inhibitors, used alone or in combination with an AR
antagonist, on tumor cell growth and viability.
Results: PIK3CA kinase mutations were highly clonal, more frequent in AR+ vs. AR- TNBC (40% vs. 4%), and often associated
with concurrent amplification of the PIK3CA locus. PI3K/mTOR inhibitors had an additive growth inhibitory effect when combined
with genetic or pharmacological AR targeting in AR+ TNBC cells. We also analyzed the combination of bicalutamide +/- the
pan-PI3K inhibitor GDC-0941 or the dual PI3K/mTOR inhibitor GDC-0980 in xenograft tumor studies and observed additive
effects.
Conclusions: While approximately one third of TNBC patients respond to neoadjuvant/adjuvant chemotherapy, recent studies
have shown that patients with androgen receptor positive (AR+) TNBC are far less likely to benefit from the current standard of
care chemotherapy regimens and novel targeted approaches need to be investigated. In this study, we show that activating
PIK3CA mutations are enriched in AR+ TNBC; and, the growth and viability of AR+ TNBC cell line models is significantly reduced
after treatment with PI3K inhibitors used in combination with an AR antagonist. These results provide rationale for pre-selection of
TNBC patients with a biomarker (AR expression) to investigate the use of AR antagonists in combination with PI3K/mTOR
inhibitors.

Title: The prognostic effect of a negative progesterone receptor (PgR) by immunohistochemistry (IHC) in luminal HER-2 negative
breast cancer (BC) by age at diagnosis: 10 years follow-up study
Siel JAR Olbrecht1, Kathleen Van Asten1, Annouschka Laenen1, Chantal Remmerie2, Wildiers Hans2, Floris Guiseppe2,
Christiaens Marie-Rose2, Vergote Ignace2 and Neven Patrick2. 1KU Leuven, Leuven, Belgium and 2University Hospitals, Leuven,
Belgium.
Body: Background
After menopause, an absent PgR predicts distant metastases (DM) in patients with ER positive HER-2 negative invasive BC.
Yamamoto et al. (JCO 2013) questioned the prognostic effect of PR in premenopausal patients, as ovarian estrogens might affect
PgR expression. We investigated the effect of PgR on DM free interval (DMFI) and BC specific survival (BCSS) by age at
diagnosis as surrogate marker for menopause.
Retrospective data of consecutive patients with a primary operable ER positive HER-2 negative BC (1/1/2000 - 31/12/2009) were
retrieved from our prospectively managed database. Cases that received neo-adjuvant chemotherapy or were operated in
another hospital, were excluded. BC with missing values for PgR, grade or lymph node status and/or lost to follow-up were
excluded in some subgroup analyses. A multivariate (MV) competing risk model for DMFI and BCSS was established considering
age at diagnosis (age 50 yrs vs >50 yrs), PgR status, tumor grade (1-3), tumor size (mm) and lymph node status (neg/pos).
Steroid receptors were considered positive if 1% stained on IHC. Differential prognostic effects of these variables according to
PR or age were tested by means of interaction effects. Only significant interactions were included.
Results
We included 3326 BCs (8 with missing PgR status); 2911 (87.5%) PgR positive [870 (26.2%) 50 yrs and 2041 (61.4%) >50 yrs]
and 407 (12.5%) PgR negative [68 (2.0%) 50 yrs and 339 (10.2%) >50 yrs]. In absolute numbers and compared to PgR positive
BCs, PgR negative cases >50yrs had more DM if grade 3 and more BCSS for all grades. In BC 50 yrs these differences were
not found, but only 7.2% were PgR negative (Tab.1). Results from the MV models showed a significant interaction with PgR on
DMFI and BCSS (respectively p=0.01 and p=0.002) resulting in higher risk of DM (HR, 1.9; 95% CI, 1.4-2.7; p=0.001) and BCSS
(HR, 2.4; 95% CI, 1.6-3.7; p<0.001) when PgR was absent for BCP >50 yrs. This difference in DMFI and BCSS according to PgR
status was not found in BC 50 yrs (Tab.2).
Table 1. Number of BCP with distant metastasis and BCSS according to age (yrs), PgR status(+/-) and tumor grade.
Distant Metastatic Disease
50 yrs
PgR+
PgR-
Breast Cancer Specific Survival
>50 yrs
PgR+
50 yrs
PgR-
PgR+
PgR-
>50 yrs
PgR+
PgR-
Grade 1-2
45/607 (7.4%)
5/42 (11.9%) 66/1512 (4.4%) 11/226 (4.9%)
28/607 (4.6%)
2/42 (4.8%) 33/1511 (2.1%) 7/226 (3.1%)
Grade 3
50/263 (19.0%) 4/26 (15.4%) 69/529 (13.0%) 35/113 (31.0%) 31/263 (11.8%) 2/26 (7.7%) 44/529 (8.3%)
26/113 (23.0%)
Table 2. Results of MV model including significant main effects and interaction effects between tumor characteristics, age and
PgR status on DMFI and BCSS.
95% Confidence interval
Interactions
Hazard ratio
Lower limit
Upper limit
P-value
A. BMFI
PgR Neg. vs Pos. |-50yrs
0.784
0.395
1.556
0.4869
PgR Neg. vs Pos. |+50yrs
1.937
1.386
2.709
0.0001
B. BCSS
PgR Neg. vs Pos. |-50yrs
0.575
0.208
1.588
0.2854
PgR Neg. vs Pos. |+50yrs
2.439
1.620
3.672
<.0001
Conclusions
A negative PgR is only prognostic for DMFI and BCSS in women aged >50 at diagnosis. There was no difference of PgR for
these endpoints in patients 50 yrs at diagnosis but PR negativity is rare in this age group.

Title: Invasive lobular carcinoma cell lines utilize WNT4 signaling to mediate estrogen-induced growth
Matthew J Sikora1, Amir Bahreini1, Caroline M Alexander2 and Steffi Oesterreich1. 1University of Pittsburgh, Pittsburgh, PA and
2
University of Wisconsin, Madison, WI.
Body: Invasive lobular carcinoma (ILC) is a histological subtype of breast cancer representing 10-15% of newly diagnosed breast
tumors. Over 90% of ILC are ER-positive, however, endocrine response and estrogen signaling are not well described in ILC.
Retrospective analyses suggest that ILC patients treated with endocrine therapy have poorer outcomes than similar invasive
ductal carcinoma (IDC) patients, and that ILC patients may not benefit from adjuvant tamoxifen. Additionally, we recently
identified ILC-specific ER-target genes and de novo tamoxifen resistance driven by ER in ILC model systems. Based on these
observations, we hypothesize that ILC-specific signaling pathways driven by ER mediate growth and endocrine resistance in ILC
cells.
Among ILC-specific estrogen-regulated genes in the ILC cell lines MDA MB 134VI (MM134) and SUM44PE (SUM44), Wnt
signaling genes were highly differentially expressed. The secreted ligand WNT4 was the most strongly estrogen-induced gene in
ILC cells. The frizzled receptor FZD7 is also strongly induced in ILC cells, but only transiently induced in the ER-positive IDC cell
line MCF-7. Among IDC cell lines, either WNT4 or FZD7 is over-expressed in ER-positive or ER-negative cells, respectively.
Conversely, MM134 and SUM44 over-express both WNT4 and FZD7. Also, we identified an ILC-specific ER binding site at
WNT4; located in intron 1, this site contains a predicted estrogen response element. Direct WNT4 regulation and parallel
regulation of pathway genes suggests that ER controls a WNT4 signaling pathway in ILC cells. In samples from the Cancer
Genome Atlas, WNT4 and FZD7 are each over-expressed in ER-positive ILC versus IDC; co-expression is also enriched only in
ILC. These observations suggest that a WNT4 signaling pathway may be specifically active in ILC tumors.
To assess whether WNT4 is necessary for estrogen-induced growth, we used siRNA to knock down WNT4. Using either of two
siRNAs, WNT4 knockdown completely blocks estrogen-induced growth in ILC cells, but not IDC cells. Consistent with this, WNT4
knockdown abrogated estrogen-regulation of a subset of ER-target genes in MM134 cells; induction or repression was inhibited
by WNT4 knockdown prior to estrogen treatment. Thus, a subset of estrogen-induced gene expression changes is mediated by
WNT4 signaling. Though Wnt signaling typically acts via the canonical, -catenin-dependent pathway, we observed that -catenin
signaling is dysfunctional in ILC cells. Additionally, WNT4 over-expression or recombinant protein cannot activate canonical Wnt
signaling in breast cancer cell lines. This suggests that WNT4 signaling mediates estrogen-induced growth in ILC cells via a novel
non-canonical signaling pathway.
Wnt signaling pathway genes including WNT4 are uniquely regulated in ILC cell lines, and are over-expressed in ILC tumors,
suggesting that a WNT4-driven pathway may be active specifically in ILC. WNT4 is necessary for estrogen-mediated growth in
ILC cells, and likely signaling via a novel non-canonical signaling pathway. Targeting WNT4 signaling represents a novel
approach to modulate endocrine response specifically for ILC patients. Future studies will focus on identifying the signaling
pathway controlled by WNT4 in order to identify novel therapeutic targets.

Title: Inhibiting androgen receptor nuclear localization decreases estrogen receptor (ER) activity and tumor growth in ER+ breast
cancer
Nicholas C D'Amato1, Britta M Jacobsen1, Dawn R Cochrane1, Nicole S Spoelstra1, Beatrice L Babbs1, Anthony Elias1 and
Jennifer K Richer1. 1University of Colorado Anschutz Medical Campus, Aurora, CO.
Body: Background: Androgen receptor (AR) is widely expressed in breast tumors, but the role of AR in estrogen receptor (ER)+
tumors is controversial. In the absence of estradiol (E2), dihydrotestosterone (DHT) increases growth of ER+ breast cancer cells
in vitro and in vivo. Anti-androgens such as bicalutamide (Bic) and enzalutamide (ENZ) inhibit this DHT-mediated proliferation.
Surprisingly we have found that ENZ, which impairs nuclear entry of liganded AR, also inhibits E2-mediated proliferation of ER+
breast cancer cells, while Bic does not.
Hypothesis: We hypothesize that nuclear localization of AR is necessary for maximal E2-mediated proliferation in ER+/AR+
breast cancer cells, and targeting AR with ENZ or other agents that impede AR nuclear entry or cause AR degradation will inhibit
growth of ER+/AR+ human breast cancer cell lines and decrease tumor burden in preclinical models.
Methods: ER+/AR+ MCF7, BCK4, and ZR-75-1 cells were treated with E2 plus or minus ER and AR antagonists and proliferation
was measured by crystal violet staining or Incucyte live cell imaging. Nuclear AR was assessed by immunocytochemistry or
nuclear/cytoplasmic fractionation. For in vivo experiments, 1x10^6 luciferase-expressing MCF7 cells were injected into the
mammary fat pad of nu/nu mice and tumor growth monitored by caliper and IVIS imaging.
Results: ENZ blocked E2-induced proliferation and showed synergistic activity with the ER antagonists 4-hydroxy-Tamoxifen
(OH-Tam) and Fulvestrant (Fulv) in vitro. E2-induced expression of ER target genes including PR and SDF1 was also inhibited by
ENZ, but not by Bic. Similarly, AR knockdown decreased baseline and E2-induced proliferation of MCF7 cells and E2-induced ER
target gene expression. Both DHT and E2 treatment induced nuclear translocation of AR, which was decreased by ENZ. Nuclear
translocation of AR in response to E2 occurs only in ER+ cell lines, further supporting a role for nuclear AR in E2-induced ER
activity. In vivo, ENZ inhibited E2-induced growth of MCF7 tumors as effectively as Tamoxifen (Tam), and the combination of ENZ
plus Tam was more effective than either drug alone. ENZ also inhibited growth of Tam-resistant MCF7 cells in vitro.
Conclusions: Our results suggest that nuclear localization of AR plays a previously-unrecognized role in E2-mediated ER activity
in ER+/AR+ breast cancer cells. Because of its ability to inhibit nuclear entry of liganded AR, ENZ may serve as an effective
therapeutic in ER+/AR+ breast cancers. Importantly, ENZ may be particularly useful in combination with current anti-estrogen
therapies (Tam or Fulv) since it affects ER, but via an indirect mechanism acting through AR. ENZ may also be effective in tumors
resistant to ER-directed therapy based on in vitro data and published clinical data indicating that many such tumors express more
AR protein than ER.
Funded by: DOD BCRP Clinical Translational Award BC120183 to JKR, American Cancer Society Postdoctoral Fellowship
PF-13-314-01 CDD to NCD.

Title: Androgen receptor expression in triple negative breast cancer
Hannah Gilmore1, Vinay Varadan1, Nicole Williams1, Cheryl L Thompson1, Stephanie Kim1, Peter Hsu1, Kristy Miskimen1, Aditi
Palkar1, Shaveta Vinayak1, Robert Lindner2 and Lyndsay Harris1. 1Case Western Reserve University, Cleveland, OH and 2Yale
University, New Haven, CT.
Body: Background: Patients with androgen receptor (AR) positive triple-negative breast cancer (TNBC) may derive a significant
benefit from anti-androgen therapy. While AR positivity is often defined by protein expression by immunohistochemistry (IHC), the
benefit of anti-androgen therapy in patients who have the Luminal Androgen Receptor (LAR) molecular subtype as defined by
genomic profiling is unclear. Our aim was to study the clinical, pathologic and molecular profiles of AR+ tumors by IHC in a
diverse set of TNBC.
Methods: Tissue microarrays from two separate, well-annotated institutional cohorts (Case Western Reserve University, Yale
University) of early-stage TNBC were evaluated for AR expression by IHC (clone SP107, Ventana Benchmark Ultra). AR positivity
was defined as greater than or equal to 10% staining in tumor nuclei. AR expression was correlated with clinical and pathologic
features such as age, race, grade, and stage within and between the two cohorts. Gene expression was analyzed with the DASL
assay (Illumina) on RNA extracted from formalin-fixed paraffin-embedded material using the Ambion RecoverAll kit (Applied
Biosystems AM1975). Pietenpol TNBC molecular subtypes were calculated using the online TNBC type tool. Co-expression of
AR with other hormone-related proteins including gross cystic disease fluid protein-15 (GCDFP-15, clone EP1582Y, Ventana
Benchmark Ultra) and GATA transcription factor 3 (GATA3, clone L50-823, Bondmax Leica) were also assessed.
Results: Overall, 22% of cases (n=192) were AR+ by IHC. There was no association between AR expression and age, race,
grade or stage within or between the two cohorts. Gene expression data was available on 88 tumors with AR staining results and
demonstrated that AR positivity by IHC was not exclusive for the TNBC LAR molecular subtype. Three of five (3/5) tumors that
were classified and LAR were in fact AR+ by IHC. Tumors that were AR+ by IHC were also identified in the basal-like 2 (BL2),
mesenchymal stem-like (MSL), immunomodulatory (IM), and unstable molecular TNBC categories with low frequency. Notably, of
cases that were rejected as ER+ by gene expression profiling despite being clinically TNBC, three of five (3/5) were in fact AR+
by IHC. Examination of other hormone-related proteins in a subset of these patients revealed that AR expression was significantly
associated with GCDFP-15 expression (p=0.0007) and was borderline significant for GATA3 expression (p=0.068).
Conclusions: AR protein expression levels by IHC were similar in two separate institutional archival cohorts of TNBC and did not
correlate with age, race, grade or stage. Comparison of AR IHC data with genomic data suggests that the AR+ phenotype is not
unique to LAR subtype of TNBC by gene expression profiling. AR protein expression may be seen in clinically TNBC patients who
have a more luminal subtype as evidenced by co-expression of GCDFP-15 and GATA3. Our results suggest that AR staining by
IHC may be necessary to capture all patients who may benefit from anti-androgen therapy.


Title: No Show

Title: Progesterone receptor (PR) blockade by antiprogestin, CDB4124 in hormone receptor positive breast cancer cells leads to
significant inhibition of G2/M cell cycle genes
Akash Gupa1, Mi Ran Choi1, Susan E Clare1, Jun Wang1, Oukseub Lee1, David Z Ivancic1, J Julie Kim2 and Seema A Khan1.
1
Feinberg School of Medicine, Northwestern University, Chicago, IL and 2Feinberg school of Medicine, Northwestern University,
Chicago, IL.
Body: Background: Several lines of evidence suggest that progesterone signaling is important in the breast cancer development,
particularly in young women. Therefore, we sought to establish the effects of the antiprogestin CDB4124 (telapristone), and to
identify PR related signature genes in hormone receptor positive (ER+ PR+) breast cancer cells.
Methods: The PR expressing breast cancer cell line T47D was used to evaluate responses to PR ligands (P4, MPA and R5020,
10nM) alone or in combination with estradiol (E2, 1nM). The effects of the antiprogestin CDB4124 (0.1 or 1M) were tested using
varying hormonal conditions. The effect on (a) PRE promoter activity by dual luciferase assay; (b) cell proliferation using the MTT
assay; (c) cell cycle by flow cytometry; (d) determination of gene expression signatures related to active PR responses. For the
gene array experiment, cells were treated with either vehicle or R5020 (10nM) or R5020 (10nM) plus CDB4124 (1M) in triplicate
for 24hr. Total RNA was isolated and converted to cDNA and human Illumina chip microarray was performed. Data obtained from
the microarray was further analyzed by Metacore Gene GO and Ingenuity Pathway Analysis. Real time PCR was performed in
triplicate to confirm the expression of those genes related to the cell cycle and proliferation. ANOVA analysis and post-hoc Sidak
test were used to determined the statistical significance of the data.
Results: The PRE reporter activity resulting from P4, MPA and R5020 stimulation was inhibited by 80-90% in the presence of
CDB4124 at 10 to 1000nM (p< 0.001). Cell proliferation was increased by PR ligands (P4, MPA and R5020) in the presence of
E2; the addition of CDB4124 caused 50% inhibition of proliferation (p< 0.01) at 72 hours. Cell cycle analysis of T47D cells treated
with P4, MPA and R5020 alone or in combination with E2 showed significant increases in S and G2/M phase and decreases in
G0/G1. These were blocked by CDB4124 at 0.1 or 1.0M (p<0.05 for all). Gene GO metacore analysis of genes identified in the
microarray revealed significant enrichment of cell cycle pathways (FDR, p< 1.0X10-11 ) upon treatment with R5020. The addition
of CDB4124 to R5020 treated samples showed inhibition of the same cell cycle pathways (FDR,p<1.0X10-14). A 16-gene panel
related to G2/M phase of cell cycle was selected based on >1.5 fold upregulation (p<0.001) during treatment with R5020,10nM
and blockade by CD4124. Real time PCR confirmed upregulation of this 16 gene panel 2.0 (p<0.05) in the presence of PR
ligands alone or in combination with E2 which were significantly blocked by the addition of CDB4124.
Conclusion: These data demonstrate that PR mediated cell proliferation occurs upon treatment with three different ligands of PR
(P4, MPA and R5020); that PR actively engages key genes involved in the G2/M phase of the cell cycle to drive proliferation of
ER+ and PR+ cells; and that the antiprogestin, CBD4124 is a potent transcriptional inhibitor for blockade of PR mediated cell
proliferation in hormone receptor positive breast cancer cells.

Title: Thyroid hormone up-regulates estrogenic and pro-carcinogenic signaling, inducing a basaloid and more motile phenotype in
only steroid receptor positive breast cancer cells
Ann D Thor1, Zeying Fan1, Reema Wahdan-Alaswad1 and Susan M Edgerton1. 1University of Colorado Anschutz Medical Center,
Aurora, CO.
Body: Background: Co-morbidity of thyroid disease and breast cancer has been recognized for over a century. Up to two-thirds of
women with breast cancer have clinical or occult thyroid disease, as compared to one fifth of unaffected females. Thyroid disease
typically proceeds breast cancer by at least a decade, and many (nearly half) are on long term thyroid hormone supplementation
at the time of diagnosis. The exogenous administration of thyroid hormone, for hypothyroidism or status post thyroidectomy for
neoplasia, has been associated with an increased risk of breast cancer and in some studies, a worse prognosis. Causality,
however, has been controversial.
Results: We studied over 800 consecutive node negative invasive breast cancer patients diagnosed between 1976 and 1993 from
a single institution. Of these, 92% were chemo-naive and only 14% received tamoxifen, reducing interactions with treatment for
the study design. We showed untoward significant interactions with both univariate and multivariate analyses, between thyroid
hormone supplementation, disease free and disease specific survival only in patients with steroid receptor positive tumors. These
interactions were strongest in pre-menopausal, as compared to post-menopausal patients. We thus hypothesized that thyroid
hormone (TH), either indirectly or directly, promoted carcinogenesis via the steroid receptors (SR) for estrogen and/or
progesterone and have interrogated mechanisms of these interactions in vitro. Estrogen (E2) induce significant cellular
proliferation, whereas TH alone (at concentrations of T4 1 x 10-6:T3 at 2.5 x 10-7 M) induced only mild cell proliferation in SR+
(MCF-7 andT47D), but not SR- breast cancer cell lines (MDA-MB-468 and SKBR3). In contrast, TH induced significant
proliferation in SR + cells at significantly lower doses (down to T4 1 x 10-12: T3 2.5 x 10-13 M) if administered with low dose
estrogen (E2: 1 x 10 -10). These mitogenic effects were demonstrated by MTS and DNA quantification assays, as well as flow
cytometry, which revealed the induction of G2/M arrest and an increase in the percentage of cells in S phase with TH alone (at
higher doses) or TH + E2 (at lower doses). Tamoxifen partially mitigated the pro-growth effects of TH + E2, whereas ICI 182,780
completely abrogated these effects. Using Western blots we have shown that that TH (with or without E2) upregulated ERa
expression and activation, Cyclin A, B,D1, and E, E2F1, MAPK/ERK1/2 but not AKT signaling. TH +/- E2 also promoted
clonogenicity, motility, a more basal phenotype (CD24+/CD44+) and mammosphere formation in cells grown under non-adherent
conditions.
Conclusions: We have demonstrated in Stage I breast cancer patients with minimal systemic treatment, that the administration of
TH was associated with a significantly shortened disease free and disease specific survival, only in patients whose tumors
expressed SR. TH administered to SR+ cell lines, particularly when co-administered with E2 at therapeutic/physiologic doses,
upregulates and activates ER, cell proliferation, a more aggressive and motile basaloid phenotype.

Title: Pathological and molecular effects of lack of PR expression in ER positive breast tumors
Rachel E Ellsworth3, Allyson L Valente1, Matthew T Hueman2 and Craig D Shriver2. 1Windber Research Institute, Windber, PA;
2
Walter Reed National Military Medical Center, Bethesda, MD and 3Henry M Jackson Foundation for Military Medicine.
Body: Background: Evaluation of hormone receptor status is standard in breast cancer diagnosis; however, prognosis and
treatment decisions are frequently based on the status of the estrogen receptor (ER) but not the progesterone receptor (PR). To
determine how PR status affects etiology and outcome of ER positive tumors, pathological and molecular characteristics of
ER+PR+ and ER+PR- tumors were assessed.
Methods: ER and PR positivity were defined as >10% staining and all patients with ER+PR+ and ER+PR- were identified.
Differences in pathological characteristics were evaluated using Chi-square tests with P<0.05 defining significance. RNA was
isolated from primary tumor cells after laser-microdissection and hybridized to U133A 2.0 arrays. Gene expression data was
analyzed by ANOVA using a false-detection rate <0.5 and >2.0-fold difference to define significance.
Results: Of the 1,139 ER+ tumors, 21% were PR-. Age at diagnosis, ethnicity, tumor size, lymph node and Ki67 status did not
differ between groups; however, patients with PR- tumors were significantly more likely to be diagnosed at an advanced stage,
with high-grade, HER2 positive tumors, and more likely to die of disease. Thirty genes were differentially expressed including
significantly higher expression levels of APOBEC3B and significantly lower expression of GREB1, MAPT and SCUBE2.
Discussion: Expression of PR significantly alters tumor biology and clinical outcomes of ER positive tumors. Genes with altered
expression in PR negative tumors are associated with more aggressive characteristics, such as increased invasion, kataegis,
regulation of ER and response to tamoxifen. Thus, PR status is critical in the diagnosis and treatment of patients with ER positive
breast tumors.


Title: Prognostic impact of Her2 and hormone receptor expression in inflammatory breast cancer: A SEER database analysis
Anna Maria Affan1, Babu P Mohan1, Kali Praveena Iruku1 and Keyvan Ravakhah1. 1St Vincent Charity Medical Center, Cleveland,
OH.
Body: INTRODUCTION: Inflammatory breast cancer (IBC) is known to be one of the most aggressive forms of breast cancer.
Patients with IBC generally present with early metastases and are frequently resistant to conventional chemotherapies. By using
the SEER database, we explored the clinical relevance of Her2 and hormone receptor (HR) expression in IBC tumors and the
possible impact it may have on management.
METHODS: Data for patients with IBC was extracted from the Surveillance, Epidemiology, and End Results (SEER 18) database
from 2010 to 2011. A sub-analysis with the available receptor status data for the same period was analyzed using CanSurv
survival software. The Actuarial method was used to construct survival curves according to the following receptor variables;
Her2+/HR+, Her2+/HR-, Her2-/HR+, and triple negative IBC groups. Demographic data including race and age was also
extracted.
RESULTS: Of 354 cases of IBC reported, 5% (n=18) of the tumors were Her2+/HR+, 7.9% (n=28) Her2+/HR-, 15.3% (n=54)
Her2-/HR+, and 12.4% (n=44) triple negative. Highest 1 year survival rates were seen in those patients with Her2-/HR+ tumors at
>95% while Her2+/HR+ had the lowest 1 yr survival rates at 65%.
Caucasian patients above 30 years represented the highest racial group at approximately 70% (n=105). Analysis of this subgroup
showed 38% (n=40) of these patients having HER2-/HR+. Despite more than 75% (n=33) of these patients presenting with
advanced disease, the 1 year survival remained highest among all the receptor combinations analyzed. This observation held
true across all racial groups even with those population numbers for IBC being less.
CONCLUSION: The Her2-/HR+ receptor combination group showed the greatest 1 year relative survival when compared to the
other groups. These results suggest that HR expression in the absence of Her2 is associated with an increased survival. This can
play in a role in the management of IBC by incorporating hormonal therapy into the treatment protocols for patients with IBC to
possibly improve survival.

Title: Ethinylestradiol treatment downregulates ER and upregulates PgR; Immunohistochemical analysis in postmenopausal
breast cancer tissues after prior long-term estrogen-deprivation therapy
Hirotaka Iwase1, Yoko Omoto1, Takashi Takeshita1, Mutsuko Yamamoto-Ibusuki1, Mitsuhiro Hayashi1, Aiko Sueta1, Saori
Fujiwara1 and Yutaka Yamamoto1. 1Kumamoto University, Kumamoto, Japan.
Body: Background: Estrogen receptor (ER) positive breast cancer can often be treated by hormone therapy; however a certain
population of ER-positive patients becomes resistant to hormone therapy after long-term hormone treatment. Ethinylestradiol
(EE2) is a derivative of estrogen which has shown promising effects on in these patients. Methods: We successfully obtained
tissue samples from 6 patients undergoing EE2 treatment and examined 13 well-known breast cancer-related factors by
analyzing their gene expression and by immunohistochemistry. Of the 6 patients, 5 responded but one patient did not.
Results: Before EE2 treatment, staining for both ER and androgen receptor (AR) was strong in the nucleus, with weak staining of
the progesterone receptor (PgR). EE2 treatment significantly down-regulated ER and up-regulated PgR while nuclear and
cytosolic AR were oppositely down- and up-regulated, respectively, by EE2. Cytosolic staining of BRCA1 was significantly
up-regulated by EE2 whereas nuclear staining tended to decrease. Individual comparisons suggested less induction of PgR and
decreasing AKT but increasing pAKT in the non-responder following EE2 treatment. Conclusion: Our observations revealed that
EE2 activated ER downstream genes, although it did not stimulate cell growth. This suggests that hormone resistant cells might
receive growth signals from a non-genomic pathway and this may be reflected in their sensitivity to EE2 treatment.

Title: Estrogen receptor agonists reduce breast cancer tumor growth in syngeneic mouse models
Cathy Samayoa1, Naveen K Krishnegowda1, Ratna K Vadlamudi1 and Rajeshwar R Tekmal1. 1University of Texas Health Science
Center, San Antonio, TX.
Body: The estrogen receptors (ER) play a significant role in breast cancer, with the majority of breast cancers expressing
estrogen receptor alpha (ER) and depending on its signaling. ER has proliferative function, however, estrogen receptor beta
(ER) has anti-proliferative functions. Recently, several selective ER agonists have been identified. The objective of this study
was to determine the effectiveness of ER agonists in inhibiting the growth of distinct breast cancer cells both in-vitro and in-vivo,
and to determine the mechanisms involved. The mouse mammary tumor cell lines, D2A1 and MM51, express both ER and ER.
Specifically D2A1 cells are dependent on estrogen for growth and have increased expression of aromatase. Additionally, D2A1
cells are very aggressive and highly metastatic; representing a good model of breast cancer progression. MM51 cells are
Her2/neu positive and are an adequate model to study the crosstalk between growth factors and ER signaling. Syngeneic mouse
tumor models were used to incorporate an intact immune system which plays a role in the tumors microenvironment. Our in-vitro
studies demonstrate that ER agonists significantly inhibit cell growth in a dose dependent manner. Our syngeneic studies show
that, in-vivo, ER agonists effectively reduce tumor volume and inhibit tumor progression. This study reveals that in addition to
acting on ER, these agonists reduce the expression of ER at both the mRNA and protein level, therefore modulating the ratio of
ER to ER. Additionally, treatment with ER agonists results in increases apoptosis through increased p53 expression, and cell
cycle arrest through p27 and cyclin D1. Together, these studies demonstrate the therapeutic potential of ER agonists for the
treatment of breast cancer.

Title: No Show

Title: Development for clinical utility of a validated predictive test of clinical response to aromatase inhibitors
Arran K Turnbull1, Laura Arthur1, Victoria Webber1, Jeremy Thomas1, Charlotte Heerlyn1, Phoebe Thornton1, Anita Dunbier2, Mitch
Dowsett3, Lorna Renshaw1, Andrew H Sims1 and J Michael Dixon1. 1University of Edinburgh, Edinburgh, United Kingdom;
2
University of Otago, Otago, Dunedin, New Zealand and 3Institute of Cancer Research, London, London City, United Kingdom.
Body: Background: Aromatase inhibitors (AIs) have an established role in the treatment of estrogen receptor alpha positive
post-menopausal breast cancer. Response rates are only 50-70% even in patients with ER-rich cancers in the neoadjuvant
setting and are lower in advanced disease. Recently we developed and validated a microarray-derived 4-gene test to predict
response to AIs in the neoadjuvant setting. Whole-genome expression analysis is impractical for clinical utility. There is a need to
translate and validate any test utilising clinically accessible and reproducible material and technologies such as polymerase chain
reaction (PCR) and immunohistochemistry (IHC).
Methods: The original microarray experiment used pre- and on-treatment (at 14 days and 3-months) biopsies from 89
post-menopausal women with ER-rich breast cancer receiving 3 months of neoadjuvant letrozole. Dynamic response was based
on periodic 3D ultrasound measurements performed during treatment. The derived 4-gene model was independently validated in
a cohort of 44 post-menopausal women with ER-rich breast cancer treated with neoadjuvant anastrozole [table 1]. RNA was
extracted from the original biopsies for RT-qPCR analysis using validated primers for the 4 genes with SYBR-green technology
normalised to the geometric mean of 3 housekeeping genes. Matched formalin-fixed paraffin embedded (FFPE) tissue sections
were used for IHC with optimised antibodies against 3 of the 4 proteins (where validated antibodies were available) using
Envision technology.
Results: PCR: Microarray and PCR expression levels for each of the four genes were well correlated (Pearson r=0.87-0.65,
p<0.0001). Application of the 4-gene model, using PCR expression levels, to a cohort of patients from the initial training set
(n=26) resulted in prediction of response with 96% accuracy [table 1]. IHC: stained FFPE tissue sections for proteins
corresponding to the 3 most informative genes in the model were independently scored using a histoscore approach (0, 1+, 2+,
3+) in a sample of patients where tissue was available (n=28). Increasing histoscores were associated with increasing gene
expression for all 3 proteins (r=0.72, 0.73, 0.60). Each sample was assessed and histoscore determined, then a positive and
negative cut-off histoscore was determined for each protein to maximise response prediction. This was then applied to an IHC
decision tree, based on the original 4-gene microarray model, and was able to categorise patients as responsive or
non-responsive with 89% accuracy.
Table 1
N
Accuracy
Sensitivity
Specificity
PPV
NPV
Microarray (Training)
89
0.96
0.89
0.98
0.98
0.96
Microarray (Validation)
44
0.91
0.80
0.97
0.92
0.90
PCR
26
0.96
0.95
1.00
1.00
0.86
IHC
28
0.89
0.86
1.00
1.00
0.89
Sensitivity and specificity of model in training (microarray, PCR, IHC) and validation datasets (microarray). PPV/NPV =
postivie/negative predictive value.
Conclusion:
A 4 gene model has been developed and validated to predict response to neoadjuvant aromatase inhibitors.
This model has been shown to work with a high degree of accuracy using both PCR and IHC technologies. Further independent
validation is currently underway.
This new test has the potential to predict accurately the benefit of endocrine therapy and has huge potential clinical value.

Title: Global characterisation of the SRC-1 transcriptome and rational drug design results in the identification of a novel peptide
targeting ADAM22 in endocrine resistance
Jarlath C Bolger1, Damian McCartan1, Damir Vareslija1, Ailis Fagan1, Christopher Byrne1, Marie McIlroy1, Peadar O'Gaora2, Arnold
D Hill1 and Leonie S Young1. 1Royal College of Surgeons in Ireland, Dublin, Ireland and 2University College, Dublin, Ireland.
Body: In spite of therapeutic advances, up to 25% of luminal breast cancers will eventually develop resistance to endocrine
therapy and develop metastatic disease. The underlying mechanism causing ER-positive, steroid responsive tumours to develop
a resistant, metastatic phenotype remains unresolved. Previous work from our group and others has identified the P160 protein
SRC-1 as a significant predictor of recurrence on endocrine therapy. The purpose of this study is to further examine downstream
SRC-1 targets in the context of endocrine resistant breast cancer.
We adopted a global approach to define the transcriptional targets of SRC-1. SRC-1 ChIP sequencing in endocrine resistant
luminal B breast cancer cells was combined with SRC-1 gene expression array analysis. This identified a number of pathways
significantly elevated following tamoxifen treatment, including a number involved in cellular adhesion. From these pathways, A
Disintegrin And Metalloproteinase-22 (ADAM22) was selected for further study.
Knockout studies confirmed ADAM22 as a tamoxifen dependent SRC-1 target gene. Functional assays including migration, three
dimensional cell culture and adhesion independence growth assays confirmed a role for ADAM22 in promoting a migratory,
aggressive phenotype. Samples from two separate TMAs comprising over 1,000 patients confirmed that ADAM22 is associated
with poor disease free survival in breast cancer patients.
LGI1 is a naturally occurring neuropeptide which acts on an inhibitory manner on ADAM22 in the central nervous system. Using
molecular modelling, a novel peptide mimetic targeting the disintegrin binding domain of ADAM22 was designed. Treatment with
this peptide mimetic restored endocrine resistant cells to a less aggressive, sensitive phenotype, similar to the effect seen with
knockdown of ADAM22. Moreover in an endocrine resistant xenograft model, treatment with the LGI1 mimetic significantly
reduced primary and metastatic tumour burden in tamoxifen treated animals.
We have used next-generation sequencing techniques to identify a novel therapeutic target in endocrine resistant, metastatic
breast cancer. Rational drug design has been used to manufacture a therapeutic peptide against ADAM22. A combination of in
vitro, in vivo and patient studies has confirmed a role for ADAM22 in metastatic breast cancer. Our novel peptide mimetic may
form a future basis for targeting ADAM22 in endocrine resistant disease.

Title: Reversal of endocrine therapy resistance with inhibitors of AKT, mTOR, or MEK as single agents or in combination
Agostina Nardone1, Gladys Morrison1, Xiaoyong Fu1, Martin Shea1, Tamika Mitchell1, Teresa Klinowska3, C Kent Osborne1,2 and
Rachel Schiff1,2. 1Lester and Sue Smith Breast Center, Baylor College of Medicine, Houston, TX; 2Dan L. Duncan Cancer Center,
Baylor College of Medicine, Houston, TX and 3AstraZeneca iMED, Macclesfeild, United Kingdom.
Body: Background: Crosstalk between estrogen receptor (ER) and growth factor receptor (GFR) downstream signaling
pathways [PI3K/AKT/mTOR and MEK\p42\44MAPK (MAPK)] has been associated with endocrine resistance. Single downstream
inhibitors like everolimus partially reverse endocrine resistance. However, more than one downstream escape pathway may
contribute to endocrine resistance. Furthermore, disruption by a single downstream inhibitor of the negative feedback loops that
balance the amplified signals from GFRs may result in compensatory pathway activation. Therefore we investigated whether dual
downstream signaling inhibitions are required to more effectively overcome tamoxifen-resistant growth, using in vitro and in vivo
models.
Materials and Methods: The effects of different kinase inhibitors (i) AZD2014 (mTORi), AZD5363 (AKTi), and AZD6244 (MEKi)
on endocrine therapy [tamoxifen (Tam) and fulvestrant (Fulv)] were tested in ER+ MCF7 and T47D Tam-resistant derivatives
(TamR). In vitro growth and apoptosis were assessed using methylene blue and c-PARP, respectively. Western blot analysis was
used to analyze the effect of each inhibitor or combination on their respective pathway substrates. Nude mice with transplantable
MCF7 TamR xenografts at a size of 200 mm3 were randomized to continued Tam, continued Tam + (mTORi, AKTi, MEKi,
mTORi+MEKi, AKTi+MEKi) or Fulv the inhibitor combinations.
Results: We found that in two ER+ models MCF7 TamR and T47D TamR in vitro, both mTORi and AKTi were effective in
restoring growth inhibition, and effective in inhibiting their respective pathways. Interestingly, inhibition of mTOR and AKT resulted
in upregulation of pMAPK. However, while MEKi did inhibit its pathway; it did not restore growth inhibition by the antiestrogens.
On the other hand, dual inhibition, adding the MEKi to either mTORi or AKTi, resulted in a more potent reduction of cell growth as
well as of downstream signaling. In the TamR derivative of MCF7, ER is still maintained and plays a role in resistance, unlike the
T47D model, where there is little to no ER expression after TamR develops. Thus, as might be expected, combining downstream
inhibitors with potent ER blockade by Fulv enhances the effect of single and dual downstream signaling inhibitors mainly in the
MCF7 TamR model. Finally, Fulv in addition to dual downstream inhibitions (mTORi+MEKi or AKTi+MEKi) delayed MCF7 TamR
xenograft growth significantly more than Fulv with single downstream inhibitors.
Conclusion: This study provides evidence that dual inhibition of GFR downstream pathways is needed to overcome activation of
escape mechanisms that are up-regulated with acquired endocrine resistance and after resistance to single pathway inhibitors.
Although the downstream inhibitors alone significantly inhibit TamR growth, combination with Fulv robustly slowed growth of
TamR tumors in vivo. Based on these results further studies combining MEK inhibition with inhibitors of the PI3K pathway and ER
downregulators are warranted.

Title: AKT antagonist AZD5363 targets estrogen receptor (ER) function in endocrine resistant breast cancer (BC) and synergises
with fulvestrant in vivo
Lesley-Ann Martin1, Stephanie K Guest1, Sunil Pancholi1, Ricardo Ribas1, Qiong Gao1, Nikiana Simigdala1, Aradhana Rani1, Barry
Davies2, Stephen Johnston3 and Mitch Dowsett3. 1Institute of Cancer Research, London, United Kingdom; 2AstraZeneca,
Macclesfield, United Kingdom and 3Royal Marsden Hospital, London, United Kingdom.
Body: AIM: To evaluate the efficacy and functional consquences of combining AZD5363 with endocrine therapy in pre-clinical
models of endocrine-sensitive and resistant ER+ BC.
BACKGROUND: The PI3K/AKT/mTOR signalling pathway plays an important role in BC. Its close interaction with ER signalling
becomes more complex and inter-dependent with acquisition of endocrine resistance. Targeting the mTOR pathway in
combination with endocrine therapy has shown clinical utility. However, a negative feedback loop exists downstream of the
PI3K/AKT/mTOR pathway with mTOR inhibition leading to increased activation of IGFR1-dependent AKT activity potentially
negating long-term benefit. Direct blockade of AKT in combination with endocrine therapy, may provide a better rationale for
treatment of endocrine-resistant BC, impacting on both cell survival/apoptosis and ER ligand-independent signaling. In this
investigation, we assessed the efficacy of AZD5363, a pan-AKT inhibitor with endocrine therapies in pre-clinical models of
endocrine sensitive and resistant ER+ BC and its impact on molecular and cellular response.
METHODS: Inhibition of AKT using AZD5363 was examined in 5 ER+ BC lines before and after adaptation to long-term estrogen
deprivation (LTED) or tamoxifen (TAMR). The effects of AZD5363 on cell proliferation were determined alone and in combination
with endocrine treatment and feedback upregulation and activation of receptor tyrosine kinases (RTKs) was examined by western
blotting. ER-transactivation was measured with an estrogen-response element (ERE)-linked luciferase reporter construct and
confirmed using chromatin-immunoprecipitation. Global gene expression analysis was used to identify pathways associated with
response. Xenografts were treated with AZD5363 fulvestrant to determine in vivo effects.
RESULTS: AZD5363 caused a dose-dependent decrease in proliferation in all cell lines tested (GI50<500nM) with the exception
of HCC1428-wild-type and LTED. Of note both T47D-LTED and ZR75-LTED, which lose expression of ER were exquisitely
sensitive (GI50100nM). AZD5363 re-sensitised the TAMR cell line to tamoxifen and acted synergistically with fulvestrant in
MCF7-Wt and MCF7-LTED (CI<1). AZD5363 decreased phosphorylation of AKT/mTOR substrates PRAS40, p70S6 and S6 in all
cell lines tested and caused a significant decrease in phosphorylation of ERser167 with an associated 50% reduction in
ER-mediated transcription and decrease in recruitment of ER and CBP to ERE on the TFF1 promoter. Furthermore, AZD5363
reduced Rb and cyclinD1. Inhibition of AKT with AZD5363 resulted in upregulation and activation of RTKs, including IGF-IR,
EGFR, ERBB2 and ERBB3, which was cell line specific. Global gene expression and pathway analysis of MCF7 and MCF7-LTED
treated with AZD5363 highlighted the relevance of ERBB2-ERBB3, ERK5 and IGF1 signalling as potential feedback loops.
Combined treatment with AZD5363 and fulvestrant showed strong synergy in an MCF7 xenograft.
CONCLUSION: These data suggest that AZD5363 plus fulvestrant may be effective in BC that is sensitive or resistant to
E-deprivation or tamoxifen and that activated AKT is a determinant of response. These data strongly support the need for clinical
evaluation.


Title: A nude mouse model of diet-induced-obesity to study the mechanisms of resistance to aromatase inhibitor letrozole in
MCF-7Ca xenografts
Amanda Schech1, Stephen Yu1, Preeti Shah1, Olga Goloubeva1, Angela Brodie1, Saranya Chumsri1 and Gauri Sabnis1. 1University
of Maryland School of Medicine, Baltimore, MD.
Body: Obesity has been identified as one of the risk factor for breast cancer progression. However the mechanisms underlying
this link are not completely understood. Specifically, breast cancer patients who are overweight, obese or have excess abdominal
fat have increased risk of local or distant (metastasis) recurrence and cancer related death. Furthermore, breast cancer patients
gain weight during treatment, especially those who receive hormone depletion therapies, are at increased risk of developing
obesity and metabolic syndrome as survivors. Importantly, the presence of obesity may influence the resistance of breast cancer
to existing treatments, such as aromatase inhibitors (AIs). In an effort to understand the effect of obesity on the growth of
hormone dependent breast cancer tumors, we fed ovariectomized athymic nude mice a diet containing 45% kcal fat (45% of the
total calories coming from the fat in the diet) and low fat (10% kcal fat) diet. We also used a standard chow diet to compare with
our previous results. We saw that mice fed 45%kcal fat diet had a higher rate of tumor growth (p=0.04) compared to mice that
were fed 10% kcal fat containing diet or chow diet. In addition, the mice fed 45% kcal fat had higher body weight, fasting insulin
(as measured by C-peptide ELISA) and high fasting glucose levels. Glucose tolerance test (IP-GTT) demonstrated that mice fed
high fat diet exhibited reduced glucose tolerance as measured by high AUCglucose in high fat versus low fat or chow diet. There
was no statistically significant difference between the chow diet and the low fat diet.
Next, we examined the effect of insulin on the growth of MCF-7Ca cells in response to E2 or letrozole. When co-treated with 2M
of exogenous insulin (in otherwise serum starved condition), the growth of MCF-7Ca cells was not stimulated by E2 at 1nM.
However, at lower does, the combination of insulin with E2 stimulated cell growth more than E2 alone. This suggests that insulin
makes MCF-7Ca breast cancer cells hypersensitive to mitogenic effects of E2. Hypersensitivity to E2 is one of the suggested
mechanisms of resistance to endocrine therapy. Furthermore, response to anti-proliferative effects of letrozole was also
abrogated in presence of insulin (2M). This was more pronounced at lower dose of letrozole. These results suggest that the
presence of insulin makes MCF-7Ca cells less responsive to E2 and letrozole. Furthermore, tumors of mice fed with high-fat diet
also had higher activation of MAPK and Akt, suggesting induction of growth factor receptor pathways. Exogenous insulin
treatment of MCF-7Ca cells also resulted in reduction of ER protein levels and increase in p-IR, p-IGFR and p-Akt. These results
suggest that diet-induced obesity may result in acquisition of resistance to letrozole due to development of hyperinsulinemia. We
hypothesized that the presence of obesity can augment adaptation of cancer cells and impact the mechanism. This may be the
result of a dramatically different hormonal milieu upon development of obesity.

Title: Determining the role of somatic ER mutations in acquired hormone (or SERM) resistance
Sean W Fanning1, Christopher Mayne2, Weiyi Toy3, Yang Shen4, Abhishek Sharma2, Srinivas Panchamukhi1, Jason Nowak5,
Kendall W Nettles5, Sarat Chandarlapaty3, John A Katzenellenbogen2 and Geoffrey L Greene1. 1University of Chicago, Chicago,
IL; 2University of Illinois, Urbana-Champaign, IL; 3Memorial Sloan Kettering Cancer Center, New York, NY; 4Toyota Technological
Institute, Chicago, IL and 5Scripps, Jupiter, FL.
Body: The estrogen receptor alpha (ER) is a member of the nuclear hormone receptor (NHR) family and is critical for the
etiology and treatment of breast cancer. Approximately 70% of breast cancers express ER and many of these are sensitive to
anti-estrogen therapies. Selective estrogen receptor modulators (SERMs), such as tamoxifen and raloxifene, are approved to
treat or reduce the risk of ER-dependent breast cancers. SERMs act by competitively binding to the ER ligand-binding domain
(LBD). Unfortunately, metastatic breast tumors recur in approximately half of patients and become SERM resistant while
remaining ER-positive in many cases. Recently, conserved somatic mutations in the ER LBD were identified in patients who
received SERM/aromatase inhibitor (AI)/selective estrogen receptor disruptor (SERD) therapy for an average of five years.
Because these mutations were observed in approximately 25% of tumors and the most frequent mutations (Y537S or D538G)
were located in or just prior to helix 12 (H12), the molecular switch that controls AF-2 activity, they represent a possible
mechanism for acquired SERM insensitivity for a significant population of patients. Further studies revealed that these mutations
conferred hormone-independent ER activity and that the inhibitory efficacy of currently approved SERMs was reduced. Our goal
is to understand the effects of these mutations on the structure and function of ER in these tumors to guide the generation of
novel compounds which bypass the effects of these somatic mutations. Here, we employ x-ray crystallography, molecular
dynamics simulations, biochemical assays in addition to breast cancer cell proliferation assays to dissect the role of somatic
mutation in acquired hormone/SERM resistance. X-ray crystal structures of the ER LBD D538G mutant in the unliganded (apo),
agonist and SERM-bound states, combined with molecular dynamics simulations, reveal a stabilized loop between H11 and H12
that allows the receptor to preferentially adopt an agonist conformation versus an antagonist conformation. The biochemical and
breast cancer cell proliferation assays reveal structural insights that may explain mutant ER function within the tumor. Further,
we use these methods to explore the utility of next generation SERMs and SERDs to inhibit these mutant ERs as well as to guide
the synthesis of additional novel compounds. Importantly, our work is expected to yield more potent and effective SERMs/SERDs
that overcome the impact of acquired activating mutations and result in improved patient survival.

Title: Effect of a novel selective progesterone receptor modulator EC312 for menopausal hormone therapy (MHT) in comparison
with Bazedoxifene in vitro
Hareesh B Nair1, Bindu Santhamma1, Walter Elger1 and Klaus J Nickisch1. 1Evestra, Inc, San Antonio, TX.
Body: Significant increase in the breast cancer risk reported from the Womens health initiative (WHI) study and other clinical
studies in postmenopausal women lead to a significantly reduced usage of estrogen/ progestin based HRT treatment in women.
A new and interesting approach to provide women with an effective treatment for vasomotor symptoms without an increased
breast cancer risk has been reported in preclinical and clinical studies. A combination of conjugated estrogens (CE) + selective
estrogen receptor modulator (SERM), bazedoxifene (BZA) offers an effective treatment for overcoming vasomotor symptoms but
preventing the proliferation of breast tissue. However, this treatment also has some drawbacks. A higher rate of induced
endometrial hyperplasia due to unopposed estroginicity with BZA+ CE treatment regimen has been noted in recent randomized
clinical trials. In addition, the risk of venous thromboembolic events has been shown not avoided completely in women who were
older at the time of BZE+CE treatment compared with those who were younger. Therefore an unmet need for new HRT regiments
that combine effective treatment of symptoms without increasing the risk of breast cancer in postmenopausal women.
A potential approach is the combination of estradiol with a partial agonistic antiproprgestin (mesoprogstin or SPRM). In order to
test the hypothesis we have compared the combinatory effect of BZA with estrogen (E2) as well as a novel SPRM, EC312+ E2 in
a cell culture model of postmenopausal breast cancer. In our model, we have found that the antiestrogenic effect of BZA on T47D
cell growth was comparable that of EC312. Both EC312 and BZA increased apoptosis dose dependently in the presence of E2 as
well as reduced E2 mediated BrdU incorporation as a read out of cell proliferation. E2 (1nM) treatment increased the expression
of anti-apoptotic genes whereas 100nM treatment of EC312 and BZA decreased the levels of anti-apoptotic genes. Expression of
E2 stimulated effect on target genes such as cMyc, pS2 and Cyclin D1 were reduced with the treatment of EC312 or BZA. It is
evident from our study that EC312 as a novel SPRM exerted no agonistic effects on human breast cancer cells and effectively
blocked the stimulatory actions of E2 or conjugated estrogens. The in vitro molecular characterization of EC312 was performed
using gene transactivation assays and confirmation of EC312 as a PRM was determined in cycling guinea pig model. Our future
animal model with breast cancer xenografts will underscore the possibility of using EC312 as a safer SPRM that antagonize the
growth promotional effect of estrogens as well as eliminating the increased breast cancer risk of progestin component of MHT.
Further rationale for choosing SPRM/mesoprogestin such as EC312 would stabilize the uterine tissue and endometrium to
minimize the oligo-menorrhea and endometrial hyperplasia associated with MHT.


Title: Glucocorticoid and aldosterone mimic progestin-induction of a therapy-resistant cytokeratin-5 positive cell population in
estrogen receptor-positive breast cancer through a Bcl6-dependent mechanism
Hallgeir Rui1, Takahiro Sato1, Amy Peck1, Melanie A Girondo1, Chengbao Liu1, Albert J Kovatich2, Jeffery A Hooke2, Craig D
Shriver2, Edith Mitchell1, Terry Hyslop3 and Chelain Goodman1. 1Thomas Jefferson University, Philadelphia, PA; 2Walter Reed
National Military Medical Center, Bethesda, MD and 3Duke University, Durham, NC.
Body: Resistance to anti-estrogen therapy remains a significant problem in patients diagnosed with estrogen receptor- (ER)
positive breast cancer. Recent progress has defined a "lumino-basal" subclass of ER-positive breast cancer characterized by
mosaic presence of a minor population of ER-negative cells expressing the basal cytokeratin-5 (CK5). The CK5-positive cells
are therapy-resistant and have increased tumorigenic potential. Initial studies have suggested that progestins but not other
steroids expand this CK5+ cell population. Unexpectedly, we discovered that at least two 3-ketosteroids other than progestins,
glucocorticoids and mineralocorticoids, are capable of inducing the CK5+/ER- cell population. CK5+ cells induced by
glucocorticoid or aldosterone showed increased clonogenicity in soft agar, expressed the stem cell marker CD44, showed loss of
ER and PR expression, and demonstrated therapy-resistance with reduced apoptosis in response to chemotherapy, and were
further enriched following adjuvant antiestrogen or chemotherapies. Induction of CK5+ cells by 3-ketosteroids was consistently
preceded by induction of Bcl6, a transcriptional repressor implicated in breast cancer progression. Suppression of Bcl6 by shRNA
or the Bcl6 suppressor, prolactin, abolished 3-ketosteroid-induction of CK5+ cells. Prolactin also blocked 3-ketosteroid-induced
colony formation in vitro and suppressed progestin-induction of the CK5+ cell population in T47D xenograft tumors in vivo.
Survival analyses with recursive partitioning revealed that CK5 and Bcl6 transcripts or protein levels in ER+ breast cancer
identify patients at high or low risk for tumor in two independent cohorts. The observations provide a mechanism by which
stress-related or pharmacologic elevation of glucocorticoids may adversely affect patients with ER+/CK5+ breast cancer, and
may justify further exploring of inhibitors to 3-ketosteroid receptors or Bcl6 for therapeutic benefit.

Title: Glutamine metabolism promotes survival through the unfolded protein response in endocrine resistant breast cancer
Ayeasha N Shajahan-Haq1, Susan Demo2 and Robert Clarke1. 1Georgetown University, Lombardi Comprehensive Cancer Center,
Washington, DC and 2Calithera Biosciences, South San Francisco, CA.
Body: About 70% of all breast cancers are estrogen receptor alpha positive (ER+). Anti-hormone therapy such as antiestrogens
(e.g.,Tamoxifen; TAM) are often used to treat ER+ breast cancer but breast cancer cells can develop resistance to these drugs
(endocrine resistance). Unfortunately, 50% percent of all antiestrogen treated tumors eventually develop endocrine resistance,
and therefore, there is an urgent need to find ways to treat this incurable disease. Endocrine resistant cells survive antiestrogen
treatment by initiating a pro-survival pathway mediated by an evolutionary conserved process call the unfolded protein response
(UPR) within the endoplasmic reticulum. Cellular stress induced by therapeutics can be initiated by the UPR leading to a
pro-survival response in the short-term and a pro-death response in the long-term, depending on the magnitude of the stress
signal. Our studies show that antiestrogens can decrease glucose uptake in ER+ breast cancer cells, and in glucose-deprived
conditions, presence of glutamine in the media can trigger the UPR through GRP78-IRE1a. Subsequently, a
glutamine-dependent pathway can induce cell death via GRP78-IREa-JNK-CHOP and survival in others via GRP78-IRE1a-XBP1.
Glutamine metabolites such as glutamate (essential substrate for many vital cellular processes) and proline (substrate for
collagen production) are significantly elevated in endocrine resistant cells. Knockdown of glutaminase (GLS1) with siRNA or small
molecule inhibitor, CB-839, significantly decreased cell proliferation in endocrine resistant cells compared with sensitive cells
within 48 h. Moreover, cell lines derived from endocrine resistant cells that are grown in glucose-free media are significantly more
sensitive to CB-839 compared with their respective controls, indicating an increased dependency on glutamine. Glutamine may
offer an alternate source of energy and raw materials in periods of glucose deprivation, and in endocrine resistant cells, this
adaptive pro-survival cellular metabolic mechanism is up-regulated. Thus, CB-839 may be a potent anti-cancer agent in treating
antiestrogen resistant breast cancers.

Title: Circulating oxysterol metabolites as potential new surrogate markers for hormonotherapy in patients with hormone
receptor-positive breast cancer? A pilot study
Florence Dalenc1, Luggi Iuliano2, Thomas Filleron3, Maud Voisin4, Henri Roch1, Sandrine Silvente-Poirot4 and Marc Poirot4.
1
Claudius Regaud Institut, Toulouse, France; 2Department of Medico-Surgical Sciences and Biotechnologies, Rome, Italy;
3
Claudius Regaud Institut, Toulouse, France and 4Claudius Regaud Institut-INSERM UMR 21037, Toulouse, France.
Body: Background: Clinicians need new predictive biomarkers of response to hormonal therapy in patients (pts) with hormone
receptor-positive (HR+) breast cancer (BC). Current treatments used the selective estrogen receptor modulator tamoxifen (Tam)
or aromatase inhibitors (AI). Several data from the literature and our preclinical results indicate that cholesterol metabolism
pathway is involved in BC oncogenesis and sensitivity/resistance to Tam. Our team has established that Tam modulates
cholesterol oxydative metabolism and modulate oxysterol (OS) levels in vitro and in vivo on BC cell lines. Tam is a potent inhibitor
of the cholesterol-5,6-epoxide hydrolase (ChEH), which led to the accumulation of cholesterol epoxides (CEs) and the inhibition
formation of cholesterol triol (CT). Importantly, we found that CEs mediated the cytotoxicity of Tam in BC. Therefore, an increase
in CEs, and no increase in CT could be markers of the efficacy of Tam reflecting its inhibition of ChEH in pts. On the other hand
27-hydroxycholesterol (27HC), another OS, is an estrogen receptor ligand with tumor promoter properties in BC. The 27HC level
could constitute a new risk factor for BC development that has to be measured. The impact of AI on cholesterol metabolism is
totally unknown, it is thus important to study its impact on OS levels in pts.
Methods: We have conducted a monocentric, prospective, clinical trial in pts who must received Tam or AI in adjuvant or
metastatic setting for a HR+ BC. The primary end point was the feasibility of detection of circulating OS (CEs, CT and 27HC) in
the serum of patients before the first administration of hormonal therapy and at 1 month. Key secondary end points were to
measure variations in the concentration of OS according to patients and treatments. 12 different OS including CEs, CT and 27HC
were quantified by GC/MS.
Results : bCE relative concentration significantly increased in the entire population (p=0.0109) while no increase in CT was
measured under HT treatment establishing that Tam inhibited ChEH in pts. It should be noted that an important inter-individual
variability in pts was observed according the OS species considered. AI stimulated the accumulation of CE (p=0.0022)
suggesting that they modulate CEs metabolism. We found that AI were not direct inhibitors of ChEH in BC cells suggesting that
they modulate CEs level through a different mechanism than Tam. Importantly, we found that letrozol, but not exemestane or
Tam, increased the blood level in 27-HC. This suggests that letrozol increased a factor of BC risk since 27HC is a tumor promoter
which may be involved in BC recurrence.
Conclusion: This pilot study provides the first evidence that circulating OS could be measured in the blood of pts with BC. The
clinical utility of OS as biomarkers of sensibility/resistance to hormone therapy needs further clinical investigations. Based on the
present study the CE/27HC ratio should be more specifically investigated. The mechanisms involved in the modulation of OS by
AI deserve further studies.

Title: Overexpression of insulin receptor substrate 4 can mediate acquired resistance to lapatinib-containing regimens in HER2+
breast cancer cells
Lanfang Qin1, Maria B Hahn1, Xiaoyong Fu1, Martin J Shea1, Mario Giuliano1, Sarmistha Nanda1, Xiaowei Xu1, Huizhong Hu1,
Sung Yun Jung1, Laura M Heiser2, Nicholas Wang2, Joe W Gray2, Susan G Hilsenbeck1, Chad Creighton1, Chad A Shaw1, Gary C
Chamness1, Dean P Edwards1, Sabrina Herrera1, Carolina Gutierrez1, C Kent Osborne1 and Rachel Schiff1. 1Baylor College of
Medicine, Houston, TX and 2Oregon Health & Science University, Portland, OR.
Body: Background: The HER2 pathway can be inhibited by potent targeting agents such as lapatinib (L), trastuzumab (T), or their
combination (LT), but acquired and de novo resistance still occur. Resistance to these drugs remains a major hurdle in the
management of HER2+ breast cancer. Consequently, elucidation of mechanisms of acquired therapeutic resistance to
HER2-directed therapies is of critical importance.
Methods: To obtain clues to the mechanisms for resistance we developed a panel of HER2+ breast cancer cell lines resistant to
L, T, or LT. Parental cells and resistant derivatives of the HER2+ BT474 cell line were characterized by RNA-seq. Genes that
were overexpressed in resistant compared to parental cells were confirmed by RT-PCR, Western blotting, and
immunohistochemistry (IHC). Cell growth and cell signaling were assessed in parental and resistant cell lines after
down-regulation (by siRNA) or overexpression (via an inducible cDNA) of IRS4 in the presence or absence of treatment. The
effect of IRS4 overexpression on L resistance was assessed in a BT474 xenograft model. The proteins that interact with IRS4
were identified by co-immunoprecipitation with IRS4 followed by separation of the associated proteins by SDS-PAGE and
microsequencing by mass spectrometry.
Results: RNA-seq analysis revealed that IRS4 was the most up-regulated gene in BT474 L or LT resistant derivatives in which
HER2 signaling is effectively inhibited, but not T alone, where HER2 signaling is reactivated. Western blotting and IHC validated
this result and identified membrane localization of IRS4. Knockdown of IRS4 in L- or LT-resistant cells reversed resistance and
restored growth inhibition. IRS4 knockdown also inhibited downstream signaling, with a reduction in pAKT but not in pMAPK.
Induction of the cell cycle regulator p27 and down-regulation of survivin were observed after IRS4 knockdown. Overexpression of
IRS4 cDNA in parental BT474 and SKBR3 cells led to resistance to L/LT, increased pAkt, and decreased the apoptotic marker
cleaved PARP in the presence of L or the LT combination. The BT474 xenograft model showed that IRS4 overexpression in the
absence of treatment had no effect on tumor growth but it significantly reduced the inhibitory effect of lapatinib (p=0.002). A group
of proteins that interact with IRS4 in BT474 L-resistant cells were identified by mass spectrometry. The roles of these proteins in
IRS4-mediated resistance to lapatinb-containing regimens are under investigation.
Conclusion: IRS4 overexpression is a critical factor in causing resistance to lapatinib-containing regimens in BT474 cells.
Investigation of IRS4 and its signaling partners in HER2+ human tumors resistant to lapatinib will be important to determine if this
mechanism is also operative in patients.

Title: Modeling chemoendocrine therapy for ER+/p53wt luminal breast cancer
Shiliang A Cao1, Elgene Lim1,2, Myles Brown1,3 and Shannon T Bailey1. 1Dana-Farber Cancer Institute, Boston, MA; 2Olivia
Newton John Cancer & Wellness Centre, Heidelberg, Victoria, Australia and 3Harvard Medical School, Boston, MA.
Body: Background
ER+ breast cancers are predominantly p53 wildtype (wt). Despite this genotype, these cancers are relatively resistant to
chemotherapy-induced apoptosis in the presence of estrogen. Thus, current clinical practice is to administer hormonal therapy
and chemotherapy sequentially rather than concurrently. Using the ER+/p53wt cell line MCF-7, we previously demonstrated that
ER bound by either the ER agonist estradiol or partial antagonist tamoxifen inhibits the expression of a set of proapoptotic p53
target genes, whereas the full ER antagonist fulvestrant, a selective ER downregulator (SERD), removes the ER-mediated
suppression of these genes, sensitizing breast cancer cells to p53-mediated cell death .We hypothesize that effective
chemoendocrine therapy requires complete ER antagonism resulting in ER downregulation, increased p53 activity, and ultimately
apoptosis.
Methods:
To model mechanisms of tumor sensitivity and resistance to therapy in vivo, therapeutic experiments were performed in
immunodeficient mice bearing ER+/p53wt tumor xenografts derived from a 61-year-old African-American female with grade 2
invasive ductal carcinoma. The mice were randomly assigned to the following treatment groups: A) fulvestrant B) tamoxifen, C)
doxorubicin, D) fulvestrant plus doxorubicin, E) tamoxifen plus doxorubicin, and F) control. Tumors were measured weekly for 42
days to determine treatment effects on growth. In addition, RNA was harvested from tumors at an early time point for RNA
sequencing to determine genes regulated by the treatments.
Results:
Combination therapy comprising doxorubicin plus fulvestrant resulted in tumor regression compared with single-agent or
doxorubicin plus tamoxifen treatment, which inhibited the growth of tumors but did not lead to their regression.
Conclusions:
Current paradigms for the treatment of ER+ breast cancer either in the adjuvant or advanced setting have involved the sequential
use of endocrine therapy and chemotherapy. This protocol is partially based on evidence of potential antagonism between
tamoxifen and chemotherapy as observed in early studies. Our published work in vitro and the preclinical study presented here
suggest that complete ER antagonism with SERDs such as fulvestrant is required to overcome the ability of ER to block
p53-mediated apoptosis. Our results suggest treatments involving fulvestrant concurrent with chemotherapies involving p53
activation should be considered for the treatment of patients with ER+/p53wt breast cancer.

Title: Divergent activation of AKT1 and AKT2 isoforms downstream of PI3K mutation impacts response of breast cancer cells to
estradiol and PI3K inhibitors
Harikrishna Nakshatri1, Chirayu Goswami1, Sunil Badve1, Luca Magnani2, Mathieu Lupien3 and Poornima Bhat-Nakshatri1.
1
Indiana University School of Medicine, Indianapolis, IN; 2Imperial College, London, United Kingdom and 3Princess Margaret
Cancer Center, Toronto, ON, Canada.
Body: Background: PI3K mutations are observed in 30% of breast cancers, which are more common in estrogen receptor (ER)
positive breast cancers compared to ER-negative breast cancers. AKT, a the major kinase downstream of PI3K, modulates ER
activity. It is unknown whether PI3K mutation leads to preferential activation of specific AKT isoform with ability to modulate ER
function.
Results: We report elevated AKT1 but not AKT2 mRNA in breast cancers with PI3K mutation. Breast epithelial cells with targeted
substitution of PI3K with PI3K-E545K or PI3K-H1047R demonstrated elevated AKT1_pS473 compared to AKT2_pS474, distinct
AKT substrate phosphorylation, and enhanced Estrogen Receptor (ER) activity. AKT1 had a dominant role in ER:estradiol
(E2)-dependent gene expression and proliferation. We have identified an unique gene expression signature in ER-positive
breast cancers that is dependent on ER, estradiol (E2), AKT1, and the pioneer factor FOXA1. Elevated expression of this
signature in ER-positive tumor was associated with better response to endocrine therapy and outcome. In addition, AKT1
determined the sensitivity to PI3-specific inhibitor BYL719 and pan-PI3K inhibitor BKM120. In contrast, AKT2 controlled global
gene expression and elevated levels of an AKT2-directed protein signature predicted poor outcome in breast cancer patients.
Conclusions: PI3K mutation favors AKT1 activation leading to imbalance in AKT isoform-specific kinome/proteome and an effect
on specific signaling pathways, particularly ER:E2 signaling network. Knowledge gained from this functional dissection of AKT
isoform activity downstream of PI3K mutation can be exploited for therapeutic stratification, particularly for anti-estrogen and PI3K
inhibitor-based therapies.

Title: The effect of HER-2/neu inhibition on prolonging clinical benefit with fulvestrant treatment for metastatic estrogen receptor
positive breast cancer patients treated with trastuzumab
Mahmoud Charif1, Elyse E Lower1, Diane Kennedy1, Harriet Kumar1, Shugufta Khan1, Neetu Radhakrishnan1 and Xiaoting
Zhang1. 1University of Cincinnati, Cincinnati, OH.
Body: Background: Fulvestrant is a well-established treatment for postmenopausal patients with estrogen receptor (ER) positive
metastatic breast cancer, and some patients experience prolonged clinical benefit exceeding one year. HER2 activation is a
major cause of endocrine resistance, and cross-talk between HER2/neu and ER coactivator MED1 regulates tamoxifen
resistance in breast cancer cells. (Cancer Res 2012 1;72(21):5625-34.). In a xenograft mouse model, suppression of MED1
enhanced tumor growth inhibition by fulvestrant in HER2/neu overexpressing breast cancer cells (PLoS One 20123 30;8(7)).
Objective of study: To determine if blocking the HER2/neu receptor with trastuzumab can improve response to fulvestrant.
Methods: This was an IRB approved record review of patients from three medical oncologists with biopsy-proven ER+ metastatic
breast cancer treated with fulvestrant. Demographic data collected included age at diagnosis, type and stage of cancer, original
and metastatic ER, progesterone receptor (PR), and HER2/neu biomarkers, and site(s) of metastasis, and primary local and
systemic treatment. All patients with HER2/neu positive primary tumors received trastuzumab. The duration of fulvestrant therapy
was calculated. Time to clinical disease progression on fulvestrant was measured as a surrogate for duration of clinical benefit.
Based on the median duration of therapy of 425 days, patients were divided into two groups: Short Treatment (< 425 days)
versus Prolonged Treatment (>425 days). Results: One hundred metastatic ER+ fulvestrant treated breast cancer patients with
documented duration of therapy were identified. There was no difference between the Short and Prolonged Treatment Groups in
regards to age, sites of metastases, or use of adjuvant endocrine or chemotherapy. Eighty five patients had recorded HER2/neu
tumor status. All 11 of 85 (13%) patients with documented HER2/neu positive primary tumors received trastuzumab. Patients with
HER2/neu positive tumors tended to have longer durations of fulvestrant therapy (772 (51-1911) days (median (range)) compared
to HER2/neu negative patients (360 (60-2739) days, p=0.059). Only 2 of 45 (4%) tumors from the Short Treatment Group were
HER2/neu positive, while 9 of 40 Prolonged Treatment Group patients with documented HER2/neu status were positive (Fishers
exact test p<0.021). Patients with HER2/neu positive tumors were more likely to experience prolonged responses to therapy with
an odds ratio of 6.2 (1.26 to 30.92 95% confidence interval, p=0.0249). Conclusion: Overexpression of HER2/neu in tumors from
ER+ metastatic breast cancer patients treated with trastuzumab was associated with a prolonged response to fulvestrant therapy.

Title: Haddows paradox revisited: Anti-estrogen withdrawal induces ER-driven apoptosis in anti-estrogen-resistant cells
Sarah R Hosford1 and Todd W Miller1. 1Dartmouth College, Hanover, NH.
Body: Anti-estrogen therapies that antagonize ER transcriptional activity have improved survival in many patients. However,
resistance to anti-estrogen therapies is common, resulting in disease recurrence in 20% of patients within 10 years of follow-up.
The current clinical strategy for managing ER+/HER2- breast cancer is treatment with anti-estrogens until disease progression,
followed by subsequent lines of anti-estrogens and chemotherapies. However, before tamoxifen was approved in the 1970s,
estrogens were used for treatment of breast cancer with response rates similar to those elicited by anti-estrogens in the advanced
setting. Similarly, withdrawal of anti-estrogen therapy has shown anti-tumor effects in some cases. In 1970, Alexander Haddow
summarized decades of observations that longer time intervals after menopause were associated with increased clinical benefit
from estrogen therapy. These studies suggest that the return of ER signaling after a prolonged period of ER suppression (via
menopause, or anti-estrogen therapy) may inhibit tumor growth. Preclinical studies have provided some insight on the mechanism
underlying estrogen-induced apoptosis, but the anti-tumor effects of estrogen remain unclear, and biomarkers to identify patients
that would benefit from estrogen therapy have not been identified.
We found that MCF-7 cells with acquired resistance to the selective ER downregulator fulvestrant (MCF-7/FR) retain ER
expression and harbor ESR1 (ER) gene amplification. Upon withdrawal of fulvestrant, these cells engage ER as demonstrated by
increased luciferase transcriptional reporter activity and re-expression of select proteins encoded by ER-inducible genes.
Following 8 days of fulvestrant withdrawal, MCF-7/FR cells show drastically decreased proliferation and increased apoptosis,
concomitant with re-engagement of ER activity. Temporal cell cycle analysis revealed a reduction of cells in G0/G1 phase
following 11 days of fulvestrant withdrawal, which paralleled decreases in cell cycle proteins E2F1 and phospho-Rb, as well as
the anti-senescence protein FoxM1. Transcriptomic analyses confirmed that fulvestrant withdrawal progressively induces gene
expression patterns indicative of stress and senescence. Similar effects were observed in long-term estrogen-deprived (LTED)
MCF-7 cells treated with 17b-estradiol. In contrast, withdrawal of fulvestrant from T47D/FR cells, which do not retain ER
expression, does not induce cell death or re-engage ER activity. Thus, re-engagement of ER in anti-estrogen-resistant cells may
be required for anti-tumor effects of anti-estrogen withdrawal. We postulate that in MCF-7/FR cells relieved of fulvestrant, high ER
levels drive aberrantly regulated transcription through ER hyperactivation, and altered ER/DNA binding patterns result in
estrogen-induced apoptosis. Gene expression signatures suggest that fulvestrant withdrawal from MCF-7/FR cells induces
downregulation of the transcription suppressive activity of the Polycomb Repressive Complex 2 (PRC2) that contains EZH2
methyltransferase. Ongoing studies are characterizing the roles of EZH2 and ER to determine therapeutic potential for
anti-estrogen-resistant breast cancer.

Title: RNA-seq reveals lncRNAs associated with hormonal influences and endocrine therapy
Nilanjana Banerjee1, Wim Verhaegh2, Patrick Cheung1 and Nevenka Dimitrova1. 1Philips Research North America, Briarcliff
Manor, NY and 2Philips Research High Tech Campus, Eindhoven, Netherlands.
Body: Background: Long non-coding RNAs (lncRNAs) are pervasively transcribed in the genome yet their role in human
disease is not well understood. LncRNAs can have regulatory effects on coding mRNAs through a number of mechanisms,
including repressing their sense-strand protein-coding partners. There is also emerging evidence that hormone signaling affects
the expression of a wide variety of lncRNAs. However, the role of lncRNAs in the response mechanisms to endocrine therapy
remains mostly unknown. We investigated MCF7 and T47D cell-line RNA-seq data to characterize changes in lncRNA expression
and evaluate their response to estradiol (E2), progesterone (P4) and to their antagonists, tamoxifen (TAM) and mifepristone
(RU486).
Methods: RNA-seq data from MCF7 cell-lines with response to E2 and TAM and T47D cell-lines with response to P4 and RU486
were obtained from publicly available short read archive (SRA: ERP000992). Expression values after hormonal or antagonist
treatments were compared to non-treated cells. We independently sequenced total RNA samples of T47D cell-lines with or
without E2 treatment and created a robust list E2-responsive lncRNAs for further analysis. The RNA-sequencing data was aligned
to a lncRNA database containing 14,572 unique lncRNAs. Significant log-fold changes in relative abundance of lncRNA
transcripts were selected using percentile-p-value. ER-responsive binding sites on or near the ER-associated lncRNAs were
investigated in a ChIP-seq study in MCF7 cells following estrogen treatment. Finally, we identified the subset of
hormonally-responsive lncRNAs that are also evolutionarily conserved.
Results: On average, in each cell-line experiment, 5000 lncRNAs were detectable. Expression of 55 lncRNAs were associated
with differing response to E2 and TAM (p<0.01) while 120 lncRNAs were associated with differing response to P4 and RU486
(p<0.01). To probe whether similar lncRNA-related response mechanisms are at play for both types of hormonal influences, we
examined the overlap between the two sets of significantly responsive lncRNAs (>3-fold change between response to agonist and
antagonist). Less than 10% of the lncRNAs were common among the two sets suggesting that while there might be some
cross-talk, E2 and P4 are generally regulating unique sets of lncRNAs. For example, XLOC_013954, a bona-fide E2-responsive
lncRNA shows 5-fold difference between response to E2 and TAM, but has a fold change of mere 1.1 between response to P2
and RU486. From the MCF7 ChIP-seq data we found that >50% of the ER-associated lncRNAs had ER binding site either
overlapping or neighboring the lncRNA. Less than 30% of the response-associated lncRNAs were evolutionarily conserved
suggesting they might be under different selection pressures.
Discussion: We have shown that lncRNA expression levels are associated with hormonal influences and endocrine therapy. We
have further established that the lncRNAs impacted by the different hormonal influences are quite exclusive and suggest separate
regulatory networks. We are also exploring lncRNA-mRNA expression for coding partners of the response-associated lncRNAs.
Understanding the regulatory effects of lncRNA expression opens up new opportunities for stratification and management of
breast cancers.

Title: A new cell panel to study oestrogen receptor loss in acquired endocrine resistant breast cancer
Julia M Gee1, Mei Yee Meng1, Richard A McClelland1, Huw J Mottram1, Susan R Kyme1, Pauline Finlay1, Lindy Goddard1, Carol M
Dutkowski1, Denise Barrow1, Walther Parson2 and Heidi Fiegl2. 1Cardiff University, Cardiff, United Kingdom and 2Innsbruck
Medical University, Innsbruck, Austria.
Body: Background: Oestrogen receptor positive (ER+) breast cancer patients can acquire endocrine resistance and 10-20%
tumours have lost ER at relapse. While growth factor pathway hyperactivity and ER promoter methylation contribute to de novo
ER negativity, ER loss in acquired resistance is largely unexplored. We have recently developed 11 lines from MCF7, T47D,
BT474 & MDAMB361 cells to model acquired resistance emerging with prolonged (3 year) endocrine treatment both in
ER+/HER2- and ER+/HER2+ disease. Here we establish prevalence of acquired ER loss in the panel, determine any association
with aggressiveness, and explore ER loss mechanisms in acquired endocrine resistance.
Methods: Authenticated acquired resistant models derived from endocrine responsive lines cultured for 3 years with 10-7M
tamoxifen (TamR), 10-7M fulvestrant (FasR) or oestrogen deprivation (5% charcoal-stripped foetal calf serum SFCSR) were
profiled for ER & PR by PCR, immunocytochemistry and Western blotting (+/- 1-2wk antihormone withdrawal), for 2nd-line
endocrine responsiveness and for migration using Boyden chamber assays vs. time-matched controls. Src kinase, EGFR, HER2,
MAPK & AKT activity were examined and whether their respective inhibition using saracatinib or gefitinib (1M), trastuzumab
(100nM), U0126 or LY294002 (5M) for 1wk restored ER. ESR1 promoter methylation was examined by bisulfite modification &
MethyLight PCR.
Results: Substantial ER mRNA & protein loss occurred in 7/11 long-term acquired endocrine resistant lines. This was irreversible
by antihormone withdrawal and paralleled by complete PR loss and endocrine growth-insensitivity. While seen in all fulvestrant
resistant lines, ER loss was less frequent with tamoxifen (in MCF7TamR & MDATamR) and only seen in oestrogen deprived
resistant T47DSFCSR cells. Increased migration accompanied acquired ER loss in ER+/HER2- derived MCF7TamR, MCF7FasR
& T47DSFCSR cells and was saracatinib-sensitive. Src and further growth factor pathway activity increased in several acquired
resistant models, and ER loss associated with increased EGFR/HER2 in the MCF7- & MDA-derived cells and with increased
MAPK activity in all lines. Weak ER recovery was seen in antioestrogen resistant models treated with saracatinib (MDATamR,
MCF7FasR), gefitinib (MDATamR, BT474FasR, MCF7FasR/TamR) or trastuzumab (MCF7TamR). ESR1 DNA methylation was
only prominent in MDATamR and MCF7TamR cells. No inhibitor restored ER in the T47D-derived cells, including T47DSFCSR
which also lacked ESR1 methylation.
Conclusions: Although ER loss is very prominent in this acquired resistant cell panel, it demonstrates there is capacity of
prolonged antihormones, chiefly antioestrogens, to promote receptor loss independent of initial HER2 status. Acquired ER loss
clinically would be expected to confer endocrine insensitivity and poorer prognosis given the panel findings. Where ER loss
emerged with antioestrogens there was some mechanistic-overlap with de novo ER negativity, including ER promoter methylation
for acquired tamoxifen resistance. Our future studies will use the panel to address if targeting these mechanisms can be
optimized for ER recovery or if further mechanisms also drive ER loss in acquired resistance, notably for prolonged oestrogen
deprivation.

Title: ESRP1 adds sp(l)ice to endocrine resistance
Yesim Gokmen-Polar1, Yaseswini Neelamraju1, Chirayu Pankaj Goswami1, Harikrishna Nakshatri1, Sarath Chandra Janga1 and
Sunil Badve1. 1Indiana University School of Medicine, Indianapolis, IN.
Body: Introduction:
De novo or acquired resistance to endocrine therapy limits its utility in a significant number of estrogen receptor (ER) breast
cancers. An increasing number of molecular assays predict the likelihood of distant recurrence in tamoxifen-treated patients with
node-negative, ER+ breast cancer. However, these do not provide the mechanistic basis for endocrine resistance. It is crucial to
identify novel targets and improve the success of endocrine therapies. We previously shown that epithelial splicing regulatory
proteins 1 and 2 (ESRP1 and ESRP2), RNA binding proteins that promote splicing, are significantly elevated in cases with high
Oncotype DX scores (innate resistance) and in ER-positive cells with acquired tamoxifen resistance (MCF-7 LCC2 cells) and
fulvestrant and tamoxifen resistance (MCF-7/LCC9 cells). The aim of this study was to investigate the ESRP1/ESRP2-regulated
alternative splicing events leading to innate and acquired endocrine resistance.
Methods:
A combinatorial bioinformatics approach was employed to identify genes altered through alternative splicing by ESRP1/ESRP2 in
endocrine resistance. Briefly, genes with ESRP1/ESRP2 motifs were screened in the human genome. This data was integrated
with tamoxifen-treated datasets, and filtered using protein-protein interactions (PPI; BIOGRID) and clinical outcome (KM plotter.
Potential target gene transcripts were further narrowed down using an algorithm based on motif binding location and the Cancer
Genome Atlas (TCGA) breast cancer datasets with low and high ESRP1 cases. The alternative transcripts were further validated
using splice variant specific-custom qRT-PCR in low and high RS Oncotype cases as surrogate for endocrine sensitivity and
resistance.
Results:
Motif analysis in combination with tamoxifen-treated datasets identified 2212 differentially expressed genes. Further collective
analysis of number of PPI (>50) and survival data from KM plotter (P<0.001) narrowed down 79 candidate genes that are
associated with tamoxifen resistance. TCGA data confirmed presence of distinct transcripts (splicing variants) based on ESPR1
expression levels. Splice specific RT-PCR confirmed alternative splicing events involving cycle related genes such as AURKA,
FZR1, and MDM2 in low ESRP1 (and low RS) and high ESRP1 (high RS) cases.
Conclusion:
ESRP1/ESRP2 by inducing alternative splicing play an important role in tamoxifen resistance and recurrence of ER+ breast
cancer. Targeting alternative splicing may offer novel avenues for combating endocrine-resistance in breast cancer.

Title: Investigation into the oncogenic potential of the androgen receptor in aromatase inhibitor resistant breast cancer
Leonie Young1, Laura Creevey1, Azlena Ali1, Arnold DK Hill1 and Marie McIlroy1. 1Royal College of Surgeons in Ireland, Dublin,
Ireland.
Body: The heterogenous nature of breast cancer, exemplified by a wide range of subtype classifications, has been well
documented. The identification of the novel molecular apocrine breast tumours (ER-, PR-ve, HER2-ve, AR+), with a gene
expression profile reflective of ER+ luminal breast cancer, has led to the proposition of an alternative role for the androgen
receptor as a pseudo ER. This proposition was borne out of global expression profiling and AR recruitment of molecular apocrine
cells, demonstrating that more than 50% of AR binding to the genome mirrors the pattern of ER binding suggesting that in the
absence of ER, AR can mimic ER in its DNA binding capabilities.
Aromatase Inhibitor (AI) therapy is the gold standard first line therapy for postmenopausal breast cancer. AIs function by
abrogating the activity of the enzyme Cyp 19 (aromatase), which is responsible for converting circulating androgens to estrogen.
By preventing this conversion step, this eradicates the main steroid driving breast cancer growth and in turn results in a more
androgenic environment. We hypothesise that AR may also act as a pseudo ER in our AI resistant (MCF7-letrozole resistant)
cells. Preliminary data demonstrates that AR is recruited to some but not all classical ER target genes as well as some AR target
genes in MCF7-LetR cells, therefore highlighting the potential of AR to act as an oncogene in this setting. Analysis of AR protein
expression also demonstrates higher levels of AR in AI resistant cells versus sensitive cells. In the evaluation of the
responsiveness of a range of breast cancer cell lines to the AR antagonist Bicalutamide, it was noted that our AI resistant cells
demonstrated responsiveness to Bicalutamide treatment.
RNA sequencing experiments following AR knockdown in AI resistant cells also provide crucial information as to the key genes
differentially regulated by AR. Given that resistance to AI therapy is an emerging clinical issue, further elucidation of the potential
mechanisms of AI resistance and potential AR targets is highly warranted.

Title: PI3K pathway as a novel target in androgen-driven aromatase inhibitor resistant breast cancer
Leonie S Young1, Azlena Ali1,2, Laura Creevey1, Fiona Bane1, Wan Afif Marzuki1,2, Arnold DK Hill1,2 and Marie McIlroy1. 1Endocrine
Oncology Surgical Research Group, Royal College of Surgeons in Ireland, Dublin 2, Ireland and 2Beaumont Hospital, Dublin 9,
Ireland.
Body: Breast cancer has the highest incidence and second highest mortality rate among all cancers in women worldwide (1). The
gold standard treatment for post-menopausal breast cancer is aromatase inhibitor (AI) therapy. Unfortunately, resistance is
inevitable in about 30-60% of these patients (2). Research from our lab has previously identified a transcription factor, HOXC11,
to be associated with breast cancer resistance and metastases (3). Prosaposin (PSAP) was then identified as a putative HOXC11
target gene in the AI resistant setting. PSAP is a secreted protein that also plays a role in metastatic prostate cancer in both a
ligand dependent and independent fashion (4,5).
We have previously shown PSAP to be an androgen responsive gene that can upregulate androgen receptor (AR) expression in
a ligand dependent manner in AI resistance. An AR antagonist was effective at slowing down cell proliferation in AI resistant cells
in vitro . However, since PSAP can also potentially act in a ligand independent fashion, this current study is focused on evaluating
the role of PSAP in PI3K pathway activation in the development of AI resistance. Treatment of cells with recombinant PSAP
protein (rhPSAP) resulted in increasing expression of p-AKT in a dose-dependent manner. To target this pathway of resistance,
we used a pan-class PI3K/mTOR inhibitor BEZ235 on our AI resistant cells. Functional studies with BEZ235 treatment showed
significant reduction in both cell motility as well as cell proliferation. Treatment of AI resistant cells with BEZ235 resulted in
decreased expression of p-AKT, with little or no effect on AR expression. These findings suggest that to prevent resistance to AI
therapy, combination treatment regimens including AR antagonists plus a PI3K inhibitor may ensure sustained response to
therapy.
References:
1. Servick K. Breast Cancer: a World of Differences. Science Magazine. Special Issue. March 2014
2. Kazi AA et al. Breast Nonhypoxic regulation and role of hypoxia-inducible factor 1 in aromatase inhibitor resistant breast
cancer. Cancer Research. 2014; 16:R15
3. McIlroy M et al. Cancer Res.2010;70(4);1585-94
4. Koochekpour Set al. Prosaposin is an AR-target gene and its neurotrophic domain upregulates AR expression and activity in
prostate stromal cells. J Cell Biochem. 2008; 104: 22722285.
5. Koochekpour S et al. Prosaposin is a novel androgen-regulated gene in prostate cancer cell line LNCaP. J Cell Biochem. 2007;
10: 631641.

Title: Novel, non-uterotrophic, selective estrogen mimics cause regression of tamoxifen-resistant breast cancer in 2D and 3D
cultures and in mouse xenograft models
Gregory RJ Thatcher1, Rui Xiong1, Hitisha K Patel1, Jiong Zhao1, Xiao Liang1, Yue-ting Wang1, Mary Ellen Molloy1 and Debra
Tonetti1. 1University of Illinois College of Pharmacy, Chicago, IL.
Body: Prior to the advent of tamoxifen therapy, estradiol (E2) and the ER agonist, diethylstilbestrol, showed efficacy in the clinical
in therapeutic treatment of breast cancer. The adverse effects of these therapeutic options led to discontinuation, despite
tamoxifen itself sharing the adverse effect of uterine growth. The perceived therapeutic action of tamoxifen in countering the
proliferative actions of E2 at the estrogen receptor (ER) in endocrine-sensitive breast tissues is not entirely compatible with the
therapeutic actions of E2 itself in breast cancer, and points to a more complex role for ER. Tamoxifen is the standard of care for
many patients with ER+ breast cancer; however significantly, 30-50% of women are resistant to tamoxifen therapy. Development
of a Selective Estrogen Mimic (SEM) that is effective in breast cancer without uterotrophic and other side effects is a novel
treatment option that we hypothesize will be effective in causing regression of tamoxifen-resistant breast cancer. We theorize that
a rational approach would be an ER ligand with pharmacological partial agonist actions, thus able to mimic E2 in regressing
breast cancer, but with antagonist actions in the face of excessive ER activation. Using the benzothiophene scaffold common to
the clinical Selective Estrogen Receptor Modulators (SERMs), raloxifene and arzoxifene, we hypothesized that an ideal SEM
could be designed by structural engineering. A library of over 30 compounds was developed and assayed in tamoxifen-resistant
cell lines, including MCF-7 and T47D. One SEM was assayed in two mouse xenograft models and shown to cause tumor
regression; impressively this SEM did not fuel growth of estrogen-dependent T47D xenografts and did not cause uterine growth.
This SEM and several other compounds were observed to act as partial agonists at ER, with variable actions at ER. To
understand this phenomenon further, the contributions of apoptotic mechanisms and cell cycle arrest to SEM activity was studied
in cell cultures. The results collected to date on SEMs that act as partial agonists at ER indicate that this is a viable, new
approach to breast cancer therapy.

Title: Adaptation to AI therapy in breast cancer can induce dynamic alterations in ER activity resulting in estrogen independent
metastatic tumours
Damir Vareslija1, Jean McBryan1, Ailis Fagan1, Aisling Redmond2, Yuan Hao3, Arnold DK Hill1 and Leonie S Young1. 1Royal
College of Surgeons in Ireland, Dublin, Ireland; 2University of Cambridge, Cambridge, Cambridgeshire, United Kingdom and 3Cold
Spring Harbor Laboratory, Clod Spring Harbor, NY.
Body: Acquired resistance to aromatase inhibitor therapy is a major clinical problem in the treatment of breast cancer. The
detailed mechanisms of how tumour cells develop this resistance remain unclear. Here, discovery of pro-survival adaptations to
the estrogen receptor (ER) signalling pathway in response to AI treatment is reported. Global ChIPseq analysis shows that the
ER retains binding activity in AI resistant cells under steroid-depleted conditions. In AI treated patients, evidence of steroid
independent adaptive ER signalling was demonstrated by the ER target gene Early Growth Response 3 (EGR3). Expression of
EGR3 decreased initially upon AI treatment but subsequently recovered following as little as 12 weeks of AI treatment. In vitro
data indicates that this increased expression of EGR3 may enhance cell growth and motility of tumour cells. Finally, evidence
from established metastatic tumours suggests that the ER signalling network may undergo further adaptations with disease
progression as EGR3 expression is routinely lost in the established metastatic tumour. Overall, these data provide evidence of a
dynamic ER response to AI treatment which may provide vital clues for overcoming the clinical issue of AI resistance.

Title: Efficacy of everolimus on multiple mechanisms of AI-resistance in vitro and xenograft, and characterization of their
everolimus-resistance
Mariko Kimura1,2, Natsu Fujiki1, Toru Hanamura1, Toshifumi Niwa1, Yuri Yamaguchi1,3, Takashi Ishikawa2, Itaru Endo2 and
Shin-ichi Hayashi1. 1Graduate School of Medicine, Tohoku University, Sendai, Miyagi, Japan; 2Yokohama City University,
Yokohama, Kanagawa, Japan and 3Saitama Cancer Center, Kita-adachi gun, Saitama, Japan.
Body: Background: mTOR inhibitor, everolimus showed remarkable clinical efficacy and has been considered as a promising
agent for breast cancer treatment especially among hormone receptor positive (HR+) postmenopausal advanced breast cancer.
Now Affinitor (everolimus) is approved for HR+ breast cancer in Japan. However, definitive biomarker has been unclear so far. In
addition, the subsequent treatment after refractory to everolimus has not been defined. In preclinical study, it has been reported
there are several resistant models of endocrine therapy and we also reported several types of models established from
MCF-7-E10, in which estrogen receptor (ER) activity could be monitored by transfected-ERE-GFP. Those EDR cells are
characterized as the followings: 1) Type 1 shows constitutive ER overexpression without estrogen, and PI3K/Akt/mTOR pathway
upregulated. 2) Type 2 shows low expression of ER (ERE-GFP negative) and IGF-1R/JNK signaling pathway upregulated (Fujiki,
2014, J Steroid Biochem Mol Biol). 3) Type4, which was cultured under the condition of estrogen depletion and addition of
testosterone. It showed using androgen metabolite as ligand and both PI3K/Akt/mTOR and MAPK signaling pathway upregulated
(Hanamura, 2013, Breast Cancer Res Tr). Methods: Efficacy of everolimus was analyzed in those cell lines in vitro and in vivo
about inhibition of cell proliferation. In vivo study, ovarectomized mice were inoculated with Type1 or Type4. Treatments were
performed as single agent letrozole, everolimus, or combination and tumor volume changes were compared. Then everolimus
resistant cells were generated from those EDR cells after long term exposure to everolimus in vitro. Though we could not acquire
everolimus resistant cell from Type4, Type1 and 2 showed its resistance. Those everolimus resistant (EVR) cells are equivalent
to resistance to the combination therapy between AI and everolimus after the first line AI treatment failure. Using those cell lines,
we investigated the different mechanism of resistance to everolimus. Results: Everolimus extremely inhibited cell proliferation in
each EDR type in vitro and in vivo. Besides, Type4 was more remarkably sensitive than parental MCF-7-E10 under testosterone
supplemented condition in vitro. Everolimus or combination treatments reached significant tumor volume reduction around 50
percent after 21 days treatment period, in contrast to increase of tumor volume in placebo or letrozole group. In vitro study of
developing resistance to everolimus, Type1-EVR cells no longer responded to everolimus and surprisingly fulvestrant also lost
sensitivity to EVR cell in spite of keeping ERE activity. Type1-EVR showed much higher expression of phosphorylation of
MAP-kinase than parental cell, however, MEK inhibitor or PI3K inhibitor alone was not effective but combination with fulvestrant
significantly effective more than parental cell. In type2, JNK inhibitor was effective for parental cell but EVR cell lost sensitivity to
JNK inhibitor, however, Src inhibitor was much more effective in Type2-EVR than in parental cell. Conclusion: Everolimus showed
remarkable efficacy to any types of EDR cells, however, the mechanism of acquired resistance to everolimus was likely to be
different between each EVR cell. This finding implies that different mechanisms of AI resistance follow different types of
everolimus resistance.

Title: Nuclear -catenin negativity predicts for late relapse in ER+, tamoxifen-treated breast cancer
Stephen Hiscox1, Chris Smith1, Robert I Nicholson1, Julia Gee1, Adrian Harris3, Judith Bliss4, Eleftheria Kalaizak4, Ivana Sestak6,
Mitch Dowsett5, Jack Cuzick6, Ian Ellis7 and Peter Barrett-Lee2. 1Breast Cancer Molecular Pharmacology Group, Cardiff
University, Cardiff, United Kingdom; 2Velindre Cancer Centre, Cardiff, United Kingdom; 3Weatherall Institute of Molecular
Medicine, University of Oxford, Oxford, United Kingdom; 4Institute of Cancer Research, Division of Clinical Studies, Sutton, United
Kingdom; 5Institute of Cancer Research, Division of Breast Cancer Research, London, United Kingdom; 6Wolfson Institute of
Preventive Medicine, Barts and The London School of Medicine and Dentistry, London, United Kingdom and 7Nottingham
University Hospitals, Nottingham, United Kingdom.
Body: Background: The annual recurrence rates of post-menopausal ER-positive breast cancers persist beyond the first 5 years
of diagnosis and treatment and the mortality rates in this period are higher versus ER-negative cancers. Extended endocrine
therapy past 5 years has been shown to be of benefit but is associated with increased toxicity and cost thus the ability to predict
patients who are at highest risk of late relapse would be of significant benefit in clinical decision making in this context. -catenin
is an intracellular protein that undergoes Wnt-mediated nuclear translocation where it transactivates genes implicated in tumour
development and progression. We have also previously reported that -catenin can play a role in aggressive resistance that
accompanies prolonged endocrine treatment in vitro. In this study, we thus investigated whether -catenin expression in ER+
breast cancer was predictive of recurrence beyond 5 years in an analysis of three separate endocrine-treated cohorts.
Methods: Associations between -catenin gene expression and relapse free survival (RFS) were performed using the online
KMplotter tool. Immunostaining for total -catenin was performed on tissue samples from 3 ER+ primary breast cancer series with
long-term follow-up data: Nottingham 2000 (n=384, tamoxifen-only); ABC (n=570; tamoxifen only); transATAC (n=743; tamoxifen
or ansatrozole). The association between subcellular (nuclear or cytoplasmic) -catenin expression and RFS was determined in
(i) the entire cohort, (ii) the first 5 years of tamoxifen treatment versus post-5 years.
Results: KMplotter analysis of -catenin in ER+, tamoxifen-treated patient samples (n=665) revealed a significant relationship with
improved RFS in the post-five year cohort [HR: 0.48 (0.34-0.68); p=0.000019) versus the first 5 years [HR: 0.91 (0.61-1.36;
p=0.64)]. No significant association was observed in untreated ER+ patients. In the Nottingham series, nuclear -catenin positivity
was significantly associated with improved survival in the 20-year follow-up for tamoxifen-treated patients (p=0.047) but not in the
first five years (p=0.239). Cytoplasmic -catenin did not associate with survival. Further analysis in the ABC series revealed an
association between nuclear -catenin positivity and improved survival for tamoxifen-treated patients in the entire cohort (HR:
0.52 (0.18, 0.55); p=0.00005). However, nuclear -catenin was more strongly associated with improved outcome in the post
5-year treatment group (HR: 0.30 (0.14, 0.66); p=0.01) versus the first five years of treatment (HR: 0.33 (0.15,0.73); p=0.06). No
significant associations were seen with cytoplasmic -catenin. The transATAC trial material again revealed an association
between presence of nuclear -catenin and reduced distant recurrence in the post 5-years endocrine-treated cohort (HR: 0.54
(0.33, 0.91): x=5.52, p=0.018) versus years 1-5 (HR: 1.16 (0.67, 2.01); x=0.29, p=0.59).
Conclusions: This is the first study to demonstrate that nuclear -catenin may represent a predictive biomarker for late relapse
following tamoxifen treatment in ER+ breast cancer where, contrary to traditional hypotheses and pre-clinical data, its nuclear
expression is strongly associated with good outcome post-five years of treatment.

Title: Exploratory analysis of single gene predictive biomarkers in TransHERA DASL cohort reveals that C8A mRNA expression
is prognostic of outcome and predictive of benefit of trastuzumab
Scooter Willis1, Varvara Polydoropoulou2, Zoi Tsourti2, Urania Dafni3, Brandon Young1, Bradley Long5, Pradip De1, Nandini Dey1,
Casey Williams1, Mitch Dowsett4 and Brian Leyland-Jones1. 1Avera Cancer Institute; 2Frontier Science Foundation-Hellas;
3
University of Athens; 4Royal Marsden Hospital and 5Scripps Florida.
Body: HERA is an international multi-center randomized trial comparing 1 or 2 years trastuzumab, given every 3 weeks, with
observation in women with HER2+ breast cancer after standard neoadjuvant or adjuvant chemotherapy. From December 2001 to
June 2005, 5102 patients were randomized. Comparing 1 year trastuzumab with observation at 8 years of follow-up, statistically
significant differences in disease-free survival (DFS) and overall survival, despite crossover to trastuzumab of observation arm
patients, were found. No benefit in DFS of 2 years compared to 1 year trastuzumab was found[Goldhirsch 2013].
To determine possible predictive single-gene biomarkers, an exploratory ITT analysis, with DFS as the primary endpoint, was
conducted using mRNA expression data from 610 TransHERA FFPE samplesprofiled on Illumina Whole-Genome DASL cubic
spline normalization was applied to the data. Outcome was obtained from the HERA database with 8 years median follow-up and
clinical cut-off date April 12, 2012. Characteristics were well balanced between treatment groups. The exploratory analysis of
20,464 genes using DFS as the endpoint identified C8A as a possible biomarker that is prognostic and predictive of response to
treatment. Cox regression was used to model DFS, with the interaction term between treatment and C8A as a continuous and a
categorical variable split on the cohort mean. The observation arm consists of 199 samples with 66 events and the trastuzumab
arm(1&2-year combined) of 411 samples with 108 events.
A statistically significant interaction between C8A mRNA and treatment was detected (p<0.001), indicating that C8A mRNA is
predictive of response to trastuzumab treatment. For the C8A low subgroup (mRNA expression lower than the cohort mean) no
significant treatment benefit is observed (p=0.73). On the other hand for the C8A high subgroup, patients in the trastuzumab arm
experience a lower hazard of a DFS event by almost 75% compared to patients from the observation arm (HR=0.25;
95%CI:0.15-0.43, p<0.001). A significant prognostic effect of C8A mRNA is also observed (p<0.001) in the observation arm,
where for the C8A high group the hazard of a DFS event is three times the respective hazard of the C8A low group (HR-=3.27;
95%CI:2.01-5.32, p<0.001).
C8A is a member of the membrane attack complex and is part of the innate immune system. C8A inserts into the membrane of
the target cell and binds with multiple copies of the pore-forming C9 leading to cell lysis. From the GeneAtlas, C8A is highly
expressed in liver tissue suggesting an advantage for tumors with high expression of C8A and innate immune response. The
Cancer Cell Line Encyclopedia indicates a wide range of C8A mRNA expression.
C8A as a single gene biomarker that is prognostic of DFS and predictive of benefit from trastuzumab has the potential to improve
the standard of care in HER2+ breast cancer. Understanding the advantage of over expression of C8A related to the innate
immune response can give insight into the mechanisms that drive cancer. We note with caution that this finding is the result of an
exploratory analysis and is being pursued in additional trastuzumab cohorts for further validation.

Title: Quantitative p95HER2 and HER2 correlations with outcome in the FinHer trial
Jeff Sperinde1, Weidong Huang1, Aki Vehtari2, Ahmed Chenna1, Pirkko-Liisa Kellokumpu-Lehtinen3, John Winslow1, Petri Bono4,
Yolanda Lie1, Jodi Weidler1 and Heikki Joensuu5. 1Monogram Biosciences, South San Francisco, CA; 2Aalto University, Aalto,
Finland; 3Tampere University Hospital, Tampere, Finland; 4Helsinki University Central Hospital, Helsinki, Finland and 5Helsinki
University Central Hospital & Helsinki University, Helsinki, Finland.
Body: Background: Expression of p95HER2 (p95), a truncated form of the HER2 receptor that lacks the trastuzumab binding
site but retains kinase activity, appears to be a prognostic biomarker for poor trastuzumab treatment outcome in HER2-positive
metastatic breast cancer. The impact of p95 expression on trastuzumab treatment efficacy in early HER2-positive breast cancer
is less clear. In the current study, p95 expression levels were measured in HER2-positive patients from the phase III FinHer
adjuvant trastuzumab trial and correlated with treatment outcome.
Methods: In the FinHer phase III trial, 232 HER2-positive early breast cancer patients were randomized to receive 9-weeks of
trastuzumab treatment versus no trastuzumab treatment (control). Quantitative p95 protein expression was measured in
formalin-fixed paraffin-embedded samples using the p95 VeraTag assay (Monogram Biosciences), specific for the M611 form of
p95. Quantitative HER2 protein expression was measured using the HERmark assay (Monogram Biosciences). Time to distant
recurrence (TDR) was used as the primary outcome measure.
Results: Sufficient tissue was available to measure p95 in 192 HER2-positive patients randomized to receive chemotherapy vs
chemotherapy plus trastuzumab. The chemotherapy only (n=97) and chemotherapy plus trastuzumab (n=95) arms were first
analyzed separately. In the chemotherapy only arm, increasing log(p95) correlated with shorter TDR (HR = 2.0; p = 0.02) when
stratified by hormone receptor status, nodal status and chemotherapy regimen. In the chemotherapy plus trastuzumab arm,
increasing log(p95) was not correlated with a shorter TDR (HR = 0.58; p = 0.19). Log(HER2) was not significantly correlated with
TDR in either arm. In a combined analysis of both treatment arms, log(p95) was significantly correlated with trastuzumab
treatment outcome in a multivariate model that included hormone receptor status, nodal status, chemotherapy regimen, log(p95)
and treatment arm. Subset analyses of hormone receptor positive and negative groups indicated that the interaction of p95
expression with trastuzumab treatment was largely driven by the hormone receptor negative subset.
Conclusions: In the FinHer phase III adjuvant breast cancer trial, HER2-positive patients with elevated breast tumor p95HER2
expression experienced poor outcomes when treated with chemotherapy alone, whereas patients with elevated p95 expression
experienced the most benefit when trastuzumab was added to chemotherapy. The different influence between hormone receptor
subsets of p95 expression on trastuzumab response resembles the effect of HER2 expression on trastuzumab response in the
NSABP B-31 trial (JNCI 105:1782, 2013).

Title: 4EVER: The impact of mTOR inhibition on bone health in postmenopausal women with hormone receptor positive (HR+)
advanced breast cancer treated with everolimus (EVE) in combination with exemestane (EXE)
Peyman Hadji15, Hans Tesch1, Oliver Stoetzer2, Thomas Decker3, Christian M Kurbacher4, Romy Neumeister5, Frederik Marm6,
Andreas Schneeweiss7, Christoph Mundhenke8, Andrea Distelrath9, Peter A Fasching10, Micheal P Lux10, Diana Lftner11,
Wolfgang Janni12, Mathias Muth13, Julia Kreuzeder13, Claudia Weiss13 and Eva-Maria Grischke14. 1Oncology Bethanien,
Frankfurt/Main, Germany; 2Haemato-Oncology, Munich, Germany; 3Oncology Ravensburg, Germany; 4Medical Centre Bonn
Friedensplatz, Bonn, Germany; 5University Hospital Magdeburg, Magdeburg, Germany; 6University Hospital Heidelberg,
Heidelberg, Germany; 7National Center for Tumor Diseases, Heidelberg, Germany; 8University Hospital Kiel, Kiel, Germany;
9
Medical Healthcare Centre East Hessen GmbH, Fulda, Germany; 10University Hospital Erlangen, Erlangen, Germany; 11Med.
Clinic for Hematology, Oncology and Tumor Immunology, Charit Campus Benjamin Franklin, Berlin, Germany; 12University
Hospital Ulm, Ulm, Germany; 13Novartis Pharma GmbH, Nuremberg, Germany; 14University Hospital Tuebingen, Tuebingen,
Germany and 15Hospital North West GmbH, Frankfurt/Main, Germany.
Body: Introduction:
The BOLERO-2 study showed significant doubling of PFS benefit with the addition of EVE to EXE in postmenopausal women with
hormone receptorpositive advanced breast cancer progressing after non-steroidal aromatase inhibitor (NSAI) therapy. The
BOLERO-2 bone sub study indicated an immediate positive influence of EVE on bone health after 6 and 12 weeks of treatment.
To further investigate the longer term effect of mTOR inhibition on bone health, we included this exploratory objective in the
phase IIIb, multi-center, open label study 4EVER for postmenopausal women with hormone receptor positive advanced breast
cancer treated with EVE +EXE.
Methods:
From May 2012 to November 2012 bone biomarker samples of 247 postmenopausal women with metastatic or locally advanced,
hormone receptor positive, HER2 negative breast cancer refractory to NSAI were collected within this phase IIIb study. Here, we
report the results of the planned exploratory analysis of biomarkers of bone turnover.
The objective of the exploratory biomarker analysis of bone resorption and formation was to assess the effect of EVE on changes
of biomarker levels of bone metabolism (CTX, P-I-NP, osteocalcin, vitamin D, testosterone, estradiol, DHEAS, SHBG, PTH,TSH
and FSH) from day 1 to weeks 4, 12 and 24 to confirm potential protective effects of EVE on bone via inhibition of bone osteoclast
resorption. Descriptive statistics were used to summarize the single bone resorption markers by visit and difference to baseline.
Results:
Trial data base lock will occur in late June 2014, therefore, the final biomarker analysis of bone resorption and formation and their
influence on ORR, PFS and safety will be presented at SABCS 2014.
The preliminary analysis on the changes of bone marker levels from baseline to week 24 included 247 patients. The measured
changes from baseline to week 24 in biomarker levels revealed increasing levels of testosterone, DHEAS and FSH, while SHBG
and PTH were significantly decreased from baseline to week 24. The combination of EVE to EXE resulted in stabilization of bone
health as documented by a decrease of absolute P-I-NP levels at week 24 compared to baseline as well as a stabilization of bone
resorption as measured by CTX.
Conclusion:
Our first preliminary results (data cut off 15 Nov 2013) on the changes in bone marker levels supported the hypothesis of EVE
having direct impact on bone health, suggesting a decrease of bone turnover and a reversal of the increase in bone resorption
associated with aromatase inhibitor therapy. The final analysis of the 4EVER sub study on bone marker levels in correlation with
the efficacy data will provide more detailed insights into possible benefits of the combination of EVE plus EXE.

Title: Evaluation of an in vitro derived signature of olaparib response (PARPi-7) as a predictive biomarker of response to
veliparib/carboplatin plus standard neoadjuvant therapy in high-risk breast cancer: Results from the I-SPY 2 TRIAL
Denise M Wolf1, Christina Yau1, Ashish Sanil2, Anneleen Daemen9, Laura Heiser8, Joe Gray8, Lamorna Brown-Swigart1, Susan
Flynn1, Gillian Hirst1, I-SPY 2 TRIAL Investigators3, Meredith Buxton1, Angela DeMichele4, Nola Hylton1, Fraser Symmans5, Doug
Yee6, Melissa Paoloni3, Laura Esserman1, Don Berry2, Hope Rugo1, Olufunmilayo Olopade7 and Laura van 't Veer1. 1University of
California, San Francisco, CA; 2Berry Consultants, LLC; 3QuantumLeap Healthcare; 4University of Pennsylvania, Philadelphia,
PA; 5University of Texas MD Anderson Cancer Center, Houston, TX; 6University of Minnesota, Minneapolis, MN; 7University of
Chicago, Chicago, IL; 8Oregon Health & Science University, Portland, OR and 9Genentech, San Francisco, CA.
Body: Background: We developed a 7-gene DNA-repair deficiency signature (PARPi-7) that predicts breast cancer cell line
sensitivity to the PARP inhibitor olaparib [PMID: 22875744]. We hypothesized that this signature would also predict response to
other PARP inhibitors including veliparib. In the I-SPY 2 TRIAL, HER2- patients were randomized to receive standard
chemotherapy or the oral PARP inhibitor veliparib in combination with carboplatin (V/C) and chemotherapy. V/C graduated in the
triple-negative (TN) signature. Here we assess the PARPi-7 as a specific biomarker of V/C response.
Methods: 115 HER2- patients (V/C: 71 and concurrent controls: 44) were considered in this analysis. The PARPi-7 signature
score is computed from Agilent 44K array data as published using expression levels of BRCA1, CHEK2, MAPKAPK2, MRE11A,
NBN, TDG, and XPA. We assess association between PARPi-7 and response in the V/C and control arms alone (Wald p < 0.05),
and relative performance between arms (biomarker x treatment interaction, likelihood ratio p < 0.05) using a logistic model. In an
exploratory analysis, we dichotomized patients by the PARPi-7 score using the published in vitro derived cutpoint (0.037). To
assess PARPi-7 in the context of the graduating signature, we added the PARPi-7 High patients to the graduating TN subset and
evaluated the treatment effect in this biomarker-positive group. Our study is exploratory with no claims for generalizability of the
data. Statistical calculations are descriptive (e.g. p-values are measures of distance with no inferential content). Our analyses do
not adjust for multiplicities of other biomarkers in the trial but outside this study.
Results: The PARPi-7 signature associates with patient response in the V/C arm (OR = 3.9, p=0.00056) but not in the control
arm (OR = 0.87, p=0.68). There is a significant biomarker x treatment interaction (OR in V/C arm relative to control arm = 4.48,
p=0.0028), which remains significant upon adjusting for HR status (p=0.0018). In an exploratory analysis, PARPi-7 dichotomized
using the published in vitro derived cutpoint yields 62 PARPi-7 Low and 53 PARPi-7 High patients. 26% of PARPi-7 High patients
are not TN. The distribution of pCR rates among PARPi-7 dichotomized groups are in Table 1.
V/C (n=71)
Control (n=44)
PARPi-7 Low (n=38)
PARPi-7 High (n=33)
PARPi-7 Low (n=24)
PARPi-7 High (n=20)
TN (n=59)
5 / 13
17 / 25
1/7
4 / 14
HR+HER2- (n=56)
2 / 25
3/8
4 / 17
0/6
When the PARPi-7 High patients are added to the graduating TN subset, the OR associated with V/C is 5.12, which is
comparable to that of the TN signature (OR: 4.29), while increasing the prevalence of biomarker-positive patients by 12%.
Evaluation of PARPi-7 in the context of the graduating signature under the I-SPY 2 Bayesian model is pending.
Conclusion: Our sample size is small. Our pre-specified analysis suggests the PARPi-7 signature shows promise for predicting
response to veliparib/carboplatin combination therapy relative to control. If verified in a larger trial, this cell-line derived signature
may contribute to the selection criteria of PARP inhibitor trials in the future.

Title: Prediction of benefit from endocrine therapy in ER+ early stage breast cancer: Correlative studies of the breast cancer
index HoxB13/IL17BR (H/I) ratio, ER, PR, and HER2 expression in the randomized Stockholm trial
Yi Zhang1, Brock E Schroeder1, Piiha-Lotta Jerevall Jannok2, Tommy Fornander3, Olle Stal4, Catherine A Schnabel1 and Dennis C
Sgroi2. 1bioTheranostics, Inc, San Diego, CA; 2Massachusetts General Hospital, Boston, MA; 3Karolinska University Hospital,
Stockholm, Sweden and 4Linkping University, Linkping, Sweden.
Body: Background: Hormone receptor expression is an indication for endocrine therapy, however, receptor status is insufficient
to account for the considerable heterogeneity of response, and additional predictive biomarkers are needed. The HoxB13/IL17BR
(H/I) gene expression ratio has previously been shown to be predictive of benefit from extended endocrine therapy for women
with estrogen receptor-positive (ER+) breast cancer treated in the MA.17 trial. In this correlative study, H/I was compared with
ER, PR, and HER2 expression in assessing response to adjuvant tamoxifen (TAM) versus untreated (UNT) in ER+,
node-negative patients from the prospective, randomized Stockholm trial.
Methods: Analysis included 600 ER+, LN- patients (317 TAM, 283 untreated) from the Stockholm cohort. Expression profiles
were generated by RT-PCR for H/I, ER, PR, and HER2, with additional IHC studies for ER, PR, and HER2. Multivariate Cox
models including age, tumor size and tumor grade, were used to assess the significance of the interaction between treatment and
each biomarker as continuous variables. 10-year risk of distant recurrence, as a function of each continuous biomarker, was
estimated from Cox model in each of the 2 treatment arms.
Results: The interaction between H/I and TAM treatment was significant (p = 0.003). Consistent with the significant interaction,
the 10-year rate of distant recurrence as a function of continuous H/I demonstrated that the reduction in recurrence rates with
TAM correlated with increasing H/I. Interaction P values for all other biomarkers were nonsignificant [ER (PCR), P=0.473; ER
(IHC), p=0.371; PR (PCR), p=0.555; PR (IHC), p=0.475; HER2 (PCR), p=0.947, and HER2 (IHC), p=0.839]. The treatment effect
of tamoxifen was unchanged across ER, PR, and HER2 expression levels.
Conclusions: Results of this study provide further support for HoxB13/IL17BR as a biomarker of endocrine response. The H/I
gene expression ratio, but not ER, PR, or HER2 expression evaluated as continuous variables, was predictive of benefit from
adjuvant TAM treatment. Findings indicate that H/I endocrine response activity is not strictly correlated with ER+ expression and
provides independent predictive information regarding estrogen signaling-driven recurrences.

Title: Alterations of intratumoral signalling in breast cancer patients receiving pre-operative trastuzumab alone or combined with
everolimus
Jean-Louis Merlin1, Maeva Lion1, Jennifer Wong2, Thomas Bachelot3, Fabrice Andre4, Isabelle Treilleux3, Delphine Loussouarn5,
Jacques Bonneterre6, Maria Rios1, Vronique Dieras2, Marta Jimenez7, Agns Leroux1 and Mario Campone5. 1Institut de
Cancrologie de Lorraine, Nancy, France; 2Institut Curie, Paris, France; 3Centre Lon Brard, Lyon, France; 4Institut Gustave
Roussy, Villejuif, France; 5Institut de Cancrologie de l'Ouest, Nantes, France; 6Centre Oscar Lambret, Lille, France and
7
UNICANCER R&D, Paris, France.
Body: Background: PI3K/AKT/mTOR and MAP kinase pathways are major signaling pathways involved in mammary
tumorigenesis and are investigated as putative targets for therapy. Multiple cross-talks exists between these two pathways,
allowing the regulation of one another by and inversely, depending on the cell conditions. It has been reported that trastuzumab
acted differently when used pre-operatively or in neo-adjuvant setting with a lower implication of signaling blockade and a higher
induction of ADCC when used alone in chemotherapy naive patients. Additionally, mTOR blockade has been experimentally
reported to activate MAPK pathway through a feed-back loop effect. The purpose of this study was to retrospectively investigate
the effect on MAPK signaling of adding everolimus to trastuzumab as preoperative therapy of HER-2 positive primary breast
cancer amenable to surgery (Unicancer RADHER Phase II trial).
Patients and methods: Formalin-fixed paraffin embedded and frozen tumor samples of primary breast cancer (n=80), were
obtained from 82 patients with infiltrating breast carcinoma randomized from July 2008 to April 2012 to receive ttrastuzumab
alone (T arm) (loading dose 4mg/kg, then 2mg/kg/week), or combined with everolimus (T+E arm) (10 mg/day) for a 6 week
pre-operative treatment. The median patient age at diagnosis (at the randomization) was 52.7 years. All patients had baseline
biopsies taken before initiation of the treatment, at cycle 4 as an option and at surgery. FFPE samples were used for
immunohistochemistry (pAKT, pS6K, eIF4E, LKB1), frozen samples were used for multiplex immunoanalysis of phosphorylated
PI3K/AKT/mTOR and MAPKinase signaling proteins analysis (p-AKT, p-GSK3, p-P70S6K, p-MEK1, p-ERK1/2, p-P90RSK).
Before being submitted to total protein extraction, all biopsies were controlled to ensure a tumor content >50%. 40 pairs
associating baseline + surgery tumor specimens or baseline + cycle 4 biopsies were eligible for protein extraction.
Results: No statistically significant relationship was observed between the expression level of any of the phosphoproteins in the
initial biopsies and neither the clinical nor the pathological response, overall. After treatment, as compared to the level of
expression measured in the initial biopsies, no significant variation of expression of either PI3 kinase or MAP kinase related
phosphoprotein was observed in T arm. In T+E arm, significant inhibition of PI3 kinase/mTOR pathway was only observed
downstream mTOR protein with decreased expression of p-P70S6 kinase and p-4EBP1 together with a significant activation of
MAPK pathway was detected with increased expression of p-MEK1, p-ERK1/2 was observed in T+E arm.
Conclusion: These results confirm that when used alone in chemotherapy naive patients, trastuzumab could not mainly act
through the blockade of signaling and therefore when combined with mTOR inhibitors could lead to the suppression of negative
feedback regulation of MAP kinase pathway.

Title: Plasma (p) biomarker results from the TANIA trial evaluating continued or reintroduced bevacizumab (BEV) after 1st-line
BEV for HER2-negative metastatic breast cancer (mBC)
Javier Cortes1, Eduard Vrdoljak2, Fabio Puglisi3, Norbert Marschner4, Joseph Gligorov5, Christoph Zielinski6, Cristian Villanueva7,
Gilles Romieu8, Istvn Lang9, Eva Ciruelos10, Corinne Veyret11, Andrea Fontana12, Mikkel Oestergaard13, Sabine de Ducla13, Ulrich
Freudensprung13 and Gunter von Minckwitz14,15. 1Vall d'Hebron University Hospital, Barcelona, Spain; 2University Hospital Split,
Split, Croatia; 3University Hospital of Udine, Udine, Italy; 4Outpatient Cancer Center, Freiburg, Germany; 5APHP Tenon,
IUC-UPMC, Paris, France; 6Comprehensive Cancer Center, Medical University Vienna and CECOG, Vienna, Austria; 7Centre
Hospitalier-Universitaire de Besanon, Besanon, France; 8Institut du Cancer Montpellier, Val d'Aurelle, France; 9National
Institute of Oncology, Budapest, Hungary; 10Hospital Universitario 12 de Octubre, Madrid, Spain; 11Centre Henri Becquerel,
Rouen, France; 12University Hospital of Pisa, Istituto Toscana Tumori, Pisa, Italy; 13F. Hoffmann-La Roche Ltd, Basel, Switzerland;
14
German Breast Group, Neu-Isenburg, Hessen, Germany and 15University Women's Hospital, Frankfurt, Germany.
Body: BACKGROUND: A potential predictive and prognostic effect of pretreatment pVEGF-A concentrations was suggested in
exploratory analyses of phase III trials in HER2-negative mBC (AVADO), gastric cancer (AVAGAST), and pancreatic cancer
(AViTA). This led to initiation of the MERiDiAN trial evaluating pVEGF-A prospectively in patients (pts) receiving 1st-line paclitaxel
BEV in HER2-negative mBC. Potential predictive value was also noted for pVEGFR-2 in AVADO, AViTA, and in early BC in
BEATRICE. We report prespecified pVEGF-A and pVEGFR-2 analyses in TANIA.
METHODS: TANIA is an open-label randomized phase III trial evaluating the addition of BEV (15 mg/kg q3w or 10 mg/kg q2w) to
the investigators choice of chemotherapy (CT) in pts with HER2-negative mBC who experience disease progression (PD)
on/after 1st-line BEV-containing therapy. Allocation to BEV or no BEV is continued with 3rd-line therapy after 2nd PD (ie no
crossover permitted). The primary endpoint is PFS from randomization to 2nd-line PD/death. Pts consenting to the optional
biomarker substudy provided 6 mL plasma samples in EDTA before drug administration at randomization, at wks 7 and 13, every
12 wks thereafter, and at 2nd PD. pVEGF-A and shedded pVEGFR-2 were measured using the IMPACT assay (v2.03). The
median concentration for each marker before 2nd-line treatment was prespecified as the cut-off between low ( median) and high
(> median) biomarker subgroups. 2nd-line PFS was analyzed in each subgroup.
RESULTS: The PFS benefit from BEV seen in the ITT population (N=494; stratified hazard ratio [HR] 0.75, 95% CI 0.610.93)
was observed consistently within subgroups of the biomarker population (N=312). However, there was no differential BEV effect
according to pVEGF-A or shedded pVEGFR-2 concentrations. The stratified HR for PFS was 0.69 (95% CI 0.461.04) in pts with
low pVEGF-A (1010.6 pg/mL [median]) and 0.80 (95% CI 0.541.18) in pts with high pVEGF-A (interaction p=0.47). As all pts
had received prior BEV, unlike earlier trials showing a potential predictive effect of pVEGF-A, we explored pVEGF-A
concentrations according to BEV-free interval. The BEV-free interval was 12 wks in 117/150 pts (78%) in the CT arm and
136/162 pts (84%) in the BEV+CT arm. Median pVEGF-A was much lower in the subgroup of 59 pts with a BEV-free interval >12
wks (45.1 pg/mL) than in pts with a BEV-free interval 12 wks (1135.3 pg/mL). Analyses of 2nd-line PFS according to shedded
pVEGFR-2 concentration showed stratified PFS HRs of 0.68 (95% CI 0.451.02) for low pVEGFR-2 (9.6 ng/mL [median]) and
0.90 (95% CI 0.601.34) for high pVEGFR-2 (interaction p=0.49).
CONCLUSIONS: The potential predictive effect of pVEGF-A in AVADO was not observed in TANIA, although high pVEGF-A
concentrations in pts recently treated with BEV complicate interpretation. pVEGF-A is being evaluated prospectively in BEV-nave
pts in the ongoing MERiDiAN trial. Shedded pVEGFR-2 showed no predictive effect in TANIA and the suggested trend was in the
opposite direction to the effect seen in AVADO and BEATRICE. Further analyses are planned to evaluate the impact of gene
expression, cell-free DNA, protein expression, and DNA mutations on treatment effect in TANIA.

Title: Plasma biomarker analysis in patients with HER2-negative locally recurrent or metastatic breast cancer (LR/MBC) treated
with first-line bevacizumab (A) and paclitaxel (T) without or with capecitabine (X)
Siu W Lam1, Nienke M Nota1, Steffen M de Groot2, Agnes Jager3, Monique MEM Bos4, Sabine C Linn5, Joan van den Bosch6,
Hans J Braun7, Ankie MT van der Velden8, Maartje Los9, Johanneke EA Portielje10, Judith R Kroep11, Aafke H Honkoop12, Carolien
H Smorenburg13, Bea Tanis14, Johanna MGH van Riel15, Jetske M Meerum Terwogt16, Marien O den Boer17, Joep Douma18, Frank
Jeurissen19, Johan Berends20, Harm van Tinteren5 and Epie Boven1. 1Medical Oncology, VU University Medical Center,
Amsterdam, Noord-Holland, Netherlands; 2Comprehensive Cancer Centre the Netherlands, Amsterdam, Noord-Holland,
Netherlands; 3Erasmus MC Daniel den Hoed Cancer Center, Rotterdam, Netherlands; 4Reinier de Graaf Hospital, Delft,
Netherlands; 5Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 6Albert
Schweitzer Hospital, Dordrecht, Netherlands; 7Vlietland Hospital, Schiedam, Netherlands; 8Tergooi Hospitals, Hilversum,
Netherlands; 9St Antonius Hospital, Nieuwegein, Netherlands; 10Haga Hospital, Hague, Netherlands; 11Leiden University Medical
Center, Leiden, Netherlands; 12Isala Clinics, Zwolle, Netherlands; 13Medical Center Alkmaar, Alkmaar, Netherlands; 14Groene Hart
Hospital, Gouda, Netherlands; 15St Elisabeth Hospital, Tilburg, Netherlands; 16Onze Lieve Vrouwe Gasthuis, Amsterdam,
Noord-Holland, Netherlands; 17Laurentius Hospital, Roermond, Netherlands; 18Rijnstate Hospital, Arnhem, Netherlands; 19Medical
Center Haaglanden, Hague, Netherlands and 20Gemini Hospital, Den Helder, Netherlands.
Body: Background
The phase II ATX trial aimed at evaluating safety and efficacy of first-line AT or ATX for HER2-negative LR/MBC (NTR1348;
BOOG 2006-06). Plasma samples were collected for investigation of circulating proteins involved in angiogenesis and their
possible association with therapy outcome. We here report the prognostic value of plasma VEGF-A, soluble (s)VEGFR-2, ANG2,
sTIE2, IL6, IL8 and CA9 at baseline (C1D1) and their changes after cycle 1 (C2D1).
Methods
312 patients were randomized 1:1 to AT (T 90 mg/m2 d1, 8, 15 & A 10 mg/kg d1, 15 q4w x 6 cycles A 15 mg/kg d1 q3w for
next cycles) or ATX (T 90 mg/m2 d1, 8, A 15 mg/kg d1 & X 825 mg/m2 bid d114 q3w x 8 cycles A & X at same dose q3w for
next cycles). Plasma proteins were measured by immunoassays (R&D Systems and MSD). The association of (continuous)
protein levels on C1D1 and their changes on C2D1 with response, PFS and OS were evaluated by Mann-Whitney U test and
univariate Cox regression analysis.
Results
The biomarker cohort (n=181) and trial cohort had similar baseline characteristics and clinical outcome. After a median follow-up
of 46 months, there were 178 (98%) PFS events and 152 (84%) deaths.
On C1D1, levels of ANG2 and sTIE2 were significantly higher in patients with hormone-receptor positive disease compared to
those with triple-negative disease (p=.025 and p=.001, respectively). A high level of VEGF-A was noted in patients with visceral
metastases (p=.028) and those with an increasing number of metastatic sites (p=.017). High levels of ANG2, IL6, IL8 and CA9
were significantly associated with poor PFS and OS (Table 1). Protein levels were not associated with response.
Table 1. Association of circulating proteins at baseline with PFS and OS
PFS HR (95%CI)
PFS p-value
OS HR (95%CI)
OS p-value
ANG2
2.3 (1.0-4.9)
.04
3.1 (1.4-7.0)
.007
IL6
1.6 (1.1-2.3)
.02
1.9 (1.3-2.8)
.001
IL8
1.9 (1.3-2.8)
.001
2.8 (1.8-4.3)
<.001
CA9
1.8 (1.2-2.7)
.006
1.9 (1.3-2.9)
.002
On C2D1, levels of all proteins, except for IL6, had significantly changed. Whereas VEGF-A, ANG2, sTIE2 and IL8 decreased
significantly, sVEGFR2 and CA9 showed a significant increase. The median relative change of both IL8 and sVEGFR2 was
significantly different between patients having CR/PR vs. SD/PD (for IL8: -19.4% vs. 22.3%, p=.001 and for sVEGFR2: 6.5 vs.
2.1%, p=.01). A large relative increase in CA9 level was associated with better PFS (HR = 0.36, 95% CI, 0.19 0.68, p=.002) and
OS (HR = 0.50, 95% CI, 0.27 0.94, p=.03). All patients had very low levels of free VEGF-A on C2D1 (median 8 pg/ml).
Conclusions
In patients with HER2-negative LR/MBC receiving first-line bevacizumab-containing chemotherapy, high baseline plasma levels
of ANG2, IL6, IL8 and CA9 indicate a high risk for poor PFS and OS. These proteins might be useful for stratification according to
prognosis. Moreover, relative decrease in IL8 and increase in sVEGFR2 on C2D1 are associated with response, whereas a large
relative increase in CA9 is associated with better PFS and OS. Changes in these proteins after one cycle might be early
indicators of efficacy of bevacizumab combined with chemotherapy.
Financial support from Roche Netherlands.

Title: Hand-foot-syndrome (HFS) is a strong predictor for OS and PFS in HER2-negative metastatic breast cancer (mBC) treated
with first-line capecitabine (CAP) + bevacizumab (BEV): Results of a subanalysis of the randomized phase III CECOG
TURANDOT trial
Zielinski C Christoph1, Lang Istvan2, Beslija Semir3, Kahan Zsuzsanna4, Moshe J Inbar5, Stemmer M Salomon6, Anghel Rodica7,
Vrbanec Damir8, Messinger Diethelm9 and Brodowicz Thomas10. 1Medical University of Vienna, Vienna, Austria; 2National Institute
of Oncology, Budapest, Hungary; 3Institute of Oncology, Sarajevo, Bosnia and Herzegowina; 4University of Szeged, Szeged,
Hungary; 5Tel Aviv Sourasky Medical Center, Tel Aviv, Israel; 6Rabin Medical Center, Petah Tikva, Israel; 7University of Medicine
and Pharmacy Buchurest, Buchurest, Romania; 8University Hospital Rebro, Zagreb, Croatia; 9IST GmbH, Mannheim, Germany
and 10Medical University of Vienna, Vienna, Austria.
Body: Background:
The randomized phase III TURANDOT trial performed by the Central European Cooperative Oncology Group (CECOG)
compared first-line BEV + paclitaxel (PAC) vs BEV + CAP in HER2-negative mBC in a prospective randomized trial [Lang, Lancet
Oncol 2013]. In this trial, HFS was the most common adverse drug reaction in the BEV + CAP arm. In the present subanalysis,
we investigated whether the occurrence of HFS constituted a predictor for improved efficacy of BEV + CAP in the study
population randomized to this treatment arm.
Methods:
In the TURANDOT trial, patients with HER2-negative mBC who had received no prior chemotherapy for mBC were randomized to
either BEV-PAC (BEV 10 mg/kg d1 & 15 + PAC 90 mg/m2 d1, 8, & 15 q4w) or BEV-CAP (BEV 15 mg/kg d1 + CAP 1000 mg/m2
bid d1-14 q3w). Only patients who were randomized to the BEV-CAP arm and received at least one dose of CAP were included in
the present analysis. Cox proportional hazard models with time-dependent covariate "HFS of any grade" and "HFS grade" were
used to analyze the association between HFS and OS or PFS. In addition, landmark (LM) analyses were performed analyzing the
impact of HFS occurrence within the first 3 treatment months on PFS and OS.
Results:
Baseline characteristics of patients with HFS (n=154) versus no HFS (n=123) were well balanced and showed no significant
differences. Table 1 shows that with the occurrence of HFS, the hazard for progression or death was reduced by > 40%, while the
risk reduction (RR) was >55% for OS. Moreover, with increasing HFS-grade an increasing RR was observed for OS. In patients
with and without HFS in the first 3 months, the median PFS after the 3 months LM were 10 and 6.2 months (logrank test
p=0.0026), respectively, and the corresponding OS rates at 2 years after the 3 months LM were 63% and 44% (log rank test,
p=0.0842), respectively. Updated OS results will be presented at the conference.
Cox Proportional Hazards Models for PFS and OS with HFS as Time-dependent Covariate
Univariate
Multivariate*
Endpoint
Time depent variable
Hazard ratio (95%Cl)
p-value
Hazard ratio (95%Cl)
p-value
PFS
HFS(Yes vs No)
0.577(0.431,0.772)
0.0002
0.558(0.409,0.760)
0.0002
HFS Grade 1
0.639(0.440,0.927)
0.0184
0.536(0.350,0.821)
0.0042
HFS Grade 2
0.481(0.321,0.720)
0.0004
0.487(0.320,0.741)
0.0008
HFS Grade 3
0.641(0.411,0.999)
0.0495
0.740(0.466,1.178)
0.2043
HFS(Yes vs No)
0.417 (0.273, 0.637)
<.0001
0.436 (0.285, 0.666)
0.0001
HFS Grade 1
0.568 (0.334, 0.968)
0.0375
0.548 (0.322, 0.934)
0.0270
HFS Grade 2
0.417 (0.227, 0.765)
0.0048
0.448 (0.243, 0.824)
0.0098
HFS Grade 3
0.220 (0.093, 0.520)
0.0006
0.247 (0.104, 0.586)
0.0015
OS
* Significant baseline factors in multivariate models: ECOG (0 vs 1-2) in all models, ER/PR status (Any positive vs Other) in both
OS models.
Conclusion:
HFS is a strong predictor for prolonged PFS and OS in mBC patients receiving BEV + CAP first-line treatment. Early occurrence
of HFS and HFS severity might be used for treatment modifications and patient motivation. Detection of biomarkers for HFS could
strengthen this approach.

Title: Homologous recombination deficiency (HRD) score predicts response to cisplatin neoadjuvant chemotherapy in patients
with triple negative breast cancer
Andrea L Richardson1, Daniel P Silver1, Zoltan Szallasi2,3, Nicolai J Birkbak2, Zhigang C Wang1, J Dirk Iglehart1, Erica L Mayer1,
Eric P Winer1, Nadine M Tung4, Paula D Ryan8, Steven J Isakoff5, William T Barry1, April Greene-Collozi1, Alexander Gutin6, Julia
Reid6, Chris Neff, Joshua Jones7, Kirsten Timms6, Anne-Renee Hartman7 and Judy E Garber1. 1Dana-Farber Cancer Institute,
Boston, MA; 2Technical University of Denmark, Lyngby, Denmark; 3Children's Hospital Boston, Boston, MA; 4Beth Israel
Deaconess Medical Center, Boston, MA; 5Massachusetts General Hospital, Boston, MA; 6Myriad Genetic Laboratories, Inc, Salt
Lake City, UT; 7Myriad Genetic Laboratories, Inc, Salt Lake City, UT and 8Fox Chase Cancer Center, Philadelphia, PA.
Body: Background A significant proportion of triple negative breast cancers (TNBC) carry homologous recombination (HR)
defects and are sensitive to therapeutics that target this pathway. Several clinical trials have demonstrated improvement in
pathologic response with the addition of platinum to standard of care regimens but at a cost of increased toxicities. Recently 3
DNA based metrics (LOH, Abkevich et al.; TAI, Birkbak et al; LST, Popova et al.) have been shown to be highly associated with
BRCA1 or BRCA2 (BRCA1/2) status or predictive of sensitivity to platinum chemotherapy. The HRD score was defined as the
sum of LOH, TAI, and LST metrics, and a threshold separating tumors with high (42) and low HRD scores was predefined on
independent datasets based on tumor BRCA1/2 mutation or methylation status. This study assesses the association of HR
deficiency, defined as HRD score 42 or BRCA1/2 mutant, with response to cisplatin neoadjuvant chemotherapy in patients with
TNBC.
Methods Archival tumor samples were obtained from 70 patients with TNBC from 2 clinical trials conducted at DFHCC under IRB
approved protocols. One trial enrolled 28 patients who received neoadjuvant cisplatin therapy (Silver et al.). The second trial
enrolled 51 patients who received cisplatin and bevacizumab therapy (Ryan, et al.). HRD score and tumor BRCA1/2 mutations
were determined. Response was categorized by the residual cancer burden (RCB) class (Symmans et al.) with pathologic
response (PR) defined as RCB0 or I and pathologic complete response (pCR) as RCB 0. Logistic regression was used to assess
HR deficiency as a predictor of response to neoadjuvant therapy. All analysis was conducted according to a pre-specified
statistical analysis plan.
Results 62 tumors provided adequate tissue and passed sequencing quality metrics. 31 (50%) tumors were HR deficient, 22
(35%) HR non-deficient, and 9 (15%) failed assays. The association of HR deficiency status with PR (RCB0/I) and pCR (RCB0)
was tested by univariate logistic regression in 50 samples with complete HR status and clinical data. HR status predicted both PR
and pCR at the 5% level (Table 1). In a multivariate model of PR, HR status retained statistical significance when combined with
clinical variables (p=0.0017, OR=12.08 (1.96, 74.4). In 51 subjects with known BRCA1/2 mutation status, BRCA1/2 mutation was
not associated with either PR (p=0.17) or pCR (p=0.14). When restricted to BRCA1/2 non-mutated tumors (n=38), HR deficiency
remained significantly associated with both PR (p=0.0039, OR=9.44 (1.69, 52.7)) and pCR (p=0.018, OR=14.79 (1.48, 2001)).
Table 1. Association of HR deficiency status with PR (RCB0/I) and pCR (RCB0).
Response
HR Deficient
Number (% response)
HR Non-deficient
Number (% response)
Odds Ratio
(95% CI)
PR = no (RCBII,III)
14
19
Reference: Non-deficient
PR = yes (RCB0,I)
15 (52%)
2 (9.5%)
10.18 (2.00, 51.89)
pCR = no (RCBI,II, III)
21
21
Reference: Non-deficient
pCR = yes (RCB0)
8 (28%)
0 (0%)
17.00 (1.91, 2249)
P-value
0.0011
0.0066
Conclusions This study demonstrates that the HRD score can be used as a tool to identify patients with breast tumors with
underlying HR deficiency, including in BRCA1/2 non-mutated tumors, that will benefit from platinum therapy.

Title: Predicting residual risk of recurrence after neoadjuvant chemotherapy- a retrospective analysis of EndoPredict in the
GeparTrio trial
Sibylle Loibl1, Jan C Brase2, Stephan Gade1, Jens Huober3, Kristin Krappmann2, Knut Engels4, Falko Fend5, Berit Maria Pfitzner6,
Joern Hilfrich7, Christoph Thomssen8, Hans Juergen Holzhausen9, Silvia Darb-Esfahani6, Christian Schem10, Keyur Mehta1, Ralf
Kronenwett2, Gunter von Minckwitz1 and Carsten Denkert6. 1German Breast Group, Neu-Isenburg, Hessen, Germany; 2Sividon
Diagnostics, Cologne; 3University Hospital, Ulm, Germany; 4Zentrum fr Pathologie, Zytologie und Molekularpathologie Neuss;
5
Institut of Pathology, University Tbingen; 6Institute of Pathology, Charit University of Berlin, Berlin, Germany; 7Eilenriede
Klinik Hannover; 8University Halle/Saale; 9University Halle /Saale and 10University Schleswig Holstein Campus Kiel.
Body: Background: Patients (pts) with breast cancer who do not achieve a pathological complete response (pCR) after
neoadjuvant chemotherapy (NACT) have a poor prognosis and might be candidates for post-neoadjuvant clinical trials
investigating novel agents. The CPS-EG score (Mittendorf et al. JCO 2011) a combination of clinical/pathological parameters is currently used as criterion for selecting patients with highest risk of recurrence after NACT. Here, we examined whether the
gene expression test EndoPredict (EP) performed on surgical specimen after NACT provides independent prognostic information
for predicting the likelihood of recurrence in breast cancer patients with ER-positive, HER2-negative (ER+/HER2-) disease.
Methods:The molecular EP score was determined by qRT-PCR in 76 available surgical specimen classified as ER+/HER2- from
breast cancer pts with residual disease after NACT participating in the neoadjuvant GeparTrio trial. The pre-specified
clinical/molecular hybrid score EPclin was determined using ypT and ypN after NACT as clinical variables. Patients were
classified as having low or high risk according to pre-defined cut-off values. CPS-EG score was calculated based on stage before
and after NACT and pretreatment grade and ER status. Primary endpoint was disease-free survival (DFS). Associations were
assessed with uni- and multivariate Cox proportional hazard models. The unbiased c-index was estimated for common
clinical/pathological parameters and EP/EPclin scores.
Results: Among the 76 ER+/HER2- breast cancer pts evaluated, EP classified 50% of all pts (n=38) as low risk. EP high-risk
patients had a significantly increased risk for relapse compared to the low-risk group (continuous EP HR 1.14 [95% CI 1.05-1.27]
p-value 0.02; log-rank p=0.015). Multivariable Cox regression and unbiased c-index estimates (using EP and CPS-EG score as
continuous variable) showed that the EP-score (HR 1.15 [95% CI 1.03-1.29 p=0.014] and CPS-EG (HR 1.51 [95% CI 1.07-2.13]
p=0.019) provide independent prognostic information. Using the composite EPclin score 13 pts (17%) were classified as being
EPclin low. The risk of relapse was significantly higher for EPclin high compared to low (continuous EPclin HR 1.78 [95% CI
1.29-2.46] p<0.001; log-rank p=0.047). Bivariate Cox regression analysis including the CPS-EG and EPclin score (both as
continuous variables) showed that only EPclin (HR 1.63 [95% CI 1.14-2.31] p=0.0068) but not the CPS-EG (HR 1.25 [95%CI
0.85-2.31] p=0.26) was significantly associated with worse DFS. Results were similar for overall survival.
Conclusions: Our study shows that EPclin performed on surgical specimen of ER+/HER2- pts after NACT is an independent
predictor of recurrence in pts not achieving a pCR after NACT. EP/EPclin low risk patients are probably sufficiently treated with
(neo-)adjuvant chemo-endocrine treatment alone, whereas EP(clin) high risk patients have an increased risk of recurrence,
despite receiving standard (neo-)adjuvant chemo-endocrine therapy. The identification of molecular luminal high-risk patients
could help to identify high risk patients as candidates for novel drug-based approaches in addition to endocrine therapy to
overcome resistance in post-neoadjuvant trials.

Title: Prospective cohort study using the breast cancer spheroid model as a predictor for response to neoadjuvant therapy The
SpheroNEO study
Kathrin ML Halfter1, Nina Ditsch1, Hans-Christian Kolberg2, Holger Fischer3, Tanja Hauzenberger4, Franz Edler von Koch5, Ingo
Bauerfeind6, Gunter von Minckwitz7, Ilona Funke8, Alexander Crispin9 and Barbara Mayer1. 1University Hospital of the LMU
Munich, Munich, Germany; 2University Hospital of the LMU Munich, Munich, Germany; 3MarienHospital Bottrop, Bottrop,
Germany; 4Klinikum St Marien Amberg, Amberg, Germany; 5Klinikum Dritter Orden, Munich, Germany; 6Klinikum Landshut,
Landshut, Germany; 7GBG Forschungs GmbH, Neu-Isenburg and University Women's Hospital Frankfurt, Frankfurt, Germany;
8
SpheroTec GmbH, Martinsried, Germany and 9Institute for Medical Informatics, Biometry and Epidemiology LMU, Munich,
Germany.
Body: (1) PURPOSE
The aim of the prospective trial was to determine if cell survival in a breast cancer spheroid model following cytostatic treatment in
vitro predicts treatment response in breast cancer patients receiving equivalent neoadjuvant therapy.
(2) PATIENTS AND METHODS
Three-dimensional spheroids were directly generated from fresh tumor biopsy samples of 78 patients eligible for neoadjuvant
therapy. Cell survival in vitro, as well baseline clinical and pathological characteristics were correlated with the outcome following
treatment of each patient to determine the factor(s) most highly associated with pathological complete response (pCR
i.e.ypT0/ypN0) at surgery.
(3) RESULTS
Cell survival after treatment in the breast cancer spheroid model proved to be a sensitive and specific predictor for pCR in
individual breast cancer patients. A mean cell survival of 21.8% was found in the breast cancer spheroid model for 22 patients
with pCR versus 63.8% in 56 patients without pCR (P = .001). A receiver operator characteristic analysis determined an area
under the ROC curve of 0.86 (95% CI: 0.77 to 0.96) for cell survival compared to classic factors i.e. negative hormone receptor
and positive Her2/neu status, and age 50 years at primary diagnosis (AUC = 0.80, 95% CI: 0.70 to 0.90). A cutoff of 35% cell
survival was proposed, which grouped patients according to likelihood for pCR. Out of the 32 patients with values below this
threshold, 21 patients (65.6%) and one patient (2.2%) with a cell survival greater than 35% achieved pCR respectively; (sensitivity
95.5% (95% CI: 0.86 to 1.00); specificity 80.4% (95% CI: 0.70 to 0.91)).The specificity was improved to 81% (95% CI: 58 to 95%)
if the patient was treated per-protocol.
A positive correlation was also found between cell survival in vitro and residual tumor size (P = .021), indicating the possibility of
the model to predict degree of response.
(4) CONCLUSION
The breast cancer spheroid model is a valuable predictor for treatment outcome in patients undergoing neoadjuvant
chemotherapy for primary breast cancer. Preclinical selection of the most efficient drugs is a prerequisite to improve pCR.

Title: Serum activin A and response to the aromatase inhibibitor (AI) letrozole versus tamoxifen in metastatic breast cancer
Meghan Jensen1, Ashley Kang1, Suhail M Ali1,2, Kim Leitzel1, Ashwani Garg1, Jaqueline Rogerio3, David Chen3, Raymond Hall3,
Scott Hofsess3, Hilary A Chaudri-Ross4, Nicholas Bade5, Walter P Carney6 and Allan Lipton1. 1Penn State Hershey Medical
Center, Hershey, PA; 2VA Medical Center, Lebanon, PA; 3Novartis Pharmaceuticals Corporation, East Hanover, NJ; 4Novartis
Pharma AG, Basel, Switzerland; 5R&D Systems Inc, Minneapolis, MN and 6Nuclea Biotechnologies/Oncogene Science,
Cambridge, MA.
Body: Introduction: The prognostic and predictive utility of pretreatment serum activin A (TGF-B superfamily ligand) was
correlated with the response of first-line metastatic breast cancer patients (MBC) in the phase 3 randomized trial of letrozole vs.
tamoxifen.
Methods: 555 ER+ first-line MBC patients had pretreatment serum available for activin A ELISA (R&D Systems). Clinical
outcomes were analyzed using Cox proportional hazards modeling. Actin A levels were analyzed using continuous and
categorical (median cutpoint) pretreatment serum activin A levels. A pretreatment serum activin A analysis was performed within
the normal (not elevated, <15 ng/ml) pretreatment serum HER2 patient subgroup since tumor HER2 status was not available in
this older clinical trial.
Results: Serum activin A had a median of 971 pg/ml and an interquartile range of 623 and 1751 pg/ml. In the total population
with available serum (n=555), patients with higher serum activin A (> median) had significantly reduced objective response rate
(ORR)(17.33% vs 34.6%, Odds Ratio (OR)=0.4; p<0.0001) and reduced clinical benefit rate (CBR)(33% vs 53%, OR= 0.45;
p<0.0001), as well as significantly shorter time to progression (TTP) [median 5.88 vs 10.98 mo, HR=1.69; p<0.0001], and
reduced overall survival (OS) (median 22.78 vs 48.59 mo, HR=2.43; p<0.0001) compared to those with lower serum activin A
(<median) (table). The results were similar in subgroup analyses within treatment arms (letrozole or tamoxifen), as well as within
the subgroup of patients with normal (not elevated, <15 ng/ml) pretreatment serum HER2 levels (table).
Outcome (Serum activin A by median cutpoint, high vs. low)
TTP
Cohort
Patients
HR
OS
p
HR
Total Population
555
1.69
<0.0001
2.43
<0.0001
Letrozole Arm
274
1.58
0.001
2.24
<0.0001
Tamoxifen Arm
261
1.92
<0.0001
2.6
<0.0001
Serum HER2 (not elevated)
395
1.55
<0.0001
2.45
<0.0001
In the total population with available serum (n=555), multivariate analysis for TTP revealed that high serum activin A was a
significant adverse prognostic factor (HR=1.46, p<0.001). Multivariate analysis for OS also revealed that high serum activin A was
an independent adverse prognostic factor (HR 1.78, p<0.0001).
Conclusions: Patients with high pretreatment serum activin A levels had a significantly reduced ORR, CBR, TTP and OS
compared to patients with low serum activin A. The results were similar within the letrozole or tamoxifen treatment arms, and
within the serum HER2 not-elevated subgroups. High pretreatment serum activin A level is associated with relative resistance to
hormone therapy in first-line metastatic breast cancer.

Title: Evaluation of HER family protein signaling network as a predictive biomarker for pCR for breast cancer patients treated with
neratinib in the I-SPY 2 trial
Julia D Wulfkuhle1, Christina Yau2, Denise M Wolf3, Rosa I Gallagher1, Ashish Sanil4, Lamorna Brown-Swigart3, Susan Flynn3,
Gillian Hirst3, I-SPY 2 TRIAL Investigators5, Meredith Buxton3, Angela DeMichele6, Nola Hylton3, William F Symmans7, Laura van't
Veer3, Douglas Yee8, Melissa Paoloni5, Laura Esserman3, Donald Berry4, Minetta C Liu9, John W Park3 and Emanuel F Petricoin
III1. 1George Mason University, Manassas, VA; 2Buck Institute; 3University of California, San Francisco, CA; 4Berry Consultants,
LLC; 5Quantam Leap Health; 6University of Pennsylvania, Philadelphia, PA; 7MD Anderson Cancer Center; 8University of
Minnesota, Minneapolis, MN and 9Mayo Clinic.
Body: Background: We hypothesize that response to the pan-ERBB inhibitor, neratinib (N), may be predicted by pre-treatment
HER2-EGFR signaling. In the I-SPY 2 TRIAL, N graduated in the HR-/HER2+ signature. All patients received at least standard
chemotherapy. For HER2+ patients, N was administered in place of trastuzumab. We evaluated 18 HER family signaling proteins
as biomarkers of N response using reverse phase protein microarray (RPMA) data from pre-treatment LCM purified tumor
epithelium.
Methods: 168 patients (N: 106, concurrent controls: 62) had RPMA and pCR data. 18 biomarkers relating to HER family signaling
were evaluated: AKT S473, AKT T308, EGFR, EGFR Y1068, EGFR Y1148, EGFR Y1173, EGFR Y992, ERBB2, ERBB2 Y1248,
ERBB3 total, ERBB3 Y1289, ERK1/2 T202/Y204, Heregulin, mTOR, mTOR S2448, PI3K p85 Y458/p55 Y199, PTEN S380, and
SHC Y317. We assessed association between biomarker and response in the N and control arms alone (likelihood ratio test), and
relative performance between arms (biomarker x treatment interaction) using a logistic model. Analysis was also performed
adjusting for HR/HER2 status. In an exploratory analysis, we selected the marker with the greatest interaction (phosphorylated
EGFR (Y1173)) to dichotomize patients optimally based on the data and assessed it in the context of the graduating signature by
adding the EGFR Y1173-High patients to the HR-/HER2+ subtype and evaluating the treatment effect in this biomarker-positive
group. Our study is exploratory with no claims for generalizability of the data and does not account for multiplicities. Statistical
calculations are descriptive (e.g. p-values are measures of distance with no inferential content).
Results: 7 HER pathway markers (EGFR Y1068, EGFR Y1173, EGFR Y992, ERBB2 total, ERBB2 Y1248, ERBB3 Y1289, SHC
Y317) are associated with response in the N but not the control arm. However, the difference in performance between arms did
not reach significance by permutation testing. Adjusting for HR/HER2 status, EGFR Y1173 shows a significant biomarker x
treatment interaction (p = 0.049). In an exploratory analysis, we dichotomized patients by their EGFR Y1173 levels and evaluated
the distribution of pCR rates (Table 1).
Neratinib (n=106)
HR-/HER2+ (n=28)
Control (n=62)
EGFR Y1173 Low

(n=31)
EGFR Y1173 High

(n=75)
EGFR Y1173 Low

(n=29)
EGFR Y1173 High

(n=33)
0/4
12 / 18
1/1
1/5
24 / 57
5 / 28
5 / 28
Not HR-/HER2+ (n=140) 3 / 27

Table 1
OR between EGFR Y1173 groups in the N relative to control arm is 10.1. When EGFR Y1173 High patients are added to the
graduating HR-/HER2+ subset, the OR associated with treatment is 3.2 and is comparable to that in the HR-/HER2+ signature
(OR: 2.1), while increasing the prevalence of biomarker-positive patients by 50%. Evaluation of EGFR Y1173 under the I-SPY 2
Bayesian model is pending.
Conclusion: Our sample size is too small to draw definitive conclusions. Our exploratory analysis reveals that HER family
phosphoproteins associate with response to N, but only phosphorylated EGFR Y1173 appears to add value to the graduating
signature. Given that this biomarker would expand the patient population that may benefit, it merits evaluation in other ongoing
trials of neratinib.

Title: Thrombocytopenia is associated with pathological complete response to neoadjuvant carboplatin/docetaxel chemotherapy
in BRCA wild type triple negative breast cancer
Priyanka Sharma1, Benjamin C Powers1, Bruce F Kimler1, Claire Ward1, Jennifer R Klemp1, Carol S Connor1, Marilee K
McGinness1, Joshua MV Mammen1, Jamie L Wagner1, Qamar J Khan1, Roy A Jensen1, Andrew K Godwin1 and Carol J Fabian1.
1
University of Kansas Medical Center, Westwood, KS.
Body: Introduction: Growing evidence demonstrates activity of neoadjuvant carboplatin in triple negative breast cancer (TNBC).
Underlying germline and somatic Homologous Recombination (HR) repair deficiency may predict response to DNA damaging
agents like platinum compounds in TNBC. Certain DNA repair machinery genes (Fanconi Anemia gene) are also involved in the
maintenance of hematopoetic stem cell (HSC) function and impaired repair of DNA double strand breaks can lead to HSC and
progenitor cell dysfunction. Thus, in presence of HR defects DNA damaging chemotherapy may lead to unique hematological
toxicity. It is also possible that in presence of HR defects breast cancer response and hematological toxicity with DNA damaging
agents will parallel each other.
Aim: To evaluate the impact of hematological toxicity on response to neoadjuvant Carboplatin/Docetaxel chemotherapy in
patients with sporadic and BRCA associated TNBC utilizing clinical and BRCA mutation data from a prospective TNBC registry.
Methods: 288 patients with Stage I (T>1cm) II and III TNBC were enrolled on a multisite prospective registry between 3/2011 to
4/2014, out of which 49 patients received neoadjuvant Carboplatin AUC 6 + Docetaxel 75mg/m2 every 21 days (4-6 cycles) and
have undergone breast surgery. Carboplatin was dosed using the modified Cockcroft-Gault formula. Hematologic toxicity was
graded using the CTCAE version 4.03. All patients received prophylactic pegfilgrastim on day 2. Pathological complete response
(pCR) was defined as absence of invasive disease in the breast and axilla. All patients underwent comprehensive
BRACAnalysis(Myriad).
Results: For the 49 eligible patients median age was 50 years, median weight was 172 lbs, 18% were African American, and 37%
had LN+ disease. 26% (13 /49) of patients carried deleterious BRCA mutation (9 BRCA1, 4 BRCA2). pCR of the cohort was 65%.
Overall 61%, 96%, and 12 % patients demonstrated > Grade1 thrombocytopenia (Tp), anemia, or neutropenia, respectively; 4%,
6%, and 6% patients demonstrated grade 3/4 thrombocytopenia, anemia, or neutropenia, respectively. There was no association
between pCR and anemia or neutropenia. Carboplatin dose reductions/delays/omission was more common in patients with Tp
compared to patients without Tp (33% vs. 5%; p=0.02). Patients with Tp were more likely to achieve a pCR compared to patients
without Tp (85% vs. 47%; p=0.013). 77% of BRCA carriers and 64% of BRCA wild type TNBC demonstrated Tp (NS). pCR rates
in BRCA wild type patients with and without Tp were 82% and 47%, respectively (p=0.041). pCR rates in BRCA mutation carriers
with and without Tp were 89% and 50%, respectively (p=0.20). On multivariable platelet count was independently associated with
pCR (p=0.001)Conclusions: Tp was associated with decreased dose delivery of carboplatin but an improved pCR in BRCA wild
type TNBC. Comprehensive assessment of HR defects beyond germline BRCA mutation status may be required to elucidate the
biological process that explains this observation. Tp may be a harbinger of underlying HR deficiency and further correlative
studies exploring this association of Tp with pathological response to carboplatin in TNBC are warranted.

Title: Multiplatform molecular profiling of BC reveals significant differences in actionable targets from matched female breast
carcinomas
Sherri Z Millis1, Joanne Xiu1, Zoran Gatalica1 and Joyce O'Shaughnessy2. 1Caris Life Sciences, Phoenix, AZ and 2Baylor
Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, TX.
Body: Introduction
Male breast cancer (MBC) is rare, comprising 1% of all breast cancers. While clinically characterized as being similar to
postmenopausal ER+ BC, MBC has been much less characterized molecularly than female BC.
Methods
60 male (ages 37-84) and 5000 female (ages 27-98) breast cancer samples were evaluated for common gene mutations (Sanger
or Illumina TruSeq), protein expression (immunohistochemistry), microarray, and/or amplification/rearrangement (CISH or FISH).
The samples were analyzed for patterns within the MBC cohort and similarities/differences compared to the female (FBC)
subtypes (TNBC, non-TNBC, HER2+, and ER+) evaluated at Caris Life Sciences.
Results
Within the MBC cohort, approximately 23% were negative for ER, PR, and HER2 (TNBC); of those 18% were also negative for
AR; 72% were ER+; 51% were both ER+ and PR+. The incidence of high ER and PR protein expression was greater (72% vs.
56%, 54% vs. 40%) but incidence of HER2 overexpression (IHC, 3+) and amplification (FISH, HER2/CEP17 ratio higher than 2)
was lower (8.8% vs. 11%, 5% vs. 14.9%) when compared to FBC overall. The rate of EGFR amplification (measured as 4
copies in 40% or more tumor cells by FISH) was not different from FBC (15%), while the percentage of MBC pts with AR protein
expression (70%) was most similar to ER, PR positive FBC patients. Other biomarkers: the rate of ERCC1 overexpression was
lower in MBC when compared to FBC (36% vs. 49), the rate of PTEN loss was lower (36% vs. 61%), and the rates of MGMT,
TLE3, and RRM1 overexpression were higher (73% vs. 64%, 70% vs. 53%, and 47% vs. 30%, respectively). In the 14 MBC
cases evaluated by NGS, no PTEN gene mutations were identified, although PIK3CA gene mutations were seen at a similar rate
(41%) as in the >50yo ER+ FBC (37%), and TP53 gene mutations (21%) were seen slightly less frequently than in the >50yo ER+
FBC (27%). Comparison of the TN MBC to the ER+ MBC cohort identified differences in the mTOR pathway (PTEN loss of 13%
vs. 28% and PIK3CA mutation rate of 13% vs. 50%, respectively), in P53 mutation rates (33% vs. 0%), and in AR protein
expression (33% vs. 82% overexpression), TLE3 (14% vs. 83% overexpression), and ERCC1 (100% vs. 77% low).
Conclusions
The gene mutation, amplification, and protein expression profiles in MBC patients, including HER2 protein
expression/amplification, AR and TLE3 protein expression and PIK3CA gene mutation, may inform standard and investigational
therapeutic options for this rare cancer population.


Title: Gene expression profiles accompanying phenotypic changes during non-malignant breast epithelial cells acini formation to
explain MRI phenotypes
Marcia V Fournier1, Alan Derr1, Alex Margulis1, Kevin Reid1, Sara Brumbaugh2, Richard Anderson3, W Fraser Symmans4, Laura
Esserman5 and Nola Hylton5. 1BIOARRAY Therapeutics Inc, Farmington, CT; 2Ceres Analytics, Redwood City, CA; 3Symmetric
Computing, Boston, MA; 4University of Texas MD Anderson Cancer Center, Houston, TX and 5UCSF Helen Diller Family
Comprehensive Cancer Center, San Francisco, CA.
Body: Magnetic resonance imaging (MRI) captures the three dimensional way in which tumors are organized in the breast,
defined as imaging phenotypes in the I-SPY 1 trial. We developed a gene set based on the way in which breast epithelial cells
aggregate and organize in three dimensional cultures. We investigated whether these organizational genes correspond to the
imaging phenotypes.
MRI phenotypes have been shown to correspond to the degree of response to neoadjuvant chemotherapy, and are used to
predict the ability to achieve breast conservation treatment (Mukhtar et al, Ann Surg Oncol, 20: 38233830, 2013). We
hypothesized that the molecular profile accompanying phenotypic changes occurring during the organization process of
non-malignant acini may explain the molecular basis of MRI tumor phenotypes.
We have developed prediction models for MRI phenotypes based on expression profiles identified during the organization
process of non-malignant breast epithelial cells in three-dimensional laminin-rich extracellular matrix (Fournier et al. Cancer Res,
66, 7095-102, 2006). We analyzed a subset of 324 organizational genes in 147 samples from stage II-III breast cancer in the
I-SPY 1 trial cohort with Agilent microarrays and MRI annotation (CALGB 150007/150012; ACRIN 6657). MRI phenotype of the
index lesion was assessed by the site radiologists using the following radiographic criteria: A) well defined unicentric mass; B)
well-defined multilobulated mass; C) area enhancement with nodularity; D) area enhancement without nodularity; E) Septal
spreading. The distribution of phenotypes in I-SPY 1 was: A: 16%, B: 33%, C: 30%, D: 15%, E: 7%. We developed predictors for
MRI phenotypes dichotomized as either "well-defined" (A and B) or "non-well-defined" (C, D and E). Using patent-pending
algorithms we selected a subset of the organizational genes with the greatest predictive power to identify MRI "well-defined"
phenotypes in the I-SPY 1 cohort. Then prediction models were developed using the 50 top ranking genes and logistic regression
methods. We followed Monte-Carlo cross validation method to make sure that the performance of a model is not a result of
over-fitting the model to the sample data using separate datasets to support the modeling. The samples were randomly
partitioned 10,000 times in a 85% training, and 15% test ratio to test models performance. The resulting performance of predictive
models in the test set had an average classification accuracy of greater than 80% (ROC statistics AUC>0.8) using gene
expression measurements from between 16 to 20 genes. For random lists of 16-20 genes, the accuracy was approximately 50%
(ROC AUC0.5). Amongst the MRI models were several genes that are known to regulate key cellular pathways such as cell
division, metabolism, and migration using MetaCore pathway analysis. Taken together, the results suggest that the MRI
phenotypes may be a manifestation of the organization genes that determine behavior in three dimensional culture. Future
research includes confirming these results using an independent dataset, defining the potential drivers of MRI phenotypes, and
determining if putative drivers provide a key contribution to tumor subtypes.

Title: A seven-gene signature can predict distant recurrence in patients with triple-negative breast cancers (TNBCs) who receive
adjuvant chemotherapy following curative surgery of the primary breast cancer
Yeon Hee Park1, Hae Hyun Jung1, In-Gu Do1, Eun Yoon Cho1, Insuk Sohn1, Sin-Ho Jung1, Won Ho Kil1, Seok Won Kim1, Jeong
Eon Lee1, Seok Jin Nam1, Jin Seok Ahn1 and Young-Hyuck Im1. 1Samsung Medical Center.
Body: BACKGROUND: Women with triple-negative breast cancers (TNBCs) represent a significant treatment challenge as they
have a relatively poor prognosis and no effective targeted therapy exists. Although TNBCs are often discussed as a single
disease entity of breast cancers, in fact they are very heterogeneous. The aim of this study was to investigate candidate genes
that might function as biomarkers to differentiate TNBCs among patients, who received adjuvant chemotherapy after curative
surgery, into those with high or low risk for distant recurrence.
METHODS: We tested whether the results of a NanoString expression assay that targeted 245 prospectively selected genes and
used mRNA extracted from paraffin wax-embedded tumor tissues would predict distant recurrence in patients with TNBC. The
levels of expression of seven genes were used in a prospectively defined algorithm to allocate each patient to a risk group (low or
high).
RESULTS: NanoString expression profiles were obtained for 203 tumor tissue blocks. Increased expressions of the five genes
(SMAD2, HRAS, KRT6A, TP63, and ETV6) and decreased expression of the two genes (NFKB1 and MDM4) were associated
favorable prognosis in this patients cohort and were validated with cross-validation. The proportions of patients categorized as
having low or high risk were 75% and 25%, respectively. The KaplanMeier estimates of the rates of distant recurrence at 10
years in the low- and high-risk groups according to gene expression signature were 62% (95% CI, 48.678.9%) and 85% (95%
confidence interval, CI, 79.290.7%), respectively. When adjusting for tumornodemetastasis (TNM) stage, the distant
recurrence-free survival (DRFS)s in the low-risk groups were significantly longer than that in the high-risk group (p<.001) in each
of TNM stages I plus II, and III. In a multivariate Cox regression model, the gene expression signature provided significant
predictive power jointly with the TNM staging system.
CONCLUSION: A seven-gene signature could be used as a prognostic model to predict DRFS in patients with TNBC who
received curative surgery followed by adjuvant chemotherapy.

Title: Next generation sequencing to find predictors for chemotherapy response in triple negative breast cancer
Esther H Lips1,2, Magali Michaut3, Marlous L Hoogstraat4,5, Lennart Mulder1, Nicolle Besselink4,5, Marco J Koudijs5,6, Emile E
Voest5,7, Rene Bernards3,5, Petra M Nederlof2, Jelle Wesseling1,2, Sjoerd Rodenhuis7 and Lodewyk FA Wessels3,5. 1The
Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands; 2The Netherlands Cancer Institute, Amsterdam,
Noord-Holland, Netherlands; 3The Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands; 4UMC Utrecht, Utrecht,
Netherlands; 5Center for Personalized Cancer Treatment, Netherlands; 6UMC Utrecht, Utrecht, Netherlands and 7The
Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands.
Body: Background: Neoadjuvant chemotherapy is the standard of care for locally advanced triple negative breast cancer (TNBC).
A complete remission after chemotherapy is associated with a good prognosis. However, patients with a poor response often
relapse and die of metastatic disease. Biomarkers are urgently needed to predict which patients will respond to standard
chemotherapy. In the current study we deep-sequenced responding and non-responding tumors to find response prediction
markers. In addition, the encountered mutations might provide clues for targeted treatment options.
Methods: Next generation sequencing (NGS) was performed for 2,000 genes involved in oncogenesis with an average coverage
of 150 reads. DNA from 56 pre-treatment TNBC-biopsies was sequenced, as was non-tumor DNA from each patient for
comparison. Biopsies were obtained from patients scheduled to receive neoadjuvant chemotherapy with
doxorubicin/cyclophosphamide. The median follow up was 2.5 years. The presence of a BRCA1 germline mutation or BRCA1
promoter methylation was also known. We assessed three variables. First, we compared samples with a pathological complete
remission (pCR) with those that did not achieve a pCR, to find mutations predicting chemotherapy response. Second, we
performed the same comparison with relapse free survival as outcome measure. Third, we compared tumors with BRCA
mutations or BRCA1 promoter methylation with tumors with a functional copy of the BRCA genes. The goal of this analysis was to
detect mutations associated with BRCAness.
Results: The mutations observed were diverse and few recurrent mutations were detected. Most mutations were in TP53, TTN,
and PIK3CA (57%, 22%, 9%). The mutation rates were similar between responders and non-responders (average mutation rate
of 13 versus 16 mutations per tumor, p=0.27). The analysis of individual genes did not reveal significant differences between
responders and non-responders. NOTCH4 mutations showed an association with relapse free survival, with three out of nine
relapsing patients bearing a mutation and no mutations occurring in the relapse-free patients (p=0.008). Interestingly, PIK3CA
mutations were only observed in patients with a functional BRCA gene (5/26 (19%) versus 0/30, p=0.017). Tumors with BRCA
impairment may develop via different routes than tumors with functioning BRCA genes, and the PI3K pathway may play a role in
the latter. This finding has also been observed in an independent dataset (Severson et al, 2014, submitted).
Conclusions: Although few recurrent mutations were found, two interesting leads for follow up studies were identified. First, three
out of nine tumors with a relapse had a NOTCH4 mutation while NOTCH4 mutations were not occurring in the relapse free group.
Second, PIK3CA mutations were associated with BRCA proficient tumors. After validation in larger series, triple negative tumors
with PI3K or NOTCH4 mutations can be candidates for agents targeting these oncogenic pathways. In this manner, a more
individualized treatment of triple negative breast cancer might become possible in the near future.

Title: Identification of ERBB2 gene variants in HER2 positive disease associated with trastuzumab response in an adjuvant
trastuzumab chemotherapy trial
Shelly Gunn1, Alexander Zien2, Markus Hartenfeller2, Francesca Diella2, Stephan Brock2, Martin Stein2, Sasha Badbanchi2, Anja
Doerks2, David Jackson2, Lina Asmar3, Yunfei Wang3 and Steve Jones1. 1MolecularHealth, Woodlands, TX; 2MolecularHealth,
Heidelberg, Germany and 3McKesson Specialty Health, Woodlands, TX.
Body: Background
Among 493 patients with early stage, lower risk (70% node negative), HER2 positive breast cancer, the 2-year DFS was 97% with
adjuvant docetaxel and cyclophosphamide plus 1 year of trastuzumab (TCH) in a phase 2 study (Jones et al, Lancet Oncology
14: 1121, 2013). The objective of this work was to investigate the presence of ERBB2 specific DNA biomarkers in HER2 amplified
tumors associated with relapse compared to those which did not recur during 3 years of followup.
Methods
The 26 coding exons of the ERBB2 gene were analyzed by next generation sequencing (NGS) on the HiSeq-2500 (Illumina)
platform using DNA samples from the primary tumors of 13 of the 493 patients who progressed. Treatment refractory cases were
analyzed in parallel with a clinically and pathologically matched group of 11 patients from the same trial with 2-year relapse free
survival (RFS).
Results
ERBB2 gene variants were detected in all 11 relapse-free patients and 10/13 patients with relapse. Heterogeneity for sub-clonal
ERBB2 variants at <1% tumor variant frequency (TVF) was pronounced. There were 126 total distinct ERBB2 SNVs across the
24 tumors. Only 4 occurrences of SNVs known to us to be activating (G309A, D769H, D769Y, V777L, V842I, R896C, S310Y,
S310F, T798M, 611M, 678M) were found: 3 times V842I and a single case with R896C. Of the investigated tumors, 5 exhibited
sub-clonal hyper-mutability with >30 ERBB2 SNVs. These sub-clonal ERBB2 SNVs were not detected in a comparison set of
non-HER2 positive solid tumors. We did find variants that seem to be associated with later relapse, of which one is statistically
significant (I655V; p=1.2%, Fisher exact test, no compensation for multiple testing), and further 5 have p<10%.
Conclusions
We find no evidence for known activating HER2 mutations to confer increased (nor decreased) risk of relapse after TCH therapy
(p=30.0%). Remarkably, I655V is found significantly less often in the relapse group. This variant is known to increase BC risk by
activating HER2 dimerization; an effect that may be muted by trastuzumab, in contrast to some other causes of BC. Moreover,
I655V mutations with high TVF (>75%, hence suggesting homozygous germline presence) are exclusively found in the no-relapse
group (p=3.1%). These biomarkers in combination with other molecular studies including immune function gene status may help
define the subset of patients at risk for relapse during TCH therapy. The hyper-mutability genotype does not seem to correlate
with later relapse, so it may be hypothized that the low TVF mutations are passengers rather than drivers. Further studies are
needed to verify and precisely define the role of these ERBB2 biomarkers in HER2 positive breast cancer.

Title: Immunoglobulin G fragment C receptor polymorphisms and clinical efficacy of preoperative chemotherapy plus trastuzumab
and lapatinib in patients with HER2-positive operable breast cancer
Antonino Musolino1, Valentina Guarneri2, Nadia Naldi1, Beatrice Bortesi1, Daniela Boggiani1, Paolo Sgargi1, Daniele G Generali3,
Federico Piacentini4, Maria V Dieci2, Massimo Ambroggi5, Katia Cagossi6, Lorenzo Gianni7, Samanta Sarti8, Giancarlo Bisagni9,
Antonio Frassoldati10, Pierfranco Conte2 and Andrea Ardizzoni1. 1University Hospital of Parma, Parma, Italy; 2Istituto Oncologico
Veneto-IRCCS, University of Padova, Padova, Italy; 3AO Istituti Ospitalieri di Cremona, Italy; 4Modena University Hospital, Italy;
5
Hospital of Piacenza, Italy; 6Ramazzini Hospital, Italy; 7Ospedale Infermi, Rimini, Italy; 8Istituto Scientifico Romagnolo per lo
Studio e la Cura dei Tumori, Italy; 9Arcispedale Santa Maria Nuova-IRCCS, Italy and 10Ferrara University Hospital, Italy.
Body: Introduction: In vitro studies have shown that lapatinib enhances the immune-mediated cytotoxicity (ADCC) of
trastuzumab. FcR polymorphisms have been associated with both ADCC and clinical activity of trastuzumab in patients with
HER2-positive metastatic breast cancer. There are no data on the relationship between these polymorphisms and the
combination of trastuzumab plus lapatinib in the early stage setting. We performed a pharmacogenomics analysis of CHER-LOB,
a randomized phase II trial of preoperative chemotherapy plus trastuzumab (arm A), lapatinib (arm B), or both (arm C) in
HER2-positive operable breast cancer.
Methods: FcRIIa-H131R and FcRIIIa-V158F polymorphisms were analyzed on DNA from peripheral blood samples. Pathologic
complete response (pCR) of genotyped cases was evaluated by FcR polymorphism and treatment arm.
Results: Genotyping was successfully performed in 73/121 (60%) patients. No deviation from the Hardy-Weinberg equilibrium
was observed. Similarly to the overall results of the CHER-LOB study, in the subset of patients genotyped in this analysis, a
significant improvement in pCR rate was observed in favor of the combination of lapatinib plus trastuzumab (arm C) compared to
arm A (OR=3.66, P=0.037), and B (OR=3.03, P=0.049). Such improvement was restricted to carriers of FcRIIIa V allele (C vs. A,
OR=5.33, P=0.043; C vs. B, OR=6.50, P=0.012), while it was not observed in patients with FcRIIIa F/F genotype (C vs. A,
OR=2.14, P=0.642; C vs. B, OR=0.71, P=0.737). Disease free survival (DFS) was not different by treatment arm in all genotyped
cases, but a trend toward significance for an interaction between FcRIIIa V allele and better DFS with the combination of
lapatinib plus trastuzumab was detected (P=0.058). No significant associations were observed by FcRIIa polymorphism.
Conclusions: Host-related immune signatures may mediate lapatinib enhanced trastuzumab-dependent ADCC. FcRIIIa
genotypes may help predict different outcomes to lapatinib plus trastuzumab in HER2-Positive Early Breast Cancer.

Title: Brief exposure to trastuzumab prior to preoperative chemotherapy confirms predictors of response to treatment
Aditi Shirish Vadodkar1, Vinay Varadan1, Kristy Miskimen1, VV Bossuyt2, MM Abu-khalaf2, I Krop3, E Winer3 and LN Harris1. 1Case
Western Reserve University, Cleveland, OH; 2Yale Cancer Center, New Haven, CT and 3Harvard Medical Center.
Body: Background: HER2-positive (HER2+) breast cancer is biologically heterogeneous; no consistent biomarkers of response to
trastuzumab (T) have been identified. Recent data suggest intrinsic subtypes applied to HER2 cohorts are able to predict
response to T-based therapy. We designed a multicenter trial (DFCI 03-311) to determine if exposure to T-alone prior to
combination T-based chemotherapy (C) could produce predictors of pathologic complete response (pCR). We previously reported
that change in AKT and IGF signatures by gene expression could predict pCR. New gene expression pathways (immune and
HER2 subtypes) were evaluated.
Design : Fresh tumor core biopsies were taken at baseline and 2 weeks after a single dose of T (8mg/m2) from 80
HER2-overexpressing, stage II/III breast cancer patients enrolled on a clinical trial of T>T+C. A total of 122 samples (50 matched
pairs and 22 single timepoints) produced useable gene expression data. Nucleic acids were extracted using Qiagen AllPrep and
were analyzed with Illumina HT12v3. Data generated was normalized and median subtracted using Illumina's Genome Studio
software; intrinsic subtyping was performed using median-subtracted PAM50 genes; Estimation of STromal and Immune cells in
Malignant Tumors using Expression data (ESTIMATE) algorithm was used to infer the fraction of stromal and immune cells in
tumor samples.
Results: ESTIMATE Immune score predicts pCR after one dose of treatment of T compared with patients with stable disease or
progression as best response ,(NOR;p=0.01) PAM50 subtyping of baseline samples showed increase in pCR in the
HER2-enriched cluster (7/17; 41%) versus the Luminal B (2/14; 14%) and HER2 basal clusters (1/8; 12.5% ; P=0.04). ). Change
in subtype between baseline and 2 weeks appeared less common in HER2-Luminal B (4/13; 31%) compared with HER2 enriched
(9/16; 56%) or HER2 Basal (5/9; 55%). Of note, HER2-Basal had the highest frequency of NOR (4/13; 31%; P=0.06), despite
having greater subtype change than HER2-luminal tumors.
Conclusions:
Biomarkers of response to T-based regimens are greatly needed as no consistent predictors beyond HER2 are available. This
data confirms (Carey et al, ASCO 2014) that the HER2 enriched subtype predicts pCR and (Harris et al, CCR 2007) that HER2
basal tumors are more likely to be non-responders. In this study, one dose of T is necessary to predict pCR using ESTIMATE
immune signature. Validation of these findings a large cohort is needed.

Title: MammaPrint High1/High2 risk class as a biomarker of response to veliparib/carboplatin plus standard neoadjuvant therapy
for breast cancer in the I-SPY 2 TRIAL
Denise M Wolf1, Christina Yau1, Ashish Sanil2, Jo Chien1, Anne Wallace3, Angela DeMichele4, Hank Kaplan5, Doug Yee6, Claudine
Isaacs7, Kathy Albain8, Rebecca Viscuzi9, Judy Boughey10, Stacey Moulder11, Steven Chui12, Qamar Khan13, Toncred Styblo14,
Kirsten Edmiston15, Donald Northfelt16, Anthony Elias17, Barbara Haley18, Debu Tripathy19, Lamorna Brown-Swigart1, Susan Flynn1,
Gillian Hirst1, Meredith Buxton1, Nola Hylton1, Melissa Paoloni20, Fraser Symmans11, Laura Esserman1, Don Berry2, Hope Rugo1,
Olufunmilayo I. Olopade21 and Laura van 't Veer1. 1University of California, San Francisco, CA; 2Berry Consultants, LLC;
3
University of California, San Diego, La Jolla, CA; 4University of Pennsylvania, Philadelphia, PA; 5Swedish Cancer Institute;
6
University of Minnesota, Minneapolis, MN; 7Georgetown University; 8Loyola University Medical Center; 9University of Arizona,
Tucson, AZ; 10Mayo Clinic; 11University of Texas MD Anderson Cancer Center, Houston, TX; 12Oregon Health & Science
University, Portland, OR; 13University of Kansas Medical Center; 14Emory University, Atlanta, GA; 15Inova Health System; 16Mayo
Clinic, Scottsdale, AZ; 17University of Colorado; 18University of Texas, Southwestern, TX; 19University of Southern California, Los
Angeles, CA; 20QuantumLeap Healthcare and 21University of Chicago, Chicago, IL.
Body: Background: Further stratification of the 70-gene MammaPrintTM signature into high and ultra-high risk groups may help
predict chemo-sensitivity. In I-SPY 2, patients were classified as MammaPrint High1 (MP1) or MammaPrint (ultra) High2 (MP2),
with MP2 defined as MP_score <-0.154. MP1/MP2 classification was added to HR and Her2 to define the cancer subtypes used
in the I-SPY 2 adaptive randomization engine. HER2- patients were randomized to receive standard chemotherapy or the oral
PARP inhibitor veliparib in combination with carboplatin (V/C) and chemotherapy. V/C graduated in the triple-negative (TN)
signature, where MP2 was not an eligible signature for graduation. Here, we assess the performance of MP1/MP2 class as a
specific biomarker of response to V/C.
Methods:115 HER2- patients (V/C: 71 and concurrent controls: 44) were considered in this analysis. We assess association
between MP1/MP2 and response in the V/C and control arms alone using Fishers exact test, and relative performance between
arms (biomarker x treatment interaction, likelihood ratio p < 0.05) using a logistic model. This analysis is also performed adjusting
for HR status as a covariate. To assess MP1/MP2 in the context of the graduating signature, we added the MP2 patients to the
graduating TN subset and evaluated the treatment effect in this biomarker-positive group. Our study is exploratory with no claims
for generalizability of the data. Statistical calculations are descriptive (e.g. p-values are measures of distance with no inferential
content). This analysis does not adjust for multiplicities of other biomarkers in the trial but outside this study.
Results: In the V/C arm vs. concurrent controls, there were 66 MP1 (V/C: 32, Control: 34) and 49 MP2 patients (V/C: 39, Control:
10), 78% of which are TN. The distribution of pCR rates among MP1/MP2 dichotomized groups are summarized in Table 1.
V/C (n=71)
Control (n=44)
MP1 (n=32)
MP2 (n=39)
MP1 (n=34)
MP2 (n=10)
TN (n=59)
3/8
19 / 30
3 / 13
2/8
HR+HER2- (n=56)
1 / 24
4/9
4 / 21
0/2
The OR between MP1/MP2 risk groups for predicting pCR is 9.71 in the V/C arm (p=6.63E-05), in comparison to an OR of 0.97 in
the control arm (p=1). There is a significant biomarker x treatment interaction (p=0.023), which remains upon adjusting for HR
status (p= 0.028). Based on the I-SPY 2 Bayesian model, a Phase III trial with 300 MP2 patients has a 95% predictive probability
of success. When the MP2 patients are added to the graduating TN subset, the OR associated with V/C is 4.36, which is
comparable to that of the TN signature (OR: 4.29), while increasing the prevalence of biomarker-positive patients by 10%.
Conclusion: In our exploratory analysis, MP2 suggests higher sensitivity to V/C combination therapy relative to controls. This
observation has prompted an investigation into the biological mechanisms distinguishing the MP1/MP2 subtype that may account
for this specificity.

Title: Measurement of domain-specific HER2 (ERBB2) expression classifies benefit from Trastuzumab in breast cancer
Daniel E Carvajal-Hausdorf1, Kurt A Schalper1, Lajos Pusztai2, Amanda Psyrri3, Konstantine T Kalogeras4, Vassiliki Kotoula5,
George Fountzilas4 and David L Rimm1. 1Yale University, New Haven, CT; 2Yale University, New Haven, CT; 3Attikon University
Hospital, Athens, Greece; 4Hellenic Cooperative Oncology Group (HeCOG), Athens, Greece and 5Aristotle University of
Thessaloniki School of Medicine, Thessaloniki, Greece.
Body: Background: The ASCO/CAP guidelines consider chromogen-based immunohistochemistry (IHC) as the primary assay to
determine HER2 status in breast cancer. Studies have shown that antibodies targeting different protein domains (intracellular
[ICD] or extracellular domain [ECD]) are equivalent using traditional methods. Here we define a new method for standardization of
domain specific measurements, then determined their effect on trastuzumab benefit in the adjuvant setting.
Methods: We measured HER2 protein using quantitative immunofluorescence (QIF) in a standardization tissue microarray (TMA)
with CLIA-lab defined HER2 status with 2 antibodies targeting the ICD (CB11 and A0485) and 2 against the ECD (SP3 and
D8F12). Cut-points were generated using Joinpoint software. HER2 IHC and FISH results were used as reference to determine
sensitivity and specificity. Finally, domain-specific HER2 levels were measured in 180 samples from a clinical trial of adjuvant
chemotherapy followed by trastuzumab (HeCOG 10/05).
Results: HER2 ICD showed slightly higher sensitivity to predict HER2 gene amplification than the ECD, while the ECD was more
specific and had higher positive predictive value. Fifteen percent of trastuzumab-treated patients from HeCOG 10/05 showed
discordant results using ICD and ECD antibodies. ECD-high status was significantly associated with longer disease-free survival
(DFS) (log-rank P=0.049, HR=0.31, 95% CI: 0.144-0.997), while ICD status was not. In patients with low ECD, there was no
difference in DFS between ICD-low and ICD-high. However, when ICD was positive, high ECD was significantly associated with
longer DFS (log-rank P=0.027, HR=0.23, 95% CI: 0.037-0.82) compared to low ECD. Since this trial was not randomized for
trastuzumab, interaction could not be tested, but neither ICD nor ECD showed prognostic value in the 462 patients that were
traditionally classified as HER2-negative and did not receive trastuzumab.
Conclusion: Determination of HER2 status in breast cancer tumor tissue samples using a standardized system and antibodies
against the ECD or ICD suggests a biological difference in the tumors. High ECD was associated with benefit from trastuzumab,
while elevated ICD alone was not. This observation could be useful in developing a new HER2 assay that can subclassify
traditionally HER2-positive patients into groups for subsequent antibody (ECD) vs tyrosine kinase inhibitor (ICD) therapies.

Title: BRCAness is important to identify TNBC subtype resistant to taxanes
Takashi Ishikawa1,2, Kazutaka Narui2, Kazuhiro Shimada2, Kumiko Kida2, Mari S Oba2, Mikiko Tanabe2, Yasushi Ichikawa3,
Sadatoshi Sugae3 and Itaru Endo3. 1Tokyo Medical Univeristy, Tokyo, Japan; 2Yokohama City Medical Center, Yokohama, Japan
and 3Yokohama City University, School of Medicine, Yokohama, Kanagawa, Japan.
Body: BACKGROUND: Triple negative breast cancer (TNBC) is heterogeneous and consists of tumors associated with basal
like, BRCA related and cancer stem cell (CSC) phenotypes. Although anti-cancer agents are substantial for treating TNBC,
existing ones do not work in some subpopulation in TNBC at all. However, it has not been reported that subdivision of TNBC is
useful for choosing ant-cancer agents.
AIM: To examine whether subdividing TNBC is beneficial for tailored chemotherapy and to identify predictive factors for existing
anti-cancer agents in TNBC.
METHODS: Sixty-six TNBC cases from a randomized phase II trial comparing TCx6 (TC6) with FEC followed by docetaxel
(FEC-D) as neoadjuvant chemotherapy for hormone receptor-negative breast cancer (Kanagawa Breast Oncology Group 1101
Study). TNBC was subdivided by 1) IHC of CK 5/6 and EGFR into basal- and non-basal subtypes, and 2) MLPA of BRCA1 into
BRCA1 and non-BRCA1 subtypes. The pCR rates were examined according to each regimen and subtype. 3) The association of
grade 3 pCR was examined with Ki-67, p53, aldehyde dehydrogenase (ALDH) 1 and topoisomerase 2A (topoII) by IHC and
TOP2A by FISH for each regimen.
RESULTS: 1) In basal subtype, the pCR rate was significantly higher for FEC-D (42.9%) compared with TC6 (13.6%) (p=0.033),
but it was equivalent in non-basal subtype (FEC-D vs TC: 25.0% vs 36.4%, p=0.554). 2) In BRCA1 subtype, it was more
significant (FEC-D vs TC: 53.8 % vs 13.3%, p=0.022). 3) An association between pCR and low ALDH1 expression was found in
both FEC-D and TC6 (OR: 3.75 and 2.73). High topo II protein expression was associated with pCR in FEC-D (OR: 3.5).
DISCUSSION: TC6 was less effective than FEC-D in basal subtype and BRCA1 subtype, showing that taxanes cannot exert their
anticancer role in tumors with BRCA1 dysfunction. Although basal subtype may contain more BRCA1-defective tumors than
non-basal subtype, MLPA of BRCA1 was better to identify subtype resistant to taxanes than CK5/6 and EGFR. ALDH1 predicted
treatment efficacy, and could therefore represent a marker of resistance to conventional chemotherapy.

Title: Alterations in HER2 status and outcome following neoadjuvant chemotherapy in HER2-positive breast cancer (BC)
Dianna Ng1, Gregor Krings1, Christina Yau2, Kristie White1, Jie Hou1, Anthony Chua1, James P Grenert1 and Yunn-Yi Chen1.
1
UCSF Medical Center, San Francisco, CA and 2Buck Institute for Research on Aging, Novato, CA.
Body: Background: Emerging data suggest that chemotherapy and anti-HER2 agents can change HER2 status in HER2+ BC
and that this may have prognostic implications. The findings suggest a role for post-treatment (tx) retesting of HER2 status in
surgical specimens. However, few studies have specifically examined this phenomenon, and direct comparisons of post-tx HER2
testing modalities are lacking. Furthermore, the biologic basis for HER2 alterations remains poorly understood. The aims of this
study were to systematically evaluate the effects of neoadjuvant chemotherapy (NAT) including anti-HER2 agents on HER2
protein expression (P-EXP) and gene amplification (G-AMP), and to examine the potential prognostic impact of HER2 alterations
on outcomes in HER2+ BC.
Design: We retrospectively identified 129 patients with HER2+ BC who received NAT at our institution. HR (ER and PR) and
HER2 P-EXP were evaluated by immunohistochemistry (IHC) and HER2 G-AMP by fluorescence in situ hybridization (FISH) in
pre- and post-tx samples. Pathologic tumor responses were categorized as no residual disease (RD; no invasive or in situ
cancer), residual ductal carcinoma in situ (DCIS) only, and residual invasive carcinoma. Pathologic complete response (pCR) was
defined as DCIS only or no RD. Association between groups was determined by Fisher exact test. For survival analysis, Kaplan
Meier curves were constructed and significance in curve separation was assessed by log rank test.
Results: Mean follow-up was 52.6 months (range 6-162). Eighty-four (65%) cases had residual invasive cancer, 21 (16%) residual
DCIS only and 24 (19%) no RD. Post-tx HER2 status was available in 70 cases with residual invasive cancer, and 27 (39%) of
these showed reduced HER2 P-EXP (staining intensity 0, 1+, 2+) by IHC. Twenty-one (78%) of 27 cases with reduced HER2
P-EXP retained HER2 G-AMP (IHC-/FISH+), and 6 (22%) were negative for both HER2 P-EXP and G-AMP (IHC-/FISH-). The
subset of IHC-/FISH- patients showed 100% 5-year recurrence free survival (RFS), whereas IHC-/FISH+ and IHC+ (intensity 3+)
cases demonstrated similarly decreased 5-year RFS of 67% and 71%, respectively. There was a trend towards reduced HER2
expression in HR+ versus HR- cases (46% vs 26%, p=0.19). HR-HER2+ tumors were more likely to achieve pCR than
HR+HER2+ cases (48% vs 26%; p=0.014, OR=0.375). In the HR- subset, RFS was significantly better in patients with no RD
compared to those with residual invasive cancer (p=0.047); this relationship was not observed in the HR+ group (p=0.693).
Conclusion: A significant fraction of pre-tx HER2+ BC demonstrate reduced HER2 P-EXP following NAT. In a subset of these,
post-tx decreased HER2 P-EXP is associated with retained HER2 G-AMP, suggestive of HER2 protein downregulation. In
another subset, decreased HER2 P-EXP with associated lack of HER2 G-AMP indicates selection for HER2 non-amplified clones
in HER2 heterogeneous tumors. Array comparative genomic hybridization studies are in progress to further elucidate these
mechanisms. Lastly, our results suggest that post-tx evaluation of HER2 status by FISH in addition to IHC may have prognostic
and/or predictive value in HER2+ BC. Further studies in larger prospective study populations are needed to confirm our findings.

Title: MammaPrint High1/High2 risk class as a biomarker of response to neratinib plus standard neoadjuvant therapy for breast
cancer in the I-SPY 2 TRIAL
Christina Yau1, Denise M Wolf1, Ashish Sanil2, Jo Chien1, Anne Wallace3, Judy Boughey4, Doug Yee5, Debu Tripathy6, Angela
DeMichele7, Rita Nanda8, Steven Chiu9, Claudine Isaacs10, Kathy Albain11, Hank Kaplan12, Stacey Moulder13, Rebecca Viscusi14,
Donald Northfelt15, Kirsten Edmiston16, Anthony Elias17, Toncred Styblo18, Barbara Haley19, Lamorna Brown-Swigart1, Susan
Flynn1, Gillian L Hirst1, Meredith Buxton1, Nola Hylton1, Melissa Paoloni20, W Fraser Symmans13, Laura Esserman1, Don Berry2,
Minetta C Liu4, John W Park1 and Laura van 't Veer1. 1University of California, San Francisco, CA; 2Berry Consultants, LLC;
3
University of California, San Diego, La Jolla, CA; 4Mayo Clinic, Rochester; 5University of Minnesota, Minneapolis, MN; 6University
of Southern California, Los Angeles, CA; 7University of Pennsylvania, Philadelphia, PA; 8University of Chicago, Chicago, IL;
9
Oregon Health & Science University, Portland, OR; 10Georgetown University; 11Loyola University Medical Center; 12Swedish
Cancer Institute; 13University of Texas MD Anderson Cancer Center, Houston, TX; 14University of Arizona, Tucson, AZ; 15Mayo
Clinic, Scottsdale; 16Inova Health System; 17University of Colorado; 18Emory University, Atlanta, GA; 19University of Texas,
Southwestern, TX and 20Quantum Leap Healthcare.
Body: Background: Further stratification of the 70-gene MammaPrintTM signature into high and ultra-high risk groups may help
predict chemo-sensitivity. In I-SPY 2, patients were classified as MammaPrint High1 (MP1) or MammaPrint (ultra) High2 (MP2),
with MP2 defined as MP_score <-0.154. MP1/MP2 classification was added to HR and HER2 to define the cancer subtypes used
in the I-SPY 2 adaptive randomization engine. Neratinib (N), one of the experimental agents evaluated in I-SPY 2, graduated in
the HR-HER2+ signature. All patients received at least standard chemotherapy (paclitaxel followed by
doxorubicin/cyclophosphamide; T->AC). HER2- patients were randomized to receive N+T- >AC vs. T->AC. For HER2+ patients,
neratinib was administered in place of trastuzumab (N+T->AC vs. H+T->AC). Here, we assess the performance of MP1/MP2
class as a specific biomarker of neratinib response.
Methods: 115 patients in the neratinib arm and 76 concurrently randomized controls had Agilent 44K microarrays and pCR data
available for analysis. We assess association between MP1/MP2 and response in the neratinib and control arms alone using
Fishers exact test, and relative performance between arms (biomarker x treatment interaction, likelihood ratio p < 0.05) using a
logistic model. This analysis is also performed adjusting for HR status as a covariate, and in receptor subsets. Our study is
exploratory with no claims for generalizability of the data. Statistical calculations are descriptive (e.g. p-values are measures of
distance with no inferential content). Our analyses do not adjust for multiplicities of other biomarkers in the trial but outside this
study.
Results: There are 133 MP1 patients (neratinib: 74, Control: 59) and 58 MP2 patients (neratinib: 41, Control: 17), 84% (49) of
which are Her2-. The distribution of pCR rates among MP1/MP2 dichotomized groups are summarized in Table 1.
Neratinib (n=115)
Control (n=76)
MP1 (n=74)
MP2 (n=41)
MP1 (n=59)
MP2 (n=17)
HER2- (n=105)
0 / 17
15 / 33
7 / 39
5 / 16
HER2+ (n=86)
22 / 57
4/8
5 / 20
0/1
MP2, one of the 10 eligible signatures, did not meet the graduation threshold; and MP1/MP2 did not show a significant biomarker
x treatment interaction (OR in neratinib relative to control arm = 1.25). The MP1/MP2 x treatment interaction remains
non-significant after adjustment for HR and HER2 status (p=0.54). In HER2- patients receiving neratinib, 45% (15/33) of MP2
patients achieved a pCR, compared to 0% (0/17) of MP1 patients. In the HER2- controls, there is a 31% pCR rate in MP2 (5/16)
vs. 18% in MP1 (7/39) patients (OR=2.14). This difference in performance between treatment arms appears significant (p=0.041).
90% of HER2+ patients are MP1, thus MP1/MP2 status x treatment interaction within the HER2+ subtype cannot be evaluated.
Conclusion: Within the I-SPY 2 population as a whole, MP1/MP2 stratification does not appear to be a specific biomarker of
response to neratinib relative to the control arm. The number of HER2- patients is small and precludes any definitive conclusion,
but these data motivate further investigation of the biological mechanisms distinguishing MP1 from MP2 to better understand
chemotherapy and/or neratanib responsiveness.


Title: Patient-derived xenografts accurately predict patient response in breast cancer patients
Lisa Stow1, Amanda Katz1, Hanna Irie2, Elisa Port3, Justin Stebbing4, Daniel Ciznadija1, Angela Davies1 and Keren Paz1.
1
Champions Oncology, Inc, Hackensack, NJ; 2Icahn School of Medicine at Mount Sinai, New York, NY; 3Mount Sinai Hospital,
New York, NY, United Kingdom and 4Imperial College, London, United Kingdom.
Body: PURPOSE/OBJECTIVES: A growing body of evidence demonstrates that patient-derived xenografts (PDXs) represent
living tumor models that accurately reflect the biology of the primary patient tumor. More importantly, we have previously shown
that PDX models show responses to therapeutic agents that are concordant with patient clinical response and can be used to
direct personalized cancer treatments (Stebbing, 2014). Here we report the ability of PDX models to predict for patient response
to drug treatment in a cohort of breast cancer patients.
MATERIALS/METHODS: Tumors were resected from patients with either primary or metastatic breast cancer and implanted into
immunodeficient mice to establish PDX models. Successfully engrafted PDX models were expanded and randomized for drug
sensitivity testing. Tumor growth inhibition and tumor regression were captured and results were correlated with a patients clinical
response. In some cases, PDX results were used to personalize cancer treatment and some patients used PDX-directed
treatments over multiple lines of therapy.
RESULTS: A total of 42 tumors from 40 patients were implanted resulting in 21 successfully engrafted PDX models (50%
engraftment rate). Notably, engraftment rates were much higher for patients with triple negative breast cancer (TNBC) and
resulted in 7 successful PDX models from 8 TNBC patients (87.5% engraftment rate). Drug sensitivity testing was offered to
patients with established PDX models. Drugs and drug combinations tested included standard and nonstandard chemotherapy as
well as biologics. At that time of this publication, 4 patients (3 TNBC and 1 HER2+) with completed drug sensitivity tests also had
clinical data available resulting in 7 clinical correlations; 4 retrospective and 3 prospective. In all 7 cases, the PDX model
accurately predicted patient clinical response demonstrating an accuracy of 100%. Five of the drug tests predicted drug sensitivity
and 2 tests predicted resistance, indicating the potential of the PDX platform to predict for both sensitivity and resistance to
therapy. The 3 prospective correlations resulted in concordant clinical benefit in 2 patients for duration greater than 6 months
each.
CONCLUSIONS: These data support the use of the personalized PDX model as a platform for therapeutic decision making that
can guide treatment for patients with breast cancer. A prospective clinical trial in TNBC is currently underway.

Title: Genetic heterogeneity for Her2 accounts for a significant percentage of breast cancers changing Her2 status following
implementation of the 2013 CAP/ASCO HER2 reporting guidelines
Monica V Estrada1, Jena M Giltnane1, Ferrin C Wheeler1, Ashwini Yenamandra1, Vandana Abramson1, Ingrid A Mayer1, Julie
Means1, Brent Rexer1, Ingrid M Meszoely1 and Melinda E Sanders1. 1Vanderbilt University Medical Center, Nashville, TN.
Body: Background: Current evidence indicates that patients with Human Epidermal Growth Factor Receptor 2 (HER2)-positive
Breast Cancer (BC) benefit from HER2-targeted therapies. Accurate determination of HER2 status is critical to ensure that the
correct patients are offered targeted treatment while patients with HER2-negative tumorswho are unlikely to benefit from
anti-HER2 therapy are spared from the adverse effects of these costly drugs. Guidelines for performance and reporting of HER2
testing, first published in 2007 by American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP), were
updated in October 2013. The new reporting criteria are based on a combination of HER2:CEP17 ratio and average HER2 copy
number.
Methods: We retrospectively reviewed HER2 dual probe FISH test results sequentially performed by the Cytogenetics Laboratory
at Vanderbilt University from January to May 2014 since implementation of the updated guidelines. The cases were then rescored
using the 2007 guidelines. The clinicopathologic features of cases with a change in Her2 status after scoring with the 2013
guidelines were examined for statistical significance.
Results: If the 266 performed HER2 FISH tests had been scored using the 2007 guidelines the results would have been the
following: amplified (n = 44, 16%), equivocal (n = 28, 10%) and not amplified (n = 194, 70%). However, using the 2013 guidelines
the cases were actually reported as follows: amplified (n=57, 21%), equivocal (n=22, 8.3%) and not amplified (n=187, 68%).
Additionally, 18 cases demonstrated genetic heterogeneity in >25% of cells. The updated guidelines resulted in a change in Her2
status in 12% of tests (32/266; p less than 0.0001); 13 changed from equivocal to amplified, 13 cases changed from not amplified
to equivocal and 6 cases changed from equivocal to not amplified. Cases with a change in Her2 status following implementation
of the new guidelines were more likely to demonstrate genetic heterogeneity (28% [9/32] vs. 4% [9/234]; p less than 0.0001).
Furthermore, hormone receptor (HR)-negative tumors scored as not amplified by the 2007 guidelines were more likely to undergo
an upgrade in HER2 status under the 2013 guidelines than HR-positive tumors (26% [11/41] vs. 7% [11/153], p less than 0.0001).
Conclusions: The 2013 reporting criteria, based on a combination of HER2:CEP17 ratio and average HER2 copy number,
increase the number of patients eligible for HER2-targeted therapies while decreasing equivocal results. Tumors that demonstrate
genetic heterogeneity for Her2 or are HR-negative account for a significant percentage of these newly eligible cases. Correlation
with clinical response is required to confirm the proposed improved analytical validity of the updated guidelines. Clinical trials to
evaluate the benefit of anti-Her2 therapy in patients with genetic heterogeneity are in the planning phase.

Title: Comparison of pathologic response evaluation systems after neoadjuvant chemotherapy among different molecular
subtypes of breast cancers
In Ah Park1, Hee Jin Lee1, In Hye Song1 and Gyungyub Gong1. 1University of Ulsan College of Medicine, Asan Medical Center,
Seoul, Korea.
Body: Background: Several ways to assess pathologic response of breast cancer after neoadjuvant chemotherapy (NAC) are
currently available. However, clinical usefulness of assessment systems in each molecular subtype of breast cancer is unclear.
Therefore, we compared four pathologic response assessment systems predicting patients clinical outcome in specific subtypes
of breast cancer.
Methods: In total, 598 tumors from 590 female breast cancer patients who received anthracycline and taxane-based NAC and
subsequent surgery from 2010 to 2012 were analyzed. Molecular subtypes were defined by immunohistochemistry (HER2 and
hormone receptor (HR): estrogen receptor and progesterone receptor). Miller-Payne grading, Residual Cancer Burden, Residual
Disease in Breast and Nodes and ypTNM stage were evaluated. Results of each assessment system were analyzed for survival
with Kaplan-Meier and Cox hazard model. Median follow-up period was 35.2 months (range 21.1-54.4 months).
Results: Pathologic complete response was achieved in 4.4% (12/275) of HR+/HER2-, 10.7% (8/75) of HR+/HER2+, 17.8%
(16/90) of HR-/HER2+, and 29.7% (47/158) of triple negative (TN) tumors. Results of all four examined assessment systems were
significantly correlated with disease-free and overall survival in all tumors. In HR+/HER2- and TN tumors, all systems were
associated with disease-free and overall survival.
Comparison of pathologic response assessment systems after neoadjuvant chemotherapy for disease-free and overall survival
Disease-free survival
Miller Payne grade
RCB class
RDBN level
ypTNM Stage
HR+/HER2-
HR+/HER2+
HR-/HER2+
TN
HR (95% CI) 0.629 (0.429-0.925)
0.310 (0.159-0.607)
0.647 (0.428-0.977) 0.451 (0.343-0.593)
P value
<0.001
.039
HR (95% CI) 2.621 (1.398-4.914)
2.454 (0.917-6.564)
1.696 (0.929-3.094) 2.966 (1.982-4.441)
P value
.074
.085
HR (95% CI) 2.892 (1.642-5.093)
2.587 (1.015-6.595)
2.189 (1.118-4.286) 3.065 (2.108-4.456)
P value
.047
.022
HR (95% CI) 2.975 (1.606-5.512)
2.062 (0.918-4.632)
2.004 (1.121-3.582) 2.950 (2.152-4.044)
P value
.080
.019
HR (95% CI) 0.416 (0.222-0.778)
0.399 (0.177-0.896)
0.633 (0.357-1.124) 0.397 (0.294-0.537)
P value
.026
.119
.018
.003
<0.001
.001
<0.001
<0.001
<0.001
<0.001
Overall survival
Miller Payne grade
RCB class
RDBN level
<0.001
HR (95% CI) 8.611 (1.924-38.531)
2.971 (0.733-12.032) 1.900 (0.771-4.680) 2.976 (1.927-4.594)
P value
.127
.005
.163
<0.001
HR (95% CI) 11.369 (2.618-49.375) 3.042 (0.801-11.550) 1.970 (0.805-4.821) 3.262 (2.145-4.961)
P value
ypTNM Stage
.006
.001
.102
.138
HR (95% CI) 5.812 (1.666-20.270)
2.623 (0.851-8.080)
1.619 (0.739-3.546) 3.040 (2.150-4.300)
P value
.093
.228
.006
<0.001
<0.001
RCB, residual cancer burden; RDBN; residual disease breast and nodes.
Especially in TN, Kaplan-Meier survival curves for disease-free survival were clearly separated by all assessment methods.
However, in HR+/HER2+ and HR-/HER2+ tumors, the association of patients survival with response assessment results was
variable according to the examined systems.

Conclusion: By the available pathologic assessment systems after neoadjuvant chemotherapy, HR+/HER2- and TN breast
cancers could be effectively classified into groups showing different prognosis. However, for the evaluation of pathologic
response of HR+/HER2+ and HR-/HER2+ tumors, the development of effective assessment methods is warranted.

Title: Platelet predominant breast cancer is a new predictor for pathological complete response to neoadjuvant chemotherapy
Satoko Ishikawa1, Masafumi Inokuchi1, Hironori Hayashi2, Katsunobu Oyama2, Hisatoshi Nakagawara2, Tomoharu Miyashita2,
Hidehiro Tajima2, Hiroyuki Takamura2, Itasu Ninomiya2, Hirohisa Kitagawa2, Sachio Fushida2, Takashi Fujimura2 and Tetsuo
Ohta2. 1Kanazawa University Hospital, Kanazawa, Ishikawa, Japan and 2Kanazawa University Hospital, Kanazawa, Ishikawa,
Japan.
Body: Background and Objective: Platelets, the smallest anucleate hematopoietic cells, are increasingly recognized as the key
regulator of tumor progression and metastasis in breast cancer. They contribute significantly to hematogenous metastasis, which
increases tumor invasiveness, by protecting tumor cells from shear stress or immune surveillance and facilitating their
extravasation to distant sites. Additionally, recent studies have demonstrated direct signaling between platelets and tumor cells
during epithelialmesenchymal transition (EMT) in the blood stream. However, the existence of platelets in primary breast cancer
and its relationship with clinicopathological factors and pathological response to chemotherapy remain poorly understood. This
study aimed to investigate (1) the presence of platelet infiltration in tumor tissue, (2) relationships between platelets and
clinicopathological features, and (3) the association of platelets with pathological complete response (pCR) to neoadjuvant
chemotherapy and EMT markers in primary breast cancer.
Methods: We retrospectively analyzed the pre-chemotherapeutic biopsy specimens from 100 breast cancer patients who
underwent surgical resection after anthracycline/taxane-based neoadjuvant chemotherapy at Kanazawa University Hospital.
Platelet subsets (glycoprotein Ib [CD42b]) and the expression of EMT markers (E-cadherin, vimentin, and -catenin) were
evaluated by immunohistochemistry and correlated with pCR after neoadjuvant chemotherapy. Platelet-predominant breast
cancer (PPBC) was defined as the presence of 10% of anucleate cells positively stained for CD42b in direct contact with tumor
cells. pCR was defined as the absence of residual invasive tumor cells in breast and lymph nodes (ypT0/is, ypN0).
Results: PPBC was observed in 48 patients (48%). The prevalence of PPBC was significantly higher in human epidermal growth
factor receptor 2 (HER2)-negative patients (HER2 [+] vs. HER2 [-]; 65% vs. 90%, p = 0.008). There was no significant association
between CD42b expression and stage, nuclear grade, histology, and estrogen receptor status. PPBC patients had a significantly
decreased pCR rate (6/48, 12.5%) compared to non-PPBC patients (33/51, 64.7%; p < 0.001). Tumor cells surrounded by
platelets exhibited EMT-like morphological changes, EMT marker expression, nuclear-staining of -catenin, loss of E-cadherin
expression, and expression of vimentin.
Conclusions: Our results suggested the presence of tumor-associated platelets in breast cancer as a new predictor of response to
neoadjuvant chemotherapy. Platelets might facilitate EMT at the primary site by potentiating tumor cell adhesion. Platelet-tumor
interaction might be a new therapeutic strategy for breast cancer. If our finding is validated in further investigations, it might serve
as the basis for novel therapeutic approaches of combining conventional chemotherapy with antiplatelet therapy. Our data also
identified new predictive parameters to stratify patients with an increased chance of response to anthracycline/taxane-based
neoadjuvant chemotherapy.

Title: Association of estrogen receptor (ER) levels and prediction of antiproliferative effect of hormone therapy (HT) in lower
ER-expressing tumors
J Michael Dixon1, Arran Turnbull2, Lorna Renshaw1, Megan P Rothney3, Cynthia A Loman3, Laura Arthur2, Jeremy S Thomas1,
Oliver Young1, Juliette Murray1, Linda Williams2, Amy P Sing3 and David Cameron2. 1Edinburgh Breast Unit, Edinburgh,
Midlothian, United Kingdom; 2Edinburgh University, Edinburgh, Midlothian, United Kingdom and 3Genomic Health Inc, Redwood
City, CA.
Body: Intro
Accurate measurement of ER in early stage invasive breast cancer (EBC) is important to identify patients likely to benefit from
HT. While immunohistochemistry (IHC) is the most common method to quantify ER, other methods can also accurately measure
ER, such as RT-PCR. ER is one of the genes included in the RT-PCR based 21-gene Recurrence Score assay (Oncotype DX,
Genomic Health, Redwood City, CA) and is also reported separately as a single gene expression. Additionally, the association
between ER expression by RT-PCR (ER-PCR) and tamoxifen benefit has been reported by Kim, et al (2011). A recent study
reported that patients with ER levels <10% by IHC were largely negative by RT-PCR (Singh, et al; 2014) and that this has
potential implications for which patients may be expected to benefit from HT. Robust measures of ER and the proliferative
response may be useful in identifying patients likely to respond to HT.
Aim
The study aims are: (1) To correlate quantification of ER in EBC as assessed by Allred Score (AS) and ER as measured by
RT-PCR in the 21-gene assay; (2) To describe changes in ER, Recurrence Score, and measures of proliferation after 2wks of an
aromatase inhibitor (AI); (3) To perform exploratory analyses of factors associated with changes in proliferation.
Methods
55 postmenopausal EBC patients with lower ER (AS 2-7) were treated with 2wks of an AI followed by wide excision. All patients
had a 21-gene assay on a pre-and post-treatment (Tx) sample. Proliferation was measured by both Ki67 by IHC (in 45 patients)
and by the proliferation gene group score (PGS) in the RT-PCR based 21-gene assay (in all patients). Proliferation response was
defined by a 20% relative decrease in Ki67 or a decrease in PGS. Changes in proliferation were correlated with AS, ER-PCR and
Recurrence Score result.
Results
The Table shows the correlation of AS with ER-PCR measured in the pre-Tx (r=0.83) samples. 94% of AS (2-3) patients and 56%
of AS (4-5) were ER(-) by RT-PCR There was a significant change (pre to post) in the average Ki67 level (18% to 11%; p<0.001)
and PGS but not Recurrence Score result. 37/45 patients showed a 20% decrease in Ki67 while only 32/55 had a decrease in
PGS. Changes in Ki67 levels were greatest in AS 6/7 patients with a 76% relative decrease vs 21% in AS 2/3 patients. The range
of PGS change was 1% increase in AS 2/3, 1% decrease in AS 4/5 and 14% decrease in AS 6/7 patients. Univariate predictors of
decrease in Ki67 were AS of 5/6/7 (vs 2/3/4), Recurrence Score result, ER-PCR (continuous or binary), and PR-PCR. The same
variables were predictors of PGS change.
Pre-Tx
AS
ER-PCR Status
(Pre-Tx)
Negative
Positive
Total
AS 2-3
17
(94%)
1
(6%)
18
AS 4-5
9
(56%)
7
(44%)
AS 6-7
0
(0%)
21
(100%)
21
Total
26
29
55
Conclusions
Results confirm earlier reports showing substantial disagreement in ER measured by IHC vs RT-PCR in patients with lower
ER-expressing tumors
The clinical implications are that a substantial number of patients with low ER by IHC may have little to no benefit from HT
The 21-gene assay may be useful in selecting patients likely to benefit from HT
Further studies in larger cohorts are required to confirm these findings.

Title: Assessment of phosphorylated HER2 protein as predictive biomarker to stratify anti-HER2 treatment in HER2 non-amplified
patients A translational study in the GeparQuattro and GeparQuinto trials
Stefanie Avril1,8, Gunter von Minckwitz2, Sibylle Gndisch1, Stephan Gade2, Katharina Malinowsky1, Peter A Fasching3, Thomas
Karn4, Arndt Hartmann5, Michael Untch6, Carsten Denkert7, Karl-Friedrich Becker1 and Sibylle Loibl2. 1Technische Universitt
Mnchen, Munich, Germany; 2German Breast Group, Neu-Isenburg, Hessen, Germany; 3University Hospital Erlangen,
Comprehensive Cancer Center Erlangen-EMN, Erlangen, Germany; 4University Hospital Frankfurt, Frankfurt am Main, Germany;
5
University Hospital Erlangen, Erlangen, Germany; 6Helios Klinikum Berlin-Buch, Berlin, Germany; 7Institute of Pathology, Charit
Universittsmedizin Berlin, Berlin, Germany and 8Case Western Reserve University and University Hospitals Case Medical
Center, Cleveland, OH.
Body: Aims: There is evidence that the benefit of anti-HER2 treatment in combination with chemotherapy may not be limited to
patients with HER2 amplification. This study tested if phosphorylated HER2 (pHER2) and co-activation of HER2 downstream
targets are predictive of response to anti-HER2 treatment in HER2 non-amplified patients.
Methods: Patients initially classified as HER2 positive (IHC, FISH) by local testing from the GeparQuattro and GeparQuinto trials
received neoadjuvant anti-HER2 treatment (trastuzumab or lapatinib) in combination with anthracycline-taxane-based
chemotherapy. However, on a central pathology review 98 out of 1060 patients were considered HER2 non-amplified. In order to
assess the potential benefit from anti-HER2 treatment in these patients the levels of pHER2 and downstream targets including
pHER3, EGFR, AKT, PI3K, ERK, PTEN, S6RP and their phosphorylated forms were quantitatively assessed using reverse-phase
protein arrays. Histopathological complete response (pCR; ypT0/is) and disease free and overall survival served as reference
standard. Optimal cutpoints for each protein were calculated to achieve maximum separation between pCR and non-pCR using
ROC analysis.
Results: 25 (26%) patients achieved pCR. The level of pHER2 was not significantly different between groups of histopathologic
responders and non-responders. S6RP and pS6RP were the only proteins significantly associated with pCR (p=0.01 and 0.03,
respectively) with a higher pretherapeutic expression in patients who subsequently achieved pCR. In contrast, low expression of
S6RP and pS6RP were associated with longer disease free (p<0.01) and overall survival (p<0.05). Expression of S6RP and
pS6RP did not correlate with that of other HER2 signaling proteins, whereas all remaining proteins were positively correlated with
each other.
Conclusions: Expression of S6 ribosomal protein (S6RP), a downstream target of S6 kinase, and its phosphorylated form
pS6RP are significantly associated with short and long-term outcome following anti-HER2 treatment in HER2 non-amplified
breast cancer patients. In contrast, activation status of the HER2 pathway reflected by pHER2 and other downstream targets was
not predictive of response, and showed no significant correlation with S6RP expression. These results suggest that S6RP and
pS6RP may be novel predictive biomarkers for response to anti-HER2 treatment in non-amplified patients, possibly through a
mechanism independent of global HER2 pathway activation.

Title: I-SPY 2 qualifying biomarker evaluation (QBE): The challenge and opportunity for interrogating predicted pathways in an
adaptive design biomarker rich trial
Christina Yau1, Denise Wolf1, Ashish Sanil2, Laura van 't Veer1, Emanuel F Petricoin3, Meredith Buxton1, Joe Gray4, Angela
DeMichele5, Mike Hogarth6, Nola Hylton1, Jane Perlmutter7, Melissa Paoloni8, Fraser Symmans9, Doug Yee10, Don Berry9 and
Laura Esserman1. 1University of California, San Francisco, CA; 2Berry Consultants, LLC; 3George Mason University; 4Oregon
Health & Science University, Portland, OR; 5University of Pennsylvania, Philadelphia, PA; 6University of California, Davis, CA;
7
Gemini Group; 8Quantum Leap Healthcare; 9University of Texas MD Anderson Cancer Center, Houston, TX and 10University of
Minnesota, Minneapolis, MN.
Body: I-SPY 2, a multicenter phase 2 neoadjuvant trial in high-risk breast cancer, uses adaptive randomization within biomarker
subtypes to evaluate novel agents added to standard chemotherapy. In addition to efficiently evaluating agent/signature pairs,
I-SPY 2 is a biomarker rich trial, where samples are profiled for gene expression, protein levels, and mutation status. Biomarkers
are classified as established, qualifying, or exploratory. Established biomarkers are those used clinically (HR/HER2 status) or
FDA cleared (MammaPrint), and,used for adaptive randomization to generate the 10 signatures from which a drug can graduate.
Qualifying biomarkers (QB) represent evidence-based, biologic pathway markers (e.g.cell line predictors, known drug targets).
QB analyses must be pre-specified and performed under CLIA. Exploratory markers are for discovery and may allow integration
of data from different technologies.
The QBE goal is to (1) evaluate biomarkers related to an agents mechanism of action to identify promising candidates for
testing/patient selection in future trials, and (2) create a resource to elucidate biological mechanisms of response.
The wealth of biomarker data is both a boon and a challenge. Our small size limits the generalizability of our findings. There are
multiple genes in each pathway measured on multiple platforms, creating the problem of multiplicity, which is compounded by the
evaluation of multiple proposals. Biomarkers may correlate with HR/HER2/MP subtypes. The adaptive randomization may
increase the prevalence of biomarker positive subsets and bias our findings. These challenges limit definitive conclusions, so our
statistics are descriptive rather than inferential, and are intended to avoid adding to the false positive biomarker literature.
Methods: Three filters are applied: 1-The difference in biomarker performance in the experimental vs control arm (biomarker x
treatment interaction) is evaluated using a logistic model under a pre-specified analysis plan 2-Biomarkers with a treatment
interaction are dichotomized. The QB-High group is added to the graduating subtype to define a novel signature and the
treatment effect in this group is evaluated 3-If the treatment effect is comparable to the graduating signature, and the prevalence
is increased, the I-SPY 2 Bayesian model is modified to include the QB to assess the novel signature.
QBE to date: Veliparib in combination with carboplatin (V/C) and neratinib (N) are the first two agents to graduate from I-SPY 2.
For V/C, we have completed initial evaluation for 5 biomarker proposals, including BRCA1/2 germline mutations and expression
signatures associated with DNA repair deficiencies. For N, 6 biomarker proposals, including HER family protein signaling
markers, have been assessed. Evaluation of the best candidates from these initial analyses in the I-SPY 2 Bayesian framework is
ongoing. Mutational analyses are pending.
Conclusions: We have developed a rigorous approach for QB analysis. A small number of QB warrant further assessment.
However, I-SPY 2 QB require validation, and should be considered preliminary efforts to effectively screen QB candidates for
evaluation in ongoing and future trials.

Title: Optical metabolic imaging predicts therapeutic response in breast cancer tumors and organoids
Alex J Walsh1, Rebecca S Cook2, Melinda E Sanders2, Carlos L Arteaga2 and Melissa C Skala1. 1Vanderbilt University, Nashville,
TN and 2Vanderbilt University Medical Center, Nashville, TN.
Body: The standard of care for breast cancer patients includes treatment with chemotherapy, antiestrogens, and targeted
inhibitors. There are more than 100 drugs approved to treat breast cancers with more in the pipeline of discovery and approval;
however, few biomarkers have been identified for individualized, patient-specific treatment planning. Currently, drug regimens are
chosen based on the presence or absence of specific receptors, including progesterone receptor (PR), estrogen receptor (ER),
and human epidermal growth factor receptor 2 (HER2) and patient response is assessed by changes in tumor size weeks after
treatment. Unfortunately, many patients do not initially respond to therapy or develop a resistance, and face a greater risk of
recurrence and death.
Herein we present a novel, light-based technology that can accurately predict individual tumor response to anti-cancer drugs
based on drug-induced changes in metabolism. Cellular metabolism is a powerful indicator response to treatment, because the
oncogenic drivers targeted by therapeutic agents often regulate cellular metabolism. In this study, we demonstrate the sensitivity
of optical metabolic imaging to predict therapeutic response in xenografts in vivo, and in primary tumor derived organoids. Optical
metabolic imaging utilizes mulitphoton fluorescence intensity and fluorescence lifetime imaging to probe the concentrations and
protein-binding dynamics of cellular NADH and FAD, two coenzymes of metabolism. A composite endpoint, the OMI index,
provides a robust, dynamic readout of cellular metabolism.
First, we probed metabolic changes within ER+/HER2+ tumors treated with trastuzumab, an anti-HER2 antibody. After 48h of a
single injection of trastuzumab (HER2 antibody), the OMI index of responsive tumors (ER+/HER2+) decreased (p<0.001), but
remained unchanged in resistant tumors. To optimize clinical utility of this technology, we developed protocols to generate
organoids from fresh biopsy samples. This allows high-throughput screening to directly test individual tumor response to a panel
of drugs. We validated this approach in trastuzumab responsive and resistant tumors: the OMI index decreased in responsive
ER+/HER2+ organoids treated with trastuzumab, paclitaxel, and XL147 (PI3K inhibitor) (p<0.05). Combination therapies resulted
in the lowest OMI index values (p<0.001). Measurements on patient-derived organoids resulted in similar findings: reductions in
ER+ tumors treated with tamoxifen and paclitaxel (p<0.05), and the greatest reductions in OMI index observed with combination
treatments. The high-resolution images of OMI allow segmentation of the cells within each organoid for cellular-level analysis of
heterogeneity. We identified heterogeneity profiles of resistant cell populations within the xenograft and human organoids and
were able to track the emergence of resistant sub-populations of cells over the first 72 hours of drug treatment. Heterogeneity
analysis is important for clinical utility of this technology to ensure the optimal drug regimen is selected. With these results, optical
metabolic imaging shows potential for development into a high-throughput screen to test the efficacy of a panel of drugs to direct
clinical therapy selection and expedite pre-clinical studies.

Title: Anaplastic lymphoma kinase (ALK) protein overexpression is not a feature of inflammatory breast cancer
Cecile Colpaert1, Melike Marsan2, Peter Vermeulen1, Luc Dirix2, Steven Van Laere2 and Inflammatory Breast Cancer International
Consortium3. 1Oncology Centre, GZA Hospitals, Iridium Cancer Net, Antwerp, Belgium; 2Translational Cancer Research Unit,
Oncology Centre, GZA Hospitals, Iridium Cancer Net, Antwerp, Belgium and 3Inflammatory Breast Cancer International
Consortium.
Body: Recent studies suggest that anaplastic lymphoma kinase (ALK) could serve as a therapeutic target for inflammatory breast
cancer (IBC) and that strategies targeting ALK should be considered for evaluation in clinical trials. Reverse phase protein arays
revealed activation of the ALK receptor tyrosine kinase and biochemically-linked downstream signalling molecules in pre-clinical
models of IBC. ALK genetic abnormalities have been reported in 80% (20/25) of IBC patients with a high prevalence of ALK
alterations in basal-like IBC (Robertson F. et al., 2013). However, ALK protein expression has not been studied in human IBC
samples.
Immunohistochemical detection of the ALK protein is increasingly being used as a screening tool to test samples for ALK
rearrangements. The biological premise is that the genetic abnormality of the ALK gene leads to overexpression of the ALK
protein and therefore overactivity of the ALK tyrosine kinase; since this kinase is the target of crizotinib, it makes sense to assess
the drug target directly.
Formalin fixed paraffin embedded tissue samples from pre-treatment surgical biopsies of 79 consecutive IBC patients were
immunohistochemically (IHC) tested for ALK protein over-expression using a validated IHC test (NCL-ALK, clone 5A4) with a
sensitivity of 93% and a specificity of 100% when the Vysis ALK-FISH break apart test is used as a gold standard. Stained tissue
sections were evaluated using the IHC histoscore proposed by Thunnissen E. et al., 2012, assessing both the proportion of tumor
cells showing cytoplasmic staining and the staining intensity. Cytoplasmic staining of appendiceal ganglion cells was used as an
on-slide external positive control; weak to moderate cytoplasmic staining of macrophages was used as an internal positive
control.
In 75 of the 79 tissue samples none of the tumor cells showed any ALK immunoreactivity. One tumor showed moderate
cytoplasmic staining in a few cells (<1%, score 2+); ALK (2p23) FISH showed no rearrangement. This tumor was hormone
receptor negative and HER2 negative. Three tumors showed minimal cytoplasmic staining in a few cells (<1%, score 1+).
These results demonstrate that ALK protein overexpression is not a feature of IBC. Although genetic abnormalities of ALK without
protein overexpression are not excluded, our results show that ALK IHC cannot be used to identify IBC patients eligible for
enrollment in clinical trials evaluating ALK targeted therapeutics.

Title: Prognostic impact of PgR and Ki-67 expression in ER-positive and HER2-negative breast cancer patients
Yasuyuki Nishiyama1, Reiki Nishimura1, Tomofumi Osako1, Yasuhiro Okumura1 and Nobuyuki Arima2. 1Kumamoto City Hospital,
Kumamoto, Japan and 2Kumamoto City Hospital, Kumamoto, Japan.
Body: Background:
The 13th St Gallen International Breast Cancer Conference (2013) proposed that conventional and new clinico-pathological
factors provided a surrogate subtype classification. Moderate or strong expression of PgR has been proposed as an additional
restriction in the surrogate definition of Luminal disease. Ki-67 level is also important for this distinction. We divided ER-positive
and HER2-negative breast cancer into 4 groups according to the PgR and Ki-67 expressions. The aim of this study was to
evaluate the clinical and prognostic significance of these markers among 4 groups.
We analyzed recurrence-free survival (RFS) in 2100 invasive cancer patients who underwent surgery at our institution between
September 2001 and December 2013 retrospectively. Patients with T4, neoadjuvant chemotherapy and adjuvant trastuzumab
therapy were excluded. The age of the patients ranged from 25 to 75 years. The subtypes were defined as LA (PgR20% and
Ki-67<20%; n=677), LB-1 (PgR20% and Ki-6720%; n=524), LB-2 (PgR<20% and Ki-6720%; n=198), LB-3 (PgR<20% and
Ki-67<20%; n=187), LH (HER2 positive; n=198), HER2 (non-luminal; n=105), and triple negative (TN, ER- and PgR- and HER2-;
n=211). RFS was compared among subtypes using Kaplan-Meier method and logrank test. The median follow-up period was
62.9 months.
Results:
In terms of RFS, patients with LA had significantly better prognosis than other subtypes (p<0.001). Patients with LB-2 had a
significantly poorer prognosis than LA, LB-1, LB-3 and HER2 (non-luminal), and the PFS of LB-2 was equivalent to LH (HER2
positive) and TN. Among ER-positive and HER2-negative patients (LA, LB-1, LB-2 and LB-3), LB-2 patients had significantly
worse prognosis than the other subtypes in node negative cases (p<0.001). On the other hand, in node positive cases, there was
no difference in PFS between LB-2 and LB-3 (p=0.298). Moreover, LB-2 patients showed a significantly worse PFS in the
endocrine therapy group (p<0.001). In the chemo-endocrine therapy group, LB-2 and LB-3 had a significantly poorer prognosis
than LA; however, there was no difference between LB-2 and LB-3 (p=0.371).
Conclusion:
In luminal type breast cancer, patients with LB-2 (PgR<20% and Ki-6720%) had the worst prognosis. On the other hand, in the
chemo-endocrine therapy group, LB-2 and LB-3 (PgR<20% and Ki-67<20%) had similar and poorer prognosis. These findings
suggested that lower expression of PgR (<20%) correlated with lower PFS and lower sensitivity to endocrine therapy, and lower
Ki-67 index (<20%) reflected lower sensitivity to chemotherapy. The expression of PgR and Ki-67 might provide a clinical
significance in the luminal type.

Title: Upfront identification of chemoresistance in locally advanced breast cancer using 99m Tc tetrofosmin scan: A low cost
technology
V Seenu1, Rakesh Kumar1 and Siddharth Dattagupta1. 1All India Institue of Medical Sciences, New Delhi, Delhi, India.
Body: With the availability of newer chemo therapeutic drugs for breast cancer with excellent efficacy; it has become essential to
identify breast cancers that are chemo resistant to doxorubicin based chemotherapy early in course of treatment so that a
different regimen can be used to improve response rates & survival. PET scan/ contrast MRI are used routinely used to assess
response to neoadjuvant chemotherapy ( NACT) in locally advanced breast cancer (LABC). However, both these modalities are
expensive & not easily available in most centres in India. 99m Tc Tetrofosmin is taken up by metabolically active breast cancer
cells, is inexpensive & easily available at most centres.This study was undertaken to evaluate the ability of Tc99 tetrafosmin scan
to identify upfront chemo resistance in LABC. Aims:Evaluate the role of Tc 99m-Tetrofosmin scan in distinguishing chemo
sensitive from chemo resistant LABC patients. Methods:N=:65 pts. Inclusion: LABC as per TNM classification .Exclusion: Pts not
fit to receive chemotherapy or refusing enrollment. Place of study: Breast Cancer Clinic, Department of Surgical Disciplines, .99m
Tc Tetrofosmin Scintimammography performed in all the patients. NACT was CAF regimen Cyclophosphomide (600mg/m2),
Doxorubicin (50mg/m2) and 5-Fluro-uracil (600mg/m2) 3 cycles @ 3 weekly intervals. Response evaluated after 3rd cycle.
Response evaluation: Scintimammo: CR: No detectable abnormal uptake on the post-chemotherapy scan.PR: Some uptake on
the post-chemotherapy scan; NR: No change in uptake when compared with pre-chemotherapy scan. evaluation. Histopathology:
CR: Lesionentirely replaced by fibrosis or when only few isolated malignant cells persist. PR: Viable invasive carcinoma persisting
in more than 25% of the area of the lesion. Chemosensitivity: A ROC curve was plotted for the T/B ratio & pathological response.
Area under the curve was found to be 84% with a C.I. between 73%-95%. Using this curve, a cut off value of 2.55 was chosen to
characterize the chemo sensitivity of the tumors.
: Scintimammographic assessment of response to NACT
Scintimammography
pathological response
CR
nouptake
11
uptake
20
Sensitivity-92%, Specificity-87%, Positive predictive value-79%, Negative predictive value-95%, LR+7, LR-0.09, LR---73.3,
accuracy-89% p value <0.000
shows T/B ratios When the T/B ratio was >2.55, the tumor can be considered to be chemo sensitive to Doxorubicin based
NACT.On the contrary, when the T/B ratio was <2.55, it is unlikely that the tumor regresses completely following Doxorubicin
based NACT. Ability of Scintimammography to assess response to NACT was done comparing post-NACT scintimammograms
with the final histopathology
T/B ratio & path response
Scintimammography
CR (tumor no)
PR(Tumor yes)
p value
T/B Ratio2.55>
18
12
<0.003
T/B ratio 2.55<<<
30
<0.003
LR+:3.3,LR-:0.07 sen:95% spe: 72%

Conclusion: Tc99 tetrofosmin can accurately identify chemo sensitive breast tumors and predict response to NACT in LABC. Its
low, cost, high accuracy and wider availability in developing nations makes it an attractive option in management of LABC.

Title: Detection of PIK3CA mutations in cell-free plasma DNA from women with early breast cancer
Kent F Hoskins1, Amer Sidani1, Tiffany Jones1, Stefan Green1, Rajyasree Emmadi1 and Anjen Chenn1. 1University of Illinois at
Chicago, Chicago, IL.
Body: Background
Activating mutations in the PIK3CA gene represent the most common molecular aberration in luminal breast cancers. Three
hot-spot mutations (E542K, E545K, H1047R) account for approximately 70% of the PIK3CA mutations identified. The presence of
a mutation is likely to become an important predictive biomarker to select patients for treatment with PIK3CA inhibitors.
Tumor-derived DNA can be detected in the systemic circulation in the form of cell-free plasma DNA (cfpDNA), and the ability to
interrogate tumor-derived DNA from an easily acquired biospecimen like cfpDNA has practical advantages over analysis of tissue
specimens. Furthermore, cfpDNA can be monitored serially and could provide a dynamic measure of tumor progression and
response to therapy. Droplet digital PCR (ddPCR) is a high-throughput technique that provides highly precise and sensitive
detection of rare alleles in samples with ultra-low abundance of target DNA.
Methods
A nested case-control study was conducted with plasma samples collected as part of a study to identify blood-based biomarkers.
Blood was collected from women scheduled for a breast biopsy at one of seven participating mammography units. All blood
samples were collected prior to the biopsy procedure and were processed using a standard protocol. Cases are women
diagnosed with invasive ductal carcinoma, stage I-III, that is positive for the estrogen receptor (ER+) and negative for HER2/neu
(0 or 1+ by IHC). Controls are women with a benign, non-proliferative lesion, matched by collection site, year of collection and
age. cfpDNA was isolated using a commercial kit (Macherey-Nagel, Bethlehem, PA). DNA was extracted from FFPE biopsy
specimens by standard techniques. Central pathologic review confirmed at least 10% tumor cellularity in biopsy specimens from
cases. A ddPCR assay was developed on a BioRad QX100 Droplet Digital PCR instrument using standard TaqManR assays with
probe pairs designed to detect the 3 PIK3CA hotspot mutations and corresponding wild type (wt) alleles. The assay was validated
using DNA from four cell lines, each positive for one of the hotspot mutations or wt at all 3 loci (Horizon Diagnostics, Cambridge,
UK). ddPCR was performed on cfpDNA and DNA extracted from FFPE biopsy specimens in cases and controls. Next generation
DNA sequencing (NGS) of exons 9 and 20 of PIK3CA was performed on biopsy specimens from cases using an Ion Torrent
Personal Genome Sequencer.
Results
Validation of the ddPCR assay with serial dilutions of cell line DNA demonstrated the assay is highly sensitive with mutant alleles
detected at an allele frequency down to 25% when input DNA template is as low as 2 ng. With DNA template input of 20 ng
(typical recovery of cfpDNA), mutations can be detected down to an allele frequency of 0.78%. We will present data on the rate of
hotspot mutation detection with the ddPCR assay performed on cfpDNA and FFPE samples in cases (n=80) and controls (n=80),
and we will report performance characteristics of the assay. Mutation status of tumors determined by NGS will serve as the gold
standard.
Conclusion
We have developed a ddPCR assay to detect PIK3CA hotspot mutations in cfpDNA, and will report the performance
characteristics of the assay in 80 cases of untreated, ER+ early breast cancer.

Title: A genetic deletion in the uridine diphosphate glucuronosyltransferase 2B17 affects the pharmacokinetics of exemestane
Harriet Johansson1, Valentina Aristarco1, Jennifer Gjerde2, Sara Gandini1, Aliana Guerrieri-Gonzaga1, Matteo Lazzeroni1, Serena
Mora1, Debora Macis1, Davide Serrano1, Antonio Toesca1, Luca Bottiglieri1, Gunnar Mellgren2, Giuseppe Viale1, Andrea DeCensi3
and Bernardo Bonanni1. 1Istituto Europeo di Oncologia, Milan, Italy; 2Haukeland University Hospital, Bergen, Norway and 3E.O.
Ospedali Galliera, Genoa, Italy.
Body: Background Exemestane (EXE) is an aromatase inactivator used in the prevention and treatment of breast cancer (BC).
The majority of EXE and its active metabolite 17-dihydroEXE are excreted as glucuronide conjugates by uridine diphosphate
glucuronosyltransferase (UGT). The UGT2B17 enzyme is the most expressed in human liver. A deletion spanning the entire
UGT2B17 gene (*2) decreased the EXE glucuronidation process by 14-fold in human liver microsomes (HLM) from
UGT2B17(*2/*2) genotype subjects compared to wild-type UGT2B17(*1/*1) HLMs.
Aim The aim of this study was to investigate whether the UGT2B17 deletion is associated with increased serum levels of EXE
and 17-dihydroEXE and to assess if this deletion predicts the anti-proliferative effect of EXE in BC tissue as measured by Ki67
changes 6 weeks apart.
Methods In a phase II pre-surgical trial, 50 postmenopausal women with histologically-confirmed ER positive BC; stage T1-2,
N0-1, M0 were assigned to EXE 25 mg/d for 6 weeks before surgery. Morning fasting blood was collected at baseline (B) and
after 6 weeks and stored at -80 C until assayed. Time of last drug intake was collected at blood draw. DNA was extracted from
whole blood (Qiagen, Italy). We used Taqman copy number variation assay (Life Technologies, Monza, Italy) for the UGT2B17
genotyping. EXE and 17-dihydroEXE concentrations were determined by mass spectrometry (MS) using Waters Xevo TQ MS
in electrospray positive ionization mode (Waters, Manchester, UK), optimized by multiple reaction monitoring mode. We used
Zorbax Eclipse Plus C18 columns and mobile phase MeOH/H2O with 0.1% formic acid. EXE pure substance was provided by
Pfizer Inc.,17-hydroxy EXE and EXE-19-d3 purchased from Toronto Research Chemicals (Toronto, Ontario, Canada).
Wilcoxon Signed Rank Test was used to assess differences in serum concentrations of EXE, its metabolites, and
post-pretreatment tissue Ki67 changes according to UGT2B17 genotypes.
Results Median age and BMI were 62 years and 26 kg/m2, respectively. The UGT2B17 genotype (n=50) was 24 homozygote wt
(*1/*1), 20 heterozygote (*1/*2) and 6 homozygote variant (*2/*2). Minor allele frequency = 0.32. Hardy-Weinberg Equilibrium
(P=0.57) was respected.
Median and interquartile ranges (IQRs) of EXE and 17-dihydroEXE at 6 weeks, by time elapsed since last drug intake
Time since last drug
EXE (nM)
17-dihydroEXE (nM)
24 hours (n=32)
10.25 (5.55, 17.25)
2.75 (1.85, 3.6)
>24 hours (n=15)
2.10 (1.20, 2.20)*
1.50 (1.20, 2.00)**
* P=.0001 vs 24h; **P=.0003 vs 24h

;
Median and IQRs of EXE, 17-dihydroEXE and Ki67 change from B by UGT2B17 genotypes
Genotype
EXE (nM)
17-dihydroEXE (nM)
Ki67 change
BMI
UGT2B17 *1/*1 (n=22)
5.3 (2.2, 11.4)
1.85 (1.4, 3.06)
-9 (-16, -5)
28 (24, 32)
UGT2B17 *1/*2, *2/*2 (n=25)
9.3 (2.6, 17.6)*
2.5 (2.0, 3.6)**
-11 (-19, -3)
26 (23, 30)
P-values between genotype groups:*P=.28; **P=.04; P=.82; P=.41

Conclusions Our study demonstrates a significant association of the UGT2B17 deletion with increased serum concentrations of
17-dihydroEXE, irrespective of time since last drug intake. Neither 17-dihydroEXE, nor genotype explained the significant
anti-proliferative effect of EXE on BC tissue. Larger studies should determine the clinical implications of this gene on EXE
efficacy.

Title: ALK and BRAF genes are associated with early failure of breast cancer (BC) who received primary neoadjuvant
chemotherapy for patients with locally advanced BCs
Yeon Hee Park1, Moon Ki Choi1, In-Gu Do1, Eun Yoon Choi1, Won Ho Kil1, Seok Won Kim1, Jeong Eon Lee1, Seok Jin Nam1, Jin
Seok Ahn1 and Young-Hyuck Im1. 1Samsung Medical Center.
Body: BACKGROUND: Neoadjuvant chemotherapy (NAC) has been used widely in patients with locally advanced breast cancer
(LABC). NAC has the added advantage of increasing breast conservation rates with similar disease-free and overall survival
compared with adjuvant chemotherapy. A subset of patients receiving NAC experiences early failure during the course of therapy
or within a short period after breast surgery. There are no established predictors for early therapy failure in LABC patients who
received NAC. This study was performed to identify candidate actionable mutation to explain early failure and refractoriness to
chemotherapy in BC patient groups that may not benefit from NAC.
METHODS: Seventy eight patients among 397 Patients with LABC (cT2-4N0-3) who were available for preoperative FFPE tumor
block for next generation sequencing (NGS) included in this analysis excluding 21 patients whose FFPE were not qualified for
Ampliseq. Early failure was defined as the development of an inoperable state caused by locoregional or systemic progression
during NAC or relapse after curative surgery within 1 year after the initiation of NAC. Patients who developed recurrence after 1
year from the start of NAC or exhibited no failure during the follow-up period were defined as control in this study. Thus, our
cohort was composed of pCR (pathologic CR to NAC), early failure, and control. The clinicopathological characteristics and
disease courses of the patients whose disease progressed within 1 year of receiving neoadjuvant chemotherapy were analyzed
to compare with the other patients. Using the Ion Torrent Ampliseq Cancer Panel v2 after DNA isolation from FFPE samples, we
sequenced 2,855 loci from 50 cancer-related genes to identify genetic mutations in 78 BC patients who received NAC for patients
with locally advanced BCs and available preoperative tumor tissue.
RESULTS: Thirty-eight of the 397 patients (9.6%) exhibited progression within 1 year after receiving neoadjuvant chemotherapy.
Among 78 patients who were available tumor tissue in this analysis, the number of patients with early failure, pCR, and control
was 22, 19, and 27, respectively. The sequencing analysis revealed TP53 mutations were observed in 95% or more patients,
evenly irrespective of patients group. Missense mutations in ALK and BRAF were found to be predominantly much higher in
patients with early failure than in other groups (p<0.01). To the contrary to this, missense mutations in JAK3 and MPL were only
found in patients with pCR than in other groups (p<0.01).
CONCLUSION: ALK and BRAF genes are associated with early failure in this analysis. Our results support that targeted
sequencing using cancer panel may function to identify actionable targets which are associated with responsiveness to NAC for
patients with LABC. Further research to figure out the functioning role of these genes for each group is warranted.

Title: Concordance/discordance rates of HER2, ER, PR, and Ki67 in matched pair samples of primary (PBC) and metastatic
breast cancer (MBC) tissues when comparing IHC with MammaTyper RT-PCR kit
Markus Wallwiener1, Andreas Hartkopf2, Thomas Deutsch1, Lakis Sotiris4, Florin-Andrei Taran2, Andreas Trumpp4, Ralph Wirtz3
and Andreas Schneeweiss4. 1University of Heidelberg, Heidelberg, BW, Germany; 2University of Tuebingen, Tuebingen, BW,
Germany; 3STRATIFYER Molecular Pathology GmbH, Cologne, Germany and 4National Center for Tumor Diseases, Heidelberg,
Germany.
Body: Background
Clinical decisions in the primary and advanced situation of breast cancer are based on immunohistochemical staining (IHC) and
semiquantitative assessment of ER, PgR, HER2 and Ki67. However, IHC carries an up to 20% risk of erroneous results.
Moreover, metaanalysis revealed variable discordance rates between primary and metastatic sites (P/M) which may have
implications for patient management (Aurilio, Eu J Cancer 2013). Herein, we undertook a pilot study aiming to investigate whether
the reported discordance between primary lesion and paired distant metastatic sites is due to technical limitations of the IHC
technique and could be improved by mRNA analysis using the MammaTyper in vitro diagnostic assay.
One 10m-thick sections from clinical routine FFPE tissues of 28 tumor samples were reexamined by RNA quantitation of ESR1,
PGR, HER2 and KI67 using the RNXtract kit for RNA extraction and MammaTyper kit for objective assessment of receptor
status. RNA levels were normalized using a synthetic in-vitro transcript for normalization according to the 40-DDCT method.
Receptor status was reported based on predefined cut-offs according to the instruction for use. mRNA and IHC results were
compared between paired primary and metastatic sites.
Results
Concordance per tumor sample between the two methods was 100% for HER2, 81.9% for ESR1 and 81.5% for PR. Among
discrepant samples (N=8), 6 were primary and among these, half (N=3) were due to negative ER by IHC. With the exception of
HER2, where no differences were observed for either method, the P/M concordance was superior by MammaTyper kit (ESR1:
92.85 vs. 76.82, PR: 71.43 vs. 69.23). Importantly, several discordant cases by IHC were ER negative in the primary site, while
being ER positive in the metastatic lesion, while being consistently ESR1 positive in both sites, when determined by the more
sensitive PCR method. The subtype discordant cases were situated in the lung (2), the liver (1) and the bone marrow (1).
Conclusion
Our data indicate that receptor status shifts less frequently when determined by PCR methods compared to IHC. While some
shifts between primary site and metastases may arise from true progression-related biologic alterations (Prat et al Nat Med 2009),
others seem to be related to technical limitations of semiquantitative and subjective IHC, with several cases being conspicuous
for ESR1. This finding is of great clinical significance and is consistent with previous reports implicating technical confounders of
IHC for large part of the observed P/M deviations (Pusztai, The Oncologist 2010). The implied superiority of RT-qPCR by
MammaTyper vs. routine IHC in detecting true P/M tumor receptor status reversals and the possible clinical impact is clearly
worth pursuing in larger datasets.

Title: No Show

Title: No Show

Title: Pharmacogenetic dosing of epirubicin in FEC chemotherapy
John R Mackey1, Edith Pituskin1, Ann Vlahadamis1, Katia Tonkin1, Karen King1, Sanraj Basi1, Maria Ho1, Judith Meza-Junco1, Anil
Joy1, Dick Au1, Sambasivarao Damaraju1 and Michael B Sawyer1. 1Cross Cancer Institute, Edmonton, AB, Canada.
Body: Background:
Epirubicin dosing affects important clinical outcomes in breast cancer, with higher dose regimens improving efficacy but
producing more myelosuppression. Epirubicin is metabolized by uridine glucuronosyltransferase 2B7 (UGT2B7). We previously
reported relationships between UGT2B7s promoter polymorphisms and epirubicin clearance and clinical outcomes in the
(neo)adjuvant breast cancer setting; we identified a trend for increased grade 3/4 neutropenia but better efficacy outcomes in
patients having at least one deficient allele (i.e. CT or CC) vs. patients who were wild type homozygotes. Patients homozygous
for the deficient allele (CC) were at statistically significant increased risk for leucopenia compared to patients who were wild type
homozygotes or heterozygotes. In this study we hypothesized patients with CT and TT genotypes would tolerate a higher
epirubicin dose compared to CC genotype patients.
We designed this study to determine the safety of pharmacogenetic-guided epirubicin dosing for each UGT2B7 genotype.
Methods:
Female breast cancer patients with histologically confirmed non-metastatic invasive breast cancer scheduled to receive at least
three cycles of FE100C in the (neo)adjuvant setting were enrolled into the study. Peripheral blood was analyzed for UGT2B7
genotype. Patients received standard dose IV FE100C during the first 21 day cycle. Based on genotype, epirubicin dosing was
escalated in the 2nd and 3rd cycles.
Epirubicin Dose Escalation Scheme
Dose of Epirubicin per Cycle (mg/m2)
Cycle
Genotype
CC
100
100
100
CT
100
115
130
TT
100
120
140
Results:
To date 32 patients are evaluable for pharmacogenetic guided epirubicin dosing (8 CC genotypes, 14 CT genotypes and 10 TT
genotypes). All 32 patients received epirubicin100 mg/m2 in cycle one and a single patient in each of the CC and CT genotypes
experienced grade 3 febrile neutropenia and were not dose escalated. All other patients with CT and TT genotypes were dose
escalated in cycle 2 and all but two patients in the CT and TT genotypes were dose escalated in cycle 3.
The incidence of febrile neutropenia was not dose dependent as all three genotypes had similar incidence in each cycle whereas
leucopenia was genotype and dose dependent. The incidence of leukopenia increased in patients with CT and TT genotypes as
their dose was increased and cycle 3 leukopenia rates were similar to patients with the CC genotype receiving standard dose
epirubicin.
Conclusions:
Pharmacogenetic guided epirubicin dosing is well tolerated. This study is ongoing and updated data will be presented.

Title: Prognostic impact of discordance in hormone receptor status after the neoadjuvant chemotherapy in primary breast cancer
Toshiaki Kurihara1, Hanako Ueno1, Masaru Takemae1, Aiko Nagayama1, Maiko Takahashi1, Tetsu Hayashida1, Hiromitsu Jinno1
and Yuko Kitagawa1. 1Keio University School of Medicine, Shinanomachi Shinjuku, Tokyo, Japan.
Body: Background: Hormone receptor (Estrogen receptor (ER), Progesterone receptor (PgR)) is an important biological marker
for predicting prognosis and making effective treatment decisions. Discordance in these biomarkers between the primary tumor
and recurrent lesions is reported frequently. However, it is not well known whether these biomarkers are affected by neoadjuvant
chemotherapy and their impacts on outcomes still remain to be elucidated. The aim of the present study is to evaluate the
changes in HR status after neoadjuvant chemotherapy in patients with operable breast cancer and their relationship with
response to treatment and prognosis.
Patients and Methods: Of 162 patients with stage II/III breast cancer patients receiving neoadjuvant chemotherapy from January
2005 to September 2012 at Keio University Hospital, 140 patients with non-pCR were analyzed. Patients were treated with
sequential anthracycline and taxane. ER and PgR were assessed in both CNB performed prior to neoadjuvant chemotherapy and
surgical samples. HR status was assessed by immunohistochemistry (IHC). ER/PgR status was determined using the Allred
score and defined as positive when score was 3 and more. HR status was considered positive in cases of ER and/or PgR
positivity. Pathological response criteria were classified as grade 0, 1, 2, or 3: grade 0 includes almost no change in cancer cells;
grade 1 includes slight or marked changes in less than two thirds of area; grade 2 includes marked changes in more than two
thirds of area; grade3 includes necrosis or disappearance of all tumor cells.
Results: ER, PgR and HR positive rates before neoadjuvant chemotherapy were 72.9%, 67.1% and 76.4%, respectively.
Changes in ER, PgR and HR status between CNB and surgical samples were 12.1% (4.3% gain; 7.8% loss), 17.1% (2.1% gain;
15.0% loss) and 9.3% (2.9% gain; 6.4% loss), respectively. In ER-discordant group, grade 2 rate of pathological response was
significantly higher than ER- concordant group (61.1% vs. 30%, p=0.033), whereas there were no significant differences of
pathological response between discordant and concordant group in PgR status. In the disease free survival (DFS), there were no
significant difference between concordance and discordance group for ER, PgR and HR (p=0.216, 0.859, 0.233) after a median
follow-up of 40.4 months. But patients with a loss in ER and/or HR status had a trend to a shorter DFS compared with the
concordant ER and/or HR-positive group (p=0.169 and 0.154)
Conclusions: After neoadjuvant chemotherapy, discordance of biomarkers was seen in 5-20%. The pathological response was
significantly associated with the change in ER status. A loss in ER and/or HR status may affect a prognosis of breast cancer after
neoadjuvant chemotherapy, but still need further investigation.

Title: Thyroid function is associated with the response to neoadjuvant chemotherapy in breast cancer patients: Results from the
NEOZOTAC trial on behalf of the Dutch Breast Cancer Research Group (BOOG 2010-01)
S de Groot1, A Charehbili1, L GM Janssen1, E M Dijkgraaf1, V THBM Smit1, L W Kessels2, A van Bochove3, H WM van
Laarhoven4, E Meershoek-Klein Kranenbarg1, A E van Leeuwen-Stok5, G J Liefers1, C JH van de Velde1, J WR Nortier1, J JM van
der Hoeven1, H Pijl1 and J R Kroep1. 1Leiden University Medical Center, Leiden, Netherlands; 2Deventer Hospital, Deventer,
Netherlands; 3Zaans Medical Center, Zaandam, Netherlands; 4Academic Medical Center, Amsterdam, Noord-Holland,
Netherlands and 5Dutch Breast Cancer Research Group (BOOG), Amsterdam, Noord-Holland, Netherlands.
Body: Background: Thyroid hormones, regulators of metabolism and development in healthy tissue, stimulate tumor growth in
vitro and are associated with breast cancer risk. We investigated the effect of chemotherapy on thyroid function and the extent to
which it can predict the pathological response in patients with HER2 negative stage II/III breast cancer taking part in the
NEOZOTAC phase III trial, randomizing between 6 cycles of neoadjuvant TAC chemotherapy with or without additional zoledronic
acid. Moreover, we examined the impact of thyroid function on chemotherapy toxicity.
Methods: Serum samples of 38 of the 105 patients who participated in the side study of the NEOZOTAC trial were available for
analyses. Serum free thyroxin (fT4) and thyroid stimulating hormone (TSH) levels were measured at baseline and compared with
fT4 and TSH levels before the 2nd and 6th chemotherapy cycle. FT4 and TSH levels were also compared between subjects with
and without pathological complete response (pCR). The relation between toxicity, per side effect of any CTC grade, and the
variation in fT4 and TSH levels during chemotherapy was tested.
Results: Serum samples at baseline, before the 2nd chemotherapy cycle and at end of treatment were available for 31, 30 and 21
patients, respectively. In the total population, the mean baseline fT4 level was 16,0pmol/L and the mean TSH level 1,11mU/L.
There were no differences between subjects solely treated with TAC chemotherapy and subjects treated with zoledronic acid as
an adjunct to TAC with respect to the mean fT4 and TSH at each time point. Baseline TSH levels tended to be higher in patients
who achieved pCR (p=0.035 univariate analysis and p=0.074 multivariate analysis) (Table1). During 6 cycles of chemotherapy,
fT4 levels decreased (p<0.000) and TSH levels increased significantly (p=0.019). Interestingly, the decrease of fT4 was
significantly greater in patients without nausea, vomiting or sensory neuropathy, than in patients with those side effects (p=0.037,
p=0.043 and p=0.050 respectively).
Characteristic
Univariate analysis
Multivariate analysis
OR
95%CI
P value
OR
95%CI
P value
N stage: N0 vs. N+
0.33
0.03-3.64
0.368
T stage: <5cm vs. >5cm
0.33
0.03-3.63
0.333
ER receptor: Pos vs. Neg
2.56
0.20-33.1
0.473
fT4
0.78
0.43-1.42
0.417
0.66
0.33-1.29
0.581
TSH
3.24
1.09-9.70
0.035
17.3
0.76-391
0.074
Table 1. Univariate and multivariate logistic regression models of baseline characteristics and TSH and fT4 predictive of pCR.
Conclusion: TSH levels at baseline were higher in breast cancer patients with pCR. Chemotherapy blunts thyroid function, and a
large decline of fT4 was associated with less side effects. These data suggest that thyroid hormones may interact with
chemotherapy to modulate treatment (side-) effects in patients with breast cancer.

Title: Atomic force microscopy of triple negative breast cancer cells: A predictive value of mechanical phenotype
Pawel A Osmulski1 and Maria Gaczynska1. 1University of Texas Health Science Center, San Antonio, TX.
Body: We applied Quantitative Nanomechanical Measurements with Atomic Force Microscopy (QNM-AFM) to test a response of
triple negative breast cancer (TNBC) cells to treatment with inhibitors of proteasome. Genetic screens identified TNBCs as
addicted to the activity of ubiquitin-proteasome pathway, the major intracellular venue of regulated protein degradation, served by
the essential protease, the proteasome. Here we tested the effects of treatment of two proteasome inhibitors on the canonical
triple-negative cell line, MDA-MB-231. The inhibitors represent two very distinct types of mechanism. Bortezomib (Velcade) is a
competitive inhibitor that targets active sites of the enzyme. On the other hand, B1 is a novel noncompetitive allosteric drug lead
that interferes with interactions between the subcomplexes of proteasome, the 20S catalytic core and the 19S regulatory particle
("cap"). In a search for test to predict the putative value of inhibitors we turned to a mechanical phenotype. Mechanical properties
of cells constitute a sensitive indicator of physiological status of cells. The best-known and most explored mechanical parameters
are elasticity ("softness") represented by the Young modulus, and surface adhesiveness ("stickiness"). It has been well
established that cancer cell are much softer and less sticky than the healthy counterparts. These changes in physical phenotype
of cancer cells are usually linked to remodeling of their cytoskeleton and altered expression of membrane proteins. Such
remodeling may be a very early indicator of the cells response to a drug, a less explored but potentially very useful feature.
Indeed, it has been shown that often treatment of cancer cells with anticancer drugs at least partially reverses their mechanical
phenotype resembling healthy cells. Here we found that 24-hrs exposure of the cells to 10 nM bortezomib increased their stiffness
and adhesiveness about two times. Even more striking, a treatment of the cells with 10 nM B1 induced almost threefold increase
of stiffness with twofold increase in adhesion. Strikingly, treatments with such concentrations of the inhibitors alone did not
significantly influence the viability of the cells, whereas their combination reduced the population of live cells to 50% or less of the
control values. The strong effects of the inhibitor treatment on the mechanical phenotype are in stark contrast with little or no
effects on cell viability. The results point at extraordinary sensitivity of the mechanical phenotype in detection of cell response to
anticancer drugs. We are exploring the potential predictive value of AFM-based cell surface studies in a search for effective drugs
or drug combinations.

Title: Circulating vitamin D concentrations and breast cancer risk: A pooled analysis of 17 cohorts
Kala Visvanathan1, Alison Mondul2, Anne Zeleniuch-Jacquotte3, Toqir K Mukhtar4, Stephanie A Smith-Warner4, Regina G Ziegler2
and On Behalf of Investigators in the Vitamin D Pooling Project of Breast and Colorectal Cancer. 1Johns Hopkins, Baltimore, MD;
2
National Cancer Institute, Washington, DC; 3New York University and 4Harvard.
Body: Optimizing vitamin D status through the use of supplementation and/or judicious sun exposure has been proposed as a
BC risk reduction strategy. This simple and non-invasive approach to BC prevention is extremely appealing given that vitamin D
concentrations in circulation are low in many individuals around the world. However, an Institute of Medicine (IOM) report found
the evidence on vitamin D and breast cancer to be inconsistent. To clarify the relationship between vitamin D and breast cancer,
data was pooled on approximately 25,000 women from 17 prospective cohorts worldwide. The association between
pre-diagnostic circulating 25(OH)D levels, the accepted measure of vitamin D status, and breast cancer incidence was examined.
For five cohorts, vitamin D status was assessed at a central laboratory (Heartland Assays, Inc.) using a direct, competitive
chemiluminescence immunoassay that measures 25(OH)D2 and 25(OH)D3 equivalently. In 12 cohorts with previously measured
25(OH)D levels, a stratified sample of 29 controls was re-assayed at Heartland Assays and used to calibrate existing levels to a
central assay using robust linear regression analyses. We standardized 25(OH) D levels for season using a periodic sine/cosine
function. Conditional logistic regression analyses were performed in each study and were then pooled to generate pooled odds
ratios by study-specific quantiles, consortium wide-quantiles, absolute cut points based on IOM guidance.
Our preliminary analyses included 10,353 cases of incident invasive breast cancer (5305 estrogen receptor (ER) positive cases
and 1311 (ER) negative cases and 12,313 matched controls. Median calibrated 25(OH)D levels in controls varied from 33 to 70
nmol/L across the cohorts. The consortium-wide median 25(OH)D among controls was 22% higher in summer as compared to
winter months. Across all studies, median age at blood draw was 41 to 70 years; and median elapsed time from blood draw to
diagnosis ranged from 2 to 13 years. The pooled odds ratio of breast cancer, comparing the highest to lowest study-specific
25(OH)D quintile, was 0.99 (95% confidence interval 0.90-1.08) after adjusting for body mass index, physical activity, menopausal
status, menopausal hormone therapy use, parity/age at first birth, and family history of breast cancer. Results were not
significantly different in analyses stratified by age of diagnosis (<50, 50-60, 60+ y).or by ER status. When calibrated circulating
25(OH)D levels were categorized based on the IOM definitions of "deficiency", "inadequacy", "adequacy", and "beyond
adequacy", risk was similar across the categories. Further analyses are ongoing to examine especially low and high 25(OH)D
concentrations, whether the vitamin D association varies according to tumor characteristics, the importance of elapsed time
between blood draw and diagnosis. These will be completed before the meeting.
In conclusion, preliminary results from the largest pooled analysis of prospective studies to date show no association between
25(OH)D levels and breast cancer risk and therefore suggest that increasing Vitamin D levels may not be an effective risk
reduction strategy for breast cancer.

Title: Circulating lipids and breast cancer risk Effects of ApoB
Talha Butt1, Olle Melander2, Peter Almgren3, Dov Shiffman4, Jonas Manjer5 and Signe Borgquist1. 1Lund University, Lund, Skne,
Sweden; 2Lund University, Clinical Research Center, Malm, Skne, Sweden; 3Lund University, Diabetes and Endocrinology,
Malm, Skne, Sweden; 4Celera-A Division of Quest Diagnostics, Alameda, CA; 5Lund University, Skne University Hospital,
Malm, Skne, Sweden and 6Lund University, Lund, Skne, Sweden.
Body: Background: Obesity in relation to breast cancer risk reveals different impact on women, warranting studies on circulating
lipids and risk for breast cancer. Apolipoprotein A (ApoA), and apolipoprotein B (ApoB) are recognized actors in cardiovascular
disease as the major protein components of High-Density-Lipoprotein (HDL) and Low-Density-Lipoprotein (LDL), respectively.
LDL and HDL particles transport cholesterol from the liver to peripheral tissues and back. Blood levels of circulating lipoproteins
and obesity are positively correlated. However the epidemiological evidence for the lipoprotein-cancer linkage has been mixed.
Aim: The aim of this study was to investigate pre-diagnostic levels of ApoA, ApoB, LDL and HDL in relation to breast cancer risk.
Methods and Material: The prospective cohort, Malm Diet and Cancer Study was initiated in 1991 and enrolled 17035 women.
These were followed-up until 31st of December 2011 and 1024 women were diagnosed with breast cancer. Blood samples were
collected at baseline. Lipid quartiles (ApoA, ApoB, HDL, LDL) were constructed based on the entire female cohort, excluding
women with a prevalent breast cancer. Associations between circulating lipids and patient characteristics; age at baseline, body
mass index (BMI) and socioeconomic index were described with ANOVA. A multivariate Cox Proportional Hazards analyses was
applied to study breast cancer risk in relation to levels of circulating lipids and a trend analysis was performed.
Results: Age at baseline was evenly distributed among quartiles of ApoA (mean 56.4 SD 8.0- 58.0 SD 7.6) and HDL (mean 57.7
SD 6.1 57.4 SD 5.8) For ApoB (mean 53.7 SD 7.2 - 60.3 SD 7.3 and LDL (mean 54.9 SD 5.8 59.5 SD 5.2), age increased
with levels of lipids. BMI was negatively associated with ApoA; women in the lowest quartile had the highest BMI (kg/m2) (mean
BMI in ApoA-quartiles 1-4 (26.5, 25.7, 25.0, 24.4). The same pattern was seen for quartiles of HDL whilst a reversed relationship
was seen for ApoB and LDL. Socioeconomic status was evenly distributed between all measured lipid quartiles. The multivariate
Cox analyses revealed a statistically significant risk reduction for women with increasing levels of ApoB. Compared with the first
quartile, the risk for incident breast cancer in quartiles 2-4 was 1.01 (95%CI, 0.85-1.20); 0.82 (95%CI, 0.68-0.99); 0.73 (95%CI,
0.60-0.89), P-trend<0.0003. A similar trend was seen for LDL, however only statistically significantly in the unadjusted analyses
(P-trend <0.016). Levels of ApoA and HDL were not significantly associated with breast cancer risk.
Conclusions: This study confirms previous studies indicating an inverse association between ApoB and breast cancer risk.
Studies addressing total cholesterol have similarly shown risk-reducing effects on breast cancer risk with increasing levels of
cholesterol. However, BMI does not seem to explain this association. Previous breast cancer studies on circulating lipids including
all separate components, ApoA, ApoB, HDL, and LDL are sparse, and the association is still insufficiently elucidated. It is possible
that ApoB and LDL as risk factors for cardiovascular events have diverse effects in breast cancer. The interaction between
circulating levels of lipids and breast cancer needs to be further investigated.

Title: Endocrine sensitivity is decisive for patient outcome in small node-negative breast cancers (BC) (pT1a,b) Results from
the Munich Cancer Registry
Thomas Kolben1, Nadia Harbeck1, Rachel Wuerstlein1, Gabriele Schubert-Fritschle2, Ingo Bauerfeind3, Simone Schrodi2 and Jutta
Engel2. 1Breast Center, Comprehensive Cancer Center of LMU, University Hospital Munich, Ludwig-Maximilians-University,
Munich, Bavaria, Germany; 2Munich Cancer Registry (MCR) of the Munich Cancer Center (MCC), Ludwig-Maximilians-University,
Munich, Bavaria, Germany and 3Klinikum Landshut, Germany, Head of the Project Group Breast Cancer of the Munich
TumorCenter (TZM), Landshut, Bavaria, Germany.
Body: PURPOSE In clinical routine, adjuvant systemic therapy (ADST) in small node-negative (N0) breast cancers is
controversial, in particular in HER2-positive disease. We aimed to define the patient subgroups with tumors <1 cm which would
benefit from ADST based on their risk of BC-recurrence and survival using population-based cancer registry data.
METHODS From 2002-2009 (median follow-up 6 years), 9,707 primary breast cancer patients with N0 tumors <2 cm (pTis,
pT1N0M0) were reported to the Munich Cancer Registry. Patients with pTis tumors (n= 1870) served as internal control. Time to
progression, observed (OS) and relative survival rates (Kaplan-Meier estimates) are presented. Cox regression analysis was
used to assess the influence of tumor size, age, HR-, and HER2-status.
RESULTS 10-year-OS in HR-positive tumors was 91.9% in pT1a (n=537), 90.6% in pT1b (n=1958), and 86.8% in pT1c (n=4513).
In HR-negative tumors, rates were 91.7% (n=78), 86.8% (n=134), and 86.8% (n=427), respectively.
In HER2-positive, it was 81.2% for pT1a (n=116), 88.1% for pT1b (n=171), and 86.7% for pT1c (n=427), in HER2-negative
tumors it was 93.1% (n=431), 90.6% (n=1751), and 86.0% (n=3942), respectively.
In the multivariate model, age, tumor size, and HR-status showed a significant impact on OS, while HER2-status was not an
independent prognostic factor.
(tumor size >1 cm vs. pTis: hazard ratio 1.94, p<0.0001 (95% CI; 1.47-2.56)
(HRneg. vs. HRpos.: hazard ratio 1.42, p=0.011 (95% CI; 1.09-1.87)
In this epidemiological registry, patients between 60 and 69 years - compared to patients aged 50 to 59 years - are at a 1.66 fold
higher risk of dying (p=0.0003; 95% CI 1.26-2.18), risk for patients aged 70-79 years was 3.99 fold (p<0.0001; 95% CI 3.03-5.27)
and for patients 80 years and older it was 15.38 fold higher (p= <0.0001; CI 11.45-20.66). However, hazard ratio for death at ages
< 50 years was 0.88; this was not statistically significant in the model (p=0.457; 95% CI 0.62-1.24). Even young age less than 40
years by itself did not turn out to be a significant risk factor.
CONCLUSION Prognosis of pN0 tumors <1 cm is excellent, especially if they are HR-positive, even in HER2-positive cases.
Weighing potential benefits vs. side-effects, there seems to be no need for chemotherapy in tumors <0.5 cm. In pT1b tumors,
chemotherapy may be considered, if tumors are triple negative or HER2-positive and HR-negative. In pT1c guideline-based
adjuvant therapy using all therapeutic options seems to be warranted.

Title: Effects of statin use on volumetric mammographic density: Results from the Karolinska mammography project for risk
prediction of breast cancer (KARMA) study
Ida Skarping1, Judith Brand2, Per Hall2 and Signe Borgquist1. 1Lund University, Lund, Sweden and 2Karolinska Institutet,
Stockholm, Sweden.
Body: Introduction:
High mammographic density is an established risk factor for breast cancer, whereas epidemiological data on statins and breast
cancer risk have been inconclusive. The aim of this study was to address the role of statins in breast cancer risk by studying their
effect on mammographic density in a large screening-based cohort.
Methods:
The KARolinska MAmmography project for the risk prediction of breast cancer (KARMA) study includes 70,876 women who
performed either a screening or clinical mammography from January 2011 to December 2013. In all, 41,102 women responded to
a web-based questionnaire, and their raw digital mammograms were stored and their volumetric mammographic density was
estimated using the Volpara system. Information on statin use was obtained through the Swedish National Prescription
Register. Analysis of covariance was used to study the effect of current statin use on mammographic density, adjusting for a large
set of potential confounders. Analyses were stratified by statin lipophilicity and exposure duration. The potential effect
modification by hormone replacement therapy (HRT) was analyzed.
Results:
Statin use was recorded in approximately 3,300 women (8.1%) of the study population of 41,102, the majority of which was
prescribed a lipophilic statin (93.4% of statin users). After multivariable adjustment, volumetric percent density was lower in statin
users than in non-users (P<0.001). Further, statin users had a larger non-dense volume than non-users (P<0.001), but no
difference in absolute dense volume was detected. No differential effects were observed according to lipophilicity of the statin or
drug duration. Interaction analyses revealed effect modification by HRT (P-interaction=0.03) with statin use being associated with
a larger dense volume among ever HRT users.
Participant characteristics
Total
Statin use
P value
No
Yes
N = 37,765
N = 3,337
Age (years)
55.0
54.2
63.8
< 0.001
BMI (kg/m2)
25.4
25.2
27.1
< 0.001
Menopausal status, % (N)
< 0.001
Premenopausal
40.2 (16,506)
43.1 (16,272)
7.0 (234)
Perimenopausal/unknown
3.3 (1,349)
3.5 (1,304)
1.4 (45)
Postmenopausal
56.6 (23,247)
53.5 (20,189)
91.6 (3,058)
Volumetric mammographic density measures by current statin use.

Model 1
Model 2
Model 3
Model 4
8.01
7.93
7.93
7.94
Statin use
Volumetric percent density (%)
No
Dense volume (cm3)
Yes
6.89
7.73
7.74
7.66
P value
< 0.001
0.001
0.001
< 0.001
No
56.8
57.1
57.1
57.0
Yes
60.7
57.6
57.5
58.0
P value
< 0.001
0.31
0.32
0.06
None
8.01
7.93
7.93
7.94
Lipophilic
6.92
7.73
7.77
7.69
Hydrophilic
6.46
7.32
7.34
7.26
P value
< 0.001
< 0.001
< 0.001
< 0.001
None
56.8
57.1
57.1
57.0
Lipophilic
60.8
57.7
57.7
58.1
Hydrophilic
58.2
55.0
54.8
55.3
P value
< 0.001
0.20
0.66
0.21
Statin type
Volumetric percent density (%)
Dense volume (cm3)
Model 1: adjusted for age Model 2: Model 1 + BMI Model 3: Model 2 + menopausal status, HRT use, parity, age at menarche,
education level, smoking, alcohol consumption and benign breast disease Model 4: Model 3 + low-dose aspirin and metformin
use
Conclusions:
Statin use was associated with a lower mammographic percent density, although no evidence was found for an effect of statins
on absolute dense volume. The observed interaction between statin and HRT use requires further investigation.

Title: Oncotype Dx use and its relationship with chemotherapy administration in the general population
Mariana Chavez-MacGregor1, Jiangong Niu1, Benjamin Smith1, Hui Zhao1, Thomas A Buchholz1 and Sharon H Giordano1.
1
University of Texas MD Anderson Cancer Center, Houston, TX.
Body: Background: Oncotype Dx is a 21-gene assay performed in paraffin-embedded tumor tissue. This test is used to assist in
estimating the likelihood of recurrence and the benefit from chemotherapy in patients with hormone receptor (HR)-positive,
node-negative breast cancer. One of he purported benefits of this assay is that by identifying low risk patients, Oncotype Dx use
will be associated with a decrease in chemotherapy use among those patients less likely to benefit from it. In this study we sought
to describe the utilization patterns of Oncotype Dx and its relationship with the use of adjuvant chemotherapy in a large
population-based study.
Methods: We identified 112,522 patients younger than 65 years old diagnosed with early stage breast cancer between 2004-2012
in the MarketScan database. Standard algorithms were used to identify our cohort and HR-positivity was defined according to
prescription information. A total of 34,245 patients older that 66 and diagnosed with early stage breast cancer between 2006-2009
were identified in the SEER-Medicare database. Descriptive statistics were used as well as logistic regression models to estimate
the impact of Oncotype Dx testing on adjuvant chemotherapy administration.
Results: In the cohort of young patients (<65 years old), 13.9% (n=15,643) underwent Oncotype Dx testing. Among those with
HR-positive disease, the percentage use was 71.8%. The use of Oncotype Dx increased according to age and year of diagnosis
(p<.001), going from 0.11% in 2005 to 21.9% in 2011. In the multivariable model, Oncotype Dx testing was associated with a
reduction in the use of adjuvant chemotherapy (OR 0.78; 95% CI 0.75-0.81).
Among the older patients (66 years old), 11.35% (n=3,017) underwent Oncotype Dx testing. Utilization was associated with
younger age, histological grade, educational level, marital status, geographic location and comorbidities (p<0.001). Oncotype Dx
use increased according to year of diagnosis going from 3.7% in 2006 to 14.11% in 2009 (p<0.001). Among the patients with
Oncotype Dx test, 85.5% had node-negative disease. In the multivariable model Oncotype Dx testing was not associated with a
statistically significant reduction in the use of chemotherapy (OR 0.95; 95% CI 0.86-1.06)
Conclusions: The use of Oncotype Dx in clinical practice is common, with an increase use in more recent years. Utilization was
determined by patient-related characteristics. Among younger patients Oncotype Dx testing was associated with a reduction in
the use of adjuvant chemotherapy, but this decrease was not observed among those 66 years old. It is possible that this
discordance is related to comorbidities and the fact that older patients are less likely to be considered candidates for
chemotherapy.

Title: Survival benefit increases after breast conserving therapy compared to mastectomy when axillary node status is positive in
early stage breast cancer: A registry based follow-up study of Norwegian women with screening and interval detected breast
cancer primary operated between 1998 and 2009
Olaf Johan Hartmann-Johnsen1, Ellen Schlichting1, Rolf Kaaresen1 and Jan Franz Nygaard2. 1Oslo Universitetssykehus HF,
Norway and 2Cancer Registry of Norway, Norway.
Body: BACKGROUND: Recent register studies on early stage breast cancer have shown better survival when women underwent
breast-conserving treatment (BCT) compared to mastectomy (MTX). It is unclear if this is due to selection effects or to a systemic
benefit of BCT compared to MTX. If there is a systemic benefit this could theoretically be more prominent in some subgroups of
breast cancer. The aim of this study is to evaluate if there are any subgroups with early stage breast cancer that benefits more of
BCT compared to MTX.
METHOD: A cohort of 6.629 women aged 50-69 with primary operable breast cancer detected in the screening programme (on
screening or between screening, interval cancer) were selected from January 1998 until December 2009 and followed up until
end of 2010. Life tables for overall survival (OS) and breast cancer specific survival (BCSS) were calculated in surgery groups
and stages. Corresponding Kaplan-Meier plots were stratified in T1N1M0 and Ductal Carcinoma. OS and BCSS were compared
using Cox proportional hazard for estimating hazard ratio between BCT and MTX in crude and adjusted analysis.
RESULTS: 5-years BCSS in T1N0M0 was 99% for both surgical cohorts (BCT and MTX). 5-years BCSS in T1N1M0 was 98% for
women undergoing BCT and 91% for women undergoing MTX. Respectively, 10-years BCSS in T1N1M0 was 96% for women
undergoing BCT and 83% for women undergoing MTX.
In adjusted analysis, HR for breast cancer death for women with stage T1N1M0 undergoing MTX compared to BCT was 3.12
(95% CI: 1.71-5.72). Respectively, 3.67 (95% CI: 1.20-11.22) stratified on grade III and 4.89 (67 (95% CI: 1.25-19.01) stratified on
hormone receptor negative disease.
CONCLUSION: The benefit of BCT compared to MTX seems to increase with increasing severity of the breast cancer disease,
positive nodal status (N1), grade III and hormone receptor negative disease. These finding indicate a systemic benefit of BCT
compared to MTX.

Title: Familial risk of breast density in extended Utah pedigrees
Karen Curtin1, Leigh Neumayer2, Matthew B Morgan3, Matthew A Stein3, Nicola J Camp1, Geraldine P Mineau1, Kerry G Rowe4
and Saundra S Buys1. 1University of Utah School of Medicine & Huntsman Cancer Institute, Salt Lake City, UT; 2University of
Utah Surgery Department & Huntsman Cancer Institute, Salt Lake City, UT; 3University of Utah, Salt Lake City, UT and
4
Intermountain Healthcare Clinical Oncology Program, Salt Lake City, UT.
Body: Background: Mammographic breast density (MBD) and family history are consistently associated with breast cancer risk,
and breast density may account for a proportion of susceptibility to this disease. MBD has been shown to correlate in small cohort
studies of twins and siblings. However, MBD has not been studied on a large scale in multi-generation families. We investigated
the familial relative risk of MBD in the Utah population, as clustering of breast density in extended relatives may provide evidence
for the role of genetics in breast density and inform screening recommendations.
Methods: Using the Utah Population Database (UPDB), an extensive genealogical database linked to medical records) we
identified 189,812 women ages 35-85 with pedigree information, who underwent digital mammography between 2005-2012, with
no history of breast/ovarian cancer and no indication of tamoxifen/aromatase inhibitor use. Individuals with unusually frequent
screening (>1/yr) or with inconsistent MBD assessments were not included. Subjects were categorized according to Breast
Imaging-Reporting and Data System (BI-RADS) composition classification at index mammogram available in radiology records
of Intermountain Healthcare and University of Utah Healthcare systems, representing the majority of mammograms in Utah, as:
(I) 0-25% fibroglandular densities (mostly fat); (II) 26-50% fibroglandular (scattered densities); (III) 51-75% fibroglandular
(heterogeneously dense); or, (IV) >75% fibroglandular densities (extremely dense). Familial recurrence risks of MBD classification
and breast cancer were estimated using Cox regression models in relatives (mothers, daughters, and sisters or 1st-degree;
aunts, nieces, grandmothers, and granddaughters or 2nd-degree; first- and second-cousins) of probands classified as BI-RADS I
(N=18,170) or BI-RADS IV (N=11,787), compared to those in the most common classifications, BI-RADS II (N=79,825) and III
(N=80,030) combined. Women in the comparison group were randomly selected and matched 5:1 to probands on birth year.
Results: Relatives of women with a history of extremely dense breasts (BI-RADS IV) were at increased relative risk (RR) of
extremely dense breasts compared to women in BI-RADS II/III: 1st-degree, RR=2.3 (95%CI 2.0-2.7, P<10-15); 2nd-degree,
RR=1.8 (95%CI 1.5-2.2, P<10-9); first cousins, RR=1.2 (95%CI 1.1-1.4, P<10-5); and second cousins, RR=1.1 (95%CI 1.06-1.2,
P<10-5). Conversely, relatives of women in BI-RADS I (mostly fat) were at decreased risk of BI-RADS IV: 1st-degree, RR=0.5
(95%CI 0.4-0.5, P<10-15); 2nd-degree, RR=0.7 (95%CI 0.6-0.8, P<10-6); first cousins, RR=0.85 (95%CI 0.8-0.9, P<10-4); and
second cousins, RR=0.9 (95%CI 0.9-<1.0, P<10-8). First-degree relatives of women in BI-RADS IV were at slightly greater risk of
breast cancer compared to women in BI-RADS II/III (RR=1.2, 95%CI 1.1-1.4; P=0.004), while women in category I showed no
increased cancer risk.
Conclusions: BI-RADS composition categories available from radiology records in the UPDB appear to be useful in
assessing familial risk of MBD at the population level. Our results may inform screening guidelines in more distant as
well as close relatives of women with a history of extremely dense breasts, whose families may be more susceptible to
breast cancer.

Title: Dont mix apples and oranges: Not all non-screen-detected breast cancers belong to the same category in a cancer registry
dataset
Elisabetta Rapiti1, Massimo Usel1, Robin Schaffar1, Hyma Schubert1 and Christine Bouchardy1. 1Geneva Cancer Registry,
Geneva, Switzerland.
Body: Background: In population-based datasets breast cancers are often classified as screen-detected or not screen-detected
(or symptoms-detected) without further information, resulting in substantial heterogeneity in the latter group. In fact, this group
includes symptoms-detected and fortuitously detected cancers,as well as interval breast cancers (i.e. detected between
screenings), all of which have different morphological characteristics than screen-detected cancers.
Objective: To compare the demographic features, tumour characteristics and survival of women with not-screen detected breast
cancer according to whether or not they reported previous mammography screening.
Methods: The study population comprised1696 women aged 50-69 years oldwith non-screen detected invasive breast cancer
recorded at the Geneva cancer registry between 1990 and 2007. According to the information on the variable "previous
mammography screening" the women were classified as those never having had a screening test and those having had one. We
comparedtumour characteristics and prognostic factors through chi square test, and calculated 5-year breast cancer specific
survival and multivariate Cox proportional hazard models to assess independent determinants of mortality.
Results:Among the non-screen detected breast cancers there were 904 womenwho never had a screening mammography and
792 who declared to have had at least one. Women previously screened were more often 55-59 years old, diagnosed after 1996,
of high social class, treated in the private sector and had a positive family history. These women had smaller tumours (mean 21.6
mm vs. 27.2, p<0.001), more often lobular cancers (18.4% vs 10.4%, p<0.001), less often poorly or undifferentiated cancers
(22.8% vs 34.2%, p<0.001), and less often stage III or IV cancers (13.4% vs. 24.3%, p<0.001). Previously screened women more
often received breast conserving surgery with radiotherapy (70.7% vs 49.1%, p<0.001), and endocrine therapy (73.0% vs 55.8%,
p<0.001).
Five year breast cancer-specific survival was 83% among never screened women (95% Confidence Intervals [95%CI] 81-86) and
91% among those previously screened (95% CI: 89-93; log rank test=20.62, p<0.001). In the Cox multivariate model adjusted for
all confounders and for prognosis and treatment variables, the difference in breast cancer mortality risk between women with and
without a previous mammography screening disappeared (Hazard Ratio for previously screened women 1.06, 95%CI: 0.73-1.54).
Conclusions:Within the category of non-screen detected breast cancers co-exist at least two distinct groups of patients, with
different tumours and, potentially, treatments and prognoses. Comparing women with screen-detected breast cancer against this
heterogeneous group of women could produce biased results. Recording information about screening use before the breast
cancer diagnosis would help to correctly classify these women.

Title: Trends and determinants of breast cancer survival among unscreened women: A population-based study
Elisabetta Rapiti1, Thomas Agoritsas1, Massimo Usel1, Robin Schaffar1, Hyma Schubert1 and Christine Bouchardy1. 1Geneva
Cancer Registry, Geneva, Switzerland.
Body: Background: Breast cancer mortality has been declining in many western countries, including Switzerland,since the late
1980s. This has largely been credited to mammography screening and improved treatment. However, mortality trends are also
decreasing among unscreened women, though most breast cancer deaths still occur in that population.
Objective: The objective of this study was to analyse trends in,and factors affecting the survival of, women whose breast cancer
was not detected by screening in the Geneva female population from 1990 to 2007. In Geneva an organized screening
programme started in 2001 while opportunistic screening has existed since the beginning of the 1990s.
Methods: The study population comprised1696 women aged 50-69 years oldwith invasive breast cancer that was not detected by
screening and that was recorded at the population-based Geneva cancer registry. We studied tumour characteristics and
prognostic factors across 6 time periods through chi square and trend tests. To assess whether breast cancer specific survival
had improved over time, we calculated 5-year specific survival and performed multivariate Cox proportional hazard models to
assess independent determinants of mortality.
Results:Median age of the women at diagnosis was 59 years. During the 18 year study periodthere was a decrease in the
proportion of diagnoses among women of low social class (25.8% in 1990-92 vs 17.3% in 2005-07, p=0.001). No change in the
distribution of stage at diagnosisor hormone receptor status was observed, while between the first and last period there was an
increase in cancers with lobular morphology (6.8% vs. 19.0%, p<0.001), and a decrease in poorly or undifferentiated cancers
(36.7% vs 28.7%, p<0.001). The use of breast conserving surgery with radiotherapy (BCS), chemotherapy and endocrine therapy
increased significantly along the period (41.2% to 58.8%, p<0.001; 41.6% to 52.8, p=0.030; and 34% to 69.9%, p<0.001,
respectively).
Five year breast cancer-specific survival was 82% in 1990-92 (95% Confidence Intervals [95%CI] 77-87) and 89% in 2005-07
(95% CI: 83-92; log rank test=8.68, p=0.122). In the Cox multivariate model there was a trend towards improved survival, but it
was only statistically significant when comparing the period 2005-07 to the first period (Hazard Ratio 0.47, 95%CI: 0.22-0.98).
Increasing age, stage, and grade, hormone receptornegative disease, and not receiving BCS or endocrine therapy were all
independently associated with a worse breast cancerspecific survival.
Conclusions:We observed an improvement in survival only in recent years among women whose breast cancers were not
detected by screening; this appears to be associated with improved treatment. This suggests that the breast cancer mortality
reduction observed in Switzerland since the late 1980s is not likely attributable to changes in treatment before 2005, but rather to
the generalization of screening.

Title: Rising incidence of primary breast lymphoma: SEER 1973-2011
Alexandra Thomas1, Mary C Schroeder1, Bradley D McDowell1 and Brian K Link1. 1University of Iowa, Iowa City, IA.
Body: Background: Primary breast lymphomas (PBL) are rare and most commonly of B-cell origin: diffuse large B-cell or
extranodal marginal zone lymphomas. Recent reports have described breast-implant associated anaplastic large T- cell
lymphoma. Other lymphomas arising from the breast have also been described less frequently, with plastic surgery. We
hypothesized that the overall incidence of primary breast lymphomas may also be increasing. This study reports the incidence of
PBL as first malignancy as recorded in Surveillance, Epidemiology and End Results Program (SEER) from 1973-2011. Mortality
and first local therapy are also reported.
Methods: Using the SEER 9 Research Data, we conducted a retrospective cohort analysis of women diagnosed with
microscopically confirmed, first primary malignant lymphoma arising in the breast from 1973 through 2011. Cases were excluded
if they were reported to the registry by a nursing home or diagnosed by autopsy or death certificate. PBLs were categorized by
the SEER lymphoid malignancy recode variable. We evaluated trends in age-adjusted incidence rates, first local therapy, and
age-adjusted all-cause and cause-specific mortality across time.
Results: The overall incidence of PBL has increased more than 3 fold, from 0.07/100,000 in 1973-82 to 0.23/100,000 in 2003-11
(p< 0.001) (Table 1).
Table 1: Primary Breast Lymphoma- Incidence (age-adjusted rate, per 100,000 women)
1973-1982
1983-1992
1993-2002
2003-2011
p value
0.1
0.1
0.2
0.2
<0.001
<50
0.0
0.0
0.1
0.05
<0.001
50+
0.2
0.4
0.6
0.7
<0.001
Hodgkins
0.0
0.002
0.003
0.002
<0.001
NHL-B-cell
0.04
0.1
0.2
0.2
<0.001
DLBCL
0.03
0.1
0.1
0.1
0.001
Follicular
0.005
0.02
0.03
0.04
0.062
Marginal Zone
0.0
0.0
0.03
0.1
<0.001
0.0
0.001
0.004
0.01
<0.001
All
Age
Lymphoma
NHL-T-cell
By age, the greatest incidence increase was for women 50 and older (0.02/100,000 in 1973-82 to 0.68/100,000 in 2003-11,
p<0.001). Both B-cell marginal zone lymphoma and T-cell lymphoma have had steep increases in incidence from 1983-1992 to
2003-2011 (0/100,000 to 0.06/100,000, p<0.001 and 0.001/100,000 to 0.008/100,000, p<0.001, respectively). The incidence of
DLBC and follicular lymphomas arising in the breast rose from 1973-1992 and subsequently stabilized. Age-adjusted, all-cause
mortality has dropped significantly since 1992, as has cause-specific mortality (Table 2).
Table 2: Primary Breast Lymphoma-Mortality and Initial Local Therapy
Age-adjusted, 5-year Mortality (% deceased)
1973-1982
1983-1992
1993-2002
2003-2011
p value
All-cause
51%
54%
33%
22%
<0.001
Cause-specific
43%
43%
25%
15%
<0.001
Initial Local therapy (% received)
Both
26%
26%
23%
12%
0.274
Radiation only
8%
11%
18%
31%
<0.001
Surgery only
59%
47%
34%
16%
<0.001
None
4%
14%
21%
40%
<0.001
For the entire cohort, no local therapy and radiation therapy, as initial local treatment, has increased over time. Surgery alone has
decreased substantially.
Conclusions: Primary breast lymphoma remains a rare diagnosis; however the incidence has risen markedly over the last four
decades, with B-cell marginal zone and T-cell lymphomas increasing significantly in the most recent time period. Further research
is required to better understand the factors contributing to these trends. Survival gains over time suggest either that lymphoma
treatments have improved or that the excess cases are treatment responsive or non-survival-threatening.

Title: Trends in clinicopathologic features and systemic therapy use in breast cancer patients: Findings from the National Cancer
Database
Anees B Chagpar1, Sarah Mougalian1, Donald Lannin1, Maysa Abu-Khalaf1, Brigid Killelea1, Tara Sanft1 and Lajos Pusztai1. 1Yale
University, New Haven, CT.
Body: Introduction: Breast cancer remains the most common malignancy affecting women in the United States. We sought to
identify temporal trends in the clinicopathologic features of breast cancer over the last 13 years, along with changes in systemic
therapy usage to treat this disease.
Methods: The National Cancer Database is a national resource maintained by the CoC and the American Cancer Society that
captures 70% of the newly diagnosed breast cancers in the US. We utilized the beta-PUF file of the NCDB, which included
1,669,679 patients with invasive breast cancer patients from 1998 to 2010, and analyzed trends in clinicopathologic features as
well as use of systemic chemo- and hormonal therapy.
Results: Over the 13 years of the study, a significant trend was noted to older age, lower grade and fewer positive lymph nodes
(all p<0.001). Yet, the proportion of patients receiving systemic chemotherapy and hormonal therapy also increased over this time
period (p<0.001). These trends remained when the time period was bisected into an early (prior to 2004) and later (after 2004)
time period (shown in the table below). Receipt of systemic therapy was positively correlated with younger age, larger tumor size,
higher grade and more positive lymph nodes over this time period (all p<0.001). There was a small increase in tumor size (<0.1
mm) in the latter period, but other factors were more favorable. Nonetheless, there has been a significantly higher utilization of
systemic chemotherapy (OR: 1.107; 95% CI: 1.017-1.098, p<0.001) and hormonal therapy (OR: 1.588; 95% CI: 1.578-1.598,
p<0.001) in more recent years, even when controlling for changes in clinicopathologic features on multivariate analysis.
Prior to 2004
After 2004
p-value
Mean age at diagnosis (years)
60.1
60.3
<0.001
Mean tumor size (mm)
20.1
20.2
<0.001
Mean number of positive nodes
1.3
1.2
<0.001
Proportion grade 3 (%)
36.8
34.6
<0.001
Proportion receiving chemotherapy (%)
41.1
42.2
<0.001
Proportion receiving hormonal therapy (%)
41.7
53.5
<0.001
Conclusions: These data demonstrate that while early detection may be increasing the number of low grade node-negative
tumors in older patients, our use of systemic therapy has become more liberal. Increasing use of hormonal therapy may be due to
more aggressive guidelines and a lower threshold for ER-positivity; increasing use of chemotherapy may be the result of more
efficacious drugs with lower toxicity being available to treat limited disease. Nonetheless, implications in terms of cost vs. benefit
of our more aggressive approach may warrant further evaluation.


Title: Outcomes of women with small, early-stage breast cancer in Manitoba from 1995-2011
Aly-Khan Lalani1, Katherine Fradette3, Rashid Ahmed3, Debjani Grenier2,3 and Marshall Pitz2,3. 1University of Manitoba, Winnipeg,
MB, Canada; 2University of Manitoba, Winnipeg, MB, Canada and 3CancerCare Manitoba, Winnipeg, MB, Canada.
Body: OBJECTIVES:
The objectives of this study were to describe the distribution, management, and outcomes of women with early-stage breast
cancer in Manitoba during the period from 1995-2011. Specifically, we looked at tumours 1cm or smaller in size (T1mic/T1a/T1b)
across the spectrum of molecular phenotypes: estrogen-receptor (ER) status, progesterone-receptor (PR) status, and human
epidermal growth factor receptor 2 (HER2) status. This is the first study to evaluate these objectives in Manitoba, and will
contribute significantly to the limited data of long-term outcomes for smaller-sized breast tumours.
METHODS:
Using the Manitoba Cancer Registry, we created a retrospective cohort of all patients with primary breast cancer of 1.0 cm or
less, between 1995 and 2011. Women with previous or synchronous contralateral breast cancers were excluded. Data included
patient demographics, diagnostic procedure, tumour size, nodal disease, treatment modalities (surgery, radiotherapy, hormone
therapy, and chemotherapy), ER status, PR status, and HER2 status. Linkage to the Manitoba Health Drug Program Information
Network (DPIN) database allowed capture of all hormonal therapies. Patient outcomes were evaluated, including risk of
recurrence, overall survival and relative survival. Risk of recurrence was illustrated using cumulative incidence of recurrence,
accounting for the competing risk of death. Kaplan-Meier curves were used to illustrate overall survival. The relative survival
estimate was used to estimate the probability of surviving from breast cancer while controlling for differences in mortality due to
other causes.
RESULTS:
Our study captured 2,341 women. Mean age at diagnosis was 62. ER+, PR+, and HER2+ status were 71% (11% unknown),
61.6% (11% unknown), and 8.6% (31% unknown), respectively. Clinically, 78% were Stage I disease, and by tumour size overall:
T1mic 11.5%, T1a 18.9%, and T1b 69.7%. Ninety-eight percent had primary surgery, 58% had primary radiation, 48% received
hormone therapy, and 16% received chemotherapy. At last follow-up, 21% of all patients were deceased. Relative survival
estimates revealed that the survival of the overall cohort was not different than the general population, when comparing based on
age groupings for Manitoban females.
Recurrence occurred in 6.6% (156) of all cases. ER+, PR+, and HER2+ status were 63% (4% unknown), 58% (4% unknown),
and 48% (5% unknown), respectively. By staging, 65% were Stage I disease and by tumour size 60% were T1b. Overall, 64% of
recurrences were node-negative. For the HER2+, node-negative subpopulation at diagnosis: 21% of the T1mic group recurred,
41% of T1a recurred, and 19% of T1b recurred. In all recurrences, 39% of patients were deceased at last follow-up.
CONCLUSIONS:
Small, early-stage breast cancers are common and a significant proportion of patients recur. HER2-positivity appears to be an
important risk factor for recurrence and may independently warrant treatment with trastuzumab, regardless of primary tumour
size.

Title: Increasing proportion of de novo compared with recurrent HER2-positive metastatic breast cancer: Early results from the
systemic therapies for HER2-positive metastatic breast cancer registry study
Debu Tripathy1, Adam Brufsky2, Melody Cobleigh3, Mohammad Jahanzeb4, Peter Kaufman5, Ginny Mason6, Musa Mayer7, Joyce
O Shaughnessy8, Hope Rugo9, Sandra M Swain10, Denise A Yardley11, Mary Beattie12, Bongin Yoo12 and Sara Hurvitz13. 1Keck
School of Medicine, USC/Norris Comprehensive Cancer; 2University of Pittsburgh Cancer Institute, Pittsburgh, PA; 3Rush
University Medical Center, Chicago, IL; 4University of Miami Sylvester Comprehensive Cancer Center Deerfield Campus; 5Norris
Cotton Cancer Center, Dartmouth-Hitchcock Medical Center; 6Inflammatory Breast Cancer Research Foundation, West Lafayette,
IN; 7AdvancedBC.org; 8Charles A. Sammons Cancer Center, Texas Oncology, The US Oncology Network; 9University of
California, Helen Diller Family Comprehensive Cancer Center, San Francisco, CA; 10Washington Cancer Institute, Medstar
Washington Hospital Center, Washington, DC; 11Sarah Cannon Research Institute and Tennessee Oncology, PLLC; 12Genentech,
San Francisco, CA and 13University of California, Jonsson Comprehensive Cancer Center, Los Angeles, CA.
Body: Introduction
The Systemic Therapies for HER2-Positive Metastatic Breast Cancer Registry (SystHERs) is a US-based prospective
observational cohort study currently enrolling patients with HER2-positive metastatic breast cancer (MBC). It began in 2012 and
aims to provide real-world insight into the disease course, treatment patterns and associated clinical outcomes, and
patient-reported experiences of disease. Here we describe the baseline characteristics of this population to date, including the
proportion of patients with de novo and recurrent HER2-positive MBC.
Methods
SystHERs aims to enroll 1000 patients with HER2-positive MBC within 6 months of first metastatic diagnosis. At enrollment,
investigators report whether patients have recurrent or de novo MBC, the latter defined as distant metastases at the time of initial
MBC diagnosis.
Results
As of February 17, 2014, data are available from 306 of 319 enrolled patients. Forty-six percent (142/306) had de novo
HER2-positive MBC. For the 54% (164/306) with recurrent HER2-positive MBC, median disease-free interval (DFI, defined here
as time from early-stage breast cancer diagnosis to diagnosis of MBC) was 43.6 (range 0.5270.2) months. For these patients
with recurrent disease, 54% (89/164) received (neo)adjuvant HER2-targeted therapy. Baseline patient and tumor characteristics
are shown below.
Characteristic
Overall (n=306)
De novo (n=142)
Recurrent (n=164)
Age, median (range)
57 (27-86)
54 (27-83)
59 (30-86)
White
233 (76)
105 (74)
128 (78)
Black
54 (18)
28 (20)
26 (16)
Other
8 (3)
3 (2)
5 (3)
Not available
11 (4)
6 (4)
5 (3)
Not Hispanic
257 (84)
120 (85)
137 (84)
Hispanic
31 (10)
17 (12)
14 (9)
Not available
18 (6)
5 (4)
13 (8)
Pre-menopausal, n (%)
61 (20)
42 (30)
19 (12)
209 (68)
90 (63)
119 (73)
Race, n (%)
Ethnicity, n(%)
Hormone receptor status, n (%)

ER and/or PR positive
ER and PR negative
92 (30)
50 (35)
42 (26)
Unknown
5 (2)
2 (1)
3 (2)
Median number of metastatic sites, (range)
2.0 (1-7)
2.0 (1-6)
2.0 (1-7)
Bone
141 (46)
76 (54)
65 (40)
Liver
120 (39)
63 (44)
57 (35)
Nodes
89 (29)
49 (35)
40 (24)
Lung
87 (28)
38 (27)
49 (30)
Chest wall
43 (14)
12 (9)
31 (19)
CNS
30 (10)
9 (6)
21 (13)
Sites of metastasis, n (%)
Conclusions
The proportion of patients with de novo MBC appears to have increased over time to 46% in SystHERs, compared with 33% in
RegistHER, a registry study that enrolled 1023 patients with HER2-positive MBC from 20032006 (Yardley et al, 2014), and
predated the broad use of HER2-targeted therapy in the (neo)adjuvant setting. The DFI for patients with recurrent HER2-positive
MBC also appears to be longer (median DFI 43.6 months in SystHERs vs. 32.6 months in registHER). We hypothesize that the
proportion of patients with de novo MBC within the metastatic population is higher due to the use of advanced screening
techniques which allow better detection of metastases, and due to a reduction in recurrences related to the availability of
HER2-targeted adjuvant therapy. Changes in the population characteristics of patients with HER2-positive MBC may impact
treatment strategies and have trial design implications. Updated data from approximately 500 patients are expected by the time of
presentation.

Title: Recent changes in breast cancer incidence and mortality trends in Mexico: A population-based study
Enrique Soto-Perez-de-Celis1, Alejandro Mohar2 and Yanin Chavarri-Guerra1. 1Instituto Nacional de Ciencias Mdicas y Nutricin
Salvador Zubirn, Mexico City, DF, Mexico and 2Instituto Nacional de Cancerologa, Mexico City, DF, Mexico.
Body: Background: Breast cancer (BC) is the most common malignancy in Mexican women since 2006, with a high number of
cases reported in the more developed northern states. However, the analysis of national BC incidence in Mexico has not been
previously reported. We sought to describe BC trends in Mexico using recent population-based data and to analyze geographical
differences in BC incidence and mortality rates.
Methods: This population-based retrospective cohort study included all incident BC cases registered in the National
Epidemiological Surveillance System and all BC deaths registered by the National Institute of Statistics and Geography in Mexico
from 2001 to 2011. Yearly populations were obtained from national census data. The age-standardized (AS) incidence rate of BC
was calculated using the female population over 15 years of age and standardized to the World Standard Population. AS
incidence rates were calculated for 3 geographic regions of the country (North, Center and South). Joinpoint regression analysis
was performed to examine trends in BC incidence and mortality. We estimated annual percentage change (APC) using weighted
least squares log-linear regression.
Results: From 2001 to 2011, 69,651 new cases of BC were registered in Mexico. The AS incidence of BC significantly increased,
rising from 14.2/100,000 person-years (PY) (Standard Error [SE] 0.23) in 2001, to 25.2/100,000 PY (SE 0.25) in 2011, with an
APC of 5.9% (95% CI 4.1-7.7, p<0.05). Regional AS incidence rates were significantly increased in the Center and in the South,
with a non-significant increase in the North (Table 1). For the same period, 48,817 deaths attributed to BC were registered. AS
mortality rate also had a significant increase, rising from 14/100,000 PY (SE 0.23) in 2001 to 14.6/100,000 PY (SE 0.2) in 2011,
with an APC of 0.4% (95% CI 0.1-0.7, p<0.05). Regional AS mortality rates were significantly increased in all three regions (Table
1).
Table 1. Regional BC incidence and mortality rates
Region
2001 AS rate per 100,000 PY (SE)
2011 AS rate per 100,000 PY (SE)
APC% (95% CI)
South
Incidence
6.1 (0.42)
16.57 (0.53)
10.5 (6-15.2)
<0.05
Mortality
7.8 (0.46)
9.56 (0.43)
2 (1-3.2)
<0.05
Center
Incidence
12.9 (0.27)
26.9 (0.33)
7.6 (5.7-9.6)
<0.05
Mortality
14 (0.29)
14.4 (0.25)
0.3 (0-0.6)
<0.05
North
Incidence
24.2 (0.64)
26.8 (0.54)
1 (-1.6-3.7)
NS
Mortality
17.6 (0.57)
18.8 (0.48)
0.7 (0-1.3)
<0.05
Conclusions: BC trends in Mexico show a continuous increase in incidence and mortality from 2001 to 2011, which could reflect
population growth, ageing, lifestyle modifications and changes in access to diagnosis and treatment. These changes could be an
expression of epidemiological transition in developing countries such as Mexico. A significant growth in both incidence and
mortality was found in both the Center and the South. In the North, incidence rates remained unchanged while mortality rates had
a significant rise, which was comparable to that of the other regions. One possible unexplored explanation for this observation
could be the recent wave of drug-related violence and high criminality rates in the north of the country, which may perhaps cause
underreporting of cases and disrupt availability of medical attention and access to healthcare in an otherwise developed and
wealthy region.

Title: Does annual mammogram screening incur lower healthcare costs for breast cancer women after diagnosis?
Su-Hsin Chang1, Lauren T Steward2, Bettina F Drake1, Sarah Lyons1, Susan Kraenzle3 and Melody S Goodman1. 1Washington
University School of Medicine, St Louis, MO; 2Washington University School of Medicine, St Louis, MO and 3Joanne Knight
Breast Health Center, The Alvin J. Siteman Cancer Center at Barnes-Jewish Hospital and Washington University School of
Medicine, St Louis, MO.
Body: Background: Breast cancer (BC) is the most commonly diagnosed cancer in women. The American Cancer Society (ACS)
guidelines recommend annual screening mammography for women of average risk beginning at age 40.
Objective: Based on the ACS recommendation, we sought to determine whether women aged 40 years and older who had annual
screening mammogram before BC diagnosis incurred lower healthcare costs after BC diagnosis compared to those who did not
follow the guidelines.
Methods: We used data from a large private health insurance claims database (Thomson Reuters Marketscan Research
Databases), 2006-2010. Our sample included women aged at least 40 years with a BC diagnosis between January 2009 and May
2010 and excluded women with cancer other than BC. The study period was chosen to allow for (1) the observation of their
screening behavior (3 years) before BC diagnosis and (2) the determination of their metastatic status. A woman was determined
to have metastasis upon diagnosis if within the period of 1 month before and 7 months after BC diagnosis, she had at least two
metastasis codes and chemotherapy (algorithm adapted from Schootman et al. 2009). The outcome variable was total healthcare
costs after BC diagnosis. Costs in MarketScan are defined as adjudicated reimbursement amounts from insurer claims data.
Women who had annual mammogram screening (ACS guidelines followers) were defined as women who had a screening
mammogram every calendar year from January 2006 until their BC diagnosis. Multivariable regression analyses of
log-transformed total costs were conducted, adjusting for age, months between BC diagnosis and December 2010, employee
classification (e.g., salary union), employment status, geographic location (e.g., northeast, south, etc.), metastatic status, and
whether they received wire localization, lumpectomy, partial mastectomy, radical mastectomy, breast reconstruction,
chemotherapy, and radiotherapy.
Results: 7,892 women with BC were identified. Among them, 27% had screening mammography every calendar year before their
BC diagnosis. In the multivariable analysis, women who had annual mammogram screening were associated with a 4.6% higher
cost (p=0.03). Women with metastatic diseases at diagnosis increased healthcare costs by 143.8% (p<.0001). Health care costs
increased by 0.9% each month after diagnosis (p<.0001). Wire localization incurred 40.4% higher costs (p<.0001). Lumpectomy
increased healthcare costs by 45% (p<.0001). Partial mastectomy was associated with 42.3% more costs (p<.0001). BC patients
undergoing radical mastectomy had 34.6% more costs (p<.0001). Breast reconstruction increased costs by 62.6%. Patients who
received chemotherapy had 189.8% higher healthcare costs compared to those who did not receive chemotherapy (p<.0001).
Radiotherapy was associated with 95.9% more costs (p<.0001).
Conclusion: This study provides evidence suggesting that even amongst an insured population, it appears that less than 30% of
breast cancer patients had annual mammogram screening before diagnosis and that these patients were associated with higher
healthcare costs in the early stage after their BC diagnosis. Future studies need to be conducted to further examine total
healthcare costs throughout breast cancer survivors life span.

Title: Rates of adherence and persistence to adjuvant endocrine therapy among women enrolled in Medicare Part D in a
four-state region of Appalachia
Gretchen G Kimmick1, Fabian Camacho2 and Roger T Anderson2. 1Duke University Medical Center, Durham, NC and
2
Pennsylvania State College of Medicine.
Body: Background: Disparities exist in breast cancer outcomes by age, geographic location, and socioeconomic status, but there
is little data regarding contributing factors to disparities within Appalachia. Underuse of adjuvant endocrine therapy for breast
cancer can contribute to disparities. We studied older women within four states of Appalachia and explored adherence and
persistence rates for adjuvant endocrine therapy.
Methods: The study group consisted of women with stage I-III breast cancer diagnosed 2008-2009 in North Carolina,
Pennsylvania, Kentucky, and Georgia, who were continuously enrolled in Part D Medicare and filled a prescription for tamoxifen,
anastrozole, letrozole, or exemestane within 1 year of diagnosis date. Adherence rate is defined by medication possession ratio
(MPR=sum of days supply for all claims during the calendar year after first prescription). Persistence rate was defined as absence
of a 90 day or greater gap in prescription coverage since the first prescription fill. Univariate analyses by Kruskall-Wallis
nonparametric test were performed. Variables included age, year of diagnosis, Charlson comorbidity score, Medicaid/Medicare
versus Medicare insurance status, rural versus urban residence, county-level economic status (Appalachia Regional Commission:
Distressed, At risk, Transitional, or Competitive), state, stage (1, 2A, 2B, or 3), and breast conserving surgery (BCS) versus
mastectomy.
Results: We identified 1,229 eligible cases. Mean age was 74.7 years (range 37-98); 97.5% were white; 46.7% had rural
residence, 9.0% lived in distressed counties, 28.2% were dual-Medicaid/Medicare insured, mean Charlson comorbidity score was
1.51, and tumor was stage 1 in 56.7%, 2 in 34.1%, and 3 in 9.2%. Mean calendar MPR was 80% (SD 0.30). Persistence rate at
6-months was 87% (SD 0.34) and at 12-months was 67% (SD 0.47).
MPR varied significantly between diagnosis years (p < .01),county-level economic status levels (p=0.03), geographical state (p <
.01), and Singh Index Area Deprivation tertiles (p=0.01). MPR was highest among those in the Competitive ARC class at median
levels of 86%, versus 76% in Distressed, 77% in At Risk, and 80% in Transitional. MPR was highest in PA (84%), compared to
OH (79%), NC(77%), and KY (73%). Greatest area deprivation tertile also coincided with lower mean and median MPR
adherence.
Persistence at 12 months varied significantly between diagnosis years (p < .01), geographical state (p = .01), and Singh Index
Area Deprivation tertiles (p=0.01). Persistence rates for PA (71%) were higher than OH (66%), NC (52%), and KY (51%).
Persistence rates for the highest area deprivation tertile were lower (62%) than rates in the lowest tertiles (71%, and 69%).
Conclusions: Among this group of women within four states in Appalachia with Medicare Part D who filled a prescription for
endocrine therapy, overall adherence and persistence rates were lower than expected. Higher adherence and persistence rates
were seen in areas with lower deprivation and economic distress.

Title: The NABON breast cancer audit; quality improvement in three years time
Annelotte C van Bommel1, Marie-Jeanne T Baas-Vrancken Peeters2, Margriet van der Heiden - van der Loo3, Thijs van Dalen4,
Emiel J Rutgers2, Michel W Wouters2, Marc B Lobbes5, Ruud M Pijnappel6, Marc A Mureau7, Pieter J Westenend8, Bart de Vries5,
Carolien H Smorenburg2, Agnes Jager7, John H Maduro9, Henk Struikmans10, Carol Richel11, Marga Schrieks11, Maike
Schepens12, Sabine Siesling13 and Vivianne C Tjan-Heijnen5. 1Leiden University Medical Center, Leiden, Netherlands;
2
Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 3Comprehensive
Cancer Centre the Netherlands (IKNL), Utrecht, Netherlands; 4Diakonessenhuis Utrecht, Utrecht, Netherlands; 5Maastricht
University Medical Centre, Maastricht, Netherlands; 6University Medical Centre, Utrecht, Netherlands; 7Erasmus MC Cancer
Institute, Rotterdam, Netherlands; 8Laboratory for Pathology, Dordrecht, Netherlands; 9University Medical Centre, Groningen,
Netherlands; 10Medical Centre Haaglanden, Hague, Netherlands; 11Dutch Breast Cancer Association, Utrecht, Netherlands;
12
Health Insurer Netherlands, Netherlands and 13Comprehensive Cancer Centre the Netherlands (IKNL), Utrecht, the Netherlands
/ University of Twente, Enschede, Netherlands.
Body: Background
The lifetime risk of developing breast cancer is 1 in 8 for women in the Netherlands. Breast cancer care in the Netherlands is of
high quality, resulting in low 5-year local recurrence rates of 1.5% after breast conserving therapy and 2.7% after mastectomy.
Although good endpoints of breast cancer treatment are obtained, it is expected that further improvement of care can be achieved
since unexplained variation in a number of treatment aspects was found between hospitals in the Netherlands. Clinical audits
provide an important tool for quality assessment.
Worldwide, only a few nationwide clinical audits for breast cancer care are currently running. In the Netherlands, a nationwide
multidisciplinary clinical audit started three years ago.
The multidisciplinary national NABON Breast Cancer Audit (NBCA) started collecting data of all Dutch hospitals in 2011,
facilitated by Comprehensive Cancer Centre the Netherlands (IKNL) and Dutch Institute for Clinical Auditing (DICA). The NBCA
has several purposes: nation-wide evaluation of quality parameters, evaluation of guideline adherence, and providing weekly
updated feedback to participating institutions.
Results
All Dutch hospitals (n=92) participate by providing data regarding delivered breast cancer care resulting in a database of more
than 42.000 breast cancer patients (5.745 DCIS and 36.396 invasive carcinomas) in three years time. Eighty-nine percent of
invasive breast cancer patients were treated with primary surgery of which 62% (n=19.885) with breast conserving surgery.
Within three years time, several quality assessments improved such as guideline compliance for pre- and postoperative
multidisciplinary team meetings, percentage of patients starting surgery within five weeks (see table). The percentage of patients
that were treated with preoperative systemic treatment (12%; 95% CI: 0 47%) and patients receiving an immediate
reconstruction after ablative surgery (19%; 95% CI 0 73%) still remained low with a large variation between hospitals. At the
conference, results will be substantiated by funnelplots. Other quality indicators will be presented as well.
Table 1. Quality indicator results and their improvement over years.
2011
2013
Preoperative MDT
81%
96%
Postoperative MDT
90%
98%
Time to operation 5 weeks (immediate reconstruction after mastectomy excluded)
80%
85%
Tumor positive margins invasive breast cancer (without PST)
6.1%
5.0%
Tumor positive margins DCIS
25%
20%
PST for invasive breast cancer
10%
12%
Immediate reconstruction after ablative surgery for invasive breast cancer
15%
19%
Immediate reconstruction after ablative surgery for DCIS
39%
44%
MDT= Multi-disciplinary team meeting; PST= Primary Systemic Treatment; DCIS= Ductal Carcinoma In Situ
Conclusions
The continuous cycle of registration and providing feedback by clinical auditing provides a powerful tool for quality monitoring and
improving breast cancer care. Improvements of monodisciplinairy surgical and pathological aspects of care have been reached in
a relatively short time period. However, for more complicated multidisciplinary issues like the use of primary systemic treatment
and immediate reconstruction, detailed analyses of the variation between hospitals is needed to further improve these aspects of
breast cancer care.

Title: Breast cancer risk perception and adherence to U.S. cancer prevention guidelines
Erin Hofstatter1 and Anees Chagpar1. 1Yale University School of Medicine, New Haven, CT.
Body: Introduction: In the United States, nearly 200,000 cancer-related deaths each year are attributed to poor diet and
physical inactivity. We sought to determine whether perceived personal breast cancer risk was associated with adherence to
healthy lifestyle habits in a nationally representative sample.
Methods: The National Health Interview Survey (NHIS) is a population-based survey conducted each year by the CDC, designed
to represent the non-institutionalized civilian population in the United States. We utilized data from the 2010 NHIS adult and
cancer supplements to compare fruit/vegetable intake, alcohol use and exercise habits among women who perceived themselves
to be at high risk of developing breast cancer versus those women who perceived themselves to be at average or low risk.
Results: In 2010, 12,055 women without a history of cancer were surveyed, representing 94,990,140 people in the population.
Perceived risk of breast cancer was not associated with improved adherence to nutrition and physical activity guidelines (see
Table). Less than 5% of all U.S. women meet the recommended fruit/vegetable intake regardless of perceived risk, and only
one-third of all women achieve recommended physical activity. Approximately 95% of women adhere to guidelines regarding
alcohol intake. On multivariate analysis, perceived risk did not affect health habits when controlling for age, ethnicity, education,
insurance status, income, region and body mass index (BMI).
Conclusion: Thousands of breast cancer diagnoses and deaths could be prevented each year through healthy diet and exercise.
Yet, a striking majority of U.S. women do not adhere to cancer prevention recommendations, including those women who
perceive themselves to be at high risk of breast cancer. These findings suggest that patient education and awareness of personal
breast cancer risk may be essential, but not sufficient, for motivating meaningful behavior change.
Univariate Analysis
Perceived High Risk-%
meeting guidelines
Fruit/Veg Intake (>=5 svg
daily)
Perceived Low/Avg Risk-%

meeting guidelines
Multivariate Analysis
p-value OR (95% CI) p-value
4.42%
0.5679
1.14
(0.78-1.66)
0.5110
Alcohol Intake (<=7 svg/wk) 94.22%
95.13%
0.3093
0.84
(0.61-1.17)
0.3076
Physical Activity (75-150 min

30.11%
vig-mod act/wk)
27.74%
0.1465
1.02
(0.87-1.21)
0.7921
4.90%

Title: Survival comparative analysis of patients with invasive breast cancer treated by a military medical center and matched
patients of the US general population
Yuanbin Ru1, Jianfang Liu1, Jamie Leigh Campbell2, Kangmin Zhu2, Albert J Kovatich2, Jeffrey A Hooke2, Leonid Kvecher1,
Brenda Deyarmin1, Audrey W Kovatich3, Frank Cammarata1, Hallgeir Rui4, Richard J Mural1, Craig D Shriver2 and Hai Hu1.
1
Windber Research Institute, Windber, PA; 2Walter Reed National Military Medical Center, Bethesda, MD; 3MDR Global, LLC,
Windber, PA and 4Thomas Jefferson University, Philadelphia, PA.
Body: BACKGROUND
U.S. military beneficiaries differ from the U.S. general population with regards to access to health care as care is provided at no or
much lower cost in the military health system. Other differences also exist. Many of these differences are known factors affecting
invasive breast cancer outcomes. Thus it is desirable to conduct a comparative analysis of breast cancer patient outcomes
between these two populations to find out whether there is any outcome difference, and if yes what the contributing factors are.
METHODS
We compared overall survival (OS), disease-specific survival (DSS), and 5-year OS and DSS rates in breast cancers between
399 patients from the Clinical Breast Care Project at the Walter Reed National Military Medical Center (CBCP-WR) and 1,000
sets of 1596 matched patients from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer
Institute. All patients were diagnosed between 2001 and 2010. Each CBCP-WR patient was randomly matched to four SEER
patients on six demographic and clinicopathologic variables (age at diagnosis in 5-year groups, race, diagnosis year, estrogen
receptor (ER), progesterone receptor, and AJCC stage).
RESULTS
The CBCP-WR cohort had better survival than the SEER population. At the whole cohort level, the mean hazard ratios (HRs)
from 1,000 matched comparisons for OS and DSS were 0.774 and 0.708, with mean log-rank P-values of 0.124 and 0.125. The
numbers of 175 and 141 comparisons showing a log-rank P-value <0.05 out of the 1,000 tests were significantly higher than what
would be expected from a random distribution of these P-values (P<0.00001, exact binomial test). By stratifying the cohorts we
identified that this survival disparity was mainly contributed by patients with a diagnosis age 50 years (for DSS, mean HR=0.550,
mean P=0.049, and 642 of 1,000 tests showed a P<0.05; for OS, mean HR=0.713, mean P=0.081, and 377 of 1,000 tests
showed a P<0.05), but not by patients with a diagnosis age <50 years. The absolute differences in 5-year DSS rates were 4.4%
(94.6% in CBCP-WR vs. 90.2% in SEER; mean P=0.010) for all matched patients and 4.8% (95.2% vs. 90.4%; mean P=0.015)
for patients diagnosed at an age 50 years. Again there was no significant difference for patients diagnosed at an age <50 years.
When stratified by race, ER, stage or grade, most of the patient subpopulations showed favorable 5-year OS and DSS rates in
the CBCP-WR cohorts.
CONCLUSION
Overall, these results suggested that breast cancer patients, especially older patients seen in the CBCP-WR, carried more
favorable outcomes than those from the general population. The findings warrant further analyses of the contributing factors, such
as health care access, treatments, population characteristics, additional pathologic characteristics, and socioeconomic statuses,
to this outcome disparity.
The views expressed in this article are those of the author and do not reflect the official policy of the Department of Defense, or
U.S. Government.

Title: Breast cancer subtype and survival in a population-based cohort of patients from California
Christina A Clark1, Scarlett Lin Gomez1, Li Tao1, Lisa Moy1, Juan Yang1, Lisa Wang2, Mary S Beattie2, Jennifer Eng-Wong2,
Melissa Brammer2 and Laura Chu2. 1Cancer Prevention Institute of California, Fremont, CA and 2Genentech, San Francisco, CA.
Body: Background: Breast Cancer (BC) is a heterogeneous disease comprising distinct subtypes defined at present by tumor
molecular markers. Over the past decades, more effective treatments targeted to specific markers have been developed and
tested in clinical trials. While clinical trials have demonstrated the impact of these targeted treatments on recurrence and mortality
rates, little is known about survival patterns associated with specific subtypes in the general population. We examined overall
survival outcomes according to tumor subtype and stage in a diverse, population-based cohort of BC patients (pts) in California.
Methods: Through the California (CA) Cancer Registry, we identified all female CA residents diagnosed with primary invasive BC
between 1/1/2005 and 12/31/2011. We classified these cancers as early breast cancer (EBC, stages I-III) vs. de novo metastatic
(MBC, stage IV). We further grouped these cancers into 4 subtypes based on HER2 and hormone receptor (HR) status. For a
subset of women with available survival data, we calculated the proportion surviving at 3 years (yrs), and median overall survival
using the Kaplan-Meier method.
Results: 118,817 EBC pts (61.4% HR+/HER2-, 9.9% HR+/HER2+, 13.3% unclassified, 4.9% HR-/HER2+, and 10.6%
triple-negative (TN)) and 6,268 MBC pts (43.7% HR+/HER2-, 13.0% HR+/HER2+, 23.3% unclassified, 8.9% HR-/HER2+, and
10.7% TN) were identified. Table 1 presents survival time in EBC vs. MBC, and for each of the receptor subtypes of BC. For EBC,
3-yr survival rate was highest (95.1%) for HR+/HER2- pts and shortest (84.3%) for TN. For the HER2- and HER2+ overall groups
(regardless of HR status), 3-yr survival was similar (93.3% vs. 92.5%, respectively). The longest survival for de novo MBC was
observed for the HR+/HER2+ subtype (median OS: 45.3 months (mos)), compared to 38.7 mos for the HR+/HER2- subtype, 23.1
mos for the HR-/HER2+ subtype and 12.7 mos for the TN subtype. For the overall HER2+ and HER2- subtypes, HER2+ MBC
had slightly better survival (3-yr rate: 47.6% and median OS: 33.6 mos) than HER2- MBC pts (3-yr rate: 44.8% and median OS:
30.9 mos).
Conclusions: This study demonstrates the relevance of subtype on the OS of BC pts in a large population of CA women. Although
HER2+ status is a negative prognostic factor, survival was similar between HER2+ and HER2- pts, likely due to available
treatments targeting HER2. TNBC pts had the shortest survival, especially when they presented with metastatic disease. There
remains an urgent unmet need for more effective treatments for TNBC.
Table 1. Survival by stage (EBC vs de novo MBC) and BC subtype
EBC
Subtype
de novo MBC
3-yr Survival (95% CI)
3-yr Survival (95% CI)
Median OS (95% CI) (months)
HR+/HER2-
38118
95.1 (94.8-95.3)
1351
52.5 (49.8-55.1)
38.7 (35.9-41.3)
HR+/HER2+
6920
94.6 (94.1-95.2)
450
55.1 (50.4-59.6)
45.3 (36.5-50.5)
Unclassified
10813
91.2 (90.6-91.7)
894
26.0 (23.1-28.9)
11.9 (9.7-13.8)
HR-/HER2+
3589
88.4 (87.3-89.4)
333
37.5 (32.3-42.7)
23.1 (19.2-27.4)
TN
7083
84.3 (83.4-85.1)
357
17.7 (13.9-21.9)
12.7 (11.2-14.1)
HER2-neg*
45550
93.3 (93.1-93.6)
1745
44.8 (42.4-47.1)
30.9 (28.6-33.6)
HER2-pos*
10635
92.5 (92.0-93.0)
797
47.6 (44.0-51.0)
33.6 (30.7-37.1)
* Pts with available HER2 data, but missing HR data were included

Title: Predictors of neoadjuvant chemotherapy use in women with breast cancer: A review of 169,329 patients from the American
College of Surgeons' National Cancer Database
Lynn J Howie1,2, Rachel Greenup1, Kevin Houck1, Julie A Sosa1,2, E Shelley Hwang1 and Jeffrey M Peppercorn1. 1Duke Cancer
Institute, Durham, NC and 2Duke Clinical Research Institute, Durham, NC.
Body: Background: Randomized controlled trials have demonstrated that neoadjuvant chemotherapy (NAC) offers equivalent
long-term outcomes when compared to adjuvant chemotherapy while improving rates of breast conservation therapy (BCT). We
sought to evaluate the national use of NAC and the patient, tumor, and provider characteristics associated with its use among
women with stage I-III breast cancer. We hypothesize that younger women with larger tumors and HER-2+ or triple negative
disease are more likely to receive NAC than their older counterparts with ER+ disease, and that use of NAC varies by region and
practice setting.
Methods: Using the American College of Surgeons National Cancer Database, which captures data on approximately 70% of new
cancer diagnoses in the U.S., we identified women with diagnosed with stage I-III invasive cancer between 2008-2013. Women
treated with both surgery and chemotherapy were included in the study. Demographic and clinical factors including race, ethnicity,
income, insurance type, region of treatment facility, treatment facility type, tumor size, hormone receptor status, HER-2 status and
Charlson Comorbidity score were analyzed to determine predictors associated with receipt of NAC. Utilization of preoperative
chemotherapy and rates of breast conserving therapy (BCT) were evaluated as outcomes.
Results: 169,329 women with stage I-III breast cancer underwent treatment with chemotherapy and surgery during the study
period. Of these, 81.0% were White, 14.4% were Black, 4.6% were classified as Other Race. 28.7% were 18-49, 46.8% were
50-64 and 24.5% were >65 years. 64.4% had private insurance while 35.6% had public insurance (Medicaid, Medicare and VA),
and 71.4% were treated at Community-Based Clinics while 28.6% were treated at Academic Medical Centers.
Patients with larger tumors (p<0.0001) and triple negative disease were significantly more likely to be treated with NAC than
those with ER+ or HER2+ disease (p<0.0001). Among women who received NAC, the median age was 54 as compared to 57 in
those receiving adjuvant chemotherapy (p<0.0001). Treatment facility type impacted rates of NAC use, with academic centers
being more likely than community-based practices to give chemotherapy preoperatively (12.1% vs. 9.8%, p<0.0001) and urban
vs. rural settings (10.4% vs 8.2%, p<0.0001). Rates of NAC utilization differed regionally with the highest rate being 14.0% and
the lowest rate 7.9% (p<0.0001).
The overall % of BCT following NAC was 36.1% compared to 59.0% for those receiving adjuvant chemotherapy. The proportion
of BCT following NAC differed significantly by subtype, 54.6% for ER+, 54.5% for Her2 +, and 59.7% for triple negative breast
cancer (TNBC) (p <0.0001).
Conclusions: In the treatment of stage I-III breast cancer across the US, variations in the utilization of neoadjuvant chemotherapy
exist across the country suggesting clinical uncertainty about its use. Further research about the use of NAC therapy and the
relationship to clinical outcomes can identify patient subsets who might obtain greatest clinical benefit from preoperative systemic
therapy.

Title: Patient characteristics, clinical and economic outcomes of women with first-line therapy for HR+/HER2- metastatic breast
cancer in a large US managed care health plan: Chemotherapy first vs. no chemotherapy first
Tanya Burton1, Stacey DaCosta Byfield1, Ying Fan1, Yiyu Fang1, Feng Cao1, Gregory L Smith2, Giovanni Zanotti2, Timothy J Bell2,
Julia J Perkins2, Ruslan Horblyuk2 and April Teitlebaum3. 1Optum, Eden Prairie, MN; 2Pfizer, New York, NY and 3Private Practice,
San Diego, CA.
Body: BACKGROUND: NCCN guidelines clearly identify when chemotherapy may be an appropriate therapeutic approach for
metastatic breast cancer (mBC). This study compared patient characteristics, mortality rates, and health care costs by initial
chemotherapy (CT) use among women with HR+/HER2- mBC.
METHODS: A retrospective cohort study design was used to analyze administrative claims data linked to clinical information for
commercial health plan enrollees with evidence of mBC between 1/2008 and 4/2013. Clinical status at diagnosis was obtained
from physician reports, including date of diagnosis, hormone receptor (HR) and human epidermal growth factor receptor 2
(HER2) status. Women with known HR+/HER2- subtype and diagnosed initially with Stage IV or Stages I-III with evidence of
progression later to metastatic disease were evaluated for at least 6 months after their Stage IV diagnosis or first metastatic
claim, or until death if sooner. Clinical characteristics were compared descriptively between women who initiated therapy
with/without chemotherapy (CT 1st vs. no CT 1st, respectively) using t-test for continuous and chi-square test for categorical
variables. Mortality was compared using the incidence rate ratio (IRR) and 95% confidence interval (CI) from a negative binomial
distribution. Total average per-member-per-month (PMPM) health care costs were compared using a generalized linear model.
Both the mortality rates and costs were adjusted for age, geographic region, stage at diagnosis, initial metastatic site(s), and initial
use of CT.
RESULTS: Of 349 women with HR+/HER2- mBC, 204 (58%) had CT 1st and 145 (42%) had no CT 1st. Median follow-up was 17
months, and median length of the first-line of therapy (1st LOT) was 4 months (including censored LOTs). The mean age of
women with CT 1st was slightly lower than those without CT 1st (52 vs. 55 years, p<0.01), although the proportion 50 years old
did not differ between cohorts (35% vs. 28%, p=0.14). Cohorts did not differ by geographic region or initial metastatic site(s) to
brain, liver, or lung. Compared to women without CT 1st, a lower proportion of women with CT 1st were diagnosed de novo Stage
IV (34% vs. 48%, p=0.01). Among women with CT 1st, 24 (12%) also received hormonal therapy (HT) during their 1st LOT. All
women with no CT 1st (100%) initiated treatment with HT, of which the most common were 52 (36%) with tamoxifen, 41 (28%)
with anastrozole, and 39 (27%) with letrozole. After adjustment for baseline characteristics, no cohort differences were found in
mortality (IRR: 0.93, 95% CI: 0.50-1.72), however adjusted total average PMPM costs were significantly higher in women with CT
1st than those without ($11,666 vs. $6,639, p<0.001).
CONCLUSION: In this study of commercially insured women with HR+/HER2- mBC, use of CT 1st (>50%) was higher than
expected. While there were minor cohort differences in patient characteristics, CT 1st does not appear to be associated with a
survival benefit, but was associated with significantly higher costs when compared to no CT 1st. Additional research is needed to
determine subset(s) of mBC women with CT 1st likely to benefit from initial HT.

Title: Clinical characteristics and treatment utilization by tumor subtype among metastatic breast cancer patients in a large US
managed care health plan
Tanya Burton1, Stacey DaCosta Byfield1, Ying Fan1, Yiyu Fang1, Feng Cao1, Gregory L Smith2, Giovanni Zanotti2, Timothy J Bell2,
Julia J Perkins2, Ruslan Horblyuk2 and April Teitlebaum3. 1Optum, Eden Prairie, MN; 2Pfizer, New York, NY and 3Private Practice,
San Diego, CA.
Body: BACKGROUND: Breast cancer is the second most common cause of cancer death among US women. Molecular profiling
of breast cancer tumors is important for assessing prognosis and optimizing treatment. This study compared clinical
characteristics and treatment utilization by tumor subtype among women with metastatic breast cancer (mBC).
METHODS: A retrospective cohort study design was used to analyze administrative claims data linked to clinical information for
commercial health plan enrollees with evidence of mBC between 1/2008 and 4/2013. Clinical status at diagnosis was obtained
from physician reports, including date of diagnosis, hormone receptor (HR) and human epidermal growth factor receptor 2
(HER2) status. Women with known HR/HER2 subtypes and diagnosed initially with Stage IV or Stages I-III with evidence of
progression later to metastatic disease were evaluated for at least 6 months after their Stage IV diagnosis or first metastatic
claim, or until death if sooner. Lines of therapy (LOTs) were identified based on the timing of claims for NCCN-recommended
chemotherapy (CT), hormonal therapy (HT), and biologics (BIO). Clinical characteristics and treatments were compared
descriptively across the HR/HER2 subtypes using t-test for continuous and chi-square test for categorical variables.
RESULTS: Table 1 presents study results by tumor subtype. There were 657 women identified (72% HR+ and 28% HR-). Median
age was 53 years, and median follow-up was 16 months. Overall, 93% initiated therapy; of which, 48% started a 2nd LOT, 4%
died, 28% discontinued or had a 90-day gap in therapy, and 20% had their 1st LOT censored due to disenrollment or end of
study (EOS) period. More than half received CT during the 1st LOT, regardless of tumor subtype. Less than half of HR+ women
initiated HT, and among HER2+ patients, most initiated a BIO agent.
Diagnosed at Stage IV*
Total
(N=657)
HR+/HER2(N=365)
HR-/HER2(N=118)
HR+/HER2+
(N=111)
HR-/HER2+
(N=63)
n(%)
n(%)
n(%)
n(%)
n(%)
265(40)
145(40)
37(31)
50(45)
33(52)
- Bone
283(43)
167(46)
49(42)
46(41)
21(33)
- Brain*
105(16)
45(12)
32(27)
17(15)
11(17)
- Liver*
171(26)
76(21)
40(34)
35(32)
20(32)
- Lung*
127(19)
64(18)
36(31)
11(10)
16(25)
- Other (incl. distant lymph nodes)
494(75)
277(76)
97(82)
76(68)
44(70)
Initial NCCN-recommended therapy

(1st LOT)
614(93)
349(96)
98(83)
109(98)
58(92)
- CT*
422(69)
204(58)
92(94)
78(72)
48(83)
- BIO*
124(20)
1(<1)
4(4)
68(62)
51(88)
- HT*
210(34)
169(48)
9(9)
31(28)
1(2)
293(48)
159(46)
45(46)
71(65)
18(31)
26(4)
10(3)
8(8)
5(5)
3(5)
-- Discontinuation/gap*
174(28)
105(30)
36(37)
15(14)
18(31)
-- Censored*
121(20)
75(21)
9(9)
18(17)
19(33)
Initial site of metastasis
- Reason for 1st LOT end

-- Initiated a 2nd LOT*
-- Died
*p<0.05. Not all groups are mutually exclusive. Percentages may not total 100% due to rounding or overlap.
CONCLUSION: In this study of commercially insured mBC women, initial therapy choices were generally consistent with NCCN
guidelines for three of the four subtypes, while the largest subtype of HR+/HER2- women had a higher than expected utilization of
CT. Future analyses of study data will investigate appropriateness of these initial treatments and their impact on clinical and
economic outcomes.

Title: 2 year survival analysis of triple negative breast cancer from SEER data
Moira Rushton1, Tinghua Zhang2 and Xinni Song3. 1University of Ottawa; 2Ottawa Hospital Research Institute, Ottawa, ON,
Canada and 3Ottawa Hospital Cancer Program, Ottawa, ON, Canada.
Body: Background
Triple negative breast cancer (TNBC) is a heterogeneous disease characterized by the lack of receptor expression (ER, PR and
Her 2/neu negative). Amongst breast cancer types TNBC has a less favourable prognosis. There is a higher incidence of TNBC in
African-American women than Caucasian women. What has not been clearly elucidated is whether survival outcomes are
different among women with TNBC from different ethnic background.
Objective
The objective of our study was to use population data to determine if significant differences exist in overall survival (OS) of TNBC
patients across various ethnicities, including but not limited towhite, black, Hispanic and Asian.
Methods
Retrospective cohort study of patients with TNBC from 1973-2011 Surveillance, Epidemiology, and End-Results (SEER)
database to examine differences in OS across ethnicities. For each case data was collected on age, race, disease stage,
treatment, insurance status, time to death and cause of death. Descriptive statistics and survival analysis was carried out on the
data. Multivariate analysis was carried out to take into account age, stage, treatments received.
Results
12894 cases of TNBC across all ethnicities were reported in the SEER database. At two years follow-up, 720 patients (5.7%) had
died of breast cancer. 9696 (78.77%) had early stage (stage 0 II) disease, 1885 (15.31%) had locally advanced/stage III
disease while 728 (5.9%) had stage IV disease. 12071 (95.53%) were insured, 11533 (91.27%) had surgery, and 5454 (43.17%)
had radiation therapy. 7746 (61.53%) patients were white, 2429 (19.29%) black, 1548 (12.30%) were Hispanic and 490 (3.89%)
were Asian. In multivariate analysis, increasing age, stage III or IV disease, lack of insurance, surgery or radiation all had
significant hazard ratios. There was no significant survival difference found between any ethnicity compared with white patients
when controlled for age, stage, insurance, surgery and radiation.
Discussion
After two-year follow-up of large cohort of TNBC patients no significant difference could be found between any ethnicity and the
white population with this disease. While there is a large population of black and Hispanic patients in this study there are small
numbers of other races. The small relatively small event rate could be masking potential differences given the majority if patients
were early stage and are still alive. Longer follow-up is needed before conclusions can be made about differences between ethnic
groups. if certain populations do worse will inform the medical oncology community of an area to focus greater research into how
to optimize therapies for that patient group.

Title: Characteristics of de novo metastatic breast cancer in California, 2005-2011
Christina A Clarke1, Laura Chu2, Li Tao1, Lisa Wang2, Lisa Moy1, Melissa Brammer2, Chunyan Song2, Marjorie Green2 and Scarlett
Lin Gomez1. 1Cancer Prevention Institute of California, Fremont, CA and 2Genentech, San Francisco, CA.
Body: BACKGROUND: Breast cancer (BC) that is metastatic at initial diagnosis (i.e. de novo metastatic or stage IV) has not
been well described, especially in the general population.
OBJECTIVE: To describe demographics, tumor characteristics and survival in a population-based cohort of patients with de novo
metastatic BC (MBC).
METHODS: We studied all 6268 de novo MBC cases diagnosed in California women between 1/1/2005 and 12/31/2011, as
reported to the California Cancer Registry. Molecular subtypes were classified according to HER2 and hormone receptor (HR,
based on estrogen and progesterone receptor) status. Median overall survival (OS) was calculated by Kaplan-Meier methods.
Cox proportional hazards regression was used to assess independent predictors of OS.
RESULTS: 5% of all newly diagnosed BC were metastatic, representing 6% of all newly diagnosed HR+/HER2+, 8% of all
HR-/HER2+, 4% of all HR+/HER2- and 6% of all triple negative BC (TNBC). Compared to patients with early BC, MBC patients
were of similar age (mean age at diagnosis, (interquartile range): 61,(51-71) vs. 60, (50-70) years)). They were slightly more likely
to be black (10% vs. 6%) or Hispanic (19% vs. 17%) but substantially more likely to be unmarried (56% vs. 40%), to live in
neighborhoods of the lowest socioeconomic quintiles (39% vs. 29%), and to have public (e.g., Medicaid) or no insurance (39% vs.
21%). Most MBC patients presented with large tumors; however, 13% of patients had tumor sizes 2 cm or less, compared with
60% of patients with early BC (of TNBC: 15% MBC vs. 44% early BC were 2 cm). A minority of patients with de novo MBC
received breast surgery (39%), with 24% receiving full or partial mastectomy, 9% lumpectomy, 3% bilateral mastectomy and 3%
other/unknown surgery . 64% of de novo MBC patients received chemotherapy and 33% received radiation.
Median survival after MBC diagnosis was 27 months (mos), but varied substantially by patient characteristics including age (<40:
40 mos, 85+: 8 mos), race/ethnicity (Asian: 34 mos, black: 16 mos), and neighborhood socioeconomic quintile (lowest: 20 mos,
highest 34 mos) and molecular subtype (HR+/HER2+: 45 mos, TNBC: 12 mos). In a multivariate Cox model including all available
variables, TNBC was the most important predictor of death (Hazard Ratio 2.8, 95% CI: 2.4-3.3 vs. HR+/HER2+). Other significant
and important predictors included HR-/HER2+ subtype (Hazard Ratio 1.6, 95% CI: 1.3-18 vs. HR+/HER2+), being unmarried,
living in low socioeconomic status neighborhoods, and high tumor grade status.
CONCLUSIONS: In this large diverse population, de novo MBC was more likely to be diagnosed for certain breast cancer
subtypes, and among minority and underserved women (black or Hispanic race, low socioeconomic neighborhood, no or public
health insurance) that may have contributed to the detection of their tumor only after it had metastasized. A high proportion of
patients with MBC are not treated surgically, but most receive chemotherapy. Median survival remains poor, with worse survival
strongly associated with tumor biology (triple negative and HR-/HER2+ molecular subtypes) and patient characteristics indicative
of low socioeconomic status.

Title: Patterns of chemotherapy (CT) use by molecularly-defined subtype among a population-based cohort of breast cancer (BC)
patients (pts) from California
Christina A Clarke1, Laura Chu2, Lisa Wang2, Chunyan Song2, Marjorie Green2, Lisa Moy1, Juan Yang1, Li Tao1 and Scarlett Lin
Gomez1. 1Cancer Prevention Institute of California, Fremont, CA and 2Genentech, San Francisco, CA.
Body: Background: Guidelines describe recommended BC treatments (tx) according to tumor subtype; however,
population-based data regarding tx utilization are limited.
Objective: To describe tx utilization by tumor HER2, hormone receptor (HR) status, and stage among representative pts from
California.
Methods: We examined patterns of CT (including biologic tx) reported as received for all 131,450 female BC pts diagnosed with
primary invasive BC in California between 1/1/2005 and 12/31/2011 in the California Cancer Registry, including n=19,918 HER2+
[65% HR+, 33% HR-]; n=91,919 HER2- [84% HR+, 15% HR-]; and n=19,613 HER2 unknown. Neoadjuvant & adjuvant CT data
was only available for a subset of 35,040 pts diagnosed 2010-2011 (4,609 HER+; 27, 604 HER2-). Descriptive statistics were
used for comparisons.
Results: CT was reported as highest among pts with metastatic breast cancer (MBC) and HER2+ status (Table 1). The proportion
of pts with early BC (EBC) receiving CT increased from 58% in 2005 to 69% in 2010, but was generally steady for MBC pts
(74.2% in 2005 to 79.6% in 2011). 20.7% of HER2+ pts received neoadjuvant CT vs 8.6% for HER2- pts. Similarly, adjuvant CT
was more common in HER2+ (55.5%) than HER2- (28.5%) pts. As expected the proportion of pts receiving CT was lower in HR+
vs HR- pts for EBC and MBC [eg. all stages HER2+ (62% vs 71%); HER2- (33% vs 68%)].
Conclusion: In this population-based cohort, 20-40% of MBC pts may not have received any CT or biologics. CT was more
commonly received by HER2+ patients compared to HER2- patients for both EBC and MBC.
Table 1. Patterns of chemotherapy (including biologics) use among California female BC pts by stage and HER2 status,
2005-2011
EBC
MBC
HER2+
HER2-
HER2+
HER2-
No
5898 (33.4%)
52467 (61.0%)
317 (22.6%)
1519 (43.7%)
Yes
11463 (64.8%)
32403 (37.7%)
1063 (75.9%)
1894 (54.5%)
322 (1.8%)
1128 (1.3%)
20 (1.4%)
60 (1.7%)
Any chemotherapy
Unknown
Any neoadjuvant therapy (2010-2011 data only)

No
3546 (76.9%)
24794 (89.8%)
Yes
953 (20.7%)
2381 (8.6%)
Unknown
110 (2.4%)
429 (1.6%)
Any adjuvant therapy (2010-2011 data only)

No
1943 (42.2%)
19296 (69.9%)
Yes
2556 (55.5%)
7879 (28.5%)
110 (2.4%)
429 (1.6%)
Unknown
* not applicable

Title: Population-based rates of in-situ breast cancer [DCIS] invasive breast cancer [BrINV], and breast cancer [BrCa] mortality
[BrMOR] over time. A case for screening mammography? [ScreenMam]
Joseph Ragaz1, Hubert Wong1, Hong Qian1, Joel Fox2, Kenneth Wilson2 and Andrew Coldman2. 1School of Population and Public
Health, University of British Columbia, Vancouver, BC, Canada and 2British Columbia Cancer Agency, University of BC,
Vancouver & Victoria, BC, Canada.
Body: INTRODUCTION:
The net benefit of ScreenMam has been questioned in recent literature. We report here the 1975-2010 population-based rates of
DCIS, BrINV, and of BrMOR in two regions of Canada, British Columbia [BC] vs the Atlantic Provinces of Canada [Atl.P],with
unequal adherence (high in BC, low in Atl.P) to ScreenMam and therapy [Th*] guidelines for BrCa [Ref. 1].
METHODS: Annual age-specific rates of DCIS, BrINV, and BrMOR were obtained for 17 age groups (ages 0-4 to 85+) in BC and
Atl.P, and averaged over each 5-year periods [1975-1979 up to 2005-2009]. Four birth cohorts were defined within each region:
women who in the 1975-79 were aged 30-34 [COHORT 1]; 35-39 [COHORT 2]; 40-44 [COHORT 3] and 45-49 [COHORT 4].
The rates [cases /100,000 person-years] of DCIS, BrINV and BrMOR in 5 year intervals between 1975 and 2009 were followed
within each birth cohort and compared between BC and Atl.P in periods 1975-1984 and 2000-2009 [see Table]. Also,
age-standardized rates [ASRs] for BrINV incidence and BrMOR were estimated for the entire population, and for ages 50-65 and
65+. The data were obtained from the Public Health Agency of Canada based on the Canadian Cancer Registry database at
Statistics Canada.
RESULTS
Table 1. Cases / 100,000 person-years, years 1975-84 vs 2000-09. BrCa Incidence [DCIS, BrINV] and mortality in BC vs Atl. P.
within four birth-cohorts.
BC vs. Atl.P [BC Atl.P]
COHORTS*: 1
1975-84
2 vs 1 [+1]**
4 vs 4 [0]
8 vs 5 [+3]
13 vs 3 [+10]
2000-09
50 vs 44 [+6]
52 vs 46 [+6]
52 vs 41 [+11]
50 vs 33 [+17]
1975-84
42 vs 37 [+5]
80 vs 66 [+14]
134 vs 106 [+24]
181 vs 149 [+32]
2000-09
271 vs 273 [-2]
304 vs 327 [-23]
325 vs 344 [-19]
341 vs 355 [-14]
1975-84
7 vs 9 [-2]
13 vs 14 [-1]
28 vs 29 [-1]
44 vs 46 [-2]
2000-09
51 vs 54 [-3]
65 vs 68 [-3]
72 vs 89 [-17]
90 vs 110 [-20]
IN-SITU
INVASIVE
MORTALITY
**[x] = Difference Cases / 100,000 person-years, BC vs Atl.P: + or

In BC, the ASR for BrINV incidence increased by 20% between 1985 and 1989, and decreased by 20% between 1990-2007. In
the Atl.P, the BrINV rates increased by 40% between 1980 and1998 [i.e.increase to higher extent / over longer time period] and
declined by 20% between 1999-2007.
Compared to 1975, corresponding ASRs for mortality show a 47% decline in BC vs 30% for Atl.P [for ages 50-65: 52% vs 46%;
for ages 65+: 37% vs 5%, respectively.].
CONCLUSIONS:
1. DCIS: Higher rates in BC than in Atl.P in the older cohorts [3 and 4] during 1975-84 likely reflect earlier clinical use of
ScreenMam in BC. Substantially higher rates maintained during 2000-09 likely reflect persistent greater usage in BC.
2. BrINV. In every birth cohort, substantially higher rates in BC than in Atl.P during 1975-1984 but substantially lower rates in
2000-2009 are likely due to detection & Th earlier in life in BC.

3. BrMOR. Rates were similar between the two regions within each cohort during 1975-84 but substantially lower in BC for older
cohorts [3 and 4] during 2000-09, likely due to earlier detection and Th in BC, concordant with more substantial ASR reduction of
BrMOR observed in BC.
SUMMARY. In view of more ScreenMam in BC, these data suggest that the higher DCIS, but lower rates of BrINV, and more
consistent mortality decline in BC than in Atl.P, could all be related - a likely reflection of earlier detection and earlier therapy. The
associations of these interactions will be discussed.
*Th : surgery, radiation, Tamoxifen for DCIS; the same + chemotherapy for early BrINV.
Ref.1: Cancer Research: December 15, 2009.

Title: Association between adjuvant chemotherapy and risk of acute kidney injury in elderly women diagnosed with early stage
breast cancer
Shuling Li1, Jiannong Liu1, Beth A Virnig2 and Allan J Collins1,2. 1Chronic Disease Research Group, Minneapolis, MN and
2
University of Minnesota, Minneapolis, MN.
Body: Background: Acute kidney injury (AKI) is a serious complication of cancer and its treatment, causing delay or interruptions
in cancer therapy and increased risk of adverse outcomes including premature death. Little is known from large population-based
cohort studies about the association between chemotherapy (chemo) and risk of AKI in elderly cancer patients (pts).
Methods: This retrospective cohort study used the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare data.
Women diagnosed with stages I-III breast cancer (BC) at ages 66-89 years between 1992-2007 were included. We performed 1-1
sequential matching on time-dependent propensity score on the day of adjuvant chemo initiation within 6 months (mos) after the
first surgery. Follow-up (F/U) began on the matching date and ended at AKI occurrence, chronic kidney disease diagnosis, death,
change in enrollment status, or 6 mos after the matching date. For pts in the matched untreated cohort, F/U time was also
censored at chemo initiation. Chemo was identified in claims through billing codes indicating drugs or administration. Regimens of
interest included anthracyclines (A), CMF, taxanes (T, no anthracyclines), and others. AKI was identified in hospital claims. The
cumulative incidence of AKI was assessed using the Kaplan-Meier method. The association between adjuvant chemo and risk of
AKI was evaluated using a Cox proportional hazards model. The analyses were repeated by regimen type.
Results: The matched study cohorts included 28,048 pts. The mean (standard deviation) F/U time was 6.0 (0.8) mos for the
chemo cohort and 4.3 (2.5) mos for the matched no-chemo cohort. The cumulative incidence of AKI at mos 1, 3, and 6 was
0.24%, 0.50%, and 0.80% for pts receiving chemo, compared with 0.05%, 0.17%, and 0.30% for no-chemo pts (P<0.001).
Adjuvant chemo was associated with a 2.7-fold increased risk of AKI (HR 2.7, 95% CI 1.8-4.1; P<0.001), despite a very low
overall incidence rate (16 and 6 per 1000 person-years in chemo and no-chemo pts, respectively). Further examination of
distribution of other diseases coded on hospital claims in AKI pts showed that septicemia occurred in 40% of chemo-treated pts
with AKI and in only 17% of untreated pts with AKI. Of chemo-treated pts, 53%, 31%, 7%, and 8% received an A-based, a CMF, a
T-based, and other regimens, respectively. Each regimen was significantly associated with increased risk of AKI, with the
strongest association for a T-based regimen (HR 4.2, 95% CI 2.2-7.8), the weakest for a CMF regimen (HR 2.2, 95% CI, 1.3-3.8),
and intermediate associations for an A-based regimen (HR 2.5, 95% CI 1.6-4.1) and others (HR 3.0, 95% CI 1.5-6.2), but the
effect of these regimens did not significantly differ from each other.
Conclusion: Adjuvant chemo is associated with increased risk of AKI in elderly women diagnosed with early stage BC. This
association may be partially explained by septicemia caused by infection/neutropenia due to use of myelosuppressive
chemotherapeutic agents, highlighting the importance of preventing serious complications of chemo in preventing AKI.

Title: Association between adjuvant chemotherapy and risk of chronic kidney disease in elderly women diagnosed with early
stage breast cancer
Shuling Li1, Jiannong Liu1, Beth A Virnig2 and Allan J Collins1,2. 1Chronic Disease Research Group, Minneapolis, MN and
2
University of Minnesota, Minneapolis, MN.
Body: Background: Chronic kidney disease (CKD) and cancer are major public health problems in the elderly population. With
the development of cancer screening and efficacious treatments including chemotherapy (chemo), the number of cancer
survivors has been increasing. In elderly cancer patients (pts), little is known about CKD as a late effect of chemo. This study
examined the association between adjuvant chemo and risk of CKD in elderly women diagnosed with early stage breast cancer
(BC).
Methods: This retrospective cohort study used the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare data.
Women diagnosed with stages I-III breast cancer (BC) at ages 66-89 years between 1992-2007 were included. We preformed 1-1
sequential matching on time-dependent propensity score on the day of adjuvant chemo initiation within 6 months after the first
surgery. Follow-up (F/U) began on the matching date and ended at CKD diagnosis, death, change in enrollment status, or
December 31, 2009. For pts in the matched untreated cohort, F/U time was also censored at chemo initiation. Chemo was
identified in claims through billing codes indicating drugs or administration. Regimens of interest included anthracyclines (A),
CMF, taxanes (T, no anthracyclines), and others. CKD was identified through diagnosis codes in Medicare claims. The
cumulative incidence of CKD was assessed using the Kaplan-Meier method. The association between adjuvant chemo and risk of
CKD was evaluated using a Cox proportional hazards model. The analyses were repeated by regimen type.
Results: The matched study cohorts included 28,048 pts. The mean (standard deviation) F/U time was 5.1 (3.4) years for the
chemo cohort and 3.3 (3.6) years for the matched no-chemo cohort. CKD rate (standard error) was 29.0 (0.6) and 29.3 (0.8) per
1000 patient-years in chemo and no-chemo pts, respectively. Overall, there was no significant difference in the cumulative
incidence of CKD between the two cohorts (chemo vs. no-chemo, 47.2% vs. 49%; P=0.91). Adjuvant chemo was not associated
with increased risk of developing CKD (HR 1.00, 95% CI 0.93-1.07). Of chemo-treated pts, 53%, 31%, 7%, and 8% received an
A-based, a CMF, a T-based, and other regimens, respectively. Though the association between adjuvant chemo and risk of CKD
varied across regimen types, these associations were not statistically significant (Table).
Regimen Type
Total, n
Mean (SD) F/U time, years
Rate of CKD (1000 pt-yrs)
Adjusted HR (95% CI)*
No chemo
14024
3.3 (3.6)
29.3
Reference
A-based
7465
4.8 (3.0)
27.7
0.96 (0.88-1.04)
CMF
4389
6.0 (3.9)
29.1
1.04 (0.95-1.14)
T-based
1030
2.8 (1.8)
39.4
0.91 (0.74-1.12)
Other
1140
5.1 (4.0)
31.2
1.04 (0.89-1.22)
*Adjusted for patient baseline characteristics and trastuzumab use

Conclusion: Adjuvant chemo in elderly women with BC may not impose additional risk for CKD. This finding suggests that
patients underlying risk factors for CKD such as diabetes, hypertension, etc. should be considered in the discussions between
clinicians and pts regarding potential risk of CKD development after chemo treatment and choice of chemo treatment.

Title: Case series of 21 patients with metastastic lobular breast carcinoma to the gastrointestinal tract
Noah J Switzer1, Andrew Lim1, Lillian Du1, Katia Tonkin2 and Dan Schiller1. 1University of Alberta, Edmonton, AB, Canada and
2
University of Alberta, Edmonton, AB, Canada.
Body: Background: Invasive lobular carcinoma comprises 5-15% of all breast cancer cases, with its incidence gradually
increasing. While it is uncommon for breast cancer to metastasize to the gastrointestinal tract, lobular carcinoma has a
disproportionately higher incidence of spread to the gastrointestinal (GI) system in comparison to other types of breast cancer. To
date, most studies of gastrointestinal metastatic lobular carcinoma have been case reports and small case series.
Aim: This study is a review of all cases of lobular breast cancer with gastrointestinal metastases seen at a University affiliated
Tertiary Cancer Institute over a five-year period, examining demographic, epidemiological, medical, and treatment factors that
may have an association with the risk of GI metastases.
Methods: This is a retrospective chart review of all patients seen at the Cross Cancer Institute in Edmonton, Alberta, Canada
between 2005-2010, with lobular breast cancer spread to the gastrointestinal tract. The outcomes of interest were: age at
diagnosis, receptor status, site of primary breast cancer, stage at initial presentation, pathology, hormone receptor status, site of
gastrointestinal metastasis, time from diagnosis of breast primary to gastrointestinal metastasis, time from diagnosis to death,
time to gastrointestinal metastasis to death, and treatment regimen for both primary and metastatic disease.
Results: 343 consecutive cases of lobular breast cancer were reviewed, and 21(6%) were found to have GI metastases. The
mean age at initial diagnosis of primary tumor was 63 years. The site of primary breast cancer was most commonly in the outer
upper quadrant. Stage at presentation of the breast primary was: Stage 1A = 17%, Stage 1B/2A = 17%, Stage 2B = 22%, Stage
3A = 11%, Stage 3B = 6%, Stage 3C = 17%, and Stage 4 = 11%. Receptor status of the primary breast cancer was as follows:
HER2+ = 5%, PR+ = 76%, ER+ = 90%.
The mean age at time of diagnosis of metastatic disease was 67 years. The main presenting symptoms of GI metastatic disease
were:: small bowel obstruction (12.5%), incidental finding on endoscopy (12.5%), and incidental finding on imaging (12.5%). Sites
of gastrointestinal spread included the stomach (52%), peritoneum (14%), duodenum (4%), jejunum (4%), transverse colon (4%),
and pancreas (4%). Five-year survival from initial diagnosis of lobular breast cancer averaged 46%. Five-year survival from
diagnosis of gastrointestinal metastasis was 29%.
Conclusions: Approximately 1 in 20 patients diagnosed with a primary breast cancer of lobular pathology will have metastatic
spread to the GI tract, presenting approximately 4 years after their initial primary diagnosis. The most common presentation of
metastatic disease is small bowel obstruction, with the most common site of spread being the stomach. There remains a paucity
of data in the literature and our project is one of the first to further characterize these patients. Future research is needed in
developing treatment regimens for these patients, as the 5-year survival is only approximately 1 in 4.

Title: Breast cancer in Hong Kong, Southern China: The population-based, ten-year analysis of epidemiological characteristics,
stage-specific, cancer-specific, & disease-free survival in breast cancer patients: 19972006
Ava Kwong1,3,4, Oscar WK Mang2, Anthony HP Tam2, Fidelia Wong1, The Hong Kong Breast Cancer Research Group3, Stephen
CK Law2 and Roger KC Ngan2. 1Univeristy of Hong Kong, Hong Kong; 2Hong Kong Cancer Registry, Hong Kong; 3Hospitals of the
Hospital Authority, Hong Kong and 4Hong Kong Hereditary Breast Cancer Family Registry, Hong Kong.
Body: Background: Breast cancer is the most common cancer and second leading cause of cancer death among women, after
lung cancer in Asia. The age-standardized incidence rates of breast cancer in Hong Kong is 61.0 per 100,000 women, after
Singapore and Taiwan. With such increase, it would be important to better understand breast cancer to guide health care
professionals and health policy makers to plan clinical management. However, to date such information is still under-reported, this
study provide a comprehensive ten-year analyses of breast cancer in Hong Kong.
Methods: A retrospective study on population database over 10-year obtained from Hong Kong Cancer Registry was performed.
A total of 20,290 female breast cancers medical records, diagnosed between January 1, 1997 and December 31, 2006, were
reviewed. Descriptive statistics were employed to describe the epidemiological, clinical, and diagnostic data. The prognostic
information for diagnostic and pathological data of relative survival (RS) was estimated using the maximum likelihood approach
with program Strel in STATA; while the overall survival (OS), cancer-specific survival (CSS) & disease-free survival (DFS) were
estimated by the Kaplan-Meier method with SPSS. Chi-squared test and Student's t-test were employed to compare variables in
the two 5-year-periods of 1997-2001 and 2002-2006, with plotted RS curves for diagnostic and pathological data between these
two 5-year-periods.
Results: 18,110 invasive breast cancer medical records in 1997-2006 were eligible for analysis, after 2,180 cases were excluding
due to incomplete data. The ages at diagnosis ranged from 16 to 105; and median age was 51 years old. There was a drop from
14.1% in 1997-2001 to only 10.6% in 2002-2006 for those were diagnosed with breast cancer at age 39 years & younger. 26.2%,
55.2%, 13.0%, & 5.6% in 1997-2001, versus 30.1%, 46.4%, 16.9%, & 6.6% in 2002-2006 had tumor staging of stages I, II, III, and
IV cancers at diagnosis, respectively. In ten-year period, the 5-year OS, RS, CSS, & DFS for the whole cohort were 80.6%,
85.6%, 87.1%, & 90.5%, respectively. The 5-year tumor stage-specific RS were 97.8%, 90.4%, 70.4%, & 21.4% for stages I, II, III,
& IV, respectively. Between the two time periods, all the stage-specific RS improved by about 1%, 4%, 6% & 2% for stages I, II,
III, & IV, respectively. There were 2,670 (14.7%) triple negative cases in 1997-2006, the ER-positive, PR-positive, &
HER2-positive cancers increased from 66.1%, 52.6%, & 25.5% in 1997-2001 to 72.0%, 57.1%, & 29.4% in 2002-2006,
respectively.
Discussion: Comprehensive analyses of breast cancer with population database from the Hong Kong Cancer Registry were
performed to provide detailed information of a baseline study cohort in Southern China for comparative studies with other Asian
regions.


Title: Breast cancer pathologic subtype and fetal microchimerism
David A Mahoney1, Veronica L Winget1, VK Gadi1,2, Christopher I Li1,3, Peggy L Porter1,5 and Jean A McDougall1,4. 1Fred
Hutchinson Cancer Research Center, Seattle, WA; 2University of Washington, Seattle, WA; 3University of Washington, School of
Public Health, Seattle, WA; 4University of Washington, Seattle, WA and 5University of Washington, Seattle, WA.
Body: Background: Stably persistent offspring-origin cells in a womans circulation and tissues decades after pregnancy, also
known as fetal microchimerism, is deficient in women with biologically unselected breast cancer and suggests a protective role.
Although the finding is in harmony with the well-known association of parity with protection against future breast cancer, it has
recently emerged that certain tumor types such as triple negative breast cancer are positively associated with parity. Here we
tested the hypothesis that fetal microchimerism is positively associated with triple negative breast cancer in young women.
Methods: Buffy coat specimens were obtained from a subset (n=450) of pathologically confirmed low-risk luminal and high-risk
triple negative breast cancer patients and control participants from a population-based cohort study of all women aged 20-44
diagnosed with invasive breast cancer in the three county Seattle-Puget Sound metropolitan area from 2004-2010. Using
quantitative PCR, DNA extracts from these specimens were tested for the presence and concentration of Y chromosome
sequence DYS14, a marker of male fetal microchimerism.
Results: At this interim analysis of 405 specimens, 29.3% (56/191) of the healthy controls tested positive. Using controls as a
reference group, 37.8% (34/90; OR .68; 95% CI .39-1.2) of women who developed luminal breast cancer tested positive for the
presence of DYS14, while 26.6% (33/124; OR 1.14; 95% CI .67-1.96) of women who developed triple negative breast cancer
tested positive.
Discussion: A preliminary analysis does not suggest an association between fetal microchimerism and breast cancer pathologic
subtype in young women. Additional analyses are pending and will be presented at the meeting. Young women may regulate fetal
microchimerism differently and surprisingly our data may suggest that as fetal microchimerism emerges later in life in the
circulation, so does the protection against breast cancer a finding consistent with population based studies of parity as a risk
factor for breast cancer.

Title: Findings from the first prospective womb to breast cancer study: New gestational biomarkers support proof of concept that
gestation is a window of susceptibility for the breast
Barbara Cohn1, Nickilou Krigbaum1, Lauren Zimmermann1 and Piera Cirillo1. 1Child Health and Development Studies, Public
Health Institute, CA.
Body: Rationale. Here we provide the first prospective evidence for proof of concept that gestation is an important window of
susceptibility for breast cancer. These findings could open the field to interdisciplinary investigation of mechanisms, and
interventions via clinical and experimental science.
Objective. We tested the hypothesis that biomarkers in gestation predict early-onset breast cancer. Data were prospectively
collected including maternal and paternal peri-conceptual body mass, tobacco and alcohol use, maternal pregnancy weight gain,
pregnancy complications and outcomes, placental morphology assessed by a standardized examination at birth, and
environmental chemicals recently assayed in archived maternal perinatal serum samples. This investigation was based on the
observation of 20,000 pregnancies beginning in 1959, with surveillance for both maternal (F0) and offspring (F1) cancer in the
Child Health and Development Studies pregnancy cohort. This report is based on the first 133 breast cancer cases in F1 that
occurred from 1992-2012, diagnosed at ages 32 to 52.
Results. We observed gestational biomarkers of breast cancer risk which were independent of maternal history of breast cancer
and race. Highlights of significant findings include independent, higher risk for women who: were born to mothers who lost weight
during pregnancy (3-fold increase in risk, p<0.03), were growth retarded in utero (3-fold increased risk, p<0.01), were born with
thick (p<0.01), but small diameter placentas (p<0.04), whose mothers had higher perinatal serum levels of environmental
chemicals including o,p'-DDT (2.5-fold increase in risk for upper quartile, p<0.03), perfluorooctanesulfonic acid (PFOS) precursors
(p<0.03) and total cholesterol (p<0.04). Maternal floor infarction of the placenta was a protective factor both for mothers and their
daughters. In ancillary studies we observed that F1 breast density at mid-life is also impacted by placental morphology during F1
gestation.
Conclusions. Here we provide a high level of evidence for the existence of gestational biomarkers for breast cancer. Prospective
design and direct clinical observation of pregnancies eliminates reporting and misclassification bias. Findings extend the
discussion of gestational biomarkers beyond birthweight and pre-eclampsia which have been previously reported. The importance
of the gestation window for breast cancer in humans is in line with toxicological evidence in animal models and strongly suggests
the existence of opportunities for primary prevention beginning before birth.

Title: Gamma-ray induced mutagen sensitivity and overall survival in young women with breast cancer
Michael C Stauder1, Simona F Shaitelman1, Pamela K Allen1, Abenaa M Brewster1, Banu K Arun1, Wendy A Woodward1, Thomas
A Buchholz1 and Li-E Wang1. 1University of Texas MD Anderson Cancer Center, Houston, TX.
Body: Background:
Hypersensitivity to radiation has been shown to be a risk factor for the development of breast cancer. We aim to determine
whether the same hypersensitivity predicts for adverse clinical outcomes in patients diagnosed with carcinoma of the breast.
Methods:
465 young, female, non-Hispanic white patients diagnosed with carcinoma of the breast at our institution from 1/1997 to 12/2005
were included in this study. All cases were histologically confirmed and all blood was drawn prior to any systemic or local therapy.
Patient age, body mass index (BMI), menopause status, tumor laterality, AJCC stage, ER status, nuclear grade, and receipt of
chemotherapy and radiation were extracted from patient medical records. A gamma-ray-induced mutagen sensitivity assay was
performed using standard published methods to evaluate individual responses to radiation. The number of simple chromatid
breaks per sample was counted from 50 well-spread metaphases. Each simple chromatid break was counted as a single break
and each isochromatid break, exchange figure, or interstitial deletion as two breaks. The mean value of chromatid breaks per cell
(b/c) was then calculated and recorded. Cox multivariable proportional hazards model was used to estimate the hazard ratio (HR)
and 95% confidence interval (CI) for the association between b/c and overall survival.
Results:
A total of 402 patients had a b/c value recorded and were included in the final analysis. The patient median age was 46 years
(range 22-55). 341 patients (84.8%) had invasive cancer and 253 patients (69.9%) had ER+ disease. AJCC stage distribution
was stage 0 (15.2%), stage 1 (41.5%), stage 2 (33.5%), stage 3 (9.5%) and stage 4 (0.3%). The median follow-up for all patients
was 97.2 months (interquartile range, IQR 83.3-119.6 months). The median b/c was 0.5 (IQR 0.38-0.62). The 5 and 10-year
survival for all patients was 92.6% and 87.5%. A statistically significant decrease in 5 and 10-year overall survival was seen in
patients with b/c greater than the median value of 0.5 (96.2% vs. 89.2%, p=.007 and 90.8% vs. 84.5%, p=.046, respectively). On
multivariable analysis (MVA), age at diagnosis (HR 0.95, CI 0.91-0.99, p=.017), BMI (HR 1.07, CI 1.03-1.12, p=.003), ER status
(HR= 0.31, CI 0.16-0.61, p=.01), AJCC stage (HR 1.91, CI 1.2-3.0, p=.006), and b/c level (HR 5.67, CI 1.5-18.2, p=.01) all
predicted for overall survival. Excluding the 61 patients with in situ disease, there remains a significant difference in survival at
both 5 and 7 years (95.5% vs. 88.5%, p=.017 and 93.5% vs. 86%, p=.021). A trend for decrease survival was seen at 10 years
(p=0.09). On MVA for patients with invasive disease, age at diagnosis (HR=0.95, 95% CI 0.91-0.99, p=.026), BMI (HR=1.06, 95%
CI 1.01-1.11, p=.023), AJCC stage (HR=2.41, CI 1.51-3.91, p=.0003), and ER status (HR=0.25, CI 0.12-0.49, p< .0001) and b/c
level (HR=3.76, CI 1.39-8.06, p=.012) were associated with overall survival.
Conclusions:
In this cohort of young, female, non-Hispanic white breast cancer patients, a greater b/c level predicted for decreased overall
survival. The use of a gamma-ray-induced mutagen sensitivity assay may be prognostic and help select for those at increased
risk of death.

Title: Expression of lipid metabolism genes in tumor and contralateral unaffected breast are conversely associated with tumor
estrogen receptor status
Ali Shidfar1, David Ivancic1, Megan E Sullivan1, Pranjal Patankar1, Seema A Khan1 and Jun Wang1. 1Northwestern University
Feinberg School of Medicine, Chicago, IL.
Body: Background: The identification of women at risk for ER- cancer would allow optimization of breast cancer prevention
strategies by guiding their recruitment to studies of agents with efficacy against ER- cancer and sparing them the toxicity of
prevention agents effective only against ER+ cancer. In our previous studies, we identified lipid metabolism (LiMe) gene set in
rFNA samples from contralateral unaffected breast (CUB) that was associated with tumor ER status. In the current study, we
further validate LiMe gene expression in tumor and CUB at the mRNA and protein levels.
Methods: Tissue samples from 56 bilateral mastectomy cases (28 ER+ and 28 ER-) and 28 healthy reduction mammoplasty (RM)
controls were used. The ER+ cases, ER- cases and controls were matched by age, race and menopausal status. We performed
laser capture microdissection of epithelial cells in fresh frozen tissues from tumor and unaffected breast. Total RNA was extracted
and LiMe genes were detected using Taqman low density gene expression arrays. The difference among groups was analyzed
using ANOVA with Sidak multiple comparison adjustment. Three proteins (HMGCS2, ACSL3 and HPGD) were detected in FFPE
sections of tumor and CUB tissues using immunohistochemistry.
Results: Among the 13 LiMe genes, 6 genes (DHRS2, HMGCS2, UGT2B7, UGT2B11, UGT2B28 and GLYATL1) were
significantly higher in CUB of ER- cases compared to CUB of ER+ cases (2.2-2.9 fold, P<0.05). In contrast, the expression of 5
genes (DHRS2, HMGCS2, UGT2B11, UGT2B28 and GSTT2) was significantly lower in ER- tumor compared to ER+ tumor
(0.11-0.37 fold, P<0.05). Immunohistochemistry of HMGCS2, ACSL3 and HPGD confirmed this pattern: protein levels were
higher in ER- CUB than ER+ CUB, but lower in ER- tumor than in ER+ tumor. When CUB samples were compared with healthy
controls, 7 genes (DHRS2, HMGCS2, UGT2B11, UGT2B28, GLYATL1, ALOX15B and SERHL) showed significantly higher
expression in CUB of ER- cases (2.5-9.6 fold, P<0.05), but not in CUB of ER+ cases. Four genes (ACSL3, APOD, AKR1B15 and
HPGD) did not show any significant difference among groups.
Conclusion: Differential expression of the LiMe genes in the CUB is associated with ER- index tumors and may characterize the
environment leading to the development of ER- breast cancer. The converse patterns in tumor and CUB by ER status suggest
that LiMe genes may be regulated by different mechanisms in benign and malignant tissues. These genes are potential risk
biomarkers of ER- breast cancer and generate novel etiologic hypotheses regarding the development of ER- versus ER+ disease.
Gene
ER-C vs ER+C
ER-T vs ER+T
ER-C vs RM
ER+C vs RM
DHRS2
2.4 (0.034)*
0.16 (0.042)*
6.0 (0.014)*
1.3
HMGCS2
2.9 (0.050)*
0.15 (0.0095)*
6.5 (0.006)*
2.2
UGT2B7
2.2 (0.004)*
0.77
1.8
1.0
UGT2B11
2.4 (0.019)*
0.11 (0.0068)*
9.6 (0.013)*
2.0
UGT2B28
2.3 (0.009)*
0.15 (0.018)*
2.8 (0.029)*
1.5
GLYATL1
2.9 (0.010)*
0.37
4.3 (0.026)*
1.4
GSTT2
1.1
0.37 (0.040)*
1.5
1.3
ALOX15B
1.6
0.56
2.5 (0.016)*
1.5
SERHL
1.8
0.29
4.9 (0.0047)*
1.4
ER-C: ER- CUB; ER+C: ER+ CUB; ER-T: ER- tumor; ER+T: ER+ tumor; RM: reduction mammoplasty

Title: No Show

Title: Aspirin and breast cancer risk for BRCA1 and BRCA2 mutation carriers
Naomi Kornhauser1, Mary Beth Terry2, Linda T Vahdat1, Irene Andrulis3, Saundra Buys4, Mary Daly5, Esther John6, John L
Hopper7 and Tessa Cigler1. 1Weill Cornell Medical College, New York, NY; 2Mailman School of Public Health, Columbia
University, New York, NY; 3University of Toronto, Toronto, ON, Canada; 4Huntsman Cancer Institute, University of Utah Health
Sciences Center, Salt Lake City, UT; 5Fox Chase Cancer Center, Philadelphia, PA; 6Cancer Prevention Institute of California,
Fremont, CA and 7University of Melbourne, Melbourne, Victoria, Australia.
Body: Background : Although epidemiologic studies have found evidence that aspirin use may be inversely associated with
breast cancer (BC) risk, little is known about whether this applies to BRCA1 and BRCA2 mutation carriers.
Methods : We compared aspirin use in 613 women with BRCA1 or BRCA2 mutations from the six centers of the Breast Cancer
Family Registry (BCFR) who were recruited at baseline and completed a questionnaire at 10 year follow-up. We defined cases as
carriers with BC (n = 215 with BRCA1 mutations and 137 with BRCA2 mutations) and controls as carriers unaffected with BC (n =
141 with BRCA1 mutations and 120 with BRCA2 mutations). We used logistic regression to estimate odds ratios and 95%
confidence intervals separately by gene mutation type.
Results : Three cases (1.4%) and 27 controls (19.1%) among the BRCA1 carriers and 3 cases (2.2%) and 25 controls (20.8%)
among the BRCA2 carriers reported ever use of aspirin-based medications before diagnosis. Aspirin use before diagnosis was
inversely associated with BC risk for both BRCA1 (OR, 0.13; 95% CI, 0.04-0.46 for ever vs. never use) and BRCA2 (OR, 0.12;
95% CI, 0.03-0.41 for ever vs. never use) carriers, after adjusting for age and center for BRCA1 carriers and age for BRCA2
carriers.
Conclusion : If replicated by larger, prospective studies, aspirin use could become an inexpensive and acceptable risk-reducing
measure for BRCA1 and BRCA2 mutation carriers.

Title: Association of mammographic density in high risk BRCA mutated tumors compared to average risk tumors and healthy
controls: Analysis of Korean Hereditary Breast and Cancer Study (KOHBRA)
Jisun Kim1, Jong Won Lee1, Sung Won Park2, Hee Jung Shin3, Hak Hee Kim3, Sae-Byul Lee1, Jong Han Yu1, Hee Jeong Kim1,
Beom Seok Koh1, Byung Ho Son1 and Sei-Hyun Ahn1. 1Asan Medical Center, College of Medicine, University of Ulsan, Seoul,
Korea; 2Health Promotion Center, College of Medicine, Asan Medical Center, University of Ulsan, Seoul, Korea; 3College of
Medicine, Asan Medical Center, University of Ulsan, Korea and 4Korean Hereditary Breast Cancer Study Group (KOHBRA).
Body: Introduction: Mammographic density is a well-known risk factor of breast cancer as a whole. Nonetheless only few
studies have examined the association of density among high risk breast cancer regarding BRCA mutation. We examined
mammographic density of 2019 breast cancer patients and 2029 healthy controls, regarding risk factors and BRCA mutation
status.
Method: Total 2019 breast cancer patients diagnosed between 1980 to 2011 were divided into two groups- high versus average
risk group. Women with 1)family history of breast/ovarian cancer or 2)younger than age 40 or 3)bilateral cases were considered
high risk group and were participants of Korean Hereditary Breast Cancer study (KOHBRA) whom undergone BRCA testing.
Density of 2029 healthy women who took screening mammogram during the same period were analyzed for comparison. Density
was measured of the unaffected contralateral CC view using computer-assisted method Cumulus by single observer (10%
randomly selected, intra-class correlation coefficient=0.96). Percent density (PD, dense area/breast area, %) among three
groups, association with BRCA mutation status and breast cancer subtypes were examined.
Results and Discussion: Percent density (PD) was significantly higher in high risk group compared to average risk and controls
in a consecutive manner. This finding was consistent after adjusting age and BMI (p*<0.001).
Mammographic percent density(PD) among three group (high risk cancer vs average risk cancer vs control)
High risk cancers
(N=1066)
Average risk cancers Control

(N=953)
(N=2029)
High vs
Average
High vs
control
Average
vs control
Age(yr)
meanSD 41.129.86
51.037.98
50.529.09
0.001
<0.001
0.141
PD (%)
meanSD 52.9012.93
45.5216.04
41.5215.80
<0.001
(<0.001*)
<0.001
(<0.001*)
<0.001
(<0.001*)
PD_median
(%)
<50
442 (41.5%)
567 (59.4%)
1414 (69.7%)
<0.001
<0.001
<0.001
50
624 (58.5%)
387 (40.6%)
615 (30.3%)
<25
26 (2.4%)
106 (11.1%)
329 (16.2%)
<0.001
<0.001
<0.001
25 <50
416 (39.0%)
461 (48.3%)
1085 (53.5%)
50 <75
564 (52.9%)
360 (37.7%)
570 (28.1%)
75
60 (5.6%)
27 (2.8%)
45 (2.2%)
PD_quartile
(%)
* adjusted p value by age and BMI

High mammographic density showed to be a significant risk factor throughout different subtypes.
Magnitude of mammographic density as a risk factor according to receptor status
25%PD<50%
(<25%(ref))
TNBC (control(ref))
Non-TNBC
Luminal
Non-Luminal
HR 4.21, 95%CI 2.12-8.38
HR 1.91, 95%CI
1.50-2.43
HR 1.77, 95%CI
1.38-2.26
HR 3.04, 95%CI
2.00-4.64
50%PD<75%
HR 8.85, 95%CI
4.45-17.61
HR 3.81, 95%CI
2.98-4.89
HR 3.89, 95%C
I3.03-5.00
HR 5.08, 95%CI
3.32-7.80
75%PD
HR 16.25, 95%CI
6.97-37.87
HR 4.09, 95%CI
2.61-6.41
HR 4.09, 95%CI
2.61-6.43
HR 8.16, 95%CI
4.41-15.07
Among the 1066 high risk group, 81.5% (869) undergone BRCA testing and 70(6.6%) had BRCA1, 78(7.3%) had BRCA2
mutations without significant difference in density. Similar strong magnitude association of mammographic density was observed
in both BRCA mutated/non-mutated tumors and among subtypes. The ongoing GWAS and whole exome analysis of this
population-subset will give insight into the tumor etiology and how density could stratify breast cancer risk for personalized
screening especially in high risk population.

Title: Demographics of breast cancer in a cohort of Afro-Caribbean women
Sophia HL George1, Talia Donenberg2, Mohammed Akbari3, Cheryl Alexis4, Gillian Wharfe5, Sook Yin6, Hedda Dyer7, Theodore
Turnquest8, Vincent DeGennaro9, Steven Narod10 and Judith Hurley11. 1Campbell Family Institute for Breast Cancer Research,
Toronto, ON, Canada; 2University of Miami, Miami, FL; 3Women's College Hospital, Toronto, ON, Canada; 4University of West
Indies, Wanstead, St Michael, Barbados; 5University of West Indies, Kingston, Jamaica; 6Cancer Society of Cayman Islands,
George Town, Grand Cayman, Cayman Islands; 7Ross School of Medicine, Portsmouth, Dominica; 8Princess Margaret Hospital,
Nassau, Bahamas; 9University of Florida, Gainsville, FL; 10Women's College Hospital, Toronto, ON, Canada and 11University of
Miami, Miami, FL.
Body: Objectives: In Latin America and the Caribbean, non communicable chronic diseases are now the leading cause of
premature mortality. The incidence of cancer has increased in the region as a result of population aging and growth but also as
more people adopt lifestyle choices like smoking, physical inactivity, and westernized diets. In women, fertility factors such as
decreased parity, earlier onset of menses and later age at time of first pregnancy are known epidemiologically to increase
incidence of hereditary and sporadic breast cancer. Recently there has also been a strong link to a genetic etiology of the breast
and ovarian cancer in the Bahamas (27% in unselected breast cancer cases). A study was designed to address the prevalence
and spectrum of BRCA1 and BRCA2 mutations in the Afro-Caribbean population.
Methods: Demographic and clinical pathologic data was collected from 347 women of Afro-Caribbean decent. The cohort
included women with breast cancer from the following countries: the Cayman Islands, Jamaica, Barbados, Dominica, Trinidad
and Haiti. Summary statistics and t-tests and ANOVA were used to analyze population characteristics. A Bahamian mutation
panel was created and detailed analyses of samples are ongoing.
Results: The mean age of onset in the cohort is 48.1 yrs with a mean BMI of 27.7. 70% of breast cancer cases ER+ (n=241
informative cases) and in Jamaica 27% (n=105) of breast cancer cases were Her2+. 67.8% cases were diagnosed at stages II/III
(n=90). TAH-BSO delayed invasive breast cancer from 48 to 53 years, p=0.005. Parity was a statistically significant factor
(p<0.0001), which delayed age of onset by 8 yrs. Additionally, pregnancy alone delayed age of onset (p<0.005) also by 8 yrs in
our cohort (n=379). Only three women out of 347 were found to have a mutation.
Summary demographics of Caribbean women with breast cancer
Country
No. of Participants
Mean age at diagnosis (yrs)
ER+
Her2+
BRCA1/2 +
BMI
Cayman Islands
66
51.4
29/44
3/43
1/27
30.3
Barbados
89
46.6
52/73
9/73
4/87
27.5
Dominica
60
52.2
9/13
2/7
0/46
28.7
Jamaica
137
48.6
69/105
28/105
28.7
Trinidad and Tobago
45.5
4/6
1/5
1/5
27.4
Haiti
34
48.5
NA
NA
NA
24.9
Reproductive Characteristics of Caribbean women with breast cancer
TAH-BSO
Parity
Mean Age
p-value
Performed
71
53
0.005
Not Performed
304
48
Nulliparous
60
42
0.0001
Pregnancy
Multiparous
319
50
None
48
43
More than 1
331
50
0.0001
Conclusions: This population-based study provides an insight into pattern of risk factors both genetic and environmental of
breast cancer incidence and subtype across the Caribbean. In conclusion 1) genetic causes of breast cancer appear rare outside
of the Bahamas, 2) fertility factors appear important in the development of breast cancer, 3) TAH-BSO is common as both a form
of contraception and because of the high incidence of fibroids in the Caribbean and it may be protective, 4) BMI may impact on
breast cancer development and 5) screening mammography is rare and the vast majority of mammography performed is
diagnostic in nature.

Title: A multicenter randomized study comparing the dose dense G-CSF-supported sequential administration of FEC followed by
docetaxel versus docetaxel plus cyclophosphamide as adjuvant chemotherapy in women with HER2-negative, axillary lymph
node-positive breast cancer
Dimitrios Mavroudis1, Nikolaos Malamos1, Pavlos Papakotoulas1, Stylianos Kakolyris1, Ioannis Boukovinas1, Elias Athanasiadis1,
Nikolaos Kentepozidis1, Nikolaos Ziras1, Kostas Kalbakis1, Charalambos Christophyllakis1 and Vassilis Georgoulias1. 1Hellenic
Oncology Research Group (HORG), Athens, Greece.
Body: Background: The dose dense sequential administration of anthracycline and taxane is very effective as adjuvant therapy in
node-positive early breast cancer. The non-anthracycline regimen docetaxel plus cyclophosphamide (TC regimen) was better
than four cycles of doxorubicin/cyclophosphamide as adjuvant therapy. This study compared the dose dense sequential regimen
versus the TC regimen as adjuvant therapy.
Methods: Women with axillary node-positive, HER2-negative early breast cancer were randomized following surgery to receive
either dose dense G-CSF-supported FEC (5FU 500mg/m2, epirubicin 75mg/m2, cyclophosphamide 500mg/m2 every 14 days for
4 cycles) followed by Docetaxel (75mg/m2 every 14 days for 4 cycles) (arm A) or 6 cycles of Docetaxel 75mg/m2 plus
Cyclophospamide 600mg/m2 every 3 weeks (arm B). The primary endpoint of the study was the 3-year disease-free survival
(DFS).
Results: Six hundred fifty patients were randomized; 326 on arm A and 324 on arm B. Of them 109 (33%) and 90 (28%) were
premenopausal, 196 (60%) and 218 (67%) had 1-3 positive nodes, 284 (87%) and 288 (89%) were hormone receptor positive in
arm A and B, respectively. Chemotherapy was completed in 97% and 93% of patients in arm A and B, respectively. After a
median follow up of 46 and 47 months, there were 37 (11.3%) and 33 (10.1%) disease relapses and the median DFS has not yet
been reached (p=0.5) while the 3-year DFS rate was 89.5% and 91.1% for arm A and B, respectively. Neutropenia grade III-IV
was more common in arm B and anemia, nausea, vomiting and fatigue grade II-III in arm A. No toxic deaths occurred.
Conclusions: The 3-year DFS rate was similar between the dose dense sequential FEC/docetaxel combination and the TC
regimen in women with node-positive HER2-negative early breast cancer.

Title: Association between definitive surgery and times to administration of adjuvant chemotherapy and outcomes in early breast
cancer: Analysis of adjuvant studies conducted by NCIC Clinical Trials Group (NCIC CTG)
Ravi Ramjeesingh1, Bingshu E Chen1, Joseph L Pater1, Liting Zhu1, Margot Burnell2, Vivien H Bramwell3, Kathleen I Pritchard4,
Lois E Shepherd1 and Wendy R Parulekar1. 1NCIC Clinical Trials Group, Queen's University, Kingston, ON, Canada; 2Atlantic
Health Sciences Corporation, Saint John, NB, Canada; 3Tom Baker Cancer Centre, Calgary, AB, Canada and 4Sunnybrook
Odette Cancer Centre, Toronto, ON, Canada.
Body: Background: For early high risk breast cancer, adjuvant chemotherapy following definitive surgery is a current standard of
care. However, the optimal time to commencement of therapy has not been well established. We evaluated the association
between time to initiation of adjuvant chemotherapy after definitive surgery (sTTC) with survival.
Methods: We retrospectively analyzed 4 NCIC CTG-led breast cancer adjuvant clinical trials involving women diagnosed from
1984 and 2005 with stage 1 to 3 breast cancer, who received adjuvant chemotherapy. Patient data were categorized into four
time groups: < 4, 4-8, >8-12, >12 weeks after definitive surgery. Outcomes measured were: overall survival (OS) and
disease-free survival (DFS).
Results: 3837 patients were included in the final analysis. In univariate analysis, an improvement was identified for patients
treated between >8 and 12 weeks:
Number of Pts
10 yr OS(%)
p-value
10 yr DFS(%)
p-value
<4 weeks
774
69.6
ref
60.9
ref
4-8 weeks
2263
71.5
0.1351
64.7
0.1043
>8-12 weeks
742
77.3
0.0002
72.6
0.0001
>12 weeks
35
77.0
0.4531
73.9
0.2124
However, in multivariate analysis there was no significant association between any sTTC time periods and either outcome (see
table below). Covariates which did show a significant association are listed in the table.
OS p-value
DFS p-value
<4 weeks
ref
ref
4-8 weeks
0.7762
0.6764
>8-12 weeks
0.4251
0.1396
>12 weeks
0.6861
0.9312
Age
<0.0001
<0.0001
BMI
0.0238
0.0298
Menopause Status
0.0126
0.1226
Performance Status 2
0.0256
0.1665
Surgery Type
<0.0001
<0.0001
Pathological Stage 3
0.0037
0.0229
Positive Nodal Status
<0.0001
<0.0001
ER Negative
<0.0001
<0.0001
Time to Adjuvant
Chemotherapy Regimen
<0.0001
<0.0001
The fact that the statistical significance of sTTC in univariate analysis was lost when covariates related to DFS and OS were
accounted for in a multivariate analysis suggested there might be a relationship between sTTC and the risk of adverse outcomes.
A disease risk score analysis was therefore carried out, but there was no indication of an advantage to an earlier sTTC within
disease risk categories.
Conclusions: Within the context of chemotherapy given within 12 weeks, we were unable to demonstrate an effect on survival
based on time to adjuvant chemotherapy in our multivariable analysis. Those treated later did not do significantly worse than
those treated earlier. Significant covariates which effected survival were consistent with predictors of poor prognosis (younger
age, poorer performance status, increased disease burden and receptor negativity). There is a potential relation of patient risk to
time to treatment which requires further study.


Title: AML and MDS after adjuvant chemotherapy: A population based study among older breast cancer patients
Aron S Rosenstock1, Sharon H Giordano1, Jiangong Niu1, Hui Zhao1, Antonio C Wolff2, Thomas A Buchholz1 and Mariana
Chavez-MacGregor1. 1University of Texas MD Anderson Cancer Center, Houston, TX and 2Johns Hopkins Sidney Kimmel
Comprehensive Cancer Center, Baltimore, MD.
Body: Background: Adjuvant chemotherapy for early breast cancer is associated with a small risk of developing MDS and AML.
The aim of this study is to determine the risk of developing AML or MDS after modern adjuvant chemotherapy in older breast
cancer patients and to further define the risk of individual chemotherapy regimens.
Methods: Patients diagnosed with stage I-III breast cancer from 2003 to 2009 were identified in the Surveillance, Epidemiology
and End Results (SEER)-Medicare linked database. Development of AML or MDS, chemotherapy use, and comorbidities were
identified using ICD-9 and HCPCS codes. A single inpatient or 2 outpatient codes 30 days apart were required for identification of
AML or MDS. Analyses included descriptive statistics, Kaplan-Meier method for 5-year AML/MDS rates and Cox proportional
hazards models to estimate the hazard of AML and MDS after adjusting for clinically relevant covariates.
Results: 68,702 patients were included, of them 166 (0.24%) developed AML and 385 (0.56%) developed MDS. The median time
from breast cancer diagnosis to AML or MDS was 924 and 943 days respectively. The 5-year AML rates were 0.23% for the no
chemotherapy group, 0.53% for anthracycline (A)-based regimens, 0.48% for anthracycline and taxane (AT)-based regimens,
0.39% for other (O) regimens and 0.09% for those who received docetaxel and cyclophosphamide (TC). 5-year MDS estimates
were 0.62%, 1.07%, 0.72%, 1.08 and 0.38% respectively. In the multivariable model using no chemotherapy as the reference
category, A (HR 3.03; 95%CI 1.75-5.24) and AT (HR 3.06; 95%CI 1.70-5.52) were the chemotherapy regimens associated with
the highest risk for developing AML, followed by O (HR 2.05; 95%CI, 1.13-3.69). No significant increase in risk was observed in
patients who received TC (HR 0.91; 95%CI 0.26-4.60). Similar results were observed when evaluating risk of MDS among
patients treated with A (HR 2.25; 95%CI, 1.50-3.37), AT (HR 1.67; 95%CI 1.06-2.65), O (HR 2.30; 95%CI 1.61-3.29) and TC (HR
1.22; 95%CI 0.60-2.66).
Conclusion: In this large cohort of patients we observed a small but significant increase in risk for AML and MDS with adjuvant
chemotherapy. The highest risk was associated with A and AT-based chemotherapy. TC was the only regimen that was not
significantly associated with an increased risk. Because of relative recent adoption of TC, longer follow-up is needed to better
estimate the risk of MDS and AML in this group of patients.

Title: Identification of optimal adjuvant chemotherapy regimen for early-stage breast cancer: A systematic review and bayesian
network meta-analysis
Takeo Fujii1,2, Fanny Le Du1, Lianchun Xiao1, Takahiro Kogawa1, Vicente Valero1, Yu Shen1 and Naoto T Ueno1. 1University of
Texas MD Anderson Cancer Center, Houston, TX and 2University of Texas Health Science Center School of Publich Health,
Houston, TX.
Body: Background: The National Comprehensive Cancer Network (NCCN) guidelines for treatment of breast cancer
recommend several adjuvant chemotherapy regimens, including sequential or concurrent anthracycline-cyclophosphamide and
taxane-based regimens (sequential AC-T or concurrent ACT, respectively); anthracycline alone; cyclophosphamide,
methotrexate, and 5-fluorouracil (CMF); and docetaxel and cyclophosphamide (TC). Sequential AC-T is the most commonly
prescribed standard regimen. Two types of regimens without anthracycline, TC and platinum-containing regimens, have been
speculated to have similar efficacy to that of ACT-based regimens. Platinum-containing regimens have also demonstrated high
efficacy in breast cancer with tolerable adverse effects in several previous studies. However, because of the limitations of
conventional meta-analysis, we do not know how these regimens compare with one another in terms of survival or adverse
events. Thus, the best adjuvant chemotherapy regimen for breast cancer is not known. Bayesian network meta-analysis allows
comparison of treatments for which there has been no head-to-head comparison by using indirect comparisons. Methods: We
searched the MEDLINE, EMBASE, and Cochrane Library databases for articles published before January 2014; the American
Society of Clinical Oncology annual meeting abstracts for 1983-2013; and the American Association for Cancer Research annual
meeting abstracts for 1916-2013. We included only randomized controlled trials of adjuvant treatments for breast cancer that
compared 2 or more of the following: observation alone; sequential AC-T; concurrent ACT; anthracycline alone; CMF; TC; and
platinum-containing regimens. We compared regimens by using a network meta-analysis approach and random-effects models.
Network meta-analysis allows derivation of hazard ratios (HRs) for death for each regimen and comparison of these HRs with
each other even when there are no direct comparisons between 2 regimens. Results: We identified 24 trials with a total of 28,853
patients. Network meta-analysis showed that OS with TC and platinum-containing regimens were similar to OS with sequential
AC-T (TC: HR=0.94; 95% CI, 0.58-1.52; platinum: HR=0.90; 95% CI, 0.69-1.19). Patients receiving CMF or anthracycline alone
had significantly worse OS than those with sequential AC-T (CMF: hazard ratio [HR]=1.62; 95% CI, 1.31-2.01; anthracycline:
HR=1.23; 95% CI, 1.07-1.43). For overall adverse events, the mean number of adverse events per patient was higher for
platinum-containing regimens (2.1) and concurrent ACT (1.22) than for sequential AC-T (1.17). The mean number of
hematological adverse events was higher for concurrent ACT (0.67) and TC (0.63), than for sequential AC-T (0.48). The mean
number of nonhematological adverse events was higher for platinum-containing regimens (1.76) than for sequential AT (0.7).
Conclusion: Our findings suggest that sequential AC-T should continue to be considered the optimal treatment for early-stage
breast cancer because of the equivalent survival benefit of concurrent ACT, TC and platinum-containing regiments but superior
adverse-event outcomes.

Title: Frequency of and reasons for deviation from therapy decisions in patients with primary breast cancer initially intended to
receive chemotherapy Results from the German prospective multi-center study BRENDA II
Lukas Schwentner1, Susanne Singer2, Reyn Van Ewijk2, Wolfgang Janni1, Maria Blettner2, Rolf Kreienberg1 and Achim Wckel3.
1
University Ulm, Germany; 2University Mainz, IMBEI, Germany and 3University Wrzburg, Germany.
Body: Objectives
This study examined the question, how often chemotherapy is not given to breast cancer patients when it is indicated according
to guidelines and when the multi-professional tumour board (TB) decided to apply chemotherapy.
Methods
In a prospective multi-center cohort study, patients with primary breast cancer were sampled consecutively over a period of four
years (2009-2012). Patients completed a questionnaire prior to surgery and prior to adjuvant therapy. This questionnaire
assessed health related quality of life (QoL) using the EORTC QLQ-C30, psychiatric comorbidity with the Patient Health
Questionnaire (PHQ), demographic characteristics (age, education), and the intensity of fear for chemotherapy.
After surgery, a multi-professional team discussed recommendation for chemotherapy and this decision was documented in a
database together with the indication for chemotherapy according to the German S3 guideline. The multi-professional team was
blinded to that algorithm-based decision. Six months later, it was documented whether the patient had received adjuvant
chemotherapy or not.
Patients were included in the analysis when chemotherapy was indicated (high risk) or possible (intermediate risk) according to
the guidelines and when the multi-professional team had decided to recommend chemotherapy. Risk group stratification is based
on St. Gallen 2007 classification.
Statistical analysis was performed with multivariate logistic regression, separately for high and intermediate subgroup.
Results
Altogether, 857 patients were included in the study of whom based on the guideline, 241 were indicated (high risk) for
chemotherapy and in 537 it was possible (intermediate risk) to apply adjuvant chemotherapy. In only 391 of those patients
(accrued from high and intermediate risk subgroup), the TB decided to recommend chemotherapy. The most important reason for
not recommending chemotherapy was somatic comorbidity not allowing adjuvant chemotherapy. Of those 391 patients, 73 (19%)
patients had not received chemotherapy (10% of high risk, 28% of intermediate risk subgroup).
If patients where CT was recommended (according to the tumorboard decision) deviations from initial therapy decision was likely
if they were old (75 years) and had poor QoL, (OR 0.003, p 0.001). There was also some evidence that patients with higher
education (OR 0.3, p 0.07) less frequently received CT.
If patients with intermediate risk (CT possible) were very afraid of chemotherapy, deviations from initial therapy decision was likely
(OR 0.4, p 0.03). In that group of patients, age, QoL, education, and comorbidity were unrelated to deviations from initial decision.
Conclusion
Even in cases where chemotherapy was indicated or possible according to guidelines and after the decision of multi-professional
team to recommend it, about 19% of patients eventually did not receive chemotherapy. In those patients, this mainly happened
associated with poor QoL in elderly patients >75. In the intermediate risk group with chemotherapy recommendation, patients
fear of chemotherapy is the main factor preventing patients from adjuvant chemotherapy.

Title: Prognostic value of immunophenotypically defined subtypes in patients treated with dose-dense sequential adjuvant
chemotherapy in the trastuzumab era. A Hellenic Cooperative Oncology Group study
George Fountzilas1, Eleni Timotheadou1, Georgia Gourgioti1, Petroula Arapantoni-Dadioti1, Sotiris Lakis1, Anna Batistatou1,
Triantafyllia Koletsa1, Olympia Tzaida1, Mattheos Bobos1, Alexandra Papoudou-Bai1, Eleftheria Tsolaki1, Sofia Chrisafi1, Elena
Fountzilas1, Ioannis Efstratiou1, Helen Gogas1, Flora Zagouri1 and Dimitrios Pectasides1. 1Hellenic Cooperative Oncology Group
(HeCOG), Athens, Greece.
Body: Background-Aim: The aim of the present study was to explore the efficacy of dose-dense sequential adjuvant
chemotherapy followed by trastuzumab in HER2-positive patients according to the immunohistochemically (IHC) defined
subtypes.
Patients and methods: A total of 771 formalin-fixed paraffin-embedded (FFPE) tumor tissue samples, prospectively collected
from 990 eligible patients with high-risk N0 or N1 operable breast cancer participating in a phase III trial (HE10/05), were centrally
assessed in tissue microarrays by IHC for 6 biological markers, that is, estrogen receptor (ER), progesterone receptor (PgR),
HER2, Ki67, cytokeratin 5 (CK5) and EGFR. All cases were further evaluated for HER2 amplification by FISH. Patients were
classified as: luminal A (ER/PgR-positive, HER2-negative, Ki67low, N=382, 49.5%); luminal B (ER/PgR-positive, HER2-negative,
Ki67high, N=136, 17.6%); luminal-HER2 (ER/PgR-positive, HER2-positive, N=125, 16.2%); HER2-enriched (ER-negative,
PgR-negative, HER2-positive, N=63, 8.2%); triple-negative (TNBC) (ER-negative, PgR-negative, HER2-negative, N=65, 8.4%);
and basal core phenotype (BCP) (TNBC, CK5-positive and/or EGFR-positive, N=53, 6.9%).
Results: At a median follow-up of 60.5 months, the 3-year disease-free survival (DFS) and overall survival (OS) rates for the total
patient population were 88.3% and 96.0%, respectively. The 3-year DFS rates for luminal A, luminal B, luminal-HER2,
HER2-enriched, TNBC and BCP patients were 91.1%, 88.2%, 86.4%, 93.7%, 87.7%, and 89.4%, respectively, while the
corresponding 3-year OS rates were 95.8%, 95.6%, 97.6%, 95.2%, 95.4%, and 95.0%, respectively. No significant differences
were detected for either 3-year DFS or OS in the immunohistochemically defined subtypes, except a trend for significantly worse
DFS in patients with luminal-HER2 tumors compared to patients with HER2-enriched tumors (log-rank, p=0.069).
Conclusions: In the post-trastuzumab era, at a relatively short follow-up, the luminal-HER2 patients show a trend for worse DFS
compared to patients with HER2-enriched tumors treated with dose-dense sequential adjuvant chemotherapy followed by
trastuzumab. No other significant differences were detected, with follow-up however being continued.

Title: Patterns of oncotype DX testing and adjuvant chemotherapy use in a tertiary care centre
Gautam Sudan1,2, Mihaela Mates1,2 and Hopman Wilma2,3. 1Cancer Centre, Southeastern Ontario at KGH, Kingston, ON, Canada;
2
Queen's University, Kingston, ON, Canada and 3KGH Clinical Research Centre, Kingston, ON, Canada.
Body: Introduction:
The Oncotype DX Recurrence Score (RS) is a 21-gene signature, retrospectively validated prognostic marker and predictor of
response to adjuvant chemotherapy (ACT) in estrogen receptor (ER) positive breast cancer (BC). In 2010 RS became publicly
funded in Ontario for patients with estrogen-positive, HER2 negative and node negative breast cancer. The aim of our study was
to explore the pattern of RS testing and use of adjuvant chemotherapy at the Cancer Centre of Southeastern Ontario (CCSEO)
since the introduction of RS testing. Additionally we compared the RS tested patients with a matched cohort treated for early BC
prior to 2010.
Methods:
A retrospective paper and electronic chart review was undertaken of patients with early BC (stage 1 and 2) with the following
pathologic features: T1-T2, N0/N1mic, ER positive and Her-2neu negative. We collected patient demographics, co-morbidities,
surgical data, tumour characteristics, ACT use, Adjuvant! mortality estimates and breast cancer outcomes. Cohort A included
patients who underwent RS testing (2010 2013) and cohort B patients treated prior to 2010. The two cohorts were compared
using chi-square tests for categorical data, and independent samples t-tests and the Mann-Whitney U for continuous data.
Results:
160 patients were included in our analysis of which 83 underwent RS testing. Compared to cohort B, cohort A was older (median
age 60 versus (vs) 48, p<0.001); had higher postmenopausal status (77 vs 34%, p<0.001); higher rates of breast conserving
surgery (88 vs 74%, p=0.024) and sentinel node biopsy (94 vs 26%, p<0.001). Cohort A also had larger tumors (T2 23 vs 5%,
p=0.010), higher stage (stage 2 - 24 vs 6%, p=0.002) and higher Adjuvant! mortality estimates above 15% (18 vs 8%, p=0.054).
Despite this the use of ACT decreased significantly (20 vs 98%, p<0.001). The majority of patients received adjuvant endocrine
treatment (> 90% in both cohorts). Of the 83 patients in cohort A 55 (66%) had low risk RS (0-17), 18 (22%) intermediate risk
(18-30) and 10 (12%) high risk RS. ACT was received by 2 of 55 patients with low RS; 6 of 18 patients with intermediate RS and
9 of 10 with high RS. Median follow-up was only 16 months in cohort A vs 74 months in cohort B. 92% of patients in cohort B
remain recurrence free. To date only 2 patients in cohort A have recurred, both of which had high RS scores and received ACT.
Conclusion:
In patients with early BC undergoing RS testing at CCSEO we observed a higher proportion of low risk RS as compared to the
literature. Additionally, despite larger tumors, higher stage disease and higher Adjuvant! mortality estimates above 15%, the
proportion of patients undergoing ACT has significantly decreased since the introduction of RS testing. So far this has not
translated into a negative long-term disease outcome although longer follow-up is needed for real-world validation.

Title: Eribulin mesylate plus capecitabine for adjuvant treatment in post-menopausal ER+ early-stage breast cancer: A phase 2,
multicenter, open-label study using 2 different dosage regimens
John W Smith II1, Svetislava Vukelja2, Anthony Hoffman3, Vicky Jones4, Kristi McIntyre5, Erhan Berrak6, Angela Teng6, David Cox6
and Joyce O Shaughnessy7. 1Compass Oncology, US Oncology, Portland, OR; 2Texas Oncology, US Oncology, Tyler, TX;
3
Eastchester Center for Cancer Care, Bronx, NY; 4Yakima Valley Memorial Hospital, North Star Lodge Cancer Center, Yakima,
WA; 5Texas Oncology, Dallas Presbyterian Hospital, US Oncology, Dallas, TX; 6Eisai Inc, Woodcliff Lake, NJ and 7Baylor Charles
A. Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, TX.
Body: Introduction: Eribulin mesylate is a novel non-taxane microtubule inhibitor. The primary objective of this study was to
explore feasibility of administering eribulin plus capecitabine as adjuvant therapy in subjects with early-stage, estrogen
receptor-positive (ER+) breast cancer.
Methods: In this phase 2, open-label, multicenter study, 67 postmenopausal women with early HER2-normal/HER2-negative,
ER+ breast cancer received four 21-day cycles of treatment with eribulin mesylate (1.4 mg/m2 IV on day 1 and day 8 of each
cycle) in combination with capecitabine (900 mg/m2 orally twice daily on days 1 through 14 of each cycle). A second dosage
regimen for capecitabine was initiated after dose reductions and treatment discontinuations were noted and attributed to
capecitabine-related toxicities, such as grade 3/4 GI events and hand-foot syndrome. As a consequence, capecitabine was
administered to an additional cohort of 10 subjects at a fixed dose of 1500 mg given orally twice daily on a 7 days on/7 days off
schedule for the 4 cycles; eribulin mesylate was administered on the same dosage schedule as the original regimen (1.4 mg/m2
IV on day 1 and day 8 of each cycle). The 7 days on/7 days off regimen for capecitabine was based on mathematical modeling
and has been shown to have an acceptable toxicity profile, including minimal gastrointestinal toxicity, when given in combination
with bevacizumab to patients with metastatic breast cancer. Feasibility was assessed based on relative dose intensity (RDI) of the
combination using prespecified criteria of 80% of subjects achieving an RDI of at least 85% of the regimen with lower 95%
confidence boundary >70%; safety and tolerability were also assessed.
Results: Among subjects on the original eribulin plus capecitabine dosing schedule (n=64), the average (SD) RDI was 90.6%
(11.94%) and the feasibility rate was 81.3% (95% lower CI: 71.4%), indicating that this dosage regimen is feasible. Dose
reductions, missed doses, and withdrawals due to adverse events were ascribed more to capecitabine (36%, 85%, and 18%,
respectively) than to eribulin (21%, 8%, and 12%, respectively), which led to higher RDI and feasibility rates for eribulin (93.5%
and 82.8%, respectively) than for capecitabine (87.8% and 71.9%). Grade 3/4 hand-foot syndrome ascribed to capecitabine only,
led to dose reductions in 11.9% of subjects under the original dosing schedule. The most common adverse events under the
original dosing schedule were alopecia (77.6%), fatigue (58.2%), and nausea (52.2%). With the new dosing schedule (n=9),
higher RDI and feasibility rates were achieved, the average RDI was 96% and the feasibility rate was 100%, indicating that this
alternative regimen is also feasible, and probably better. (Detailed feasibility and safety data with new dosing regimen will be
available at the time of the presentation.)
Conclusions: Adjuvant use of the combination of eribulin plus capecitabine is feasible in patients with early,
HER2-normal/HER2-negative, ER+ breast cancer. The combination had an acceptable safety profile under the original dosing
schedule and has the potential to be improved by use of a 7 day on/7 day off regimen of capecitabine.

Title: Randomized phase III trial of taxanes versus S-1 as first-line chemotherapy for metastatic breast cancer (SELECT BC:
CSPOR- MBC01)
Takanori Watanabe1, Kojiro Shimozuma2, Kentaro Imi3, Hiroyoshi Doihara4, Hiromitsu Akabane5, Hiroaki Ueo6, Shinji Ohno7,
Masahiro Kashiwaba8, Atsushi Fukuuchi9, Kenichi Watanabe10, Michiko Tsuneizumi11, Hirotsugu Isaka12, Yukari Uemura13, Yasuo
Ohashi14 and Hirofumi Mukai15. 1National Hospital Organization Sendai Medical Center, Sendai, Japan; 2Ritsumeikan University,
Kusatsu, Japan; 3School of Medicine Kyorin University, Mitaka, Japan; 4Okayama University, Okayama, Japan; 5Hokkaido
P.W.F.A.C. Asahikawa-Kosei General Hospital, Asahikawa, Japan; 6Ueo Breast Cancer Hospital, Oita, Japan; 7National Hospital
Organization Kyushu Cancer Center, Fukuoka, Japan; 8Iwate Medical University, Morioka, Japan; 9Mitsui Memorial Hospital,
Tokyo, Japan; 10National Hospital Organization Hokkaido Cancer Center, Sapporo, Japan; 11Shizuoka General Hospital,
Shizuoka, Japan; 12Kugayama Hospital, Tokyo, Japan; 13University of Tokyo, Tokyo, Japan; 14Chuo University, Tokyo, Japan and
15
National Cancer Center Hospital East, Kashiwa, Japan.
Body: Background: Treatment goals of metastatic breast cancer (MBC) are to prolong survival and improve health-related quality
of life (HRQOL). Current standard first-line chemotherapy for MBC are the taxanes or anthracyclines; however treatment-related
adverse events greatly reduce HRQOL. S-1 is an oral 5-fluorouracil derivative, and phase II trials showed good clinical efficacy
and tolerability. We conducted a phase III randomized controlled trial to establish non-inferiority of S-1 in overall survival (OS) and
superiority in HRQOL to taxanes, when given as first-line chemotherapy for MBC.
Methods: Patients with HER2-negative non-life-threatening MBC, nave to chemotherapy for metastatic disease, were randomly
assigned to the taxane or S-1 groups. In the taxane group, patients received docetaxel 60-75mg/m2 q3w, paclitaxel
80-100mg/m2 q1w, or paclitaxel 175 mg/m2 q3w according to institutional policy. In the S-1 group, patients received S-1 4060
mg twice daily based on body surface area using a 28 days on;14 days off regimen. Treatment was repeated until tumor
progression or for at least 6 cycles (taxane) or 4 cycles (S-1). After failure of the first-line protocol therapy, another cytotoxic agent
was administered, based on the investigators discretion. HRQOL was assessed with the European Organization for Research
and Treatment of Cancer (EORTC) QLQ-C30, the Patient Neurotoxicity Questionnaire (PNQ) and the EQ-5D at baseline and 3, 6,
12 months after the start of the treatment. The primary endpoint was OS. Secondary endpoints were time to treatment failure
(TTF), adverse events, and HRQOL.
Results: A total of 618 women were enrolled. After a median follow-up of 34.6 months, median OS was 37.2 months in the taxane
group (n=309) and 35.0 months in the S-1 group (n=309) (hazard ratio [HR] 1.05, 95% confidence interval [CI] 0.861.27,
non-inferiority test p=0.015). Median TTF was 8.9 months in the taxane group and 8.0 months in the S-1 group (HR 1.10, 95% CI
0.931.30, p=0.022). The incidence of the following grade 3-4 adverse events, allergic reaction, edema and sensory neuropathy,
were statistically significantly more frequent in the taxane group (p=0.038, 0.0013 and 0.0077, respectively). Hematologic and
non hematologic toxicities except above did not differ significantly between the two groups. The results of the EORTC QLQ-C30
under study treatment indicated that the S-1 was better than the taxanes in global health status/QOL (p=0.044), physical
functioning (p=0.002), role functioning (p=0.002), emotional functioning (p=0.004), cognitive functioning (p=0.026), social
functioning (p<0.0001), pain (p=0.042) and financial difficulties (p=0.003). EQ-5D utility scores were significantly higher in the S-1
group (p=0.033) during the first year. PNQ sensory and motor scores were significantly better in the S-1 group (p<0.0001 and
p=0.0002, respectively).
Conclusions: This study clearly demonstrated that S-1 was superior to taxanes in terms of HRQOL and toxicity, without
compromising the prolonged OS. S-1 should be considered as a new standard for first-line chemotherapy for MBC. We are
conducting another similar trial (UMIN000005449) that compares first-line anthracycline with S-1 in terms of OS and HRQOL.

Title: Gemcitabine with cisplatin or paclitaxel in metastatic triple-negative breast cancer
Xichun Hu1, Binghe Xu2, Li Cai3, Zhonghua Wang1, Biyun Wang1, Jian Zhang1, Yuee Teng4, Zhongsheng Tong5, Yueyin Pan6,
Yongmei Yin7, Changping Wu8, Zefei Jiang9, Xiaojia Wang10, Guyin Lou11, Donggeng Liu12, Jifeng Feng13, Jianfeng Luo14, Jiong
Wu1, Zhimin Shao1 and Joseph Ragaz15. 1Fudan University Shanghai Cancer Center, Shanghai Medical College, Fudan
University, Shanghai, China; 2Cancer Hospital, Chinese Academy of Medical Sciences, Beijing, China; 3The Affiliated Tumor
Hospital of Harbin Medical University, Harbin, Heilongjiang, China; 4The First Hospital of China Medical University, Shenyang,
Liaoning, China; 5Tianjin Medical University Cancer Institute and Hospital, Tianjin, China; 6The First Hospital, Anhui Medical
University, Hefei, Anhui, China; 7The First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, China; 8The Third
Affiliated Hospital of Soochow University, Changzhou, Jiangsu, China; 9Beijing 307 Hospital of PLA, Beijing, China; 10Zhejiang
Cancer Hospital, Hangzhou, Zhejiang, China; 11Breast Cancer Center, RuijinHosiptal, Shanghai Jiaotong University School of
Medicine, Shanghai, China; 12Cancer Center, Sun Yat-sen University, Guangzhou, Guangdong, China; 13Jiangsu Cancer Hospital,
Nanjing, Jiangsu, China; 14School of Public Health, Fudan University, Shanghai, China and 15University of B.C., School of
Population & Public Health, Faculty of Medicine, Vancouver, BC, Canada.
Body: Background: There is still no standard chemotherapy for patients with metastatic triple-negative breast cancer (mTNBC).
Our previous phase II pilot trial with first-line gemcitabine and cisplatin combination (GP) in patients with mTNBC (clinicaltrials.gov
Identifier: NCT00601159) showed a median progression-free survival (PFS) of 6.2 months.
In this Chinese Breast Cancer Study Group (CBCSG) 006 trial (clinicaltrials.gov Identifier: NCT01287624) we explored in a
randomized trial the role of the less costly GP regimen versus the standard GT [Gemcitibine + paclitaxel] chemotherapy for the
metastatic breast cancer as a first line treatment for mTNBC.
Trial objectives: progression free survival [PFS]; overall survival [OS]; and toxicity.
Methods: In the trial with a hybrid trial design incorporating a formal test of superiority as well as noninferiority, mTNBC patients
with no previous chemotherapy for metastatic disease were randomly assigned to receive either GP regimen (G/P: 1250 mg/m2
d1,8/ 75 mg/m2 d1) or the GT regimen (same G; T: 175 mg/m2 d1).
Results: Between Jan. 2011 and Nov., 2013, 236 patients were randomized [118 patients / arm], and all received at least one
dose of assigned chemotherapy. As of Mar. 20, 2014, the intent-to-treat analysis showed 201 recurrences and 97 deaths.
Objective response rates of GP vs GT were 67.9% vs. 50.4% (P= 0.008), with median PFS of 232 vs. 194 days (HR=0.692, 95%
CI 0.523-0.915; P= 0.009). Overall survival of patients from the GP vs. the GT arms was median 672 vs. 556 days (HR=0.902,
95% CI 0.605-1.344; P= 0.611).
Significant differences in grade 3/4 adverse events were seen for nausea, vomiting, anemia and thrombocytopenia [GP vs. GT,
6.8 vs. 0.8%; 11.0 vs. 0.8%; 33.1 vs. 51.0%; and 32.2 vs. 2.5%, respectively].
In addition, assessment of adverse events of any grade showed the GP regimen had more anorexia, constipation,
hypomagnesemia and hypokalemia, while GT regimen had significantly more alopecia and peripheral neuropathy.
The delivered relative dose intensity was > 90% for all three drugs, with the total number of delivered cycles of chemotherapy in
GP and GT arms being 654 and 648 [average 5.54 and 5.49 /patient], respectively.
Conclusions:
1.The Gemcitabine + Platinum is superior to Gemcitabine + Paclitaxel in terms of objective response rates and duration of PFS.
2.While grade 3 / 4 nausea & vomiting, and anemia, were heavier for the GP combination, the delivery of chemotherapy and
average number of cycles delivered were comparable between the two arms.
3.Overall survival data will be updated on the conference to indicate the long-term effect of the somehow more toxic GP regimen,
which shows nevertheless superiority of response rates and of the PFS over the more costly GT regimen.

Title: Etirinotecan pegol target-specific pharmacodynamic biomarkers in circulating tumor cells from patients with metastatic
breast cancer in the phase 3 BEACON study
Edith A Perez1, Katie Caygill2, Alison L Hannah3, Javier Cortes4, Ahmad Awada5, Joyce O Shaughnessy6, Christopher
Twelves7, Hope Rugo8, Seock-Ah Im9, Kenna Anderes10, Darren W Davis10 and Ute Hoch2. 1Mayo Clinic, Jacksonville, FL; 2Nektar
Therapeutics, San Francisco, CA; 3Consultant, Sebastopol, CA; 4Vall D'Hebron University Hospital, Barcelona, Spain; 5Jules
Bordet Institute, Brussels, Belgium; 6Baylor Sammons Cancer Center, US Oncology, Dallas, TX; 7University of Leeds, St James's
University Hospital, Leeds, United Kingdom; 8University of California, San Francisco, CA; 9Seoul National University Hospital,
Seoul, Korea and 10ApoCell, Houston, TX.
Body: Background: Etirinotecan pegol (EP) is the first long-acting topoisomerase 1-inhibitor providing sustained level of active
metabolite that concentrate in the tumor for an entire chemotherapy cycle. In a Phase 2 study (Lancet Oncology 2013), EP
demonstrated a 29% overall response rate in patients with metastatic breast cancer, leading to the Phase 3 global BEACON
study in this population. Detection of circulating tumor cells (CTCs) using immunomagnetic EpCAM-based methods have been
conceptually accepted as a "liquid tumor biopsy". To circumvent the limited recovery of CTCs these methods provide for
molecular profiling applications, we isolated CTCs based on an antibody-independent, continuous flow dielectrophoresis (DEP)
technology (ApoStream). We developed quantitative multiplex immunofluorescent assays for target-specific pharmacodynamics
(PD) biomarkers for EP in CTCs isolated pre- and post-treatment. Here we present the distribution of the PD biomarkers in CTC
samples collected from BEACON patients.
Methods: Donation of blood samples for CTC analysis was voluntary. BEACON patients participating in the substudy had serial
(pre-dose, Cycle 2 Day 1, Cycle 4 Day 1, End of Treatment) 7.5-mL whole blood samples drawn in EDTA tubes and shipped
ambient to ApoCell (Houston, TX) within 96 hrs for processing. PBMCs were separated by Ficoll gradient, and CTCs were
isolated using ApoStream technology. Isolated cells were deposited on three slides and stained for DAPI, CD45, and cytokeratin
markers, as well as 2-3 of the seven PD markers, and analyzed by iCys laser scanning cytometer equipped with image analysis
software. CTC count, mean fluorescence intensity (MFI) of PD biomarkers, and % marker-positive cells were analyzed for their
distribution.
Results: 75% of the 852 BEACON patients participated in the CTC substudy, yielding 649 pre-treatment CTC samples. As of
Mar14, 538 Cycle 2 Day 1, 281 Cycle 4 Day 1, and 394 End of Treatment samples were collected. 98% of pre-treatment blood
samples were successfully processed. 97% had detectable CTCs, with a median of 504 CTCs (interquartile range: 128-1432). PD
markers could be analyzed in 83-92% of samples after implementing a minimum requirement of 10 CTCs analyzed. Staining was
positive for Top1, Top2, H2Ax, Rad51, ABCG2, Ki67, and TUNEL in 82%, 71%, 14%, 30%, 20%, 56%, and 94% of samples,
respectively. Percent of marker-positive cells varied from 0.2-69%, with 6- to 37-fold interquartile ranges. Median MFI in these
cells varied between 25x103 and 100x103, with 1.2- to 1.6-fold interquartile ranges.
Conclusions: CTC collection was successfully incorporated into the BEACON study, with 75% patient participation. CTC
detection rate using ApoStream was high, permitting evaluation of biomarkers at baseline and over time. EP target-specific PD
biomarkers can be measured in CTCs isolated from patients participating in BEACON and will be assessed for their potential to
predict clinical response. Final BEACON efficacy and safety results are expected in early 2015, which will allow further analysis of
CTCs with patient outcome (response, PFS, OS) and change in CTCs at time of progression.

Title: A open-label, randomized, parallel, phase III trial to evaluate the efficacy and safety of Genexol-PM compared to
Genexol(conventional paclitaxel with cremorphor EL) in recurrent or metastatic breast cancer patients
Jung Sil Ro1, Joo Hyuk Sohn2, Sung Bae Kim3, Keun Seok Lee1, Joo Seop Chung4, Jae Hoo Park5, Soo Hyeon Lee2, Tae You
Kim6, Kyung Hae Jung3, Eun Kyung Cho7, Yang Soo Kim8, Hong Suk Song9, Jae Hong Seo10, Hun Mo Ryoo11, Sun Ah Lee12, So
Young Yoon13, Chul Soo Kim14, Yong Tai Kim15, Hwa Jung Sung16, Si Young Kim17 and Yong Jin Lee18. 1National Cancer Center,
Goyang, Gyeonggi-do, Korea; 2Severance Hospital, Seoul, Korea; 3Asan Medical Center, Seoul, Korea; 4Pusan National
University Hospital, Busan, Korea; 5Ulsan University Hospital, Ulsan, Korea; 6Seoul National University Hospital, Seoul, Korea;
7
Gachon University Gil Medical Center, Incheon, Korea; 8Kosin University Gospel Hospital, Busan, Korea; 9Keimyung University
Dongsan Medical Center, Daegu, Korea; 10Korea University Guro Hospital, Seoul, Korea; 11Daegu Catholic University Medical
Center, Daegu, Korea; 12Daegu Fatima Hospital, Daegu, Korea; 13Konkuk University Medical Center, Seoul, Korea; 14Inha
University Hospital, Incheon, Korea; 15National Health Insurance Service Ilsan Hospital, Goyang, Gyeonggi-do, Korea; 16Korea
University Ansan Hospital, Ansan, Gyeonggi-do, Korea; 17Kyung Hee University Medical Center, Seoul, Korea and 18Samyang
Biopharmaceuticals Corporation, Seoul, Korea.
Body: Background: The rationale for developing an alternative paclitaxel formulation concerns Cremophor EL-related side
effects, and a novel paclitaxel delivery system might augment its therapeutic efficacy. Genexol-PM is a novel polymeric micelle
formulated paclitaxel free of Cremophor. The polymeric micelle formulation is composed of hundreds of low molecular weight,
non-toxic, and biodegradable amphiphilic diblock copolymers which include monomethoxy poly (ethylene glycol)-block-poly
(D,L-lactide). This multicenter phase III study was designed to evaluate the non-inferiority of efficacy of Genexol-PM compared to
conventional CrEL-based paclitaxel.
Methods: In this phase III study, 213 patients were enrolled onto the study and randomly assigned (1:1) to treatment group
according to prior recurrent or metastatic breast cancer chemotherapy. The study evaluated the objective response rate for the
primary objective, and others including overall survival (OS), progression free survival (PFS), time to tumor progression (TTP),
duration of overall response and adverse events. Eligible patients were randomly assigned to receive either Genexol-PM or
standard paclitaxel. Genexol-PM or standard paclitaxel was administered on the first day of every 3 weeks. Measurable disease
was assessed by imaging using the RECIST 1.0 criteria.
Results: The objective response rate (ORR) was higher by the administration of the study drug (39.05% v 24.30% in ITT, 56.92%
v 39.29% in PP). One-sided 95% upper confidence limit was -4.36%, which is lower than the non-inferiority threshold (7%),
indicating that the study group is not inferior to the control group. OS, PFS, TTP and duration of overall response were analyzed
in the ITT population. The analysis of OS showed no significant difference (p=0.5878) (859 days, 95% CI : 732.001,025.00 v 726
days, 95% CI : 553.00 -). Median PFS periods were 232 days (95% CI: 164.00274.00) vs. 191 days (95% CI: 159.00237.00).
Median TTPs were 233 days (95% CI: 165.00286.00) vs. 191 days (95% CI: 159.00241.00) between the groups. Difference in
PFSs and TTPs between the groups were not statistically significant. (p=0.2407, 0.2076, respectively) Genexol-PM was not
significantly different from the comparator in terms of safety.
Conclusion: Genexol-PM demonstrated non-inferior efficacy and comparable safety profile compared with standard paclitaxel in
this patient population. Of note, Genexol-PM permits the delivery of a higher paclitaxel dose without additional toxicity achieved
with CrEL-based formulation. In the absence of CrEL, no filter or special tubing is required that conventional PVC infusion sets
can be used.

Title: Phase III trial evaluating paclitaxel plus carboplatin versus paclitaxel plus epirubicin as first-line treatment for metastatic
breast cancer
Zhongsheng Tong1, Shufen Li1, Yehui Shi1, Xu Wang1, Chunfang Hao1, Lihong He1, Guolei Dong1, Xiaorui Wang1, Yongsheng Jia1
and Li Zhang1. 1Tianjin Medical University Cancer Institute and Hospital, Tianjin, China.
Body: Background: To compare survival and side-effects between patients with metastatic breast cancer (MBC) treated with
paclitaxel/carboplatin (TP) or paclitaxel/epirubicin (TE) chemotherapy.
Patients and methods: From June 2009 to April 2012, 121patients were randomly assigned, 61 of whom were randomized to TP
and 60 to TE. Baseline characteristics were relatively well-balanced in the two treatment arms.
Table 1 Baseline patient characteristics
Characteristic
TP(n=61)
TE(n=60)
P(2 test)
Median age,years(range)
53.5(35-68)
51.8(32-69)
0.614
ECOG performance status <2
61(100%)
60(100%)
Node positive at first diagnosis
27(44.3%)
25(41.7%)
0.773
ER/PR-positive
42(68.9%)
43(71.7%)
0.735
HER2-positive
18(29.5%)
15(25.0%)
0.578
33(58.1%)
35(55.2%)
0.639
Visceral
40(65.6%)
42(70.0%)
0.602
Lymph
25(41.0%)
28 (46.7%)
0.529
Lobular
12(19.7%)
11(18.3%)
0.851
Ductal
44(72.1%)
42(70.0%)
0.796
Other
5(8.2)
7(11.7%)
0.523
Yes
28(45.9%)
25(41.7%)
0.639
No
33(54.1%)
35(58.3%)
Yes
13(21.3%)
11 (18.3%)
No
48(78.7%)
49(81.7%)
Metastatic sites
Histology
Adjuvant endocrine therapy
Adjuvant radiotherapy
0.681
Results: After a median follow-up of 21.3 months, there was no significant difference between the two treatment arms in objective
response rate(ORR) (68.5 % vs. 73.3%, respectively;). And both the progression-free survival (P=0.704)and overall survival
(P=0.403)were very similar between the two arms. Both regimens were well tolerated. The main toxicities were
myelosuppression, gastrointestinal reactions and alopecia. There was more grades 34 alopecia and more nausea with
TE(p<0.05). For the QLQ-C30 questionnaire statistically significant changes after treatment were fatigue(0.031) in TP,
nausea/vomiting(p=0.041) and lose of appetite(p=0.048) in TE.
Table 2 Summary of adverse events per patient; p-values are given overall
TP (n=61)
TE (n=60)
P value
Adverse events
Grades 1-4
Grades 3-4
Grades 1-4
Grades 3-4
Neutropenia
46
13
38
0.149
Leukopenis
50
11
40
0.054
Anemia
0.981
Thrombocytopenia
0.714
Lymphocytopenia
0.391
Vomiting
15
20
0.289
Nausea
27
41
0.010
Diarrhea
0.714
Alopecia
44
15
60
31
0.000
Cardiotoxicity
0.523
Hypersensitivity reaction
ALT increase
0.523
Peripheral neurotoxicity
20
19
0.895
Overall p value for the comparison of grades 1-4 are given

Conclusion: This study did not found significant differences in ORR, PFS and OS between TP and TE arms. Both arms were well
tolerated. TP has some advantages in less side-effects, especially alopecia and nausea.This combination can be safely
prescribed to patients in which anthracyclines have the potential of being harmful such as those previously exposed to
anthracyclines in the adjuvant setting.

Title: Real-world efficacy and safety outcomes of nab-paclitaxel (nab-P) in patients (pts) with metastatic breast cancer (MBC):
Results from a US health insurance database
Debra A Patt1, Caihua Liang2, Ling Li2, Amy Ko3, Cindy Duval Fraser3, Deyanira Corzo3 and Cheryl Enger4. 1McKesson Specialty
Health/US Oncology, Houston, TX; 2Optum Epidemiology, Waltham, MA; 3Celgene Corporation, Summit, NJ and 4Optum
Epidemiology, Ann Arbor, MI.
Body: Background: nab-P, an albumin-bound formulation of paclitaxel, was approved in 2005 for the treatment of breast cancer
after failure of combination chemotherapy for metastatic disease or relapse within 6 mo of adjuvant chemotherapy based on
clinical trials. nab-P demonstrated efficacy and safety when administered weekly in phase II trials and every 3 weeks (q3w) in an
international phase III trial. Little is known about the treatment patterns and outcomes of nab-P in the real-world setting. Using
health insurance claims data, this study was conducted to characterize efficacy and safety of nab-P in pts with MBC treated in US
clinical practices.
Methods: This retrospective claims analysis used data in the Optum Research Database (United Health affiliate). Data were
supplemented by Social Security Death Index sources. The analysis included women aged 18 y with MBC diagnosis ( 2 claims
of BC diagnosis separated by 30 d and 2 claims of metastatic spread) prior to nab-P initiation. Pts had 6 mo of continuous
enrollment in the health plan from January 2005-September 2012, complete medical coverage and pharmacy benefits, no other
primary malignancy, and no prior chemotherapy. Cohorts were determined by line of therapy, nab-P regimen, and schedule.
Endpoints included treatment patterns, time to next therapy or death (TNTD), overall survival (OS), and safety.
Results: Of the 664 eligible pts, most were between 40-69 y of age (88%) and had received nab-P as second-line therapy
(74%), monotherapy (61%), and weekly dosing (71%). In combination, nab-P was most often given with bevacizumab (58%) or
human epidermal growth factor receptor 2 (HER2)targeted therapy (24%) vs another cytotoxic agent (19%). Median TNTD and
OS were 6.1 and 17.4 mo, respectively. By line of therapy (first, second, and third), TNTD was 7.1, 6.6, and 5.3 mo, and OS
was 22.7, 17.4, and 15.1 mo. The OS data are comparable with published clinical trial results (Table). In a subgroup of pts (n =
400) with aggressive disease features ( 50 y of age or having 3 metastases), median OS was 15.6 mo. These data are
comparable with a retrospective analysis of pts with visceral dominant metastasis (VDM) or a short disease-free interval (SDFI;
Table). Toxicities reported in healthcare claims were consistent with those previously published.
Conclusions: Consistent with clinical trial data, outcomes of this analysis demonstrated the efficacy and safety of nab-P across
lines of therapy in a real-world population of patients with MBC.
Clinical Trial Experience in MBC for Pts Treated With nab-P
nab-P dose (mg/m2) and schedule
Trial
Median OS, mo
CA0121 (Ph III)

ITT (all lines)
260 q3w
229
15.0
second line
260 q3w
131
13.0
300 q3w
76
27.7
100 qw 3/4
76
22.2
150 qw 3/4
74
33.8
CA0242 (Ph II, first line)
Median OS, mo
Subgroups With Aggressive Disease Features
VDM
SDFI
CA0123
(first line)
260 q3w
74
15.1
42
14.6
CA0243
(first line)
300 q3w
61
27.7
20
16.6
100 qw 3/4
60
19.6
21
19.1
150 qw 3/4
qw 3/4, weekly for the first 3 of 4 weeks.
1
Gradishar WJ, et al. J Clin Oncol. 2005;23:7794-7803.

Gradishar WJ, et al. Clin Breast Cancer. 2012;12:313-321.
3 OShaughnessy J, et al. Breast Cancer Res Treat. 2013;138:829-837.
2
59
32.1
14
18.6

Title: Phase I trial of eribulin in combination with S-1 for advanced and recurrent breast cancer patients
Tsutomu Iwasa1, Junji Tsurutani1, Satomi Nishida1, Yoshifumi Komoike2 and Kazuhiko Nakagawa1. 1Kinki University Faculty of
Medicine, Osaka-Sayama, Osaka, Japan and 2Kinki University Faculty of Medicine, Osaka-Sayama, Osaka, Japan.
Body: Introduction: Eribulin is a non-taxane microtubule dynamics inhibitor that has been proved to prolong overall survival in
patients with advanced and recurrent breast cancer compared to treatment of physicians choice. S-1 is an oral fluoropyrimidine
anticancer agent that combines tegafur as the effector drug with two modulators, gimeracil, and oteracil potassium, and has
shown promising antitumor activity as a single agent for the treatment of advanced breast cancer. In the basis of these results, we
aimed to assess the efficacy of the combination of these two drugs in patients with advanced and recurrent breast cancer. In this
trial, we determined maximum tolerated dose (MTD) and phase II recommended dose of Eribulin and S-1 combination therapy.
Methods: A traditional 3+3 dose escalation design was implemented. 12 patients pre-treated with anthracycline and taxane were
enrolled. As an initiation level, patients received s-1 50 mg/m2 from day1 to day14, and Eribulin 1.1mg/m2 on day1, and day8. In
level 2, Eribulin dose was increased to 1.4mg/m2. In level 3, S-1 dose was increased to 80 mg/m2.
Results: In level 1 or 2, no dose limiting toxicity (DLT) was observed. In level 3, grade 3 hypokalemia occurred in 1 case. The
level 3 dose level was determined as phase II recommended dose. Neutropenia was observed in all cases, 3 of which was grade
3 febrile neutropenia. However, other non-hematological toxicity was mild, and the toxicity of the both drugs
Safety (N=12)
Adverse Event
Grade 3/4
Leukopenia
3 (25%)
Neutropenia
10 (83%)
Febrile Neutropenia
3 (25%)
Hypokalemia
1 (8%)
was not enhanced by the combination. Partial response

Efficacy (N=12)
CR
0 (0%)
PR
5 (41%)
SD24 weeks
2 (17%)
SD<24 weeks
3 (25%)
PD
2 (17%)
Clinical Benefit
7 (58%)
Disease Control Rate
10 (83%)
is confirmed in 5/11 patients with measurable lesions.

Conclusion: We have shown that treatment with Eribulin and S-1 is active and well tolerated in patients with advanced and
recurrent breast cancer, suggesting that this combination therapy may have potential as a new treatment option. A phase II study
is being conducted to evaluate the efficacy and safety.

Title: No Show

Title: A Phase II trial of doxil, carboplatin and bevacizumab in metastatic triple negative breast cancer and molecular correlates of
response
Kim M Hirshfield1, Shou-En Lu1, Serena Wong1, Antoinette Tan1, Laurie Kirstein1, Thomas Kearney1, Weichung Shih1, Shridar
Ganesan1 and Deborah L Toppmeyer1. 1Rutgers Cancer Institute of New Jersey, New Brunswick, NJ.
Body: Background. Despite improved outcomes for other breast cancer subtypes, triple negative breast cancer continues to
display the worst prognosis with conventional, standard of care chemotherapy. A growing body of evidence suggests that
platinating agents may offer superior outcomes in a subset of triple negative breast cancers and that genomic alterations may
identify those tumors most likely to respond.
Patients and Methods. Eligibility included previously untreated, stage IV ER, PgR, and Her2neu negative breast cancer with
measurable disease. Thirty-one patients with ECOG PS2 and adequate bone marrow, renal and hepatic function were enrolled
to receive doxil (30 mg/m2), carboplatin (AUC 5) and bevacizumab (15 mg/m2) every 4 weeks in an open-label, multicenter,
investigator-initiated Phase II trial from 2008-2012. The primary endpoint was median progression free survival (PFS). Secondary
endpoints were response rate based on RECIST criteria, survival time, and toxicity profile.
Results. Thirty-one patients received a median of 5.6 cycles (range 1-13). Prior adjuvant or neoadjuvant anthracycline or taxane
was given in 38.7% or 41.9%, respectively. Clinical benefit rate (CBR= CR+PR+SD 6months) was 38.7% where 9/18 (50%)
with CBR<6 months and 4/13 (30.7%) with CBR>6 months had prior taxane. Median progression free survival (PFS) was 5.6
months, 95% CI [4.4-6.9] and 6-month PFS rate was 41.9%, 95% CI [24.6-59.3]. Median overall survival was 11.9 months, 95%
CI [8.8-21.8] and 1-year survival rate was 47.3%, 95% CI [29.5-65.1]. Thirteen patients were alive at 10 months or longer. Grade
3 non-hematologic and hematologic toxicities included fatigue (n=1), hand-foot skin reaction (n=1), and neutropenia (n=1). Grade
4 hematologic toxicity included thrombocytopenia (n=1). Grade 4 hypertension (n=2) and thrombosis (n=1) were attributable to
bevacizumab and this drug was discontinued from the protocol. Neither significant change in cardiac function nor alopecia were
observed. Next-generation sequencing from tumors revealed p53 alterations as the most common alteration. For three patients
with serial pre- and post-platinum specimens, gain of genomic alterations were observed at time of progression of disease
occurring at 5 months, 8 months, and 20 months (eight months of these were off study but receiving doxil and carboplatin).
Conclusions. Results demonstrate that the combination of doxil, carboplatin, and bevacizumab is an active and well-tolerated
regimen in previously untreated, metastatic, triple negative breast cancer. We anticipated presentation of additional genomic
tumor profiling results that may yield insights into markers of sensitivity and mechanism of resistance to anthracycline and/or
platinum.

Title: Neoadjuvant chemotherapy with or without bevacizumab or everolimus: Survival analysis of The HER2-negative cohort of
the GEPARQUINTO study (GBG 44)
Bernd Gerber1, Sibylle Loibl2, Michael Untch3, Holger Eidtmann4, Mahdi Rezai5, Peter A Fasching6, Hans Tesch7, Holm
Eggemann8, Iris Schrader9, Kornelia Kittel10, Claus Hanusch11, Jens Huober12, Christine Solbach13, Christian Jackisch14, Georg
Kunz15, Jens-Uwe Blohmer16, Maik Hauschild17, Tanja Fehm18, Valentina Nekljudova2 and Gunter von Minckwitz2,13.
1
Unifrauenklinik Rostock; 2German Breast Group, Neu-Isenburg, Hessen, Germany; 3Helios Klinikum Berlin-Buch; 4Unifrauenklinik
Kiel; 5Luisenkrankenhaus Dsseldorf; 6Unifrauenklinik Erlangen; 7Chop GmbH Frankfurt; 8Unifrauenklinik Magdeburg;
9
Henriettenstiftung Hannover; 10Praxisklinik Berlin; 11Rotkreuzklinikum Mnchen; 12Unifrauenklinik Ulm; 13Unifrauenklinik
Frankfurt/Mail; 14Sana Klinikum Offenbach; 15St Johannes Hospital Dortmund; 16St Gertrauden Berlin; 17Frauenklinik Rheinfelden
and 18Unifrauenklinik Duesseldorf.
Body: BACKGROUND: The GeparQuinto study showed that adding bevacizumab to 24 weeks of anthracycline-taxane-based
neoadjuvant chemotherapy increases pathological complete response (pCR) rates from 14.9% to 18.4% (P=0.04) overall;
specifically in patients with TNBC (27.9% to 39.3% (P=0.003) (von Minckwitz et al, NEJM 2012). No difference in pCR rate was
observed for adding everolimus to paclitaxel patients who had no early response to neoadjuvant chemotherapy (Huober et al, Eur
J Cancer 2013). Here, we present disease-free (DFS) and overall survival (OS) analyses.
PATIENTS AND METHODS: Patients (n = 1948) with HER2-negative tumors of a median tumor size of 4 cm were randomly
assigned to neoadjuvant treatment with 4xEC 4x docetaxel with or without bevacizumab, 15 mg/Kg q3w before surgery. 408
patients not clinically responding to EC Bev were randomized to 12x weekly paclitaxel with or without everolimus 5mg/day.
Patients with HR-positive tumors received endocrine treatment after surgery. 379 events are required to show a HR of 0.75
(=0.05, =0.8) between the bevacizumab arms. 397 relapses and 234 deaths were observed after a median follow up of 3.8
years overall, of those 115 relapses and 75 deaths occurred in the non-responding cohort.
RESULTS: Overall, 3-year DFS was 80.8% and 3-year OS was 89.7%. Outcome was not different for patients receiving
bevacizumab (HR 1.03; P = 0.780 for DFS and HR 0.974; P = 0.842 for OS) compared to patients receiving chemotherapy alone.
Patients with TNBC showed no improvement in DFS (HR 0.991; P = 0.948) and OS (HR 1.02; P = 0.891) when treated with
bevacizumab. No other predefined subgroup (HR+/HER2-; locally advanced (cT4 or cN3) or not; cT1-3, cT4a-c, cT4d; pCR or
not, CR, PR, NC after first 4 cycles chemotherapy) showed a benefit from bevacizumab. No difference in DFS (HR 0.997;
P=0.987) and OS (HR 1.11; P=0.658) was observed for patients who had no early response to neoadjuvant chemotherapy
receiving paclitaxel with or without everolimus overall as well as in subgroups.
CONCLUSIONS: Long-term results finally do not support the neoadjuvant use of bevacizumab in addition to an
anthracycline-taxane-based chemotherapy or everolimus in addition to paclitaxel for patients who had no early response to

Title: Predictors of complete pathological response to neoadjuvant chemotherapy for breast cancer: 19,310 cases from the
national cancer database treated in 2010 and 2011
Vicky Q Yang1, Gang Han1, Sarah S Mougalian2, Brigid K Killelea2, Nina Horowitz2, Anees B Chagpar2, Brandon Hayse2, Lajos
Pusztai2 and Donald R Lannin2. 1Yale School of Public Health, New Haven, CT and 2Yale University School of Medicine, New
Haven, CT.
Body: Background: Complete pathological response following neoadjuvant chemotherapy for breast cancer indicates an
excellent prognosis and may influence subsequent local or systemic treatment. However it is not clear which patients should
undergo neoadjuvant chemotherapy.
Methods: The National Cancer Database is a joint project of the Commission on Cancer of the American College of Surgeons
and the American Cancer Society and contains about 80% of the cancer cases in the United States. Patients undergoing
neoadjuvant chemotherapy for breast cancer in 2010 and 2011 were identified and the pathological response was compared
across multiple demographic and tumor associated variables. Complete pathological response (pCR) was defined as no invasive
cancer remaining in the breast or lymph nodes at the time of subsequent surgery.
Results: Out of 334,447 cases of breast cancer diagnosed in 2010 and 2011, 29,534 (8.8%) underwent neoadjuvant
chemotherapy. This included 2052 (6.9%) who also received neoadjuvant hormonal therapy and 616 (2.1%) who also received
neoadjuvant radiation therapy. The pathological response was known for 19,310 (65%) and 6244 (32%) had a complete response
(pCR), 11,522 (60%) had a partial response, and 1544 (8%) had no response. In logistic regression analysis, demographic and
tumor related variables were significant (p<0.05) in both univariable and multivariable models. Estimated odds ratio of having
pCR and p-values of demographic and tumor related variables are shown in Table 1, where a multivariable logistic regression
was used to fit the data of 15686 patients.
Conclusions: These data can be used to construct a nomogram that estimates the chance of pCR based on patient age, tumor
histology, grade, molecular type, clinical T stage, and clinical N stage. This may be useful to help select suitable candidates for
Predictors of Complete Pathological Response : Multivariate Logistic Regression
N
OR
95% CI
p value
> 50
8500
Reference
< 50
7186
1.27
Ductal
12,926
Reference
Lobular
837
0.65
0.52 - 0.80
<0.001
Mixed ductal/lobular
577
0.71
0.57 - 0.89
0.003
Other
1346
0.82
0.72 - 0.94
0.003
872
Reference
5222
1.09
0.90 - 1.33
0.363
9592
1.76
1.45 - 2.13
<0.001
ER/PR+, Her2-
6738
Reference
ER/PR+, Her2+
2662
2.62
2.37 - 2.90
<0.001
ER/PR-, Her2+
1757
4.03
3.58 - 4.53
<0.001
Age
<0.001
1.19 - 1.37
<0.001
Histology
<0.001
Grade
<0.001
Molecular type
<0.001
ER/PR-, Her2-
4529
2.32
2.12 - 2.55
<0.001
T1
2147
Reference
T2
7167
0.71
0.64 - 0.79
<0.001
T3
3513
0.48
0.43 - 0.54
<0.001
T4
2859
0.50
0.44 - 0.57
<0.001
N0
5713
Reference
N1
7307
0.77
0.72 - 0.84
<0.001
N2
1646
0.68
0.60 - 0.78
<0.001
N3
1020
0.70
0.60 - 0.82
<0.001
Clinical T stage
<0.001
Clinical N stage
<0.001

Title: A randomized, phase 2 trial of preoperative MM-121 with paclitaxel in triple negative (TN) and hormone receptor (HR)
positive, HER2-negative breast cancer
Frankie A Holmes1, Kristi J McIntyre2, Ian E Krop3, Cynthia R Osborne4, John W Smith II5, Manuel R Modiano6, Manish Gupta7,
Leona B Downey8, Rita Nanda9, Mansoor N Saleh10, Jonathan R Young11, Kerry E Horgan11, William Kubasek11, Gavin
MacBeath11, Michael A Danso12 and Joyce A O'Shaughnessy4. 1Texas Oncology, Memorial City, Houston, TX; 2Texas Oncology,
Dallas Presbyterian Hospital, Dallas, TX; 3Dana-Farber Cancer Institute, Boston, MA; 4Texas Oncology, Baylor Charles A
Sammons Cancer Center, Dallas, TX; 5Northwest Cancer Specialists, P.C., Portland, OR; 6ACRC/Arizona Clinical Research
Center, Tuscon, AZ; 7Texas Onocology, Garland, Garland, TX; 8Arizona Oncology Associates, Tucson, AZ; 9University of Chicago
Medicine, Chicago, IL; 10Georgia Cancer Specialists, Atlanta, GA; 11Merrimack Pharmaceuticals, Inc, Cambridge, MA and
12
Virginia Oncology Associates, Norfolk, VA.
Body: Background: MM-121 is a fully human monoclonal antibody that targets ErbB3 and blocks heregulin (HRG), the principal
ligand for ErbB3, from binding. HRG-driven ErbB3 signaling has been widely implicated in the development of resistance to
conventional chemotherapies and targeted agents. Recently, as part of a broader Phase 2 program designed to identify
responders to MM-121, HRG was identified as a key biomarker that correlates with poor response to standard-of-care therapies
and benefit from co-treatment with MM-121 in the metastatic setting in patients with NSCLC, ovarian cancer and breast cancer.
Here we present data on the impact of HRG and MM-121 co-treatment in the pre-operative setting in HER2- breast cancer.
Methods: Patients enrolled had locally advanced HR+, HER2-negative invasive breast cancer (Group 1) or TN (Group 2), no
distant metastatic disease, no prior treatment for the disease under study, and T2. Patients were randomized in a 2:1 fashion to
receive MM-121 plus paclitaxel or paclitaxel alone followed by doxorubicin and cyclophosphamide, followed by surgery. The
primary endpoint of this study was to describe pCR rates, defined as the absence of invasive cancer in the breast and lymph
nodes (i.e. ypT0, ypN0). Residual cancer burden index (RCBI) was also recorded for each patient. Pre- and post-treatment tumor
biopsies were collected in order to assess the levels of HRG and other potential biomarkers.
Results: 200 patients (101 HR+, 99 TN) entered the study. For Group 1 (HR+), 96 of the 101 patients were Efficacy Evaluable
(EE) and the pCR rate in the MM-121 treatment arm was 7/66 (10.6%, 95% CI [5.2%, 20.3%]) compared to 1/30 (3.3%, 95% CI
[0.6%, 16.7%]) on the control arm. For Group 2 (TN), 85 of the 99 patients were Efficacy Evaluable (EE) and the pCR rate in the
MM-121 treatment arm was 23/56 (41.1%, 95% CI [29.2%, 54.1%]) compared to 14/29 (48.3%, 95% CI [31.4%, 65.6%]) on the
control arm. Preliminary analysis of pretreatment biopsies (52% of samples analyzed) suggests a potential link between HRG
levels and both pCR rates and RCBI. For both groups, the overall incidence of adverse events, regardless of relationship, was
comparable between the two arms. A higher frequency of any grade AEs (Treatment vs. Control) was reported for diarrhea, rash,
stomatitis, fatigue, epistaxis, nail disorder and dysgeusia. A higher frequency of Grade 3 or higher AEs (Treatment vs. Control)
was reported for febrile neutropenia (HR+: 10.4% vs. 3.0%), diarrhea (HR+: 7.5% vs. 0%, TN: 7.8% vs. 0%), fatigue (HR+: 6.0%
vs. 3.0%), anemia (HR+: 7.5% vs. 3.0%, TN: 7.8% vs. 3.1%), hypokalemia (HR+: 7.5% vs. 3.0%), infusion-related reactions (TN:
4.7% vs. 0%), pulmonary embolisms (TN: 4.7% vs. 0%), and hyperglycemia (TN: 4.7% vs. 0%).
Conclusion: Although a potential signal of benefit from MM-121 was observed in the HR+ group, the same was not observed in
the TN group, and overall the results are inconclusive. Biomarker analyses, including pharmacodynamics studies, are ongoing.
The observed safety profile is consistent with the expected toxicities associated with ErbB inhibitors and weekly paclitaxel, and
did not impact exposure or compliance on treatment.

Title: Effects of neoadjuvant chemotherapy with or without zoledronic acid on pathological response: A meta-analysis of
randomised trials
Judith R Kroep1, Ayoub Charehbili1, Rob E Coleman2, Rebecca L Aft3, Yoshie Hasegawa4, Gerrit-Jan Liefers1, Matthew C Winter2,
Katherine N Weilbaecher3, Kohei Akazawa5, Samantha Hinsley6, Hein Putter1, Hans WR Nortier1 and Norio Kohno7. 1Leiden
University Medical Center, Leiden, Netherlands; 2University of Sheffield, Sheffield, United Kingdom; 3Washington University
School of Medicine, St Louis, MO; 4Hirosaki Municipal Hospital, Aomori, Netherlands; 5Niigata University, Niigata, Japan;
6
University of Leeds, Leeds, United Kingdom and 7Kobe Kaisei Hospital, Kobe, Hyogo, Japan.
Body: Background:
The addition of bisphosphonates to adjuvant systemic therapy reduces bone metastases and improves survival in
postmenopausal patients with early breast cancer. We have conducted four randomised trials of neoadjuvant chemotherapy (CT)
+/- zoledronic (ZA) in stage II/III breast cancer to investigate the potential for enhancing pathological response within the breast
and axilla. We report here a meta-analysis of these studies to enable more reliable evaluation of treatment effects and investigate
activity in pre-defined subgroups of interest.
Methods:
Individual patient data from four prospective randomized clinical trials reporting the effect of the addition of ZA on pathological
response after neoadjuvant CT were pooled. Primary outcomes were pathological complete response in the breast (pCRb) and in
the breast and lymph nodes (pCR). A fixed effects Mantel-Haenszel meta-analysis was performed using odds ratios (OR) from
multivariate analyses in each study correcting for T-status and ER-status after testing for heterogeneity. Predefined subgroup
analyses were performed for postmenopausal women and patients with triple-negative breast cancer.
Results:
pCRb and pCR data were available in 736 and 553 women respectively. A summary of results is shown in the table below. In the
total study population, ZA addition to neoadjuvant CT did not increase pCRb/pCR rates (12.9% for CT only vs. 16.0% with CT+
ZA; OR 1.34, 95% C.I. 0.86-2.08). However, in postmenopausal women, the addition of ZA resulted in a doubling of the pCRb
rate (8.5% for CT only vs. 17.0% with CT+ ZA; OR 2.72, 95% C.I. 1.15-6.42) and pCR rate (7.8% for CT only vs. 13.6% with CT+
ZA; OR 2.48, 95% C.I. 0.80-7.69). There was significant interaction between menopausal status and ZA benefit (P for
interaction=0.047). In patients with triple-negative breast cancer a trend was observed favouring CT+ZA.
Pathological response
CT
CT + ZA
Odd's Ratio
95% C.I.
pCRb/total
Percentage
pCRb/total
Percentage
Total population
48/373
12.9
58/363
16.0
1.34
0.86-2.08
Postmenopausal patients
11/130
8.5
24/141
17.0
2.72
1.15-6.42*
Pre/perimenopausal patients
34/232
14.7
33/214
15.4
1.03
0.59-1.79
Triple-negative breast cancer
13/70
18.6
21/67
31.3
1.99
0.85-4.66
pCR/total
Percentage
pCR/total
Percentage
Total population
28/279
10.0
33/274
12.0
1.31
0.73-2.34
Postmenopausal patients
7/90
7.8
14/103
13.6
2.48
0.80-7.69
Pre/perimenopausal patients
19/178
10.7
18/163
11.0
1.00
0.49-2.08
Triple-negative breast cancer
9/52
17.3
15/51
29.4
1.80
0.63-5.21
pCR in breast
pCR in breast and lymph nodes
Overview of pCRb and pCR in both treatment arms in the total population and predefined subgroups
Conclusion:
This meta-analysis shows that no overall benefit from the addition of ZA to neoadjuvant CT. However, as has been seen in the
adjuvant setting, the addition of ZA to CT in postmenopausal women with early breast cancer appears to specifically increase the
effects of CT with increases in pCRb and pCR. Further translational research is warranted to explore the mechanism of these
observations and the potential treatment interaction with menopause.

Title: Neoadjuvant chemotherapy in invasive lobular breast carcinoma: Comparison of response, surgery and disease free
survival with invasive ductal carcinoma
Hilary L Martin2,3, Geraldine Walsh1, Komel Khabra1, Tabitha Skinner1 and Ian E Smith1. 1Royal Marsden Hospital, London, United
Kingdom; 2Royal Perth Hospital, Perth, WA, Australia and 3University of Western Australia, Perth, WA, Australia.
Body: Background
Lobular breast cancer (ILC) has been reported to be associated with a lower rate of pathologic complete response (pCR) and
breast conserving surgery following neoadjuvant chemotherapy compared with ductal carcinoma (IDC). ILC is predominantly
oestrogen receptor positive, HER-2-ve and histologic grade 1 or 2. Oestrogen receptor positive breast cancer has been shown to
have poorer responses to neoadjuvant chemotherapy. HER-2 positivity is associated with better responses to anthracycline
based neoadjuvant chemotherapy than HER-2 negative, and substantially improved pathological complete response (pCR) rates
relative to HER-2 negative disease with the use of HER-2 targeted agents in this patient group. This study investigates response
and disease free survival of ILC compared with IDC.
Methods
Patients with ILC and IDC treated with neoadjuvant chemotherapy were identified from the prospectively collated breast unit
database at the Royal Marsden Hospital. Demographic, tumour, treatment and outcome data was obtained from the database
and hospital electronic patient record system. Statistical analysis was undertaken comparing IDC and ILC. Clinical response
rates, pCR rates, breast conserving surgery rates, and disease free survival were analysed. Clinical response was defined as
those with partial response (50% reduction in the product of breast tumour size from baseline measurement), or complete
clinical response (no clinically palpable breast tumour). Subgroup analysis was performed on ER+ve, HER-2ve and HER-2
unknown cases.
Results
A total of 1017 patients were identified. Of these, there were 920 IDC and 97 ILC cases. Median age at commencement of
neoadjuvant chemotherapy was 47.9 years (range 23.6-74.9). Median follow-up was 7.6 years (range 0.2-26.5). Pathologic
response rates were available for 717 patients. 119 (19%) IDC had pCR, and 1 (1.1%) ILC (p <0.001). Of the patients with clinical
response data available, 762 (82.8%) of the IDC and 74 (76.3%) of the ILC had a clinical response (p 0.19). Of the 804 IDC
patients who underwent surgery, 480 (59.7%) had breast conserving surgery, while of the 89 lobular patients 33 (37.1%)
underwent breast conserving surgery (<0.0001). Median disease free survival was 10.9 years (95%CI 8.6-13.2) for IDC and 10.1
years (95%CI 7.3-12.9) for ILC (p 0.318).
Within the ER+ve, HER2-ve/unknown group there were 403 IDC and 82 lobular patients. pCR results were available for 148
(30%) of the patients. 41 (15.8%) in the IDC and 1 (1.3%) in the ILC group had pCR (p=0.001). 335 (83.1%) in the IDC group and
64 (78.0%) in the ILC group had a clinical response (p 0.27). Breast conserving surgery rates were 212 (57.6%) in the ductal
group and 30 (39.0%) in the lobular group (p 0.003).
Conclusion
Within the cohort entire rates of pCR and of breast conserving surgery were lower for those with ILC compared with IDC. Despite
lower rates of pCR and breast conserving surgery in ILC, there was no difference in clinical response rates or in disease free
survival compared with IDC. Significant differences in pCR and breast conserving surgery rates between ILC and IDC persisted
on subgroup analysis of those with ER+ve HER2-ve/unknown breast cancer.

Title: Bevacizumab in combination with docetaxel+trastuzumab +/- non-pegylated liposomal doxorubicin: Final results of
ABCSG-32, a prospective, randomized phase II-study
Guenther G Steger1, Richard Greil2, Michael Hubalek3, Michael A Fridrik4, Christian F Singer1, Rupert Bartsch1, Marija Balic5,
Peter Dubsky1, Daniel Egle3, Simon P Gampenrieder2, Georg Pfeiler1, David Mayr4, Theresa Czech3, Gabriel Rinnerthaler2, Ruth
Exner1, Andreas L Petzer6, Paul Sevelda7, Alois Lang8, Margaretha Rudas1, Barbara Krause2, Michael Seifert1, Sophie Frantal9,
Christoph C Zielinski1 and Michael Gnant1. 1Medical University of Vienna, Vienna, Austria; 2Paracelsus Medical University,
Salzburg, Austria; 3Medical University of Innsbruck, Innsbruck, Austria; 4General Hospital of Linz, Linz, Austria; 5Medical
University of Graz, Graz, Austria; 6Hospital of the Sisters of Mercy, Linz, Austria; 7Hietzing Hospital, Vienna, Austria; 8Feldkirch
Hospital, Feldkirch, Austria and 9Austrian Breast and Colorectal Cancer Study Group, Vienna, Austria.
Body: BACKGROUND
Pathological complete response (pCR) after neoadjuvant systemic therapy is correlated with better prognosis in HER2-positive,
early breast cancer. Bevacizumab (B) was shown to add efficacy to neoadjuvant systemic treatment in HER2-negative, early
breast cancer in terms of pCR-rate but failed in the adjuvant setting of both, HER2-negative and HER2-positive disease. The role
of B in the neoadjuvant treatment of HER2-positive, early breast cancer is not defined. Thus, ABCSG-32 was designed as a
randomized phase II trial to evaluate the efficacy, the non-cardiac safety, and the cardiac toxicity of B when added to docetaxel +
trastuzumab (DT) +/- non-pegylated liposomal doxorubicin (N) in the neoadjuvant setting of early, HER2-positive breast cancer.
METHODS
100 patients with biopsy-proven, invasive, early, HER2-positive breast cancer were stratified according to major risk factors
including estrogen receptor (ER)-status, histology, tumor grade, and center and were randomized to receive 6 cycles (q 21 days)
of either D100 mg/m2+T8/6mg/kg (DT, n=25), DT+B15mg/kg (DTB, n=25), D75T+N50 mg/m2 (DTN, n=26), or D75TN+B15
mg/m2 (DTNB, n=24). All patients received pegfilgrastim 6 mg sc on day 2. pCR was defined as the absence of invasive tumor
cells in the breast and total pCR (tpCR) was defined as the absence of invasive tumor cells in the breast and axillary nodes after
NST. Left ventricular (LV) ejection fraction (EF) was measured at baseline, before each treatment cycle, and before sugery. A
cardiac
toxicity event (CTE) was defined as the occurrence of either symptomatic LV dysfunction NYHA II-III, or an asymptomatic drop of
EF (adEF) of >15% from baseline, or an adEF <50%, or the appearance of significant arrhythmias requiring treatment. The trial
was designed to detect a difference in the incidence of CTE of 8% in the control group (DT) vs. 44% in each of the experimental
groups (power: 80%, two-sided alpha: 0.05).
RESULTS
The overall rate of pCR in all patients was 52% (DT: 36%, DTB: 51%, DTN: 63%, DTNB: 62%). In the ER-negative subgroup
(n=43) the overall pCR rate was 63% (DT: 31%, DTB: 70%, DTN: 75%, DTNB: 88%) and the overall tpCR rate was 60% (DT:
31%, DTB: 60%, DTN: 75%, DTNB: 88%). Cardiac toxicity was low with a CTE in only3 patients (DT:0, DHB:1, DTN:1, DTNB:1).
Non-cardiac toxicity/patient as evaluated by the incidence of serious adverse events (SAE,n=50) and significant safety events
(SSE, n=114) was acceptable (SAE: DT:8, DTB:12, DTN:14, DTNB: 16; SSE: DT: 23, DTB: 31, DTN: 29, DTNB: 31). No
differences in the incidence of non-serious AE and no new safety signals for B and N were detected. In 8 patients the treatment
was terminated early (DT: 0, DTB: 3, DTN: 2, DTNB: 3).
CONCLUSIONS
Our data show that all regimens tested are effective in inducing pCR in HER2-positive early breast cancer. The addition of B
and/or N to DT may lead to pCR/tpCR-rates of 51-88% with the highest activity seen in the ER-negative subgroup. 6 cycles of
these regimens can safely be administered to patients with HER2-positive early breast cancer and further phase III-evaluation
appear to be warranted.

Title: Changes in PgR and Ki-67 in residual tumour and outcome of breast cancer patients treated with neoadjuvant
chemotherapy
Emilia Montagna1, Vincenzo Bagnardi1,2, Giuseppe Viale1,3, Nicole Rotmensz1, Andrea Sporchia1, Giuseppe Cancello1, Alessandra
Balduzzi1, Viviana Galimberti1, Paolo Veronesi1,3, Alberto Luini1, Mauro G Mastropasqua1, Chiara Casadio1, Claudia Sangalli1,
Aron Goldhirsch1 and Marco Colleoni1. 1European Institute of Oncology, Milan, Italy; 2University of Milano-Bicocca, Milan, Italy and
3
University of Milan, School of Medicine, Milan, Italy.
Body: PURPOSE We assessed the frequency of the change in biological features between the diagnostic biopsy and final
surgery in breast cancer patients with residual disease after neoadjuvant chemotherapy. We also valued the impact of change on
outcome in terms of disease free survival (DFS) and overall survival (OS)
PATIENTS AND METHODS We collected information through the institutional clinical database on all consecutive breast cancer
patients treated with neoadjuvant chemotherapy at the European Institute of Oncology (EIO), Milan, Italy, between 1999 and
2011. We selected patients who did not achieved pathological complete response (pCR) at final surgery. All patients had
pathological evaluation, including ER, PgR, HER2 protein and ki-67 expression performed at EIO both at diagnostic core biopsy
and final surgery.
RESULTS We identified a total of 904 patients. The 5% of patients with ER positive at diagnostic biopsy had ER negative residual
tumor at final surgery. For PgR expression, 67% of patients, whose tumors had a PgR> 20% at diagnostic biopsy had a
PgR<20% at final surgery. The ki-67 expression changes from >20% to <20% in 40% of patients. At the multivariate analysis the
decrease of PgR -immunoreactive cells correlated with improved outcome in terms of DFS (HR 0.73; CI 0.54-1.00 p 0.046). In
addition, the decrease of ki-67 expression to < 20% of the cells at final surgery was found to be associated with better outcome
both in terms of DFS (HR 0.52; CI 0.40-0.68 p <0.0001) and OS (0.45 CI 0.32-0.64 p<0.0001)
CONCLUSION The decrease of PgR and Ki-67 expression after preoperative chemotherapy have a prognostic role in breast
cancer patients with residual disease and should be considered in the adjuvant therapeutic algorithm.

Title: Eribulin/cyclophosphamide (ErC) versus docetaxel/cyclophosphamide (TC) as neoadjuvant therapy in locally advanced
HER2-negative breast cancer: A randomized phase II trial of the Sarah Cannon Research Institute
Denise Yardley1, Karyn Dyehouse2, Aruna Mani3, Carmen Calfa3, David Drosick2, Kathleen Yost4, Diana Shipley5, Raven Quinn6,
Robyn Young7, Lindsey Finney6, Chris Earwood6, Mythili Shastry6 and John Hainsworth1. 1Sarah Cannon Research Institute and
Tennessee Oncology, PLLC, Nashville, TN; 2Oncology Hematology Care/SCRI, Cincinnati, OH; 3Memorial Cancer Institute,
Hollywood, FL; 4Grand Rapids Oncology Program, Grand Rapids, MI; 5Tennessee Oncology, PLLC/SCRI, Nashville, TN; 6Sarah
Cannon Research Institute, Nashville, TN and 7Center for Cancer and Blood Disorders, Fort Worth, TX.
Body: Background: Pathologic complete response (pCR) following neoadjuvant chemotherapy for locally advanced breast cancer
strongly correlates with improved disease-free survival, rates of breast-conserving surgery, and provides an early indicator of
treatment efficacy. Standard taxane-containing neoadjuvant combinations result in pCR rates of 18% in unselected patient (pt)
populations. Eribulin (Er) is a novel inhibitor of microtubule dynamics. In the EMBRACE trial, Er demonstrated a survival
advantage in anthracycline and taxane pretreated breast cancer. This study evaluated non-anthracycline regimens of Er in
combination with cyclophosphamide (ErC) and docetaxel/cyclophosphamide (TC) as neoadjuvant therapy in HER2-negative
breast cancer.
Methods: Adult women with histologically confirmed invasive HER2-negative (defined as IHC 0-1+ or FISH/SISH negative),
clinical stage T1-3, N0-2, M0 (pN3a disease allowed) adenocarcinoma of the breast were eligible. Additional eligibility criteria
included: ECOG PS 0,1 or 2; normal cardiac function, known hormone receptor status for stratification; adequate hematologic,
liver, and renal function. Following a 10 pt lead-in to confirm the safety and feasibility of ErC, pts were randomized in a 2:1 ratio:
Arm 1, Er 1.4 mg/m2 IV (Days 1 & 8) and C 600 mg/m2 IV (Day 1); Arm 2, T 75 mg/m2 IV and C 600 mg/m2 IV on Day 1, both
regimens administered q 21 days x 6 cycles followed by surgery. Locoregional radiation and/or antiestrogen therapy were
prescribed postoperatively according to standard guidelines. Tumor samples were collected at baseline and from residual disease
noted at the time of surgery. A pCR rate 18% in pts treated with ErC warranted further evaluation.
Results: Enrollment was completed April 2014 (76 pts); 66 pts were randomized (Arm 1, 44; Arm 2, 22). Median age was 52
years (range, 23-77); 88% ECOG 0; 79% invasive ductal adenocarcinoma; median baseline primary tumor size 3cm (range,
0.4-10cm; 30% were T3); axillary nodes clinically positive in 51%. Thirty-three percent were triple negative. 55 pts (72%) received
neoadjuvant therapy and underwent surgery. 16 pts continue to receive study treatment. At this time, pts who underwent surgery
have similar pCR rates with ErC and TC (14%, 5/37 vs. 11%, 2/18 respectively). 3 of 51 hormone receptor positive pts (6%) and 4
of 25 triple negative pts (16%) had pCR. Neutropenia (40%) was the most common grade 3/4 toxicity. Common treatment-related
toxicities included: fatigue (66%), alopecia (57%), and nausea (54%). Peripheral neuropathy incidence is currently lower with ErC
than with TC (30% vs 45%) with 3 grade 3 events on TC. Prophylactic growth factor use was higher in TC than with ErC (77% vs
24%).
Conclusions: Eribulin in combination with cyclophosphamide was well tolerated with no unexpected toxicities. At present, pCR
rates with both regimens are within the range previously reported with docetaxel/cyclophosphamide and other standard
taxane-containing neoadjuvant regimens. Final efficacy and toxicity data will be presented. A planned exploratory correlative
tissue analysis may identify tumors more likely to derive benefit from eribulin-based therapies.

Title: Elevated peripheral blood lymphocyte-to-monocyte ratio predicts favorable response and prognosis in locally advanced
breast cancer after neoadjuvant chemotherapy
Lei Wang1,2, Xiao-Jian Ni1,2, Guo-Wei Qian1,2, Qian-Wen Ou-Yang3, Xiao-Lan Zhang4, Yi-Zhou Jiang1,2, Sheng Chen1,2, Xin Hu1,2
and Zhi-Ming Shao1,2. 1Shanghai Cancer Center and Cancer Institute, Fudan University, Shanghai, China; 2Shanghai Medical
College, Fudan University, Shanghai, China; 3The Third Hospital of Nanchang, Nanchang, Jiangxi, China and 4Changzhou No 2
People's Hospital Affiliated to Nanjing Medical University, Changzhou, Jiangsu, China.
Body: Neoadjuvant chemotherapy (NAC) is the standard of care for locally advanced breast cancer (LABC). However, the lack of
efficient method to predict the treatment response and prognosis limits the clinical evaluation of the patient eligibility. Elevated
lymphocyte-to-monocyte ratio (LMR) has been reported to be associated with favorable prognosis in some hematology
malignancies and nasopharyngeal carcinoma, but has not been studied in breast cancer. The purpose of this study is to evaluate
whether the LMR is a predictor of the curative impact of LABC patients after NAC and short/long-term mortality. A retrospective
cohort of 542 LABC patients who received NAC was recruited between May 2002 and August 2011. The counts for peripheral
absolute lymphocyte and monocyte before NAC were retrieved, and the LMR was calculated. Receiver operating characteristic
curve analysis, univariate and multivariate Cox proportional hazards analyses were applied to evaluate the associations of LMR
with overall survival (OS) and disease-free survival (DFS) respectively. Univariate analysis revealed that higher LMR level (4.25)
was significantly related to better overall response rate (P = 0.005). Univariate analysis revealed that higher LMR level (4.25)
was significantly associated with superior OS (P = 0.014), DFS (P = 0.009). The higher lymphocyte count (1.5109/L) was
significantly associated with better OS (P = 0.035), while the lower monocyte count (<0.4109/L) was associated with better OS
(P = 0.007) and DFS (P = 0.01), respectively. Multivariate Cox proportional hazard analysis showed that higher LMR level was a
significantly independent predictor for superior OS (hazard ratio or HR = 0.636, 95% confidence interval or 95% CI = 0.432-0.937;
P = 0.022), DFS (HR = 0.666, 95% CI = 0.495-0.896; P = 0.007) respectively. The traditional predictive factors, including tumor
size, lymph node status, hormone receptor status and HER2 status, were also independent prognostic factors of OS and DFS.
The elevated pre-NAC peripheral LMR level was a significant favorable factor for NAC LABC prognosis and this easily accessed
variable may serve as a potent marker to predict the outcomes of LABC patients after NAC.

Title: Response to subsequent therapies after failure to achieve pathologic complete response (pCR) after neo-adjuvant
chemotherapy (NAC) in patients (pts) with triple negative breast cancer (TNBC)
Shalu Pahuja1, Kandace McGuire1, Nancy Davidson1, Adam Brufsky1, Priya Rastogi1, Rachel Jankowitz1, Barry Lembersky1 and
Shannon Puhalla1. 1Magee-Women's Cancer Hospital, University of Pittsburgh Cancer Institute, Pittsburgh, PA.
Body: Background: More than half of pts with TNBC do not achieve a pCR after NAC, which carries a poor prognosis. It is
unclear whether these patients respond to subsequent therapies on relapse. We sought to determine the time to progression
(TTP) on subsequent therapies for recurrent disease in TNBC patients who did not achieve pCR to neo-adjuvant therapy.
Methods: We retrospectively identified 117 TNBC pts who received NAC from 2009 through 2012 using an electronic database at
the Magee Womens Breast Cancer Program at the University of Pittsburgh. Patient records were then assessed for pts who did
not achieve a pCR for time to local or distant recurrence (TTR) and time to progression (TTP) on susbequent therapies.
Results: 86 of 117 TNBC pts did not achieve pCR to NAC defined as residual disease present in either the breast and/or axilla.
Median follow up was 2 yrs (range 1-4yrs). Out of these 86 pts, 21 pts(25%) had recurrence in form of distant metastasis or local
recurrence. Nearly all pts(95%) received anthracycline and taxane based NAC. Median TTR was 19 months (4-29 months).
Therapies received in the metastatic setting most commonly included single agent capecitabine, paclitaxel, nab-paclitaxel, and
eribulin. The most commonly received doublets included carboplatin with paclitaxel or gemcitabine. Median TTP on systemic
chemotherapy was 15 weeks (3 -39 weeks), 6 weeks (1-24 weeks) and 6 weeks (2-9 weeks) in first, second and third line setting,
respectively. Greater than 80% pts were able to receive two lines of systemic chemotherapy but <50% received third line
treatment. Two pts died before initiating any systemic treatment.
Conclusions: Response to subsequent chemotherapeutic regimens in metastatic setting in TNBC pts who do not achieve pCR to
anthracycline and taxane based NAC is limited. This highlights the importance of identifying new agents and targeted therapies
for this poor prognosis sub-type.

Title: Dedicated breast PET (dbPET) a unique functional new tool to accurate quantify response to neoadjuvant therapy in breast
cancer
Michel Herranz1, Ines Dominguez1, Alvaro Ruibal1, Elena Brozos1, Sze Yiun Teo2, Carmela Rodriguez1, Jasper Chaal3, Maria
Teresa Curiel1, Juan Cueva1, Beatriz Fernandez1, Gabriel G Pavn4, Marielle Fortier1 and Rafael Lopez1. 1Complejo Hospitalario
Universitario de Santiago de Compostela, Santiago de Compostela, Galicia, Spain; 2Kerbang Kerbau (KK) Women and Childrens
Hospital, Singapore, Singapore; 3Clinical Imaging Research Centre, Singapore, Singapore and 4Oncovision, Valencia,
Comunidad Valenciana, Spain.
Body: Background: Breast cancer is one of the most common cancers in women. Approximately 232,670 new cases of breast
cancer (14% of all new cancer cases) and 40,000 breast cancer deaths (6,8% of all cancer deaths) are expected to occur among
US women in 2014. Since 1990, breast cancer death rates have dropped by 34%. Continued progress in the control of breast
cancer will require sustained and increased efforts to provide high-quality screening, diagnosis, and treatment Recently, the
MAMmography with Molecular Imaging (MAMMI) dedicated breast PET (dbPET) has emerged as an additional imaging tool for
breast cancer diagnosis, clarification of complex lesions and therapy follow-up. This study is aimed to determine whether
correlations exist between physiological images with 18FDG of pre, post-2-cycles and post Neoadjuvant Chemotherapy, with a
predictive value of response.
Methods: Twenty-five patients, and three scan points: pre, 2 cycle and post Neoadjuvant Chemotherapy were included in this
analysis A prone position high-resolution dedicated breast PET (MAMMI-dbPET) was performed 60 min after administration of
90-120 MBq of 18F-FDG. Maximum standardized uptake value (SUVmax) quantification, volume characterization, positioning in
all three space-axes, distances to reference points (proximal breast limit, nipple areola complex) were registered.
Results: When treatment was successful, a significant difference was found between pre and post neoadyuvant chemotherapy
status and the SUVmax (p < 0.001) of breast tumors. Pre Neoadjuvant (mean SUVmax, 11.4) demonstrated a significantly higher
SUVmax than did post 2 cycles tumors (median SUV, 4.7) (p= 0.025). No statistical significant difference was found for SUVmax
of post-2 cycles vs. post lesions with a mean SUVmax of 4.7 and 3.9 (p=0.25) respectively. A statistically significant difference
was found for volume measurement of pre vs. post-2 cycles vs post Neoadjuvant therapy lesions. A clear qualitative difference by
three different observers has been reported among dbPET and MRI volume characterization. A 21% of volume (measurement)
discrepancies was found, always with a volume over-estimation by magnetic resonance (structural vs functional imaging). An
exquisite -and unexpected- millimetric correlation with post-surgical pathology at the end of neoadjuvant theraphy has been found
in dbPET images.
Conclusions: dbPET MAMMI has proven to be an excellent tool for quantification, 3D spatial localization and monitoring of
neoadjuvant therapy, showing, thanks to its functional nature, earlier and better precision and accuracy than conventional
techniques (MRI). Post 2-cycle and Post Neodayuvant Chemotherapy breast cancer tumors consistently display lower 18FDG
uptake than pre treatment tumors when treatment is successful. This data suggest that SUVmax measurements of
18FDG-dedicated breast PET can provide valuable information about therapy efficiency. Such an association might be of relevant
importance to treatment continuity or adjustement.

Title: Vacuum assisted core-needle biopsy after neoadjuvant therapy in breast cancer to predict pathologic response
Regina Grosse1, Jana Saegenschnitter1, Thi-Dao Nguyen2, Susanne Unverzagt3, Joerg Buchmann4, Eva J Kantelhardt1, Nancy
Papendick1 and Christoph Thomssen1. 1Martin Luther University of Halle-Wittenberg, Halle (Saale), Germany; 2Martin Luther
University of Halle-Wittenberg, Halle (Saale), Germany; 3Institute for Medical Epidemiology, Biometry and Informatics, Martin
Luther University of Halle-Wittenberg, Halle (Saale), Germany and 4Martha-Maria Hospital, Halle-Doelau, Halle (Saale), Germany.
Body: Introduction: Recently published neoadjuvant breast cancer therapy trials have documented impressive pathologic
complete response (pCR) rates of up to 50% and above, depending on grade, hormone receptor and HER2 status and type of
therapy. The purpose of our prospective study is to examine the predictive value of a standardized vacuum assisted core-needle
biopsy (VAB) of the tumor site after neoadjuvant therapy with respect to pCR status. This is a presentation of the first case series
of this single institution study.
Methods: As of 2011, all patients who had completed neoadjuvant breast cancer therapy at our institution were informed about
this trial regardless of their clinical response status. In case of consent, a mammography-guided VAB of the initially tagged tumor
site was performed directly before the operation. In case of breast conserving surgery, mammographic wire localisation was
performed immediately thereafter. Final surgery (breast conservation or mastectomy) was performed according to German
guidelines. Ethical approval was obtained from the institutional review board. Pathologic complete response was defined as ypT0
ypN0.
Results: Of 70 patients completing neoadjuvant therapy, 44 consented to the trial. The pCR rate was 47% (n=21). Pre-operative
VAB correctly predicted residual disease in 19 of 23 cases (82,6%). We observed incorrect prediction of pathologic response if
the tagging clip was not placed centrally, or if the radiologist did not perform the VAB in the tagged area. In 3 cases of pathologic
partial response (pPR), all residual tumor (DCIS) was removed by VAB. In these cases, no further malignant lesions were
detected in the resulting surgical specimens.
Conclusions: Preliminary observations of this ongoing study are encouraging, although the correct prediction rate of residual
disease is nowhere near the sensitivity needed to change clinical practice. However, if these results can be improved and
confirmed in a larger series, it may be possible to reduce operative procedures after neoadjuvant therapy based on the results of
a preoperative standardized VAB.

Title: Change of Ki-67 after long course of neoadjuvant chemotherapy as prognostic factors in patients with stage II, III triple
negative breast cancer
Yaewon Yang1, Tae-Yong Kim1,2, Seock-Ah Im1,2, Bhumsuk Keam1,2, Kyung-Hun Lee1,2, Sae-Won Han1,2, Do-Youn Oh1,2, Tae-You
Kim1,2, Hyeong-Gon Moon2,3, Wonshik Han2,3, In Ae Park2,4 and Dong-Young Noh2,3. 1Seoul National University Hospital, Seoul
National University College of Medicine, Seoul, Korea; 2Cancer Research Institute, Seoul National University, Seoul, Korea;
3
Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea and 4Seoul National University
Hospital, Seoul National University College of Medicine, Seoul, Korea.
Body: Background: Triple negative breast cancer (TNBC) shows poor prognosis despite their higher response rate to
neoadjuvant chemotherapy (NAC). Response to NAC such as pathologic complete response (pCR) is a well-known useful
prognostic marker for relapse and overall survival. Pre-chemotherapy baseline higher Ki-67 is a positive predictive marker for
higher pCR rate but a negative prognostic marker for relapse free survival (RFS) and overall survival (OS). After short course of
NAC, Ki-67 index changed in response to the treatment. The purpose of this study is to investigate the prognostic role of Ki-67
index change between before and after the long course of NAC in TNBC patients.
Patients and Methods: Patients with clinical stage II or III breast cancer treated with NAC at Seoul National University Hospital
were enrolled and their clinicopathological findings including Ki-67 index were reviewed retrospectively. The definition of Ki-67
threshold of high was 14% in current study. TNBC definition was adopted from 2011 St Gallen Consensus Panel.
Results: From Jan 2009 to Dec 2010, a total of 183 patients received NAC. Among them, 46 (25.1%) were TNBC. The pCR rate
(19.6% vs 9.5%, P=0.069) and the objective response rate (complete response + partial response) (76.1% vs 70.8%, P=0.153)
were higher in TNBC group than in non-TNBC group. The RFS was not different between the two groups (3 year 77.3% vs
75.4%, P=0.877). The change of Ki-67 index from high (14%) to low (<14%) was significantly associated with longer RFS in only
in TNBC group. As for TNBC patients with high Ki-67 index at diagnosis, RFS of those whose Ki-67 index changed to low after
NAC were significantly longer than that of whose Ki-67 index remained high (3 year RFS 87.2% vs 54.5%, P=0.036). In
non-TNBC group, change of Ki-67 index was not significantly associated with RFS (P=0.245). The initial Ki-67 index itself (high
vs. low) was not associated with RFS in both TNBC and non-TNBC group (P=0.726 and P=0.523 respectively).
Conclusion : The change of Ki-67 index from high to low is associated with longer RFS in TNBC. Ki-67 index change between
before and after the NAC can be used for further discrimination of TNBC as a prognostic marker.

Title: Is pathologic complete response (pCR) a valid marker of outcome even in large breast cancer? Clinical results from a
neoadjuvant trial using a combination of epirubicin, docetaxel and bevacizumab (PROMIX)
Thomas Hatschek1, Judith Bjhle1, Elisabet Lidbrink1, Tobias Lekberg1, Niklas Loman2, Anna von Wachenfeldt Vppling1, Martin
Sderberg2, Zakaria Einbeigi3, Lena Carlsson4, Henrik Lindman5, Irma Fredriksson6, Jan Frisell6, Lars Lfgren7, Lisa Rydn8, Mats
Hellstrm1, Mrten Fern9 and Jonas Bergh1. 1Karolinska University Hospital, Stockholm, Sweden; 2Skne University Hospital,
Lund, Sweden; 3Sahlgrenska University Hospital, Gothenburg, Sweden; 4Sundsvall Hospital, Sundsvall, Sweden; 5Uppsala
University Hospital, Uppsala, Sweden; 6Karolinska University Hospital, Stockholm, Sweden; 7St Gran Hospital, Stockholm,
Sweden; 8Skne University Hospital, Lund, Sweden and 9Lund University, Lund, Sweden.
Body: Background: Several randomized trials have proven that preoperative chemotherapy is equivalent to adjuvant treatment,
and allows for clinical and radiological assessment of efficacy in vivo. Recent results show that response-guided neoadjuvant
therapy is a favorable policy for hormone receptor positive tumors, while pCR predicts prognosis for triple-negative (TNBC) and
HER2 positive breast cancer (von Minckwitz 2013; Cortazar 2014). Since 2012 the FDA accepts pCR (ypT0 ypN0) as endpoint
for accelerated drug approval.
Methods: In total, 150 women 18 years or older with verified HER2 negative breast cancer suitable for primary medical treatment
were included in the trial between September 2008 and December 2011. The patients received two courses of epirubicin and
docetaxel (Taxotere), both 75mg/m2 for the 1st two courses, followed by the same treatment with addition of bevacizumab
(Avastin) 15 mg/kg for 4 additional courses. Clinical and radiological evaluations with mammography and ultrasound were
performed before start and after courses 2, 4 and 6. Core biopsies were taken before start, after 2 courses, and at time of
surgery. Blood samples were drawn before and 24 hours after the first 4 courses.
Results: Median age was 49 years, range: 27 to 70 years; 73% were reported as ductal, 15% as lobular, and 12% as rare
histological types. Mean tumor size was 59 mm, median 55 mm, range: 20-180 mm; 3 tumors (2.0%) were reported as
inflammatory, and 13 (8.7%) presented with skin involvement (T4b). Enlarged axillary nodes were found in 102 patients (68%)
before start of treatment, 77 of these (64%) verified as metastatic. SNB in cases of normal axillary status was performed in 16
cases, in 9 cases (8%) with positive finding. Supra- or infraclavicular node involvement was verified in 20 cases (13%). 25% of all
tumors were ER- and/or PR-negative, tumor grade based on a diagnostic biopsy was evaluable in only 83 cases. Of these, 4
(5%) were grade I, 46 (55%) grade II, and 33 (40%) grade III. Mean proliferation count (Ki67) was 37%, median 30%, range
1-90%. Breakdown into intrinsic subtypes based on immunohistochemistry defined 68 (46%) as luminal A-like, 36 (24%) as
luminal B-like, and 44 (30%) as TNBC. pCR was achieved in 20 cases, 3 (2%) luminal A-like, 5 (3.4%) luminal B-like and 12
(8.2%) TNBC. After 2.2 years of follow-up, 35 patients (23.3%) have experienced recurrence and 18 of these (12%) have
deceased due to breast cancer, among these 6 despite pCR, 2 classified as luminal B-like, and 4 as TNBC. The molecular
subtype of the tumor predicted outcome, but pCR was not in our material, even after adjustment for tumor size at diagnosis, a
predictor of favorable outcome. The number of events in relation to molecular subtypes is however limited. Updated outcome data
will be presented.
Conclusions: The present trial does not confirm previously reported observations that pCR is a marker of favorable prognosis.
One possible explanation is that unfavorable biological characteristics, particularly heterogeneity, may increase with tumor
burden. Genomic and proteomic analyses are currently ongoing.

Title: Different clinical usefulness of AJCC response criteria according to the subtypes in stage II/III breast cancer patients after
neoadjuvant chemotherapy
Yaewon Yang1, Seock-Ah Im1,2, Tae-Yong Kim1,2, Kyung-Hun Lee1,2, Bhumsuk Keam1,2, Sae-Won Han1,2, Do-Youn Oh1,2, Tae-You
Kim1,2, Wonshik Han2,3, Hyeong-Gon Moon2,3, In Ae Park2,4 and Dong-Young Noh2,3. 1Seoul National University Hospital, Seoul
National University College of Medicine, Seoul, Korea; 2Cancer Research Institute, Seoul National University, Seoul, Korea;
3
Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea and 4Seoul National University
Hospital, Seoul National University College of Medicine, Seoul, Korea.
Body: Background: Response to neoadjuvant chemotherapy (NAC) is a useful prognostic marker for relapse and overall survival
but their prognostic role is distinct in different breast cancer subtypes. Pathologic complete response (pCR) is a current gold
standard for evaluating response to NAC. Despite its clinical usefulness, discrimination into pCR and non-pCR is too simple
because non-pCR includes broad range of actual responses from very good partial response to even progressive disease.
American Joint Committee on Cancer (AJCC) 7th edition proposed response criteria for NAC based on precise clinical stage
before NAC and pathologic staging after NAC. The purpose of this study is to evaluate the clinical usefulness of AJCC response
criteria in each breast cancer subtypes.
Patients and Methods: Patients with clinical stage II or III breast cancer treated with NAC at Seoul National University Hospital
were enrolled and reviewed retrospectively. AJCC response criteria for NAC were adopted from the AJCC 7th edition: complete
response (CR), absence of invasive carcinoma in both breast and lymph node; partial response (PR), decrease in either or both T
or N stage; no response (NR), no change or increase in either or both T or N stage. The definition of each breast cancer subtypes
were adopted from 2011 St Gallen Consensus Panel for current study.
Results: From January 2009 to December 2010, 183 patients were enrolled in the study with median follow up period of 38.0
months. AJCC response was significantly associated with relapse free survival (RFS) (P<0.001), whereas pCR was not
(P=0.120). In subgroup analysis, AJCC response was a significant prognostic factor for RFS in luminal B (P<0.001), HER-2
enriched (P=0.035) and triple negative breast cancer (P=0.037) groups, but not in luminal A group patients (P=0.101). On the
contrary, pCR was not significantly associated with RFS in any groups.
Conclusion : AJCC criteria is an easily reproducible tool for response evaluation for breast cancer patients in NAC setting
compared with classically used pCR in all breast cancer subgroups except luminal A.

Title: S0800: Nab-paclitaxel, doxorubicin, cyclophosphamide, and pegfilgrastim with or without bevacizumab in treating women
with inflammatory or locally advanced breast cancer (NCI CDR0000636131)
Zeina A Nahleh1, William E Barlow2, Daniel F Hayes3, Anne F Schott3, Julie R Gralow4, Edith A Perez5, William M Sikov6, Sudhathi
Chennuru7, Hamid Mirshahidi8, Sarah Vidito9, Danika L Lew2, Lajos Pusztai10, Robert B Livingston11 and Gabriel N Hortobagyi12.
1
Texas Tech University Health Sciences Center, Paul L. Foster School of Medicine, El Paso, TX; 2SWOG Statistical Center,
Seattle, WA; 3University of Michigan, Ann Arbor, MI; 4Seattle Cancer Care Alliance, University of Washington, Seattle, WA; 5Mayo
Clinic, Jacksonville, FL; 6Warren Alpert Medical School of Brown University, Providence, RI; 7Hem/Onc Consultants,
Inc/Columbus CCOP, Westerville, OH; 8Loma Linda University, Loma Linda, CA; 9Upstate Carolina CCOP/ Spartanburg Regional
Medical Center, Spartanburg, SC; 10Yale University, New Haven, CT; 11Arizona Cancer Center, Tucson, AZ and 12MD Anderson
Cancer Center, Houston, TX.
Body: Background. Locally advanced breast cancer (LABC) and inflammatory breast cancer (IBC) remain difficult challenges
despite progress in multimodality treatment. The SWOG trial S0800 (clinicaltrial.gov NCT00856492) compared bevacizumab in
combination with weekly nab-paclitaxel followed by dose-dense AC to nab-paclitaxel followed or preceded by AC as neoadjuvant
treatment for HER2-negative IBC/LABC. The rationale was based on the proposed role of angiogenesis, potential role of
improved flow and oxygenation in enhancing the delivery of chemotherapy agents, and the proapoptotic effect with certain
chemotherapeutic agents, particularly taxanes.
Methods. This was a randomized open-label Phase II trial with an accrual goal of 200 patients equally allocated to either
bevacizumab (Arm 1) or no bevacizumab.Two patients were ineligible and five withdrew consent leaving 208 for analysis. The no
bevacizumab group was further randomized to two sequences (Arm 2: nab-paclitaxel - AC+PEG-G versus Arm 3: AC+PEG-Gnab-paclitaxel) with 50 patients expected in each sequence. The primary endpoint of this study was pathologic complete
response (pCR) defined as no evidence of invasive tumor at the primary site and axillary lymph nodes in the surgical specimen.
The power for the primary comparison of bevacizumab (bev) versus no bevacizumab (no bev) ignoring sequence was 80% with a
1-sided = 0.10. Randomization was stratified by hormone receptor status and type of disease (IBC or LABC).
Results 215 patients were accrued May 2010 - September 2012. Most had LABC (88%) versus IBC (12%) and most tumors were
hormone-receptor (HR) positive (67%). Fourteen (7%) patients had no definitive surgery (included as no pCR); 135 (65%) had
residual disease (no pCR) and 59 (28%) had pCR. The bevacizumab pCR rate was higher (35/96; 38%) than that in the
non-bevacizumab arms (24/112; 21%) (exact p=0.021; stratified p=0.015). In HR-positive disease there was slight improvement
that was not statistically significant (bev 25% vs. non-bev 18%; p=0.41) while the difference was larger in HR-negative disease
(bev 59% vs. non-bev 28%; p=0.014). In LABC the overall pCR rate was 29% with a higher rate in the bevacizumab patients
(37% vs. 22%; p=0.035). For IBC there was improvement (30% vs. 14%), but not statistically significant (p=0.61) in a small
sample. Overall, Grade 3 and 4 events were common in both (bev 67%; non-bev 65%), but did not differ by treatment. There
were 21 deaths with 3-year overall survival (OS) of 87% and 83% for bevacizumab and non-bevacizumab, respectively (log-rank
p=0.57).
Conclusion Compared with combination anthracycline-taxane neoadjuvant chemotherapy, the Bev-Nab-paclitaxel-AC regimen
significantly improved pCR rate overall, primarily for triple negative (TNBC) patients. This neoadjuvant regimen could be a good
choice for TNBC/IBC . The observed pCR rate in ER negative disease (59%) suggests that the addition of bevacizumab to a
standard chemotherapy backbone may improve outcome in this subset, and justifies further testing of such an approach.
Correlative science studies to further delineate the biology of TNBC and the effects of bevacizumab are ongoing.

Title: Impact of treatment on quality of life (QOL) and menstrual history (MH) in the NSABP B-36: A randomized phase III trial
comparing six cycles of 5-fluorouracil (5-FU), epirubicin, and cyclophosphamide (FEC) to four cycles of adriamycin and
cyclophosphamide (AC) in patients (pts) with node-negative breast cancer
Patricia A Ganz1,2, John W Wilson1,3, Hanna Bandos1,3, Andr Robidoux1,4, Alexander HG Paterson1,5, Johnathan Polikoff1,6, Luis
Baez-Diaz1,7, Adam M Brufsky1,8, Louis Fehrenbacher1,9, Aroop Mangalik1,10, Patrick J Ward1,11, Louise Provencher1,12, John T
Hamm1,13, Philip J Stella1,14, Robert L Carolla1,15, Richard G Margolese1,16, Henry R Shibata1,17, Edith A Perez18,19 and Norman
Wolmark1,20. 1National Surgical Adjuvant Breast and Bowel Project (NSABP), Pittsburgh, PA; 2University of California, Jonsson
Comprehensive Cancer Center, Los Angeles, CA; 3Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA;
4
Centre Hospitalier de l'Universit de Montral (CHUM); 5Tom Baker Cancer Centre; 6Kaiser Permanente San Diego Mission;
7
MBCCOP, San Juan; 8University of Pittsburgh, Magee Womens Hospital, PA; 9Kaiser Permanente, Northern CA Region;
10
MBCCOP, University of New Mexico; 11Oncology/Hematology Care Clinical Trials; 12Centre Hospitalier Affilie Universitaire de
Quebec (CHA); 13Norton Cancer Institute; 14CCOP, Michigan Cancer Research Consortium Community Clin Onc Program;
15
CCOP, Ozark Health Ventures; 16Jewish General Hospital, McGill University; 17Royal Victoria Hospital; 18Mayo Clinic,
Jacksonville, GL; 19NCCTG/ALLIANCE and 20Allegheny Cancer Center at Allegheny General Hospital.
Body: Background
NSABP B-36 compares 6 cycles of FEC-100 with 4 cycles of standard AC in pts with node-negative breast cancer. As reported
separately, no significant differences between the two treatment arms were observed in the primary endpoint of disease-free
survival or in the secondary endpoints of overall survival, recurrence-free, or distant recurrence-free intervals. Greater toxicity was
reported in pts who received FEC compared to AC. We present here the results of the QOL and MH studies obtained
prospectively in conjunction with the treatment study. We hypothesized that pts on FEC would experience greater treatment
toxicity in the first 12 months of the study, and would have greater rates of amenorrhea at 18 months compared to pts on AC.
Methods
Among the 1,357 pts enrolled in the QOL study, 1,332 (675 AC, 657 FEC) submitted the baseline form and had QOL follow-up
(fup) information. Pts completed: 1) the FACT-B instrument; 2) a symptom checklist; and 3) the SF-36 Vitality Scale, all at
baseline, day 1 of cycle 4, and at 6, 12, 18, 24, 30, and 36 months after random assignment. FACT-B Trial Outcome Index (TOI),
symptom severity, and vitality scores were compared between the two treatment arms using a mixed model for repeated
measures analysis with adjustment for the baseline scores, type of surgery, and hormone receptor status, examining the first 12
months and the later time points separately. Menstrual status was collected at baseline for all enrolled pts, with subsequent
assessments on day 1 of cycle 4, and at 6, 12, 18, 24, 30, and 36 months for pts with menstrual bleeding within 12 months prior
to random assignment and not having had a hysterectomy and/or bilateral oophorectomy (1, 096 pts). Post-chemotherapy
amenorrhea was defined as the lack of menstrual periods during the 12 months preceding the 18-month fup evaluation. Data from
921 pts (475 AC, 446 FEC) were available for analysis. Logistic regression, adjusted for type of surgery and hormone receptor
status, was used to test for association of amenorrhea status and treatment.
Results
Both TOI and vitality scores were worse for pts on FEC compared to those on AC at 6 months (p<0.01) with no significant
difference at 12 months and beyond. No significant differences in symptom severity between the two treatment arms were
observed. The rates of post-chemotherapy amenorrhea were significantly different between FEC and AC (66.8% vs. 58.7%,
p=0.01) with positive hormone receptor status as an independent risk factor (p=0.03).
Conclusions
Women receiving FEC had diminished QOL at 6 months after random assignment, but no difference at 12 months or later.
Premenopausal women receiving FEC experienced a higher rate of post-chemotherapy amenorrhea than women receiving AC.
Support
NCI grants U10-CA-12027, -37377, -69974, -69651 and -44066-26, and Pharmacia & Upjohn Company, a subsidiary of Pfizer,
Inc.

Title: Lifestyle interventions combined with acupuncture-like transcutaneous electrical nerve stimulation in managing vasomotor
symptoms induced by breast cancer treatment: Results of a phase 2 randomized controlled trial
Margaret Forbes1,2, Raimond Wong1,2, Stephen M Sagar1,2, Jim A Julian1,2,3, Mark N Levine1,2,3 and Joseph Hayward1,2. 1McMaster
University, Hamilton, ON, Canada; 2Juravinski Cancer Centre/Hamilton Health Sciences, Hamilton, ON, Canada and 3Ontario
Clinical Oncology Group, Hamilton, ON, Canada.
Body: Background: Women with breast cancer can experience significant treatment induced vasomotor symptoms (TIVS).
Non-hormonal strategies for TIVS (e.g. acupuncture, Venlafaxine) provide some relief but may be intolerable because of the
invasive nature of the treatment and possible side effects. As such, women often prefer lifestyle strategies (LS) that can be
self-administered. Acupuncture-like transcutaneous electrical nerve stimulation (ALTENS) is a non-invasive needleless technique
that uses specific electrical parameters to stimulate selected acupoints to achieve clinical response with minimal toxicity. ALTENS
can be administered with minimal training. It is amenable to quality assurance and can allow for self treatment by women. This
study aimed to evaluate the efficacy of ALTENS in addition to LS in relieving TIVS in women with breast cancer.
Methods: Eligible subjects were postmenopausal women with Stages 0-3 breast cancer who had completed cancer treatment
and were experiencing hot flashes for one month with a Hot Flash Score (HFS) 15 in one week prior to consent. Anti-estrogen
therapy was permitted. Non-hormonal drug therapies were prohibited. Subjects were randomized to either LS (control) or LS with
concurrent ALTENS (combined). LS consisted of standardized lifestyle strategies (e.g. environmental control, managing hot flash
triggers) counseling delivered by a specially trained nurse practitioner at week 0 with reinforcing counseling at weeks 12 and 24.
ALTENS was given twice weekly for 12 treatments over an 8-week period. The HFS, Hot Flash Related Daily Interference Scale
and the Short Form version 2 health survey were administered at weeks 0, 12 and 24. Heart rate variability was measured at
weeks 0 and 12. The primary study endpoint was the number of responders, defined as women who had > 50% reduction in their
HFS between weeks 0 and 12.
Results: 71 eligible subjects with a median age of 52 (range: 40-87) were randomized to combined arm (n=36) and control
(n=35). At 12 weeks there were 11 (30.6%) responders in the combined arm versus 2 (5.7%) in the control (p=0.012).The results
at 24 weeks were 14 versus 4, respectively (p=0.013). Arms were balanced for anti-estrogen use. Two subjects chose to
discontinue ALTENS after experiencing symptoms improvement. There were no serious adverse events.
Conclusion: ALTENS in combination with lifestyle strategies is a promising non-pharmacologic approach that showed
improvement in managing treatment induced vasomotor symptoms in women with breast cancer. Our trial results support its
evaluation in phase 3 studies.

Title: Risk of infectious complications in breast cancer patients treated with trastuzumab: A meta-analysis
Tomohiro Funakoshi1, Maya Suzuki2 and Hyman B Muss1. 1UNC Lineberger Comprehensive Cancer Center, Chapel Hill, NC and
2
Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background
Serious infections related to the anti-HER2 monoclonal antibody trastuzumab has been reported in clinical trials. It is not yet clear
whether trastuzumab increases infection risk or not. We performed a systematic review and meta-analysis to assess the risk of
infections associated with trastuzumab.
We searched PubMed and the ASCO online database of meeting abstracts up to January 2014 for relevant clinical trials. Eligible
studies included randomized controlled trials (RCTs) of trastuzumab for breast cancer patients that reported adequate safety data
for grade 3-4 infection, febrile neutropenia, neutropenia, or leukopenia. Grade3 and 4 infection is defined as "IV antimicrobial
agents, interventional radiology or operative intervention indicated" and "life-threatening consequences (e.g., septic shock)",
respectively. The summary incidence, relative risk (RR) and 95% confidence intervals (CIs) were calculated. RRs and summary
incidence were calculated using random-or fixed-effects models depending on the heterogeneity of included studies. Prespecified
exploratory subgroup analyses were performed according to treatment setting (neoadjuvant or adjuvant vs. metastatic setting)
and therapy used with trastuzumab (combination vs. best supportive care (BSC) or single agent therapy) with regard to the RRs.
Results
A total of 10,094 patients from 13 trials of trastuzumab were included. The use of trastuzumab was associated with an increased
risk of high-grade infection (RR, 1.21; 95% CI, 1.07-1.37; P=0.002) and febrile neutropenia (RR, 1.28; 95% CI, 1.08-1.52;
P=0.004). The incidence of grade 3-4 infection and febrile neutropenia related to trastuzumab and control was 8.5% (95% CI
4.5-15.4%) vs. 6.9% (95% CI 4.011.6%) and 12.0% (95% CI, 8.1-17.4%) vs. 9.9% (95% CI, 6.6-14.6%), respectively. There was
no significant increase in a risk of high-grade neutropenia (RR, 1.07; 95% CI, 0.98-1.68; P=0.15) or leukopenia (RR, 1.07; 95%
CI, 0.94-1.23; P=0.31) in patients receiving trastuzumab. The results of exploratory subgroup analyses were shown in table1.
RR of high-grade infection and febrile neutropenia according to treatment setting and therapy used with trastuzumab.
High-grade infection
Febrile neutropenia
No. of trials
RR (95%CI)
No. of trials
RR (95%CI)
Neoadjuvant/Adjuvant setting
1.21 (1.06-1.38)
0.004
1.27 (1.06-1.52)
0.008
Metastatic setting
1.21 (0.82-1.79)
0.34
1.38 (0.79-2.42)
0.26
Combination therapy
1.18 (1.04-1.35)
0.01
1.27 (1.06-1.52)
0.008
BSC/Single agent therapy
1.47 (1.02-2.12)
0.04
1.38 (0.79-2.42)
0.26
Treatment setting
Therapy used with trastuzumab
BSC, best supportive care; RR, relative risk

Conclusions
Treatment with trastuzumab is associated with a 1.21-fold increased risk of high-grade infection and a 1.28-fold increased risk of
febrile neutropenia compared with control. In order to obtain the maximum therapeutic benefit from trastuzumab, clinicians should
be aware of the increased infection risk and perform careful monitoring especially when it is used with combination chemotherapy
in the neoadjuvant and adjuvant setting.

Title: P53 based strategy to reduce hematological toxicity of chemotherapy: A Pilot Phase II study
Chul S Ha1, Richard Elledge1, Kevin R Kelly5, Suthakar Ganapathy2, Hang Su1, Carol A Jenkins1, Athanassios Argiris3, Ronan
Swords4, Anand Karnad1, Martin Goros1, Bradley Pollock1 and Zhi-Min Yuan2. 1University of Texas Health Science Center, San
Antonio, TX; 2Harvard University School of Public Health, Boston, MA; 3Hygeia Hospital, Athens, Greece; 4University of Miami
Miller School of Medicine, Miami, FL and 5University of Southern California Keck School of Medicine, Los Angeles, CA.
Body: Background: P53 activation is a major pathway by which normal tissues respond to DNA damaging agents such as
chemo and radiotherapy. We have shown that the use of very low dose arsenic (LDA) for 3 days before chemotherapy in animal
models temporarily and reversibly suppresses p53 activation for about 5 days. LDA-mediated protection requires functional p53
and thus is selective to normal tissues, as essentially every cancer cell has dysfunctional p53. Arsenic Trioxide (ATO) is currently
used to treat acute promyelocytic leukemia (APL) at much higher dose (50-fold higher than the dose we used for suppression of
p53). The primary objectives of this trial were to: 1) define the lowest safe dose of ATO that blocks p53 activity in vitro as
measured in patients(pts) peripheral lymphocytes and 2) assess the potential of LDA to decrease hematological toxicity in pts
receiving chemotherapy.
Methods: Pts with malignancies other than leukemia who were to receive at least 4 cycles of myelosuppressive chemotherapy
given at least 2 weeks apart were eligible. Pts had to have no baseline p53 activation in peripheral lymphocytes but p53 had to be
responsive as measured by an in vitro assay. For objective 1, dose escalation was performed at the starting dose of ATO
0.005mg/kg/day for 3 days. For objective 2, ATO 0.005mg/kg /day x 3 was given prior to chemotherapy cycles 2, 4 and 6 only.
WBC, ANC, HgB and platelet counts were obtained on chemotherapy days 1 (prior to chemotherapy), 8, 15 and 22, normalized to
the corresponding counts from day 1 and compared between cycles with a paired t-test. The presented analysis compares cycles
1 and 2.
Results: Thirty-three evaluable pts were accrued. Chemotherapy was: TC 8, AC 7, 8 other regimens 18 pts. ATO at a dose of
0.005mg/kg/day for 3 days blocked p53 activity in vitro as measured in pts peripheral lymphocytes and was not associated with
any toxicity. For WBC and ANC, potential protective effect was most pronounced on day 8. The mean normalized WBC counts
were 0.54 in cycle 1 and 0.72 in cycle 2 on day 8 (p=0.01). 64% of pts had higher normalized WBC in cycle 2 over cycle 1, 24%
remained the same (5%) and 12% lower value. The mean normalized ANC counts were 0.53 in cycle 1 and 0.85 in cycle 2 on
day 8 (p=0.02). 59% of pts had higher normalized ANC in cycle 2 over 1, 31% same, and 9% lower. There was no significant
change detectable for HgB as less than 10% of pts had > grade 1 anemia. The mean normalized platelet counts were 0.91, 1.11
and 1.21 in cycle 1 and 0.77, 0.83 and 1.03 in cycle 2 on days 8,15 and 22 respectively (corresponding p values=0.001, <0.001,
and 0.04). It is worth noting that >96% of pts had normal platelet counts or grade I thrombocytopenia. On an explorative subset
analysis using the exact Wilcoxon signed rank test, the chemotherapy regimen that gained the most protective effect was AC
(doxorubicin and cyclophosphamide) used for breast cancer, with mean normalized WBC counts of 0.39 and 0.46 (p=0.03) and
ANC counts of 0.39 and 0.49 (p=0.02) on day 8 of cycle 1 and 2 respectively.
Conclusions: Our data suggest that LDA pretreatment confers significant protection against chemotherapy induced leukopenia
and neutropenia. Further study is needed to confirm its beneficial effect on AC chemotherapy.

Title: Reduction of ovarian reserve in young early breast cancer patients: 24 months follow up of a prospective cohort trial
Christoph Mundhenke1, Antonia S Wenners1, Jana Grambach1, Juliane Koss1, Ulrike von Hehn2, Walter Jonat1 and Andreas G
Schmutzler3. 1OB/GYN, Breast Unit, UKSH, University of Kiel, Kiel, Germany; 2Medistat, Kiel, Germany and 3OB/GYN,
Reproductive Medicine, UKSH, University of Kiel, Kiel, Germany.
Body: Background: Sideeffects of adjuvant chemotherapy on ovarian function are inadequately studied by now. Premenopausal
women undergoing chemotherapy are at risk for sexual hormone deficiency and impaired fertility. This prospective cohort study
searches for predictive parameters of ovarian reserve after chemotherapy. Methods: 51 premenopausal patients (average 38
years; 28-46 years) with primary breast cancer have been included. They all received neo- or adjuvant chemotherapy:
anthracycline based "A"(FEC)(n= 18); anthracycline/taxane based "T"( TAC or FEC/Doc or EC/Pac)(n=30) or anthracycline- free
"AF"(n=3). After chemotherapy patients received endocrine therapy (ET) for hormone receptor (HR) positive disease (tamoxifen
(n=34) or goserelin/tamoxifen (n=7)). For HR negative disease (n=20) no ET was given. Before (visit 1) and 6, 12 and 24 months
after initiation of chemotherapy (visit 2-4) age and chemotherapy related changes in hormone (LH, FSH, E2 and Anti-Muellerian
hormone (AMH)) levels, antral follicle count and amenorrhea as parameters of endocrine function and fertility were assessed.
Also quality of life (QL) and additional impact of parity, BMI and nicotine use on ovarian reserve were evaluated. Results: Antral
follicle count is decreasing dramatically after chemotherapy and is not fully recovering after 2 years. Antral follicle count before
and 1 year after chemotherapy are significantly correlated (p=0.012). Follicle count and age before chemotherapy are negatively
correlated (p=0.004). At visit 4 the type of chemotherapy (A, T, AF) has no influence on antral follicle count or AMH levels.
Tamoxifen is correlated with a significant decrease in follicle count (vs. no endocrine therapy) (p=0.039). Smoking is significantly
detrimental on antral follicle count after 24 months (p=0.001). AMH levels at visits 2-4 are significantly correlated with levels at the
1 visit and are negatively correlated with age (p=0.032). Tamoxifen therapy has no impact on AMH levels. LH and FSH are
increasing between visit 1 and 2 and decreasing at visits 3 and 4 (but stay above basic levels). At visit 4 LH is not and FSH is
correlated with visit 1 levels (p=0.005). LH and FSH are correlated with age (p=0.0001) and are not influenced by type of
chemotherapy or tamoxifen .E2 levels at visit 4 are not influenced by type of chemotherapy, but significantly decreased (p=0.003)
by continued smoking. At visit 4 31/51 patients have stayed amenorrheic, 19 resumed their menstruation. This was not influenced
by type of chemotherapy or age. Non- smokers are 13 times more likely to resume their menstruation (vs. smokers). Tamoxifen is
correlated with a higher probability of permanent amenorrhea (p=0.058, n.s.). BMI, nicotine abuse, age, type of chemotherapy
and tamoxifen have no influence on duration of amenorrhea. QL is significantly lower at visit 2, but recovers to basic values at
visits 3 and 4. Conclusions: Our study contributes to a better prediction of ovarian reserve before chemotherapy. We suggest
personalized counselling on fertility preserving measures before chemotherapy especially at higher age, with low AMH levels or
low antral follicle counts. To stop smoking could enhance chances for ovarian preservation.

Title: Subpopulations of peripheral sensory neurons are differentially sensitive to the microtubule-targeting agent, paclitaxel
Peter M LoCoco1, Teresa C Chavera1, Raehannah J Jamshidi1, Susan L Mooberry1, Kelly A Berg1 and William P Clarke1.
1
University of Texas Health Science Center, San Antonio, TX.
Body: Paclitaxel (PTX), a microtubule-targeting anticancer agent, produces a debilitating peripheral neuropathy that is
accompanied by neuropathic pain. Patients develop localized pain in their distal extremities, and more frequently complain of
increased sensitivity to cold rather than hot temperatures. These clinical observations led us to hypothesize that PTX differentially
damages subpopulations of peripheral sensory neurons. The purpose of this study was to evaluate the effects of PTX on
bradykinin (BK) and eicosanoid (EP) receptor-expressing sensory neurons. PTX (11.7 mg/kg) or vehicle (Cremophor
EL/EtOH/PBS; 1:1:6) was administered intraperitoneally to Sprague-Dawley rats (250-300 g) every other day for 5 days.
Following treatment, PTX-treated rats exhibited increased sensitivity to cold and mechanical stimuli, but decreased sensitivity to
heat stimulation as compared to vehicle-treated rats. Intraplantar (i.pl.) injection with BK (5 g) produced a prolonged allodynia to
cold, but a diminished allodynia to heat in PTX-treated rats. Interestingly, following treatment with PGE2 (0.3 g, i.pl.),
PTX-treated rats displayed no statistically significant differences in the allodynia to cold or heat compared to vehicle-treated rats.
Efficacy of BK (1 pM - 1 M) to stimulate inositol phosphate (IP) production in cultured primary sensory neurons from PTX-treated
rats was significantly reduced. By contrast, stimulation of cAMP production by PGE2 (0.01 pM - 100 nM) was unaffected by PTX
treatment. Preliminary mechanistic studies treating nave cultures of rat primary sensory neurons directly with PTX (5 nM, 0 - 48
hr) yielded a time-dependent, tri-phasic effect on BK-stimulated IP accumulation and reduced PGE2-stimulated cAMP production,
suggesting PTX differentially disrupts BK- and PGE2-mediated signaling. Collectively, these data provide evidence that
subpopulations of peripheral sensory neurons are differentially sensitive to PTX. Understanding why particular subclasses are
less sensitive, or resistant, to damage could reveal novel approaches to protect sensory neurons from chemotoxicity and
ultimately prevent chemotherapy-induced peripheral neuropathy.
Supported by USPS grant #8ULITR000149, Greehey Fellowship, and Translational Science Training Across Disciplines Program
at UTHSCSA
The project described was supported by Award Number 8UL1TR000149 from the National Center for Advancing Translational
Sciences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the
National Center for Advancing Translational Sciences.

Title: A multicenter randomized double-blind phase III clinical trial to evaluate the efficacy and safety of GCPGC, a novel
pegylated G-CSF in patients receiving DA or TAC chemotherapy for breast cancer compared to peg-filgrastim
Joo Han Lim1, Ki-Hyeong Lee2, Keon-Uk Park3, In-Hae Park4, Eun Kyung Cho5, Moon Hee Lee1, So-Young Yoon6, Jee-Hyun
Kim7, In-Sil Choi8, Jae-Hoo Park9, Young-Jin Choi10, Hee-Jun Kim11, Kyung Hae Jung12, Si-Young Kim13, Do-Youn Oh14 and
Seock-Ah Im14. 1Inha University Hospital, Incheon, Korea; 2Chungbuk National University Hospital, Korea; 3Keimyung University
Hospital, Korea; 4National Cancer Center, Korea; 5Gachon University Gil Medical Center, Incheon, Korea; 6Konkuk University
Medical Center, Seoul, Korea; 7Seoul National University Bundang Hospital, Korea; 8SMG-SNU Boramae Medical Center, Korea;
9
Ulsan University Hospital, Korea; 10Pusan National University Hospital, Korea; 11Chung-Ang University Hospital, Korea; 12Asan
Medical Center, University of Ulsan College of Medicine, Korea; 13Kyung Hee University Hospital, Korea and 14Seoul National
University Hospital, Korea.
Body: Introduction
Treatment with cytotoxic chemotherapy is known to be associated with a significant risk of febrile neutropenia. GCPGC(Green
Cross Co., Korea) is a novel long acting recombinant human granulocyte colony-stimulating factor (G-CSF) analog that reduces
the severity and duration of neutropenia. We conducted a phase III trial to evaluate the efficacy and safety of GCPGC compared
to pegfilgrastim.
Methods
A total of 117 patients were enrolled in this multicenter, phase III, double-blind randomized trial between Feb 2012 and May 2013.
They were randomly assigned to receive either GCPGC or pegfilgrastim during the maximal 6 cycles of chemotherapy which
consisted of DA (docetaxel and doxorubicin) or TAC (docetaxel, doxorubicin, and cyclophosphamide). Based on the results of the
phase II dose-finding study for GCPGC, the dose of 6 mg was selected. Both medications were administered 24 hours after the
completion of each cycle of chemotherapy. The primary efficacy endpoint of this study was the duration of grade 4 neutropenia
(Absolute Neutrophil Count (ANC) <500/mm3) during the first cycle of chemotherapy. Secondary endpoints included the time to
ANC recovery during the first cycle of chemotherapy, which was defined as the number of days required for neutrophil counts to
exceed 2,000/mm3, the rate of febrile neutropenia, the rate of severe neutropenia which persisted for more than three days during
first cycle of chemotherapy, the depth of ANC nadir, the ANC level on day 7 after each cycle of chemotherapy, the frequency of
dose reduction or delay, the number of hospitalization related to febrile neutropenia after the first cycle of chemotherapy, and the
number of treatment with intravenous antibiotics.
Results
A total of 116 patients were evaluable for safety and 115 patients were evaluable for efficacy. The intention-to-treat analysis
showed that there was no statistical difference between GCPGC and pegfilgrastim in terms of the duration (days) of grade 4
neutropenia in chemotherapy cycle 1 (1.641.18 vs 1.801.05; difference -0.151.11 [97.5% C.I. -, 0.26]; the pre-specified
non-inferiority margin of 1.0). Notably, patients treated with GCPGC had a statistically significant reduction in the time to recovery
from neutropenia (ANC >2,000/mm3), one of the secondary endpoints of the study, compared to those treated with pegfilgrastim
(8.85 1.45 vs. 9.83 1.20 days; p <0.0001). There was no difference between groups regarding the other secondary endpoints.
In addition, no significant differences were observed between the safety profiles of the two groups.
Conclusion
Collectively, GCPGC is not inferior to peg-filgrastim and represents an effective alternative for reducing the duration of
neutropenia in breast cancer patients receiving TAC or DA chemotherapy.

Title: Association of serum prealbumin (transthyretin) with anthracycline chemotherapy toxicity in patients with invasive breast
cancer
A Armengol-Alonso1, K Zamudio-Osuna1, I Cortes-Franco1, P Milke-Garca1 and E Len-Rodrguez1. 1Instituto Nacional de
Ciencias Mdicas y Nutricin "Salvador Zubirn" (INCMNSZ), Mexico City, DF, Mexico.
Body: Background: In Mexico about 60% of invasive breast cancer (IBC) cases are diagnosed in advanced or locally-advanced
stages that require systemic treatment. The chemotherapy toxicity (CTox) depends largely on the free fraction of the drug, i.e. the
fraction unbound to plasmatic proteins. There is evidence that the Prealbumin (P-Alb) is a good marker of hepatic protein
production. Some studies relate reduced serum proteins with CTox. The aim of the study was to correlate serum P-Alb with the
degree of hematologic, gastrointestinal and skin toxicity.
Patients and Methods: Prospective cohort from January 2010 to December 2011 which recruited 41 patients with newly
diagnosed IBC who received chemotherapy FAC-50 (5-fluorouracil, Adriamycin and Cyclophosphamide). Anthropometric
measurements and serum determination of blood count, total protein, albumin and prealbumin were performed at baseline. The
presence and degree of toxicity was evaluated on day +12 and +21 of first cycle. Toxicity was evaluated by CTCAE (Common
terminology criteria for adverse events) v.4.0 We used Mann-Whitney s U test and Spearman correlation. The cutoff points for
P-Alb were calculated by ROC curves.
Results: Median age was 50 years, 56% and 17% were diagnosed in cTNM stage II-III. Eighteen patients (43.9%) had grade (G)
3-4 neutropenia and 2 (4.9%) had febrile neutropenia. Nine patients (21.9%) had G. 3-4 vomiting and 3 (7.3%) patients had G.
3-4 diarrhea. We classified the patients according to neutropenia, group 1 (N= 23 patients) (56%) without or G. 1-2 toxicity; and
group 2 (N=18 patients) (43.9%) with G. 3-4 toxicity. No significant differences in age, cTNM stage, performance status, total
body surface area, and body mass index between the groups were identified. Lower basal P-Alb (23.1 mg/dl) was observed in
group 2 compared to group 1 (28.3 mg/dl). (p = 0.001). We observed significant differences between the P-Alb medians and
degree of neutropenia, vomiting and diarrhea (p = 0.003, p = 0.001 and p = 0.025). The G. of neutropenia, vomiting and diarrhea
were negatively correlated with P-Alb seric concentration (r = -0594, p = <0.0001, r = -0418, p = 0.006 and r = -0642, p =
<0.0001) respectively. A cutoff of P-Alb 25 mg/dl for G. 3-4 neutropenia had a sensitivity (S) of 77% and (E) specificity 87%.
(AUC 0.84). For diarrhea of any G. a cutoff of P-Alb 26.8 mg / dl had a S 78% and E 88% (AUC 0.80). A cutoff P-Alb 26 mg/dl
for vomiting of any G. had a S 90% and E 76% (AUC 0.86). HR for grade 3-4 neutropenia was 23.3 (95% CI 18.8-144) (p =
0.011); diarrhea any G. HR 28.5 (95% CI 3.03-268) (p = 0.0001) and vomiting any G. HR 28.8 (95% CI 4.89-169) (p = 0.011).
Conclusion: To our knowledge this is the first prospective study describing that P-Alb is negatively correlated with the presence
and grade of hematologic and gastrointestinal toxicity secondary to anthracycline breast cancer chemotherapy. A cutoff value of
prealbumin <26 mg/dl has a good diagnostic accuracy for predicting CTox. (neutropenia, vomiting and diarrhea). The baseline
P-Alb can identify patients at increased risk of severe CTox. Our results should be validated in other cohorts with a larger number
of patients.

Title: Factors influencing aromatase inhibitor induced musculoskeletal syndrome: Roles of menopause timing and osteoporosis
therapy
Clinton R Morgan1, Zsolt Kulcsar1, Jonathan D Jones1, William F Rigby1 and Peter A Kaufman1. 1Dartmouth-Hitchcock Medical
Center, Lebanon, NH.
Body: Background: Aromatase Inhibitor (AI) therapy is the most effective hormonal treatment in post-menopausal estrogen
receptor (ER) positive breast cancer. Side effects such as arthralgias - termed aromatase inhibitor induced musculoskeletal
syndrome (AIMSS) - limit their use in some patients. We evaluated factors associated with AIMSS and explored possible
therapeutic options in a large cohort of patients.
Methods: We performed an IRB-approved retrospective review of breast cancer patients seen in the Norris Cotton Cancer
Center clinics from April 2011 to January 2013. 378 patients were included in our chart review on the basis of taking an AI for
breast cancer with follow up documented in the electronic health record. Statistical analysis was performed by chi squared test for
dichotomous variables and students t-test for continuous variables.
Results: In our cohort 91% of patients were taking an AI as adjuvant therapy (9% for metastatic disease) with 41% (n=153)
reporting new or worsening arthralgias after initiation of an AI. AIMSS was 42.5%(95%CI: 0.375 to 0.478) in the adjuvant and
22.7%(95%CI: 0.101 to 0.434) in the metastatic groups. The median time to symptom onset was 120 days. 2.1% (n=8)
discontinued AI therapy due to AIMSS. There was no association with prior chemotherapy, baseline arthralgia, BMI, or statin use.
We found an increased risk of developing AIMSS with more recent menopause (p=0.055), and therapy in the adjuvant setting
(p=0.067). We also note a potential association of baseline osteoporosis and osteoporosis therapies with lower rates of AIMSS.
Treatments included temporary discontinuation of AI, switching between AI, and non-steroidal anti-inflammatory therapy
(NSAIDs). Of those attempting such treatment, all had improvement with temporary discontinuation, 25% improved after AI
switch, and 85% had symptomatic benefit on NSAIDs.
Conclusions: The incidence of AIMSS in our review was 41%. Patients treated in the metastatic setting may have a lower rate of
AIMSS. Our cohort revealed that more recent menopause did seem to be a risk factor. Baseline osteoporosis and osteoporosis
treatments have a potential association to be explored. Management options included switching between AIs, temporary
discontinuation, and NSAID treatment. Updated analysis will be presented.
Potential risk factors for the development of AIMSS
Characteristic
(-)Arthralgia N(%)
(+)Arthralgia N(%)
p-value
Type of AI Used
Anastrozole
204
112(55.9)
92(45.1)
Letrozole
131
83(63.4)
48(36.6)
Exemestane
36
23(64.9)
13(36.1)
LMP < 5 years prior to AI start
151
79(52.3)
72(47.7)
LMP 5-10 years prior to AI start
40
23(57.5)
17(42.5)
LMP > 10 years prior to AI start
155
102(65.8)
53(34.2)
Adjuvant
348
200(57.5)
148(42.5)
Metastatic
22
17(77.3)
5(22.7)
Normal (T-score 0 to -1.49)
112
55(49.1)
57(50.9)
Osteopenia (T-score -1.5 to -2.49)
171
96(56.1)
75(43.9)
0.19
Menopause timing
0.055
Type of therapy
0.067
Baseline T-score by DEXA Scan

0.049
Osteoporosis(T-score < -2.5)
40
29(72.5)
11(27.5)
(-) Therapy
218
118(54.1)
100(45.9)
(+) Therapy
127
81(63.8)
41(32.3)
On active osteoporosis therapy

0.03
Bisphosphonate/denosumab; LMP=Last menstrual period, AI=Aromatase Inhibitor, DEXA=Dual-energy x-ray absorptiomotry

Title: Calcium and magnesium infusion for the prevention of taxane induced neuropathy in early stage breast cancer
Theresa Shao1, Joshua Kra1, Paula Klein1, Anupama Goel2, Stephen Malamud1, Tiffany Xing1, Johnny Chan2 and Michael L
Grossbard1. 1Mount Sinai Beth Israel Medical Center, New York, NY and 2Mount Sinai St Luke's and Roosevelt Hospitals, New
York, NY.
Body: Background: Taxane is an active drug in the treatment of breast cancer, but peripheral neuropathy is a major dose limiting
side effect. There are currently no effective drugs or treatment modalities for the prevention or treatment of taxane-related
neuropathy. We examined whether calcium and magnesium (Ca/Mg) infusions can reduce the incidence of neuropathy in patients
with early stage breast cancer who are treated with paclitaxel.
Methods: This was a pilot study evaluating the feasibility of Ca/Mg infusion to prevent taxane induced neuropathy in women with
early stage breast cancer receiving adjuvant or neo-adjuvant paclitaxel treatment, either given every 2 weeks for 4 cycles or every
week for 12 weeks. All patients received calcium gluconate and magnesium sulfate infusion, 1 g of each agent immediately
before and after each dose of paclitaxel. The primary endpoint was paclitaxel-related neuropathy grade 2 or greater as measured
by NCI Common Terminology Criteria Version 3 compared with historical controls. Secondary endpoints included other measures
of neuropathy and quality of life such as the Functional Assessment of Cancer Therapy-Taxane (FACT-Tax) score, taxane-related
neuropathic pain as measured by the Brief Pain Inventory-Short Form (BPI-SF). The endpoints were assessed in patients midway
through treatment, at the end of treatment and 4 weeks after finishing taxane therapy.
Results: We enrolled 50 patients, 47 patients were evaluable, and 3 patients were taken out of the study due to non-neuropathy
related side effects or progression of disease. Median age: 50.8 (range 27-71), White/Hispanic/Black/Asian/Other: 17/16/12/3/2.
Two patients received paclitaxel every 2 weeks, while the remainder received weekly therapy. Eight of 47 patients (17%) had
grade 2 neuropathy four weeks after treatment completed, while no patients had grade 3 or 4 neuropathy. This rate of neuropathy
is significantly lower compared to that seen in historical control where approximately 30% of patients develop grade 2 or greater
neuropathy. There were no significant changes in the quality of life measurements. There were no observed toxicities related to
the Ca/Mg infusion.
Discussion: Our study showed a decreased incidence of paclitaxel-related neuropathy in patients receiving Ca/Mg infusions when
compared to historical controls. The infusions are well tolerated without any side effects. Randomized studies are warranted to
further evaluate Ca/Mg infusion for the prevention of paclitaxel-related neuropathy.

Title: Cardiac toxicity in breast cancer patients: A single centre, retrospective review
Moira Rushton1, Freya Crawley2, Jeffrey Sulpher2 and Susan Dent2. 1Departments of Experimental Radiation Oncology;
2
Hematopathology, Ottawa, ON, Canada; 3Experimental Therapeutics; 4Pathology; 5Biostatistics; 6Radiation Oncology; 7Baylor
College of Medicine, Houston, TX, USA; 8Breast Medical Oncology, University of Texas MD Anderson Cancer Center, Houston,
TX, USA and 9Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, University of Texas MD Anderson
Cancer Center, Houston, TX.
Body: Background: While advances in breast cancer treatment, including targeted therapies (e.g trastuzumab) have improved
patient outcomes, such treatments may also be associated with both short and long term toxicity. In 2008, the Ottawa Hospital
Cardiac Oncology Clinic (COC) was established with the goal of providing timely access and care to cancer patients experiencing
cancer therapy-related cardiac toxicity. The purpose of this observational study is to review demographics, baseline
characteristics, referral patterns and clinical outcomes of breast cancer patients referred to a dedicated COC.
Methods: Breast cancer patients referred to the COC between October 2008 and December 2012 were reviewed. Data collected
included: age, sex, date of diagnosis, date of referral to COC, stage of breast cancer at diagnosis and referral, cardiac history,
medications, cancer therapy, treatment delays and completion rates, cardiac complications, cardiac test results and cardiology
interventions. Descriptive statistical analysis was performed.
Results: 263 breast cancer patients were assessed at the COC (median age = 57; range 23-87). The majority of patients (64%)
had early stage (I-II) disease and 215 (81.4 %) of patients had at least one identifiable cardiac risk factor. The most common
cardiac risk factors included: smoking (99/263, 37.6%), hypertension (87/263, 33.1%) and dyslipidemia (64/263, 24.3%). Reasons
for referral to the COC included reduced left ventricular ejection fraction (LVEF) (137/263, 52.1%) followed by pre-treatment
assessment (33/263, 12.5%) and cardiac arrhythmia (21/263, 8.0%). The majority of patients (243/263, 92.4%) received
chemotherapy, 188 (77.4%) with first line anthracycline regimens, of those a minority 10/263 (3.8%) stopped chemotherapy for
cardiac related toxicity, while 21/263 (8.0%) patients experienced cardiac related chemotherapy delays. The majority of patients
160/263 (60.8%) received first-line trastuzumab and of these, 66/160 (41.2%) suffered cardiac toxicity, 55/66 due to reduced
LVEF. 23/160 (14.4%) discontinued therapy due to cardiac toxicity.
164/263 patients were on cardioprotective medications at time of COC referral. 100/263 referred patients had new cardiac
medications initiated, most commonly ACE inhibitors (n=67), beta-blockers (n=54) and angiotensin receptor blockers (n =15). 15
patients (5.7%) had invasive testing/procedures, such as cardiac catheterization. After COC assessment/treatment, 117/263
(44.5%) of patients experienced full recovery of LVEF; 27/263 (10.3%) experienced partial recovery, 104/263 (39.5%) had no
change, and 13/263 (4.9%) experienced continued decline. 39/263 (14.8%) patients were hospitalized for cardiac toxicity and
3/263 (1.1%) died of cardiac complications. Discussion: While the majority of breast cancer patients referred to a dedicated COC
complete systemic therapy, there remains a significant population who require delay or discontinuation of treatment secondary to
cardiac dysfunction. Tools to identify patients at higher risk of developing cardiac toxicity are urgently needed, so that appropriate
monitoring and treatment can be initiated. Future studies will also determine the impact of this clinic on the delivery of cancer
therapy and cardiac health, compared with non-COC referred patients.

Title: Lurbinectedin (PM01183) activity in BRCA1/2-associated or unselected metastatic breast cancer. Interim results of an
ongoing phase II trial
Judit Balmaa1, Cristina Cruz1, Judy Garber2, Jose A Perez Fidalgo3, Ana Lluch3, Nadine Tung4, Silvia Antolin5, Cristian
Fernandez6, Carmen Kahatt6, Sergio Szyldergemajn6, Arturo Soto Matos6, Sonia Extremera6, Jose Baselga7 and Steven J
Isakoff8. 1Hospital Vall d'Hebrn, Barcelona, Spain; 2Dana-Farber Cancer Institute, Boston, MA; 3Hospital Clnico de Valencia,
Valencia, Spain; 4Beth Israel Deaconess Medical Center, Boston, MA; 5Complejo Hospitalario Universitario A Corua, A Corua,
Spain; 6Pharma Mar S.A. Sociedad Unipersonal, Colmenar Viejo, Madrid, Spain; 7Memorial Sloan Kettering Cancer Center, New
York, NY and 8Massachusetts General Hospital, Boston, MA.
Body: Background: Metastatic breast cancer (MBC) is a clinically heterogeneous disease in which selective approaches are
needed to identify patients (pts) who will benefit the most from available therapies and avoid unnecessary toxicities.
Lurbinectedin (PM01183) is a new anticancer drug that binds to the DNA minor groove inducing double-strand breaks and
blocking transcription. It has significant in vitro and in vivo antitumor activity, particularly in breast cancer models. PM01183 is
more active against homologous recombination-deficient cell lines. Hence, MBC pts with deleterious germline BRCA mutations
might be more sensitive to PM01183 than those with sporadic tumors.
PM1183 activity was shown in different tumor types in clinical trials, especially in pts with platinum-resistant ovarian cancer
(objective response rate [ORR]: 30%).
Methods: MBC pts < 75-years-old with ductal or lobular carcinoma pretreated with 3 chemotherapy regimens for MBC,
measurable disease per RECIST v1.1, performance status (PS) 1 and adequate major organ function are being treated with
PM01183 7.0 mg flat dose i.v. every 3 weeks.
The primary aim is to evaluate the clinical efficacy of PM01183 in two cohorts of MBC pts: cohort A: BRCA+ (known germline
BRCA1/2 mutation) and cohort B: unselected (BRCA1/2 wild type or unknown mutation status).
The primary endpoint is confirmed ORR by RECIST v1.1. A futility analysis was planned when 20 and 30 pts were recruited in
cohort A and B, respectively. If at least 4 pts in cohort A and 3 pts in cohort B, achieve a response, recruitment in that cohort will
continue up to 53 and 64 total pts, respectively.
Results: As of June 2014, 56 pts had been enrolled. Cohort A/B (n: 21/35): Median age 40/52-years-old; Cohort A(%)/B(%): PS
0: 48/66; >2 metastatic sites: 58/40; most common sites of metastasis: lymph node 79/33, liver 48/60 and bone 42/48; triple
negative: 57/46; hormonal receptor +: 38/49; prior anthracyclines: 95/91, taxanes: 100/94, platinum: 52/26, PARPi: 29/0; cohort A:
BRCA 1/2 (%): 52/48.
Cohort A-BRCA 1/2+ (n=21)
Cohort B-unselected (n=35)
Median cycles (range)
4 (1-20)
3 (1-14)
Best overall response
(n= 17 evaluable)
(n= 34 evaluable)
CR
1 (6%)
PR
6 (35%)
3 (9%)
SD
6 (35%)
16 (47%)
PD
4 (24%)
15 (44%)
ORR (95%CI)
41% (18-67%)
9% (1.9-23.7%)
Median duration of response Monts (95%CI)
5.0 (0.1-12.8)
3.3 (1.4-5.1)
In an exploratory analysis, ORR in cohort A was higher in PARPi nave pts: 64% (7/11 pts).
Grade (G) 3-4 related adverse events occurred in >5% pts were myelosuppression (neutropenia 69%, febrile neutropenia 7%,
thrombocytopenia 14%); G3 fatigue 7%, transient transaminase increase 19% (G4: 2%), nausea 6% and dyspnea 6% (3 pts, 2 of
them due to pneumonitis). One patient with massive liver involvement and impaired function died on C1D4 due to multiorgan
failure possibly related to the study drug.
Conclusions: PM01183 has promising activity in pretreated MBC pts with BRCA mutation. Safety profile appears to be mostly
predictable and with non-cumulative toxicity. Treatment-related neutropenia was manageable with G-CSF and/or dose reduction.
After futility analysis, targeted activity has been met in cohort A (BRCA+), and recruitment will continue up to 53 evaluable
BRCA+ pts.

Title: Phase II randomised clinical study of first line chemotherapy plus metformin versus first line chemotherapy alone in HER2
negative, non diabetic, metastatic breast cancer patients: Final results of the MYME study
Alessandra Gennari1, Oriana Nanni2, Andrea DeCensi1, Samanta Sarti2, Andrea Freschi3, Alessandra Bologna4, Lorenzo Gianni5,
Laura Amaducci6, Francesco Rosetti7, Filippo Giovanardi8, Anna Fedeli9, Massimo Ambroggi10, Paolo Bruzzi11 and Dino Amadori2.
1
SC Oncologia Medica, EO Ospedali Galliera, Genoa, Italy; 2IRCCS Istituto Scientifico Romagnolo per lo Studio e la Cura dei
Tumori (IRST), Meldola (FC), Italy; 3Centro di Riferimento Oncologico, Aviano, Italy; 4Arcispedale S. Maria Nuova-IRCCS, Reggio
Emilia, Reggio Emilia, Italy; 5Ospedale degli Infermi, U.O. Oncologia Medica, Rimini, Italy; 6Ospedale degli Infermi, U.O.
Oncologia Medica, Faenza, Italy; 7AULSS n.13 di Mirano, U.O. Oncologia Medica, Mirano (VE), Italy; 8DH Oncologico, Ospedale
di Guastalla, Guastalla (RE), Italy; 9P.O.M. Bufalini, U.O. Oncologia Medica, Cesena (FC), Italy; 10Ospedale Civile di Piacenza,
U.O. Oncologia Medica, Piacenza, Italy and 11IRCCS San Martino IST, Genoa, Italy.
Body: Background: Epidemiological studies indicated that the presence of insulin resistance is an adverse prognostic factor in
MBC. Recently increasing interest has focused on metformin, an oral insulin- sensitizing drug widely prescribed for type 2
diabetes; unexpensive and well tolerated, metformin has also been shown to have direct antiproliferative properties in breast
cancer. We present here the final analysis of a phase II comparative multicentric study on the addition of metformin to first line
chemotherapy in MBC non diabetic patients.
Methods:
This is a Phase II randomized study of HER-2 negative MBC patients with measurable or non-measurable disease; no prior
chemotherapy for MBC was allowed. Patients were allowed to have had prior endocrine therapy for MBC and prior adjuvant
chemotherapy if completed at least 1 year prior to study entry. Patients were stratified by HOMA Index (>2,5 vs </= 2,5) and
center. Patients were randomly assigned to Arm A, non pegylated liposomal doxorubicin 60 mg/sqm plus cyclophosphamide
600mg/sqm (AC) plus metformin (M) 1,000 mg PO QD or to AC alone. Treatment was administered for 8 3-weekly cycles in both
arms, M was administered until disease progression. The primary endpoint was progression free survival (PFS). Secondary
objectives included activity, safety and evaluation of metabolic profile. Correlative studies evaluated 1) circulating tumor cells and
2) modulation of insulin-related genes in mRNA isolated from CTCs. Planned sample size was 112 patients (98 events).
Results:
As of June 8th, 2014, 108 patients had been randomised. Median age was 60 yrs (range 36-77); 87% of patients were ER+, 60%
had received prior adjuvant CT, with antracyclines in 51% of patients. Prior endocrine therapy for MBC was used in 39% of the
patients. Measurable disease was present in 74% of the patients. 48% of the patients were insulin resistant by HOMA Index >2.5
and 60% were overweight (BMI > 25: 16% were obese, BMI >30). At a median follow up of 16 months (range 1 48), median
PFS (ITT) was 9 months (95% CI 8-14) with AC + M and 11 months (95% CI 7-16) with AC alone, p=.84. No significant
interaction was detected between HOMA Index and treatment arm (p = 0.15). Median OS was 30 months (95% CI 14-NE) in Arm
A versus 27 (95% CI 17-33) in Arm B, p = .58. The most common toxicities observed were G3/4 neutropenia in 51.5% of patients
in arm A vs 69.6% in arm B, with Febrile Neutropenia observed in 2,2% and 5.4% of patients, respectively. As expected G2
diarrhea was reported by 11.1% of patients in Arm A.
Conclusions: The addition of M to AC in MBC patients receiving first line chemotherapy did not improve PFS compared with AC
alone. M seems to have a protective effect on hematological toxicity. Final results including translational data will be available at
SABCS 2014.


Title: Efficacy and safety of eribulin in combination with capecitabine in patients with metastatic breast cancer: an open-label,
phase II dose-confirmation study
Chris Twelves1, Muhammad Y Nasim2, Alan Anthoney1, Claudio I Savulsky3, Shuxin Yin4 and TR Jeffry Evans2. 1Leeds Institute of
Cancer and Pathology and St James's Institute of Oncology, Leeds, United Kingdom; 2Beatson West of Scotland Cancer Centre,
University of Glasgow, Glasgow, United Kingdom; 3Eisai Ltd, Hatfield, United Kingdom and 4Eisai Ltd, Woodcliff Lake.
Body: Background: Two phase III trials have demonstrated the efficacy of eribulin monotherapy in patients with metastatic
breast cancer (MBC) who have progressed after previous anthracycline and taxane chemotherapy. Capecitabine is also an
effective agent in patients with MBC. A phase Ib dose-escalation study has investigated two dosing schedules of eribulin in
combination with capecitabine to determine the recommended phase II dose. The aims of this open-label, non-randomized,
phase II, dose-confirmation study were to determine efficacy and safety of this combination in patients with MBC. Interim data to
April 2014 are provided here; final results will be presented at the meeting.
Methods: Eligible patients were females with MBC who had received up to 3 previous chemotherapy regimens in any setting
(including an anthracycline and a taxane, but excluding capecitabine), an ECOG performance status (PS) 1, and adequate
hematological, renal and hepatic function (at study entry, 20 [47.6%] patients had liver metastases; 24 [57.1%] had lung
metastases). Patients received 1.4 mg/m2 eribulin mesylate (equivalent to 1.23 mg/m2 eribulin [expressed as free base]) i.v. on
days 1 and 8 plus oral capecitabine 1000 mg/m2 b.i.d. on days 114 (21-day cycles) until disease progression, unacceptable
toxicity or death. Disease assessments (CT scans) were performed every 6 weeks. The primary outcome was the objective
response rate (ORR); secondary outcomes included clinical benefit rate (CBR), progression-free survival (PFS), and safety and
tolerability.
Results: 42 females (median age: 52.5 years [range: 32 74]; ECOG-PS of 0: n = 18 [42.9%]) were enrolled; patients received
study drug as a 1st-line (23.8% [n = 10]), 2nd-line (45.2% [n = 19]) or 3rd-line (31.0% [n = 13]) chemotherapy. A median of 8
(range: 130) treatment cycles were administered (59.5% of patients received 6 cycles), with 94.4% of eribulin doses and 94.5%
of capecitabine doses given as planned. The ORR was 42.9% and the CBR was 57.1% (Table), with a median PFS of 7.1 months
(95% CI: 4.4, 9.8). The most common grade 3/4 adverse events (AEs) were neutropenia (66.7%), leukopenia (14.3%), febrile
neutropenia (7.1%) and anemia, fatigue, nausea and peripheral sensory neuropathy (each 4.8%). AEs of special interest (any
grade) included: handfoot syndrome (26.2%), peripheral neuropathy (23.8%), and diarrhea (21.4%). The incidences of AEs
leading to study drug dose adjustment were: 47.6% of patients (dose interruption), 42.9% (dose reduction), and 2.4% (drug
withdrawal).
Table: Tumor response (investigator assessment)
Patients
N = 42
n (%)
n (%)
Complete response (CR)
1 (2.4)
Partial response (PR)
17 (40.5)
Stable disease (SD)
16 (38.1)
Progressive disease (PD)
3 (7.1)
Not evaluable
5 (11.9)
Objective response rate (CR + PR)
18 (42.9)
95% CI
27.7, 59.0
Clinical benefit rate (CR + PR + SD 6 months)
24 (57.1)
95% CI
41.0, 72.3
Conclusions: This phase II study suggests that eribulin in combination with capecitabine is efficacious in patients with MBC, with
a safety and tolerability profile consistent with previous data. Phase III studies of this combination in patients with MBC are
warranted.

Title: Change in survival of stage III and IV breast cancer patients from an institutional cohort: 1990-2007
Henry G Kaplan1, Judith A Malmgren2 and Mary K Atwood1. 1Swedish Cancer Institute, Seattle, WA and 2HealthStat Consulting
Inc, Seattle, WA.
Body: Background: Breast cancer (BC) survival has significantly improved over the last twenty years. Our investigation compares
change in BC survival over time for patients presenting with Stage III and IV disease to evaluate the impact of treatment changes
during this time period.
Methods: We examined breast cancer specific survival among a cohort of stage III and IV breast cancer patients age 40-64 years,
identified and tracked for mortality from 1990-2007 at our institution (n=710). Patients were all biopsy proven BC with diagnosis,
treatment and follow up recorded and tracked in our breast cancer specific registry database. Breast cancer staging was
converted to AJCC 7 to remove inconsistency over time. Pearson chi square tests were used for bivariate comparisons. Kaplan
Meier method was used for survival analysis with breast cancer death as the outcome for disease specific survival (DSS) and the
log rank test for survival comparisons.
Results: Mean age of patients was 51.32 years with 85% stage III (n=603) and 15% stage IV (n=107). The percentage of stage IV
patients presenting at our institution did not change over time (p = .197). Stage IV patients were less likely than stage III patients
to have surgery, receive radiation or adjuvant chemotherapy (all p<.001) (table). Stage IV patients that did receive systemic
therapy were less likely to receive taxane therapy (p = .005) and were more likely to receive non-anthracyline regimens (p = .001)
(table). There was no difference in hormone therapy given for the two groups. Five year disease specific survival (DSS) for all
years combined was 79% for stage III vs. 51% for stage IV [10 year DSS stage III = 62%, stage IV = 24%] (p<.001). When
stratified by diagnosis year, stage III patients had significant change over time for 5 and 10 year disease free survival with a 13%
improvement for 5 year DSS and 9% improvement for 10 year DSS over the 17 year time period [5 year DSS: 1990-94 = 71%,
1995-99 = 78%, 2000-07 = 84%; 10 year DSS: 1990-94 = 56%, 1995-99 = 60%, 2000-07 = 65% (p = .037). No change was
observed over time for stage IV breast cancer [5 year DSS: 1990-94 = 50%, 1995-99 = 46%, 2000-07 = 52%; 10 year DSS:
1990-94 = 17%, 1995-99 = 33%, 2000-07 = 20% (p = .461)].
Conclusions: The significant improvement in stage III survival is most likely related to treatment improvements for breast cancer in
the studied time period. The lack of improvement in survival for stage IV patients may be due to a biologic change in the nature of
metastatic tumors that impacts response to treatment or possibly the less aggressive use of systemic therapy in stage IV relative
to stage III patients.
Treatment Characteristics stage III and IV Breast Cancer (n=710)
Stage III
Stage IV
variables
N (%)
N (%)
p value
chemotherapy = no
14 (2%)
27 (25%)
<.001
radiation = no
59 (10%)
59 (55%)
<.001
surgery = no
1 (.2%)
52 (49%)
<.001
taxanes = no
270 (46%)
50 (63%)
.005
non-anthracycline regimen
108 (18%)
27 (34%)
.001
hormone therapy = yes
397 (66%)
70 (65%)
.933

Title: Efficacy of eribulin in patients with invasive lobular carcinoma of the breast: data from a pooled analysis
Javier Corts1, Jos Prez1, Yi He2 and Otto Metzger-Filho3. 1Vall d'Hebron Institute of Oncology, Vall d'Hebron University
Hospital, Universitat Autnoma de Barcelona, Barcelona, Spain; 2Eisai Inc, Woodcliff Lake, NJ and 3Dana-Farber Cancer
Institute, Boston, MA.
Body: Background: Invasive lobular carcinoma (ILC) represents the second most common breast cancer (BC) subtype and is
usually characterized as hormone-receptor positive, low-to-intermediate histologic grade, and human epidermal growth factor
receptor (HER) 2-negative. In the early-stage setting, ILCs are associated with lower rates of pathological response to
preoperative chemotherapy compared with invasive ductal carcinoma (IDC). This exploratory analysis investigated the magnitude
of benefit of single-agent eribulin for the treatment of advanced ILC using data from three clinical trials in women with advanced
BC. We also describe the patterns of response and survival outcomes compared with IDC.
Methods: Individual patient (pt) data from the experimental arms of two phase III studies (305 and 301) and a single-arm, phase
II study were pooled for the present analysis. Study 305 (EMBRACE) randomized pts treated with 2 lines of chemotherapy for
advanced BC to receive eribulin or treatment of physician's choice. In study 301, pts treated with 2 lines of chemotherapy for
advanced BC were randomized to receive eribulin or capecitabine. In the phase II study, pts who had received 3 lines of
chemotherapy were treated with eribulin. Overall survival (OS) and progression-free survival (PFS) analyses were adjusted by
study, estrogen receptor (ER) and HER2 status, and number of lines of therapy for advanced disease.
Results: The three studies included 1353 eribulin-treated pts. Of the 1152 pts included in the present analysis, 118 were
classified as ILC and 1034 as IDC. Median age of ILC and IDC pts was 58 years and 55 years, respectively. ER and/or
progesterone receptor (PgR) positivity was more common in ILC (ER = 69%, PgR = 55%) than IDC (ER = 60%, PgR = 48%),
while HER2 positivity was less frequent in ILC than IDC (9% vs 16%). A total of 52.5% of ILC and 61.4% of IDC pts received 3
lines of chemotherapy (for any stage BC) prior to eribulin. Pts with ILC and IDC had similar median OS (13.4 vs 13.5 months;
hazard ratio [HR] = 1.10; 95% confidence intervals [CIs] 0.87, 1.38) and PFS (4.1 vs 3.6 months; HR = 0.91; 95% CIs 0.72, 1.14).
Investigator-evaluated tumor response rates are shown in the table.
Table: Best overall tumor responses (by investigator review)
Response, %
Invasive lobular carcinoma, n =

110
Invasive ductal carcinoma, n =

985
Objective response rate: CR + PR (95% CIs)
15.5 (9.3, 23.6)
14.8 (12.7, 17.2)
Complete response (CR)
0.4
Partial response (PR)
15.5
14.4
Stable disease (SD)
53.6
52.7
Progressive disease (PD)
26.4
28.9
Clinical benefit rate: CR + PR + SD 6 months (95%

CIs)
29.1 (20.8, 38.5)
28.6 (25.8, 31.6)
Not evaluable
4.5
3.6
Conclusions: In this exploratory, pooled analysis, magnitude of benefit from single-agent eribulin did not differ between the ILC
and IDC cohorts. While there was a limited numbers of pts with ILC, response rates, PFS, and OS were similar for the two pt
groups. The results with eribulin for advanced ILC contrast with data for other agents in early-stage settings, where ILC is
generally less responsive to chemotherapy than IDC. These findings may, however, underline changes in the disease biology
after exposure to previous therapies or changes inherent to disease progression.

Title: Does treatment response to new agents differ between visceral and non-visceral metastatic breast cancer: A systematic
literature review of registration trials
Rachel Wrstlein1, Maximilian Bardenhewer1, Alexander Knig1, Thomas Kolben1, Caroline Gehring1 and Nadia Harbeck1. 1Breast
Center, University of Munich (LMU), Munich, Germany.
Body: Differential efficacy of newly registered therapies in subgroups of metastatic breast cancer (mBC) is an important
consideration for their subsequent use in clinical practice. Unfortunately, such subgroup analyses are often exploratory, rarely
statistically adequately powered and may thus be misleading. In a systematic literature review, we evaluated differences in
outcome (taking into account the diversity of available study data, inhomogeneous in- and exclusion criteria and
subgroup-analyses) regarding progression free survival (PFS), time to progression (TTP), overall survival (OS) and visceral
versus non-visceral disease. The impact of HER2- and hormone receptor-status was also considered.
Methods: A systematic literature search (6362 hits) in the meta-Database PubMed (U.S. National Library of Medicine) was
performed for the last 20 years. 257 studies (n=126.291) were included for further analysis. 69 studies had published data for
visceral (i.e. presence of visceral metastases independent of the presence of other metastasis sites) vs. non-visceral (i.e.
non-visceral metastases in the absence of visceral involvement) disease including phase III trials plus studies that had further
used their data. Out of these 69 studies we selected n=16 studies (n=13.083) by considering (A) the information given in the
medical product's professional information and (B) the decision of the U.S. Food and Drug Administration or the European
Medicine Agency for the approval of the respective therapeutic agents which are now used in treatment of mBC. All selected 16
studies had looked at the endpoints mentioned above. In order to achieve comparability, we extracted the information of hazard
ratios (HR), confidence intervals (CI) and times in weeks (if available) for PFS, TTP, OS of the entire study population, which was
divided into three groups: HER2-positive, HER2-negative, unknown HER2 status.
Results: No statistically significant difference in treatment response was found in mBC patients with visceral vs. non-visceral
metastasis looking at HRs and CIs. Relevant, yet not statistically significant differences were found in the specific response of
visceral metastases to modern combination therapies, especially in HER2-positive breast cancer: There was an increased benefit
regarding OS using Lapatinib combined with Trastuzumab or Trastuzumab and Docetaxel combined with Pertuzumab.
Additionally, in two chemotherapy trials, there was a numerical difference between therapy response in visceral vs. non-visceral
metastases regarding PFS in the unknown HER2 group, and regarding OS in the HER2-negative group.
Conclusions: Using a systemic literature search, we stratified published studies of the last 20 years considering HER2 and
hormone receptor status with respect to metastasis pattern. In the subgroup analyses, we did not find any significant differences
in response rates for visceral vs. non-visceral metastasis. For targeted therapies based on a biomarker, there seems to be a
beneficial effect of combination therapies regarding OS in visceral disease. At the present time, metastasis localization should not
be used as a predictive marker for choice of systemic therapy in mBC.

Title: Modification of the response to chemotherapy of HER2 negative metastatic breast cancer by lipids of marine origin: A
controlled, randomized, double blind dietary supplementation trial
Philippe Bougnoux1, Jacques Bonneterre2, Anne Mercier-Blas3, Patrick Souli4, Hlne Simon5, Franck Priou6, Christine
Piprot-Choffat7, Christelle Levy8, Caroline Goupille9 and Virginie Berger10. 1Inserm U 1069, CHU Bretonneau, Tours, France;
2
Centre Oscar Lambret, Lille, France; 3CHP Saint Grgoire, Saint Grgoire, France; 4ICO Paul Papin, Angers, Pays de la Loire,
France; 5CHU Morvan, Brest, France; 6CHD Les Oudairies, La Roche sur Yon, France; 7CHU, Amiens, France; 8Centre Franois
Baclesse, Caen, France; 9Inserm U 1069, CHU Bretonneau, Tours, France and 10ICO Paul Papin, Angers, Pays de la Loire,
France.
Body: Background: Long chain n-3 Polyunsaturated Fatty Acids (n-3 PUFA) of marine origin (docosahexaenoic acid, DHA and
eicosapentaenoic acid, EPA) have been shown to increase sensitivity of cell lines to anthracyclins or taxanes. Preclinical studies
and a phase II clinical trial have shown that increasing DHA level in the body tissues through a dietary supplementation improves
chemotherapy efficacy. We have set up a phase III controlled, double blind study to determine whether a dietary supplementation
with n-3 PUFA would increase PFS in patients receiving chemotherapy for metastatic breast cancer (MBC).
Methods: Inclusion criteria were patients with a ductal or lobular breast carcinoma, positive hormone receptors, Her2 (-), who
have developed visceral metastasis, and were due to receive a 1st or 2d line chemotherapy. Dietary intervention was carried out
during chemotherapy. Patients had to take daily at each meal a can* (medical food) containing either fish oil (1.56 g/d of DHA and
2.64 g/d EPA, experimental arm) or coprah oil (short chains fatty acids, control arm). Principal endpoint was PFS, secondary were
objective response, overall survival, dietary tolerance to cans, lipids plasma levels of n-3 PUFA.
Results: Sixty five patients with MBC have been prospectively enrolled in the DHALYA trial (NTC01548534). The first 45 patients
who completed chemotherapy along with dietary supplementation are evaluable. Plasma fatty acid level was measured at
baseline, at C1 after 10 days loading dose, at C3, and at C6 or at withdrawal. Level of EPA and DHA increased in half of the
patients, thus allowing a putative allocation into either arm: arm A with induced elevation of PUFA levels (N=22), and arm B
without any change in PUFA levels (N=23). No difference was observed in the quality of the dietary intervention (number of cans,
duration) among arms. Distribution of patients according to demographics (age, menopausal status, histology, SBR grade), or
type or quality of chemotherapy received (anthracycline- or taxane-based, number of cycles, length) was similar between arms.
However a significantly greater proportion of patients with poor prognosis at staging (larger tumor size and greater axillary lymph
nodes involvement) were observed in arm A. There was no difference in side effects (grade 2) between arms. In terms of
efficacy, median PFS was 14.3 (arm A) and 12.5 (arm B) months, not statistically different yet.
Conclusion: A dietary intervention targeted on marine-derived PUFA during systemic chemotherapy for MBC is safe and
feasible. A longer follow-up is required in order to know whether chemotherapy efficacy can be increased.
This study was supported by a grant from the French Ministry of Health, PHRC-11-153. *Cans were graciously provided by
Nutrialys Medical Nutrition, Saint Grgoire, France.

Title: Personalized therapy in breast cancer: Brazilian experience
Raphael Brandao Moreira1, Matheus Bongers Alessandretti1, Antonio Carlos Buzaid1 and Marcelo Rocha Cruz1. 1Centro
Oncolgico Antonio Ermrio de Moraes, So Paulo, Brazil.
Body: INTRODUCTION
Cancer biology is characterised by genomic alterations. There is an ongoing revolution in cancer research due to the emergence
of novel technologies based on next generation sequencing (NGS).
OBJECTIVES
To identify abnormalities in individual patients with the aim of providing targeted therapy matched to individuals genomic
alterations.
METHODS
From April 2013, to February 2014, we included patients who had metastatic breast cancer with lesion accessible for biopsy. The
tumor tissue stored in paraffin was evaluated by the method of next-generation sequencing (NGS) where 236 cancer-related
genes were sequenced.
RESULTS
Fourteen patients were included and NGS was feasible. An actionable genomic alteration was identified in 12 (85,7%) patients,
most frequently in TP53 (10 [83,3%], PIK3CA (8 [66,6%]), AKT (5 [50%]), FGFR1 (4 [33%]), CCND1 (3 [25%]), PTEN (3[25%])
and BRCA2 (2[16%]). Therapy could be personalized in 12 of 14 patients. Of the 8 patients who received targeted therapy, 5
presented objective response.
Results
Patients
Target Therapy
Objective Response
Mutations
Everolimus
NO
ERBB2, PIK3CA
Everolimus
Yes
PI3KCA, FGFR, AKT
Everolimus
Yes
PI3KCA, FGFR
Everolimus
Yes
AKT1, PI3KCA
Everolimus
Yes
AKT1
Everolimus
Yes
AKT1, FGFR
Everolimus
Yes
AKT1. FGFR
Everolimus
No
ERBB2, PI3K
Table1
Mutations Prevalence
Mutation
Prevalence
TP53
83,30%
PIK3CA
60,60%
FGFR
33%
CCND1
25%
PTEN
25%
BRCA2
16%
AKT
50%
Table2
CONCLUSION
. NGS was able to find genomic alterations in the majority of breast cancer tissues;
. Most patients did not have "actionable" mutations for which a specific therapy could be offered;
. We are just at the dawn of exploring this strategy and NGS can effectively contribute for increasing the understanding of the
disease.
REFERENCES
1. Jeffrey. Ross, Massimo Cristofanilli et al, JClin Oncol 3, 2013 (suppl; abster 1009).

Title: The maintenance treatment after xeloda contained regimens with xeloda or endocrine for HR positive and HER2 negative
MBC patients
Zefei Jiang1, Fan Qi1, Shaohua Zhang1, Li Bian1, Tao Wang1 and Lei Li1. 1Affiliated Hospital of Academy of Military Medical
Science, Beijing, China.
Body: Objective To discuss the optimizing selection of maintenance treatment after xeloda-contained chemotherapy regimen
being applicated in HER-2 negative, HR positive relapsing metastatic breast cancer(MBC) patients, and contrast the curative
effect of maintenance treatment with xeloda or endocrine. Methods 119 patients with HER-2 negative, HR positive were enrolled
during 2009-2013.All patients taking the xeloda contained chemotherapy regimens to obtain CR,PR or SD were randomly divided
into group A(65 cases, maintenance capecitabine therapy 800-1000mg/ m2 bid, orally, Days 1 14, q3w)and group B(54 cases,
a switch to maintenance endocrine therapy). Endpoints include overall survival (OS) and progression-free survival (PFS). Results
The ages, menopausal status, number of metastases and treatment line numbers had no statistical differences between two
groups, and the p value of two therapy co-operative groups was 0.002. There were 37 patients(56.9%)using TX and
28(43.1%)using NX in therapy co-operative group A. There were 39 patients (72.2%)using TX, 9(16.7%)using NX and 6(6%)
using regimen containing xeloda in therapy co-operative group B. Progression-free survival was 8 months in patients who follow
by xeloda and 12 months in those who follow by endocrine (p<0.01).Conclusion Endocrine therapy after xeloda-contained
chemotherapy regimen tend to be more effective than single xeloda therapy for patients with HER-2 negative, HR positive. This
results request to be verified by advanced OS data, which was needed to in-depth study in clinic and explore relevant biological
indicators.

Title: Response to treatment and prognosis of recurrent breast cancer patients with receptor discordance
Hanako Ueno1, Hiromitsu Jinno1, Takamichi Yokoe1, Toshiaki Kurihara1, Masaru Takemae1, Aiko Nagayama1, Maiko Takahashi1,
Tetsu Hayashida1, Kaori Kameyama2 and Yuko Kitagawa1. 1Keio University School of Medicine, Tokyo, Japan and 2Keio
University School of Medicine, Tokyo, Japan.
Body: BACKGROUND
Estrogen receptor (ER), progesterone receptor (PgR) and human epidermal growth factor receptor -2 (HER2) statuses are
clinically used to select treatments. Several studies reported that discordance between primary and metastatic lesions lead to
detrimental outcome. Although biopsy of recurrent breast cancer has been recently recommended by international clinical
guidelines, prognostic relevance remains to be elucidated. The aim of the present study is to evaluate response to treatment and
prognosis of patients with receptor discordance, compared with patients with receptor concordance.
Patients and METHODS
We retrospectively identified recurrent breast cancer patients who had biopsies or resections of recurrent lesions between
January 2007 and April 2012 at Keio University Hospital. HR status was assessed by immunohistochemistry (IHC) and
determined using the Allred score. HR status was defined as positive when score was 3 and more. HER2 status was assessed by
IHC and fluorescence in situ hybridization (FISH) analysis. We defined HER2 positivity as 3+ staining intensity by IHC or the
presence of HER2 gene amplification by FISH. Tumors were classified as luminal (HR+ and HER2-), luminal/HER2 (HR+ and
HER2+), HER2 (HR- and HER2+), or triple negative (HR- and HER2-). Treatment was decided according to the receptor status of
recurrent tumors.
RESULTS
Among 38 patients undergoing biopsy or resection, 13.2% (5) were loco-regional recurrences and 86.8% (33) were distant
metastases (lung 21; liver 6; brain 3; pleura 1). Overall, 10 patients (26.3%) changed subtypes at recurrent lesions (Table 1) and
all of them had a change in treatment plan.
Discordance in subtype between primary tumor and recurrent lesion
Recurrent lesion (%)
Primary tumor (%)
Total
Luminal
Luminal/HER2
HER2
Triple negative
Discordance rate (%)
Luminal
18 (66.7)
2 (8.3)
1(4.2)
3(12.5)
6/24 (25.0)
Luminal/HER2
1 (14.3)
6 (85.7)
1/7 (14.3)
HER2
1 (100.0)
1/1 (100.0)
Triple negative
1 (16.7)
1 (16.7)
4(66.7)
2/6 (33.3)
20
10
10/38 (26.3)
Changes in management included the addition of trastuzumab in patients with gain of HER2 (n=3), the use of chemotherapy in
those with loss of HR (n=6) and provision of endocrine therapy for those with gaining HR (n=3). Response rate of discordant
group and concordant group were 10.0% and 21.4%, respectively (p=1.000). Clinical benefit rate of discordant group and
concordant group were 40.0% and 67.9%, respectively (p=0.150). There is no significant difference of time to progression (TTP)
between the discordant and concordant groups (169.5days vs.319.5days, p= 0.081).
CONCLUSION
Patients with receptor discordance tended towards worse response rate and shorter TTP, leading to the poor prognosis of the
recurrent breast cancer patients with receptor discordance.

Title: Circulating tumor cell (CTC) enumeration and HER2 assessment as predictors of breast cancer outcomes in the ALTTO
(BIG 2-06, Alliance N063D) Trial
Minetta C Liu1, Brigitte Rack2, Amylou C Dueck4, David W Hillman1, Michael B Campion1, Monica M Reinholz3, Kevin C Halling1,
Christos Sotiriou5, Franoise Roth5, Marion Maetens5, Ghizlane Rouas5, Wolfgang Janni6, Antonio C Wolff7, Lyndsay N Harris8,
Julie R Gralow9, Kathleen I Pritchard10, Susan Ellard12, Nguyet A Le-Lindqwister13, Frances Boyle14, Evandro De Azambuja5,
Martine J Piccart-Gebhart5, Michail Ignatiadis5 and Edith A Perez11. 1Mayo Clinic, Rochester, MN;
2
Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany; 3Ventana Medical Systems, Inc, A Memberof the Roche Group,
Tuscon, AZ; 4Mayo Clinic, Scottsdale, AZ; 5Institut Jules Bordet, Brussels, Belgium; 6Universitts-Frauenklinik Ulm, Ulm,
Germany; 7Johns Hopkins Hpsital/Sidney Kimmel Comprehensive Cancer Center, Baltimore, MD; 8Case Western/Seidman
Cancer Center, Cleveland, OH; 9Seattle Cancer Care Alliance, Seattle, WA; 10Sunnybrook Odette Cancer Centre and University
of Toronto, Toronto, ON, Canada; 11Mayo Clinic, Jacksonville, FL; 12British Columbia Cancer Agency, Southern Interior, Kelowna,
BC, Canada; 13Illinois Cancer Care-Peoria, Peoria, IL and 14Patricia Ritchie Centre for Cancer Care and Research, University of
Sydney, Mater Hospital, North Sydney, Australia.
Body: Background: CTCs are associated with clinical outcomes in metastatic breast cancer irrespective of ER/PR/HER2 status.
Some data support the prognostic relevance of serial CTC enumeration relative to adjuvant chemotherapy in early stage breast
cancer. However, data from a large scale study focused on HER2 directed therapy for HER2+ disease have been lacking. We
therefore sought to prospectively evaluate the effect of trastuzumab +/- lapatinib on CTCs and assess the prognostic/predictive
value of CTC monitoring in HER2+ early stage breast cancer patients (pts).
Methods: The Adjuvant Lapatinib and/or Trastuzumab Treatment Optimisation (ALTTO; NCT00490139) Trial is an international,
randomized, open-label phase III study of two targeted agents for HER2+ breast cancer. From June 2007 to July 2011, 8381 pts
were randomised from 946 sites in 44 countries to 1 of 4 arms with sequential or concurrent chemotherapy: (i) 52 wks of
trastuzumab (T); (ii) 52 wks of oral lapatinib (L); (iii) 12 or 18 wks of T followed by a washout and then 34 or 38 wks of L; or (iv) 52
wks of L+T. 540 (6%) pts provided optional informed consent and up to 30 mL peripheral blood suitable for CTC analyses at
baseline with additional collections at 13 or 19 wks, 52 wks, 18 mos, 24 mos, and recurrence. CTC analyses are being conducted
in three laboratories (Mayo Clinic Rochester, n=431; Institut Jules Bordet and University of Munich, n=109). 2-3 x 10 mL
CellSave samples are pooled and processed at each time point for CTC enumeration and HER2 expression using the
immunomagnetic/immunofluorescence assay (CellSearch). A round-robin concordance project was done between Mayo Clinic
Rochester and Institut Jules Bordet before embarking on the primary correlative work.
Results: At baseline, 20% pts had detectable (i.e., 1) EpCAM+/CK+/DAPI+/CD45- CTCs, and 16% pts had detectable
EpCAM+/CK+/DAPI+/CD45-/HER2+ CTCs. Correlative analyses with clinical outcome are ongoing with plans for completion by
Fall 2014.
Conclusions: CTCs were detected in 20% of pts with HER2+ early stage breast cancer. This is similar to the frequency of
detection in mixed early stage breast cancer populations relative to ER/PR and HER2 status. Concordance of enumeration and
HER2 assessments between the two experienced laboratories, and correlation between disease free survival and CTC findings
(from serial samples collected at baseline, during the course of HER2 directed therapy, and at set intervals of follow-up) will be
reported.

Title: Prospective characterization of HER2-positive circulating tumor cells in patients with HER2-negative metastatic breast
cancer
Ian E Krop1, Erin Macrae2, Sarah R Galler1, Farideh Bischoff3, Romeo Fauni3, Edgar Sales3, Lan Huynh3, Christine Mitchell3,
Trisky Clarin-Tamayo3, Mark Anderson3, Leslie Abad3 and Eric Winer1. 1Dana-Farber Cancer Institute, Boston, MA; 2Ohio State
University, Columbus, OH and 3Biocept, Inc, San Diego, CA.
Body: Background: We have previously shown feasibility by CLIA validation of HER2 FISH on circulating tumor cells (CTCs).
Several small retrospective studies have identified HER2 amplification in CTCs in a subset of patients (pts) with clinically HER2
negative metastatic breast cancer (MBC). While these findings potentially have profound implications for CTCs as a predictive
biomarker, prospective validation and functional characterization of this subgroup is necessary.
Methods: We enrolled a prospective cohort of pts with MBC that was HER2 negative by IHC and/or FISH on all available primary
and metastatic biopsies. Pts had 1 line of prior chemotherapy for MBC. Blood samples were collected at study entry. CTCs were
enumerated based on standard criteria (Cytokeratin (CK)+/CD45- staining) as well as by FISH for HER2 using the
OncoCEE-BR CTC test (Biocept, Incorporated). This test employs a proprietary antibody cocktail for capture followed by
CK/CD45 staining, and FISH analysis directly within a microfluidic device. Samples were reported as positive if the HER2/CEP17
ratio was 2.0. HER2+ CTCs were classified into two categories: CK+/CD45-/HER2+ and CK-/CD45-/HER2Results: CTCs were observed in 208 out of 323 pts (64%). Median number of CTCs was 10 (range 1 to > 34195). 75 pts (23%)
had HER2+ CTCs, with a median number of 3 HER2+ CTCs (range 1 to 21). 36% (27/75) of these pts had CK+/HER2+ CTCs
and 43% (32/75) of pts had only CK-/HER2+ CTCs. The remaining pts (21%) had both CK+/HER2+ and CK-/HER2+ CTCs
present.
Conclusion: HER2 amplified CTCs are present in a subset (23%) of pts with clinically HER2-negative breast cancers.
Interestingly, we observed a high prevalence of pts with only CK-/HER2 amplified CTCs (32 out of 323; 10%). The unique
multi-antibody CTC capture method used here thus allows for detection of a prevalent population of CK-/HER2+ CTCs that may
be largely undetected by other current adopted technologies. The functional significance of CK-/HER2+ and CK+/HER2+ CTCs in
patients with clinically HER2 negative breast cancer is currently being evaluated in a prospective study with HER2-directed
therapy.

Title: Prognostic relevance of circulating tumor cells across different molecular subgroups in the adjuvant SUCCESS-A study
Wolfgang Janni1, Andreas Schneeweiss3, Lothar Hberle2, Peter A Fasching2, Lukas Schwentner1, Mahdi Rezai8, Jrn Hilfrich10,
Hans Tesch9, Georg Heinrich12, Helmut Forstbauer11, Thomas Friedl1, Fabienne Schochter1, Susanne Albrecht1, Bernadette
Jger1, Julia Jckstck4, Tanja Fehm5, Volkmar Mller6, Klaus Friese4, Werner Lichtenegger7, Matthias B Beckmann3 and Brigitte
Rack4. 1University of Ulm, Ulm, Germany; 2University Hospital Erlangen, Friedrich-Alexander University Erlangen-Nuremberg,
Comprehensive Cancer Center Erlangen-EMNUniversity Hospital, Erlangen, Germany; 3National Center for Tumor Diseases,
University of Heidelberg, Heidelberg, Germany; 4University Hospital Ludwig-Maximilians-University, Munich, Germany;
5
Duesseldorf University Hospital, Duesseldorf, Germany; 6University Medical Center Hamburg-Eppendorf, Hamburg, Germany;
7
Charit Berlin, Berlin, Germany; 8Luisenkrankenhaus, Dsseldorf, Germany; 9Onkologische Gemeinschaftspraxis am
Bethanien-Krankenhaus, Frankfurt, Germany; 10Eilenriede Klinik Brustzentrum, Hannover, Germany;
11
Hmatologisch-Onkologische Schwerpunktpraxis, Troisdorf, Germany and 12Schwerpunktpraxis fr Gynkologische Onkologie,
Frstenwalde, Germany.
Body: Aim: The prognostic value of circulating tumor cells (CTCs) in the adjuvant setting has recently been demonstrated in the
SUCCESS A Study (Rack et al. JNCI 2014). As breast carcinomas depend on partly different pathways for progression, the
relevance of CTCs could differ between molecular intrinsic subtypes of breast cancer. Aim of this study was therefore to analyze
the prognostic impact of CTCs in molecular subtypes of a large patient cohort.
Methods: Within the adjuvant SUCCESS A Study, patients were treated either with 5-Flourouracil, Epirubicin and
Cyclophosphamid (FEC) followed by Docetaxel (D) or with FEC followed by D and Gemcitabine (DG). There was no restriction
with regard to molecular subtype, however a high recurrence risk was required for study entry. In addition patients were assessed
prospectively for the presence of CTCs before chemotherapy. Molecular subtypes were defined as: triple negative (TN), hormone
receptor positive and grading 1/2 (LUM A like), hormone receptor positive and grading 3 (LUM B like), HER2 positive (HER2 like).
We studied whether the addition of CTC status (0 CTC vs > 0 CTCs) to well-known predictors such as age, BMI, tumor size,
lymph node status improved the prediction of overall survival (OS) and disease free survival (DFS) across all patients and
especially within molecular subtypes using likelihood ratio tests, which compared multivariable Cox regression models with and
without CTC and the interaction between CTC and molecular subtype.
Results: Information about molecular subtype and CTCs was available in a total of 1994 patients. At least one CTC was seen in
422 (21.2%) of patients. 383 (19.2%) were TN, 798 were LUM A like (40.0%), 328 (16.4%) were LUM B like (16.4%) and 485
(24.3%) were HER2 like. The effect of CTC on overall survival had a HR of 2.50 (95%: 1.75 to 3.58) for the entire cohort.
However as the effect was different across subtypes (p=0.04, likelihood ratio test), subtype specific HR were calculated. The
effects on OS were most prominent in LUM B like patients (HR=3.96; 95%CI: 1.93 to 8.14) and LUM A like patients (HR=3.57;
95%CI: 1.81 to 7.03), less strong in HER2 like (HR=2.35; 95%CI: 1.04 to 5.32) and not present in TN patients (HR=1.18; 95%CI:
0.62 to 2.24). CTC status had a clear effect on DFS as well (HR=1.93, 95%CI: 1.48 to 2.52). It could not be shown that this effect
was different across subtypes (p=0.07, likelihood ratio test). However, the effect size was similarly distributed like the ones for
OS.
Conclusion: With regard to OS the prognostic effect of CTCs in this study cohort seems most prominent in patients with
hormone receptor positive disease. It is still significant in HER2 positive, but not in TN breast cancer patients. Results with regard
to DFS trended into the same direction, differences within subgroups could however not be shown, possibly due to power
reasons.

Title: Circulating tumor cells (CTC) are associated with defects in innate and adaptive immunity in inflammatory breast cancer
(IBC) patients
Michal Mego1,2, Hui Gao1, Evan N Cohen1, Simone Anfossi1, Antonio Giordano1, Sanda Tin1, Tamer M Fouad1, Wendy A
Woodward1, Ricardo H Alvarez1, Vicente Valero1, Naoto T Ueno1, Gabriel N Hortobagyi1, Massimo Cristofanilli1,3 and James M
Reuben1. 1University of Texas MD Anderson Cancer Center, Houston, TX; 2Comenius University, Faculty of Medicine, Bratislava,
Slovakia (Slovak Republic) and 3Thomas Jefferson University, Kimmel Cancer Center, Philadelphia, PA.
Body: Background: CTCs play a crucial role in tumor dissemination and are prognostic factor in primary and metastatic breast
cancer patients. Immune cells in peripheral blood (PB) contribute to an unfavorable microenvironment for the CTCs survival. As
such, effective host innate and adaptive immune surveillance systems could adversely influence tumor dissemination whereas
dysfunctional immune systems could provide a favorable microenvironment for the dissemination of CTCs and cancer
progression. This study aimed to correlate CTCs with the functions of innate [natural killer (NK) cells] and adaptive (T-cells)
immune effector cells in PB of IBC patients.
Methods: This prospective study included 65 IBC (21 non-metastatic, 14 de novo metastatic and 30 recurrent metastatic) patients
treated between October 2008 and April 2012 at the MD Anderson Cancer Center. CTCs were enumerated before patients
started a new line of chemotherapy using the CellSearch system, 33 (50.8%) of patients were treatment nave at the time of
blood collection. The phenotype of T cells, their ability to secrete cytokines following activation through the T-cell receptor (TCR)
and the NK cell subsets were analyzed by multiparameter flow cytometry and the results were correlated with CTCs and clinical
outcome. For survival analysis immune cell counts were dichotomized to low or high category using the median count.
Results: At least 1 CTC ( 1) or 5 CTCs per 7.5 mL of PB was detected in 40 (61.5%) or 21 (32.3%) of patients, respectively.
Patients with at least 1 CTC or 5 CTCs had a significantly inferior overall survival (OS) [HR=2.48, p=0.003 and HR=1.85,
p=0.045] than patients with no CTCs or with <5 CTCs, respectively. There was no correlation between CTCs count and total
lymphocytes; however, patients with at least 1 CTC or 5 CTCs had significantly lower percentages of CD3+ and CD4+ T-cells
compared with patients with no CTCs or < 5 CTCs, respectively. Patients with 1 CTC, had a lower percentage of TCR-activated
CD8+ T-cells producing TNF- (p=0.03) and IFN- (p=0.08), and a higher percentage of T regulatory lymphocytes (p=0.05)
compared to patients with undetectable CTCs. Moreover, CTCs 5 was inversely associated with the percentage of the following
NK cells subsets: non-ADCC NK (Spearman rho' = -0.30, p=0.02), ADCC NK (rho' = -0.15, p=0.20) and exhausted NK (rho =
-0.24, p=0.04). We also observed increased prognostic value of CTCs in the context of adaptive immune cells, with worse OS for
patients with 5 CTCs and low count of TCR-activated CD8+ T cells producing TNF- (HR=6.72, p=0.0007) compared with
patients with < 5 CTCs and high count of TCR-activated CD8+ that produced TNF-.
Conclusions: IBC patients with CTCs in PB had abnormalities in both innate and adaptive immunity as evidenced by low
percentages of NK cell subsets, and low percentage of TCR-activated CD8+ T cells producing TNF-, respectively. These data
illustrate an inverse relationship between CTCs and both innate and adaptive immune cells in the PB microenvironment that could
potentially impact tumor cell dissemination and initiation of the metastatic cascade. Moreover, immune cell profiling could add
further prognostic value to CTCs in IBC patients.


Title: Circulating and disseminated tumor cells from breast cancer patient-derived xenograft-bearing mice as a novel model to
study metastasis
Mario Giuliano1,2,3, Sabrina Herrera1,2, Pavel Christiny4, Chad Shaw2,5, Chad J Creighton2,6, Tamika Mitchell1,2, Raksha Bhat4,
Xiaomei Zhang1,2, Sufeng Mao1,2, Lacey Dobrolecki1,2, Ahmed Al-rawi4, Fengju Chen1,2, Bianca M Veneziani3, Xiang H Zhang1,2,6,7,
Susan G Hilsenbeck1,2,6, Alejandro Contreras1,2,8, Carolina Gutierrez1,2,8, Rinath Jeselsohn1,2,9, Mothaffar F Rimawi1,2,6, C Kent
Osborne1,2,6, Michael T Lewis1,2,6, Rachel Schiff1,2,6,7 and Meghana V Trivedi1,2,4,6. 1Lester and Sue Smith Breast Center, Baylor
College of Medicine, Houston, TX; 2Dan L. Duncan Cancer Center, Baylor College of Medicine, Houston, TX; 3University Federico
II, Naples, Italy; 4University of Houston College of Pharmacy, Houston, TX; 5Baylor College of Medicine, Houston, TX; 6Baylor
College of Medicine, Houston, TX; 7Baylor College of Medicine, Houston, TX; 8Baylor College of Medicine, Houston, TX and
9
Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA.
Body: Introduction: Real-time monitoring of biological changes in tumors that metastasize may be possible by investigating the
transitional cells such as circulating tumor cells (CTCs) and disseminated tumor cells in bone marrow (BM-DTCs). However, the
small numbers of CTCs and the limited access to bone marrow aspirates in patients with localized disease pose major hurdles.
The goal of this study was to determine if breast cancer (BC) patient-derived xenograft (PDX) mice could provide a constant and
renewable source of CTCs and BM-DTCs, thereby representing a unique system for the study of metastatic processes.
Methods: CTCs and BM-DTCs, isolated from BC PDX-bearing mice, were identified by immunostaining for human
pan-cytokeratin and nuclear counter staining of RBC-lysed blood and bone marrow fractions, respectively. The lung metastasis
(LM) rate was previously reported in these lines. Associations between the presence of CTCs, BM-DTCs, and LM were assessed
by the Fishers Exact and Cochran-Mantel-Haenszel tests. Two separate genetic signatures associated with the presence of CTC
clusters and with lung metastatic potential were computed using the gene expression arrays of primary tumors from different PDX
lines and were subsequently overlapped to identify common genes.
Results: A total of 18 BC PDX lines were evaluated. CTCs and BM-DTCs, present either as single cells or as clusters, were
detected in 83% (15/18) and 62.5% (10/16) of the lines, respectively. There was a positive association between the presence of
CTCs and BM-DTCs within the same mice. LM was previously found in 9 out of 18 (50%) lines, of which all 9 had detectable
CTCs. The presence of LM was strongly associated with the detection of CTC clusters but not with individual cells or detection of
BM-DTCs. Overlapping of the 2 genetic signatures of the primary PDX tumors associated with the presence of CTC clusters and
with lung metastatic potential identified 4 genes (HLA-DP1A, GJA1, PEG3, and XIST). This 4-gene profile predicted distant
metastases-free survival in publicly available datasets of early BC patients.
Conclusion: This study suggests that CTCs and BM-DTCs detected in BC PDX-bearing mice may represent a valuable and
unique preclinical model for investigating the role of these rare cells in tumor metastases.

Title: Different prognostic value of circulating and disseminated tumor cells in primary breast cancer. Influence of bisphosphonate
intake?
Sabine Kasimir-Bauer1, Bahriye Aktas1, Siegfried Hauch2, Rainer Kimmig1 and Oliver Hoffmann1. 1University Hospital Essen,
Essen, Germany and 2AdnaGen AG, Langenhagen, Germany.
Body: Background: Disseminated tumor cells (DTCs) in the bone marrow and circulating tumor cells (CTCs) in blood of breast
cancer patients (pts) were shown to provide independent prognostic information and can be regarded as an early indicator of
therapy failure. Here we demonstrate a different prognostic value of DTCs and CTCs and explain these findings by early
bisphosphonate intake.
Patients and Methods: 10 ml blood and two bone marrow aspirates of 488 pts with first diagnosis of breast cancer between Aug
2006 and Dec 2010 were studied for DTCs and CTCs. CTCs were detected using the AdnaTest BreastCancer (AdnaGen AG,
Hannover, Germany) for the detection of EpCAM, MUC-1 and HER2 transcripts. DTCs were analyzed by immunocytochemistry
using the pan-cytokeratin antibody A45-B/B3. In addition to chemo-, radio- and anti-hormonal therapy, bisphosphonates were
given to DTC-pos pts [clodronate (2520 mg per day) for at least two years] and 69 DTC-neg, postmenopausal pts [zoledronic
acid (4 mg, twice a year) for three years]. Results: After a median follow-up time of 48 months (range: 5 to 87 months), the
overall survival rate was 93% (456/488 pts) and relapses occurred in 11% (56/488 pts) of cases. CTCs were detected in 109/488
pts (22%) and significantly correlated with reduced disease free survival (DFS; p=0.02). Negative prognostic relevance was
predominantly related to the lobular subtype (p=0.0047), the ER-pos subtype, (pER+/Her2=0.01; pER+/Her2+=0.25; pER-/Her2+=0.5) and
pts who had received chemotherapy (p=0.015) or radiation therapy (p=0.0083) but not to pts with anti-hormonal regimens
(p=0.06). DTCs were detected in 162/488 pts (33%). In contrast to CTCs, no prognostic significance could be shown with regard
to DFS (p=0.49), all clinical parameters as well as treatment regimens. In contrast, only in CTC-pos pts, the presence of DTCs
significantly correlated with DFS (p=0.04). We further investigated whether this lack in prognostic significance was due to
bisphosphonate intake in case of DTC-positivity. Predominantly, in the CTC-neg group, bisphosphonate treatment significantly
influenced DFS in both histological subtypes (pCTC-/ductal=0.04; pCTC-/lobular=0.0021). However, in the CTC-pos group of pts, no
such effect was observed (pCTC+/ductal=0.76; pCTC+/lobular=0.51).Conclusion: Here we show, that CTCs, rather than DTCs were
significantly prognostic for early relapse. The ER-pos subtype was mostly affected which is in concordance with our previous
studies, demonstrating that CTCs show EMT and tumor stem cell characteristics and, thus, down regulate ER which probably
makes anti-hormonal treatment less effective. In addition, the presence of these CTC-subtypes might also explain the negative
prognostic impact of CTCs in pts receiving chemo- or radiation therapy which we could verify in a subgroup of our pts (n=238)
showing a significant correlation of CTCs and EMT/stem cell like CTCs (p=0.017; data not shown). The lack of prognostic
significance of DTCs can be related to bisphosphonate intake in all DTC-pos pts, a subgroup of DTC-neg pts as well as in
DTC-pos/CTC-neg pts, underlining our assumption that CTCs might be a high risk indicator for already ongoing metastasis not
limited to bone metastasis.

Title: Persistence of circulating tumor cells immediately after and two years after systemic adjuvant chemotherapy in patients with
early breast cancer Results of the German SUCCESS trials
Bernadette AS Jaeger1, Ulrich Andergassen2, Julia K Neugebauer2, Marianna Alunni-Fabbroni2, Carola A Melcher3, Carsten
Hagenbeck3, Susanne Albrecht1, Ralf Lorenz4, Thomas Decker5, Georg Heinrich6, Tanja Fehm3, Andreas Schneeweiss7, Matthias
W Beckmann8, Klaus Pantel9, Klaus Friese2, Peter A Fasching8, Thomas WP Friedl1, Wolfgang Janni1 and Brigitte K Rack2.
1
University Hospital Ulm, Ulm, Germany; 2Klinikum der Ludwig-Maximilians-Universitaet, Munich, Germany;
3
Heinrich-Heine-Universtitaet, Duesseldorf, Germany; 4Gemeinschaftspraxis Dr. Lorenz / Hecker /Wesche, Braunschweig,
Germany; 5Studienzentrum Onkologie Ravensburg, Ravensburg, Germany; 6Praxis Dr. Heinrich, Fuerstenwalde, Germany;
7
National Center for Tumor Disease and University Hospital Heidelberg, Heidelberg, Germany; 8University Hospital Erlangen,
Friedrich-Alexander University Erlangen-Nuremberg, Comprehensive Cancer Center Erlangen-EMN, Erlangen, Germany and
9
Institute for Tumor Biology, Center of Experimental Medicine, University Medical Center, Hamburg, Germany.
Body: Background
There is growing evidence that circulating tumor cells (CTCs) have prognostic impact in patients (pts) with early breast cancer
(EBC). In this study the persistence of CTCs immediately after and two years after chemotherapy (Ctx) was prospectively
evaluated according to molecular subtypes within the German multicentre SUCCESS trials.
Methods
SUCCESS A and C were randomized Phase III studies including pts with node positive or high-risk node negative EBC. In each
trial two different adjuvant Ctx regimen were compared: FEC-DOC (3 cycles of FEC followed by 3 cycles of Docetaxel) to
FEC-DG (3 cycles of FEC followed by 3 cycles of Docetaxel/Gemcitabine) in SUCCESS A and in the SUCCESS C study
FEC-DOC to an anthracycline-free Ctx regimen (6 cycles of Docetaxel/Cyclophosphamide). Both studies involved a second
randomization after Ctx: 2 vs. 5 years of zoledronic acid treatment (SUCCESS A) or 2-years of an individualized
lifestyle-intervention program vs. general lifestyle recommendations (SUCCESS C). Adequate endocrine treatment and treatment
with trastuzumab as indicated were included in both trials.
As part of the translational research program, 23ml of peripheral blood were drawn to isolate CTCs using the CellSearch System
(Veridex, USA). After immunomagnetic enrichment with an anti-EpCam-antibody, cells were labelled with anti-CK8/18/19 and
anti-CD45 antibodies to distinguish epithelial cells from leucocytes. The cut-off for CTC-positivity was 1 CTC.
Molecular subtypes were defined as luminal-A-like (hormone-receptor positive, G1 or 2), luminal-B-like (hormone-receptor
positive, G3), HER2-positive and triple-negative.
Results
CTC analyses were performed for 3344 blood samples collected immediately after Ctx and for 1352 blood samples two years
after Ctx. After Ctx 17.5% (584/3344) of the pts were CTC-positive (range 1 124 CTCs), and two years after Ctx the positivity
rate for CTCs was 17.2% (233/1352, range 1-99).
CTC positivity as assessed immediately after Ctx differed significantly among molecular subtypes (chi-square test, p < 0.001): Pts
with HER2-positive tumors were more likely to have CTCs in the blood (26.3%, 105/400) as compared to pts with luminal-A-like
tumors (15.4%, 283/1842), luminal-B-like tumors (17.7%, 142/802), or triple-negative tumors (18.0%, 54/300).
Two years after Ctx CTC-positivity did not differ significantly among molecular subtypes (chi-square test, p = 0.463).
CTC-positivity rates were 15.7% (96/613) for luminal-A-like tumors, 19.1% (49/256) for luminal-B-like tumors, 17.2% (51/296) for
HER2-positive tumors, and 19.8% (37/187) for triple-negative tumors.
Conclusions
The data of this study confirm previous findings that CTCs may persist after standard adjuvant therapy. Immediately after Ctx
CTCs seem to be more frequent in pts with HER2-positive tumors as compared to other molecular subtypes, while two years after
Ctx no differences in CTC positivity among molecular subtypes were detected. These results might indicate good efficacy of
HER2-targeted therapies on CTCs.

Title: Exploring the intra-patient PIK3CA mutational heterogeneity of circulating tumour cells by massive parallel sequencing in
patients with metastatic hormone receptor-positive breast cancer
Bram De Laere1, Dieter JE Peeters1, Salgado Roberto1,2, Vermeulen B Peter1,2, Van Dam A Peter1, Dirix Y Luc1,2 and Van Laere J
Steven1,3. 1CORE Faculty of Medicine and Health Sciences University of Antwerp, Antwerpen, Belgium; 2GZA
Sint-Augustinus, Wilrijk, Antwerpen, Belgium and 3Catholic University Leuven, Leuven, Vlaams-Brabant, Belgium.
Body: Introduction: Circulating tumour cells (CTCs) are a real-time reflection of the ad hoc relevant subpopulation in patients
with progressive disease. The study comprises the clinical application of a semi-automated methodology to determine the
PIK3CA mutational status at a single cell level.
Methods: Using CellSearch and DEPArray, we purified single (n=172) and groups (n=93, ranging 5120 cells) of CTCs from
peripheral blood in 18 patients with metastatic ER+/PR+/HER2- breast cancer, of whom archival primary tumour (PT) material
was available. Isolation of WBCs served as internal negative control. Recovered cells were subjected to
whole-genome-amplification (WGA). During CTC blood draw, an extra SST tube was filled for isolation of circulating cell-free DNA
(cfDNA) from serum as comparator. All samples from different compartments underwent gene mutation analysis via targeted
amplification of exons 9 and 20 of the PIK3CA gene and downstream 454 massive parallel sequencing (MPS) on the GS Junior
system.
Results: WGA of single and group CTC samples had a success rate of 8322 % and 968%, respectively. MPS resulted in an
average throughput of 135.99624.423 high quality reads with a normal-distributed mean fold coverage depth of 1257703 and
1494897 reads for exon 9 and 20, respectively. PIK3CA mutations were present at a relatively high frequency in archival PT
(20/29 (68,9%)) and showed poor and moderate agreement with cfDNA (n=24; 33,3% disparity; kappa=0,03) and CTCs (n=14;
11,28% disparity; kappa=0,329), respectively. Comparison of CTCs and temporally matched cfDNA samples revealed substantial
agreement (n=14; 14,3% disparity; kappa=0,536). A concordant PIK3CA status across all compartments (PT, cfDNA and CTCs)
was observed in 11/15 (73,3%) samples. At the used sequencing depth, cfDNA failed to the detect PIK3CA mutations in 2 cases
(13,3%), which were present in the respective PT and corresponding CTCs. CTCs failed to detect the mutant PIK3CA gene in
one case (6,7%), which was present in PT and cfDNA. Gain of mutation was observed in 2/15 patients (13,3%), with a wild-type
PT and mutant cfDNA and CTCs at progression. A wild-type PIK3CA sequence in recovered WBCs of all patients (n=15)
indicates a high specificity and tumorigenic nature of the picked up variants. In depth intra-patient analysis of mutant CTCs on a
single cell level reveals PIK3CA mutational heterogeneity with the presence of both mutant and wild-type CTCs (average
%MT/%WT ratio of 50/50). Additionally, unique double-mutated CTCs were detected in 5/15 (30%) cases as well.
Conclusions: We report the frequent occurrence of PIK3CA hotspot mutations in metastatic HR+ breast cancer. Intra-patient
mutational heterogeneity was observed in the single CTC samples of the same patient. Thereby, this study provides evidence of
both concordance and discordance of the PIK3CA genotype in the intravascular compartment (down to the single cell level) with
comparison to the temporally unmatched PT. The study presents the utilization of a liquid biopsy, thereby paving the way towards
the application of a more personalized medicine in the management of patients with metastatic cancer.

Title: Predictive impact of circulating tumor cells with an epithelial-to-mesenchymal transition phenotype in patients with primary
breast cancer treated with primary systemic therapy
Fanny Le Du1,2, Dzifa Y Duose3, Elisha J Dettman3, Jackson A Summer1, Mariana Chavez-MacGregor1, Carlos H Barcenas1,
Abenaa M Brewster1, Alvarez H Ricardo1, Vincente Valero1, Ana M Gonzalez-Angulo1, James M Reuben1 and Naoto T Ueno1.
1
MD Anderson Cancer Center, University of Texas, Houston, TX; 2Centre Eugne Marquis, Rennes, France and 3ApoCell, Inc,
Houston, TX.
Body: Background: Tumor cells with a mesenchymal phenotype include cancer stem-like cells (CSCs), which are known to
contribute to metastasis. Circulating tumor cells (CTCs) are epithelial cells in peripheral blood that are detected using an
anti-EpCAM antibody; however, CTCs undergoing epithelial-mesenchymal transition (EMT) may not be detected using this
method. We have developed an antibody-independent CTC enrichment platform, Apostream, which does not rely on
EpCAM-based capture. We used this instrument to determine the clinical relevancy and feasibility of measuring EMT-CTCs in
Methods: Blood samples from newly diagnosed breast cancer patients were prospectively collected at baseline (T0), after primary
systemic therapy (T1), and after definitive surgery (T2) and processed using the Apostream system. Isolated cells were stained
with antibodies to cytokeratin (anti-CK), leukocytes (anti-CD45), and the nuclear stain, 4,6-diamidino-2-phenylindole (DAPI), to
identify CTCs. These CTCs were also stained with additional markers and examined on a laser scanning cytometer to measure
protein expression levels of epithelial (EpCAM, E-cadherin), mesenchymal (-catenin, vimentin) and CSC-markers (CD44, CD24).
Pathological complete response (pCR) and residual cancer burden (RCB) statuses after preoperative treatment were obtained to
correlate baseline CTCs and marker expression with treatment response.
Results: The study enrolled 33 patients (10 with early-stage, 19 with locally advanced, and 4 with metastatic breast cancer); 32,
12, and 10 patients provided samples at T0, T1, and T2, respectively. Of the 20 patients who underwent surgery, 6 patients
achieved pCR. CTCs were detected ( 1 cell in at least one of the three samples) in 47%, 75%, and 80% of the T0, T1, and T2
samples, respectively. EMT markers (either vimentin or -catenin) were detected in 67%, 11%, and 38% of these CTCs,
respectively. The mean number of CTCs per mL detected at T0 was 0.43 (range, 0-3.7). CTC detection was correlated with a
high histological Nottingham index (P=0.039). Conversely, vimentin-positive CTC detection was inversely correlated with clinical
stage at T0 (P=0.034). No significant correlation was observed between CTCs detected at baseline and breast cancer subtypes.
However, there was a trend for CTCs detected at T0 to be predictive of chemotherapy response, as 62% of patients with CTCs at
T0 achieved a pCR, whereas only 11% of patients without CTCs at T0 had a pCR (P=0.057). The other EMT and CSC markers
we tested did not predict response.
Conclusions: Apostream was successful in detecting EMT-CTCs in this prospective study. There was a trend for the presence
of CTCs at baseline to predict pCR. We will present our final data analysis of 50 patients.
CK+CD45- CTCs
pCR (n=6)
RCB-I (n=2)
RCB-II (n=5)
RCB-III (n=4)
P-value
Negative (n=9)
1(11)
2(22)
3(33)
3(33)
0.146 0.057*
Positive (n=8)
5(62)
0(0)
2(25)
1(13)
All data are no. of patients (%) Two did not have an RCB status available, and 1 did not have a T0 sample. * Comparing pCR
status (pCR vs no-pCR) between CTCs negative and positive.

Title: Circulating tumor cells (CTCs) detect HER2+ status and phenotypic heterogeneity in metastatic breast cancer (MBC)
Laura Austin1, Zhaomei Mu1, Tiffany Avery1, Rebecca Jaslow1, Carmela Paolillo1, Angela Toss2, Paolo Fortuna1, Ye Zhong1,
Hushan Yang1 and Massimo Cristofanilli1. 1Thomas Jefferson University Hospital, Philadelphia, PA and 2Universtiy of Modena and
Reggio Emilia, Modena, Italy.
Body: Background
Circulating tumor cells (CTCs) are epithelial cells that can be found circulating in the blood of MBC patients and may represent a
heterogeneous population including epithelial cells and cancer stem cells (CSC) shed from the tumor. Their detection and
enumeration has prognostic significance and can be used for longitudinal monitoring of response to treatment. Recent advances
have allowed for the detection of HER2 protein expressing CTCs using the CellSearch platform. HER2 expression has been
associated with CSCs phenotype in absence of gene amplification (Korkaya et al, Oncogene 2008) and particularly in Luminal B
disease (Ithimakin et al, Cancer Res, 2014). We hypothesized that HER2+/CTCs are detectable in patients with MBC irrespective
of their HER2 status and this information can be potentially be used for treatment selection. Targeting HER2+ CSCs may result in
clinical benefit and improved outcome.
Methods
This is a prospective analysis of 40 patients with locally advanced or MBC whose blood was analyzed for the baseline detection
of CTCs as part of their disease initial evaluation. Blood was drawn for CTC detection on eligible patients at initiation of a new line
of therapy; CTCs monitoring was repeated at progression or change in therapy. The 7.5mL of whole blood was collected in a
CellSave Preservative Tube, and CTC isolation, enumeration and characterization were performed using the FDA-approved
CellSearch System (Janssen Diagnostics, USA). The CellSearch tumor phenotyping reagent HER-2/neu
(Fluorescein-conjugated) was used to determine CTC with HER-2/neu expression using the CellTrack Analyzer II.
Results
Most patients in this study had metastatic disease (90%). According to disease subtype, 43% of patients were ER+/HER2(Luminal A), 17% ER+/HER2+ (Luminal B), 20% ER-/HER2+ (HER2) and 20% ER-/HER2- (TNBC). Moreover, 55% had a clinical
diagnosis of Inflammatory Breast Cancer (IBC). Twenty-two patients had CTCs detected (55%), the average number of CTCs
was 8.6 (0-135) with a median follow up of two months. Of the patients who had CTCs detected and had HER2+ disease
(IHC/FISH), 83% (5/6) had concordance in HER2+ CTCs. Interestingly, in the subset of patients who had HER2 negative disease
(ICH/FISH) and had detectable CTCs, there was discordance in HER2 status: 44% (7/16) had HER2+ CTCs and all but one had
Luminal A disease. Two patients have been started on HER2 targeted therapy based on finding HER2+ CTCs. One of these
patients had 22 CTCs, 8 of which were HER2+, was initiated on HER-2 combined regimen and at a repeat evaluation in 3 months
demonstrated 0 CTCs and clinical response.
Conclusions
CTCs offer a new and innovative approach to detect HER2+ cells in MBC. Tissue analysis with IHC and FISH has been the gold
standard but these methods are unable to account for disease phenotyopic hetereogeity and identify patients who have CSCs
(HER2+) and therefore would benefit from HER2 targeted therapy. This warrants further investigation in a prospective trial in
Luminal disease to formally compare these methods and correlate the results with clinical outcomes.

Title: Circulating tumor cells (CTC) and endothelial cells (CEC) changes in HER2 negative metastatic breast cancer (MBC)
patients treated with first line weekly paclitaxel and bevacizumab: Preliminary results of a prospective cohort from the French
Breast Cancer InterGroup Unicancer: COMET study
Jean-Yves Pierga1, Isabelle Vaucher1, Maya Gutierrez2, Olivier Tredan3, Sverine Guiu4, Gilles Romieu5, Anthony Goncalves6,
Marc Debled7, Christelle Levy8, Jean-Marc Ferrero9, Christelle Jouannaud10, Elisabeth Luporsi11, Marie-Ange Mouret-Reynier12,
Florence Dalenc13, Bernard Asselain1, Jerome Lemonnier14 and Francois-Clement Bidard1. 1Institut Curie, Paris, France; 2Institut
Curie, St Cloud, France; 3Centre Lon Brard, Lyon, France; 4Centre Georges Franois Leclerc, Dijon, France; 5ICM Val
D'Aurelle, Montpellier, France; 6Institut Paoli-Calmettes, Marseille, France; 7Institut Bergoni, Bordeaux, France; 8Centre Franois
Baclesse, Caen, France; 9Centre Antoine Lacassagne, Nice, France; 10Institut Jean Godinot, Reims, France; 11ICL Alexis Vautrin,
Vandoeuvre les Nancy, France; 12Centre Jean Perrin, Clermont-Ferrand, France; 13Institut Claudius Regaud, Toulouse, France
and 14R&D UNICANCER, Paris, France.
Body: Background: increased levels of circulating tumor cells (CTC) are associated with worse progression-free survival (PFS)
and overall survival (OS) in patients (pts) with MBC (Bidard FC et al, Lancet Oncol 2014). The failure of chemotherapy to reduce
CTCs to levels below five CTCs per 7.5 mL whole blood at first follow-up after initiating a new systemic therapy for MBC predicts
shorter time to progression and OS. It has been hypothesized that bevacizumab could modify CTC prognostic value due to
extravasation or epithelio-mesenchymal transition induction. CEC variations to predict benefit of anti-angiogenic treatment is still
controversial.
Patients & methods: the French cohort COMET is a prospective study including first line HER2 negative pts receiving weekly
paclitaxel and bevacizumab according to EMEA approved combination. The aim of this cohort is to evaluate clinical, biological
and radiological parameters associated with pts outcome (CTC, CEC, VEGFA levels, ctDNA, pharmacogenomic polymorphisms,
metabolomic parameters, visceral fat assessed by initial CTscan, serum estradiol level, and quality of life). We present here the
first planned analysis on 206 patients evaluated for CTC and CEC using the FDA cleared CellSearch method.
Results: inclusions started in 09/2012. At time of analysis, 219 patients were included, 211 were evaluable for CTC at baseline
(failure rate 4%) and 207 for CEC at baseline (failure rate 5%). Due to short follow-up, 173 pts and 166 pts were evaluable for
both CTC and CEC at baseline and first day of second cycle of CT (D1C2) respectively. At baseline, 100/211 (47%) pts had 5
CTC (median 4 (range 0-30,000) and 30% had no detectable CTC (0 CTC). After one cycle of chemotherapy (D28) 38 pts (22%)
had still 5 CTC: 37 pts with initial high level and only one patient with low CTC at baseline had increased CTC above 5. Median
number CEC was 21 (0)-2231) at baseline and 22 (1-881) at D1C2. CEC increased in 16% and decreased in 15% of the cases.
CTC & CEC changes after one cycle
CTC Baseline
CTC at D1C2
CEC Baseline
CEC at D1C2
37 (21%)
>20
>20
65 (39%)
<5
45 (26%)
>20
20
25 (15%)
<5
1 (<1%)
20
>20
27 (16%)
<5
<5
90 (52%)
20
20
49 (30%)
Total
173
166
CTC number at baseline line and CTC D1C2 were correlated (p<0.01) (Spearman test). There was no correlation between CEC
at baseline or at D1C2 with CTC or CTC changes. Final analysis will be completed when 206 couples for both CTC and CEC at
baseline and D1C2 will be available. Accrual is still ongoing.
Conclusion: this 22% rate of failure to reduce CTC < 5 after one cycle of first line CT in a homogeneously bevacizumab-treated
cohort of MBC patients did not differ from previous series. As second lines of chemotherapy do not improve the poor prognosis of
this group of patients according to the SWOG 500 study results (Smerage et al, JCO 2014), trials of novel therapeutic agents
should be considered at the time of progression.

Title: CTC enumeration and characterization has predictive and prognostic implications in patients with metastatic breast cancer
treated with exemestane plus the mTOR inhibitor everolimus
Sofia Agelaki1,2, Dimitris Mavroudis1,2, Maria Spiliotaki2, Eleni Politaki2, Maria A Papadaki2, Stella Apostolaki2, Christos Nikolaou1
and Vassilis Georgoulias1,2. 1University Hospital of Heraklion, Heraklion, Crete, Greece and 2Laboratory of Tumor Cell Biology,
School of Medicine, University of Crete, Heraklion, Crete, Greece.
Body: Background: The utility of CTC enumeration in predicting patient (pt) outcome has been demonstrated in metastatic
breast cancer (MBC) treated with chemotherapy or endocrine therapy. In this study we evaluated the clinical impact of CTC
assessment in terms of both enumeration and characterization in breast cancer pts treated with exemestane plus everolimus.
Patients and methods: Thirty-nine pts with hormone receptor (HR)-positive, HER2-negative MBC, received exemestane plus
everolimus. CTC enumeration in peripheral blood (7.5 ml) was performed before treatment (n=39), post cycles 1 (n=39) and 3
(n=29), on disease re-evaluation and on relapse, whichever occurred first, using the CellSearch System. CTC characteristics
were determined at the same time points by immunofluorescence (IF) analysis of PBMC cytospins (106 cells), triple stained with
pancytokeratin (CK) antibody along with Ki67 and M30 as proliferation and apoptosis markers, respectively, using the Ariol
System. Patients were assessed by CT scans and bone scan, every 3 months or as clinically indicated. Results: At the cut-off of
1 CTC, 25 of 39 (64%) pts had detectable CTCs at baseline, 12 (31%) of 39 post-1st and 10 (34.5%) of 29 post-3rd cycle. Ten
(25.6%) pts remained CTC(+) and 12 (30.8%) CTC(-) both at baseline and post-1st cycle; 15 (38.5%) CTC(+) pts turned to CTC(-)
and 2 (5%) CTC(-) turned to (+). CTC positivity after the first cycle was associated with shorter median progression-free survival
(PFS) compared to CTC(-) status (3.9 vs 8 mo, p=0.031). Shorter PFS was also recorded for pts that remained CTC(+) at both
time points compared to all other (p=0.02). At the cut-offs of 2 and 5 CTCs, 16 (41%) and 9 (23%) pts were CTC(+) at
baseline, respectively; post-1st cycle, 7 (18%) and 4 (10%) pts were CTC(+) (at 2 and 5 CTCs, respectively). Post-3rd cycle
the positivity rate was 17% for both cut-offs and these pts had significantly shorter PFS compared to CTC(-) pts (3.7 vs 8.7
months, p=0.048). Efficacy assessment revealed partial response in 3 (7.7%) pts, stable disease in 27 (69.23%) and progressive
disease (PD) in 8 (20.5%); 1 pt was non-evaluable for response. Among pts determined CTC(+) post-1st cycle (cut-off 2 CTCs),
57% progressed compared to 13% of CTC(-) pts (p=0.02). In addition, at the post-3rd cycle evaluation, pts with PD had
significantly higher CTC counts compared to non-progressors (mean SEM; 10 5.78/pt vs 1.620.83/pt, p=0.027). By the use
of IF 43%, 44% and 40% of CTC(+) pts had proliferative [Ki67(+)/M30(-)] CTCs at baseline, post -1st and -3rd cycles, respectively
(cut-off 1 CTC); 67%, 50% and 50% of those pts, respectively, experienced PD. Apoptotic [Ki67(-)/M30(+)] CTCs were detected
in 14%, 22% and 60% of CTC(+) pts at baseline, post -1st and -3rd cycles, respectively; none of the pts with apoptotic CTCs
experienced PD. Conclusions: CTC enumeration and characterization in terms of proliferation and apoptosis during the course
of treatment has significant predictive and prognostic implications in patients with MBC receiving the combination of exemestane
plus everolimus.

Title: Whole exome sequencing of circulating and disseminated tumour cells in patients with metastatic breast cancer
Dieter JE Peeters1,2, Ken Op De Beeck1,3, Anja Brouwer1,2, Geert Vandeweyer3, Patrick Pauwels1,4, Marc Peeters1,2, Peter B
Vermeulen1, Peter A van Dam1,2, Steven J Van Laere1,5, Guy Van Camp3 and Luc Y Dirix1. 1Center for Oncological Research
(CORE), University of Antwerp, Antwerp, Belgium; 2Antwep University Hospital, Edegem, Belgium; 3Center for Medical Genetics,
University of Antwerp, Antwerp, Belgium; 4Antwerp University Hospital, Edegem, Belgium and 5University of Leuven, Leuven,
Belgium.
Body: Introduction:
Circulating tumour cells (CTC) found in the blood of patients with cancer offer the potential to provide a repeatedly accessible
source of tumour cells for the real-time assessment of tumour characteristics in patients with metastatic breast cancer (MBC).
Questions remain to what extent CTC are truly representative of the actually present tumour mass in a patient at a specific
moment in time and the molecular heterogeneity within the CTC population is only now being explored. Here, we report on the
first results of an ongoing comparative study of mutation profiles of CTC and synchronously isolated disseminated tumour cells
(DTC) from metastatic effusions or biopsies of solid metastases of patients with clinically progressive MBC.
For this project CTC are isolated from 7.5 ml blood samples of patients with MBC using the CellSearch system. CellSearch
enriched CTC samples are subsequently further purified and sorted into several batches of 1-125 CTC per patient using the
DEPArray system. DNA is isolated and amplified using the Ampli1 whole genome amplification (WGA) kit and subjected to whole
exome paired-end sequencing (WES). DTC from metastatic effusions, fresh frozen tissue from solid metastases or the primary
tumour, or - in patients with extremely high CTC counts (>10.000/7.5 ml) - pooled CTC from the CellSearch Profile sample, are
sequenced as a comparator for mutation profiles. DNA from the buffy coat of white blood cells are sequenced to enable somatic
mutation analysis.
Results:
Eight samples of 1-125 CTC and a CellSearch Profile sample of one patient with MBC who had ca. 30.000 CTC/7.5 ml of blood
(patient 1) and 4 CTC samples of 5-10 CTC, 2 temporally matched DTC samples of 10 and 20 DTC from a pleural effusion and a
fresh frozen tissue sample of the primary tumour of a second patient (patient 2) have been sequenced so far. Average base
coverages were 13.6x (patient 1) and 11.8x (patient 2) for CTC/DTC samples and 175x and 120x for the CellSearch profile
sample (patient 1) and the primary tumour sample (patient 2) respectively. Between 29.64% and 53.57% of the exomes of
amplification products of CTC/DTC DNA were uncovered, probably due to technical limitations of the WGA procedure. Overall, if
adequately covered, good concordances were observed for variants identified with MuTect in 28 frequently mutated genes in
breast cancer between samples of amplification products of 1-125 CTC and the CellSearch Profile sample of patient 1. In patient
2, the same H1047R PIK3CA mutation was identified in the primary tumour and all CTC and DTC samples. In-depth analyses of
the full exome data are being conducted.
Discussion:
Our data provide insight into clinically relevant questions to what extent CTC reflect mutational profiles in temporally matched
metastatic tumour cells, and by analysing multiple CTC samples of the same patient genetic heterogeneity between CTC in
patients with MBC. Sample accrual and analysis is being expanded and updated results will be presented at the conference.

Title: Assay Development for detection of estrogen responsive gene histone acetylation in breast cancer circulating tumor cells
S A Litherland1, Robert Reynolds1, Louis Barr1,2, Alvin JO Almodovar1 and David A Decker1. 1Florida Hospital Cancer Institute,
Orlando, FL and 2Florida Hospital Center for Specialized Surgery, Orlando, FL.
Body: Research Objectives: To develop a tumor-specific assay for assessment of histone acetylation and histone deacetylase
(HDAC) effects in circulating tumor cells (CTC).
Rationale: Therapy-resistant tumor cells arise in breast cancer patients during hormone therapy. Anti-deacetylation drugs such
as Entinostat, Vorinstat, and Romidepsin block histone deacetylases (HDAC), preventing removal of epigenetic acetyl chromatin
modifications which allow for expression of estrogen receptor (ER). Clinical trials such as ECOG E2112 are investigating a
histone acetylase inhibitor (HDACi) therapy in combination with standard endocrine therapy to overcome hormone resistance. A
drawback for HDACi clinical trials in the past has been the lack of a sensitive assay to identify tumor specific HDAC effects
contributing to hormone resistance.
Methods: We testing a histone acetylation assay using Fluorescence Activated Cell Sorting (FACS) peripheral blood CTC
isolation and histone acetylation/HDAC specific chromatin immunoprecipitation (ChIP) with realtime PCR analysis to identify
ERalpha and GREB-1 gene activation. We report the initial findings from 11 consented subjects (IRB approved protocol#
372522).
Results: Our findings indicate that GREB-1, a gene essential to breast cell ER expression in response to estrogen, can be
regulated via a histone acetylation epigenetic control mechanism. Using our CTC-ChIP analysis, we found both ER+ and ERCTC arising from ER+ tumors and that the proportion of ER- CTC mirrored clinical hormone resistance. GREB-1 activation and
expression is detectable in both ER+/ER- CTC when enough CTC could be collected for ChIP analysis; however, ERalpha
acetylation and mRNA expression was below the level of detection.
Table of Preliminary Results
Patient
Group/N
Tumor
status
HT
Status
CTC
ER+/million
(meanSD)
ER+ HST
Metastatic Sensitive 130379260109
sensitive/4
CTC
CTC ER+GREB-1
ER-/million(mean Acetylation+(mean
SD)
SD)
CTC ER- GREB-1

Acetylation(mean SD)
PREDICTED
ESTROGEN
RESPONSE
336594
0.95*0.08
0.015*0.02
Sensitive
ER+>ER-
ER+ HST
Metastatic Resistant
resistant/2
82429.7
73346421
0.87*0.13
1*
Resistant
ER+<ER-
ERMetastatic Resistant
metastatic/1
18758
21498
1*
ND
Resistant
ER+<ER-
34000000
89500000000126600000000
Sensitive
ER+>ERactive
GREB-1
0*
0*
ERprimary/2
Healthy
controls/2
Primary Unknown 277251379512 112754159451
None
None
0.50.7071
*Insufficient cells for accurate ChIP analysis

Conclusions: Our preliminary data suggest that by using a combination of CTC based epigenetic biomarker analyses; it is
possible to detect the presence of hormone resistant ER+ and ER- CTC subpopulations arising from ER+ breast cancer tumors.
Monitoring of longitudinal changes in ER+/ER- CTC and the epigenetic activation of GREB-1 and ERalpha responsiveness could
provide support in deciding which patients may benefit from the use of anti-deacetylation drug therapy in combination with
hormonal therapy to promote ER alpha expression, reinstating estrogen dependency in both ER- and ER+ subpopulations; and
thereby, re-sensitizing them to hormonal therapy.

Title: Detection of EMT, anoikis and stem cell markers in metastatic breast cancer patients under different lines of treatment
Elisabeth K Trapp1, Brigitte Rack1, Leonie Majunke1, Julian Koch1, Simone Hofmann1, Thomas WP Friedl2, Julia Neugebauer1,
Julia Jckstock1, Bernadette Jger2, Jens Huober2, Wolfgang Janni2 and Marianna Alunni-Fabbroni1.
1
Ludwig-Maximilians-University, University Hospital, Munich, Germany and 2University Hospital, Ulm, Germany.
Body: Background: Metastasis are thought to be induced by occult spreading of tumor cells already during the early phases of
the disease. Circulating Tumor Cells (CTCs) are regarded as precursors of distant metastasis, while detaching from the primary
tumor and originating micrometastases in distant organs.
Recent evidences pointed to CTC heterogeneity, showing that CTCs can present different phenotypes. Goal of this study was to
identify in metastatic breast cancer (mBC) patients CTCs with EMT features and to further characterize them with respect to
cellular heterogeneity.
Methods: This prospective ongoing study included mBC patients (n=12) with a median age of 58,5 years (range: 35-78 years),
enrolled in a time frame of 7 months while undergoing therapy. The majority of patients were estrogen and/or progesterone
receptor positive (11/12) and HER2 non-amplified (10/12). Patients had metastatic lesions in viscera as well as bone (6/12), only
bone (2/12), only viscera (2/12), only viscera in combination with locoregional recurrence (1/12), or visceral and bone metastases
in combination with locoregional recurrence (1/12). Current therapy was endocrine therapy (4/12), chemotherapy (3/12),
chemotherapy and HER2-targeted therapy (2/12), HER2-targeted monotherapy (1/12), antiangiogenic therapy (1/12), or surgical
therapy only (1/12). Blood samples, withdrawn at any time point during treatment, were depleted of EpCAM+ cells and CD45+
white blood cells (EpCAM/CD45 depleted fraction) (Mego et al., Int J Cancer 2012;130(4):808-816) and expression of epithelial
markers (EpCAM, E-Cadherin, Cytokeratin 8,18,19), mesenchymal markers (n-Cadherin, Vimentin), EMT-inducing factors
(Twist1, Snail1, SLUG, Zeb1 and FoxC2), anoikis markers (TrkB1, Bcl2) and stem cell markers (CD24, CD44, CD133) were
analyzed by qRT-PCR. CTC counting with the CellSearchTM system (Veridex, Raritan NJ) was run in parallel. Healthy donors
(n=10) were included in the study as negative controls.
Results: The data collected so far showed that 50% of the patients were positive for CTCs in the EpCAM+ fraction as detected
with the CellSearchTM system, while 33% were still CTC positive in the EpCAM/CD45 depleted fraction. 50% of the patients,
found CTC negative with CellSearchTM, were nevertheless positive for the epithelial and EMT markers in the EpCAM/CD45
depleted fraction (EpCAM 25%, E-Cadherin 25%, CK8 16,6%, CK18 16,6%, CK19 16,6%, SLUG 8,3%, Zeb1 25%, Twist1 8,3%,
Vimentin 58,3%). N-cadherin, FoxC2, Snail1, TrkB1, CD24, CD44 and CD133 were never detectable.
Conclusions: These preliminary results suggest that mBC patients undergoing different lines of therapy present heterogeneous
CTCs. 50% of the patients with undetectable EpCAM+ CTCs, were found CTC positive in the EpCAM/CD45 depleted fraction.
mBC patients might be insensitive to treatment due to a selection of resistant CTCs subpopulations with different phenotypes.
Additional patients with the same clinical characteristics will be analyzed in the next 6 coming months and updated results will be
included.

Title: Expression of circulating tumor cells (CTC) and CK-19 mRNA (CK19) as prognostic factors in heavily pretreated metastatic
breast cancer
Serafin Morales1, Ana Velasco1, Jose Vidal3, Anna Serrate2, Joan Valls2, Juan Carlos Samame1, Rafael Gisbert3, Desamparados
Moral3, Antonio Llombart3, Antonieta Salud1 and Xavier Matias-Guiu2. 1Hospital Arnau de Vilanova, Lleida, Spain; 2Institut de
Recerca Biomedica Lleida, Spain and 3Hospital Arnau de Vilanova, Valencia, Spain.
Body: Introduction: The expression of CTC and CK19 holds prognostic value for patients with breast cancer but their clinical
significance remains still controversial.
Methods: This clinical observational study included 58 preteated metastatic breast cancer patients who started a standard new
treatment line. CTC and CK19 was measured with CellSearch and RT-PCR respectively at inclusion time. Progression-free
survival (PFS) was defined as the time elapsed between the initiation of the treatment and either the date of clinical or radiological
progression or death or the last follow-up. Cox proportional hazards regression model was used to assess the univariate
prognose value of CTC and CK19 on PFS, and Kaplan-Meier estimates. A multivariate Cox model was also performed to
additionaly account for ER and visceral disease. CTC and CK19 positivity was considered when a value of 1 or more was
observed.
Results: Mean age was 59.68 (range 35-86), the average number of previous treatments was 2.98 (range 1-10), 38 patients
(65.52%) were ER+ and 22 (37.93%) had visceral disease. Median PFS was 7 months (CI 95% 4-9). Univariate analyses showed
a significant effect of the positivity of CK19 (HR=2.22, CI 95% 1.15-4.29, p=0.01) but did not reach statistical significance for CTC
(HR=1.86, CI 95% 0.93-3.72, p=0.07). The estimate disease-free survival rate at 6 and 12 months were 63.6% and 43.4% for
patients with CK19<1 and 47.1% and 10.6% for patients with CK19>=1, respectively. The estimate disease-free survival rate at 6
and 12 months were 77.8% and 28.8% for patients with CTC=0 and 43.6% and 21.5% for patients with CTC>=1, respectively. In
the multivariate analysis the effect of CK19 and CTC were similar (HR=2.12 and 1.86, p=0.03 and 0.08 respectively), ER was
statistically significant (HR=0.48, p=0.05, + vs -) but visceral diseases appeared not significant (HR=1.57, p=0.17, yes vs no).
Conclusions: The expression of CK19 appeared clinically meaningful in pretreated metastatic breast cancer patients, even when
adjusting by ER and visceral diseases. CTC showed a similar but minor effect and not statistically significant. These results
support CK19 as an interesting biomarker for predicting clinical response in metastatic breast cancer.

Title: The relationship between components of tumour inflammatory cell infiltration and hematogenous metastasis in patients with
primary breast cancer
Naping Wu1, Jue Wang1, Tiansong Xia1, Wenbin Zhou1, Zhao Liu1, Yi Zhao1, Xiaoan Liu1, Xiaoming Zha1 and Shui Wang1. 1The
First Affiliated Hospital with Nanjing Medical University, Nanjing, Jiangsu, China.
Body: Background: Tumors are infiltrated by inflammatory cells (e.g. macrophages, B and T lymphocytes), which may promote
tumor growth and invasion. However, the clinical correlation between hematogenous metastasis and relative amount of
inflammatory cells is still unknown.
Methods: The prevalence of circulating tumor cells (CTC) in peripheral blood of 133 breast cancer patients was detected by a
multimarker comprised of cytokeratin 19 (CK19), small breast epithelial mucin (SBEM) and human mammaglobin (hMAM) real
time quantitative RT-PCR (QPCR) platform. Meanwhile, CD68+macrophages, CD20+B lymphocytes and CD3+T lymphocytes at
the peritumor and intratumor were also counted by immunohistochemical examination, in corresponding primary invasive breast
carcinomas. The relationship between CTC in peripheral blood and inflammatory cell infiltration was analyzed.
Results: The positive rate of CTC was 39.8% (53 out of 133). The prevalence of CTC in peripheral blood was only observed in
invasive breast carcinomas. There were more inflammatory cells at the peritumor than the tumor center (P < 0.001).In addition,
higher total numbers of infiltrating both CD68+macrophages and CD3+T lymphocytes were associated with significantly the
prevalence of CTC in peripheral blood (P < 0.05).
Conclusion: Our data support a hypothesis that CD68+macrophage and CD3+T lymphocyte infiltrating in the breast cancer
microenvironment may promote tumor cells disseminating through circulation system.

Title: The combined detection of CTC and serum HER2 ECD predict PFS for HER2-positive advanced breast cancer patients
Zefei Jiang1, Jinmei Zhou1, Tao Wang1, Yi Liu1, Lei Li1, Huiqiang Zhang1, Shaohua Zhang1, Li Bian1 and Santai Song1. 1Affiliated
Hospital of Academy of Military Medical Sciences, Beijing, China.
Body: Background: Circulating tumor cell (CTC) and serum HER2 ECD can all reflect an aggressive tumor behavior. We
performed this prospective, monocenter, double-blinded study to investigate the potential clinical significance of combined
detection of CTC and serum HER2 ECD for advanced breast cancer patients with histological HER2-positivity.
Methods: A total of 88 eligible patients were enrolled in the present study from April 2012 to October 2013. We used Cell search
system and ADVIA Centaur System to detect CTC and serum HER2 ECD respectively. Patients received systemic treatment
according to national and international guidelines.
Results: Twenty nine (33%) patients had 5 CTC, seventy three (83%) patients had serum HER2 ECD values of at least 15ng/ml,
twenty seven (30.7%) patients had both elevated CTC and ECD values and fourteen (15.9%) patients had both normal CTC and
ECD values. Patients with both normal CTC and serum HER2 ECD values exhibited a significantly longer median PFS than
patients with both elevated values (9.0 months versus 2.8 months, p=0.023) and exhibited a trend toward longer PFS compared
with patients with elevated CTC or ECD values (9.0 months versus 4.2 months, p=0.065), patients with both or one elevated
values showed similar median PFS (2.8 months versus 4.2 months, p=0.211) (Figure1).
Conclusions: The combined detection of CTC and serum HER2 ECD showed prognostic significance for HER-2 positive
advanced breast cancer patients, patients with both normal values exhibited longer median PFS than others.

Title: A novel microfluidic system for the detection, enumeration and molecular analysis of circulating tumor cells (CTCs) in
metastatic breast cancer (MBC)
Carmela Paolillo1, Zhaomei Mu2, Angela Toss3, Priyadarshini Gogoi4, Saedeh Sepehri4, Yi Zhou4, Kalyan Handique4, Ye Zhong2,
Hushan Yang2, Ettore Capoluongo, Massimo Cristofanilli2 and Paolo Fortina2. 1Catholic University of the Sacred Heat Rome,
Rome, Italy; 2Thomas Jefferson University, Kimmel Cancer Center, Philadelphia, PA; 3University of Modena and Reggio Emilia,
Modena, Italy and 4DeNovo Sciences, Inc, Plymouth, MI.
Body: Background: In recent years blood testing for circulating tumor cells (CTC) has gained increasing interest in cancer
research. CTC detection and enumeration can serve as a liquid biopsy and an early marker of response to systemic therapy.
Different analytical systems for CTC detection and isolation have been developed, but the CellSearch is currently the only
FDA-approved technology. We aimed to evaluate CTCs detection by a novel microfluidific technology, a size- and
deformability-based capture system. This unique platform not only allows flexibility in the selection of antibody markers but also
segregates the CTCs in their own chambers, thus, enabling morphological, immunological and genetic characterization of each
CTC at the single cell level.
Methods: We performed a prospective study to compare the detection of CTCs using the CellSearch (Janssen Diagnostics) vs.
the new microfluidic platform. Enumeration by the CellSearch was performed according to standard protocol. For the
microfluidific-device capture (De Novo Sciences, MI) peripheral blood from MBC patients was diluted 1:1 with PFA 0.8% and PBS
1%. Prior to sample loading, the microfluidic device was coated with priming buffer. After cells were fixed using 4.0% PFA and
were subsequently stained with pancytokeratin, Zym 5.2 and CD45. Nuclei were counterstained with Hoechst-33342. CTCs
were identified as round and bright green cells not stained with red (CK+/CD45-/DAPI+). Data were analyzed using
non-parametric methods: Cohens kappa, Chi2 test, Spearman rank correlations and Mann-Whitney test. Associations with CTCs
were calculated in two ways: CTCs as a continuous variable normalized for ml of blood and CTCs categorized as < 5 versus 5.
Results: The two methods was concordant in 88.2% of patients with a Cohens kappa of 0.743 when the detection of a single
CTC is considered like positive. We also found a concordance of 85% (k=0.70) when we use the CTC cut-off level of 5 cells per
7.5 ml of blood to identify patient with higher risk for disease progression. Thirty-one patients with MBC were tested for CTCs
using both methods. CTC detection by microfluidific platform was positively associated with Her2 positive patients if we consider
as categorical variable; a weak association was seen also with the CellSearch.
CTCs Association
CTC/ml of blood Continuous
DeNovo
median [min, max]
Her 2
CTC categorized (5)

p-value
n (%)
OR ( 95% CI)
0.562
0.038
No
0.5 [0, 62]
19
7 (36.8)
ref
Yes
1.75 [0, 4]
12
9 (75.0)
5.14 (1.03, 25.6)
n (%)
OR ( 95% CI)
CellSearch
median [min, max]
Her 2
p-value
p-value
0.704
p-value
0.056
No
0.13 [0, 335.7]
19
6 (41.2)
ref
Yes
0.95 [0, 3.33]
12
8 (34.7)
4.33 (0.93, 20.2)
Conclusions: The enumeration of CTCs showed strong prognostic significance in MBC raising interest in a more accurate
molecular characterization to achieve the possibility for a dynamic molecular monitoring. Introducing novel methodologies should
demonstrate comparable detection rate for epithelial cells. This new microfluidic system showed accuracy and flexibility to move
to the next phase of molecular testing with the intent of assessing the value as liquid biopsy.

Title: Analysis and single-cell retrieval of circulating tumor cells to monitor treatment response and assess genotype in
triple-negative breast cancer
Arturo Ramirez1, Daniel Campton1, Elisabeth Mahen2, Sibel Blau3, Anthony Blau2, Eric Kaldjian1 and Jackie Stilwell1. 1RareCyte,
Inc, Seattle, WA; 2Center for Cancer Innovation, University of Washington, Seattle, WA and 3Northwest Medical Specialties,
Puyallup, WA.
Body: Introduction: We used a high-recovery rare cell analysis and single-cell picking system to enrich, visualize, and isolate
circulating tumor cells (CTCs) for genomic analysis from the blood of patients with advanced triple-negative breast cancer (TNBC)
undergoing treatment with cisplatin as part of a study to intensively characterize TNBC. CTCs were evaluated regularly during
treatment to monitor CTC burden and characteristics that could be associated with treatment response or disease progression,
and perform single-cell mutational analysis to inform clinical decision making. Methods: Patients were enrolled in the study at the
University of Washington Center for Cancer Innovation after informed consent for participation in investigation of their disease,
including molecular analysis of multiple biopsies of accessible tumor. CTCs were evaluated prior to treatment and tracked
longitudinally. Density-based enrichment of blood cells was performed using the AccuCyte tube, float and collector system.
Collected cells were processed and applied to microscopic slides. Fluorescently labeled antibodies to cytokeratin, CD45 and
EpCAM, and a nuclear dye were applied to samples using an automated slide stainer. Slides were scanned on a digital
microscope and candidate CTCs identified using image analysis software. CTCs were verified by appropriate morphology and
expression of epithelial and nuclear stains without CD45 expression. Other antibodies used to characterize cells included Her2,
EGFR, and Ki-67. A mutation hypothesized to lead to the activation of ROS1 was identified in the cancer cells isolated from the
bone marrow of one patient. CTCs were picked using our integrated semi-automated system and evaluated for the ROS1 variant
using whole genome amplification followed by nested PCR and Sanger sequencing. Results: Seven patients have been enrolled
to date. At least 1 CTC/mL has been found in all patients. Pre-treatment CTC levels in the patient with the ROS1 mutation were
extremely high (1500/mL). One week after treatment, CTC levels spiked to more than 5000/mL. CTC counts then dropped
exponentially to 9/mL after 4 months. CTC clusters and Ki-67 positive cells also decreased during therapy. Treatment with
cisplatin was discontinued in this patient due to toxicity and progression, and CTC levels increased to nearly 9000/mL over 4
months. The ROS1 mutation was found in approximately 50% of individually picked CTCs before treatment with crizotinib, a
ROS1 inhibitor. A second patient was found to have somatic loss of BRCA1, and was therefore treated with the PARP inhibitor,
veliparib. CTC levels increased during veliparib treatment up to 13/mL. The same patient was subsequently treated with
ponatinib, an FGFR inhibitor, based on the identification of two linked somatic missense mutations involving FGFR2 (S252W and
Y375C). After beginning ponatinib, CTCs fell to undetectable levels. Conclusions: Analysis of CTCs may provide a non-invasive
measure of cancer progression/response and the molecular evolution of tumor cells in patients with TNBC. Single-cell CTC
retrieval after slide-based immunofluorescent visualization is compatible with whole genome amplification and sequencing
methods.

Title: High-depth sequencing of circulating tumor DNA to interrogate the genetics of residual micro-metastatic disease prior to
relapse in early breast cancer
Charlotte KY Ng1, Britta Weigelt1, Isaac Garcia-Murillas2, Gaia Schiavon2, Sarah Hrebien2, Rosalind J Cutts2, Peter Osin3,
Ashutosh Nerurkar3, Iwanka Kozarewa2, Javier Armisen Garrido2, Mitch Dowsett2, Ian E Smith3, Jorge S Reis-Filho1 and Nicholas
C Turner2. 1Memorial Sloan Kettering Cancer Center, New York, NY; 2Breakthrough Breast Cancer Research Centre, The
Institute of Cancer Research, London, United Kingdom and 3Breast Unit, Royal Marsden Hospital, London, United Kingdom.
Body: Background: The identification of early stage breast cancers that are at high risk of relapse after apparently curative
treatment of the primary tumor would allow tailored adjuvant therapy approaches to prevent relapse. We sought to define whether
high-depth targeted sequencing of circulating cell-free plasma DNA could be used to interrogate the genetics of residual
micro-metastatic disease (RMD) persisting after neoadjuvant treatment in patients with early stage breast cancer.
Methods: In a cohort of 31 patients with early breast cancer receiving neoadjuvant chemotherapy, with no evidence of metastasis
at presentation, we collected tumor tissue at baseline with the metastatic relapse if it occurred, and serial plasma samples at
baseline, post-surgery, and every 6 months in follow-up. Serial plasma samples were subject to digital PCR mutation tracking to
examine for circulating tumor DNA (ctDNA). Four patients had ctDNA detectable prior to clinical relapse. Primary tumor-derived
DNA (from 4 cases), serial plasma DNA samples (from 4 cases), and metastases derived DNA (from 2 cases) were subjected to
massively parallel sequencing (Illumina HiSeq2000) targeting all exons of 273 genes frequently mutated in breast cancers and/ or
a custom AmpliSeq cancer panel (IonTorrent).
Results: Targeted sequencing revealed the presence of 1 to 20 somatic mutations in the early breast cancers at baseline,
including 3 PIK3CA H1047L/R and 2 TP53 pathogenic mutations, all of which were detected in the subsequent plasma DNA
samples taken prior to relapse. However, in one case an ESR1 E380Q mutation found in the baseline primary breast cancer was
undetectable by targeted sequencing in all 3 subsequent plasma samples and in the metastasis; this mutation could not be
identified in the metastasis by digital PCR. In two cases plasma DNA sequencing revealed no additional mutations to those
identified in the primary tumor, whereas in the other two cases, plasma DNA targeted capture sequencing revealed divergence in
the genetics of RMD. In these cases, 1 and 5 somatic mutations were found in the plasma DNA but were not detected in the
respective baseline tumors. In particular, an activating FGFR1 K656E mutation, which was not detected by targeted capture
sequencing in the primary breast cancer at 355x sequencing depth, was found at a mutant allele fraction (MAF) of 2.6% in the
plasma 12 months post-surgery, and was subsequently detected in the distant metastasis at a MAF of 43.4% by targeted
sequencing and digital PCR.
Conclusions: Our results provide evidence of clonal shifts in response to neoadjuvant systemic therapy of early breast cancers.
In addition, we demonstrate that high-depth targeted capture massively parallel sequencing analysis of plasma ctDNA may help
predict the genotype of recurrence prior to the onset of clinically overt metastasis.

Title: Targeted molecular characterization of serum derived cell free DNA from metastatic breast cancer patients treated with
first-line tamoxifen
Maurice Jansen1, Corine Beaufort1, Jean Helmijr1, Ronald van Marion1, Niels Krol1, Kim Monkhorst1, Marian van Fessem1, Ron
van Schaik1, Marcel Smid1, Marion Meijer van Gelder1, Maxime Look1, Anita Trapman1, Diana Ramirez-Ardila1, Irene Lurkin1, Ellen
Zwarthoff1, John Martens1, John Foekens1, Winand Dinjens1, Stefan Sleijfer1 and Els Berns1. 1Erasmus MC Cancer Institute,
Rotterdam, Netherlands.
Body: Background: Evaluation of cell-free DNA (cfDNA) is a very attractive tool to serve as "liquid biopsy" to define and
establish mutational changes in circulating tumor DNA (ctDNA) non-invasively during therapy.
Aim: To identify tumor specific mutations in serum cfDNA associated with resistance against tamoxifen in metastatic breast
cancer.
Materials & Methods: Ten metastatic ER-positive breast cancer patients treated with first-line tamoxifen of whom blood sera was
available at start therapy, during treatment and at disease progression were selected. DNA was isolated from normal (nDNA) and
primary tumor (ptDNA) tissue and from sera (cfDNA). DNA was analyzed with next generation sequencing by the ion-PGM
system (Life Technologies) for a panel of 45 cancer genes. This panel included the most frequently mutated genes for breast,
colon, prostate and ovarian cancer reported in the Catalogue Of Somatic Mutations In Cancer (Cosmic database, Release 67;
cancer.sanger.ac.uk/). In total 1242 exons (255kb) were sequenced with 200 to 5000x reads depth coverage. The panel
analyzed all exons for 39 genes and only hotspot exons for 6 oncogenes. Variant Caller software (Life Technologies; version
4.16) was applied to detect non-synonymous and stop-gain single nucleotide variants (SNVs) within the sequenced DNA. Hotspot
mutations detected in PIK3CA exons 9 and 20 were validated with snapshot multiplex assays.
Results: Variant Caller analyses revealed in total 252 SNVs within the 40 DNAs from tumor tissue and/or serum analyzed which
were not detected in normal tissue. Of these, 229 SNVs were novel and not yet reported in the Cosmic database. Mutations were
detected for 10 genes in both ptDNA and cfDNA, which enabled us to characterize ctDNA in serum of nine out of ten patients.
The discovered mutations were already reported in Cosmic for PIK3CA, TP53, and NF1, but not for CDH1, APC, SMAD4, MLL,
AKAP, CREBBP and MLL2. The PIK3CA, TP53 and APC mutations were observed in tumor and sera for 2, 3 and 2 patients,
respectively. Mutations in the other genes, except MLL, were unique for individual patients. The mutation for MLL was seen in
almost all patients at low frequencies (1-5%). The hotspot mutations in PIK3CA were confirmed with snapshot assays in the
sequenced samples and in additional sera. In four patients, Cosmic reported mutations in BRCA1, KAT6B, MAP3K1, MLL2,
PTCH1 and PTEN occurred in ctDNA at disease progression while they were not found in the primary tumor nor in preceding
sera. The mutation frequencies ranged from 3% for BRCA1 and MAP3K1 to 6% for MLL2. Moreover, mutations for MED12 (3%)
and MLL3 (4%), not yet reported in Cosmic, were each observed in two sera at disease progression for two additional patients.
The remaining SNVs are currently verified for their authenticity and occurrence in ptDNA and/or ctDNA.
Conclusion: Molecular characterization of tumor and serum derived DNA with targeted next generation sequencing enabled us
to identify ctDNA in serum and to detect mutations at disease progression that might play a role in resistance to first-line
tamoxifen.

Title: Detection of single nucleotide variations and copy number variations in breast cancer tissue and ctDNA samples using
single-nucleotide polymorphism-targeted massively multiplexed PCR
Robert J Pelham1, Bernhard G Zimmermann1, Eser Kirkizlar1, Ryan K Swenerton1, Bin Hoang1, Onur Sakarya1, Joshua E
Babiarz1, Nicholas Wayham1, Tudor Constantin1, Styrmir Sigurjonsson1, Matthew Rabinowitz1 and Matthew Hill1. 1Natera, Inc.
Body: Genomic instability, the hallmark of cancer, presents with a variety of mutation types, most commonly single nucleotide
variations (SNVs) and copy number variations (CNVs), which traditionally have required different methods for identification. It has
proven challenging to simultaneously achieve sufficient breadth to detect CNVs and depth to detect SNVs on samples of limited
input amount. The objective of this study was to validate a new methodology for detection of SNVs and CNVs in a single assay.
We used a massively multiplex PCR/NGS approach combining an SNV panel covering 585 point mutation hotspots in breast
cancer (Cosmic) and a CNV panel targeting 28,000 SNPs designed to detect copy number at chromosomes 1, 2, 13, 18, 21, and
X, and focal regions 4p16, 5p15, 7q11, 15q, 17p, 22q11, and 22q13. We applied these panels to breast cancer cell lines and
fresh frozen (FF) breast tumor samples; the presence of CNVs in circulating cell-free tumor DNA (ctDNA) in the plasma of breast
cancer patients was also investigated.
The CNV assay methodology was validated using genomic DNA isolated from 96 human samples with known karyotype;
sensitivity to single region deletions or duplications was 100% (71/71) and specificity was 100% for normal regions in the same
samples. Single-molecule sensitivity for the detection of CNVs was established by analyzing isolated single cells. Performance of
the mutation assay was demonstrated with the analysis of 5 matched tumor and normal cell lines, with 24 out of 27 SNVs known
to be present in these cell lines detected. The 3 undetected SNVs were determined to be a result of assay design failure. Also,
multiple somatic CNVs (median: 13) were detected in all 5 tumor cell lines. Analysis of the normal cell lines found no cancer
related SNVs or CNVs.
In 32 FF tumor samples, 78.1% (25/32) had SNVs detected; of samples with SNVs, 88% (22/25) had SNVs in TP53 or PIK3CA.
Of the same 32 FF breast tumor samples, 96.9% (31/32) showed full or partial CNVs in at least 1 and up to 15 regions; of the 31
samples with detected CNVs, 93.5% had a CNV of either 1q or 17p, two of the three most prevalent breast cancer CNVs (the 16q
region was not represented in this panel). Overall, a combination of SNV and CNV testing allowed identification of genetic
changes in 100% of the breast tumor samples, a significant improvement in diagnostic yield than using SNV detection alone.
Of the 12 breast cancer patients with matched tumor tissue and plasma samples, 83.3% (10/12) had CNVs detected in tissue.
The CNVs present in each primary tumor sample were identified in corresponding plasma ctDNA samples (1 stage IIa, 7 stage
IIb, and 2 stage III). The ctDNA fractions in these samples ranged from 0.58 to 4.33%; detection required as few as 86
heterozygous SNPs per CNV.
Analysis of ctDNA for cancer-associated mutations may allow earlier, safer and more accurate profiling and monitoring of breast
cancer. Thus, this targeted PCR approach offers the promise of an assay able to detect both cancer-associated SNVs and CNVs
in the same sample with good sensitivity and specificity, and improved detection rates compared to assays that only detect SNVs.

Title: The significance of serum HER2 levels at diagnosis on outcome of breast cancer patients by molecular subtype
Moo Hyun Lee1, Siew Kuan Lim2, Ji Young You1, Eun Jin Song1, So-Youn Jung1, See Youn Lee1, Han-Sung Kang1 and Eun Sook
Lee1. 1National Cancer Center, Goyang, Gyeonggi-do, Korea and 2Changi General Hospital, Singapore.
Body: Background/purpose: Increased serum HER2 levels have been shown to correlate with higher tumor burden and poorer
clinical outcomes in metastatic breast cancer, but there is little data regarding the significance of serum HER2 levels at diagnosis
in the operable breast cancer. We evaluated the association between baseline serum HER2 levels and clinicopathologic
parameters, and the correlation of baseline serum HER2 levels with clinical outcome by molecular subtype in operable breast
cancer patients.
Methods: We included patients with stage I-III breast cancer diagnosed at our center between October 2004 and December
2011. Baseline serum HER2 levels were measured by chemiluminescence immunoassay at diagnosis. Patients were categorized
into 4 molecular subtype groups by their hormone receptor (HR) status and HER2 status: HR+/HER2-, HR+/HER2+, HR-/HER2+
and HR-/HER2-. HR and HER2 status were determined by immuhohistochemistry (IHC) in all tumors, and fluorescence in situ
hybridization (FISH) assay was performed whenever the HER2 status was equivocal.
Results: There were 439 consecutive stage I-III breast cancer patients, of which 271 (61.7%) were HR+/HER2-, 75 (17.1%) were
HR+/HER2+, 30 (6.8%) were HR-/HER+ and 63 (14.4%) were HR-/HER2-. A total of 192 patients underwent neoadjuvant
chemotherapy and the remaining 235 patients did not. Of these 235 patients, 204 underwent adjuvant chemotherapy. High serum
HER2 levels (15 ng/ml) were reported in 51 patients (11.6%) and HER2-positive status in tumor tissue was observed in 105
patients (23.9%). High serum HER2 levels were significantly associated with tumor size > 2cm (p=0.032), high histologic grade
(0.048), negative HR status and neoadjuvant chemotherapy usage. Patients who had high baseline serum HER2 levels had a
worse disease free survival (DFS) (p<0.001). Especially, high baseline serum HER2 levels were associated with worse DFS in
HR+/HER2-, HR+/HER2+ and HR-/HER2+ subtypes (p=0.005, 0.001 and 0.027 respectively). However there was no significant
correlation between baseline serum HER2 and DFS in the HR-HER2- subtype (p=0.614).
Conclusions: This is the largest study in the evaluation of serum HER2 levels at diagnosis on the prognosis of patients with
operable breast cancer. Our results showed that baseline serum HER2 level is a good prognostic marker in patients with operable
breast cancer. High serum HER2 levels at diagnosis were associated with worse DFS, especially in patients of HR+/HER2-,
HR+/HER2+ and HR-/HER2+ subtypes.

Title: A blinded multicenter phase II study of a panel of plasma biomarkers for the detection of triple negative breast cancer
Karen S Anderson1, Margaret Pepe2, Jeffrey Marks3, Paul Engstrom4, Christos Patriotis5, Richard Zangar6, Steven Skates7, Paul
Lampe2, Joshua LaBaer1 and Christopher I Li2. 1Biodesign Institute, Arizona State University, Tempe, AZ; 2Fred Hutchinson
Cancer Research Center, Seattle, WA; 3Duke University Medical Center, Durham, NC; 4Fox Chase Cancer Center, Philadelphia,
PA; 5National Cancer Institute, Rockville, MD; 6Pacific Northwest National Laboratories, Richland, WA and 7Massachusetts
General Hospital, Boston, MA.
Body: Background: Triple negative breast cancers (TNBC) comprise 15-20% of all breast cancers and frequently present as
interval cancers with high proliferative rates and increased risk of mortality. There is a clinical need for biomarkers for the early
detection of TNBC to complement radiologic imaging. No plasma biomarkers for TNBC currently exist. The purpose of this study
is to evaluate a panel of novel plasma biomarkers for TNBC for the discrimination of TNBC and benign breast disease as a crucial
step in identifying a panel of plasma biomarkers for early detection.
Methods: In a multicenter collaboration between the NCI EDRN and the CPTAC consortium, we conducted a prospective blinded
phase II biomarker study that evaluated 76 candidate TNBC plasma biomarkers. Plasma samples collected at the time of
diagnosis from 65 TNBC cases and 195 matched controls with benign breast disease without atypia were identified from multiple
clinical sites. The samples were distributed as blinded aliquots to the biomarker laboratories for protein and autoantibody
detection. Candidate protein (n=54) and autoantibody (n=22) biomarkers were selected and ranked prior to evaluation. All results
were centrally analyzed. The sensitivity at 95% specificity was calculated for each biomarker. Effects of age, race, and specimen
source on biomarkers were evaluated. Logistic regression was used to assess complementarity of biomarkers. The top three
biomarkers underwent verification with an independent set of 60 TNBC cases and 180 matched controls from women undergoing
mammography.
Results: Statistically significant differences in case versus control signals were observed for 3 biomarkers with sensitivities of
17-23% at 95% specificity (p <0.008, unadjusted) and areas under the curve of 0.55-0.57. These three biomarkers were
confirmed in the independent set of 60 cases and 180 controls with sensitivities 12-23% at 95% specificity and cross-validated
combined sensitivity of 27%. In the subset of women ages 50-79 (cases n=38, controls n=119), five biomarkers were identified
with sensitivities of 15.7-16.2% at 95% specificity (p<0.10).
Conclusion: We have developed a pipeline strategy for the validation of plasma biomarkers for detection of breast cancer. At
least three biomarkers for TNBC were confirmed in this study. Further evaluation of these biomarkers for early detection is
ongoing.

Title: No Show

Title: No Show

Title: Global quantitative measures using next-generation sequencing for breast cancer presence outperform individual tumor
markers in plasma
Ellen M Beasley1, Richard D Abramson1, Gregory E Alexander1, David Chan2, Kristen Bradley2, Francois Collin1, Michael Crager1,
Andrew Dei Rossi1, Joseph Dorado1, Adam Friedman1, William J Gibb1, Jennie Jeong1, Col Jones1, C J Ku1, Yan Ma1, John
Morlan1, Kunbin Qu1, Aibing Rao1, Aaron Scott1 and Haluk Tezcan1. 1Genomic Health Inc, Redwood City, CA and 2Cancer Care
Associates, TMPN, Redondo Beach, CA.
Body: Background: Analytically and clinically validated non-invasive blood tests that quantify breast cancer burden and clinical
drug response/resistance are greatly needed. Many groups have successfully detected tumor markers in blood using a variety of
technologies, including next generation sequencing (NGS). We performed a comprehensive NGS study on a small number of
patients to evaluate the value of global versus individual markers for the quantitation of tumor-derived cell free DNA (cfDNA) in
plasma.
Methods: DNA isolated from formalin-fixed primary tumor, buffy coat cells, and plasma from 2 patients with metastatic breast
cancer were characterized simultaneously for copy number aberrations (CNAs) and differentially methylated regions (DMRs)
using whole genome bisulfite sequencing (WBGS), and targeted sequencing-based genotyping of 346 cancer-associated single
nucleotide variations (SNVs). CNA and DMR regions were identified from log normalized, GC content corrected counts and DMR
data using Poisson and binomial distribution theory and false discovery rate controlling methods. Percent tumor in cfDNA was
estimated from the normalized ratio (plasma: primary tumor) of CNA or DMR compared to buffy coat, aggregating over genomic
regions. Sample sets from 8 non-metastatic patients were also profiled using the targeted SNV panel in order to compare SNVs
between samples and estimate percent tumor cfDNA.
Results: WGBS detected tumor specific alterations in each primary tumor compared to buffy coat. By analyzing the genome
using 100 Kb bins, we observed over 1000 bins with detectable CNA signal and, among 56 million CpG sites, over 30,000 DMRs.
As expected, 5 or fewer informative somatic SNVs were detected in each patient. Analysis of these somatic changes in plasma
revealed that the tumor fraction estimated from SNV detected in cfDNA varied widely between sites originally discovered in the
patients primary tumor. In contrast, similar estimates of tumor fraction in cfDNA were obtained using CNA and DMR profiles
within each patient; both methods yielded similar estimates of over 50% in one patient and less than 10% in the other. For the
patient with high tumor fraction, both CNA and DMR profiles contained examples of individual large genomic regions that
displayed additional clear aberrations in the plasma compared to the original tumor, such as a striking loss of a >25 Mb region of
chromosome 4.
Conclusions: Although individual somatic SNV in cfDNA can be detected in metastatic disease, calculated allelic fraction based
on individual SNVs varies greatly within the same patient. Measuring and integrating CNA or DMR across the genome provided
more consistent and reliable estimates of tumor DNA fraction in plasma, and also revealed alterations in plasma from patients
with metastatic disease that were not prominent in the primary tumor.

Title: RUNX3 hypermethylation in circulating tumor DNA is a biomarker of breast cancer distant metastasis
Bodour Salhia1, Gerald C Gooden1, Kyle N Johnson1, Waibhav Tembe1, Garrick Wallstrom2, Winnie Liang1, Manmeet Ahluwalia4,
Steven N Kalkanis5, Tom Mikkelsen5 and Steven A Toms3. 1Translational Genomics Research Institute, Phoenix, AZ; 2Arizona
State University, Tempe, AZ; 3Geisinger Health System, Danville, PA; 4Cleveland Clinic, Cleveland, OH and 5Henry Ford Hospital,
Detroit, MI.
Body: Breast cancer detection has not resulted in the expected decrease in mortality suggesting that current screening strategies
are not finding metastasis-prone lesions before metastasis occurs and instead are detecting many lesions that will not progress to
metastatic disease. Although several improvements in the management of breast cancer have been made, an estimated 90% of
breast cancer related deaths are due to the development of distant metastasis to organs such as lung, bone, liver, and brain.
Metastatic breast cancer affects 10-15% of breast cancer patients, is often refractory to therapy and associated with poor
outcomes. However, a blood test indicating a high-risk population at the time of breast cancer diagnosis might shift the current
strategy and improve outcomes. Efforts for identifying predictive markers of breast cancer metastasis have been impeded by a
lack of data defining the genetic and epigenetic determinants of the disease. Numerous studies suggest that DNA methylation
could be a useful biomarker for improving the clinical management. Recently circulating tumor (ct)DNA has attracted attention for
clinical use in the context of risk prediction, prognostication and prediction in human cancer. Various types of DNA alterations
have been reported in ctDNA but methylated DNAs are chemically stable and are among the most sensitive and specific ctDNA
cancer markers. The objective of our study was to identify and validate a blood-based DNA methylation biomarker for breast
cancer patients at high risk of distant metastasis. Using the HumanMethylation27K bead array, we recently identified RUNX3
hypermethylation involving 3 CpG probes in 32 breast brain metastasis tumor tissue compared to a series of 50 unmatched early
stage breast cancer samples. We calculated the area under the curve (AUC) to exceed 0.8, with a higher AUC achieved in
Her2-positive samples. In an independent, first of its kind analysis, we performed whole genome bisulfite sequencing (WGBS) of
ctDNA from plasma samples representing three pools of 40 women with distant metastatic breast cancer to various organs, 40
women with metastasis-free breast cancer, and 40 non-breast cancer controls. In this unbiased analysis, these data
demonstrated RUNX3 hypermethylation in the plasma of patients with metastatic breast cancer in the same region corresponding
to the CpG probes of the methylation array. We are now performing bisulfite pyrosequencing of each individual plasma sample,
and in an additional cohort of 50 archival primary/brain metastasis tumor pairs with matching serum and clinical outcomes data.
To determine regional specific differences we have also selected two additional regions of RUNX3 not differentially methylated for
validation by bisulfite pyrosequencing. Determining the extent to which RUNX3 hypermethylation occur in the context of brain
metastasis versus other distant sites are ongoing studies. Together these studies point towards RUNX3-regional specific
methylation as a potential blood-based biomarker of breast cancer metastasis. Such a biomarker will help establish surveillance
methods and eventually preventative regimens that will increase the likelihood of long-term progression-free survival for breast
cancer patients at risk of distant relapse.

Title: Detection of H1047R and E545K PIK3CA mutations from peripheral blood in ER positive breast cancer patients
Jose Angel Garcia-Saenz1, Daniel Acosta1, Fernando Moreno1, Paticia Ayllon1, Miguel Sotelo1, Trinidad Caldes1, Eduardo
Diaz-Rubio1 and Atocha Romero1. 1Hospital Clnico San Carlos, Instituto de Investigacin Sanitaria del Hospital Clnico San
Carlos (IdISSC), Madrid, Spain.
Body: Background: Digital polymerase chain reaction (PCR) is a new technology that enables detection and quantification of
cancer DNA molecules from peripheral blood. Detecting tumor-specific mutations in circulating plasma DNA may potentially be
useful to select systemic therapies for solid tumors. The aim of our study is to evaluate the feasibility of detecting PIK3CA
mutations from plasma of breast cancer patients.
Methods: We have designed an allele-specific PCR assay and used a digital PCR system for the detection of PIK3CA mutations
H1047R and E545K. Formalin-fixed paraffin-embedded (FFPE) tumor samples were analyzed by COBAS and by digital PCR.
The matched plasma samples were then analyzed by digital PCR. Kappa Cohens coefficient () was used to test agreement
between methods
Results: 37 ER positive breast cancer and matched plasma were evaluated. 31 (84%) patients were stage IV disease. E545K
mutation was detected in 4 out of 37 (11%) tumor specimens and H1047R mutation was identified in 8 out of 37 (22%) tumor
samples. The proportion of observed agreement between H1047R and E545K detection in tumor samples by COBAS and digital
PCR was 100%. Regarding E545K mutation assay we found that the proportion of observed agreement between mutation status
in FFPE samples and circulating tumor DNA (ctDNA) was 94.6% (kappa= 0.770; Sensitivity=100%; Specificity=93.9%). In
H1047R assay the proportion of observed agreement between FFPE samples and ctDNA was 83.78% (Kappa=0.561;
Sensitivity=60%; Specificity=92.59%).
Conclusions: The agreement between methodologies when assessing the PIK3CA status in tumor samples was perfect.
However, a moderate-to-fair agreement was found between FFPE samples and ctDNA which might be due to the heterogeneity
of the disease. The methodology presented in this study is a feasible approach for PIK3CA mutation detection in blood derived
samples.

Title: Circulating methylated DNA of pregnancy-associated plasma protein A as a clinically useful biomarker in breast cancer
Richard Sainsbury1, Karsten Koehler2, Catherine Page2 and Saroj Velamakanni2. 1UCL, London, United Kingdom and 2Fahy
Gurteen Labs, Cambridge, United Kingdom.
Body: Background: A recent report suggests that Pregnancy-Associated Plasma Protein A (PAPPA) plays a pivotal role in normal
cell division and is targeted early in breast cancer development. Epigenetic silencing of PAPPA is highly prevalent in precursor
lesions and invasive breast cancer, correlating with loss of PAPPA protein expression. In normal breast tissue PAPPA promoter
methylation was not observed (Journal of Pathology 344-356: 2014). We sought to test the hypothesis that PAPPA promoter
hypermethylation can be detected in cell-free tumour specific DNA (ctDNA) in the bloodstream of patients with metastatic breast
cancer.
Methods: Plasma derived from 12 patients with metastatic breast cancer was examined for ctDNA. A semi-quantitative real-time
PCR based MethyLight assay was used for detection of PAPPA methylation. Spiking experiments with breast cancer cell lines
were performed to develop the assay and confirm linearity of sodium bisulfite treatment.
Results: Percentage methylated reference (PMR) values from spiking experiments with PAPPA methylated and non-methylated
DNA in plasma (matrix) demonstrate good linearity for the PAPPA gene and the endogenous control Col2A. PAPPA methylation
was detected in 10 out of 12 (83%) of the metastatic breast cancer samples. In stark contrast, epigenetic silencing of PAPPA was
not detected in any of the plasma samples from non-cancer bearing patients (n=10).
Conclusions: Given the potential clinical impact of our data, this study is being extended to include early stage breast cancer
patients with poor prognostic clinic-pathological indicators. Our proof-of-concept findings indicate that testing for PAPPA promoter
methylation in ctDNA may be used as a biomarker of disseminated disease, helping clinicians with treatment decisions and
adjusting those choices as conditions change. By developing and refining more sensitive techniques, PAPPA methylation testing
in ctDNA could also find practical utility in breast cancer screening.

Title: Role of disseminated tumor cell (DTC) in bone marrow (BM) detected at primary diagnosis on overall survival and central
nervous system (CNS) recurrences in a cohort of 620 early breast cancer patients after 11 years follow up
Delphine Loirat1, Franois-Clment Bidard1, Lonor Chaltiel1, Frdrique Berger1, Vronique Dieras1, Yann de Rycke1, Sverine
Alran1, Youlia Kirova1, Paul Cottu1, Anne Vincent- Salomon1, Xavier Sastre-Garau1 and Jean-Yves Pierga1. 1Institut Curie, Paris,
France.
Body: Background: BM DTC detection is known to be a prognostic factor for distant metastasis and overall survival in early
breast cancer. However, distant relapses may occur in patients with no DTC detected at time of primary cancer. In this study, we
studied the impact of early breast cancer pathological features (including BM DTC status) on the incidence of CNS metastases.
Methods: In a cohort (1998-2005) of 620 early breast cancer patients in whom BM DTC were detected using an anti-cytokeratin
antibody (A45B/B3) (Bidard et al., 2008, Clinical Cancer Research, outcome updating (median follow up 11y) on overall survival
(OS) and distant metastasis free survival (DMFS) was assessed by log rank test and uni- and multivariate analyses. 137 patients
of this cohort have a later metastatic relapse. Chi2 test were performed to compare patient with CNS metastasis and patients with
other metastases locations. CNS metastasis survivals were estimated using Kaplan-Meier method and compared by log-rank
test.
Results: At eleven-year median follow-up, the prognostic value of DTC detection was confirmed in the 620 patients cohort for OS
(p=0.03 in multivariate analysis), but DTC were probably less involved in late metastatic events.
Distant metastases were diagnosed in 137 (22%) patients. In the course of the metastatic disease, 55 of these patients (40%)
have been diagnosed with CNS metastases (parenchymal=30, leptomeningeal=10, both localizations=15), as first metastatic or
later event. Patients with CNS metastasis were mostly less than 50 years old (60% vs 38%, p=0.01), pN0 (31% vs 10%,
p=0.002), hormonal receptor status negative (54% vs 30%, p=0.007) and HER2 status positive (50% vs 22%, p=0.01). Strikingly,
although DTC detection was associated with development of distant metastasis in the whole cohort, the occurrence of CNS
metastasis was found to be higher in patients who had no DTC detected at primary diagnosis (p=0.016). From CNS metastasis,
median survival was 7.8 months (8.3 months for parenchymal and 2.4 months for leptomeningeal recurrences). For HER2+
patients, the median survival after diagnosis of CNS metastasis was 16.6 months (N=15) and 4.1 months for Her2- patients
(N=15).
Conclusions: After a long-term follow up, DTC remain an adverse prognostic factor in early breast cancer. Our study suggests
that cancer cells with epithelial differentiation, as detected in the BM, are less prone to CNS dissemination. This study also
confirmed previously reported determinants of CNS metastases (age, RH-, HER2+) in the pre-adjuvant trastuzumab era, but also
showed that the outcome of HER2+ patients with CNS relapse was longer, probably due to HER2 targeted therapy during
metastatic course.

Title: A new breast cancer classification scheme based on novel classes of tumor stroma
Crista Thompson1, Nicholas Bertos1, Tina Gruosso1, Greg Finak1,2, Robert Lesurf1,2, Sadiq M Saleh1,2, Hong Zhao1, Margarita
Souleimanova1, Sarkis Meterissian3, Atilla Omeroglu4, Michael T Hallett1,2 and Morag Park1. 1Goodman Cancer Research Centre,
Montreal, QC, Canada; 2McGill Centre for Bioinformatics, Montreal, QC, Canada; 3McGill University, Montreal, QC, Canada and
4
McGill University, Montreal, QC, Canada.
Body: A major challenge in cancer treatment is the heterogeneous nature of the disease. This is particularly evident in breast
cancer where gene expression profiling of whole tumours has identified multiple intrinsic subtypes of breast cancer. These
subtypes are associated with differential prognoses and are correlated with previously identified clinical biomarkers (i.e., ER and
HER2 status) used to stratify patients for targeted therapy. Despite recent studies demonstrating that elements within the tumour
microenvironment can affect breast cancer progression and outcome, and that information contained within this compartment
carries significant prognostic information for patient stratification above and beyond the information supplied by the intrinsic
subtypes and existing therapeutic biomarkers, a limited understanding of stromal heterogeneity across the population has
hindered the development of effective prognostic tools and targeted therapies directed against these processes. Here we perform
expression profiling of the microdissected tumour-associated stromal components of 49 human breast tumours, and demonstrate
that stromal heterogeneity can be captured by categorization into six classes which bear distinct molecular phenotypes. These
stromal classes exhibit distinct biological functions and carry prognostic information independent of existing tumor-intrinsic
biomarkers and molecular breast cancer subtypes. Specific combinations of stromal class and tumour subtype are significantly
over- and under-represented; furthermore, simultaneous stratification of tumors in external datasets by both tumour subtype and
stroma class identifies good- and poor-outcome cohorts within four of the five molecular breast cancer subtypes. The stroma
classes identified here form the basis for an improved breast cancer classification scheme which takes the contribution of the
microenvironment into account.

Title: Cell death and efferocytosis generate a pro-metastatic landscape during mammary gland involution that increase
dissemination of post-partum breast cancers
Rebecca S Cook1, Shelton Earp2 and Jamie Stanford1. 1Vanderbilt University, Nashville, TN and 2UNC Lineberger Comprehensive
Cancer Center, Chapel Hill, NC.
Body: Although pregnancy at a young age decreases a womans lifetime breast cancer risk, the first five years following a
pregnancy at any age are associated with increased breast cancer risk, a risk that continues to increase with age. Given societal
trends to postpone child-birth later into womens lives, the incidence of malignant post-partum breast cancer is expected to
increase. Currently, breast cancers diagnosed 2-5 years post-partum account for nearly 25% of all pre-menopausal breast
cancers. Importantly, breast cancers diagnosed 2-5 years post-partum are diagnosed more frequently as metastatic disease and
correlate with decreased disease-free survival (DFS) versus breast cancers occurring in other young pre-menopausal women. In
contrast, many studies show that breast cancers diagnosed during pregnancy do not correlate with decreased DFS, although
there remains some debate about this conclusion. The observation that post-partum breast cancers correlate with reduced DFS
suggests that events occurring in the post-partum breast augment the malignant severity of tumors existing therein. This
hypothesis is supported by studies in which breast cancer cells transplanted into involuting mouse mammary fatpads grow and
invade more rapidly than cells transplanted into mammary glands of nulliparous mice, demonstrating a unique mammary
micro-environment during involution versus other reproductive stages. We aimed to study how the changing landscape of the
post-partum breast accelerates cancer progression.
Although the molecular mechanisms underlying the malignant severity of post-partum breast cancers (ppBCs) are unclear, they
relate to stromal wound healing events during post-partum involution, a dynamic process characterized by widespread cell death
in milk-producing mammary epithelial cells (MECs). Using both spontaneous and allografted mammary tumors in fully
immune-competent mice, we discovered that post-partum involution increased mammary tumor metastasis 10-fold. Widespread
cell death occurred not only in MECs, but also in tumor epithelium. Dying tumor cells were cleared through MerTK-dependent
efferocytosis, which robustly induced transcription of wound healing cytokines interleukins (IL)-4, -10, and -13, and transforming
growth factor (TGF)-beta. Genetic models of MerTK ablation and pharmacologic MerTK inhibition impaired efferocytosis in
post-partum tumors, reduced M2-like macrophages without altering total macrophage levels, decreased TGF-beta expression and
reduced post-partum tumor metastasis to levels seen in nulliparous mice. TGF-beta blockade reduced post-partum tumor
metastasis. These data suggest that widespread cell death during post-partum involution triggers efferocytosis-induced wound
healing cytokines in the tumor microenvironment that promote metastatic tumor progression.

Title: No Show

Title: Estrogen signaling through astrocytes promotes migration and invasion of ER-negative brain metastatic breast cancer cells
Diana M Cittelly1, Hazel Cruz1, Natalie J Serkova3, Borges F Virginia2, Kabos Peter2, Steeg S Patricia4 and Sartorius A Carol1.
1
University of Colorado Anschutz Medical Campus, Aurora, CO; 2University of Colorado Anschutz Medical Campus, Aurora, CO;
3
University of Colorado Anschutz Medical Campus, Aurora, CO and 4National Institute of Health, Women's Malignancies Branch,
Bethesda, MD.
Body: Approximately 16% of patients with breast cancer develop symptomatic brain metastases (BMs) and the majority [80%]
will die subsequently. Symptomatic BMs are more prevalent in breast tumors overexpressing Human Epidermal Growth Factor
Receptors (HER1, EGFR) or 2 (HER2) and triple-negative (TN) tumors (lacking estrogen receptor (ER), progesterone receptor
(PR) and HER2). Although the majority of breast cancer cells colonizing the brain lack ER, central nervous system (CNS)
metastases are more frequent in pre-menopausal women with high levels of circulating estrogens. Since astrocytes are ER+ cells
that surround and infiltrate brain metastases, we hypothesize that estrogens act in a paracrine manner on reactive astrocytes to
promote BMs.
Results: To determine the effect of estrogen in brain colonization, a brain-seeking sub-line of human MDA-MB-231 TN breast
carcinoma cells (231Br) cells were injected intracardially in ovariectomized female nude mice supplemented with placebo (n=5) or
1mg 17-estradiol (E2) pellets (n=5), and metastases were detected by gadolinium-enhanced magnetic resonance imaging (MRI)
5 weeks later. 100% (5/5) of mice from E2 group showed large and multiple BM when compared to 3/5 (60%) of placebo mice.
Further, E2-treated mice showed decreased survival (41.6%, n=12) by 3 weeks following IC injection of 231Br cells compared to
E2-depleted mice (84%, n=12) (P=0.03). Since astrocytes are the most abundant brain cell type expressing ERs, we focused on
their interaction with 231BR cells. Using global gene expression arrays and RT-PCR we found that E2 upregulated Egf and Tgf-a
in primary mouse astrocytes and human astrocytes (derived from neural stem cells), and that anti-estrogens blocked this effect.
Concentrated conditioned media (CM) from E2-treated astrocytes increased P-EGFR levels in 231Br cells compared to vehicle
(EtOH), or E2+ICI astrocytic CM. Moreover, CM from E2-treated astrocytes significantly increased migration and invasion of
231Br cells as compared to EtOH-treated CM, and ERs inhibitors abolish this effect. Treatment of 231Br cells with lapatinib
abolished EGFR activation in response to E2-astrocytic CM, and decreased migration and invasion of 231BR cells. These data
suggest that EGFR-ligands upregulated by E2 in astrocytes activated brain metastatic cells-EGFR resulting in increased
migration and invasion. To identify the mechanisms of increased migration and invasion we performed global gene expression
profiling of 231Br cells co-cultured with E2- or EtOH-stimulated-astrocytes. Co-culture with E2-treated astrocytes significantly
increased expression of metastatic mediators Matrix-metalloproteinase-9 (MMP9) and S100 Calcium-binding protein A4 (S100A4)
in 231Br cells, and treatment of astrocytes with ER inhibitors abolished this effect. Furthermore, MMP9 or S100A4 knockdown in
231Br cells significantly decreased migration and/or invasion in response to E2-astrocytic CM, suggesting that MMP9 and
S100A4 are functional mediators of the paracrine effect of E2. Conclusion: These studies suggest that EGFR activation by
astrocytic ligands is at least one mechanism by which E2 contributes to the promotion of BM of ER- breast cancer cells and
suggests a novel role for ER in breast cancer BM.

Title: Contributions of STAT3 to pro-tumorigenic alterations within the microenvironment during mammary tumorigenesis
Pavlina Chuntova1, Laura R Bohrer1 and Kaylee Schwertfeger1,2. 1University of Minnesota, Minneapolis, MN and 2Masonic Cancer
Center, University of Minnesota, Minneapolis, MN.
Body: Interactions between tumor cells and their surrounding microenvironment are critical for the development and progression
of breast cancer. Specifically, oncogenic alterations such as aberrant activation of growth factor receptors in epithelial cells lead
to a pro-tumorigenic inflammatory response characterized by increased production of inflammatory factors and increased
macrophage infiltration. Our studies focus on identifying novel mechanisms through which tumor-derived cytokines act in an
autocrine manner to regulate tumor cells and in a paracrine manner to regulate infiltrating inflammatory cells. In recent studies,
we have shown that following activation of the fibroblast growth factor receptor 1 (FGFR1) oncogene, mammary epithelial cells
produce high levels of interleukin 6 (IL-6) family members. These cytokines activate signal transducer and activator of
transcription 3 (STAT3) in both epithelial cells and macrophages. STAT3 is constitutively activated in up to 70% of primary breast
tumors, and has been shown to enhance proliferation and angiogenesis as well as aid in immune suppression. Using a novel
mouse model of FGFR1-driven tumor growth, we demonstrated that inhibition of STAT3 led to decreased tumor growth.
Histological examination revealed high levels of STAT3 activation in both tumor cells and macrophages. Thus, further studies
focused on identifying mechanisms through which STAT3 activation in different cellular compartments contributes to mammary
tumor growth. Initial studies focused on the consequences of STAT3 activation in epithelial/tumor cells. We found that epithelial
cell STAT3 activation resulted in accumulation of the extracellular matrix component hyaluronan, which contributed to epithelial
cell proliferation, survival and migration. Meanwhile, STAT3 activation in macrophages led to expression changes within a variety
of molecules which are known to enhance tumor cell survival, tumor invasion and regulation of the anti-tumor immune response.
Thus, it becomes evident that systemic inhibition of STAT3 function can potentially inhibit tumor growth and progression by acting
on both tumor cells and on the surrounding stroma. As STAT3 is being considered as a viable target for breast cancer therapy,
understanding the mechanisms through which STAT3 promotes tumor growth and progression will be critical for developing
optimal therapeutic strategies.

Title: CD146 positive and negative stroma direct breast tumor estrogen receptor levels, therapeutic response and metastatic
potential
Heather M Brechbuhl1, Jessica J Finlay-Schultz1, Tomomi M Yamamoto1, Austin E Gillen1, Anthony Elias1, Carol A Sartorius1 and
Peter Kabos1. 1University of Colorado Denver Anschutz Medical Campus, Aurora, CO.
Body: Background: Cellular heterogeneity within all breast cancer subtypes remains a major cause of treatment failure and
development of metastatic disease. Currently, both preclinical studies and drug development efforts focus almost exclusively on
the epithelial component of breast cancers. Despite strong preclinical data novel therapies often fail in clinical testing. We propose
that this failure is, in part, due to taking tumor cells out of their context and using pre-clinical models that fail to capture the
complexity of human disease. We hypothesize that tumors hijack normal components of the tissue microenvironment and use it to
their advantage. Here we demonstrate that similar to the normal hematopoetic niche, two major subtypes of breast cancer stroma
can be defined by CD146 expression. We further show that the ratio of the stromal subtypes alters the response to therapy and
increases the metastatic potential of breast cancer cells (BCC).
Results: Tumor associated stroma, from all breast cancer subtypes, contains a mixture of CD146+ and CD146- fibroblasts. We
isolated and derived pure human CD146+ and CD146- clonal lines. Both subtypes expressed markers of activated fibroblasts and
clustered by gene expression profiling with normal stromal cell lines HS27A (CD146+) or HS5 (CD146-) according to CD146
expression. Although both stromal subtypes were derived from tumor associated tissue, CD146+ breast cancer stroma clustered
with normal breast associated stroma and correlated with good clinical outcome (Finak et. al. Nature Med 2008). CD146- stroma
clustered with breast cancer associated stroma and predicted worse outcome.
Using cell line and patient-derived xenograft models of estrogen receptor (ER) positive breast cancer, we demonstrated that
CD146+ compared to CD146- stroma supported significantly higher ER expression in BCCs. BCC co-cultured with CD146+
stroma responded more robustly to estrogen treatment and anti-endocrine therapy with tamoxifen.
Next we used expression profiling data to predict stromal influence on treatment response. CD146+ stroma expressed 3-fold
more TGF than CD146- stroma. Inhibiting TGF decreased proliferation 2-fold in BCCs grown in media conditioned by CD146+
stroma, but not by CD146- stroma. Conversely, HBEGF expression was 3-fold higher in CD146- stroma compared to CD146+
stroma. Inhibiting EGFR decreased proliferation 1.5-fold in BCCs grown in conditioned by CD146- stroma, but not by CD146+
stroma. In addition, intracardiac injection of stroma resulted in distant metastases in our primary orthotopic PDX model of breast
cancer.
Lastly we confirmed our preclinical observation using a small set of clinical samples. Patients with high CD146+ to CD146stromal ratio had better clinical outcomes than patients with high CD146- to CD146+ stromal ratio.
Conclusion: We conclude that stromal subtypes defined by CD146 expression direct the heterogeneity of ER expression,
response to therapy and the metastatic potential of breast cancer cells.

Title: Breast cancer tissue context determines whether inflammation with bacterial/psoriasin signature is pro- or
anti-carcinogenesis
Helong Zhao1, Mohd W Nasser1, Tasha Wilkie1 and Ramesh K Ganju1. 1Ohio State University Wexner Medical Center, Columbus,
OH.
Body: Objective: It is often neglected that breast tissue microbiota and systematic bacterial infection actively and potently
influence the development of breast cancer. With reports of both pro-cancer and anti-cancer roles of bacterial inflammation, the
actual effect of bacterial inflammation to breast cancer and its mechanism remain unclear and are of our interests.
Rationale: Different from other human tissues, breast tissue microbiota is majorly Gram-negative at both normal and pathological
statuses. We discovered that bacterial inflammation in breast tissue has a signature of Psoriasin (S100A7) impact which
resembles psoriasis of the skin. However, this Psoriasin mediated inflammation can either promote or inhibit breast cancer
development. We hypothesize that different tissue context determines the differential effects of bacterial/Psoriasin inflammation
on breast cancer development.
Results: We first verified the presence of commensal Gram-negative microbiota (and LPS) in breast tissues under normal and
pathological circumstances using mouse models. We observed that in both normal and cancerous breast tissues, bacterial factors
(such as LPS) found in breast triggered secretion of Psoriasin by mammary adenocytes and Psoriasin mediated inflammation.
This type of inflammation is featured by macrophage recruitment and ductal infiltration induced by Psoriasin. However, in normal
tissues, macrophages matured into M1 type (iNOS positive); whereas in cancerous tissues, macrophages matured into M2 type
(Arginase-1 positive). In normal tissues, Psoriasin and M1 caused an unresolved accumulation of inflammation in mammary ducts
without lactation, which might lead to accelerated cancer initiation. In tumors, Psoriasin and M2 increased tumor growth and
metastasis in immune-competent mammary epithelial specific expression mouse models. This was further confirmed in human
patient samples and large cohort bioinformatic analysis. Using human breast cancer cell lines and nude mice, we elucidated that
inflammation related Psoriasin upregulation and secretion in cancer cells had differential effects on tumor growth depending on its
tissue source. Psoriasin increased cell proliferation and tumor growth in basal-like cancer cells; whereas it decreased cell
proliferation and tumor growth in non-basal-like cancer cells. And this difference was mediated through differential regulation of
the NF-B/miR-29b/p53 pathway. in vitro molecular study and in vivo models verified the important role of miR-29b switch in the
differential effects of Psoriasin in different tissue types of breast cancer cells.
Conclusion: Bacterial/Psoriasin inflammation differentially influences breast cancer development depending on the tissue context,
and it involves the effects on macrophages and cancer cells. Our work provides novel potentials to improving prognosis and
targeting inflammation to treat breast cancer in personalized medicine.

Title: EphA4-deleted microenvironment regulates cancer development of isografted 4T1 murine breast cancer cells via reduction
of an IGF1 signal
Xuefeng Jing1, Takashi Sonoki1, Masayasu Miyajima1, Takahiro Sawada1, Nanako Terada1, Kenryo Furushima1, Daiki Arai1,
Kazuki Kawakami1 and Kazushige Sakaguchi1. 1Wakayama Medical University, Wakayama, Japan.
Body: Background:
EphA4 (the erythropoietin-producing hepatocellular receptor A4) belongs to a large family of receptor tyrosine kinases that play
critical roles in cancer progression. We have previously reported that the absence of EphA4 expression decreases the amount of
IGF1 in the circulation and locally in tissues, which contributes to the short stature (Cell Reports 2012). It is known that 4T1
murine breast cancer is a good model of human metastatic breast cancer. 4T1 cells are also known to produce a large amount of
granulocyte colony-stimulating factor which can cause extramedullary hematopoiesis associated with a poor recipient prognosis.
Aim:
To investigate whether EphA4-deleted microenvironment affects growth of primary tumors, tumor metastasis, and extramedullary
hematopoiesis via a novel EphA4-mediated IGF1 synthesis pathway.
Methods:
We isografted 105 mouse breast cancer cells (4T1) into the left inguinal mammary fatpad of both EphA4-knockout (KO) and
control wild type (WT) female littermate mice. The parameters evaluated in vivo were growth of the primary tumors, distribution of
metastatic foci, number of peripheral blood leukocytes (PBL), and splenomegaly. To examine the effect of IGF1 treatment on
these parameters, recombinant human IGF1 (5 mg/kg body weight (BW)/day) was subcutaneously injected into the EphA4-KO
mice for 9 weeks starting 4 weeks before grafting 4T1 cells. And the corresponding WT control mice were treated with saline
alone for the same period. We evaluated the extent of metastasis by counting the number of metastatic foci larger than 1 mm in
diameter in the lung, heart, liver, kidney, adrenal glands, lymph nodes, peritoneum, pleura and ovary.
Results:
In the absence of IGF1 injection, both the weight of primary tumors and the number of metastatic tumor foci were significantly
reduced in EphA4-KO mice as compared with those in control WT littermate mice (n=6 of each genotype, t-test P<0.005).
Splenomegaly (n=6 of each genotype, t-test P<0.001) and the number of PBL (n=6 of each genotype, t-test P<0.005) were also
markedly decreased in EphA4-KO mice. When treated with IGF1, EphA4-KO mice showed significant weight gain of the primary
tumors to almost the level of WT mice, had a greatly increased number of metastatic tumors, and showed an enhanced PBL
number to the level of WT mice. However, IGF1 treatment could not enhance splenomegaly.
Conclusions:
EphA4 expression in the tumor microenvironment plays an important role in tumor growth, metastasis, leukemoid reaction and
splenomegaly. Deletion of EphA4 in the microenvironment delays primary tumor growth and metastasis and reduces leukemoid
reaction mainly by regulating the amount of IGF1 in the circulation and tissues. However, splenomegaly might not solely be
mediated by an IGF1 signal. Our findings may prove a new therapeutic target for breast cancer.

Title: Mesenchymal stem cells and macrophage interactions promote inflammatory breast cancer cell invasion and self renewal
Adam R Wolfe1, Rachel Atkinson1, Bisrat G Debeb1, Yan Zhang1, Brian Ruffel2, James M Reuben1, Naoto T Ueno1 and Wendy A
Woodward1. 1MD Anderson Cancer Center, Houston, TX and 2Ohio State Univeristy.
Body: Background: Inflammatory breast cancer (IBC) is responsible for 10% of breast cancer deaths. The hallmarks of IBC are
skin involvement and a high propensity to metastasize. Our lab has shown previously, "normal" breast tissue from women with an
IBC diagnosis had significantly greater macrophage infiltration and increased cells with stem cell markers compared to non IBC
"normal" breast tissue. These changes were present prior to diagnosis in two patients where pre-IBC biopsies were available.
Therefore, we hypothesized changes in the normal breast microenvironment prior to tumor formation contributes to the IBC
phenotype.
Methods: To study our hypothesis we used a co-culture system to measure the interactions between normal macrophages (Raw
264.7 cell line), bone-marrow derived mesenchymal stem cells (MSCs), and IBC cells (SUM 149, IBC3, and KPL4). Conditioned
media (CM) from MSC culture was added to macrophages overnight. The macrophages were subsequently analyzed for their
surface markers and cytokine production. Reciprocally, MSCs were "educated" by macrophages by adding CM from polarized M2
macrophages to the MSC culture media at a 1:1 ratio. MSCs and cancer cells were co-cultured in trans-wells (Boyden chambers)
for 24 hours. Migration and invasion in vitro was determined by adding MSCs or IBC cells to the insert of the trans-well and
cultured in combination with either polarized macrophages or macrophage educated MSCs for 24 hours. After co-culture, IBC
cells were analyzed for their ability to form mammospheres.
Results: Mouse macrophages polarized (using IL-4) into a M2 phenotype (Tumor associated macrophage) secreted 3 fold more
IL-6 compared to unstimulated or M1 polarized macrophages. The addition of MSC CM to the macrophage culture for 24 hours
polarized the macrophages into a similar M2 phenotype described above with 4-fold increase in IL-6 secretion compared to
unstimulated macrophages (P<0.01). When M2 macrophages at the bottom of the co-culture system were co-cultured with MSC,
the number of MSCs migrating increased 2 fold with the addition of M2 macrophages compared to media alone, or unstimulated
macrophages (P<0.05). IBC cells showed a 2 fold enhancement of invasion towards M2 educated MSCs compared to either
uneducated MSCs, or media alone (P<0.05). IBC cells co cultured with M2 educated MSCs grew 3 fold more mammospheres
compared to IBC cells grown with uneducated MSCs (P<0.01). IBC cells added to the bottom of the co-culture system also
enhanced migration of MSCs and did so to a greater extent than non-IBC cells (P<0.01). Lastly, the addition of IL-6 neutralizing
antibody inhibited IBC induced MSC migration.
Conclusions: Herein we demonstrate reciprocal tumor interactions between normal cells in the IBC microenvironment. MSC and
macrophages can influence each other to increase the tumor promoting influence of each on IBC cells. Our results suggest IL-6 a
mediator of these tumor promoting influences and is important for the IBC induced migration of MSCs. Currently we are
investigating the in vivo interactions between macrophages and MSCs in an orthotopic IBC mouse model.

Title: No Show

Title: Comparing tissue compositions of within-individual mammographically high and low dense breast tissue
Cecilia W Huo1,2, Dexing Huang7, Grace L Chew2, Prue Hill3, Wendy V Ingman4, Michael Henderson2, Kara L Britt5 and Rik
Thompson1,2,6,7. 1University of Melbourne Medical school; 2St Vincent's Hospital; 3St Vincent's Hospital; 4School of Medicine,
University of Adelaide; 5Peter MacCallum Cancer Centre; 6Institute of Health and BioMedical Innovation, Queensland Institute of
Technology and 7St Vincent's Institute, Australia.
Body: Abstract
Purpose
Mammographic density (MD) refers to the extent of radio-opaque breast tissue on a mammogram. Adjusted for a womens age
and body mass index (BMI), it is a strong and independent risk factor for breast cancer (BC). The presentation of
mammographically dense breast is not uncommon in the normal female population, and patients awareness of its association
with BC risk has been growing following mandatory reporting of MD by physicians in several states of America. Women with
breasts that have over 75% dense tissue are 4 to 6 times more likely to develop BC than women with breasts that have less than
10% dense tissue. However, the biological basis of how high MD raises BC risk remains elusive. We aimed to examine the
histological and molecular differences between high and low dense breast tissues of healthy women, using specimens accrued
from prophylactic mastectomy procedures.
Method
48 women between 2008 and 2013 underwent prophylactic mastectomy at St Vincents Hospital and Peter MacCallum Cancer
Centre due to a high BC risk profile. Of these, 41 were eligible for analyses. Tissue slice resected from the mastectomy specimen
was X-rayed, and high (HD) and low dense (LD) regions were dissected based on the radiological appearance. The histological
composition, immunohistochemistry and proliferation status were assessed on matched high and low MD tissue of the same
breast. Signed rank test and paired t test were used for quantitative analyses of potential differences between HD and LD tissue.
Result
HD tissue demonstrated a significantly greater proportion of stroma (p<0.0001) and epithelium (p<0.0001), and less amount of fat
(p<0.0001) than LD tissue (n=41 women). Epithelium from HD region also demonstrated epithelial-mesenchymal transition (EMT)
plasticity, which was evident as the co-expression of cytokeratin (CK)-19 and vimentin in the glandular area. There was no
significant difference with regards to oestrogen receptor (ER) (p= 0.2772), progesterone receptor (PR) (p= 0.9910), and Ki-67 (p=
0.6028) expression between HD and LD tissue.
Conclusion and Significance
We found that increased stroma and epithelium proportions contribute to the dense appearance on mammogram. Moreover,
dense tissue did not demonstrate differed hormonal receptor expression or proliferation status from non-dense tissue, but showed
a preponderance of EMT in the form of co-localisation by both CK-19 and vimentin in some of the epithelial cells. Our study is the
first to report EMT phenomenon in benign mammary tissue, and suggests that investigations of the stromal micro-environment,
and their interactions with epithelium may be key to improving our understanding on MD-mediated BC risk.

Title: Both metabolic syndrome and statin use are more common in women with breast inflammation
Neil M Iyengar1, Ayca Gucalp1, Xi K Zhou2, Louise R Howe2, Patrick G Morris1, Dilip Giri1, Kotha Subbaramaiah2, Priya Bhardwaj2,
Samuel S Park1, Michael Pollak3, Monica Morrow1, Clifford A Hudis1 and Andrew J Dannenberg2. 1Memorial Sloan Kettering
Cancer Center, New York, NY; 2Weill Cornell Medical College, New York, NY and 3McGill University, Montreal, QC, Canada.
Body: Background: Metabolic syndrome is associated with increased breast cancer (BC) risk. After menopause, obesity is
associated with increased risk of hormone receptor (HR)-positive breast cancer. We demonstrated an obesity inflammation
aromatase axis in breast white adipose tissue (WAT) where inflammation is defined by the presence of crown-like structures
(CLS) consisting of a dead or dying adipocyte encircled by macrophages. CLS in the breast (CLS-B) are associated with elevated
body mass index (BMI), postmenopausal status, increased adipocyte size, and increased tissue levels of proinflammatory
mediators and aromatase. As obesity is a component of the metabolic syndrome and is associated with inflammation, we
compared levels of relevant circulating factors in women with and without breast WAT inflammation.
Methods: We prospectively collected paired WAT and fasting serum and plasma from women undergoing mastectomy at
MSKCC. WAT inflammation was detected by CD68 immunohistochemistry to identify CLS-B by light microscopy. Plasma levels of
glucose, insulin, hsCRP, leptin, adiponectin, and IL-6 were measured by ELISA. Serum levels of total cholesterol, triglycerides
(TG), HDL and LDL cholesterol were determined. Insulin resistance (IR) was assessed using fasting plasma glucose and insulin
levels via the updated Homeostasis Model Assessment (HOMA2-IR). Associations between CLS-B and clinicopathologic
features, including medication usage, were analyzed by logistic regression and Fishers exact test. Differences in serum/plasma
endpoints between subjects with and without CLS-B were evaluated using Student t-test and nonparametric Wilcoxon rank-sum
test.
Results: From 11/2011 3/2013 we accrued 100 patients (pts); median age 47 years (range 27 70). Overall, CLS-B were
found in 52/100 (52%) pts: 18/19 (95%) obese pts, 17/33 (52%) overweight pts, and 17/48 (35%) normal BMI pts. A clinical
diagnosis of dyslipidemia was present in 14/52 (27%) pts with CLS-B and 1/48 (2%) pts without CLS-B (P<0.001). CLS-B were
found in 10/11 (91%) statin users, but in only 42/89 (47%) non-users (P=0.008). Fasting glucose, insulin, LDL, TG, leptin, hsCRP,
and IL-6 levels were higher in pts with CLS-B (Table). HOMA2-IR was greater in pts with CLS-B (mean 0.63 0.34) versus those
without CLS-B (mean 0.46 0.23; P=0.006).
Fasting level, mean (SD)
No CLS-B, N=48
CLS-B +, N=52
Glucose, mg/dL
73.2 (8.0)
84.3 (37.6)
0.04
Insulin, mU/L
4.3 (2.1)
5.6 (2.9)
0.01
LDL cholesterol, mg/dL
105.9 (31.0)
119.4 (32.4)
0.04
HDL cholesterol, mg/dL
71.9 (15.8)
62.1 (16.1)
0.003
Triglycerides, mg/dL
69.2 (26.3)
104.9 (50.6)
<0.001
Leptin, pg/mL
12.0 (10.1)
22.6 (19.7)
<0.001
Adiponectin, ng/mL
13.7 (5.2)
10.4 (5.4)
0.002
hsCRP, ng/mL
1.0 (1.4)
2.3 (2.7)
0.003
Conclusions: Breast WAT inflammation, which we have previously linked to increased aromatase activity, is associated with
biochemical changes that occur in the metabolic syndrome, a risk factor for BC. Statin use is more common in patients with
breast WAT inflammation and metabolic syndrome. Clinically, statin use may be a surrogate identifier of a population that is at
increased baseline risk of BC. These findings may account for the variability in results of prior studies examining statin use and
breast cancer risk due to elevated risk in users compared to non-users.

Title: Cancer-derived miR-105 promotes tumor invasion and destroys the natural barriers against metastasis
Shizhen Emily Wang1, Weiying Zhou1, Miranda Y Fong1, Yongfen Min2 and Pengnian Charles Lin2. 1City of Hope Beckman
Research Institute and Medical Center, Duarte, CA and 2National Cancer Institute, Frederick, MD.
Body: Cancer-secreted miRNAs are emerging mediators of cancerhost crosstalk. In this study, we set out to identify
cancer-derived miRNAs that participate in cancer metastasis by adapting the niche cells. We show that miR-105, which is
characteristically expressed and secreted by metastatic breast cancer cells, is a potent regulator of migration through targeting
the tight junction protein ZO-1. In epithelial and endothelial monolayers, exosome-mediated transfer of cancer-secreted miR-105
efficiently destroys tight junctions and the integrity of these natural barriers against metastasis. Overexpression of miR-105 in
non-metastatic cancer cells induces metastasis and vascular permeability in distant organs, whereas intervention of miR-105 in
highly metastatic tumors alleviates this effect. By testing clinical specimens from breast cancer patients, we further show that
miR-105 can be detected in the circulation at the pre-metastatic stage, and its levels in the blood and tumor are associated with
ZO-1 expression and metastatic progression in early-stage breast cancer. Our results demonstrate for the first time the dual roles
of miR-105 in inducing the migratory potential of cancer cells as well as destroying the epithelial and endothelial barriers in the
cancer niche by targeting the cellular tight junctions. In breast cancer patients, increased levels of miR-105 in the circulation can
be detected at the pre-metastatic stage and predict the occurrence of metastasis. Anti-miR-105 treatment suppresses metastasis
and abolishes the systemic effect of tumor-derived miR-105 on niche adaptation. Therefore, these pre-clinical observations
strongly suggest clinical applications of miR-105 as a predictive or early-diagnostic blood-borne marker as well as a therapeutic
target for breast cancer metastasis.

Title: High mammographic density is associated with deposition of organised fibrillar collagen and increased stiffness in
periductal breast stroma
Ashu Gandhi1, Cliona C Kirwan2, James C McConnell3, Oliver V O'Connell4, Michael J Sherratt3 and Charles H Streuli4.
1
Manchester Academic Health Sciences Centre, University Hospital of South Manchester, University of Manchester, Manchester,
Greater Manchester, United Kingdom; 2Institute of Cancer Sciences, University Hospital of South Manchester, University of
Manchester, Manchester, Greater Manchester, United Kingdom; 3Centre for Tissue Injury & Repair, Faculty of Medical and Health
Sciences, University of Manchester, Manchester, Greater Manchester, United Kingdom and 4University of Manchester, Faculty of
Life Sciences, Manchester, Greater Manchester, United Kingdom.
Body: Introduction
High mammographic density (MD) in women is strongly associated with breast cancer risk. However the structural and
compositional differences between dense and non-dense breast tissues are not well defined. We determined the relationship
between MD, collagen deposition and fibril alignment, and tissue micro-stiffness, in similarly aged individuals without adjacent
cancer.
Methods
Fresh tissue samples were collected from post-menopausal women undergoing breast screening. Collagen deposition and fibril
organisation were analysed using light microscopy of wax-sections stained with H&E, Trichrome or Picrosirius Red (with
polarising light), and quantified using ImageJ. Local tissue stiffness was measured using atomic force microscopy (AFM) of
hydrated tissue. For AFM, 3 x 25 m2 areas of each sample were indented at 400 equally spaced points with a 1 m diameter
spherical probe at a loading rate of 1 Hz (Reduced Modulus; YM).
Results
Volumetric MD (VolparaTM) was determined in 22 women (54-66y) undergoing risk-reducing surgery or mastectomy. Localised
regions of elevated density, determined from digital mammograms, were isolated from patients of low and high overall MD, using
a new collaborative workflow linking radiologist, surgeon, pathologist, and tissue biobank.
All elevated-density regions contained considerable amounts of stromal connective tissue. However, there were significant
differences in these regions from women with low vs high overall MD. Picrosirius Red staining of the localised areas of density
revealed that the percentage organised fibrillar collagen content, particularly in the periductal breast stroma, strongly correlated
with overall MD.
AFM showed that the localised micro-stiffness of dense areas increased significantly in the breast stroma of patients with high
overall MD (Volpara score > 15) compared with those of low overall MD (Volpara score < 5).
Conclusions
High MD is a significant risk factor for breast cancer, yet its molecular determinants in the normal, non-cancerous breast are
poorly defined. We have shown that high MD is associated with more organised fibrillar collagen, leading to increased stiffness of
the periductal breast stroma. Women with low and high MD all have regions with localised density, which contain both stromal
connective tissue and epithelial ducts/lobules. However, our results show that these localised areas have differences in collagen
organisation and tissue micro-mechanics. We now hypothesise that in the connective tissue of women with high MD, altered
synthesis, deposition and turnover of stromal proteins alters the local biomechanical properties within the breast, providing a
stiffer microenvironment, and contributing to cancer onset.

Title: Hierarchy of breast cancer associated fibroblasts communicate with cancer cells via microRNAs to drive breast cancer
progression
Sanket H Shah1, Phil Miller1, Zheng Ao1, Emilio Issa1, Katherine Drews Elger1, Ram Datar1, Marc Lippman1 and Dorraya
El-Ashry1. 1University of Miami, Miami, FL.
Body: Background: Increasing evidence has demonstrated that stromal cells play a pivotal role to promote breast cancer
progression and metastasis. Breast cancer stroma is comprised mainly of Cancer Associated Fibroblasts (CAFs). Upon
interaction with tumor cells, CAFs promote tumor progression by providing paracrine oncogenic signals mediated by activation of
various pathways including developmental pathways, integrin signaling, and the MAPK pathway in tumor cells. CAFs have also
been shown to promote the survival of CTCs and help them in metastasis at distant sites. Using breast cancer patient tumor
datasets, we have previously identified a microRNA signature reflective of hyperactive MAPK signaling and that is significantly
associated with reduced recurrence-free and overall survival. We have established 3 primary breast CAF lines, one from a
Luminal A breast cancer, one from an ER-/Her2 amplified cancer, and one from a triple negative cancer, along with several
primary tumor-derived dissociated tumor (DT) culture models that are tumorigenic in vivo and vary in metastatic ability. The CAFs
express several hMAPK-microRNAs preferentially compared to the DTs. In addition to the paracrine interaction of stromal and
tumor cells mediated by chemokines or hormones, miRNA cross talk between stromal and tumor cells can also occur.
Results: To further investigate the connection between our miRNA signature and stroma, we analyzed the TCGA and METABRIC
breast cancer datasets and found that the hMAPK-miRNA identifies tumors that with signficant stromal cell infiltrate. To
investigate the role of specific expression of hMAPK-miRNAs in the CAFs, CM was isolated from CAFs from "aggressive" tumors,
from the "indolent" tumor, and from normal human mammary fibroblasts (HMFs) and analyzed for exosome and microRNA
secretion. CAFs from the aggressive tumors secrete more exosomes and more hMAPK-microRNAs into the CM than do HMFs or
CAFs from the indolent tumor. Importantly, conditioned media (CM) from the "aggressive" CAFs activate MAPK and repress ER
protein, mRNA and ER 3UTR-reporter activity in ER+ MCF-7 breast cancer cells, while HMFs and "indolent" CAFs did not.
Exosomes from the "aggressive" CAFs were responsible for the ER repression. To determine if the secreted miRNA differences
exhibited by the CAFs could be seen in patients, serum from breast cancer patients with metastatic breast cancer and patients
without metastases was analyzed for microRNA expression. Differentially expressed circulating hMAPK-miRNAs were identified
in serum from metastatic breast cancer patients compared with patients without metastatses. Further analyses of the serum for
CTCs and CAFs show that serum samples from metastatic patients had a significantly higher number of CTCs with CAFs
compared to serum from patients without metastases.
Conclusions: Collectively these data suggest that different CAF populations have distinct abilities to influence the phenotype and
behavior of associated cancer cells and that CAF secreted hMAPK-miRNAs may play important roles in breast cancer
progression. They further suggest that these CAF secreted miRNAs can be found in patient serum along with circulating CAFs.

Title: Obesity-associated systemic interleukin-6 promotes pre-adipocyte aromatase expression via increased breast cancer cell
prostaglandin E2 production
Laura W Bowers1, Andrew J Brenner2, Stephen D Hursting3, Rajeshwar R Tekmal2 and Linda A deGraffenried1. 1University of
Texas, Austin, TX; 2University of Texas Health Science Center, San Antonio, TX and 3University of North Carolina, Chapel Hill,
NC.
Body: Introduction: Obesity is associated with a worse breast cancer prognosis, particularly in estrogen receptor alpha (ER)
positive, postmenopausal patients. Resistance to aromatase inhibitor treatment may be one mechanism mediating this effect, as
obese patients have a lower response rate to this class of drugs and higher breast tissue aromatase expression. We
hypothesized that obesity-associated circulating factors promote breast cancer cell cyclooxygenase-2 (COX-2) expression and
prostaglandin E2 (PGE2) production, resulting in an elevation in pre-adipocyte aromatase expression that enhances breast
cancer cell estrogen receptor activity and proliferation.
Methods: We utilized an in vitro model of the obese patients tumor microenvironment in which cultured MCF-7 breast cancer
cells and pre-adipocytes were exposed to pooled serum from obese (OB; BMI30.0 kg/m2) or normal weight (N; BMI 18.5-24.9
kg/m2) postmenopausal women.
Results: Exposure to OB versus N patient sera significantly increased MCF-7 cell COX-2 expression and PGE2 production. This
was coupled with 89% greater pre-adipocyte aromatase expression following culture in conditioned media (CM) from MCF-7 cells
exposed to OB versus N patient sera, a difference nullified by treatment of the MCF-7 cells with the COX-2 inhibitor celecoxib
during CM generation. Previous analysis of the sera revealed significantly higher interleukin 6 (IL-6) concentrations in the OB
versus N patient samples. Depletion of IL-6 from the sera resulted in neutralization of the difference between OB and N
CM-stimulated aromatase expression. N CM generated with the addition of IL-6 treatment at the time of N patient sera exposure
also induced pre-adipocyte aromatase expression that was statistically equivalent to the OB CM condition. Finally, CM from
pre-adipocyte/MCF-7 cell co-cultures exposed to OB patient sera stimulated greater MCF-7 and T47D breast cancer cell ER
activity and proliferation in comparison to N.
Conclusions: This study indicates that obesity-associated systemic IL-6 indirectly enhances pre-adipocyte aromatase expression
via increased breast cancer cell PGE2 production, thereby promoting greater cancer cell ER activity and proliferation.
Investigation regarding the efficacy of a COX-2 inhibitor/aromatase inhibitor combination therapy in the obese postmenopausal
patient population is warranted.

Title: Tumor-infiltrating neutrophils in breast cancer subtypes: A retrospective cohort study
Enrique Soto-Perez-de-Celis1, Mariana A Tenorio-Serralta1, Yanin Chavarri-Guerra1, Eucario Leon-Rodriguez1 and Armando
Gamboa-Domnguez1. 1Instituto Nacional de Ciencias Mdicas y Nutricin "Salvador Zubirn", Mexico City, DF, Mexico.
Body: Background: Tumor infiltrating neutrophils (TINs) are related to aggressiveness and poor prognosis in several human
cancers. Despite evidence pointing towards the pivotal role of inflammatory cells in the initiation, progression and response to
treatment of breast cancer (BC), the relevance of TINs in BC has not been systematically investigated. We sought to determine if
TINs were present in BC and to compare their presence among different subtypes. We hypothesized that the presence of TINs
would correlate with more aggressive histologic subtypes, particularly triple negative (TN) tumors.
Methods: This retrospective cohort study analyzed patients with BC treated at our institution from 2006 to 2012. Women with
treatment-nave stage I-III BC and with available medical records were included. Those with metastatic disease and/or lacking
material for pathologic review were excluded from analysis. Tumors were divided into three subtypes: luminal type (LT)
(hormone-receptor[HR]-positive, HER2-negative); HER2-positive and TN. Clinical and pathological characteristics were recorded.
The primary outcome was the amount of TINs, defined as neutrophils in direct contact with tumor cells, within each tumor.
Hematoxylin & eosin stained sections were examined by a dedicated breast pathologist blinded to tumor subtype, and the number
of TINs per 10 high power fields (HPF, 40x) was recorded. To evaluate the capacity of TIN measurements in predicting HER2+ or
TN tumors and to identify an optimal cut-off value for TIN positivity, ROC curve analysis was performed. Fishers exact test was
used to test for independence between qualitative variables, and logistic regression models were used for quantitative variables
and multivariate analysis.
Results: 133 patients were assessed for inclusion. 28 were excluded because of missing material, incorrect staging or erroneous
pathological diagnosis and 105 were analyzed. 72 tumors (69%) were classified as LT, 15 (14%) as HER2+ and 18 (17%) as TN.
The mean TIN count was 1.82 x 10 HPF (0-28). ROC analysis determined a cut-off value for positivity of >1 TIN x 10 HPF (AUC
0.85; 95% CI 0.76-0.93, p<0,001). Tumors with >1 TIN x 10 HPF were considered TIN-positive. 27% of all tumors were TIN+
(n=28). 16 of 18 TN (88%), 8 of 15 HER2+ (53%) and 4 of 68 LT tumors (5%) were TIN+ (p<0.001). 79% of HR-ve tumors (19 of
24) were TIN+, in contrast with 11% of HR+ve (9 of 81) (p<0.001). HER2 expression (p=0.023); tumor grade (p<0.001) and Ki67
(p<0.001) were also associated with TIN positivity. Age, menopausal status and T and N stage were not significant for the
presence of TINs. On multivariate analysis, only Ki67 (OR 1.05, 95% CI 1.01-1.1, p=0.008) and HR negativity (OR 10.6; 95% CI
2.5-45.8, p=0.002) were associated with a higher likelihood of TIN positivity.
Conclusions: Although TINs are uncommon in BC, they are present in most TN and in half of HER2+ tumors. In fact, the
absence of HR expression was the strongest predictor for TIN positivity. These results raise the question as to whether TINs, as
part of the tumor microenvironment, have a role in the aggressiveness and progression of TN tumors and warrant further
investigation of TINs in this BC subtype, particularly in relation with response to treatment and prognosis.


Title: Recruitment of tumor associated macrophages in patients with breast cancer is not host dependent but tumor grade
dependent
Ann Smeets1, Giuseppe Floris1, Kathleen Lambein2, Sigrid Hatse1, Annouschka Laenen3, Ines Nevelsteen1, Hans Wildiers1 and
Marie-Rose Christiaens1. 1University Hospitals, Leuven, Belgium; 2Ghent Universty Hospital, Ghent, Belgium and 3KU Leuven,
Leuven, Belgium.
Body: Background:
It has been shown that tumor related inflammation plays a crucial role in breast cancer progression. However, it remains unclear
whether the density of immune cells in the tumor micro-environment is determined by tumour or by host characteristics. To
answer this question, we measured tumor associated macrophage (TAM) density in 52 patients with a synchronous bilateral
breast cancer with different lymph node status on both sites.
TAMs were identified by CD68 and CD163 antibodies. Tumour stromal macrophages were counted using a semi-quantitative
method.
Grade
Description
no TSM
scanty TSM
II
small foci of TSM
III
large foci of TSM
IV
diffuse dense infilatration of TSM
TSM = tumor stromal macrophages

Macrophage count was scored by two independent pathologists who were blinded from the clinicopathological data during the
assessment.
To evaluate the relative impact on TAM density exerted by the tumor itself and by the host, we used the likelihood ratio tests for
the variance components related to host and tumor in determination of CD68 and CD163 levels.
Results:
In 60% of the tumors, the scores for CD68 and CD163 of both pathologists were identical resulting in a moderate concordance
according to the weighted kappa coefficient te st (0.534; CI 0.418-0.649, p<0.0001).
The TAM density strongly correlated with the number of mitosis and secondly with tumor grade. The results for CD68 and CD163
were similar. The association of clinicopathological variables with CD163 are shown.
Association of clinicopathological variables with CD163
Variable
Score (No(%))
0
II
p-value
III
IV
Histological subtype
0.0883
IDC
148
3 (2%)
64 (43%)
58 (39%)
20 (14%)
3 (2%)
ILC
31
0 (0%)
21 (68%)
7 (23%)
3 (10%)
0 (0%)
Mixed
0 (0%)
0 (0%)
7 (88%
1 (13%)
0 (0%)
Mitosis
0-10
0.0005
143
2 (1%)
76 (53%)
50 (35%)
15 (10%)
0 (0%)
10-20
19
1 (5%)
9 (47%)
7 (37%)
1 (5%)
1 (5%)
>20
25
0 (0%)
0 (0%)
15 (60%)
8 (32%)
2 (8%)
Grade
0.0226
45
0 (0%)
22 (49%)
19 (42%)
4 (9%)
0 (0%)
II
98
2 (2%)
53 (54%)
31 (32%)
12 (12%)
0 (0%)
III
44
1 (2%)
10 (23%)
22 (50%)
8 (18%)
3 (7%)
pT
0.2312
93
1 (1%)
49 (53%)
32 (34%)
10 (11%)
1 (1%)
71
1 (1%)
27 (38%)
30 (42%)
11 (15%)
2 (3%)
21
1 (5%)
9 (43%)
10 (48%)
1 (5%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
2 (100%)
0 (0%)
pN
0.1376
92
2 (2%)
44 (48%)
39 (42%)
6 (7%)
1 (1%)
95
1 (1%)
41 (43%)
33 (35%)
18 (19%)
2 (2%)
LVI
0.8007
No
162
3 (2%)
73 (45%)
64 (40%)
19 (12%)
3 (2%)
Yes
25
0 (0%)
12 (48%)
8 (32%)
5 (20%)
0 (0%)
ER
Negative
0 (0%)
0 (0%)
1 (25%)
3 (75%)
0 (0%)
Positive
179
3 (2%)
83 (46%)
70 (39%)
20 (11%)
3 (2%)
PR
0.4849
Negative
25
0 (0%)
10 (40%)
10 (40%)
5 (20%)
0 (0%)
Positive
156
3 (2%)
73 (47%)
60 (38%)
17 (11%)
3 (2%)
Her-2
0.0903
Negative
166
3 (2%)
79 (47%)
62 (37%)
20 (12%)
3 (2%)
Positive
0 (0%)
0 (0%)
6 (75%)
2 (25%)
0 (0%)
Focality
0.8170
Unifocal
1141
1 (1%)
66 (47%)
55 (39%)
16 (11%)
3 (2%)
Multifocal
46
2 (4%)
19 (41%)
17 (37%)
8 (17%)
0 (0%)
Age at diagnosis
0.8936
<55
60
1 (2%)
18 (31%)
35 (57%)
6 (10%)
0 (0%)
55-65
56
2 (4%)
32 (57%)
17 (30%)
4 (7%)
1 (2%)
>65
77
0 (0%)
34 (44%)
29 (38%)
14 (18%)
0 (0%)
For both CD 68 and CD163, the tumor introduces important clustering or correlation, whereas the host does not introduce any
additionale correlation. TAM recruitment is therefore primarily determined by tumor-related characteristics.

Title: Tumor infiltrating lymphocytes in inflammatory breast cancer
Cecile Colpaert1, Melike Marsan2, Peter Vermeulen1, Luc Dirix2, Steven Van Laere2 and Inflammatory Breast Cancer International
Consortium3. 1Oncology Centre, GZA Hospitals, Iridium Cancer Net, Antwerp, Belgium; 2Translational Cancer Research Unit,
Oncology Centre, GZA Hospitals, Iridium Cancer Net, Antwerp, Belgium and 3Inflammatory Breast Cancer International
Consortium.
Body: Inflammatory breast cancer (IBC) is a rare and aggressive form of breast cancer with poor prognosis. Multimodal treatment
including neo-adjuvant chemotherapy is the treatment of choice for IBC patients, with pathological complete response (pCR)
being one of the most important prognostic factors.
Tumor infiltrating lymphocytes (TILs) are an independent predictor of response to neo-adjuvant chemotherapy in breast cancer
(Denkert C. et al., 2010). In a search for multigene predictors of pCR in IBC using DNA micro-arrays, a 107 gene signature was
found that distinguished between responders and non-responders. This signature was enriched for immunity-related genes
showing that in IBC, as in non-IBC, response to neo-adjuvant chemotherapy is associated with immunity related processes
(Bertucci F. et al., 2014).
Standard H&E stained sections of formalin-fixed paraffin-embedded pre-treatment tumor tissue of 77 IBC patients were used to
evaluate the presence of TILs according to the Denkert method: mononuclear cell infiltrates were considered to be intratumoral
TILs when making direct contact with the tumor cells, while stromal TILs are mononuclear cells being present in the tumor stroma
without making direct contact with tumor cells.
Intratumoral TILs were observed in 10,4% (8/77) of the patients. Stromal TILs (strTILs ) were present in all tumors: in 35% (27/77)
strTILs occupied less than 5% of the tumor stroma, in 62 % (48/77) strTILs occupied between 5 and 50% of the tumor stroma and
2,6% (2/77) were "lymphocyte predominant breast cancers (LPBC)" with strTILS occupying more than 50% of the tumor stroma.
One LPBC was HER2 positive, the other one was triple negative. There was a significant association of strTILs with response to
neo-adjuvant chemotherapy: pCR was obtained in 13% of the patients with strTILs <5%, in 20% of the patients with strTILs
between 5 and 50% and in both patients with LPBC (Chi 2 p=0,018).
Tertiary lymphoid structures - lymphoid follicles with germinal centres, known to be associated with better clinical outcomes
(Gu-Trantien C. et al., 2013) - were detected in 2.6% of the patients (2/77); both patients had a disease free survival of more than
3 years.
This study shows that IBC tumors do not contain more TILs than non-IBC tumors and that, as in non-IBC, there is a positive
association of stromal TILs with pCR in IBC. Further validation of these results and further study of lymphocyte subsets in IBC is
warranted.

Title: Host response to microbes found in the tumor microenvironment and healthy adjacent tissue
Andrew W Chung1, Marian Navarrete1, Peter A Sieling1 and Delphine J Lee1. 1Dirks/Dougherty Laboratory for Cancer Research,
John Wayne Cancer Institute at Providence Saint John's Health Center, Santa Monica, CA.
Body: Next generation sequencing methods have defined the bacterial microbiome in human breast milk and breast tissue, but
the role of bacteria or their components in breast cancer has not been well defined. There is growing evidence in humans and
animals that alterations of the gut microbiome may contribute to host susceptibility to cancer. However, the presence of bacteria
at the site of the tumor microenvironment may also contribute to carcinogenesis (e.g. by promoting chronic inflammation) or
immune surveillance (e.g. by stimulating antitumor pathways). Previously, we found that bacterial load in breast cancer tissue is
inversely correlated to the progression of disease and that antimicrobial gene expression was comparatively higher in breast
tissue from healthy subjects. Our investigation of the microbiome of breast tissue showed that overall microbiome signatures in
breast cancer patients are similar when comparing their healthy adjacent tissue vs. the tumor tissue. However, our published
preliminary studies did identify some differences in the abundance of specific bacteria between the two tissues. Here we
investigated how those bacteria in healthy adjacent vs. tumor tissue may differentially stimulate cells found in breast tissue
including breast ductal epithelium, adipocytes, and immune cells. Our preliminary data suggest that bacteria found in healthy
adjacent vs. tumor tissue stimulate differential cytokine responses that may shape the local tumor microenvironment and beyond
and may also contribute to local/regional immune surveillance.


Title: Epithelial and stromal components are equivalently affected in breast fibroepithelial progression. A gene expression
analysis of phyllodes and fibroadenomas tumors
Angela F Logullo1, Mabel Planila3, Elisa Napolitano2, Fernando A Soares2 and Dirce Carraro2. 1Universidade Federal de So
Paulo, So Paulo, Brazil; 2A.C.Camargo Cancer Center, So Paulo, Brazil and 3Universidad de Concepcin, Chile, Conception,
Chile.
Body: Rationale-Phyllodes tumours (PTs) account for 0,5 to 1% of all breast lesions and, analogous to fibroadenomas (FA), are
considered true biphasic neoplasms. However, unlike in FA, their stromal element is more cellular and may outgrow the
epithelium, and PT may recur locally and can undergo malignant progression to sarcoma and metastasize. Recent evidence
emphasize that epithelium and stroma interactions are critical for tumour development and progression, but molecular alterations
in both, epithelial and stromal cells, in the arising of PT onset is not well characterized. Thus, in order to contribute for the
identification of the molecular events involved in PT tumorigenesis, we performed gene expression analysis of both types of cells
captured from PT and FA.
Methods- Epithelial (EpC) and stromal cells (StC) from 13 PT and 3 FAs were separately submitted to laser capture
microdissection (LCM) by CapSure HS Arcturus. Procedures were performed in 10 min to prevent RNA degradation. RNA
was extracted using the Pico Pure RNA Isolation Kit Arcturus and was twice amplified the first as described in Castro et al.,
2008 with adaptation and the second using low input Quick-Amp Labeling kit* and hybridized to the microarray slide SurePrint G3
Hmm 60K*. Signals were captured by Scanner Agilent Bundle model B* and Agilent Feature Extraction (v 10.7.1) program*and
data was analyzed by genespring GX12.1 software* (*Agilent Technologies,USA). A Fold Change 2,0 and p0,05 were
considered significant.
RESULTS: Flag quality filter determined 8.067 coding probes. Differentially expressed genes (DEG) between PT and FA were
assessed with data from EpC and StC captured cells separately. A total of 3386 DEG were identified in EpC, and 1910 were up
regulated in PT. SC samples revealed 2.619 probes with significant difference among them 1.616 up regulated in PT. Hierarchical
clustering based on the DEG from EpC and StC, separately, could not discriminate between TP and FA samples. When low
grade PT (n=7) were compared to high grade PT (n=6) the universe of DEG was smaller: EpC derived samples revealed 30
probes with only 14 up regulated in high grade PTs, and StC samples revealed 6 up regulated genes within only 10 probes with
DEG in PT. Conclusion: EpC are genetically altered as much as StC in breast fibroepitelial lesions carcinogenesis. Low grade
and high grade PTs are similar in RNA and gene expression level. DEGs identified are potential diagnostic targets. Moreover,
EpC and StC components seem to share main affected gene pathways, suggesting a common carcinogenesis process.

Title: Metastatic lymph node fibroblasts maintain similar profile of stromal biomarkers registered in primary breast carcinomas
Fiorita GL Mundim1, Fatima S Pasini2, Suely Nonogaki3, Fernando A Soares3, M Mitzi Brentani2,3 and Angela F Logullo1. 1Escola
Paulista de Medicina EPM-UNIFESP, So Paulo, SP, Brazil; 2Faculdade de Medicina da Universidade de So Paulo, So Paulo,
SP, Brazil and 3Hospital A.C. Camargo, So Paulo, SP, Brazil.
Body: Background: Lymph node metastasis is the predominant dissemination route of breast carcinoma (BC) and remains as an
important prognostic indicator of this disease. Several studies demonstrated a similar gene profiling or gene/biological markers
assessed by immunohistochemistry in neoplastic epithelial cells at both sites. Recent evidence pointed that cancer associated
fibroblasts (CAFs) play a critical role in breast cancer progression and migration. The evidence of CAFs in close association to
epithelial tumor cells within lymph node metastasis raised the possibility of a common source of CAFs in these two lesions.
However, little is known about the differentially expressed genes or proteins between fibroblasts of the primary tumor and its
nodal metastasis. The contribution of CAFs in lymph node colonization is still unknown.
Objective: Our aim was to compare the profile of biomarkers between fibroblasts in lymph nodes and those found in the primaries.
We compared differences in frequency of activated CAFs through the expression of -SMA and S100A4 in 43 matched pairs of
invasive breast carcinomas and respective macro-metastatic lymph nodes arrayed in TMAs. We also investigated a panel of
immunohistochemical biomarkers associated to TGF1 (that induces -SMA and myofibroblast differentiation in CAFs) including
CXCR4, pAKT, pm-TOR and c-myc. In addition the frequency of all these markers were assessed in samples of matched disease
free lymph nodes obtained from the same patients (which contain fibroblastic reticular cells, limited to the exterior capsule).
Results: CAFs characterization confirmed the positive expression of -SMA (approximately 50%) and S100A4 (100%) on
fibroblasts of both sites in contrast to uninvolved lymph nodes, which were uniformly negative. As expected, CAFs were ER, PR,
Her-2, CD9 and p53 negative at both sites. As KI67 labeling rate was always lower than 10% in fibroblasts they were considered
negative. Comparison between fibroblasts from the stromal component of the primaries and respective lymph nodes indicated
that the frequency of TGF1, CXCR4 and pAKT positivity was similar whereas a significant lower rate of c-myc and pm-TOR
frequency was observed in the metastatic lymph nodal fibroblasts. All biological markers were negative in normal lymph nodes
highlighting their association with activated fibroblasts. None of the markers showed association with presence/absence of lymph
node metastasis or with other clinic-pathological parameters. Primary carcinomas lacked any potential associations among the
evaluated biological markers in fibroblast cells, however, in the stromal tissue of lymph nodes, significant relationships between
TGF1 and CXCR4, pAKT and c-myc, were observed.
Conclusion: Our results indicated the presence of activated fibroblasts expressing -SMA and S100A4 in fibroblasts of breast
cancer primaries as well in their matched metastatic lymph nodes. High concordance for the expression of the analyzed
biomarkers in matched pairs of breast cancer primaries and metastatic lymph nodes suggested the maintenance of a similar
supportive microenvironment in metastatic sites. Financial Support: FAPESP and CNPq.

Title: Naked mole rat hyaluronan synthase 2 displays similar effects as human hyaluronan synthase 2 and promotes tumor
growth in a mouse xenograft model
Chunmei Zhao1, Susan Zimmerman1, Calvin T Vu1, Michael Shepard1 and Zhongdong Huang1. 1Halozyme Therapeutics, San
Diego, CA.
Body: Accumulation of hyaluronan (HA) is correlated with poor prognosis in many human cancer including breast cancer and
pancreatic cancer. Hyaluronan synthase 2 (HAS2) is one of the three HA synthase genes (HAS1-3) found in mammals.
Up-regulation of the HAS2 gene has been observed in breast cancer patients; and HAS2 expression correlates with the incidence
of metastasis and lower rate of overall patient survival (Okuda et al 2012). Interestingly, a recent study suggested that the naked
mole rat (nmr) HAS2 enzyme may function differently than the HAS2 proteins from mice and humans, and its expression had a
tumor suppressor effect (Tian et al 2013). The nmrHAS2 protein sequence was distinct from other mammals in the cytoplasmic
loop where the active site resides, which led to the hypothesis that nmrHAS2 may contribute to the production of extremely high
molecular weight HA species. Furthermore, nmrHAS2 is required for the resistance of nmr skin fibroblasts to malignant
transformation. In this study, we examined whether the nmrHAS2 gene can also render human cells resistant to tumor
progression. We have previously observed that HA level in the pancreatic cancer model AsPC1 can be modulated by the
over-expression of the human HAS3 protein (Osgood et al 2014). To compare and contrast the effect of nmrHAS2 and huHAS2
on HA production and tumor growth, nmrHAS2 and huHAS2 AsPC1 models were engineered using a recombinant lentivirus
encoding the nmrHAS2 and huHAS2 genes respectively. Over-expression of nmrHAS2 and huHAS2 both significantly increased
the HA level in AsPC1 cells. Both nmrHAS2 and huHAS2 promoted the growth of AsPC1 in an intramuscular xenograft tumor
model. Ex vivo analysis of tumor xenografts showed that nmrHAS2 and huHAS2 AsPC1 tumors contain elevated levels of HA,
and the size range of the HA in the nmrHAS2 tumors is similar to that in the huHAS2 AsPC1 tumors. Therefore, nmrHAS2
functions similarly as huHAS2 gene and promotes tumor growth in the human pancreatic tumor model AsPC1.
References:
Okuda H, Kobayashi A, Xia B, Watabe M, Pai SK et al. Hyaluronan synthase HAS2 promotes tumor progression in bone by
stimulating the interaction of breast cancer stem-like cells with macrophages and stromal Cells. Cancer Research 72:537-47,
2012.
Tian X, Azpurua J, Hine C, Vaidya A, Myakishev-Rempel M et al. High-molecular-mass hyaluronan mediates the cancer
resistance of the naked mole rat. Nature 499:346-349, 2013.
Osgood RJ, Skipper JF, Cowell JA, Chen Y, Zhu L et al. Pegylated recombinant human hyaluronidase PH20 (PEGPH20)
enhances Nab-Paclitaxel plus gemcitabine efficacy in human pancreatic cancer xenografts. AACR Special Conference Pancreatic
Cancer 2014.

Title: Down-regulation of p16INK4a inhibits miR-146b-5p and modulates IL-6 in breast stromal fibroblasts
Mysoon M Al-Ansari1 and Abdelilah Aboussekhra2. 1King Saud University, Riyadh, Saudi Arabia and 2King Faisal Specialist
Hospital and Research Centre, Riyadh, Saudi Arabia.
Body: Breast cancer is a major health problem that threatens millions of womens lives each year worldwide. Cancer-associated
fibroblasts (CAFs), which constitute the major component of the tumor stroma, have been reported to actively contribute to tumor
cells proliferation and invasion. Recently, we have shown down-regulation of the tumor suppressor p16INK4a protein in breast
cancer-associated fibroblasts. Moreover, p16INK4a deficiency led to the activation of the stromal fibroblasts, which
express/secrete elevated levels of IL-6, a major player in breast carcinogenesis. We have shown here that p16INK4a negatively
regulates the IL-6 expression and secretion in breast stromal fibroblasts. Furthermore, we have shown that IL-6 is playing a major
role in mediating the paracrine pro-carcinogenic effect of p16-deficient fibroblasts. We have also shown that p16INK4a inhibits the
IL-6 expression in a miRNA-146b-5p-dependent manner. Importantly, we present clear evidence that miR-146b-5p inhibition
activates breast stromal fibroblast. Indeed, miR-146b-5p inhibition increased the migration/invasion abilities of breast stromal
fibroblasts, and the paracrine effect of these cells on the migration/invasion of breast cancer cells. Furthermore,
miR-146b-5p-deficient stromal fibroblasts triggered epithelial to mesenchymal transition in breast cancer cells in a paracrine
manner. In addition, we have shown that miR-146b-5p is down-regulated in CAFs as compared to their adjacent counterpart
fibroblasts. These results indicate that p16INK4a negatively regulates IL-6 through the activation of miR-146b-5p, which plays a
major role in repressing breast stromal fibroblasts and inhibiting their pro-carcinogenic effects. This indicates that miR-146b-5p
has cell-non-autonomous tumor suppressor function. Therefore, this miRNA could be of great therapeutic value.

Title: Chemokines released by breast cancer-associated fibroblasts induce epithelial to mesenchymal transition in MCF7 breast
cancer cells
Patsy S Soon1, Edward Kim2, Diana E Benn2 and Robert Baxter2. 1University of New South Wales, Sydney, NSW, Australia and
2
Kolling Institute of Medical Research, Sydney, NSW, Australia.
Body: Purpose: The tumour microenvironment plays a critical role in tumour progression, with cancer-associated fibroblasts
(CAFs) in particular being involved in this process (1). Treating breast cancer cells with conditioned media from CAFs has been
shown to induce epithelial-mesenchymal transition (EMT) of the breast cancer cells compared to conditioned media from normal
breast fibroblasts (NBFs) (2). This study aims to identify factors released by CAFs which induce EMT in breast cancer cells.
Methodology: Primary cultures of six pairs of CAFs and NBFs were established from breast cancer patient samples. Serum-free
conditioned media from these cells were collected for chemokine array analysis. MCF7 breast cancer cells were treated with
recombinant chemokines and their corresponding receptor antagonists. Vimentin levels, as determined by immunoblot, and cell
invasiveness, as measured by transwell assay, were used as indicators of EMT.
Results: Chemokine array showed that CAFs secreted more CXCL1, CXCL8, CXCL12 and CCL2 compared to matched NBFs.
Recombinant CXCL8, CXCL12 and CCL2, but not CXCL1, were able to induce vimentin and to increase invasiveness in MCF7
cells. Antagonists for receptors of CXCL8, CXCL12 and CCL2 counteracted the effect of EMT induction on MCF7 cells by
recombinant chemokines.
Conclusion: CXCL8, CXCL12 and CCL2, which are secreted at higher levels by CAFs compared to NBFs, induced EMT and
increased invasiveness in MCF7 cells. These results suggest that therapeutic targeting of these chemokines or their receptors
may be inhibitory to metastasis in some breast cancers.
References:
1. Bhowmick, N A et al. (2004) Nature 432, 332-337
2. Soon, P S et al. (2013) Endocr Relat Cancer 20, 1-12.

Title: Weight loss reversed the carcinogenic effect of obesity on basal-like breast cancer
Yuanyuan Qin1, Sneha Sundaram1, Liyang Zhao1, Luma Essaid1, Kirk K McNaughton3, Katharine M Bendt2,6, David B Darr2,6,
Melissa A Troester2,7,8 and Liza Makowski1,2,4,5. 1Gillings School of Global Public Health and School of Medicine, University of
North Carolina at Chapel Hill, Chapel Hill, NC; 2Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel
Hill, NC; 3University of North Carolina, Chapel Hill, NC; 4UNC Nutrition Obesity Research Center, University of North Carolina,
Chapel Hill, NC; 5University of North Carolina at Chapel Hill, Chapel Hill, NC; 6Mouse Phase I Unit, Lineberger Comprehensive
Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC; 7University of North Carolina, Chapel Hill, NC and
8
University of North Carolina at Chapel Hill, Chapel Hill, NC.
Body: Epidemiologic studies demonstrate that obesity is associated with a subtype of breast cancer called basal-like breast
cancer (BBC), which is fast proliferating and aggressive, with no targeted therapies. Using the C3(1)-TAg murine model of BBC,
we have previously shown that obese adult mice displayed significantly decreased tumor latency compared to lean mice with
induction of the oncogenic HGF/c-Met pathway. Reducing adiposity is predicted to lower incidence of BBC in human population
studies. Thus, the objective of this study was to investigate how a diet-switch to reduce obesity affected BBC onset and early
progression. C3(1)-TAg mice were placed on a control low fat diet (LFD) and remained lean, high-fat diet (HFD) to induce obesity,
and a group made obese by HFD that were then switched to LFD to induce weight loss. Just two weeks after weight loss, mice in
the diet switch group had weights and adiposity similar to lean mice. Weight loss caused a 51.4% decrease in adiposity
(P<0.0001) compared to obese mice. Importantly, mice that lost weight displayed significantly delayed tumor latency (17 week)
compared to obese mice on HFD (15 week), with no differences detected in tumor burden or growth. Therefore the obese
microenvironment that promotes early tumor onset can be reprogrammed with weight loss and restoration of a lean phenotype.
Ongoing studies are examining HGF/cMet expression by immunohistochemistry in normal mammary glands and tumors to
determine the contribution of this obesity-responsive pathway. Metabolomics and RNAseq analysis will identify novel
obesity-dependent pathways relevant to BBC. In conclusion, we demonstrated that loss of adiposity protected against early BBC
onset. Further research will identify important biomarkers associated with obesity and weight loss that can be compared through
conserved biology approaches.

Title: Genetic analysis of the role of Brca1 in suppression of basal-like breast cancer
Xin-Hai Pei1, Feng Bai1, Ho Lam Chan1, Alexandria Scott1, Matthew D Smith2, Cheng Fan2, Jason I Herschkowitz2,3, Charles M
Perou2, Alan S Livingstone1, David J Robbins1 and Anthony J Capobianco1. 1Miller School of Medicine, University of Miami,
Miami, FL; 2University of North Carolina, Chapel Hill, NC and 3University at Albany, Rensselaer, NY.
Body: BRCA1 mutation carriers are predisposed to developing basal-like breast cancers with high metastasis and poor
prognosis. However, the mechanism underlying the function of BRCA1 in suppressing basal-like breast cancer remains unclear.
We previously reported that deletion of p18Ink4c (p18), an inhibitor of CDK4 and CDK6, functionally inactivates the RB pathway,
stimulates mammary luminal stem cell proliferation, and leads to spontaneous luminal tumor development whereas germline
mutation of Brca1 alters the fate of luminal cells and causes luminal-to-basal mammary tumor transformation. We report here that
disruption of Brca1 by both germline and epithelium-specific mutation of Brca1 in p18 deficient mice activates
epithelial-to-mesenchymal transition (EMT) and induces basal-like breast tumors. Introduction of BRCA1 suppresses expression
of EMT-inducing transcription factors (EMT-TFs) and inhibits EMT in mammary tumor cells. Knockdown of BRCA1 in human
luminal cancer cells activates EMT and increases tumor formation potential. Consistently, EMT features are inversely correlated
with BRCA1 in human breast cancers. These findings provide genetic evidence suggesting that BRCA1 suppresses EMT and
inhibits basal-like tumor development.

Title: Overexpression of the chemokine receptor CCR2 in the breast epithelium is associated with progression of DCIS
Benford Mafuvadze1, Wei Bin Fang1 and Nikki Cheng1. 1University of Kansas Medical Center, Kansas City, KS.
Body: Ductal carcinoma in situ (DCIS) is the most common type of non-invasive breast cancer diagnosed in women and an
immediate precursor to invasive ductal carcinoma (IDC). It is largely unclear why some cases of DCIS progress to IDC, while
others remain non-invasive. Current cyto- and histopathological approaches do not accurately predict disease progression,
resulting in patients being under-treated or over-treated for DCIS. Our long-term goals are to identify key factors that lead to IDC
that will enable the development of a molecular based approach to predict the risk of IDC, and a more tailored approach to treat
DCIS. The CCL2/CCR2 chemokine pathway is best known for regulating macrophage recruitment to late stage breast tumors. In
recent studies, we found that CCR2 and its binding ligand CCL2 were overexpressed in breast ductal carcinomas, correlating with
tumor grade and poor patient prognosis. To further investigate the significance of CCR2 and CCL2, we performed flow cytometry
analysis of breast epithelial cell lines, and found that increased CCR2 overexpression but not CCL2, was associated with invasive
potential of the cell lines. CCL2 was found to be increased in stromal cells, particularly in cancer associated fibroblasts. Using a
mammary intra-ductal injection model, we further examined for expression patterns of CCR2 and CCL2 in DCIS lesions derived
from DCIS.com and Sum225 cells. Increased CCR2 was detected in DCIS.com lesions, which progressed to IDC within 10 weeks
of injection, while CCR2 expression was found to be lowly expressed in non-invasive Sum225 lesions. CCL2 was detected in the
stroma of both DCIS.com and Sum225 lesions. Overexpression of CCR2 in cultured non-invasive 67NR mammary carcinoma
cells enhanced cell survival and increased wound closure. CCL2 treatment further enhanced cell survival and wound closure of
CCR2 overexpressing 67NR cells. Knockdown of CCR2 in DCIS.com cells in 3D Matrigel: collagen cultures inhibited cellular
invasion into the matrix. These studies indicate that CCR2 overexpression in ductal carcinoma cells, enhances intracellular
signaling, and contributes to development of IDC. These data suggest that targeting the CCR2 pathway may be promising
strategy for prevention or treatment of IDC.

Title: Cytoplasmic PELP1 promotes breast cancer initiation via NF-B signaling
Julie H Ostrander1, Brian J Girard1, Todd Knutson1, Bethanie Kuker1 and Victoria Seewaldt2. 1University of Minnesota,
Minneapolis, MN and 2Duke University Medical Center, Durham, NC.
Body: Progress in breast cancer prevention is limited by an inability to reliably predict which women will develop breast cancer
and which high-risk women will respond to chemoprevention therapies. Thus, there is a critical need to determine the molecular
mechanisms that promote breast cancer initiation in order to 1) identify biological markers of disease susceptibility and 2) to
develop novel targeted chemoprevention agents. A promising target for preventing breast cancer initiation is proline, glutamic
acid, and leucine rich protein 1 (PELP1). PELP1 was first identified as an estrogen receptor (ER) co-activator in ER+ breast
cancer cell lines. Subsequent studies found that PELP1 functions in ER+ and ER- breast cancer cell lines and is overexpressed
in 80% of invasive breast cancers. Interestingly, while PELP1 is primarily localized to the nucleus in the normal breast epithelial
cells, in about 40% of invasive breast tumors a significant amount of PELP1 is localized in the cytoplasm. Interestingly, in a
mammary gland specific transgenic mouse model, PELP1-cyto expression induced mammary gland hyperplasia. We therefore
analyzed breast needle aspirates from asymptomatic high-risk women, and found cytoplasmic PELP1 in 4/11 (36%) samples.
These findings suggest that altered localization of PELP1 may be an early event in breast cancer initiation.
The objective of this research project is to identify the molecular mechanisms associated with PELP1-induced breast cancer
initiation. We performed global gene expression analyses of our in vitro human mammary epithelial cell (HMEC) models to
identify potential downstream pathways regulated by PELP1-cyto. Interestingly, genes regulated by PELP1-cyto differ greatly
from those regulated by nuclear PELP1. Our results showed that PELP1-cyto induces expression of cell survival genes, and
inflammatory cytokines and chemokines, and activates the NF-B signaling pathway. Thus, we hypothesize that cytoplasmic
PELP1 induces activation of inflammatory cytokines and chemokines via NF-B, which promotes tumor initiation and a
pro-tumorigenic microenvironment.
In support of our hypothesis we have found that PELP1-cyto expression promotes phosphorylation of the non-canonical IKK,
TBK1, and the NF-B subunit p65. Additionally, we found that conditioned medium from PELP1-cyto cells, induced expression of
IL-1, TNF, and IL-8 in THP-1 cells that had been differentiated into macrophages, suggesting that PELP1-cyto HMECs release
cytokines that promote the activation of macrophages. The goal of future studies is to validate these findings in vitro and in vivo
and identify the cytokines and/or chemokines induced by PELP1-cyto that promote breast cancer initiation and activation of tumor
associated macrophages.

Title: MECP2 is a frequently amplified oncogene and potential therapeutic target in TNBC
Daniel P Silver1, Manish Neupane1, Allison P Clark1, Marc Vidal1 and David E Hill1. 1Dana-Farber Cancer Institute, Harvard
Medical School, Boston, MA.
Body: Background: To identify new oncogenes that drive cancer development, we conducted an unbiased genome-scale
screen for genes that can substitute for activated RAS in oncogenic transformation. We focused attention on one of the new
potential oncogenes identified in this screen, Methyl CpG Binding Protein 2 (MECP2), which has no previously described role in
malignancy and is amplified across 20% of all human cancers, and 30% of Triple-Negative Breast Cancer (TNBC). MECP2 is
an X-linked gene known to bind methylated cytosines, and can act as a transcriptional repressor in this context. Recent studies
show that it also acts as a transcriptional activator, likely through binding to another epigenetic modification of DNA,
5-hydroxymethylcytosine (5hmC).
Results: MECP2 is as potent as activated RAS in conferring anchorage independent growth upon primary human mammary
epithelial cells (hMECs) previously transduced with the SV40 early region and hTERT (NRAS hMECs). MECP2 partially rescues
the growth inhibition of RAS-addicted human cancer cell lines after the shRNA-mediated suppression of RAS. MECP2 expresses
two spliced isoforms; experiments showed the short isoform activates the MAPK pathway, while the long isoform activates the
PI3K pathway. Neither isoform alone can cause growth of NRAS hMECs as a xenograft in nude mice; together, they can. The
transformation and growth factor pathway induction activities of MECP2 absolutely require its DNA-binding activity.
A number of TNBC cell lines have amplified, overexpressed MECP2, and of the first 13 TNBC patient-derived xenografts
examined, 4 have MECP2 overexpression. Several TNBC cell lines that have amplified, overexpressed MECP2 show significant
growth inhibition after shRNA-mediated downregulation of MECP2 (MECP2 addiction), while a breast cancer cell line without
MECP2 overexpression showed no such inhibition. NRAS hMECs transformed with MECP2 are an order of magnitude more
sensitive to either of the DNA methylation inhibitors 5-azacytidine or decitabine than isogenic cells transformed by activated RAS,
or isogenic cells without an additional transforming gene. Further, we find that combined treatment with the DNA methylation
inhibitor 5-azacytidine and the HDAC inhibitor Trichostatin A is synergistic in our hMEC experimental system.
Conclusion: MECP2 is a commonly amplified and overexpressed oncogene whose two splicing isoforms together recapitulate
the major oncogenic functions of activated RAS. Because MECP2 requires DNA binding to methylated or hydroxymethylated
cytosines for its tumor-promoting activities, DNA methylation inhibition with FDA-approved drugs may be therapeutic for tumors
overexpressing MECP2.

Title: A novel TP53-dependent role for TLR4 in driving breast cancer
Svasti Haricharan1 and Powel H Brown1. 1MD Anderson Cancer Center, Houston.
Body: Breast cancer is a leading cause of cancer-related death. Toll-like receptor (TLR)-4 is an important mediator of cytokine
secretion that is expressed in breast cancer cells, and is frequently mutated in a subset of breast tumors associated with high
mutation load and poor patient survival. We show here that TLR4 plays an important role in driving breast cancer growth in a
TP53 context-dependent manner. Using siRNA mediated knockdown, pharmacological inhibition and hyperactivation of TLR4 we
demonstrate that TLR4 inhibits growth in breast cancer cells with wildtype TP53 by inducing G0/G1 arrest and decreasing mitotic
entry. Conversely, in breast cancer cells with mutant TP53, TLR4 acts as an oncogene and drives cell growth by inducing mitosis.
Furthermore, we identify the underlying mechanism for this dual TP53-dependent effect of TLR4 on breast cancer cell growth:
differential interferon gamma (IFN-) secretion by tumor cells into their microenvironment. Hyperactivated TLR4 signaling in TP53
wildtype breast cancer cells results in increased IFN- secretion, identified by an unbiased cytokine array and validated by ELISA,
thus inhibiting growth in an autocrine/paracrine fashion. Moreover, secreted IFN- is both necessary and sufficient for
TLR4-induced growth inhibition in TP53 wildtype breast cancer cells. Finally, the dual TP53-dependent effect of TLR4
extrapolates to several other cancer types in silico, attaching potentially global significance to these results. Taken together, the
data presented in this paper strongly suggest a novel role for TLR4 as a suppressor of cell growth in TP53 wildtype tumors, and
identify differential IFN- secretion as the underlying mechanism. The results of this study are also translationally relevant,
delineating the TP53 mutant breast cancer subset as a group that can benefit from pharmacological TLR4 inhibition. Most
importantly, these results indicate a need for studying the effect of drivers of cancer growth pleiotropically rather than as isolated
events.

Title: Elevated frequency of a deleterious deep intronic BRCA2 splice variant in non-Caucasian Americans
Charles M Strom1, S Rivera1, M Peng1, Wendy Conlon1, B Crossley1 and W Sun1. 1Quest Diagnostics Nichols Institute, San Juan
Capistrano, CA.
Body: Introduction: In 2012, Anczukow et al (Clin Cancer Res 2012;18:4903-9) described a deep intronic BRCA2 variant
between exons 12 and 13 that was discovered because a cell line from a patient with hereditary breast and ovarian cancer
(HBOC) produced an aberrant mRNA. The variant (c.6937 +594 T>G; rs191253965), which further improved splice site sequence
to consensus, was also identified in 8 additional families. Segregation analysis of 13 affected relatives from 6 families revealed
that all affected family members had the c.6937 + 594 T>G variant, providing strong evidence that it is deleterious. All resided in
France, but ethnicity was not specified. This variant was also seen in 5 individuals in the 1000 genomes project. Since this is a
deep intronic variant, standard BRCA2 sequencing tests that determine only a limited number of bases into each intron would not
detect it. For these reasons, we included a bait tile for the intron harboring this variant in the design of our hybrid capture-based
next generation sequencing (NGS) test for comprehensive BRCA1 and BRCA2 analysis. Here we investigate the frequency of
this variant in samples submitted for comprehensive BRCA1 and BRCA2 testing, and assessed carrier frequency in samples
submitted for non-BRCArelated genetic testing.
Methods: We reviewed the de-identified results of the first 2300 clinical samples submitted to our reference laboratory for
comprehensive BRCA1 and BRCA2 testing to assess the frequency of the c.6937 + 594 T>G variant.
Results: This deep intronic variant was observed in 12 of the initial 2300 patient samples. All 12 individuals were affected with
HBOC and were Hispanic. To estimate the population frequencies of this variant in various ethnic groups, we anonymized
samples submitted for Cystic Fibrosis Carrier Detection and subjected them to a single-site assay for this variant. In all, 344
patients self-identified as Caucasian, 355 as Hispanic, 140 as African American, and 73 as Asian. The frequency of c.6937 +594
T>G heterozygotes was 0.30% (1 individual) in Caucasians, 1.1% (4) in Hispanics, 0.8% (1) in African Americans, and 1.4% (1) in
Asians.
Conclusions: In an initial series of comprehensive BRCA analyses, we discovered that a deep intronic variant known to cause
aberrant splicing and to segregate with HBOC is present at elevated frequencies in non-Caucasian Americans. We have already
observed this variant in 12 Hispanic HBOC patients. Physicians could consider testing for this variant in women of non-Caucasian
ancestry with HBOC who have had negative results on BRCA assays that do not test for c.6937 +594 T>G.

Title: Functional genomics of TP53 mutations and its impact in breast cancer progression
Anasuya Pal1, Laura Gonzalez-Malerva1, Seron Eaton1, Mayra Guzman1, Donald Chow2, Hongwei Yin2, Jin Park1, Karen
Anderson1 and Joshua LaBaer1. 1Biodesign Institute, Arizona State University, Tempe, AZ and 2TGen Drug Development (TD2),
Scottsdale, AZ.
Body: Somatic TP53 mutations are prevalent in basal-like breast cancer (BLBC) tumors. Patients with BLBC tumors have fewer
treatment options and respond poorly to current therapies. The majority of TP53 point mutations occurs in the DNA binding
domain and can be categorized as either DNA contact or structural mutations. TP53 mutation results in a dominant negative
phenotype with neomorphic activity. We predicted that different p53 mutations may lead to different phenotypic characteristics. To
investigate this, we generated MCF10A stable transduced cell lines over-expressing the ten most frequent TP53 point mutations
associated with breast cancer located in the DNA binding domain of TP53. Ectopic expression of TP53 in these stable cells has
been confirmed by qRT-PCR and immunoblot. To assess the impact of mutation on carcinogenesis, we developed a series of
high-throughput quantitative assays that measure several hallmarks of cancer, including proliferation, escape from apoptosis,
epithelial to mesenchymal transition (EMT), cell migration and invasion, anoikis and morphology in 3D. We observed that one
DNA contact mutation with the substitution of a positively charged amino acid with hydrophobic side chains such as R248W, and
two structural mutants Y234C and H179R are resistant to apoptosis in presence of doxorubicin, are the most invasive displaying
a mesenchymal phenotype characterized by the presence of disrupted B-catenin and E-cadherin staining, with reported worst
clinical outcome, suggesting that these are the most aggressive phenotypes. Interestingly, the DNA contact mutants (R248Q,
R273H, R248W, and R273C) had a growth advantage in absence of growth factors while structural mutants (R175H, H179R,
Y220C, Y234C and Y163C) were more resistant to apoptosis after the cells were challenged with doxorubicin. G245S is
comparable to the MCF10Ap53wt and is less proliferative, sensitive to apoptosis, and neither migratory nor invasive. In
comparison, R248W which is one of the most aggressive mutants, together with R273C, and H179R resist anoikis; but Y234C,
requires matrix for attachment in order to be invasive. In conjunction, these results confirmed our hypothesis that different TP53
point mutants have distinct phenotypes and functional effects on hallmarks of cancer due to distinct underlying cellular programs.

Title: Cellular localization dictates the role for c-Abl in breast cancer
Chevaun D Morrison1, Hannah Gilmore2 and William P Schiemann2. 1Cleveland Clinic Lerner College of Medicine of Case
Western Reserve University, Cleveland, OH and 2Case Western Reserve University, Cleveland, OH.
Body: c-Abl is a ubiquitously expressed non-receptor tyrosine kinase that regulates numerous aspects of normal mammary
epithelial cell (MEC) physiology. However, the exact role of c-Abl during mammary tumorigenesis remains controversial. Clinical
trials utilizing the c-Abl antagonist, imatinib mesylate, for the treatment of metastatic breast cancer (BC) patients failed due to
disease progression. Additionally, our previous studies overexpressing a constitutively active c-Abl mutant in aggressive murine
triple negative breast cancer (TNBC) cells abrogated tumorigenesis when injected into syngeneic mice. Through data mining we
found c-Abl expression to be down-regulated across BC subtypes and this correlated with a significantly worse relapse free
survival rate for patients of triple negative and luminal A BC subtypes. Furthermore, through in silico analyses we determined that
BC patients with high c-Abl expression responded to docetaxel treatment, whereas their low c-Abl expressing counterparts did
not. Taken together these findings support a tumor suppressive role for c-Abl in BC and that its targeted activation could inhibit
BC tumorigenesis as well as sensitize BC patients to docetaxel treatment. We show here that shRNA mediated depletion of c-Abl
elicited docetaxel resistance in malignant MECs through three dimensional (3-D) culture and clonogenic assays on plastic.
Furthermore, we show through data mining that c-Abl expression is down-regulated in response to docetaxel treatment in a
cohort of BC patients. This finding was recapitulated in malignant MECs through which docetaxel treatment selected for cells that
retained lower c-Abl expression highlighting the importance of c-Abl in this cell death pathway. Targeted activation of c-Abl
utilizing a novel small molecule activator, DPH, presents a novel therapeutic strategy for sensitizing BC patients to docetaxel
treatment. However, we show DPH-mediated c-Abl activation to inhibit the proliferation of luminal A MCF7 BC cells, but enhance
proliferation of the TNBC MDA-MB-231-L2 cell line in vitro. The cellular localization of c-Abl is crucial for its tumor suppressive
functions. In Philadelphia chromosome positive chronic myelogenous leukemia (CML) the constitutively active BCR-Abl gene
fusion has a strict cytoplasmic localization, however when nuclear expression of BCR-Abl is enforced CML cells undergo
apoptosis. Cellular fractionation of normal murine MECs cells indicate that c-Abl expression is detected both in cytoplasmic and
nuclear compartments before and with greater amounts detected in the nucleus upon activation. A similar trend is seen in the
luminal A MCF7 BC cell line, however c-Abl is predominantly expressed in the cytoplasm basally in the TNBC MDA-MB-231-L2
cell line and retains a strict cytoplasmic localization upon activation with DPH. These findings suggest a similar mechanism as
CML of cytoplasmic isolation as a means of circumventing the tumor suppressive role of c-Abl across different BC subtypes. Data
from ongoing immunohistochemical analyses of a cohort of BC patient tissue microarrays will inform whether nuclear versus
cytoplasmic c-Abl will correlate with available patient outcome data to determine if cellular localization dictates a tumor
suppressive or tumor promoting role for c-Abl.

Title: Matrin 3: A novel micro-tubule associated RNA binding protein that acts as a potent tumor suppressor
Panneerdoss Subbarayalu1,2, Subapriya Rajamanickam1,2, Suryavathi Viswanadhapalli3, Nicholas Dybdal-Hargreaves4, Santosh
Timilsina1,2, Sanjay Bansal1, Hima Bansal1, Tabrez Mohammad1, Yidong Chen1, John C Herr5, Susan L Mooberry3,4 and Manjeet K
Rao1,2. 1Greehey Children's Cancer Research Institute, University of Texas Health Science Center, San Antonio, TX; 2University
of Texas Health Science Center, San Antonio, TX; 3University of Texas Health Science Center at San Antonio, San Antonio, TX;
4
University of Texas Health Science Center, San Antonio, TX and 5University of Virginia, Charlottesville, VA.
Body: Microtubules are highly dynamic components of the cytoskeleton that play an important role in a wide range of cellular
processes including cell division, cell motility and intracellular transport. Increasing evidence suggests that alterations in
microtubule dynamics are critical for cancer growth and metastasis. Microtubule associated proteins (MAPs) regulate microtubule
dynamics and consequently can affect sensitivity of cancer cells to microtubule targeting drugs. We discovered a novel
microtubule associated protein "Matrin 3 (MATR3)" that is known to bind to RNA and play a critical role in RNA transport and RNA
stabilization. Immunofluorescence analyses revealed that although MATR3 is predominantly a nuclear protein, it translocates to
the cytoplasm and interacts with microtubules when breast cancer cells are treated with paclitaxel. We show that MATR3
associates with stabilized microtubules and it mediates microtubule polymerization in a taxol-dependent manner. Interestingly, our
results reveal that MATR3 acts as a effective tumor suppressor as it inhibits breast cancer colony formation, migration and
invasion of breast cancer cells in addition to suppressing breast tumor growth in vivo. Analysis of breast cancer samples showed
a significantly lower expression of MATR3 when compared to normal adjacent tissues. Our results indicate the possibility that
MATR3 mediates its tumor suppressor function by binding and regulating proteins that are known to affect microtubule dynamics.
We believe that nuclear-cytoplasmic shuttling of MATR3 is critical for stability of key proteins that might regulate
paclitaxel-dependent microtubule dynamics and subsequently cellular effects in cancer cells.

Title: PRKCQ, a novel protein kinase C preferentially expressed in triple negative breast cancer, drives oncogenic growth,
survival and migration
Hanna Y Irie1, Gwyneth Halstead-Nussloch1 and Koichi Ito1. 1Icahn School of Medicine at Mount Sinai, New York, NY.
Body: Background/Rationale: PRKCQ, a member of the novel protein kinase C family, was identified in a functional genomic
screen for regulators of anoikis resistance/anchorage-independent survival (Irie et al., 2010). Interestingly, it is preferentially
expressed in the triple negative/basal subtype of breast tumors compared to Luminal or Her2+ tumors. We sought to determine
the functional role of PRKCQ in triple negative breast cancer and evaluate PRKCQ as a candidate therapeutic target for this
subtype. Materials and Methods. Consistent with its expression in triple negative patient breast tumors, PRKCQ is also highly
expressed in several triple negative breast cancer cell lines. We utilized both gain- and loss-of-function approaches in these cell
lines to determine the requirement for PRKCQ in oncogenic growth and survival using in vitro and in vivo models. Results.
Isoform-specific downregulation of PRKCQ using shRNA vectors severely impaired growth of triple negative breast cancer cells in
2D monolayer and 3D Matrigel cultures. In 3D cultures, PRKCQ downregulation not only inhibited growth, but impaired the
formation of invasive branching. PRKCQ downregulation also inhibited the growth of MDA231 primary tumor xenografts. All of
these data support a requirement for PRKCQ expression in triple negative breast cancer cell growth. To determine if PRKCQ is
sufficient to drive oncogenic phenotypes, we overexpressed PRKCQ in a non-transformed immortalized breast epithelial cell line,
MCF-10A. PRKCQ expression conferred EGF-independent growth, migration and anoikis resistance. In PRKCQ-expressing
MCF-10A cells, EGFR phosphorylation and activation were preserved in the absence of exogenous EGF ligand addition. We are
currently elucidating the mechanisms responsible for PRKCQ-mediated, sustained activation of EGFR. Conclusions. PRKCQ is
critically required for growth of triple negative breast cancer cells. Increased expression of PRKCQ is sufficient to drive oncogenic,
growth factor-independent growth, survival and migration. Interestingly, PRKCQ could play a dual role in the development of triple
negative breast cancer, as it may also function in the immune microenvironment to support the growth of these tumors. Therefore,
PRKCQ is an attractive candidate therapeutic target for patients with triple negative breast cancer.

Title: CENPU acts as a new proto-oncogene to regulate tumorigenesis and cancer metastasis in breast cancer
Jin Zhang1,2,3, Xiaobei Zhang1,2,3, Yunhui Hu1,2,3, Jiao Li1,2,3, Jingjing Liu1,2,3, Sheng Zhang1,2,3 and Wei Su1,2,3. 1Tianjin Medical
University Cancer Institute and Hospital, National Clinical Research Center of Cancer, Tianjin, China; 2Key Laboratory of Breast
Cancer Prevention and Therapy of Ministry of Education, Tianjin, China and 3Key Laboratory of Cancer Prevention and Therapy,
Tianjin, China.
Body: Background: Centromere protein (CENPU) gene locus is at chromosome 4q35.1, encoding a protein with some classic
functional domains; recently some research groups proved CENPU protein is a constitutive kinetochore component and regulates
cell-cycle status by recruitment of function-specific proteins. Homologous gene in murine hematopoiesis system represents early
erythroblasts-specific expression. Our previous research had found that about 25 % of transgenic mice amplified CENPU would
suffer breast cancer in body surface which induce us to hypothesize that CENPU gene and its protein played an important role in
breast cancer occurrence and development.
Methods: CENPU protein expression was examined in normal mammary tissue, DCIS tissue, primary invasive breast cancer and
different human breast cancer cell lines. Cell type and epitope-dependent subcellular-specific CENPU staining pattern in normal
mammary gland epithelium and cancer biopsies were correlated to molecular and clinical parameters. A CENPU-specific small
inhibitory RNA (siRNA) was introduced into MDA-MB-231 human breast cancer cell lines to investigate its effect on cancer cell
growth, migration and invasion. Distribution of cell cycles were examined with flow cytometry. Test the tumorigenicities of
si-control and si-CENPU cells by soft agar colony formation assay. Migration was observed by wound healing and transwell
migration assays. The expression of related pathway proteins were determined by Western blotting analysis.
Results: Compared with normal mammary and DCIS tissues, CENPU protein was highest expression in primary invasive breast
cancer tissue (P<0.001).Then we compared CENPU expression lever between no metastasis and metastasis patients during
three year, and found that CENPU expression in recurrence group is higher than no-metastasis group (P<0.01). CENPU protein
expression of different human breast cancer cell lines were detected, which found the CENPU in triple negative breast cancer cell
line, MDA-MB-231, was the highest expression. Knockdown of CENPU by specific siRNA in MDA-MB-231 reduced the ability of
colony formation and induced the G2/M phase arrest. Moreover, the migration rate of si-CENPU MDA-MB-231 cancer cells was
significantly reduced as compared with si-controls (P<0.01).With the knockdown of CENPU, the E-cadherin expression was
up-regulated and the N-cadherin, AKT1 and NF-B were downp-regulated.
Conclusion: We propose that CENPU functions as a novel proto-oncogene to regulate oncogenesis and metastasis. In further, in
vivo study is needed to evaluate the biologaical importance of CENPU in breast cancer.


Title: PTEN deletion is linked to adverse phenotype and poor prognosis in breast cancer
Eike Burandt1, Martina Kluth1, Valerie Kopperschmidt1, Anja Mittenzwei1, Annette Lebeau1, Volkmar Mller1, Isabell Witzel1, Fritz
Jnicke1, Stefan Geist2, Peter Paluchowski2, Christian Wilke3, Uwe Heilenktter4, Rainer Krech5, Albert von d Assen6, Ronald
Simon1 and Patrick Lebok1. 1University Medical Center Hamburg-Eppendorf, Hamburg, Germany; 2Regio Clinic Pinneberg,
Pinneberg, Germany; 3Regio Clinic Elmshorn, Elmshorn, Germany; 4Clinical Centre Itzehoe, Itzehoe, Germany; 5Clinical Centre
Osnabrck, Osnabrck, Germany and 6Breast Centre Osnabrck, Osnabrck, Germany.
Body: Deletions of chromosome 10q23, including the PTEN (phosphatase and tensin homolog) locus, are known to occur in
breast cancer, but systematic analyses of its clinical relevance are lacking. We thus analyzed a tissue microarray (TMA) with
2,197 breast cancers by fluorescence in-situ hybridization (FISH) using a PTEN-specific probe, and found deletions in 19% of no
special type, 9% of lobular, 46% of medullary and 4% of tubular cancers. 98.7% of deletions were heterozygous and 1.3% were
homozygous. PTEN deletion was significantly linked to advanced tumor stage (p=0.0054), high tumor grade (p<0.0001), high
tumor cell proliferation (Ki67 Labeling Index; p<0.0001), and shortened overall survival (p=0.0090). PTEN deletions were
inversely associated with features of luminal type breast cancers (ER/PR positivity, CCND1 amplification). PTEN deletions were
strongly linked to amplification of genes involved in the PTEN/AKT pathway such as MYC (p=0.0430) and HER2 (p=0.0065).
Remarkably the combined analysis of MYC, HER2, and PTEN aberrations suggested that aberrations of multiple PTEN/AKT
pathway genes have a strong additive effect on breast cancer prognosis. Cancers with two or three of these aberrations behaved
significantly worse than cancers with none or one of these changes (p<0,0001). The particularly poor prognosis of patients with
HER2 amplification and PTEN deletions challenges the concept of PTEN deletions interfering with trastuzumab therapy. In
conclusion, PTEN deletion occurs in a relevant fraction of breast cancers, and is linked to aggressive tumors. Reduced PTEN
function cooperates with MYC and HER2 activation in conferring aggressive phenotype to cancer cells.

Title: The PI3KCA mutations are frequent and remain along with recurrence of liver metastases from breast cancer in women
who underwent up to 3 liver resections
Archie Ruiz1, Mylne Sebagh2, Raphal Saffroy3, Marc-Antoine Allard1,3, Philippe Bouvet De La Maisonneuve3, Nelly Bosselut3,
Ren Adam1, Antoinette Lemoine3 and Jean-Franois Morere4. 1Centre Hpato-Biliaire-Chirurgie Digestive, Hpital Paul Brousse,
Villejuif, France; 2Hpital Paul Brousse, Villejuif, France; 3Hpital Paul Brousse, Villejuif, France and 4Hpital Paul Brousse,
Villejuif, France.
Body: The Liver resection associated with anticancer and hormonal treatments for metastatic breast cancer results in improved
patient survival. However, many patients have tumor recurrence. The frequently observed mutations in PIK3CA that have been
associated with resistance to chemotherapy and to anti-HER2 or anti-estrogen therapies treatment, could be involved in the
recurrence of liver metastases (and targeted in the future). Therefore, we have analyzed the incidence of the two major 'hot spot'
mutations in the helical and catalytic domains of PI3KCA in the liver metastases from breast cancer (LMBC) isolated in a selected
cohort of 20 women who underwent at least 2 liver resections with curative intent. Twelve of them had a third hepatectomy. The
first hepatectomy was between 1 to 6 yrs after the diagnosis of primary tumors. A total of 73 LMBC were analyzed.
The characteristics of the LMBC at the first hepatectomy : the median age was 52 yrs (range: 30 to 64). Five patients had a
solitary tumor whereas the others had up to 5 nodules. The tumor stages varied from stage IA to IIIA and were ductal tumors for
80% of them. Hormonal status varied from the negativity for the receptors to estrogen (ER), progesterone (PR) and HER2 (n=4),
to the clearcut positivity ER (n=2), PR (n=3), HER2 (n=4). Out of 20 patients, 6 had LMBC carrying PI3KCA mutations (1
H1047R, 6 E545K). When LMBC were multinodular, all the nodules except one harbored the PI3KCA mutation.
The characteristics of the recurrent LMBC: out of the 6 LMBC with PI3KCA mutations, 4 recurred and underwent a third
hepatectomy. One patient whose LMBC did not harbor any PI3KCA mutations at the first liver resection had mutant PI3KCA in
the LMBC from the second to the fourth liver resection. All the nodules of the 5 women harbored the same PI3KCA mutation. In
parallel, 7 women underwent a third liver resection with no PI3KCA mutations in any of the nodules analyzed. There was no
relationship between the recurrence of LMBC with PI3KCA mutations and the surgical margins. The frequency of women whose
LMBC carry PI3KCA mutations increased at the third hepatectomy
In conclusion, PI3KCA mutations are frequently observed in LMBC in all the nodules and persist along with the recurrence.
However, PI3KCA could constitute a new target in the neoadjuvant setting before hepatectomy.

Title: Examining the role of Nm23-H1 in the metastatic profile of triple negative breast cancer (TNBC)
Tanisha Z McGlothen1, Rachel Tobin1, Tiffanie Alcaide1, LaTonia Taliaferro-Smith1, Liu Tongrui1 and O'Regan Ruth1. 1Emory
University, Atlanta, GA.
Body: Background: Triple negative breast cancer makes up 10-20% of all mammary tumors. It is so named because it exhibits
low or no expression of both the estrogen receptor (ER) and progesterone receptor (PR), and has low or no expression of the
human epidermal growth factor (HER2) receptor. With no current molecular targets identified, treatment options for TNBCs are
limited to conventional chemotherapy, which has been shown to be only moderately effective. Furthermore, diagnosis is
correlated with high rates of relapse and metastatic disease, low survival rate five years past diagnosis, and overall poor
prognosis.
Nm23-H1 is the most well characterized in a class of metastasis suppressor genes that have received increased attention as a
potential therapeutic target in breast cancers. Nm23-H1 mRNA is expressed at relatively low levels in highly metastatic tumor
cells compared to normal and non-neoplastic tissues. Low expression levels correlate with increased metastasis and poor clinical
prognosis in several type cancers, including breast cancer. Nm23-H1 inhibits a class of Rho small GTPases, including Rac1, Rho
and cdc42, which mediate cell-to-cell adhesion and cytoskeletal reorganization, which ultimately modulates the metastatic profile,
i.e. metastasis, cellular motility, and invasion.
Methods: Western blotting was used to examine the basal level of expression in a panel of mesenchymal and epithelial TNBC cell
lines. Transient siRNA was used to silence Nm23-H1 in MDA MB 231 cells to determine the effects on downstream Nm23-H1
targets IQGAP2, Rac1, Rho and Cdc42. The effects of Nm23-H1 inhibition on the invasive potential of TNBC cells were assessed
using Matrigel chamber assays. Spheroid migration assays were employed to assess the effects of Nm23-H1 inhibition on the
migratory potential of TNBC cells.
Results: Our studies indicate that Nm23-H1 protein is expressed in all TNBC cell lines tested but to varying degrees, with
mesenchymal cells expressing higher levels compared to epithelial cells. Immunoblotting showed that silencing Nm23-H1
resulted in an increase in phosphorylated Rac and Rac and the cytoskeletal intermediate filament vimentin. Silencing also
decreased the invasion and migration of TNBC cells compared to control cells.
Conclusion: These results suggest a paradoxical role of the metastasis suppressor protein Nm23-H1 as an oncogene in TNBC
cells. Nm23-H1 plays a vital role in the promotion of TNBC cellular migration, invasion, and the expression of proteins associated
with motility and invasion. These data strongly suggest an oncogenic potential of Nm23-H1 in a subset of TNBCs and warrants
further investigation of this protein a potential molecular marker for metastatic TNBCs.

Title: No Show

Title: No Show

Title: No Show

Title: Epiregulin-induced matrix metalloproteinase-1 contributes to early stages of breast tumorigenesis
Kathryn L Schwertfeger1, Mariya Farooqui1, Laura R Bohrer1 and Andrew C Nelson1. 1University of Minnesota, Minneapolis, MN.
Body: The earliest stages of malignant transformation require the acquisition of multiple phenotypes, such as proliferation,
survival and migration, which contribute to tumor growth and progression. Identifying the key cellular and molecular factors that
drive the formation of early breast cancer lesions will ultimately result in the development of preventive approaches for patients
that are at high risk for developing invasive breast cancer. The epidermal growth factor (EGF) family of ligands has been
implicated in promoting breast cancer growth and progression. The contributions of EGF family ligands and their receptors to
breast cancer are complex, and the specific mechanisms through which different ligands regulate different stages of breast tumor
initiation and growth are not well-defined. These studies focus on the EGF family member epiregulin (EREG) as a mediator of
early stages of breast tumorigenesis. In comparison with other EGF family members, EREG was found to be highly expressed in
MCF10DCIS.com cells compared with the non-transformed MCF10A cells. Increased EREG expression was also found in
significantly more patient samples of human ductal carcinoma in situ (DCIS) than in samples of normal epithelium (64.5% vs.
29.4%, p<0.05). Therefore, further studies were performed to identify mechanisms through which EREG promotes the formation
of DCIS lesions.
Experimental studies demonstrated that treatment of MCF10A cells with exogenous EREG enhanced multiple tumorigenic
phenotypes, including proliferation, migration and survival. Examination of EREG-induced signaling pathways demonstrated that
EREG promotes survival of MCF10A cells primarily through activation of the signal transducer and activator of transcription 5
(STAT5) pathway. Analysis of potential STAT5 target genes revealed that the EREG-STAT5 pathway induces expression of
matrix metalloproteinase-1 (MMP-1), a novel STAT5 target gene that was subsequently found to promote survival in response to
EREG treatment. To assess the consequences of loss of EREG expression on tumor formation, EREG was knocked down in
MCF10DCIS.com cells using shRNA strategies. Loss of EREG led to decreased tumor growth in vivo, which corresponded with
decreased cell survival and decreased MMP-1 expression. To determine the relevance of these findings in human tumors,
samples of DCIS were analyzed for both EREG and MMP-1 expression. MMP-1 was significantly induced in DCIS lesions in
comparison with normal breast epithelium, and EREG and MMP-1 were significantly correlated in a subset of DCIS samples
(p=0.011). Together, these studies suggest that EREG contributes to early stages of breast tumorigenesis through regulation of
MMP-1. Based on these studies, we propose that EREG contributes to the formation of preneoplastic lesions in a subset of breast
cancers and further studies are focused on characterizing the mechanisms through which the EREG-MMP-1 pathway contributes
to tumor initiation and growth. Understanding these mechanisms will ultimately lead to the identification of biomarkers of
aggressive disease and novel targets for therapeutic strategies for breast cancer patients.

Title: Mammaglobin is a potent regulator of breast cancer processes leading to disease progression
Roxann Guerrette1, Nadia Picot1 and Gilles Robichaud1. 1Universit de Moncton, Moncton, NB, Canada.
Body: Metastasis is the major cause of death in women suffering from breast cancer. To provide a better understand breast
cancer progression, we have studied the role of mammaglobine-1 (MGB1) gene in breast cancer pathogenesis. MGB1 has been
extensively studied as a diagnostic biomarker due to its abundant expression in mammary cancer cells. Yet, MGB1's role in
disease progression is still unknown. Our experimental results demonstrate for the first time that MGB1 in a pivotal regulator on
breast cancer malignancy. More precisely, loss of MGB1 expression correlates with a decrease in proliferation, spheroid
formation, migration, and invasion capacities of breast cancer cells. Concomitantly, we also observe that MGB1 expression
activates pro-malignant signaling cascades such as MAPKs, focal adhesion kinase (FAK) and NFkB pathways. Moreover, MGB1
promote epithelial to mesenchymal (EMT) features which coincide with our findings. Our study provides the first evidence for
MGB1 as regulator of breast cancer malignancy and disease progression.

Title: Insulin receptor targeting in breast cancer through yeast surface display
Jie Ying Chan1, Benjamin Hackel1 and Douglas Yee1. 1University of Minnesota, Minneapolis, MN.
Body: Selective estrogen receptor modulators (SERMs) such as tamoxifen have been a conventional therapy for breast cancer
patients expressing estrogen receptor, representing the most common breast cancer type. The development of secondary
resistance to SERM is an unsolved clinical dilemma, leading to the exploration of more targeted therapies. The insulin-like growth
factor (IGF) system is well documented and is implicated as a contributor to endocrine resistance. Unfortunately, antibodies
directed against type I IGF receptor (IGF1R) failed to demonstrate efficacy in endocrine resistant breast cancer. We have
previously shown that tamoxifen resistant (TamR) breast cancer cells lose IGF1R yet retain expression of insulin receptor (InsR),
a closely related receptor to IGF1R. Thus, we hypothesized that InsR may serve as a compensatory pathway to the loss of IGF1R
in TamR cells and InsR may be a target in the therapy of endocrine resistant breast cancer. Indeed, our preliminary results show
that TamR breast cancer cells were more sensitive to insulin stimulation. As compared to the parental cells, TamR cells showed
stronger PI3K/AKT and MAPK/ERK activation, greater MTT proliferation growth and anchorage-independent growth upon insulin
treatment. Knocking down InsR in TamR cells with shRNA was able to attenuate their sensitivity towards insulin-mediated
PI3K/MAPK activation and growth, suggesting InsR targeting may be necessary in endocrine resistant breast cancer. To develop
new agents to target InsR, we used yeast surface display to develop an InsR-selective protein scaffold the 10th type III domain
of human fibronectin (Fn3). This technique has arisen as an alternative method for the development of specific binders of cell
surface proteins. Compared to antibodies, the Fn3 proteins are smaller (10kDa), thermally more stable, lack disulfide bonds and
have the ability to bind a large variety of proteins. Using error-prone polymerase mutagenesis, we identified improved Fn3
proteins with increased InsR affinity, specificity, and stability. Our current results show that InsR is an important target in
endocrine resistant breast cancer cells. Development of engineered Fn3 peptides with increased InsR binding specificity and
affinity for InsR compared to IGF1R could be used to disrupt signaling through this pathway. Engineered Fn3 will further be
evaluated as a potential imaging and therapeutic tool.

Title: Novel oncogene CNOT2 regulates proliferation and tumor angiogenesis targeting VEGF in MDA-MB-231 breast cancer
cells
Sunghoon Kim1, Eunjung Sohn1 and Deokbeom Jung1. 1Kyung Hee University, Seoul, Korea; 2Kyung Hee University, Seoul,
Korea and 3Kyung Hee University, Seoul, Korea.
Body: CCR-NOT (CNOT2) complex plays an important role in regulating the mRNA decay and translation repression. Here, we
investigated the role of CNOT2 in human breast cancers. We found that CNOT2 was overexpressed in breast, pancreas, or
prostate cancer cell lines and tissues. Proliferation of MDA-MB231 cells was suppressed by siRNA and shRNA mediated
silencing of CNOT2. Knockdown of CNOT2 by siRNA transfection in MDA-MB-231 breast cancer cells decreased cell motility of
MDA-MB231 cells, while overexpression of CNOT2 enhanced motility. Inhibition of CNOT2 significantly reduced clonogenicity,
while overexpression of CNOT2 enhanced. Furthermore, silencing of CNOT2 attenuated caner related genes such as cyclin D1,
VEGF, or IB as well as ERK, AKT in MDA-MB-231 breast cancer cells . Overall, our findings suggest that novel oncogene
CNOT2 promotes tumor proliferation and motility via enhancement of survival genes and VEGF in MDA-MB-231 breast cancer
cells.

Title: Targeting the insulin receptor with a small peptide (S961) in cancer
Kelly LaPara1 and Douglas Yee1. 1University of Minnesota, Minneapolis, MN.
Body: The insulin-like growth factor (IGF)/insulin receptors are members of the superfamily of growth factor receptor tyrosine
kinases. The insulin and type I IGF-I receptors (IGF-1R) have very similar structure and share high homology (84%) in the
kinase domain. Their dimeric structure allows for the formation of either holo or hybrid receptors. Thus, targeting of any individual
receptor has been a challenge. We have previously shown that the insulin receptor is expressed in tamoxifen resistant breast
cancer cell lines. Therefore, specific targeting might be a useful strategy to block this pathway. The antagonist S961 is a single
chain peptide that binds insulin receptor and has been reported to have partial antagonist activity. The affinity of S961 for the
insulin receptor is comparable to that of insulin and the selectivity of the insulin receptor versus IGF1 is higher than that of insulin
itself (Sturis et al). To determine if S961 has activity in breast cancer cells, we tested its ability to disrupt insulin and IGF-I
signaling and growth in a panel of cancer cell lines (MCF-7L, MCF-7L TamR, MDA-231, and HepG2 cells). MCF-7L cells express
high levels of IGF1, while HepG2 express mostly insulin receptor. Pre-incubation with S961 significantly suppressed p-Akt and
p-MAPK after insulin stimulation in the HepG2, but not in MCF-7L. In contrast, insulin receptor stimulation in MCF-7L TamR cells,
which do not express IGF1R, was inhibited by S961. Even a 1000-fold increase in S961 was unable to suppress insulin receptor
activation in cells expressing both insulin receptor and IGF1R. S961 inhibits insulin-stimulate cell cycle progression in MCF7L
TamR cells. In these TamR cells, S961 also suppresses colony formation in anchorage independent growth assays. These data
suggest S961 is an effective inhibitor of insulin receptor activation but only when little IGF1R is expressed. Thus, insulin receptor
targeting might be useful in the management of endocrine resistant breast cancer.

Title: Characterization of the constitutive and tumor microenvironment-mediated changes in the expression, localization and
function of CREB in HER-2/neu-imnduced tumor cells
Barbara Seliger1, Andre Steven1, Sandra Leisz1, Bernhard Hiebl2, Claudia Wickenhauser3, Chiara Massa1, Manuela Iezzi4, Davide
Bedognetti5, Francesco Marincola5 and Rolf Kiessling6. 1Institute of Medical Immunology, Halle, Germany; 2Center for Basic
Medical Research, Halle, Germany; 3Institute of Pathology, Halle, Germany; 4Center for Aging Sciences, Chieti, Italy; 5Sidra
Medical Research Center, Doha, Qatar and 6Karolinska Institutet, CCK, Stockholm, Sweden.
Body: Despite the cAMP responsive element-binding protein (CREB) is involved in tumorgenicity, a link between
HER-2/neu-induced transformation, CREB activation and alterations of the tumor micromilieu and cellular localization has not yet
been established, which might contribute to the pathogenesis of HER-2/neu-associated tumors including mammary carcinoma.
Using in vitro models of HER-2/neu-overexpressing and HER-2/neu-negative/silenced tumor cells as well as human mammary
carcinoma lesions with defined HER-2/neu status a correlation of HER-2/neu overexpression with an increased expression and
activation of CREB accompanied by an enhanced signal transduction and increased angiogenesis was detected both in vitro and
in vivo. Hypoxia revealed a HER-2/neu-dependent CREB activation, which was associated with an upregulation of HIF-1,
GLUT1 and VEGF expression and altered microenvironment such as enhanced cell migration and matrix
metalloproteinase-mediated invasion as well as increased extracellular lactate concentrations. The CREB-dependent hypoxic
response was further accompanied by ubiquitination and mitochondrial colocalization of CREB and an altered mitochondrial gene
expression. Furthermore, CREB downregulation in HER-2/neu-transformed cells by shRNA and by the treatment with the
inhibitors KG-501 and lapatinib caused morphologic changes, a reduced cell proliferation and G0/G1 cell cycle arrest, which could
be rescued by CREB expression. This was accompanied by a reduced cell migration, wound healing, an increased fibronectin
adherence, invasion and matrix metalloproteinase expression. In vivo shCREB HER-2/neu+ cells, but not control cells exerted a
significantly decreased tumorgenicity associated with a decreased proliferative activity, an enhanced apoptosis rate and an
increased frequency of T lymphocytes in peripheral blood mononuclear cells. Thus, CREB seems to play an important role in the
HER-2/neu-mediated transformation process of (mammary) tumor cells by altering their in vitro and in vivo growth characteristics
as well as their immunogenicity.

Title: Targeting the PI3K-AKT-mTOR and RAF-MEK-ERK pathways in HER2 amplified breast cancer models
Pradip De1, Yuliang Sun1, Jennifer Carlson1, Lori Friedman2, Nandini Dey1 and Brian Leyland-Jones1. 1Avera Research Institute,
Sioux Falls, SD and 2Genentech, San Francisco, CA.
Body: Background: The PI3K-AKT-mTOR and the RAF-MEK-ERK signaling pathways are critical for normal physiology and
also commonly dysregulated in several cancers including breast cancer. Inhibition of one pathway can still result in the
maintenance of signaling and tumor progression via the other reciprocal pathway. Here, we hypothesize that due to the existence
of such "escape" mechanisms a dual targeting of these pathways may lead to superior therapeutic activity in HER2+ breast
tumors. Methodology: This study evaluated the combination of a pan-PI3K inhibitor GDC-0941 and a MEK1/2 inhibitor
GDC-0973 in different HER2+ breast cancer cell lines. HER2+/trastuzumab (T)-sensitive (BT474), HER2+/trastuzumab-resistant
(BT474HerR) and HER2+/PIK3CA mutated (HCC1954) cells were used for this study. Growth inhibition/survival, apoptosis, signal
transduction and antitumor efficacy were examined using 3D-ON-TOP assays, annexinV staining and mouse xenografts
respectively, in T-sensitive, T-resistant and HER2+/PIK3CA mutated breast cancer cell lines. Results: 1) treatment with
GDC-0941 caused dose dependent inhibition of phosphorylation of AKT (S473, T308), P70S6K, S6RP, and 4EBP1 (T37/46) and
the combination of GDC-0941 plus GDC-0973 more effectively blocked p-S6RP and p-4EBP1, especially in BT474HerR and
HCC1954 cells, 2) GDC-0973 inhibited activation of ERK and this inhibition was sustained when combined with GDC-0941, 3) all
cell lines showed an increase in annexin V positivity (index of apoptosis) following the treatment of GDC-0941 alone and a
combination of GDC-0941 plus GDC-0973 more effectively induced apoptosis in HER2+ cell lines, 4) similar to anti-proliferative
and pro-apoptotic signals, a combination of GDC-0941 plus GDC-0973 more effectively abrogated anchorage-independent colony
formation in 3D-ON-TOP assay and 5) xenograft data exhibited that the combination of T plus GDC-0941 has an enhanced
anti-tumor effect in T-sensitive (80%), T-resistant (82%) and PIK3CA mutated models (58%), and most importantly, tumor
regression was more pronounced in all three models when mice were treated with T plus GDC-0941 plus GDC-0973.
Conclusions: Our findings suggest that simultaneous administration of MEK1/2 inhibitor may enhance the antitumor activity of
PI3K targeted drug and may delay the appearance of resistance of HER2-targeted therapy in HER2 amplified breast cancer.

Title: Prognostic value and clinicobiological associations of the EGFR / PI3K pathway in 204 consecutive localized sporadic triple
negative breast cancers
William Jacot1, Caroline Mollevi1, Anne-Claire Laberenne1, Nicolas Lozano1, Catherine Viglianti1, Frdric Bibeau1, Gilles Romieu1
and Pierre-Jean Lamy1. 1ICM Val d'Aurelle Montpellier Cancer Institute, Montpellier, France.
Body: Background: Triple negative breast cancer (TNBC) is characterized by the lack of expression of hormone receptors and
HER-2 antigen. This lack of drugs targets prompted the search for additional targeted therapies. The transmembrane EGFR
receptor, also known as HER1, and the phosphatidylinositol 3 phosphate kinase (PI3K) appears frequently dysfunctional in
TNBC. It thus appears important to simultaneously evaluate these biomarkers of interest in order to better characterize this
population of tumors and the opportunity of dedicated biomarkers-oriented targeting in a European population of TNBC patients.
Material and methods: A total of 1,695 consecutive patients with breast cancer referred to the Val dAurelle Cancer Institute
between 2002 and 2010 were prospectively entered into the database of a tumour DNA and cytosol biobank. 204 cases with a
localized TNBC were selected for the present study.
PIK3CA PCR amplification and high-resolution melting (HRM) analysis designed to span exon 9 and 20 mutations were
performed on a Rotor-Gene 6000 using the LightCycler 480 HRM PCR Master Mix kit. Bidirectional sequencing was
performed by using PCR primers and an Applied Biosystems 3730xl DNA Analyzer.
Regarding HER1 CNV determination, qPCR reactions were performed on an ABI Prism 7700 sequence detection apparatus.
HER1 levels were normalized to ACBT and GAPDH expression. Measurements were performed in duplicate. The data were
expressed as the HER1/ACBT or HER1/GAPDH relative copy number ratio. EGFR cystosol quantification was performed with
EGFR Elisa Kit (Calbiochem). Proteins were quantified with the pierce method. Results were expressed in nmol/mg of protein.
Results: Regarding the PIK3CA gene, we identified 14 (6.9%) mutations in exon 9 and 17 (8.3%) mutations in exon 20. Exon 9
mutations were associated with SBR grade I-II (p=0.04) and exon 20 mutations were associated with size 3-4 (p=0.03). Overall
mutations were only associated with SBR I-II (p=0.05). HER1 gene was deleted in 11 cases (5.3), normal in 154 cases (75.9%),
amplified in 18 cases (8.9%) and presented a polysomy in 20 cases (9.9%). HER1 amplification was clearly associated with an
increased EGFR protein content (p=0.03), while polysomic cases presented an EGFR content non-significantly different of the
normal HER1 cases. After a median follow-up of 5.6yrs, the 5-year Disease Free Survival ( DFS) rate was 76.1% (95% CI [69.2;
81.6]) and the 5-year Overall Survival (OS) rate was 81.3% (95% CI [74.8; 86.3]). PI3KCA exon 9 mutations were a significant
pejorative prognostic factor in univariate and multivariate analysis, while a high (upper quartile) EGFR protein content was
associated with a better prognosis, in uni- and multivariate analysis.
Conclusions: In this large cohort of localized TNBC, 15.2% of patients presented mutations in both exon 9 or 20 of PI3KCA, and
8.9% of the tumors presented HER1 amplification. Our study shows the negative prognosis value of exon 9 PI3KCA mutations.
We confirm then the clinical importance to detect PI3KCA mutations in TNBC. In addition, a high EGFR tumoral content was
associated with a better prognosis. The exact mechanism of this association warrant further studies.

Title: Growth hormone induction of oncogenic signaling promotes survival of endocrine resistant breast cancer cells
Heather C Beckwith1 and Douglas Yee1. 1University of Minnesota, Minneapolis, MN.
Body: Targeting endocrine resistant breast cancer cells has been successfully accomplished by the use of mTORC1 inhibitors in
combination with either aromatase inhibitors or tamoxifen. While there was preclinical data suggesting that targeting of the type I
IGF receptor (IGF1R) might also be a target for endocrine resistant cells, clinical trials using IGF1R monoclonal antibodies
showed no advantage and even suggested potential harm for the combination. IGF1R inhibition results in the disruption of an
endocrine feedback loop and results in elevation of the pituitary growth factor growth hormone (GH). GH is involved in normal
human growth, development, metabolism, and longevity. Growth hormone released from the pituitary stimulates production of
IGF-I from the liver. The GH/IGF-1 axis is known to promote mammary gland hyperplasia and breast cancer carcinogenesis.
Previous studies have shown that GH promotes carcinogenesis independently of insulin-like growth factor. The role of GH in
specific subtypes of breast carcinoma remains to be defined. Our laboratory has found that estrogen receptor positive breast
cancer cell lines up regulated oncogenic STAT5, Akt, and MAPK signaling pathways in response to GH. Tamoxifen resistant
(TamR) and long-term estrogen deprived (LTED) breast cancer cell lines derived from MCF-7 and T47D cells demonstrated
increased oncogenic signaling in response to GH compared to the corresponding parental cell lines from which they were
derived. In contrast, GH stimulation of a triple negative breast cancer cell line (MDA-MB-231) failed to induce signaling via these
oncogenic pathways. Although GH treatment of parental, TamR, and LTED breast cancer cells had minimal effect on cell
proliferation, cell cycle progression, or migration; GH signaling protected cells from cytotoxic chemotherapy induced apoptosis.
This data shows that GH signaling is found in endocrine sensitive and resistant cells. Furthermore, GH may function in protecting
cells from cell death induced by either endocrine or cytotoxic drugs. GH receptor blockade may be a valid treatment option for
endocrine resistant breast cancer.

Title: Expression of growth hormone releasing hormone receptor ( GHRH-R ) in primary and metastatic mammary carcinomas
Mehrad Nadji1, Norman L Block1, Andrew V Shally1, Jonathan Lara2, Rick Michaelson2, Suhail M Ali3,4, Kim Leitzel3, Syed M Rizvi3,
Mhd Yaser Al-Marrawi3 and Allan Lipton3. 1University of Miami, Miami, FL; 2Saint Barnabas Medical Center, Livingston, NJ; 3Penn
State Hershey Medical Center, Hershey, PA and 4Lebanon VA Medical Center, Lebanon, PA.
Body: In addition to its nominative function as a neurohormone acting on the pituitary, Growth Hormone Releasing Hormone
(GHRH) has been shown to modify the growth behavior of numerous cancers, including breast. GHRH is produced by tumor
cells, acts in an autocrine/paracrine manner, and requires the presence of GHRH receptor (GHRH-R) on the tumor cells to exert
its effects. As this work has been done predominantly on tumor cell lines and xenografts, we set out to examine the clinical
analog.
Matched primary and metachronous metastases from 50 breast cancers were included in this study of GHRH-R in breast cancer.
Immunohistochemistry for GHRH-R (AbCam) was performed on paraffin sections and the staining results were assessed semi
quantitatively from 0 (negative) to 3+ (strongly positive). A section from normal pituitary was used as positive control. Forty-three
of the primary breast cancers (86%) that ultimately relapsed or metastasized showed moderate to strong immunohistochemical
expression of GHRH-R. Tumors from the metastatic foci also showed strong immunoreactivity in 78%, 71%, and 44% of liver,
brain, and bone foci, respectively. The lower intensity of staining in bone samples may be due to the effect of the decalcification
process routinely performed before staining. Whenever present, the non-neoplastic glands adjacent to breast tumors showed
either negative or 1+ reaction for GHRH-R. We conclude that the great majority of mammary carcinomas at primary and
metastatic sites express GHRH-R. This finding could potentially serve as a basis for therapeutic approaches using peptide
receptor antagonists. By receptor blockade using GHRH receptor antagonists, with minimal pharmacologic side-effects, we have
been able to control the growth in a number of tumor cell lines (HCC1806, MDAMB468, MDAMB435S, MCF7, T47D, HCC1937,
BT474, and MX1s). Based on the presence of GHRH receptors in these clinical tumors, we conclude that clinical trials evaluating
GHRH receptor antagonists are indicated.

Title: Calorie restriction normalizes global microRNA expression by preventing the loss of dicer expression during mammary
tumorigenesis
Kaylyn L Devlin1, Tiffany Sanford2, Elizabeth Mambo2 and Stephen D Hursting3. 1University of Texas, Austin, TX; 2Asuragen, Inc,
Austin, TX and 3University of North Carolina, Chapel Hill, NC.
Body: MicroRNA (miRNAs) are small, post-transcriptional regulators that play an integral role in maintenance of cellular functions
and whose dyregulation has been shown to promote many types of cancer, including breast cancer. Global downregulation of
miRNAs has emerged as a common theme in human breast tumors and has been shown to contribute to oncogenesis. One of
the primary mechanisms through which miRNAs are globally dysregulated is downregulation of enzymes involved in miRNA
biogenesis. We aim to establish how calorie restriction (CR), which potently inhibits breast cancer progression, regulates global
miRNA expression and miRNA biogenesis enzyme functionality in mammary tumors. To address these questions, 100 female
Sprague Dawley rats were administered either dimethylbenz(a)anthracene (DMBA) or vehicle control at 50 days of age, then
randomized to receive either control (AIN-76A) diet ad-libitum (n=40) or a 30% CR diet regiment (n=60). Resultant mammary
tumors were allowed to develop for 12 weeks. Calorie restriction significantly increased survival to study endpoint relative to
control diet (75% vs 35%, respectively) (p=0.0047). Furthermore, of the animals that developed tumors, CR significantly
decreased median tumor area by 56% compared to control diet (109.4 mm2 vs 250.9 mm2, respectively) (p=0.0286). Global
miRNA expression was analyzed through miRNA-specific sequencing. Calorie restriction had a broad effect on miRNA
expression, illustrated by the fact that of all the miRNAs with a greater than two-fold expression difference between CR and
control, 80% are overexpressed in CR tumors compared to control tumors. These results can be explained by the additional
finding that CR was able to prevent the loss of Dicer expression, a key miRNA biogenesis enzyme, observed in control-fed
mammary tumor tissue compared to normal tissue. This important finding suggests that global miRNA normalization through the
retention of Dicer expression during cancer progression could be a contributing mechanism to CRs anticancer effects. We plan to
further these investigations by exogenously manipulating Dicer expression in Rama25 cells, which were originally derived from a
DMBA-induced mammary tumor from a Sprague Dawley rat, and analyzing the resultant tumorigenic potential in vitro and in vivo.
The results obtained will provide insights into the mechanisms of breast cancer progression and how CR inhibits progression
through microRNA modulation.

Title: A large integrated-gene profiling analysis identifies prognostic microRNAs and correlated DNA repair genes in estrogen
receptor positive and negative breast cancers
Balazs Gyorffy12, Mate Kormos1, Giulia Bottai3, Kristine Kleivi4, George A Calin5, Anne-Lise Borresen-Dale4 and Libero Santarpia3.
1
MTA TTK Lendlet Cancer Biomarker Research Group, Budapest, Hungary; 2Semmelweis University, Budapest, Hungary;
3
Humanitas Clinical and Research Institute, Rozzano, Milan, Italy; 4Institute for Cancer Research, Oslo University Hospital,
Radium Hospitalet, Oslo, Norway and 5University of Texas MD Anderson Cancer Center, Houston, TX.
Body: Background: MicroRNAs (miRNAs) are small non-coding RNAs involved in the pathogenesis of breast cancer (BC).
Estrogen receptor (ER) is one of the most important factors influencing BC outcome. In this study we aim to assess the
prognostic value in a cross-project analysis for all known miRNAs in ER-positive (ER+) and ER-negative (ER-) BC separately.
Materials and Methods: We assessed the prognostic value of miRNA expression in three independent BC datasets: TCGA
(n=782), Metabric [M] (n=1,293) and GSE40267 (n=181) including 472, 853, and 475 miRNAs, respectively. Overall survival (OS)
data was available for all patients. Statistical analysis was performed with R software. Survival analysis was performed using Cox
proportional hazards regression in each dataset separately. To define an mRNA expression fingerprint for each miRNA,
correlation between miRNA and mRNA expressions was computed using Spearman rank correlation in the Metabric dataset.
Subsequently, specific metagenes were established using significant correlated mRNAs ("Metabric mRNA", M-mRNA) for each
miRNA. Metagenes were also used in the Cox regression analysis. The entire analysis was performed for each miRNA in
ER-positive and ER-negative BCs separately.
Results: In the ER+ BC, 283 miRNAs reached statistical significance in at least one dataset. The best performing and significant
miRNAs in all datasets were miR-195 (M: HR=1.73, p=3.5E-06; TCGA: HR=2.17, p=3.5E-06, M-mRNA: HR=0.54, p=2.0E-10),
miR-199b (M: HR=1.74, p=3.8E-06; TCGA: HR=0.46, p=0.028, M-mRNA: HR=0.56, p=2.8E-09), and miR-210 (M: HR=0.7,
p=0.003; TCGA: HR=0.45, p=0.014, Metabric mRNA: HR=1.34, p=0.002). These 3 miRNAs retained their prognostic significance
also in a multivariate analysis paired with HER2 status, lymph node status, size, grade and MKI67 expression. In the ER- group
73 significant miRNAs were identified. The best performing miRNAs in at least two datasets were miR-155 (GSE40267: HR=2.21,
p=0.034, M: HR=1.67, p=0.008) and miR-381 (GSE40267: HR=0.41, p=0.026; M: HR=0.67, p=0.038). However, the miRNAs in
ER- BC in a multivariate analysis were not independent from tumor size. The metagenes identified were significantly enriched in
DNA repair genes (e.g. AURKB, TEK) and exhibited an inverse correlation to survival as compared to miRNA expression.
Conclusions: This is the first large-scale miRNA expression analysis demonstrating different prognostic miRNAs in ER+ and ERBCs. Correlated miRNA-metagenes are involved in the regulation of DNA repair and genomic stability.

Title: MicroRNAs correlating with outcome in patients treated with first-line bevacizumab for metastatic breast cancer
Simon P Gampenrieder1, Frank Hamacher1, Gabriel Rinnerthaler1, Clemens Hufnagl1, Hubert Hackl2, Franz Romeder1, Claudia
Mu1, Cornelia Hauser-Kronbeger3, Brigitte Mlineritsch1 and Richard Greil1. 1Salzburg Cancer Research Institute with Laboratory
of Immunological and Molecular Cancer Research and Center for Clinical Cancer and Immunology Trials, Paracelsus Medical
University, Salzburg, Austria; 2BioCenter, Innsbruck Medical University, Innsbruck, Austria and 3Paracelsus Medical University,
Salzburg, Austria.
Body: Background: Biomarkers predicting response to bevacizumab containing therapy in breast cancer are of urgent need.
MiRNAs, small non-coding RNAs, can be involved in tumor evolution including regulation of angiogenesis and development of
treatment resistance. Therefore, miRNAs can provide both prognostic and predictive information in different cancer entities.
Patients and methods: A genome-wide miRNA profiling using high-throughput TaqMan Array Human MicroRNA Cards
enabling quantification of 754 unique human miRNAs was performed. Formalin-fixed paraffin-embedded specimens from 60
patients treated with bevacizumab as first-line therapy at our institution were analyzed. Based on the median overall
progression-free survival (PFS) patients were divided into a responder (G1) and a non-responder group (G2). Differentially
expressed miRNAs were selected considering a more than two-fold change and a false discovery rate (FDR) < 10% as
significant. Further interesting miRNAs were selected by a multivariate logistic regression approach using LASSO (Least Absolute
Selection and Shrinkage Operator) regularization.
Results: Overall median PFS was 9.3 months with a median PFS of 17.5 and 5.0 months in G1 and G2, respectively. Eight
miRNAs (miR-19b-3p, miR-21-5p, miR-9-5p, miR-590-5p, miR-106b-5p, miR-20a-5p) were significantly differentially expressed
between these groups (FDR < 10%). Their expression levels were all negatively associated with G1. Additionally, four miRNAs
(miR-210-3p, miR-224-5p, miR-155-5p, miR-28-5p) might be interesting predictive biomarkers as they were included in a logistic
regression classification model providing an optimal separation between responder and non-responder group (evident from the
area under curve [AUC] and 100 x 5-fold cross validation). Currently these data is being validated in a patient cohort not treated
with bevacizumab in order to detect the predictive value of these miRNAs for a bevacizumab-based treatment. These validated
results will be presented at the meeting.
Conclusion: Twelve miRNAs in breast cancer tissue could be identified showing promising predictive value for
bevacizumab-based therapy. Although these data need to be confirmed, differences in miRNA expression could help identifying
patients with greater benefit from anti-VEGF agents.

Title: Immunohistochemical determination of arylamine N-acetyltransferase 1 (NAT1) as the target of miR-1290 and prognostic
biomarker of breast cancer
Yumi Endo1, Hiroko Yamashita2, Satoru Takahashi3, Shinya Sato3, Nobuyasu Yoshimoto1, Tomoko Asano1, Yukari Hato1, Mina
Yamaguchi1, Yu Dong1 and Tatsuya Toyama1. 1Immunology and Surgery, Nagoya City University Graduate School of Medical
Sciences, Nagoya, Aichi, Japan; 2Breast and Endocrine Surgery, Hokkaido University Hospital, Sapporo, Hokkaido, Japan and
3
Nagoya City University Graduate School of Medical Sciences, Nagoya, Aichi, Japan.
Body: Introduction: There are large-scale molecular differences between estrogen receptor (ER)-positive breast cancer and
ER-negative breast cancers. In ER-positive breast cancer, recent analyses have shown that ER-positives can be divided into
two subtypes such as luminal A and luminal B. These subtypes differ in characteristics such as response to endocrine therapy
and chemotherapy, and prognosis. In a previous study, we identified a microRNA, miR-1290, that was significantly
down-regulated in luminal A tumors and its potential target arylamine N-acetyltransferase 1 (NAT1). The aim of this study was to
clarify whether NAT1 is a bona fide target of miR-1290, and to investigate the impact of NAT1 on breast cancer prognosis.
Methods: Luciferase reporter assay was employed to validate NAT1 as a putative miR-1290 target gene. Protein expressions of
NAT1, ER, progesterone receptor (PgR) and HER2 were analyzed in 394 breast cancer samples by immunohistochemistry.
Results: NAT1 was shown to be a direct target of miR-1290 by luciferase reporter assay. Expression levels of NAT1 were
positively correlated with expression levels of ER (P < 0.0001) and PgR (P < 0.0001), whereas expression levels of NAT1 were
negatively correlated with both tumor grade and tumor size (P < 0.0001). Kaplan-Meier analysis showed that NAT1 presence was
significantly associated with increased overall survival (OS) (P = 0.0416) in breast cancer patients (n=394). Similarly, significant
associations of NAT1 presence were shown with disease-free survival (DFS) (P = 0.0048) and OS (P = 0.0055) in patients who
received adjuvant endocrine therapy with tamoxifen (n = 176). Moreover, NAT1 presence was also significantly associated with
increased DFS (P = 0.0025) and OS (P = 0.0007) in lymph node-positive breast cancer patients (n=147). Univariate and
multivariate analyses showed significant associations between expression levels of NAT1 and DFS (P = 0.0005 and 0.019,
respectively). Conclusion: We reported that miR-1290 directly targets NAT1 3-UTR and NAT1 protein expression is correlated
with improved OS of breast cancer. Moreover, NAT1 is a possible prognostic biomarker for lymph node-positive breast cancer.
Thus, miR-1290 and its potential target NAT1 are associated with characteristics of breast cancer.

Title: Exosome-mediated trafficking of microRNAs by breast cancer cells
Doireann P Joyce1, Claire L Glynn1, James Brown1, Emma Holian1, Peter Dockery1, Michael J Kerin1 and Roisin M Dwyer1.
1
National University of Ireland Galway, Galway, Ireland.
Body: Introduction
Cellular communication in the primary tumour micro-environment is known to play a key role in tumour development and
progression. Exosomes are membrane-derived nanovesicles that are actively secreted by cells. Exosomes have been implicated
in cell-to-cell communication through the transfer of genetic material including messenger RNA (mRNA), and more recently
microRNA (miRNA). miRNAs are small non-coding RNAs approximately 22 nucleotides in length. They play an important role in
posttranscriptional regulation of gene expression. Recent reports suggest exosome-mediated trafficking of microRNAs between
cells. The aim of this study was to identify the panel of exosome-encapsulated microRNAs secreted by breast cancer cells in
vitro.
Methods
Four breast cancer cell lines, MDA-MB-231, BT-20, Sk-Br-3 and T47D were cultured in exosome-depleted media for 48 hours.
Exosomes secreted by the cells were isolated through a process of differential centrifugation, microfiltration and
ultracentrifugation. The presence of exosomes was first confirmed by Transmission Electron Microscopy (TEM) and Western Blot
analysis. Global miRNA array analysis of RNA extracted from exosomes was performed to identify the panel of miRNAs secreted
by these cells. miRNA targets of interest were further validated using relative quantification PCR (RQ-PCR). Transfer of Red
fluorescent protein (RFP)-labelled exosomes between cell populations was visualized using Confocal microscopy.
Results
TEM analysis of secreted exosomes revealed vesicular bodies of 40-100nm in size. Immunoblotting confirmed the presence of
the exosome-associated protein CD63. MicroRNA array analysis of exosome fractions targeting 2089 miRNAs, revealed
secretion of between 324 and 394 miRNAs by the cell lines. The miRNAs appeared to cluster in a biologically relevant fashion.
282 miRNAs were common to exosomes from all 4 cell lines, while a small selection were specific to individual cell lines.
Exosome-mediated trafficking of miRNA targets of interest including miR-451a and miR-744-5p was successfully validated using
RQ-PCR. Further validated targets included miR-10b, miR-145, miR-492 and miR-498, all of which have established roles in
regulation of proliferation and apoptosis in cancer. Confocal microscopy supported visualization of miRNA-enriched exosome
release from donor cells and subsequent uptake of the RFP-labelled exosomes by recipient cells.
Conclusions
A distinct panel of miRNAs are actively and selectively packaged into exosomes and secreted by breast cancer cells. This
transfer of functional miRNAs between cells may play an important role in intercellular communication in the primary tumour
microenvironment. The data presented also has potentially important implications in the identification of a circulating miRNA
signature for breast cancer detection.

Title: miR-9 expression, retinoids and their potential role in trastuzumab resistance
Karen Howe1, Brigid C Browne1, Sinead Aherne1, John Crown2 and Norma O'Donovan1. 1Molecular Therapeutics for Cancer
Ireland, National Institute for Cellular Biotechnology, Dublin City University, Dublin, Ireland and 2St Vincent's University Hospital,
Ireland.
Body: HER2 positive breast cancer accounts for approximately 20-25 % of breast cancers. Trastuzumab, a humanised
monoclonal antibody, is approved for the treatment of HER2 positive breast cancer. However, the majority of metastatic HER2
positive breast cancers progress on trastuzumab treatment due to either innate or acquired resistance. The aim of this study is to
investigate the role of microRNAs (miRNAs) in a cell line model of acquired trastuzumab resistance.
The trastuzumab resistant cell line, SKBR3-T was established by continuous exposure to trastuzumab (10 g/ml) for 6 months.
miRNA extracted from SKBR3 and SKBR3-T was profiled using Taqman low density arrays (TLDA). Differentially regulated
miRNAs were selected using >2-fold change and a P-value of <0.05. Individual quantitative RT-PCR (qRT-PCR) was performed
to confirm miRNA alterations. Proliferation assays were performed using the acid phosphatase method. Taqman gene expression
assays for retinoic acid receptor alpha (RARA) were performed using GAPDH as an endogenous control. Immunodetection of
RARA was performed using -tubulin as a control.
Six differentially regulated miRNAs were identified in the SKBR3-T cells. qRT-PCR assays confirmed that four were significantly
altered in SKBR3-T compared to SKBR3 cells including miR-9 which was 2.2 fold up-regulated (p=0.04).
Utilising miRWalk, we identified RARA as a target for miR-9. We confirmed that RARA mRNA and protein expression are reduced
in SKBR3-T cells compared to SKBR3 cells. Treatment with all-trans retinoic acid (ATRA) (0.2 M0.025 M) alone inhibited
growth of the SKBR3 cell line in a dose dependent manner but not in the SKBR3-T cell line. Combined treatment with
trastuzumab (10 g/uL) and ATRA (0.2 M) had a significantly greater growth inhibitory effect on the SKBR3 cell line (90.15.2
%) (p=0.0002) than either trastuzumab (39.14.0 %) (p= 0.0004) or ATRA alone (57.94.2 %) (p=0.001). Interestingly, despite
relative insensitivity to ATRA in the SKBR3-T cells (23.75.0%) (p=0.86), combined treatment with trastuzumab produced
significant growth inhibition in the SKBR3-T cells (74.25.1%) compared to either trastuzumab (24.14.4%) (p=0.0001) or ATRA
(23.75.0%) (p=0.86) alone.
miR-9 is up-regulated in the acquired trastuzumab resistant SKBR3-T cell line, RARA expression is downregulated, and SKBR3-T
cells are resistant to ATRA treatment compared to SKBR3 cells. However, SKBR3-T cells are sensitive to combined treatment
with trastuzumab and ATRA. Thus combined treatment with ATRA and trastuzumab may overcome acquired resistance to
trastuzumab in HER2 positive breast cancer.

Title: A MAPK microRNA signature significantly associated with poor outcome and predictive of response to tamoxifen therapy in
ER+ breast cancer
Philip Miller1, Jennifer Clarke2 and Dorraya El-Ashry1. 1University of Miami Miller School of Medicine, Miami, FL and 2University of
Nebraska, Lincoln.
Body: Background: Hyperactivation of ERK1/2 MAPK (hMAPK) leads to loss of estrogen receptor (ER) expression and poor
outcome in breast cancer. Recent evidence suggests that microRNAs (miRNAs) play important regulatory roles and serve as
biomarkers of disease. We previously reported a miRNA signature indicative of hyperactivation of MAPK signaling in breast
cancer and its associations with pathological features of breast cancer and breast cancer clinical outcomes. This hMAPK-miRNA
signature identified ER+ breast cancers with reduced recurrence-free and overall survival. Here we report on a leave-one-out
analysis of this hMAPK-miRNA signature to narrow down and identify those miRNAs that may be critically important to poor
clinical outcomes associated with hyperactive MAPK signaling among patients with ER+ breast cancer.
Methods: We performed a leave-one-out analysis of the full hMAPK-miRNA signature, wherein we determined the impact that
removal of individual members of the signature had on the ability of the hMAPK-miRNA signature to predict poor disease survival
in patients from the METABRIC breast cancer dataset with ER+ cancers. We identified a subsignature of 21 miRNAs that is very
significantly associated with poor clinical outcome in patients with ER+ breast cancer from a large patient cohort, according to
multivariate Cox proportional hazard analysis.
Summary of Results: Of 57 miRNAs in our original hMAPK-miRNA signature, 21 were retained following leave-one-out analysis.
"High-hMAPK" status as described by this signature significantly correlated with adverse pathological characteristics of breast
cancer, including ER-negativity, increased tumor grade, and enrichment for Basal and HER2 subtypes. Kaplan-Meier survival
analysis of primary breast cancers from the METABRIC dataset indicate that the 21 miRNA subsignature improves upon the
prognostic capability of the overall signature in this large patient cohort, and particularly in providing prognostic capability in breast
cancers of the luminal-A and luminal-B molecular subtypes. Furthermore, multivariate Cox proportional hazards analysis suggests
that classification of breast cancers as "high-hMAPK" based on this leave-one-out miRNA signature is an independent risk factor
for poor disease outcome, and is a more significant indicator of poor disease status than hormone receptor status, tumor grade,
molecular subtype, and HER2 status. Patients with ER-positive breast cancer who were treated with hormone therapy and
classified as "high-hMAPK" by this 21 miRNA signature displayed significantly poorer disease survival compared to patients
classified as "low hMAPK". Kaplan-Meier survival analysis and multivariate analysis of independent breast cancer cohorts with
miRNA expression data confirm these observations.
Conclusions: We report a subset of 21 of the hMAPK-miRNAs that are important prognostic factors in ER-positive breast cancer,
and that may have predictive value in estimating whether an ER-positive breast cancer may be resistant to hormone therapy.
Additionally, these 21 miRNAs may be potential effectors of MAPK signaling, and could serve as novel biomarkers or therapeutic
targets in breast cancer.

Title: Circulating miRNAs as surrogate markers for hormonosensibility in patients with hormone receptor-positive metastatic
breast cancer? A pilot study
Florence Dalenc1, Monia Ouali2, Thomas Filleron2, Laurence Gladieff1, Henri Roch1, Jean-Louis Lacaze1, William Jacot3, Anna
Durigova3, Anne Pradines4, Pauline Bringer4, Magali Farella4 and Gilles Favre4. 1Claudius Regaud Institut, Toulouse, France;
2
Claudius Regaud Institut, Toulouse, France; 3Institut of Cancer of Montpellier, Montpellier, France and 4Claudius Regaud Institut,
Toulouse, France.
Body: Backgroud: Clinicians need new predictive biomarkers of response to therapy hormonal in patients with hormone
receptor-positive (HR+) breast cancer (BC). Tumor-associated miRNAs are interesting new markers. Several data indicate the
extensive alterations in miRNAs regulation upon estrogen pathway and suggest the utility of considering miRNAs expression in
the understanding of hormonal therapy efficacy.
Methods: We have conducted a bicentric, prospective clinical trial in patients who must received an anti-estrogen (Tamoxifen) or
an aromatase inhibitor (+/- LHRH agonist) in first line for a HR+ metastatic BC. Plasma of patients were collected before the first
administration of hormonal therapy (T=0) and after 1 (T=1), 3, 6 months and therafter in case of objective response or/and
progression disease. After extraction from plasma expression i) of 372 miRNAs was analyzed on microRNA Ready-to-Use PCR,
Human panel I, V3 from Exiqonand ii) of a selection of candidate miRNAs described in the litterature to be associated with
estrogen pathway and hormonotherapy response in tissues were analyzed using the BioMark 96.96 Dynamic Array (Fluidigm
Corporation) with Exiqon primers. The primary end point of our study, was the feasibility of detection of circulating miRNAs as
biomarkers in plasma of patients. Key secondary end points were to compare i) the concentration of determinated miRNAs
according patients, under therapy hormonam and according the efficacy or not ii) the profile of miRNAs between patients at T=0 +
T=1 month and in the same patient under treatment.
Results : From March 2012 to January 2014, 39 patients were enrolled (5 under tamoxifen and 34 under aromatase inhibitor). At
first concentration of circulating miRNAs in pools of patient plasmas from T=0 and T=1 was compared by RTqPCR. miRNAs with
a significative fold change ratio together with several miRNAs from the litterature were then further analyzed for individual patient
plasmas at T=0 and T=1. Several miRNAs and notably miRNAs previously described in breast tumors to be associated with
estrogen pathway and hormonotherapy response are well-detected in plasma and, some of them show difference between T=0
and T=1. After univariate and multivariate analysis, the plasma miRNAs significantly associated with a therapy response will be
next examined at 3 and 6 months.
Conclusion : This pilot study provides that tumor-associated circulating miRNAs could be measured in the plasma of patients
and that alterations in miRNAs concentration upon hormonal therapy could be observed. More results will be presented at the
SABCS meeting.

Title: Breast cancer specific expression of miRNA deciphered using next generation sequencing of LCM procured cells
Kelli S Bramlett1, James L Wittliff2, Jose G Cienfuegos1, Sarah A Andres2 and Jeoffrey J Schageman1. 1Thermo Fisher Scientific,
Austin, TX and 2University of Louisville, Institute for Molecular Diversity & Drug Design, Louisville, KY.
Body: The objective of this study is to decipher miRNA expression profiles of laser capture microdissection (LCM)-procured
carcinoma cells compared to those of intact serial sections of a breast cancer biopsy. Our hypothesis is that miRNA signatures
discerned from specific carcinoma cell populations more precisely correlate with clinical behavior than that provided by
conventional biomarkers of intact tissue biopsies. De-identified frozen biopsies of invasive ductal carcinomas of known grade and
biomarker status containing 35-70% cancer were selected from an IRB-approved Biorepository (JLW). Serial tissue sections were
stained with H & E and 12-15,000 carcinoma cells were collected from an adjacent section. RNA was extracted using PureLink
RNA Mini Kit (Invitrogen), evaluated (Agilent Bioanalyzer) and sequenced for miRNA expression using the Ion Torrent
System (Thermo Fisher). Total RNAs were enriched for small RNA species using mirVana miRNA kits (Thermo Fisher) and
RNA libraries were constructed from 5 ng of enriched RNA using Ion Total RNA-Seq Kit v2. Barcodes were utilized to multiplex
libraries for template preparation and sequencing on Proton PI chips as twelve-plex library pools. Each library was sequenced
to an average of 30M reads on the Ion Protonsequencer with the Ion PI chip. Sequence reads were aligned to miRNA
precursor hairpins available from the miRBase (miRBase.org) miRNA repository. Aligned reads to each miRBase reference
miRNA were then reported. Using the R statistical software package, DESeq (Bioconductor), counts for all libraries were
normalized and relative expression was calculated. Mapping statistics (e.g., aligned reads (range 59-79%) and miRBase matches
(range 69-85%)) were assessed for each library. Comparison of expressed miRNAs from intact tissue sections with those of
cognate carcinoma cells procured by LCM revealed, in general, that smaller defined miRNA gene sets were expressed in isolated
populations of carcinoma cells. miRNA expression patterns of experimental pairs (intact vs LCM-procured) using MA-plots were
highly variable in carcinomas with different grades, suggesting a relationship to disease status. Gene frequency plots, comparing
expression from intact tissue sections to that of LCM-procured cell population, revealed subsets of differently expressed miRNAs.
To increase statistical power, a follow-up experiment was performed with triplicate libraries from 4 different representative
carcinoma samples. In addition to miRNA sequencing, targeted RNA sequencing with an Ion AmpliSeq RNA panel was used to
capture gene expression information from the12 additional samples. From these replicated libraries, we are able to combine
mRNA and miRNA expression information to create an expected profile from these breast carcinoma tissue samples. Application
of Next Generation Sequencing of miRNAs and Ion AmpliSeq RNA panels using LCM-procured cells and intact tissue provides
an innovative approach for assessing differential expression of miRNA and mRNA levels involved in breast cancer behavior.
Supported in part by a grant from the Phi Beta Psi Charity Trust (JLW & SAA) and a CTSP Award from the Commonwealth of
Kentucky (JLW). For research use only.

Title: MiRNAs are important regulators of Pax-5 expression and function during breast cancer progression
Jason M Harquail1,2 and Gilles A Robichaud1,2. 1Universit de Moncton, Moncton, NB, Canada and 2Atlantic Cancer Research
Institute, Moncton, NB, Canada.
Body: Recent studies have enabled the identification of important factors regulating cancer progression, one of these being the
Pax-5 gene. Pax-5, an essential developmental factor of B cells, is aberrantly expressed in various B cell cancer lesions and solid
tumors such as breast carcinoma. Although Pax-5 downstream activity is relatively well characterized, the regulation of aberrant
Pax-5 expression in a cancer specific context is poorly understood. To investigate the regulation of Pax-5 expression, we turned
our attention to micro-RNAs (miRNAs). MiRNAs are highly conserved, small non-coding RNA molecules that regulate key
biological processes. Extensive studies also show their deregulation in multiple cancer lesions. In this study, we aim to elucidate a
causal link between differentially expressed miRNAs in cancer cells and their putative targeting of Pax-5-dependent cancer
processes. With the help of biobank data and bioinformatics analyses, we observe that miRNAs 484 and 210 are aberrantly
expressed in breast cancer cells and cross-reference with their predicted capacity to target the Pax-5 mRNA 3 untranslated
region (3UTR). Using anti- or pre-miRNAs transfected into Pax-5 expressing breast cancer cell lines (MCF-7 and MB231), we
demonstrate that miRNAs 484 and 210 are capable of regulating Pax-5 expression. In addition, miRNA-regulated Pax-5
expression resulted in a concomitant alteration in Pax-5-mediated phenotype and cancer processes. This is the first study
demonstrating the regulation of Pax-5 expression and function by non-coding RNAs in cancer cells. We believe that the aberrant
expression of Pax-5 in cancer cells is in part due to deregulated miRNA expression profiles. This study will bring insight in regards
to cancer regulating processes associated with miRNA and Pax-5 deregulations and help us better understand aberrant Pax-5
expression levels within cancerous states. This study can therefore provide the eventual possibility of earlier, more efficient
diagnostics as well as more targeted treatments for cancer patients.

Title: No Show

Title: Re-introduction of tumor suppressor miR-34a harbors therapeutic efficacy in triple negative breast cancer
Brian D Adams1, Lajos Pusztai1, David L Rimm1 and Frank J Slack1. 1Yale University, New Haven, CT.
Body: Place Holder Abstract
Date Expected for Obtaining Final Information for Abstract: 9/14 - 10/14
Triple negative breast cancer accounts for a disproportionate share of the total breast cancer morbidity because of its aggressive
behavior, increased incidence in younger women, and lack of effective targeted therapies. MicroRNAs are global regulators of
survival and proliferation pathways important in cancer development and tumor maintenance. MicroRNAs function as tumor
suppressors and oncogenes, and are highly dysregulated in cancer. Here, we identified miR-34a to be aberrantly lost in triple
negative cancer cell lines when compared to both a luminal cancer subtype as well as normal breast cells. Re-introduction of
miR-34a in triple negative cancer lines results in inhibition of cell proliferation, reactivation of senescence, and enhances
sensitivity to apoptotic-inducing agents. Furthermore, intratumoral delivery of miR-34a into subcutaneously implanted tumors in
nude mice delays tumor growth as compared to a scrambled control. In conclusion re-introduction of miR-34a in triple negative
breast cancer promotes anti-tumorigenic phenotypes both in vitro and in vivo, and could therefore be used as a therapeutic agent
to treat the disease.

Title: Assessing the safety and feasibility of efficient hypothesis testing in patients with metastatic triple negative breast cancer
C Anthony Blau1, Colin Pritchard1, Michael O Dorschner1, Sibel Blau2, Brigham Mecham3, Elisabeth Mahen1, VK Gadi1, Wayne
Monsky1, Kimberly Burton1, Arturo Ramirez4, Jackie Stilwell4, Eric Kladjian4, Carol Collins4, Jeannine S McCune1, William S
Noble1, Julie Gralow1, Frank Senecal2, Linda Dhaene2, Nicole Kuderer1, Jennifer Specht1, Chaozhong Song1, Carla Grandori7,
Nathan Price6, Mary Goldman5, Aime Radenbaugh5, David Haussler5 and Jingchun Zhu5. 1Center for Cancer Innovation,
University of Washington, Seattle, WA; 2Northwest Medical Specialties, Puyallup and Tacoma, WA; 3Trialomics, Seattle, WA;
4
Rarecyte, Seattle, WA; 5University of California, Santa Cruz, CA; 6Institute for Systems Biology, Seattle, WA and 7Fred
Hutchinson Cancer Research Center, Seattle, WA.
Body: We hypothesize that new insights into how cancers progress and respond to treatment will come from clinical trials that i)
extensively characterize the molecular features of a patients cancer; ii) use results to predict drug susceptibilities; iii) treat in
accordance with these predictions; and iv) learn from individual patient outcomes to iterate and improve over time. To investigate
the feasibility of this type of clinical study, we launched the "Intensive Trial of OMics in Cancer" (ITOMIC) for patients with
metastatic triple negative breast cancer (TNBC) (Clinicaltrials.gov ID: NCT01957514). Eligible patients have metastatic TNBC,
are platinum-naive, and are scheduled to receive Cisplatin. Biopsies are performed under carefully controlled conditions prior to
Cisplatin starting all subjects on a common treatment path, and uncoupling the time needed for specimen analysis from
immediate therapy. Biopsies are repeated upon completion of Cisplatin and following subsequent therapies. A subset of
specimens is chosen for whole Exome Sequencing, deep sequencing of a panel of cancer associated genes, and
RNA-sequencing. De-identified results are placed on a web-based server for analysis and discussed at a meeting of the ITOMIC
tumor board. A report describing results and potential therapies is provided to the subjects oncologist. Treatment decisions are
left to the discretion of the oncologist. If a decision is taken to pursue treatments identified in our report we offer assistance in
accessing those treatments.
Ten patients have been screened and seven have enrolled. Subjects range in age from 40 to 77 years and all but one has
received extensive prior treatment for metastatic TNBC. All seven underwent an initial set of biopsies, targeting between two and
five metastatic sites. For most metastatic sites, multiple core needle passes are performed. All subjects tolerated the biopsies well
without significant adverse events, and all started treatment with Cisplatin. Three subjects completed Cisplatin and underwent a
second round of biopsies.
Potential targets for therapy were identified in 5 of the first 6 subjects, and three subjects have received four predicted therapies:
1) a patient with somatic loss of BRCA1 and two linked FGFR2 activating mutations, who was treated first with Veliparib through a
single-patient IND and then switched to Ponatinib which produced a partial response; 2) a patient with a novel missense ROS1
mutation treated with crizotinib; and 3) a patient with CYP3A4 copy gain treated with cyclophosphamide.
Conclusion: Our early experience indicates that this approach is feasible and may increase the efficiency of learning from
patients with advanced cancer.

Title: PAPAyA The genome informatics framework for oncology applications and other clinical domains
Nilanjana Banerjee1, Konstantin Volyanskyy1, Vartika Agrawal1, Yong Mao1, Yee-Him Cheung1, Sitharthan Kamalakaran1, Laurens
Boekhorst2, Arjan Claassen2 and Nevenka Dimitrova1. 1Philips Research, Briarcliff Manor, NY and 2Philips High Tech Campus,
Eindhoven, Netherlands.
Body: Background: Genomics is expected to transform oncology clinical practice. However, converting genomic data into
clinically actionable information is a daunting task. Genomic high-throughput technologies produce massive amounts of raw data,
and its complexity presents a formidable challenge for clinical adoption. While algorithms exist to convert genomic data into
meaningful biological information, they are geared towards the bioinformatics expert user, lack clinical annotation and
interpretation, and are not addressing the needs of clinical experts. Furthermore, the main problem is that sequencing results
should not be treated like a test as they are geared towards precision diagnostics which require patient clinical data.
Methods: We present PAPAyA, a genome informatics platform that provides an overall solution to the genomic data overload
which includes analysis of whole transcriptome and exome data, secure storage and management of this data and the interactive
presentation of patient genome information in a contextualized manner. It is a continuum of analytics and user experiences with
deep understanding of the clinical questions and the workflow. PAPAyA is a framework for hosting multiple genome informatics
applications that bring information that is Connected, Digital and in Real Time. Connected in many dimensions - across hospital
systems, across time, across many hospitals and their affiliates. Digital means retrievable (without complicated SQL queries) not
just stored bits across modalities (genomics is a single vertical that pulls other information towards precise treatment). Real time
means up to date ACTGs get converted into actionable information fast, and get to the right clinical expert without manual actions
and printing reports. The framework sits on top of a digital health platform that offers core capabilities such as: persisting data and
provisioning service (elastic computing), security, auditing, business workflow engine, reports service, user management service,
patient identity service, provider registry service.
At its core, PAPAyA provides full pipelines that analyze genomic high-throughput data, and further extract and prioritize
clinically-meaningful information from the patient's genome. Furthermore, PAPAyA enables storage and secure management of
this complex data which allows clinicians to query the data in a clinical action-oriented framework, as opposed to data
aggregation and reporting framework.
Results: We demonstrate the usability of PAPAyA using public data from TCGA breast cancer cohort. We show the utility of
executing in-silico assays applied on RNAseq data, such as ER/PR/Her2, differential expression of long-noncoding RNAs, gene
fusions, subtyping, hypoxia index, and other known signatures from the biomedical literature. In parallel, we annotate full exomes
in order to find clinically relevant information for variants of unknown significance as well as variants with known disease and
response phenotype using biological, functional, and clinical annotation resources.
Conclusion: We have implemented an automated open learning system for processing genomic data to aid in clinical decision
making. We already started the pilot phase and clinical evaluation with key academic collaborators.

Title: Demonstration of the evolutionary dynamics of the progression from breast hyperplasia to cancer using the duct epithelial
agent based model (DEABM)
Joaquin Chapa1, Gary An2 and Swati A Kulkarni2. 1University of Chicago Pritzker School of Medicine and 2University of Chicago,
Chicago, IL.
Body: Introduction: Abnormal proliferation in the breast is clinically seen ranging from hyperplasia to atypia to in situ carcinomas
(DCIS) to invasive cancers. Whether these states comprise points along an evolutionary continuum is a subject of debate, but has
significant implications for prevention, prognosis, and treatment. Due to the length of time involved, tumor evolution is difficult to
study in vivo and in vitro models. We hypothesize a continuum from hyperplasia to cancer can be demonstrated by examining
patterns of accumulated mutations in tumors. We employ a previously developed computational model of ductal epithelial
dynamics, the DEABM (1), to simulate and track accumulated mutations during the progression from a normal epithelial
population through hyperplastic states to the development of invasive tumors
Methods: Simulation experiments were performed using an expanded DEABM, which incorporates the functions of luminal and
myoepithelial cells, their progenitors and fibroblasts. DNA damage can potentially disrupt 12 functional pathways regulated by
oncogenes and tumor suppressors implicated in breast cancer (BRCA1, P53, E-cadherin, RUNX3, Myc, ER, TGF-b, MMP-3,
EGFR, HER-2 and Telomerase). The expanded DEABM allows for sequential tracking of genetic lesions in each cell. 3,000
simulations in both wild-type (WT) and BRCA1-mutated states were run over 40 years of simulated menses. Cell populations
were characterized as normal, hyperplastic or malignant based on quantitative cell expansion from baseline, as well as by
mutational profiles, sequential order of mutations and ER status.
Results: Tumor profiles approximated epidemiologic rates of cancer incidence and ER/HER2 status. 2.6 % of WT developed
malignancy vs. 45.9% of BRCA1. WT tumors were 54% ER+/HER2-, 17% ER+/HER2+, 6% ER-/HER2+ and 22% ER-/HER2-.
BRCA1 tumors were 29% ER+/HER2-, 7% ER+/HER2+, 13% ER-/HER2+ and 50% ER-/HER2-. Hyperplastic populations carried
more mutations than non-hyperplastic populations (p>.01) and were more likely to carry mutations in telomerase, E-cadherin and
genes related to ER expression (TGFB, RUNX3, p<.01), similar to the early mutations found in ER+ tumors (RUNX3, TGFB,
telomerase p<.01). Early P53 mutations were common in all tumors (p<.01). ER- tumors were more likely to carry early mutations
in BRCA1, MYC and genes associated with epithelial-mesenchymal transition (MMP-3, p<.01). ER- tumors carried significantly
more mutations than ER+ (p<.01), corresponding to data on increased genomic instability in ER- and BRCA1-associated tumors.
Conclusion: The DEABM generates diverse tumors that express tumor markers consistent with epidemiologic data. The DEABM
also generates non-invasive, hyperplastic populations, analogous to atypia and/or DCIS, via mutations to genes known to be
present in hyperplastic lesions and early mutations in breast cancers. The results demonstrate that agent-based models are
well-suited to studying tumor evolution through stages of carcinogenesis and have the potential to be used to develop prevention
and treatment strategies.
Ref: Chapa J, et al. "Examining the Pathogenesis of Breast Cancer Using a Novel Agent-Based Model of Mammary Ductal
Epithelium Dynamics." PloS one 8.5 (2013).

Title: Impact of adjuvant chemotherapy on clinical and biological ageing in older breast cancer patients
Barbara Brouwers1, Sigrid Hatse1, Lissandra Dal Lago5, Patrick Neven2,4, Peter Vuylsteke6, Guy Debrock7, Heidi Van den Bulck8,
Ann Smeets3,4, Oliver Bechter1, Evalien Swerts1, Jithendra Kini Bailur9, Cindy Kenis1, Annouschka Laenen10, Bruna Dalmasso1,
Patrick Schffski1, Graham Pawelec9 and Hans Wildiers1,4. 1Laboratory of Experimental Oncology, KU Leuven and University
Hospitals, Leuven Cancer Institute, Leuven, Belgium; 2University Hospitals, Leuven, Belgium; 3University Hospitals, Leuven,
Belgium; 4Multidisciplinary Breast Center, University Hospitals, Leuven, Belgium; 5Institut Jules Bordet, Universit Libre de
Bruxelles, Belgium; 6Ziekenhuis Oost Limburg, Genk, Belgium; 7Imelda Ziekenhuis, Bonheiden, Belgium; 8Center for Medical
Research, ZMF, University of Tbingen Medical School, Waldhrnlestr, 22 D-72072 Tbingen, Germany and 9Leuvens
Biostatistiek en Statistische Bioinformatica Centrum, KU Leuven, Leuven, Belgium.
Body: Background. This prospective observational study aimed to evaluate the impact of adjuvant chemotherapy on clinical and
biological markers of ageing and frailty.
Materials and methods. Eligible patients were females 70y with early invasive breast cancer for whom adjuvant chemotherapy
(4 x docetaxel-cyclophosphamide +/- trastuzumab and Granulocyte-Colony Stimulating Factor) was planned (ChemoG). The
control group consisted of breast cancer patients for whom adjuvant chemotherapy was not indicated (ControlG). Patients were
enrolled after surgery, underwent blood sampling and received full geriatric assessment (GA) and Quality of Life, (QoL) evaluation
at baseline, at 3 months (3m) and 1 year (1y). GA results were summarized in a single score, LOFS (Leuven Oncology Frailty
Score), ranging from 10 (very fit) to 0 (very frail). Chemotherapy administration and toxicity were recorded. An extensive (ageing)
biomarker analysis is being performed, but this abstract focuses on mean leukocyte telomere length (TL), and circulating
inflammatory cytokines (IL-6, IL-10, IGF-1, TNF-alfa, MCP-1, and RANTES). The primary endpoint was to assess whether
adjuvant chemotherapy induces accelerated telomere attrition at 1y. Secondary endpoints were the evolution of other ageing
biomarkers, LOFS and QoL during chemotherapy; correlations between ageing biomarkers, chronological age and LOFS at
inclusion; and the predictive value of ageing biomarkers for functional decline, QoL decline, and chemotherapy toxicity.
Results. 57 patients were included (ChemoG), and 53 (ControlG), with mean age at diagnosis of 73,8y and 76,8y, and mean
LOFS of 7,5 (SD 2,3) and 6,8 (SD 1,6). TL was similar in both groups at baseline, and decreased at 3m (p 0,05) and 1y (p
0,0009) in ChemoG, with a similar evolution in ControlG indicating no difference in evolution between both groups (test for
interaction p 0,88). RANTES showed a similar decline at 1y in both groups. The other 5 markers remained stable in ControlG
while significantly changing in ChemoG (significant time interaction): IL-6 decreased at 3m in the ChemoG (p 0,01) and returned
to baseline values at 1y ; MCP-1 strongly decreased at 3m (p <0,0001) but increased above baseline value at 1y (p <0.0001) ;
IGF-1 had a similar initial decline at 3m (p <0,0001) yet with only slight recovery at 1y (p 0,006). On the other hand, IL-10
increased at 3m (p 0,04) but decreased at 1y (p <0,0001), and TNF-alfa increased at 3m (p 0,001) and 1y (p <0,0001). LOFS
declined in ChemoG at 3m (p 0,0007) but returned to baseline level at 1y (p 0,6) while remaining stable over time in ControlG.
Global QoL decreased slightly at 3m in ChemoG and returned to baseline while remaining stable in ControlG. IL-6 correlated
most strongly with chronological age (p 0,0008) and LOFS (p 0,03) while associations were less clear for the other biomarkers. In
ChemoG, MCP-1 and RANTES were associated with functional decline (IADL 1 point decline at 1y) but no biomarkers were
associated with QoL decline and grade II-III-IV toxicity.
Conclusions. TL does not evolve differently over time in patients treated with or without chemotherapy. Chemotherapy has
measurable impact on clinical frailty and other ageing biomarkers at 3 months, but these effects disappear at 1y follow-up.

Title: Evaluation of the role of EBV in breast cancer
Rachel E Ellsworth1, Jill D Henning2, Nikki Oakes2, Allyson L Valente3, Brednda Deyarmin3, Jeff Meyer3, Matthew T Hueman4 and
Craig D Shriver4. 1Henry M Jackson Foundation for the Advancement of Military Medicine; 2University of Pittsburgh, Johnstown,
PA; 3Windber Research Institute, Windber, PA and 4Walter Reed National Military Medical Center, Bethesda, MD.
Body: Background: A number of factors, both modifiable and non-modifiable, are associated with increased risk of developing
breast cancer. Although these factors may help predict risk and prognosis, these known factors are imperfect and suggest that
additional factors exist that promote the development and behavior of breast cancer. The Epstein-Barr virus (EBV), which is
associated with tumor types such as Burkitts lymphoma and gastric cancer, may contribute to breast cancer development and/or
progression, however, current research has led to mixed results, with some groups finding a positive correlation between EBV
and breast cancer and others failing to find an association.
Methods: The Clinical Breast Care Project database was queried to identify patients with high-quality, research-grade frozen
tumor specimens and serum samples available. Previous exposure to EBV was determined using the Epstein-Barr virus (EBNA)
IgG Human ELISA Kit. EBV, B-cells and macrophages were detected by immunofluorescence (IF) in seropositive patients using
probes for EBNA1, CD-68 and CD-20, respectively. EBV and B-cell status was scored as present or absent while macrophages
were classified as sparse, focal (weak, moderate or strong expression) or disseminated (macrophages detected throughout the
tumor and stroma).
Results: ELISA analysis on 211 serum samples found that 202 (96%) patients were seropositive for past EBV exposure. IF data
has been generated for 31 tumors, from which six had no detectable B-cells. Of the remaining B-cell positive tumors, 32% (8/25)
were positive for EBNA1 within the B-cells. Distribution of intrinsic subtypes was similar between both EBV+ and EBV- tumors,
however, diagnosis with a late-stage tumor and lymph node metastasis was higher in EBV+ (50% stage III/IV, 88% lymph node
positive) compared to EBV- tumors (24% stage III/IV, 47% lymph node positive). In addition, EBV+ tumors were more likely to
have sparse/focal(weak) macrophage staining (63%) compared to EBV- tumors (47%).
Conclusions: These data demonstrate that latent EBV is present within B-cells in a significant number of invasive breast tumors.
Decreased detection of macrophages within the tumor component supports the theory that expression of EBV proteins impairs
the function and maturation of macrophages, thus impairing immune surveillance and creating a pro-tumorigenic environment.
This altered immune response may contribute to the later stage and increased frequency of metastasis in patients with EBV
positive B-cells within the tumor microenvironment.

Title: Prognostic impact of discordance between different risk assessment tools in early breast cancer (recurrence score, central
grade, Ki67): Early outcome analysis from the prospective phase III WSG-PlanB trial
Ulrike Nitz1,2, Oleg Gluz1,2, Ronald E Kates1, Daniel Hofmann1, Hans H Kreipe3, Matthias Christgen3, Steve Shak4, Michael
Clemens5, Stefan Kraemer6, Bahriye Aktas7, Sherko Kuemmel8, Toralf Reimer9, Manfred Kusche10, Volker Heyl11, Fatemah
Lorenz-Salehi12, Marianne Just13, Cornelia Liedtke1,14, Rachel Wuerstlein1,15 and Nadia Harbeck1,15. 1West German Study Group,
Moenchengladbach, Germany; 2Ev. Hospital Bethesda, Breast Center Niederrhein, Moenchengladbach, Germany; 3Medical
School Hannover, Institute of Pathology, Hannover, Germany; 4Genomic Health Inc, Redwood City, CA; 5Mutterhaus der
Borromerinnen, Breast Center, Trier, Germany; 6University Clinics Cologne, Breast Center, Cologne, Germany; 7University
Clinics Essen, Women Clinics, Essen, Germany; 8Clinics Essen-Mitte, Women Clinics, Essen, Essen, Germany; 9Clinics
Suedstadt, Rostock, Germany; 10MarienHospital Aachen, Women Clinics, Aachen, Germany; 11Asklepios Paulinen Clinics,
Wiesbaden, Germany; 12Dr. Horst-Schmidt Clinics, Women Clinics, Wiesbaden, Germany; 13Oncologic Practice, Bielefeld,
Bielefeld, Germany; 14University Clinics Schleswig-Holstein/Campus Luebeck, Luebeck, Germany and 15University Clinics
Grohadern, Breast Center of the LMU, Munich, Germany.
Body: Background: In early HR+, HER2-negative breast cancer (BC), the 2013 St. Gallen Consensus recommends adjuvant
chemotherapy (CT) for patients with nodal involvement, Recurrence Score (RS) >25, grade G3, or high Ki67. However, risk
assessment by these factors may be discordant; e.g., in PlanB, about 60% of centrally G3 tumors did not have RS >25. Here, we
present first cumulative outcome data from PlanB for evaluation of different risk assessment tools.
Methods: The WSG-PlanB trial was designed to evaluate anthracyline-free adjuvant CT, 6 x TC vs. 4 x EC - 4 x Doc in
HER2-negative BC. Since an early amendment (August, 2009), HR+ patients with 0-3 involved nodes and RS11 were selected
to omit CT, receiving only adjuvant endocrine therapy. The primary trial endpoint is event-free survival (EFS, events: relapse,
second malignancy, death); secondary endpoints include relapse-free (RFS) and overall survival (OS). Central grade and luminal
B classification were centrally assessed by an independent trial pathologist.
Results: From April 2009 to December 2011, 3198 patients were recruited; of these, 2449 were randomized to chemotherapy.
Median age was 56 years; 84.1% were HR+ by local pathology, 60.8% node-negative. The central tumor bank population
reported here included 3071 cases. RS was available in 2566/2741 cases registered as HR+; of these, 18% had Recurrence
Score of 0-11, 60.4% RS 12-25, and 11.6% RS >25. In 343 patients (14.1% of pN0-1 patients after amendment), CT was omitted
based on RS 11.
By central assessment, in HR+ disease, grade was distributed as follows: G1/G2/G3: 5.3%/62.6%/32.1%; only 43.5% of central
G3 tumors were locally G3; overall concordance between central and local grade was 65.6%. 41.7% of (central) HR+ patients,
had "luminal B" tumors (central Ki6720% and/or PR20%).
After 35 months median follow-up, 131 events, including 103 relapses, have been documented; 3-year EFS and RFS in the
no-chemotherapy group were 98.4% and 99.0%, respectively.
In the central HR+ population, EFS was substantially poorer in patients with RS >25 than in others (3y EFS: 92% vs. 98% in both
RS 12-25 and RS 0-11; p<.001) (n.b.: all patients with RS 12 received CT). RFS was lower in luminal B than in luminal A
patients (3y RFS: 96% to 99%, p=.03); EFS did not differ significantly. Involved lymph nodes, Ki67, central grade, tumor size, and
RS were univariate prognostic factors for EFS. In multivariate analysis (EFS) in central HR+ disease including these factors,
tumor size (fractionally ranked), central G3 (vs. G1 or G2), lymph nodes (2 vs. <2), and RS (fractionally ranked) were all
significant predictors for poor EFS.
Discussion: In spite of receiving no adjuvant CT, patients with RS 0-11 (HR+ HER2- pN0-1) had excellent 3y-EFS. The excellent
outcome of patients with RS 12-25 receiving CT suggests potential CT overtreatment in a subgroup. The ongoing WSG-ADAPT
trial addresses this issue. Early results of WSG-PlanB suggest that quality-assured pathology together with Oncotype DX are
essential in identifying high-risk patients to avoid undertreatment.

Title: Worse breast cancer prognosis in insulin treated diabetic patients - A population based registry study in Sweden
Hkan Olsson1,2, Rickard Einefors1, Per Broberg1 and Mona Landin Olsson3. 1Clinical Sciences, Lund University, Lund, Sweden;
2
Lund University, Lund, Sweden and 3Clinical Sciences, Lund University, Lund, Sweden.
Body: Background. Diabetes may be linked to incidence of different tumor diseases and prognosis through various mechanisms
such as the disease itself, hyperglycemia, obesity and anti-diabetes therapy.
Material and methods. The study includes all women with BC diagnosed in Sweden between 2000 through 2008 (n=54406). The
women had no previous cancer diagnosis during the period of 1958-1999. Dates of birth, BC diagnosis and TNM-stage where
directly extracted from the cancer registry. The womens anti-diabetes therapy was gathered from the Swedish Prescribed Drug
Registry. Information regarding the cause of death and date of death was obtained from the Cause of Death Registry and tbe
Swedish Population Register up until the 31st of December 2012 and 31st of December 2013 respectively. Analyses have been
restricted to patients receiving insulin therapy (n=2463) and their breast cancer prognosis has been calculated in comparison with
breast cancer patients without diabetes. All analyses were adjusted for TNM-stage and age at diagnosis.
Results. Patients with insulin treated diabetes had a worse prognosis compared with other women with breast cancer (HR 1.7,
95%CI 1.5-2.0). The worse prognosis could be seen both for patients with ER+ and ER- tumors. The worst prognosis was seen
for patients treated with NPH insulins (HR 2.8, 95% CI 2.4-3.3) while patients treated with long-acting insulin analogs had an
intermediate prognosis (HR 1.6, 95% CI 1.2-2.2). Those women treated with NPH insulins and metformin had a slightly worse
prognosis (HR 1.4, 95% CI 1.0-1.8). The results for breast cancer specific survival and total survival were similar.
Conclusion. Our results imply that insulin treated breast cancer patients have a worse survival compared with other women with
breast cancer regardless of tumor stage. Metformin therapy may partially counteract the association.

Title: Androgen receptor expression is an independent marker of lower residual risk in the TACT2 trial (CRUK/05/019)
Jane Bayani1, James Morden2, Sunil Skaria6, Peter Bliss7, Robert Grieve8, Adrian Harnett9, Chris Bradley10, Diana Ritchie11, Peter
Barrett-Lee3, Peter Canney4, David Cameron5, Judith Bliss2 and John Bartlett1. 1Transformative Pathology, Ontario Institute for
Cancer Research, Toronto, Canada; 2Institute of Cancer Research Clinical Trials & Statistics Unit (ICR-CTSU), London, United
Kingdom; 3Velindre NHS Trust Cancer Centre, Cardiff, United Kingdom; 4Beatson Oncology Centre, Glasgow, United Kingdom;
5
Edinburgh Cancer Research Centre, University of Edinburgh and NHS Lothian, Edinburgh, United Kingdom; 6Mid Essex Hospital
Services NHS Trust, Chelmsford; 7South Devon Healthcare NHS Foundation Trust, Torquay; 8University Hospitals Coventry and
Warwickshire NHS Trust, Coventry; 9Norfolk & Norwich University Hospital, Norwich; 10Bradford Teaching Hospitals, NHS Trust,
Bradford and 11Beatson West of Scotland Cancer Centre, Glasgow.
Body: Introduction:
TACT2 (CRUK/05/019), a multicentre randomized phase III trial in patients with node +ve or high risk node -ve invasive EBC with
E-CMF as control tested two hypotheses in a 2x2 factorial design, with previously presented results showing: i) no evidence of a
benefit from accelerated 2-weekly epirubicin (aE) compared to standard 3-weekly epirubicin (E) (Cameron 2012); and ii)
capecitabine (X) gives equivalent efficacy but preferential side-effect profile to CMF (Canney 2014).
Studies suggest that Androgen receptor (AR) expression may be associated with improved outcome in early breast cancer. We
performed an analysis of AR expression in TACT2 patients to test the hypothesis that AR would represent an independent
predictor of residual risk following adjuvant therapy.
Methods:
Tumour samples were collected prospectively from 3803/4391 TACT2 patients, Tissue micro-arrays were constructed as per
published guidelines and central AR, ER, PgR, HER2, Ki67, CK5/6, EGFr and BCL2 staining performed and quantified by imaged
analysis for ER, PgR, HER2, Ki67 and BCL2. EGFr and CK5/6 were dichotomised by light microscopy evaluation of cores. After
planned biomarker analyses had been performed, a subset of TACT2 cases for whom remaining tissue was available were
analysed for AR expression. Data for 1878 cases was available for this analysis (49% of those consenting to tissue collection).
AR expression was dichotomised as AR-ve (<10%) vs AR+ve (>10%). Log-rank tests were used to explore the prognostic value
of AR on time to recurrence (TTR). Cox-regression models were used to test the independent prognostic value of AR in the
presence of tumour size, grade, nodal status, and biological subtypes. Marker by treatment interaction terms were included in
models to test the predictive value of AR for both randomised treatment comparisons.
Results:
1398/1878 (74%) patients were classed as AR+ve and 480/1878 (26%) as AR-ve. The proportion of AR+ve patients differed
significantly between biological subtypes (Luminal A 269/293 (92%); Luminal B 784/928 (84%); HER+ve 86/111 (77%); Basal-like
37/216 (17%); 5-marker-ve 37/93 (40%); 2 p<0.001)
117/480 (24%) AR-ve patients had a TTR event compared with 183/1398 (13%) AR+ve patients (HR for AR-ve compared to
AR+ve =2.05 95%CI 1.63-2.59; p<0.001). AR expression remained independently prognostic following adjustment for nodal
status, grade, tumour size and biological subtype (HR1.68, 95%CI 1.26-2.25; p<0.001). Exploratory analysis suggested the
prognostic impact of AR might be predominantly in luminal B and 5-marker-ve breast cancers; however the interaction between
subtype and AR was not statistically significant.
No differential treatment effect between AR subtypes was observed for either randomised treatment comparison (E/aE or
CMF/X).
Conclusion:
AR is an independent prognostic marker for residual risk following chemotherapy in this large study which included patients with
both luminal and non-luminal cancers. AR expression patterns differ between molecular subtypes of early breast cancer, with
evidence suggesting the impact of AR on residual risk may also differ between subtypes. AR is currently being explored as
another potential target for therapy so these data could have future clinical relevance.

Title: Intrinsic subtypes and BCL2 as predictive and prognostic biomarkers in the TACT2 trial (CRUK/05/019)
James Morden1, Judith Bliss1, Jane Bayani2, Robert Laing6, Rajiv Agrawal7, Jeremy Thomas8, Andrew Goodman9, Vivienne Loo10,
Peter Clark11, Peter Canney3, Peter Barrett-Lee4, John Bartlett2 and David Cameron5. 1Institute of Cancer Research Clinical Trials
& Statistics Unit (ICR-CTSU), London, United Kingdom; 2Transformative Pathology, Ontario Institute for Cancer Research,
Toronto, Canada; 3Beatson Oncology Centre, Glasgow, United Kingdom; 4Velindre NHS Trust Cancer Centre, Cardiff, United
Kingdom; 5Edinburgh Cancer Research Centre, University of Edinburgh and NHS Lothian, Edinburgh, United Kingdom; 6St Lukes
Cancer Centre, Guildford; 7Shrewsbury & Telford Hospitals NHS Trust, Shrewsbury; 8Western General Hospital, Edinburgh,
United Kingdom; 9Royal Devon and Exeter NHS Foundation Trust, Exeter; 10Colchester Hospital University NHS Trust, Colchester
and 11Warrington and Halton Hospitals, NHS Trust, Warrington.
Body: Introduction:
TACT2, a multicentre randomized phase III trial in patients with node +ve or high risk node-ve invasive EBC with E-CMF as
control tested two hypotheses in a 2x2 factorial design, presented results showing: i) no evidence of benefit from accelerated
2-weekly epirubicin (aE) compared to standard 3-weekly epirubicin (E) (Cameron 2012); and ii) capecitabine (X) gives equivalent
efficacy but preferential side-effect profile to CMF (Canney 2014).
Here we present prognostic and potential predictive value of translational biomarkers. We address two main hypotheses: i) aE is
less effective in patients with luminal A than patients with other subtypes (Coates 2012) and ii) BCL2 is an independent prognosis
marker (Callagy 2006). We also explore the relationship between CK5/6, EGFR and BCL2 and residual risk following
chemotherapy.
Methods:
Tumour samples were collected prospectively from 3803 patients (86.6% of the 4391 TACT2 patients and 94.5% of those
consenting). Tissue microarrays were constructed as per published guidelines and central ER, PgR, HER2, Ki67, CK5/6, EGFR
and BCL2 staining performed and quantified by imaged analysis for ER, PgR, HER2, Ki67 and BCL2. EGFR and CK5/6 were
dichotomised by light microscopy evaluation of cores. 94.5-97.6% of cases were stained and successfully analysed for individual
biomarkers by IHC. Patients were categorised into 4 BC subtypes by central ER/PgR/HER2/Ki67 (Luminal A, B, HER2+ve,
Triple-ve (TN)) with TN further divided into basal-like (CK5/6 or EGFR +ve) and 5-marker-ve (both CK5/6 and EGFR -ve).
Log-rank tests assessed prognostic effect of each marker individually and cancer subtypes on time to tumour recurrence (TTR).
Cox-regression models tested independent prognostic value of BCL2 in the presence of tumour size, grade, nodal status, and
biological subtype.
Results:
No evidence of a difference in the efficacy of aE compared with E between the 4 patient subtypes was observed (Luminal A:
n=608, HR (for aE compared with E) 0.80 (95% CI 0.47-1.38); Luminal B: n=1804, HR=0.97 (95% CI 0.76-1.23); HER2+ve:
n=219, HR=1.03 (95% CI 0.53-1.99); TN: n=638, HR=1.00 (95% CI 0.72-1.39); Test for heterogeneity p=0.84). When subdividing
the TN group into basal-like and 5-marker-ve, HRs were 0.66 (95% CI 0.45-0.98) and HR 1.56 (95% CI 0.82-2.96) respectively,
post-hoc analysis for heterogeneity between these 2 subtypes p=0.025. No differential effect between subtypes was observed for
the comparison of CMF and X.
82/434 (18.9%) patients with low BCL2 expression (10%) had a TTR event compared with 444/3158 (14.1%) with high BCL2
(>10%), univariate HR 1.39, 95% CI 1.10-1.76, p=0.006. However this difference was no longer seen after adjustment for clinical
factors and biological subtype (HR 1.17 95% CI 0.90-1.52, p=0.25).
Conclusion:
We found no statistical evidence that luminal A cancers are associated with reduced benefit from aE vs E. A hypothesis
generating observation that benefit from aE vs E might be different between basal-like and 5-marker-ve cancers should be
interpreted with caution due to the small numbers of cases and the retrospective nature of the analysis. In this study BCL2 did not
provide independent prognostic information when corrected for conventional histopathological features.

Title: Does androgen receptor (AR) expression impact on residual risk? A TEAM pathology study
John MS Bartlett1, Cassandra L Brookes2, Fu J Yan1, Mary Anne Quintayo1, Jane Bayani1, Jane Starczynski3, Cornelis JH van de
Velde4, Annette Hasenburg5, Dirk G Kieback6, Christos Markopoulos7, Luc Dirix8, Caroline Seynaeve9 and Daniel W Rea2.
1
Ontario Institute for Cancer Research, Toronto, ON, Canada; 2Cancer Research UK Clinical Trials Unit, School of Cancer
Sciences, University of Birmingham, Birmingham, United Kingdom; 3Heart of England NHS Foundation Trust, Birmingham, United
Kingdom; 4Leiden University Medical Center, Leiden, Netherlands; 5University Hospital, Freiburg, Germany; 6Klinikum Vest
Medical Center, Marl, Germany; 7Athens University Medical School, Athens, Greece; 8St Augustinus Hospital, Antwerp, Belgium
and 9Erasmus MC Daniel den Hoed Cancer Center, Rotterdam, Netherlands.
Body: Introduction: Several studies have suggested that AR expression, particularly in luminal cancers following endocrine
therapy, may be associated with improved outcome in early breast cancer. We performed an analysis of AR expression in the
TEAM pathology cohort to test the hypothesis that AR would represent an independent predictor of residual risk following
adjuvant endocrine therapy.
Methods:
Triplicate 0.6mm2 TMA cores from the TEAM pathology cohort (N=4,598) were stained using AR clone ER179(2) on a Ventana
automated staining platform and analysed by image analysis using the Ariol Image analysis platform. Continuous histoscores
were generated as previously described (Bartlett et al, JCO 2011).
Results: AR histoscores were generated from 3866/4598 (84%) of available cases. Median AR histoscores were 227 (interquartile
range 195-262). In a univariate Cox proportional hazard model with AR histoscore as a continuous variable, increased AR
histoscores were significantly associated with a reduced hazard of distant disease relapse or death from breast cancer. The
hazard ratio (HR) associated with a 50 unit increase in the histoscore was 0.88, 95% confidence interval (95%CI) 0.83-0.93,
P<0.0001. There was evidence that a log transformation of the histoscore resulted in a better fitting model (P<0.0001) resulting in
the following model estimates HR= 0.93, 95% CI 0.89-0.98, P=0.006. However, a multivariate model of AR histoscore including
other known prognostic factors such as age, grade, tumour size, number of positive nodes, HER2 status, ER and PgR
histoscores found AR histoscore was not independently prognostic for distant relapse or death (HR=1.00, 95% CI 0.94-1.06,
P=0.96). There was no significant interaction between AR expression and type of endocrine treatment (Tamoxifen exemestane
versus exemestane alone) in either univariate (HR 1.003 95%CI 0.91-1.11, p=0.96) or multivariate (HR 0.92, 95%CI 0.81-1.04,
p=0.18) analysis.
Logistic regression analysis was performed to investigate the association between AR histoscore (2 groups above and below the
median) and the known prognostic factors mentioned above. Increased AR histoscore is associated with good risk factors; young
age, low grade, small tumours, decreasing ki67 and increasing ER and PgR histoscores.
Conclusion: AR expression is common in luminal breast cancers. However, in this study AR histoscore does not add residual risk
information beyond what can already be assessed using conventional prognostic factors. High AR expression is associated with
good prognostic factors, including young age, low tumour grade, small tumour size, lower Ki67 and higher ER/PgR expression.

Title: TLE3 is not a predictive biomarker for taxane sensitivity in the NCIC CTG MA.21 clinical trial
John MS Bartlett1, Torsten O Nielsen2, Dongxia Gao2, Karen A Gelmon3, Mary Anne Quintayo1, Jane Starczynski4, Lei Han5,
Margot J Burnell6, Mark N Levine7, Lois E Shepherd5 and Judy-Anne W Chapman5. 1Ontario Institute for Cancer Research,
Toronto, ON, Canada; 2University of British Columbia, Vancouver, BC, Canada; 3BC Cancer Agency, Vancouver, BC, Canada;
4
Heart of England NHS Foundation Trust, Birmingham, United Kingdom; 5NCIC Clinical Trials Group, Queen's University,
Kingston, ON, Canada; 6Saint John Regional Hospital, Saint John, NB, Canada and 7McMaster University, Hamilton, ON,
Canada.
Body: Background: TLE3, a nuclear transcriptional repressor downstream of the WNT signaling pathway, has been identified as
a candidate predictive biomarker of taxane benefit in early breast cancer. However, robust clinical evidence is required before
implementing novel diagnostic biomarkers. We tested the hypothesis that TLE3 predicts for benefit from taxane containing
polychemotherapy in the NCIC CTG MA.21 clinical trial.
Methods: MA.21 accrued 2104 patients [701 each to cyclophosphamide, epirubicin, and 5 fluorouracil (CEF) and epirubicin and
cyclophosphamide with filgrastim and epoetin alfa followed by paclitaxel (EC/T), 702 to doxorubicin and cyclophosphamide
followed by paclitaxel (AC/T)] who were followed median 8 years by the final analysis. EC/T and CEF were not significantly
different (p= 0.69) while AC/T was inferior to both EC/T and CEF (respectively, p=0.001 and p=0004). Tissue microarrays were
constructed from 1097 of the 2104 patients. Patient characteristics were well balanced between those included in the TLE3
analysis and the full trial population. Up to four 0.6 mm tumor cores were stained for TLE3 expression by immunohistochemistry
using a previously validated methodology. Continuous visual TLE3 score was the average % positive stain across all cores, while
continuous automated score was sum of cells with positive stain/ total cells assessed in all cores. The primary objective used the
EC/T (taxane-containing) and CEF (non-taxane; similar dose-density) arms for a test of predictive effect of TLE3 on relapse free
survival. TLE3 was positive if >30% of cells stained, the established cut-point, with data available from at least 1 core/tumor. We
also examined quartile cut-points, multivariate effects of TLE3, and compared AC/T and CEF.
Results: MA.21 patients had 83.2% TLE3+ tumors by visual score and 80.6% TLE3+ by automated image analysis greater than
the predicted rate of TLE3 positivity (58.6%) based on prior series and adjusting for clinicopathological features. TLE3 expression
was significantly positively associated with ER expression (91.2% of ER+ were TLE3+; p<0.0001). There was no evidence of a
predictive effect of TLE3 expression with respect to taxane benefit using the established 30% cut-point, nor quartile cut-points.
The treatment and TLE3 interaction term for the EC/T by CEF comparison was not significant (stratified p=0.68, for manual TLE3;
p=0.44, for automated TLE3).
Conclusions: MA.21 patients had a much higher proportion of TLE3+ tumors than anticipated. Multiple assessments of TLE3
cut-points yielded no evidence that it was predictive of taxane benefit.
In our trial TLE3 expression was not a biomarker for taxane benefit (EC/T vs CEF) when using either previously established or
common quartile cut-offs for expression in breast cancer.

Title: Comparison of multiparameter tests in the UK OPTIMA-Prelim trial
John MS Bartlett1, Robert C Stein2, Jane Bayani1, Andrea Marshall3, Janet A Dunn3, Amy F Campbell3, Carrie Cunningham4,
Monika Sobol4, Peter Hall5, Leila Rooshenas6, Adrienne Morgan7, Christopher Poole8, Sarah E Pinder9, David A Cameron10, Nigel
Stallard11, Jenny Donovan6, Christopher McCabe12, Luke Hughes-Davies13, Andreas Makris14 and on Behalf of the OPTIMA Trial
Management Group15. 1Ontario Institute for Cancer Research, Toronto, ON, Canada; 2National Institute for Health Research
University College London Hospitals BioMedical Research Centre, London, United Kingdom; 3Warwick Clinical Trials Unit,
University of Warwick, Coventry, United Kingdom; 4Cancer Research Centre, University of Edinburgh, Edinburgh, United
Kingdom; 5Leeds Institute of Health Sciences, University of Leeds, Leeds, United Kingdom; 6University of Bristol, Bristol, United
Kingdom; 7Independant Cancer Patients' Voice, London, United Kingdom; 8University of Warwick, Coventry, United Kingdom;
9
King's College London, London, United Kingdom; 10University of Edinburgh, Edinburgh, United Kingdom; 11Warwick Medical
School, Univeristy of Warwick, Coventry, United Kingdom; 12University of Alberta, Edmonton, AB, Canada; 13Addenbrooke's
Hospital, Cambridge, United Kingdom; 14Mount Vernon Cancer Centre, Northwood, Middlesex, United Kingdom and 15Various.
Body: Introduction
All published adjuvant chemotherapy trials in breast cancer have made the assumption that the proportional benefits of
chemotherapy apply uniformly across molecular subgroups. However, it can be argued that chemotherapy effectiveness for
luminal A breast cancer is low in comparison to other subtypes irrespective of tumour stage. A logical extension of this argument
is that novel multiparametric tests that use biological features of breast cancers to assess risk may also inform chemotherapy
benefit in luminal cancers. The OPTIMA trial is a multi-centre, partially blinded, randomised clinical trial with a non-inferiority
endpoint, and an adaptive design, to compare standard treatment (chemotherapy followed by endocrine therapy) with
multi-parameter test-guided treatment allocation to either chemotherapy followed by endocrine therapy or endocrine therapy
alone. OPTIMA-prelim aimed to compare the predicted risk stratification, sub-type classification and cost effectiveness of different
multiparameter tests performed on the same patient population.
Methods
Over 20 months of recruitment 285 patients were randomised to OPTIMA-prelim. Tissue was collected centrally, ER and HER2
status confirmed and samples provided for testing with Oncotype DX, Prosigna (PAM50), Mammaprint, Mammatyper,
IHC4-AQUA and IHC4 using conventional biomarkers. Sub-type classification was provided by Blueprint, Mammatyper and
Prosigna. Each test was performed at central diagnostic laboratories (OncotypeDx, Mammaprint/Blueprint, Mammatyper) or in a
central laboratory (Prosigna/IHC4) strictly according to GLP practices.
Results
Samples from 181 patients randomised by January 2014 were tested and data analysed for this study. Patients were categorised
as low/intermediate or high risk using predetermined cut-offs for each test. Oncotype DX predicted a proportion of low-risk
tumours (79%; 95% CI 73-85%) similar to that predicted as either low or intermediate risk using Prosigna ROR_P (71%; 95% CI
64-78%) and IHC4 (69%; 95% CI 62-76%), whilst MammaPrint identified the fewest low-risk tumours (59%; 95% CI 52-66%).
Strikingly, a comparison between tests showed modest agreement between tests when dichotomising results between high vs
low/intermediate risk. Disagreement between different tests, in assigning individual tumours to risk categories, is not uncommon;
for the four tests [Oncotype DX, MammaPrint, Prosigna ROR_P (low/int) and IHC4 (low/int)], only 71 (39%) tumours were
classified as low/intermediate risk for all four tests and only 17 (9%) tumours were high risk for all four tests, 93 (52%) tumours
were assigned to different risk categories by different tests. Similarly all three subtypes tests (Blueprint/Prosigna/Mammatyper)
each assigned 59% of tumors to luminal A subtype but only 70% of these cases were classified as luminal A by all three assays.
Conclusion
Existing evidence on the comparative prognostic information provided by different tests suggests current multiparameter tests
provide broadly equivalent risk information for the population of women with luminal breast cancers. However, for the individual
patient, tests may provide differing risk categorisation or indeed subtype information.
Acknowledgement
This project was funded by the NIHR Health Technology Assessment (HTA) Programme (project 10/34/01). The opinions
expressed therein are those of the authors and do not necessarily reflect those of the HTA programme, NIHR, NHS or
Department of Health.

Title: Pathological assessment of discordant cases for molecular (BluePrint and MammaPrint) vs clinical subtypes for breast
cancer, among 6,694 patients from the EORTC 10041/BIG 3-04 (MINDACT) trial
Giuseppe Viale1, Leen Slaets2, Femke A de Snoo3, Jan Bogaerts2, Laura J van 't Veer4, Emiel J Rutgers5, Martine J
Piccart-Gebhart6, Jeroen van den Akker3, Lisette Stork-Sloots3, Leila Russo1, Patrizia Dell'Orto1, Fatima Cardoso7 and On Behalf
of the TRANSBIG Consortium & the MINDACT Investigators. 1European Institute of Oncology and University of Milan, Milan, Italy;
2
European Organisation for Research and Treatment of Cancer, Brussels, Belgium; 3Agendia, Amsterdam, Noord-Holland,
Netherlands; 4Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA; 5Netherlands
Cancer Institute, Amsterdam, Noord-Holland, Netherlands; 6Institut Jules Bordet, Universit Libre de Bruxelles, Brussels, Belgium
and 7Champalimaud Cancer Center, Lisbon, Portugal.
Body: Background
Biology has become the main driver of breast cancer (BC) therapy. Biological subtypes have been recommended as a guide for
treatment selection. Molecularly identified subtypes can be determined by gene expression profiling. Alternatively, pathology can
be used to define surrogates of these subtypes. These methodologies are not always concordant, which can lead to different
systemic therapies. The purpose of this preplanned translational research is to investigate the concordance between molecular
based BluePrint and MammaPrint breast cancer subtypes and pathological surrogates (based on ER, PgR, HER2 & Ki67) and to
characterize the discordant cases.
Methods
MINDACT is an international, prospective, randomized, phase III trial investigating the clinical utility of MammaPrint in selecting
patients with early BC for adjuvant chemotherapy (CT), which enrolled 6,694 patients. Molecular subtyping data were obtained by
MammaPrint and BluePrint (Agendia, Amsterdam, the Netherlands) on frozen samples (n=6,694) classifying patients in the
following subtypes: Luminal A (Luminal-type/MammaPrint Low Risk); Luminal B (Luminal-type/MammaPrint High Risk);
HER2-type; and Basal-type. ER, PgR, HER2 and Ki67 protein status were centrally assessed on FFPE blocks by IHC and/or
FISH in the European Institute of Oncology, Milan, Italy (n=5,740; 86%). Patients were also classified according to the St. 2013
Gallen recommendations [Goldhirsch et al. 2013], which recognizes an additional category (Luminal B-like HER2+).
Results
Ki67 is often used as biomarker to distinguish Luminal A from Luminal B subgroups. The concordance between MammaPrint and
centrally assessed Ki67 in Luminal-type patients is 60%, with a score of 0.26 (95% CI 0.240.28) indicating that Ki67 and
MammaPrint cannot reliably substitute for each other. When using a cut-point of 20% instead of 14% the concordance increased
to 78%, with a score of 0.44 (95% CI 0.410.47).
There is a relatively large group of clinical HER2+ cases that are BluePrint Luminal-type (208 out of 541; 38%) indicating that
tumor expression of the Luminal profile is dominant compared with expression of the HER2 profile.
These patients have high IHC ER results and all except for 1 fall into the group that St Gallen separately defines as Luminal B-like
HER2-positive.
These patients may have lower response to trastuzumab [von Minckwitz et al. JCO 2012].
98 out of 622 BluePrint Basal-type patients are clinical Luminal HER2-. 2/3 of these patients have low centrally assessed IHC PR
expression and 1/3 have low centrally assessed ER expression (1% and <10%).
Conclusions
Marked differences are observed between BluePrint and MammaPrint (microarray based) breast cancer subtypes and centrally
re-assessed pathological surrogates (based on ER, PR, HER2 & Ki67). The greatest discordance is seen in the sub-stratification
of Luminal patients, and in the HR+/HER2+ patients. The observed subtype discrepancies may have an important impact on
treatment decision making.

Title: Comparison of immunohistochemical residual risk panels to predict risk in early breast cancers treated with endocrine
therapy
Jacqueline Stephen1, Gordon Murray1, David Cameron2, Jeremy Thomas2, Ian Kunkler2, Wilma Jack2, Gill Kerr2, Tammy Piper2,
Cassandra Brookes4, Daniel Rea4, Cornelis van de Velde5, Annette Hasenburg6, Christos Markopoulos7, Luc Dirix8, Caroline
Seynaeve9 and John Bartlett3. 1Centre for Population Health Sciences, University of Edinburgh, Edinburgh, United Kingdom;
2
Edinburgh Cancer Research Centre, University of Edinburgh, Edinburgh, United Kingdom; 3Transformative Pathology, Ontario
Institute for Cancer Research, Toronto, Canada; 4Cancer Research UK Clinical Trials Unit, University of Birmingham,
Birmingham, United Kingdom; 5Leiden University Medical Center, Leiden, Netherlands; 6University Hospital, Freiburg, Germany;
7
Athens University Medical School, Athens, Greece; 8Oncology Center, St Augustinus, Antwerp, Belgium and 9Erasmus
MC-Daniel den Hoed Cancer Center, Rotterdam, Netherlands.
Body: Background: We compare two residual risk models combining immunohistochemical (IHC) biomarkers, IHC4 and
Mammostrat, in the Edinburgh Breast Conservation Series (BCS) and in the Tamoxifen versus Exemestane Adjuvant
Multinational (TEAM) trial.
Materials and Methods: The primary cohorts comprised 831 and 2,513 estrogen receptor (ER)-positive patients who did not
receive adjuvant chemotherapy from the Edinburgh BCS and TEAM cohorts respectively. We evaluated prognostic scores for
distant recurrence-free survival (DRFS).
Results: Low scores for both IHC4 and Mammostrat are associated with better DRFS. In multivariate Cox regression analyses
the addition of both scores to clinical factors provided independent information on residual risk (p<0.05). In the larger TEAM
cohort, IHC4 was the stronger predictor of DRFS but additional information was gained from including the Mammostrat score for
all ER-positive patients (p<0.001).
Conclusion: The results showed that the scores have different capabilities in predicting DRFS depending on the study and
subgroup of patients. However, significant benefit in estimating residual recurrence risk after treatment was observed from a
combined use of both marker panels. This provides support for investigating their combined use for risk stratification of
ER-positive early breast cancer patients.

Title: Time dependence of biomarkers: Non-proportional effects of immunohistochemical panels predicting relapse risk in early
breast cancer
Jacqueline Stephen1, Gordon Murray1, David Cameron2, Jeremy Thomas2, Ian Kunkler2, Wilma Jack2, Gill Kerr2, Tammy Piper2,
Cassandra Brookes3, Daniel Rea3, Cornelis van de Velde4, Annette Hasenburg5, Christos Markopoulos6, Luc Dirix7, Caroline
Seynaeve8 and John Bartlett9. 1Centre for Population Health Sciences, University of Edinburgh, Edinburgh, United Kingdom;
2
Edinburgh Cancer Research Centre, University of Edinburgh, Edinburgh, United Kingdom; 3Cancer Research UK Clinical Trials
Unit, University of Birmingham, Birmingham, United Kingdom; 4Leiden University Medical Center, Leiden, Netherlands; 5University
Hospital, Freiburgh, Germany; 6Athens University Medical School, Athens, Greece; 7Oncology Center, St Augustinus, Antwerp,
Belgium; 8Erasmus MC-Daniel den Hoed Cancer Center, Rotterdam, Netherlands and 9Transformative Pathology, Ontario
Institute for Cancer Research, Toronto, Canada.
Body: Background: We investigate the impact of follow-up duration to determine whether two immunohistochemical prognostic
panels, IHC 4 and Mammostrat, provide information on the risk of early or late distant recurrence using the Edinburgh Breast
Conservation Series and the Tamoxifen versus Exemestane Adjuvant Multinational (TEAM) trial.
Methods: The multivariable fractional polynomial time (MFPT) algorithm was used to determine which variables had possible
non-proportional effects. The performance of the scores was assessed at various lengths of follow-up and Cox regression
modelling performed over the intervals 0-5 years and > 5 years.
Results: We observed a strong time-dependence of both the IHC4 and Mammostrat scores with their effects decreasing over
time. In the first five years of follow-up only, the addition of both scores to clinical factors provided statistically significant
information (p<0.05) with increases in R2 between 5 and 6% and increases in D-statistic between 0.16 and 0.21.
Conclusion: Our analyses confirm that the IHC4 and Mammostrat scores are strong prognostic factors for time to distant
recurrence but this is restricted to the first 5 years after diagnosis. This provides evidence for their combined use to predict early
recurrence events in order to select those patients who may/will have benefit from adjuvant chemotherapy.

Title: CADER prognostic gene signature for disease free survival in hormone receptor positive breast cancer: NCIC CTG MA.12
phase III placebo-controlled tamoxifen trial
Matthew J Ellis1, Bingshu Chen2, Judy-Anne W Chapman2, Jingqin Luo1, Diana Ma2, Lois Shepherd2, Jeremy Hoog1, Samuel
Leung3, Elaine Mardis1, Sherri R Davies1, Torsten O Nielsen3 and Vivien Bramwell4. 1Washington University School of Medicine,
St Louis, MO; 2NCIC Clinical Trials Group, Queen's University, Kingston, ON, Canada; 3University of British Columbia,
Vancouver, BC, Canada and 4Tom Baker Cancer Centre, Calgary, AB, Canada.
Body: Background: The copy number aberration derived endocrine response (CADER) gene signature was developed as a
prognostic marker for hormone receptor positive breast cancer; it combines information derived from both gene expression and
copy number aberration and was trained on samples from both the neoadjuvant and adjuvant settings (see separate CADER
nanostring assay methodology abstract, Luo et al). We examined the effects of CADER on outcome in the MA.12 trial to evaluate
the association of CADER and PAM 50 with relapse.
Methods: From 1993 to 2000, MA.12 accrued 672 women to a placebo-controlled pre-menopausal trial of adjuvant tamoxifen
with locally assessed hormone receptor +/- tumors. The 9.7 years median follow-up was used here. A secondary endpoint was
disease free survival (DFS), defined as time from randomization to the earliest date of recurrence or death; censoring was at the
last date the patient was known to be alive. Stratification factors were type of chemotherapy (CMF vs CEF vs AC), hormone
receptor status (ER and/or PgR positive vs ER and PgR negative), and nodal status (0 vs 13 vs 49 vs 10+). Exact Fisher tests
were used to compare baseline characteristics of those assessed, or not, for CADER. We examined the effects of the CADER
classifications (sensitive, indeterminate, resistant) for hormone receptor positive patients. The prognostic effects of CADER, and
preexisting qPCR PAM50 ROR-S, and PAM50 intrinsic subtype (Chia et al, CCR 16, 4465, 2012) on DFS were examined with
stratified multivariate Cox regression, adjusted for treatment and baseline patient characteristics. An interaction test with
treatment was used to assess evidence for CADER as a predictive signature. Graphical depiction was with adjusted Cox survivor
plots.
Results: CADER was assessed in 434 (65%) of the 672 patients: 213 patients on the tamoxifen arm; 221, on the placebo arm.
Proportionately more patients profiled for CADER underwent CEF adjuvant chemotherapy (p=0.03) so we performed only
stratified analyses. Of the 434 patients, 317 (73%) had hormone receptor breast tumors. The CADER classifications significantly
impacted DFS (p=0.04): hazard ratio (HR) of indeterminate CADER to sensitive=2.29 (95% CI 0.93-5.62; p=0.07); HR of resistant
to sensitive=3.72 (95% CI 1.33-10.42, p=0.01). Patients with low ROR-S had longer DFS (p=0.04), while PAM50 intrinsic subtype
did not significantly impact DFS (p=0.51). CADER and ROR-S were not predictive factors.
Conclusions: In MA.12 hormone-receptor patients, both CADER and ROR-S had significant prognostic effects on disease free
survival. Neither CADER nor PAM50 ROR were predictive in this placebo-controlled tamoxifen trial. We confirmed here the earlier
clinical prognostic relevance of CADER that was seen during the signatures development.

Title: MammaPrint accurately predicts long-term survival (25 years) and adjuvant tamoxifen therapy benefit in lymph node
negative patients
Linda S Lindstrom1, Christopher C Benz2, Christina Yau3, Laura J van't Veer3, Carlie K Thompson3 and Laura J Esserman3.
1
Karolinska Institutet, Stockholm, Sweden; 2Buck Institute for Research on Aging, Novato, CA and 3University of California, San
Francisco, CA.
Body: Background: Adjuvant endocrine therapy is a standard of care for early stage estrogen receptor positive (ER+) patients,
but analyses show the majority of benefit accrues after 10 years of treatment. Recent changes to the American Society of Clinical
Oncology recommendations now include the option to extend tamoxifen treatment duration to 10 years. In contrast, screening trial
data with 25 years of follow-up suggest that 50% of screen-detected cancers may never come to clinical attention (Miller et al
BMJ 2014). Better tools are needed to identify patients who benefit from any, 5 or 10 years of adjuvant endocrine intervention.
MammaPrint (MP) is an FDA-cleared gene expression signature that categorizes untreated patients as Low (LR) or High risk
(HR) for distant recurrence (DR) at 10 years, independent of ER and HER2 status. Herein, we examined the performance of MP
to predict long-term (25 years) survival benefit from adjuvant endocrine therapy in early stage breast cancer patients randomized
after surgery to receive 2 or 5 years tamoxifen vs observation only (control arm).
Methods: The Stockholm Tamoxifen Trial (STO) is a uniquely annotated trial with long-term follow-up. 733 samples were made
available from the 1774 lymph node negative breast cancer patients, age <71, with a tumor size <30 mm, enrolled between
November 1976 and May 1990. This group was randomized to adjuvant tamoxifen versus no adjuvant therapy (surgery alone).
Long-term breast cancer specific survival was determined according to MP status (LR or HR as categorized by Agendia; or
UltraLow, an additional threshold proposed in 2011 as predicting exceptionally good distant disease free survival >10 years) for
all patients as well as for each of the randomized treatment arms. MP status was successfully assessed for 656 patients; 3
patients were excluded due to missing randomization data. Multivariate proportional hazards analyses (Cox) was performed, with
the multivariate model adjusted for ER, PR and HER2 status (original assessment), tumor grade and size, and treatment cohort.
Results: A statistically significant difference in survival by MP risk categories was seen for all patients (Plog rank=0.0013), as
well as for the tamoxifen treatment arm (Plog rank=0.011) and the control arm (Plog rank=0.047). By multivariate
proportional hazards (Cox) analyses and after adjustment, women with MPHR had a statistically significant twofold
increased risk of dying from breast cancer by 25 years (Hazards ratio, HR = 1.68, CI 1.132.48), compared to women with
MPLR. In this STO trial, the UltraLow threshold (MP index of > 0.355) was associated with a statistically significant survival
advantage across all patient groups with a 95% breast cancer specific survival at 15 years.
Conclusions: MP accurately predicts for a statistically significant longterm survival benefit in MPLR patients across all STO
patients and, specifically, for those who received adjuvant tamoxifen. Separation of 'UltraLow' risk patients from the MP-LR group
may facilitate clinical decisions for the duration of hormonal therapy. Additional analyses are underway to evaluate the timing of
risk with and without endocrine treatment stratified by MP risk categories.

Title: Validation of the preoperative endocrine prognostic index in the ACOSOG (Alliance) Z1031 neoadjuvant aromatase inhibitor
trial
Souzan Sanati1, Vera J Suman2, Rodrigo Goncalves1, Katherine DeSchryver1, Cynthia X Ma1, Jeremy Hoog1, Erika Crouch1,
Michael Barnes3, Gary Unzeitig4, A Marilyn Leitch5, Kelly K Hunt6 and Matthew J Ellis1. 1Washington University School of
Medicine, St Louis, MO; 2Mayo Clinic and Alliance Statistical Center, Rochester, MN; 3Ventana Medical Systems, Inc, Tucson,
AZ; 4Doctors Hospital of Laredo, Laredo, TX; 5University of Texas Southwestern, Dallas, TX and 6MD Anderson Cancer Center,
Houston, TX.
Body: Background: The Preoperative Endocrine Prognostic Index (PEPI) is a method to predict outcome after neoadjuvant
endocrine therapy that integrates Ki67, estrogen receptor (ER) analysis and pathological stage from the surgical specimen (Ellis,
JNCI 100:1380, 2008). We sought to further develop the PEPI for use in clinical trials by: a) establishing an efficient SOP for Ki67
analysis, b) determining the effect of simplifying the score by removing the ER component (modified "mPEPI") and c) independent
validation of mPEPI in the ACOSOG (Alliance) Z1031 neoadjuvant aromatase inhibitor trial (Ellis, M, JCO 29:234, 2011).
Methods: Ki-67 assay development focused on reproducing a 2.7% Ki-67 cut point (CP) required for PEPI. Ventana Image
analysis (IA) to replace labor-intensive visual point counting (VPC) was assessed to increase scoring efficiency. Discordant
scores led to a triage approach where cases with complex histological features that could not be resolved by IA were flagged for
VPC. The Ki-67 scoring approach was preliminarily validated on T1/2 N0 cases from the P024 and POL trials (SABCS 2013,
abstract P3-05-190). Models with and without ER suggested ER was dispensable. A locked SOP for mPEPI was subsequently
applied to the Z1031A trial. The primary endpoint was time from the date of surgery to local, regional, or distant recurrence in the
mPEPI-0 group (T1/2 N0, Ki67 <2.7%) versus the mPEPI>0 group.
Results. mPEPI by IA was evaluated on 202 of 377 (53%) patients enrolled into Z1031A (6% of IA results were triaged to VPC).
All patients have been followed a minimum of 2 years (median: 5 years; max: 7 years). Only 10 patients in the mPEPI-0 group
(22.7%) received adjuvant chemotherapy, versus 78 in the mPEPI>0 group (49.4%). Time to breast cancer recurrence was
decreased among those with mPEPI>0 status relative to those with mPEPI-0 status (log rank p=0.012). Only one disease event
among 44 (2%) cases with mPEPI-0 was observed versus 26 of 158 cases with mPEPI>0 (16.5%)
Conclusions. mPEPI-0 status can identify patients at low risk of relapse after neoadjuvant endocrine therapy: therefore mPEPI-0
status has operational characteristics similar to pCR after chemotherapy for ER-negative disease. mPEPI is undergoing
prospective validation in the Alliance ALTERNATE trial that will assess whether Fulvestrant increases the mPEPI-0 rate and also
will prospectively determine whether patients with mPEPI-0 status can safely be managed without adjuvant chemotherapy
treatment.

Title: Molecular response in breast cancer treated with neoadjuvant chemotherapy with and without bevacizumab: Results from
NeoAva - a randomized phase II study
Olav Engebraaten1,9, Laxmi Silwal-Pandit2,9, Marit Krohn2,9, Elen K Mller2,9, Silje Nord2,9, Thomas Fleischer2,9, Hedda von der
Lippe Gythfelt1,9, Elin Borgen3,9, ystein Garred3,9, Anne Fangberget4, Marit Muri Holmen4, Ellen Schlichting5,9, Helle Skjerven6,
Steinar Lundgren7, Vessela N Kristensen8,9, Ole Christian Lingjaerde10, Erik Wist1,9, Bjrn Naume1,9 and Anne-Lise
Brresen-Dale2,9. 1Oslo University Hospital, Oslo, Norway; 2Oslo University Hospital, Oslo, Norway; 3Oslo University Hospital,
Oslo, Norway; 4Oslo University Hospital, Oslo, Norway; 5Section for Breast and Endocrine Surgery, Oslo University Hospital,
Oslo, Norway; 6Vestre Viken Hospital Trust, Drammen, Norway; 7St Olav Hospital and Norwegian University of Science and
Technology, Trondheim, Norway; 8Akershus University Hospital, Lrenskog, Norway; 9KG Jebsen Center for Breast Cancer
Research, University of Oslo, Oslo, Norway and 10University of Oslo, Oslo, Norway.
Body: The NeoAva study is a phase II clinical trial of patients with HER2 negative primary tumors of 25 mm treated with
neoadjuvant chemotherapy (4 x FEC100 + 12 weeks of taxane-based therapy) and randomized (1:1) to receive bevacizumab or
no bevacizumab. Mammography, ultrasound and MR imaging were used for response evaluation, in addition to final pathology
assessment.
Tumor response were evaluable in 131 patients; of which 66 received bevacizumab in addition to chemotherapy. Tumor material
was obtained at screening, 12 weeks into treatment and at surgical removal of tumors at 25 weeks. mRNA expression profiling
was performed on Agilent 8x60K platform and the tumors were classified into LuminalA, LuminalB, Her2-enriched, Basal and
Normal-like subtypes using the PAM50 classifier. Ratio of the tumor size at final pathology assessment, and at inclusion (by
radiology assessment) was calculated to obtain a continuous scale of response reflecting the percentage of tumor shrinkage in
response to therapy. Genomic Grade Index (GGI scores) based on expression profiles of 97 genes (including cell-cycle and
proliferation genes) were calculated.
There were no significant differences in the tumor size, lymph node, hormone receptor status or PAM50 subtypes between the
treatment arms. pCR in breast and axilla were obtained in 14 (21.1%) patients in the chemo+bev arm, and in 7 (10.6%) patients
in the chemo-only arm. Tumors that obtained pCR were in higher number ER negative and TP53 mutated and exhibited
Basal-like phenotype. The overall pCR rates were higher in the ER negative tumors compared to ER positive tumors {39.1% (9 of
23) vs 11.1% (12 of 108)}. However, addition of bevacizumab seemed to improve pCR in the ER positive patient group (9 vs 3)
and not in ER negative patient group (5 vs 4).
On evaluating the continuous response variable, ER status, TP53 mutation status and PAM50 subtypes were significantly
associated to response (p < 0.001). GGI scores were highly correlated to response (p< 0.001), i. e tumors with higher GGI scores
showed better response. Importantly, when the chemo+bev and the chemo-only arms were evaluated separately, although similar
trend of associations was observed in both arms, the associations were found to be enhanced in the chemo+bev arm.
Next, we evaluated a shift in PAM50 subtypes across the timepoints. A shift towards a better prognosis group, i.e Luminal A or
Normal-like profile was observed in response to therapy. Distribution of Luminal A and Normal-like tumors at week 25, (and not at
screening or week 12) was significantly different in the chemo+bev vs chemo-only group (p = 0.026, Fishers exact test).
GGI scores regressed across timepoints reflecting the loss of aggressive and proliferating component of the tumors in response
to therapy. GGI scores in the chemo+bev group became significantly lower (p < 0.01) already at week 12. This suggests that the
removal of the proliferating component of the tumors by chemotherapy is accelerated and improved by addition of bevacizumab.
These results, with potentially important clinical relevance will be further investigated with respect to subtypes and the molecular
changes induced by antiangiogenic therapy.

Title: Risk stratification within luminal B breast cancer using a second generation prognostic RNA signature
Giancarlo Pruneri1,5, Vincenzo Bagnardi2,4, Davide Disalvatore2, Giuseppe Curigliano3, Nicole Rotmensz2, Carmen Criscitiello3,
Darl D Flake II6, Susanne Wagner6, Alexander Gutin6, Jerry Lanchbury6, Massimo Barberis1, Francesca Lombardi1 and Giuseppe
Viale1,5. 1European Institute of Oncology, Milan, Italy; 2European Institute of Oncology, Milan, Italy; 3European Institute of
Oncology, Milan, Italy; 4University of Milan-Bicocca, Milan, Italy; 5University of Milan, School of Medicine, Milan, Italy and 6Myriad
Genetic Laboratories, Inc, Salt Lake City, UT.
Body: Background: Indication to adjuvant chemotherapy in early stage ER-positive breast cancer is usually based on
clinico-pathological parameters, and has been recently improved by the introduction of prognostic expression profiles. Clinical or
molecular features are efficient in identifying low-risk (pT1, histological grade 1, pN0, low RS, Luminal A subtype) and high-risk
patients (>pT1, grade 3, N2, high RS). However, a large group of patients with intermediate clinical or molecular characteristic
(grade 2, intermediate RS, luminal B) fall into the category between distinctly low and distinctly high risk and receive no treatment
guidance from current decision tools.
Methods: From a population of 1929 chemo-nave, hormone treated, luminal B (Her-2 negative), pT1-pT3, pN0-N1a breast
cancer patients diagnosed and treated at the European Institute of Oncology from 1997 to 2005, we selected a random subcohort
of 555 cases, in a case-cohort design. All the patients with local or distant metastasis which were not already included were
added to the subcohort, leading to a total of 704 patients (208 with local or distant recurrences and 496 random controls). Luminal
B status was determined by the immunohistochemical analysis of ER, PgR, HER2 and Ki-67, according to the 2011 St. Gallen
criteria. FFPE sections of the primary tumor were analyzed for the mRNA expression of 43 genes by multiplex quantitative PCR.
A molecular score (MS) was calculated from the average expression of 23 cell cycle progression genes, the average expression
of seven lymphocyte specific genes and the expression of PR and ABCC5, based on a model derived from an independent
training cohort. A combined score of MS and the clinical variables of tumor size, grade and node status was modeled in the
training cohort and applied to the Luminal B set. The association between MS and the risk of distant metastasis was evaluated in
a weighted multivariable Cox regression model, adjusted for traditional clinical factors and Ki-67 labeling index (LI).
Results: 640 samples, including 102 distant metastasis, had full clinical and expression data. In the 500 samples from the
subcohort, median Ki67 LI was 21% (IQR=11%, Q1=16%, Q3=27%). Either one unit increase of Ki-67 LI (HR 1.06, 95%CI
(1.04-1.08), p<0.0001) and of MS (HR 3.4, 95%CI (2.5-4.6), p<0.0001) were highly significant predictors of distant recurrence in
univariable analysis. In multivariable analysis, the MS provided independent significant prognostic information after adjustment for
Ki-67 LI, tumor size, grade and node status (HR 4.3, 95%CI (2.5-7.3), p=<0.0001). Using the combined score of MS and clinical
variables, 383 patients or 77% of the subcohort had an estimated 10 year risk of distant recurrence of 10%. Similar results were
obtained when samples were re-defined according to 2013 St. Gallen guidelines.
Conclusions: The MS provides important prognostic discrimination beyond traditional clinico-pathological characteristics,
including Ki-67 LI, in Luminal B breast cancer, and contributes in identifying a subset of patients which may be successfully
treated with endocrine therapy only.

Title: Low serum adiponectin level is an independent risk factor of DCIS in postmenopausal women at increased risk of breast
cancer
Aliana Guerrieri-Gonzaga1, Debora Macis1, Sara Gandini1, Valentina Aristarco1, Harriet Johansson1, Giorgia Bollani1, Teresa
Roth1, Maria-Teresa Sandri1, Jill Knox2, Jack Cuzick2 and Bernardo Bonanni1. 1European Institute of Oncology, Milan, Italy and
2
Queen Mary University of London, London, United Kingdom.
Body: Background:The assessment of breast cancer (BC) risk is a key step for an effective preventive treatment. Besides the
established risk assessment models, validation of independent predictive factors such as circulating biomarkers would improve
patient selection and treatment efficacy.
Obesity and metabolic imbalance play an important role in BC risk in menopausal women. The role of adipocytes in energy
homeostasis is currently under investigation for their emerging relationship with BC. Adipokines (such as leptin and adiponectin)
are linked to insulin sensitivity and have been related to BC risk and prognosis. Adiponectin, a peptide hormone secreted by the
adipose tissue, has been inversely related to BC risk both in observational studies and in a phase II chemoprevention trial in
premenopausal women at increased risk.
Aim:We measured baseline serum adiponectin and leptin levels as well as HOMA index, in 534 postmenopausal women enrolled
in 16 Italian centers and randomized in one of the two international phase III trials for BC prevention -the IBIS-II(Prevention) and
IBIS-II(DCIS) trials- to assess whether these biomarkers were different in the healthy women at increased risk for BC cohort
compared to the DCIS cohort.
Methods:Healthy postmenopausal women (aged 40-70) at increased risk for BC (on an age-dependent risk model; n=186) or
DCIS patients who underwent radical surgery in the previous 6 months (n=348) were eligible according to the two separate
protocol entry criteria. At baseline, fasting blood was collected, processed and stored at -80C till biomarkers measurement.
Insulin and glucose levels were measured with the Architect system (Abbott Laboratories, Abbott Park, IL 60064 USA). Serum
adiponectin and leptin levels were determined with Immunoassays by R&D (Minneapolis, USA).
Results:Participant characteristics and biomarker levels (median, IQ range) by disease status are reported below.
Table 1
Healthy (n=186)
DCIS (n=348)
Age at entry
59 (55, 63)
60 (56, 65)
BMI (kg/m2)
25.2 (22.9, 28.4)
25.0 (22.4, 28.1)
Adiponectin (ng/mL)
13063 (10279, 18157)
11498 (7722, 16909)
Leptin (pg/mL)
16181 (9594, 26391)
17284 (9675, 26173)
L/A ratio
1.33 (0.60, 2.09)
1.46 (0.69, 3.19)
Glucose (mg/dL)
89 (81, 97)
93 (86, 103)
Insulin (U/mL)
6.6 (5.0, 9.6)
7.5 (5.4, 10.6)
HOMA index
1.39 (1.04, 2.02)
1.79 (1.18, 2.61)
L/A, Leptin/Adiponectin ratio

Adiponectin was significantly negatively correlated with leptin, L/A ratio, HOMA index and BMI, while leptin was positively
correlated with L-A ratio, BMI, HOMA.
Logistic regression has been used and Odds Ratios (ORs) have been calculated to assess the association of DCIS with
biomarkers.DCIS patients were significantly more frequent in the lowest quartile of adiponectin compared to the highest quartile
(60% vs 75%; OR=2.49; 95% CI, 1.39-4.44, p= 0.0003) adjusting for center, BMI, HOMA index and age.
Conclusions:Low serum adiponectin levels in postmenopausal women are more frequent in DCIS patients compared to healthy at
risk subjects independently of BMI, HOMA index and age and results are similar to premenopausal women.
Future investigations in both trials will assess whether adiponectin is also associated with BC events.
Acknowledgments: J.Forbes and T.Howell, Co-Chairmen,IBIS-II Steering Committee.

Title: Validation of a multi-marker test that predicts recurrence in patients diagnosed with ductal carcinoma in situ (DCIS) treated
with breast-conserving surgery (BCS)
Steven P Linke1, Troy M Bremer1, Pat Whitworth8, Alfred Lui3, Jess Savala3, Yelina Noskina3, Todd Barry6, Stephen Lyle2,
Stephanie C Walters2, Cherie Taglienti2, Karl Simin2, Wenjing Zhou4, Karin Jirstrm5, Rose-Marie Amini4 and Fredrik Wrnberg4.
1
Prelude Corporation, Laguna Hills, CA; 2University of Massachusetts Medical School, Worcester, MA; 3PersonalizeDx, Lake
Forest, CA; 4Uppsala University, Uppsala, Sweden; 5Lund University, Lund, Sweden; 6Spectrum Pathology, Mission Viejo, CA and
7
Nashville Breast Center, Nashville, TN.
Body: Background: DCIS is diagnosed in 54,000 women/year in the US. Identifying patients who are most likely to recur has
been deemed one of the most important unmet needs in breast cancer treatment. The goal of this study was to develop and
blindly validate a multi-marker risk stratification test in DCIS patients treated with BCS.
Material and Methods: A variety of clinical, pathological, immunohistochemical, and in situ hybridization data were derived from a
set of patients diagnosed with DCIS (without evidence of invasive breast cancer) between 1986 and 2004 in Uppland and
Vstmanland, Swedenthe Uppsala University Hospital (UUH) patient set. Separate models to predict DCIS and invasive event
risk were developed using statistical pattern recognition and modeling methods on UUH patients treated with BCS in the absence
of adjuvant therapy (n=200). In addition, an "overall" risk model was created by combining the DCIS and invasive models. The
risk models consist of algorithms that combine data on p16/INK4A, Ki-67, COX-2, PR, HER2, FOXA1, SIAH2, CD31, patient age,
necrosis, tumor size, palpability, and/or margin status, along with predefined thresholds that create low, intermediate, and high
risk groups ("elevated" risk groups combine the intermediate and high risk groups).
The models were then tested blindly on a set of patients diagnosed with DCIS and treated with BCS from 1999 through July 2008
at the University of Massachusetts Memorial Hospital, Worcester, MA (UMass patient set). Molecular marker data was collected
with CLIA-validated assays and Board-certified pathologists, and other data was collected from medical records. Testing was
done according to a predefined statistical analysis plan in the UMass patients equivalent to those in the UUH patient training
set--those with complete marker data and treated with BCS, and were either PR-negative or not treated with hormone therapy
(n=155). Event rates were assessed for up to 10-year outcome using Kaplan-Meier survival analysis. Hazard ratios (HRs) were
determined using Cox proportional hazards analysis.
Results: The 10-year overall event rate in the full population was 12% (6% invasive plus 6% DCIS). For the invasive risk model,
the low (n=29) and elevated (n=97) risk groups had 0% and 7% 10-year invasive event rates, respectively. For the DCIS risk
model, the low (n=76) and elevated (n=50) risk groups had 2% and 15% 10-year DCIS event rates, respectively. For the overall
risk model (combined risk of DCIS or invasive events), the low (n=20), intermediate (n=65), and high (n=41) risk groups had 0%,
7%, and 22% overall 10-year event rates, respectively (HR=8.0/bin, p=0.001). The overall risk model maintained significance
when adjusted for nuclear grade, tumor size, patient age, necrosis, and margin status, while none of these clinicopathologic
factors demonstrated significance in the presence of the model.
Discussion: This study indicates that the present approach to risk stratification modeling can accurately identify patients at risk for
DCIS or invasive events after a primary DCIS diagnosis. The models presented here are the basis of a comprehensive
multi-marker panel undergoing formal validation.

Title: FOXA1 and PR predict ipsilateral event risk and identify a group with strong radiation response in ductal carcinoma in situ
(DCIS)
Steven P Linke1, Troy M Bremer1, Pat Whitworth7, Aflred Lui3, Jess Savala3, Yelina Noskina3, Todd Barry6, Stephen Lyle2,
Stephanie C Walters2, Cherie Taglienti2, Karl Simin2, Wenjing Zhou4, Karin Jirstrm5, Rose-Marie Amini4 and Fredrik Wrnberg4.
1
Prelude Corporation, Laguna Hills, CA; 2University of Massachusetts Medical School, Worcester, MA; 3PersonalizeDx, Lake
Forest, CA; 4Uppsala University, Uppsala, Sweden; 5Lund University, Lund, Sweden; 6Spectrum Pathology, Mission Viejo, CA and
7
Nashville Breast Center, Nashville, TN.
Body: Background: Identification of biomarkers in DCIS is critical to help guide treatment decisions, particularly for patients
receiving breast-conserving surgery (BCS). The goal of this study was to assess FOXA1 levels in the context of PR status in
primary DCIS to predict ipsilateral invasive and DCIS events, given the established roles of these endocrine signaling factors in
breast cancer.
Material and Methods: Patients included in this study (n=518) were women diagnosed with DCIS without evidence of invasive
cancer treated with BCS, and for whom tumor tissue was evaluable for both PR and FOXA1. An Uppsala University Hospital
(UUH) set was diagnosed in 1986-2004 (117 treated and 122 not treated with adjuvant radiation therapy [RT]); and a University of
Massachusetts Memorial Hospital (UMass) set was diagnosed in 1999-2008 (195 treated and 84 not treated with RT).
Tumors were immunohistochemically stained for PR and FOXA1 and scored by pathologists for percentage (0-100) and intensity
(0-3), with the product being calculated for FOXA1 immunoscore (0-300). Patients were considered PR+ when 10% of cells
stained positively, and immunoscore thresholds of 100 and 270 were used to separate patients into FOXA1 low, intermediate,
and high groups. Event rates were assessed for 10-year outcome using Kaplan-Meier survival analysis. Hazard ratios (HR) were
determined using Cox proportional hazards analysis.
Results: In non-RT-treated patients, neither FOXA1 nor PR were prognostic as independent factors for either invasive or DCIS
event risk. However, in the PR+ subset, the invasive event rate increased with increasing FOXA1 (0%, 10%, and 13%,
respectively, HR=2.9/bin, p=0.046), while the invasive event rate was lower in the PR- subset with elevated FOXA1 levels (3% in
FOXA1 int./high vs. 25% in FOXA1 low, HR=9.7, p=0.006). In contrast, the DCIS event rate increased in the PR- subset with
increasing FOXA1 (3%, 15%, and 22%, respectively, HR=2.1/bin, p=0.040), while the DCIS event rate was lower in the PR+
subset with elevated FOXA1 levels (8% in FOXA1 int./high vs. 28% in FOXA1 low, HR=3.7, p=0.011).
In the full population, RT-treated patients (n=312) fared better than the non-irradiated (n=206) with invasive event rates of 6% and
10%, respectively. However, the event rate in patients with elevated marker-based invasive event risk based on PR and FOXA1
status (n=195) was reduced from 22% to 4% with RT. In contrast, the remainder of patients (n=323) showed no evident RT
response. In a multivariate Cox analysis, the high-risk group (HR=5.0, p=0.005) and RT response within that group (HR=0.13,
p=0.013) were significant when adjusted for clinicopathologic factors, but baseline RT response was not (p=0.29).
Discussion: These results indicate a complex interaction between PR and FOXA1, in which the prognosis for invasive and DCIS
events flips both within and between the event types. Thus, the biology that drives these events differs and, in order to predict
both event types, risk models must include biomarkers in context. In addition, PR/FOXA1 identify a risk group with remarkably
strong RT response with the remaining patients exhibiting no measurable response.

Title: Integration of breast cancer index (BCI) with clinicopathological factors for prediction of distant recurrence in ER+ breast
cancer
Ivana Sestak1, Yi Zhang2, Brock E Schroeder2, Paul E Goss3, Mitch Dowsett4, Dennis C Sgroi3, Catherine A Schnabel2 and Jack
Cuzick1. 1Centre for Cancer Prevention, Queen Mary University, London, United Kingdom; 2bioTheranostics, Inc, San Diego, CA;
3
Massachusetts General Hospital, Boston, MA and 4Royal Marsden Hospital, London, United Kingdom.
Body: Background: BCI is a genomic signature that significantly predicts risk of both early (0-5 years) and late (5-10 years)
distant recurrence (DR) in hormonal receptor-positive, lymph node negative (LN-) breast cancer. In LN- disease, tumor size (TS)
and grade (TG) are important independent prognostic factors of breast cancer recurrence. In this analysis, the effect of integrating
these traditional clinicopathological factors on the ability of BCI to predict distant recurrence was evaluated in the translational
arm of the Arimidex, Tamoxifen, Alone or in Combination trial (TransATAC).
Methods: 709 primary tumor samples from hormonal receptor-positive, LN- patients treated with 5 years of tamoxifen or
anastrozole were examined. Multivariate Cox proportional hazards regression was used to fit 2 models: 1) BCI+TS; 2)
BCI+TS+TG. To facilitate comparison across models, cut points were chosen based on the pre-specified 10 year DR rates of
<10% (low), 10-20% (intermediate) and >20% (high). Kaplan-Meier (KM) estimates of 10 year DR and hazard ratios (HR) were
examined. Change in likelihood ratio 2 (LR-2) values were used to quantitate relative prognostic information beyond standard
clinicopathological variables (CTS, Clinical Treatment Score).
Results: In the univariate analysis, all models were significantly prognostic for 10 year DR risk; BCI was somewhat less predictive
vs BCI+TS, whereas TG did not provide significant additional value beyond BCI+TS [HR (95% CI): 3.26 (2.29-4.63), 2.72
(2.11-3.50), 2.72 (2.11-3.50); LR-2 (p value): 45.54 (p<0.0001), 54.71 (p<0.0001), 57.27 (p<0.0001) for BCI, BCI+TS,
BCI+TS+TG, respectively]. Adjusted HRs beyond CTS demonstrated highly significant and comparable prognostic ability [HR
(95% CI): 2.35 (1.61-3.42), 2.06 (1.48-2.86); LR-2 (p value): 20.75 (p<0.0001), 18.96 (p<0.0001)] for BCI and BCI+TS,
respectively. BCI and BCI+TS categorized similar proportions of patients into respective risk groups, and KM analysis comparing
BCI vs BCI+TS risk categories showed similar rates of 10 year DR (Table 1).
Comparison of Risk Categorization and 10 year DR
BCI
BCI+TS
% Patients
10 year DR
% Patients
10 year DR
Low Risk
53.7%
5.3%
54.7%
5.2%
Intermediate Risk
29.3%
15.5%
29.2%
16.5%
High Risk
17.0%
28.0%
16.2%
27.3%
In cross stratification analysis between BCI and BCI+TS, no significant re-classification was observed (Table 2), however 14.5%
of BCI and 9.7% of BCI+TS intermediate risk patients were re-stratified as low risk.
Re-classification of BCI and BCI+TS for all patients
BCI
BCI+TS
Low
Intermediate
High
Total
Low
358
30
388 (54.7%)
Intermediate
20
163
23
206 (29.1%)
High
14
98
115 (16.2%)
Total
381 (53.7%)
207 (29.2%)
121 (17.1%)
709
Discussion: Integration of tumor size, but not grade, statistically enhanced the prognostic ability of BCI to predict 10 year DR risk
in patients with ER+, LN- early stage breast cancer. However, there was limited clinical impact on risk stratification, indicating that
the prognostic information provided by these clinicopathological factors is effectively captured by BCI alone.

Title: No Show

Title: Prevalence of circulating tumor cells (CTCs) after five years of zoledronate treatment in the adjuvant SUCCESS-A study
Brigitte Rack1, Peter A Fasching2, Lothar Hberle2, Thomas Friedl3, Mahdi Rezai4, Jrn Hilfrich5, Hans Tesch6, Georg Heinrich7,
Helmut Forstbauer8, Julia Neugebauer1, Elisabeth Trapp1, Susanne Albrecht3, Bernadette Jger3, Tanja Fehm9, Volkmar Mller10,
Andreas Schneeweiss11, Klaus Friese1, Werner Lichtenegger12, Matthias W Beckmann2 and Wolfgang Janni3.
1
Ludwig-Maximilians-University, Munich; 2University of Erlangen; 3University of Ulm; 4Luisenkrankenhaus Dsseldorf;
5
Diakoniekrankenhaus Henriettenstiftung Hannover; 6Practice Prof. Tesch Frankfurt; 7Practice for Obstetrics and Gynecology
Fuerstenwalde; 8Practice Dr. Forstbauer Troisdorf; 9University of Duesseldorf; 10University Hamburg-Eppendorf; 11University of
Heidelberg and 12University Medical Center Charite, Berlin, Germany.
Body: Aim: The prognostic value of CTCs at primary diagnosis has recently been confirmed by the SUCCESS A Study (Rack et
al. JNCI 2014). Key questions on the role of adjuvant bisphosphonate treatment, including patient populations deriving benefit
and optimal timing/scheduling of therapy are still controversial. Aim of this study was therefore to evaluate CTCs in the context of
two different zoledronate regimens.
Methods: The SUCCESS A trial is a large, randomized, open-label, 2x2 factorial design Phase III study in patients with high risk
breast cancer (stage N1 or T2T4 or grade 3 or age 35 or hormone-receptor negative). Patients were first randomized to
adjuvant chemotherapy treatment with 3 cycles of Epirubicin-Fluorouracil-Cyclophosphamide (FEC) followed by either 3 cycles of
Docetaxel or 3 cycles of Gemcitabine-Docetaxel. In addition, patients were randomized to 2 years vs. 5 years of zoledronate
treatment. CTC status 5 years after primary diagnosis was assessed using the FDA-approved CellSearch System (Veridex,
USA), and CTC positivity was defined as 1 CTC. We studied the influence of zoledronate treatment duration on CTC
prevalence at 5 years in addition to well-known patient and tumour characteristics using a multiple logistic regression analysis on
the ITT population.
Results: Data on CTC status at 5 years after primary diagnosis were available for 728 (19.4%) out of 3754 randomized patients.
310 patients had been randomized to 2 years of zoledronate treatment and 418 patients had been randomized to 5 years of
zoledronate treatment. In these patients a difference in CTC positivity after 5 years could not be shown between patients
randomized to 2 or 5 years of zoledronate (p = 0.13, Wald test). The adjusted odds ratio for 2 years vs 5 years was 0.65 (95%CI:
0.37 to 1.13).
Conclusions: The final survival analysis of the SUCCESS A trial will give further insight, whether CTCs can be used as early
predictive marker for the efficacy of adjuvant zoledronate treatment.

Title: High expression of LMTK3 is an independent factor indicating a poor prognosis in Japanese estrogen receptor -positive
breast cancer patients
Tatsuya Toyama1, Tomoko Asano1, Shinya Sato2, Nobuyasu Yoshimoto1, Yumi Endo1, Yukari Hato1, Dong Yu1, Satoru
Takahashi2 and Yoshitaka Fujii1. 1Immunology and Surgery, Nagoya City University Graduate School of Medical Sciences,
Nagoya, Japan and 2Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan.
Body: Over 70% of breast cancers are estrogen receptor (ER)-positive, and endocrine therapy targeting estrogen action
decreases mortality from breast cancer. However, their efficacy of endocrine therapy is limited by intrinsic and acquired
resistance. Recently, a novel protein kinase that regulates ER activity, lemur tyrosine kinase-3 (LMTK3) has been identified. In
this study, we investigated whether LMTK3 mRNA expression and its polymorphisms are associated with prognosis in Japanese
breast cancer patients during long-term follow-up. First, we investigated the relationship between mRNA expression of LMTK3
and patient outcome in 242 breast cancers (median follow-up, 6.8 years). The effects of several variables on survival were tested
by Cox proportional hazards regression analysis. Next, we performed to analyze LMTK3 rs9989661 and rs8108419 genotyping in
641 breast cancer tissues (median follow-up, 9.2 years) to clarify the prognostic role of these polymorphisms. We showed that
high expression levels of LMTK3 mRNA were significantly associated with a shorter overall survival (OS) in all patients. We then,
analyzed the impact of LMTK3 mRNA expression on the prognosis of breast cancer according to ER status. Both disease-free
survival and OS were significantly shorter in ER-positive patients with high LMTK3 mRNA expression receiving adjuvant
endocrine therapy than in those patients with low LMTK3 mRNA expression. Notably, the level of LMTK3 mRNA expression was
not associated with prognosis in ER-negative breast cancer patients. Univariate and multivariate Cox regression analysis of
factors associated with OS revealed that high LMTK3 mRNA expression was an independent poor prognostic factor in
ER-positive breast cancer patients.
Univariate and multivariate Cox regression analysis of factors associated with overall survival in ER-positive breast cancer
patients
Variables
n (%)
2cm
51 (31)
> 2cm
112 (67)
Negative
79 (47)
Positive
72 (43)
1,2
128 (77)
27 (16)
Low
129 (77)
High
38 (22)
Univariate
Multivariate
P value
P value
HR (95% CI)
Tumor size
1 (reference)
0.264
0.228
2.75 (0.57-23.5)
Node status
1 (reference)
0.003
0.002
5.25 (1.35-34.7)
Grade
1 (reference)
0.485
0.547
1.57 (0.32-6.04)
LMTK3 mRNA expression

1 (reference)
0.002
0.012
4.94 (1.44-19.4)
HR, hazard ratio; CI, confidence interval.

We did not find any correlation between LMTK3 genotypes and prognosis of breast cancer patients in our series. Our results
suggest that higher LMTK3 expression might be one possible mechanism for endocrine resistance in ER-breast cancer patients,
to be confirmed by further research.

Title: Membranous ER-36 expression is an independent predictor of poor prognosis in operable breast cancer
Loay Kassem1, Soleilmane Omarjee2, Sylvie Chabaud3, Emilie Lavergne3, Christelle Faure4, Frdric Beurrier4, Olivier Tredan5,
Laura Corbo2, Isabelle Treilleux6 and Muriel Le Romancer2. 1Cairo University Teaching Hospital, Cairo, Egypt; 2Inserm
U1052/CNRS UMR5286/University of Lyon, Centre de Recherche en Cancrologie de Lyon, Lyon, France; 3Biostatistics Unit,
Centre Lon Brard, Lyon, France; 4Centre Lon Brard, Lyon, France; 5Centre Lon Brard, Lyon, France and 6Centre Lon
Brard, Lyon, France.
Body: Background
ER-36 is a splice variant of ER- with molecular weight of 36-kDa that lacks transactivation domains, and is expressed in the
cytoplasm and cell membrane of ER (ER66) negative as well as ER66 positive breast cancer cells. It is also thought to predict
resistance to tamoxifen therapy. Here we investigate its prognostic significance, its association with other clinico-pathologic
factors and correlation with other biomarkers of the PI3K/AKT/mTOR pathway.
Methods
We studied ER-36 expression on TMA blocks prepared from samples of 160 consecutive operable breast cancer patients who
presented at CLB between 1998 and 2001. The intensity of the staining and the percentage of tumor cells stained for each
biomarker (ER-36, PI3K, pAKT, p4EBP1, pS6RP and LKB1) were integrated into a single score and a cutoff was defined for
high versus low expression. Correlations were done between ER-36 expression and the clinico-pathological parameters and
other biomarkers using Pearsons chi-square test. Kaplan-Meier method was used to estimate distant metastasis free survival
(DMFS), disease free survival (DFS) and overall survival (OS) and the difference between the groups was evaluated with log-rank
test. Cox regression model was used to adjust for other prognostic parameters in the multivariate analysis.
Results
Median age at diagnosis was 56.9 years (range: 30 to 87 years). The maximum tumor size was larger than 2 cm in 57.5% of
cases and axillary lymph nodes (LN) were positive (N1a to N3) in 52.5% of cases. 16.3% of the patients had SBR grade I, 44.4%
had grade II and 39.4% had grade III tumors. ER66 was positive in 91.2%, PgR in 74.7% and HER2 was over-expressed in 15%
of the cases. High ER-36 expression in the cell membrane was observed in 65 patients (40.6%). ER-36 expression was
independent of the ER66, PgR or HER2 expression and was not associated with age, tumor size, SBR grade or axillary LN
invasion. There was no correlation between ER-36 expression and PI3K, pAKT, p4EBP1, pS6RP or LKB1 expression.
ER-36 expression in tumor cells was a predictor of poor prognosis regarding DMFS (HR=2.02; 95% CI: 1.2 to 3.4; p=0.008),
DFS (HR=1.7; 95% CI: 1.05 to 2.7; p=0.031) and OS (HR=1.8; 95% CI: 1.02 to 3.2; p=0.043). In the multivariate analysis and
after adjustment for age, tumor size, SBR grade and LN invasion, ER-36 remained an independent predictor of shorter DMFS
(p=0.016) and DFS (p=0.052) in addition to SBR grade and axillary LN metastasis. The ER-36 expression predicted shorter
DMFS for patients who received tamoxifen as the only adjuvant systemic treatment (p=0.022) and also for those who received
other hormonal therapy and adjuvant chemotherapy (p=0.039).
Conclusion
Immunohistochemically detected membranous ER-36 expression can be a poor prognostic factor for patients with operable
breast cancer that is independent from the traditional clinico-pathologic parameters and from PI3K/AKT/mTOR pathway activation
status.

Title: TIF1 interacts with the TGF1/SMAD4 signaling leading to poorer outcome in operable breast cancer
Loay Kassem1, Laurent Fattet2, Jonathan Lopez2, Goulvent Thibault2, Emilie Lavergne3, Sylvie Chabaud3, Nicolas Carrabin4,
Nicolas Chopin4, Thomas Bachelot5, Germain Gillet2, Isabelle Treilleux6 and Ruth Rimohk2. 1Cairo University Teaching Hospital,
Cairo, Egypt; 2Inserm U1052/CNRS UMR5286/Universit de Lyon, Centre de Recherche en Cancrologie de Lyon, Centre Lon
Brard, Lyon, France; 3Biostatistics Unit, Centre Lon Brard, Lyon, France; 4Centre Lon Brard, Lyon, France; 5Centre Lon
Brard, Lyon, France and 6Centre Lon Brard, Lyon, France.
Body: Background
The Transforming growth factor (TGF) signaling has a paradoxical role in cancer development and outcome. It protects against
tumorigenesis by inhibiting cell growth and promoting apoptosis, but at advanced stages, it promotes tumor progression. Besides,
the prognostic significance of the TGF1, SMAD4 in breast cancer patients is also an area of many contradictions. Transcriptional
intermediary factor 1 (TIF1) is thought to interact with the TGF/SMAD signaling through different mechanisms. Our study
aimed at defining the prognostic significance of TGF1, SMAD4 and TIF1 expression in breast cancer patients in addition to
detection of possible interactions among those markers that might affect the outcome and explain the contradictory results.
Methods
Immunohistochemistry was performed on TMA blocks prepared from samples of 248 operable breast cancer patients who
presented at CLB between 1998 and 2001 using. The intensity and the percentage of stained tumor cells were integrated into a
single score (0-6) and a cutoff was defined for high or low expression for each marker. Correlation was done between the TGF1,
SMAD4 and TIF1 expression with the clinico-pathologic parameters using Pearsons chi-square test. Kaplan-Meier method was
used to estimate distant metastasis free survival (DMFS), disease free survival (DFS) and overall survival (OS) and the difference
between the groups was evaluated with log-rank test.
Results
223 cases were assessable for TIF1, 204 for TGF1 and 173 for SMAD4. Median age at diagnosis was 55.8 years (range: 27 to
89 years). Tumors were larger than 20 mm in 49.2% and 45.2% had axillary lymph node (LN) metastasis (N1a to N3). 19.4% of
the patients had SBR grade I tumors, 46.8% grade II tumors and 33.9% grade III tumors. ER was positive in 85.4%, PR in 75.5%
and Her2-neu was over-expressed in 10% of the cases. Nuclear TIF1, cytoplasmic TGF1, nuclear and cytoplasmic SMAD4
staining was high in 35.9%, 30.4%, 27.7% and 52.6% respectively. TIF1 expression was associated with younger age (p=0.006),
higher SBR grade (p<0.0001), more ER negativity (p=0.035), and tumors larger than 2 cm (p=0.081), while TGF1 was not
associated with any of the traditional prognostic factors.
TGF1 expression in tumor cells was a marker of poor prognosis regarding DMFS (HR=2.28; 95%CI: 1.4 to 3.8; p=0.002), DFS
(HR=2.00; 95% CI: 1.25 to 3.5; p=0.005) and OS (HR=1.89; 95%CI: 1.04 to 3.43; p=0.037). TIF1 expression carried a tendency
towards poorer DMFS (p=0.091), DFS (p=0.143) and OS (p=0.091). In the multivariate analysis TGF1 remained an independent
predictor of shorter DMFS, DFS and OS after adjustment for age, tumor size, SBR grade and LN invasion. Moreover, the
prognostic significance of TGF1 was more obvious in the TIF1 high patient subgroup than in the patients with TIF1 low
expression. The subgroup expressing both markers had the worst DMFS (HR=3.2; 95%CI: 1.7 to 5.9; p<0.0001), DFS (HR=3.02;
95%CI: 1.6 to 5.6; p<0.0001) and OS (HR=2.7; 95%CI: 1.4 to 5.4; p=0.005).
Conclusion
There is a crosstalk between the TIF1 and the TGF1/SMAD4 signaling that deteriorate the outcome of operable breast cancer
patients and when combined together they can serve as an effective prognostic tool for those patients.

Title: Reappraisal of conventional risk stratification for local recurrence based on clinical outcomes in 285 resected phyllodes
tumors of the breast
Cha Kyong Yom1, Wonshik Han2, Sung-Won Kim3, So Yeon Park4, In-Ae Park5 and Dong-Young Noh2. 1Myongji Hospital,
Goyang, Korea; 2Seoul National University College of Medicine, Seoul, Korea; 3Seoul National University Bundang Hospital,
Seongnam, Korea; 4Seoul National University Bundang Hospital, Seongnam, Korea and 5Seoul National University College of
Body: Background A second resection has been recommended, to ensure a surgical margin of 1cm for the effective treatment
of PTB, but the outcomes of an extensive series of cases casts this clinical approach in doubt.
Objective To identify the local recurrence risk factors of phyllodes tumor of the breast (PTB) and determine future optimal surgical
treatment according to verified risks.
Methods All cases given a diagnosis of PTB and resected between 1989 and 2008 were retrospectively evaluated.
Clinicopathologic data and clinical outcomes were analyzed and stratified according to the risks for local recurrence (LR).
Results All 285 cases occurred and 200 (70.2%) categorized as benign, 51 (17.9%) as borderline, and 34 (11.9) as malignant.
Median follow-up was 6.7 years and during follow-up, there were 20 LRs. All benign PTBs recurred as benign PTB lesions.
Mitosis (p = 0.007) and tumor size (p = 0.029) were independent prognostic factors for LR in multivariate analysis. Neither margin
status (p = 0.773) nor type of surgery (p = 0.922) had any significance for LR.
[table 1] Local relapse-free survival by Cox-regression
Subgroup of patients
No recurrence(%)
Recurrence(%)
Univariate
HR
Mitosis
Multivariate
p
HR
0.025
p
0.007
14/10HPF
170(96.0)
7(4.0)
59/10HPF
33(84.6)
6(15.4)
4.259
0.009
4.297
0.014
10/10HPF
15(88.2)
2(11.8)
3.461
0.122
12.211
0.007
Size
0.035
0.029
2cm
75(98.7)
1(1.3)
2cm< and 5cm
136(89.5)
16(10.5)
8.599
0.037
5.630
0.010
>5cm
51(98.1)
1(1.9)
1.564
0.752
0.360
0.519
Operation
0.957
0.922
WLE
229(93.1)
17(6.9)
VABS
23(92.0)
2(8.0)
1.247
0.769
1.396
0.814
Mastectomy
13(92.9)
1(7.1)
1.055
0.959
1.472
0.744
Margin
0.846
Clear
200(93.5)
14(6.5)
Close/Involvement
39(92.9)
3(7.1)
1.132
0.773
1.253
In the risk stratification for LR, PTBs sized 2 - 5cm with 10 mitoses / 10 HPF had the highest LR rate (60%) compared with all
other groups (p < 0.001).
Conclusions Only PTB with 2 - 5 cm and frequent mitoses, is it recommended to follow to ascertain a wide excision and clear
margin of 1 cm, it necessary by means of a 2nd surgery could be considered to avoid the risk of LR in this distinct group.

Title: No Show

Title: Prospective comparison of conventional clinicopathological factors, uPA/PAI-1 and EndoPredict-clin score (EPclin) for
adjuvant clinical decision making in ER-positive, HER2-negative breast cancer: Progesterone receptor expression is strongly
associated with risk stratifikation according to EP clin
Johannes Ettl1, Kirsten Groe Lackmann1, Alexander Hapfelmeier2, Evelyn Klein1, Stefan Paepke1, Christoph Petry3, Katja
Specht4, Laura Wolff1, Heinz Hfler4 and Marion Kiechle1. 1Klinikum rechts der Isar, Technische Universitt Mnchen, Munich,
Germany; 2Institute of Medical Statistics and Epidemiology, Klinikum rechts der Isar, Technische Universitt Mnchen, Munich,
Germany; 3Sividon Diagnostics GmbH, Cologne, Germany and 4Institute of Pathology, Klinikum rechts der Isar, Technische
Universitt Mnchen, Munich, Germany.
Body: Background: Recently, EPclin, a second generation multigene test has been introduced into clinical practice. Aim of this
prospective study is to compare conventional clinicopathological factors, uPA/PAI-1 and EPclin and to determine, how these
parameters influence adjuvant treatment decisions. Methods: 217 consecutive cases of ER-pos, HER2neg, breast cancer cases
were enrolled in this study. Conventional clinicopathological factors, uPA/PAI-1 and EPclin were obtained by central pathology
assessment of the patients' surgical specimen. Decision for type of adjuvant treatment was made twice once with and once
without EPclin result- after case discussion in an interdisciplinary tumor conference. Results: 217 Patients (pt) have been
evaluated. Tumor grading within the presented cohort was as follows: G1: 44 pt (20%), G2: 146 pt (68%), G3: 27 pt (12%). 59 pt
(27%) had positive axillary lymph node involvement. 57 pt (26%) were pT1a/b. EPclin could be assessed in all patients. 85 pt
(39%) were classified as "high risk" whereas 132 pt (61%) were classified as "low risk". uPA/PAI was obtained from 123 pt (57%).
64 pt (52%) out of these had high uPA/PAI-1 levels whereas 59 pt (48%) showed low uPA/PAI-1 levels. Level of progesterone
receptor (PR) expression was obtained from 216 pt. 39 pt (18%) had low or absent PR expression (PR 20%). 26 pt (67%) of
this low PR expression group were classified as EPclin high risk. Furthermore 118 pt (67%) of the high PR expression group (177
pt) were classified as "low risk" via EPclin (p<0,0001). In 89 cases (41%) treatment decision was influenced by EPclin: In 73pt
(34%) adjuvant chemotherapy (ctx) was omitted whereas in 16 pt (7%) ctx was added following the EPclin result. Conclusions:
This prospective study shows for the first time, that PR expression is strongly associated with risk score evaluated by EPclin
testing. In comparison with uPA/PAI-1 and conventional clinicopathological factors EPclin is the more powerful tool to help reduce
unnecessary adjuvant chemotherapy.

Title: Clinical significance of adiponectin receptor 1 (AdipoR1) expression in invasive breast cancer
Younok Lee1, Kang Yool Lee1, Young Ah Lim2, June Ho Kim1 and Lee Su Kim1. 1Hallym University Sacred Heart Hospital,
Anyang, Gyeonggi, Korea and 2Hallym University Dongtan Sacred Heart Hospital, Dongtan, Gyeonggi, Korea.
Body: Background: Recent studies have demonstrated that obesity is associated with an increased risk of breast cancer, but the
mechanisms underlying this relationship remain to be fully elucidated. Adiponectin is one of major adipokines secreted from
adipose tissue. Adiponectin serum levels have been shown to be inversely correlated with breast cancer risk. This protein is
believed to act through AdipoR1 and has been suggested to play an important role in cancer development. The purpose of this
study was to quantitatively evaluate the expression of AdipoR1 in invasive breast cancer tissue compared to normal breast tissue.
And then, we analyzed clinical significance of AdipoR1 in invasive breast cancer.
Materials and Methods: Tissues were obtained from 269 patients who underwent curative surgery with no prior treatment for
invasive ductal carcinoma from Jan. 2003 to Dec. 2008 in Hallym Sacred Heart Hospital. A tissue microarray (TMA) containing
269 invasive ductal carcinomas as well as 269 adjacent normal breast tissues was established from paraffin-embedded archived
tissue for further analysis. AdipoR1 expression was investigated in epithelium and stroma by immunohistochemistry, using 1:1600
dilution of rabbit anti-adiponectin receptor antibody, and correlated with clinical and pathologic tumor parameters.
Results: In 269 patients, median follow-up period was 57 months. AdipoR1 was detected in epithelial and stromal component of
both normal breast and invasive ductal carcinoma tissues. In epithelium, immunoreactivity for AdioR1 was much lower in cancer
tissue than normal one (24.5% versus 72%). This trend was quite similar in stroma, although the gap between cancer and normal
was a bit narrow (48.7% versus 77.6%). AdipoR1 was more expressed in stroma than in epithelium among invasive breast cancer
(48.7% versus 24.5%). In clinicopathologic features, mean age at diagnosis of AdipoR1 expression group in both epithelium and
stroma was older than negative group. Furthermore, in epithelial component, HER-2/neu overexpression rate was slightly higher
in AdipoR1 negative group than in positive one (27.8% versus 15%, p=0.059). And, Ki67 was more expressed in AdipoR1
negative group than in positive one (49.5% versus 34.5%, p=0.051). However, in stroma component, there was no other
difference between AdipoR1 expression and clinicopathologic parameters. In survival analysis, AdipoR1 expression group in
stroma showed significantly better disease free survival (DFS) than negative group (p=0.02). DFS curve according to AdipoR1
expression in epithelium showed a quite similar trend with ones in stroma (p=0.06).
Conclusions: This study showed that AdipoR1 expression was suppressed in both epithelium and stroma of invasive breast
cancer tissue compared to normal breast tissue, and AdipoR1 expression group in stroma showed significantly better DFS than
negative group. Although it is earlist days yet to make conclusion, loss of AdipoR1 expression in stroma, possibly epithelium
either, appears to play a role of independent risk factor to predict disease progression, somehow.

Title: Quantitative measurement of HER2 levels by multiplexed mass spectrometry from FFPE tissue predicts survival in patients
treated with anti-HER2 based therapy
Paolo Nuciforo1, Sheeno Thyparambil2, Claudia Aura1, Ana Garrido-Castro1, Vicente Peg1, Jose Jimenez1, William Hoos2, Jon
Burrows2, Todd Hembrough2, Jose Perez-Garcia1, Javier Cortes1 and Maurizio Scaltriti3. 1Vall d'Hebron Institute of Oncology,
Barcelona, Spain; 2Oncoplex Diagnostics, Rockville, MD and 3Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Introduction
Approximately 20% of breast cancer patients overexpress HER2 and are treated with anti-HER2 therapies. However, there is a
great deal of disparity of HER2 levels in the patients that are classified as HER2 positive (IHC3+). Techniques like FISH or IHC
do not allow for HER2 quantification and a significant proportion of patients are wrongly classified as HER2 positive. Liquid
Tissue-Selected reaction monitoring (LT-SRM) is a multiplexed mass spectrometric technique that can objectively quantify levels
of Her2 and other targets from formalin fixed paraffin embedded (FFPE) sections. Given the different available anti-HER2
therapies (trastuzumab, TDM1, lapatinib and pertuzumab) with different modes of action, it would be beneficial for a clinician to
understand the levels EGFR and HER3 so as to personalize the therapy. In this work, we have used LT-SRM to quantitate HER2,
EGFR and HER3 from FFPE samples (one slide) of patients treated with anti-HER2 agents and correlated the levels of these
proteins to clinical outcome.
Methods
FFPE sections from 60 HER2 positive (IHC3+) primary breast cancer patients were microdissected and proteins were solubilized
and digested by trypsin in Liquid Tissue buffer. Of the 60 samples, 24 were from metastatic setting and 36 from adjuvant
setting. After trypsin digestion, internal standards were added and absolute quantitation for multiple proteins was performed using
selected reaction monitoring (SRM) mass spectrometry. In addition to LT-SRM, FISH for HER2 was also conducted.
Results
HER2 quantitation by LT-SRM revealed receptor level ranges from 283 to 14938 amol/g. ROC analysis was conducted and a
cut-off of 2758.75 amol/g gave the optimal sensitivity and specificity. Survival analysis revealed statistically significant DFS (4.40
years vs 3.38 years; p=0.013) and OS (4.43 years vs 4.03 years; p= 0039) in patients expressing 2758.75 amol/g in the
adjuvant setting and also statistically significant OS (5.51 years vs 3.37 years; p=0.037) in the metastatic setting. Correlation of
HER2 FISH and levels of HER2 is ongoing. Approximately 41% of samples expressed EGFR (range 45 to 2317 amol/g) and
similarly 51% of the samples expressed HER3 (range 84 to 360 amol/g) with 18% of samples expressing all three targets.
Correlation of EGFR/HER3 expression with clinical outcome is ongoing.
Conclusion
We used an objective multiplex non-antibody based method to quantify multiple targets from FFPE tissue. Clinical correlation
analysis of HER2 revealed improved OS and DFS in samples with high HER2 protein levels. Currently, we are expanding these
studies to a larger set of samples and taking into account also the expression of other markers such as EGFR and HER3. This
approach can potentially identify those tumors that are more dependent on these receptors for survival and also those patients
that are exquisitely sensitive to anti-HER2 therapy.

Title: No Show

Title: Inverse relationship between ER+/PR- breast cancer and OncotypeDx scores and concordance between IHC and RT-PCR
Lubna N Chaudhary1, Zeeshan Jawa1, Aniko Szabo1, Alexis Visotcky1 and Christopher R Chitambar1. 1Medical College of
Wisconsin, Milwaukee, WI; 2Medical College of Wisconsin, Milwaukee, WI; 3Medical College of Wisconsin, Milwaukee, WI;
4
Medical College of Wisconsin, Milwaukee, WI and 5Medical College of Wisconsin, Milwaukee, WI.
Body: Background
Estrogen receptor-positive/progesterone-receptor negative breast cancer (ER+/PR- BC) is recognized as a distinct clinical entity
that is associated with potentially more aggressive tumor biology and early resistance to endocrine therapy when compared to
ER+/PR+ BC. The OncotypeDx assay provides risk stratification and prognostic information for hormone receptor (HR)-positive
invasive BC; however, the association between the PR status determined by immunohistochemistry (IHC) and by OncotypeDx
genomic analysis and the impact on the recurrence score (RS) is less clear.
Methods
We designed an analysis to determine whether OncotypeDx scores differ between ER+/PR+ and ER+/PR- tumors. Three
hundred and fifty patients (pts) with HR positive, node-negative invasive BC who underwent OncotypeDx testing at our institution
between Dec 2006-Oct 2013 were included in our study. We examined whether there was discordance in the HR status on IHC
staining reported by our pathologists and the RT-PCR analysis reported as part of the OncotypeDx assay. The data were
analyzed by ANOVA-F test and t-test. Chi-square test was used to analyze the discrepancies between the two methods of tumor
genomic profile assessment.
Results
The mean age at diagnosis was 58 yrs (SD 10.1 yrs). Three hundred and one pts had ER+/PR+ and 47 pts had ER+/PRtumors by IHC. Pt characteristics are shown in Table 1. PR- tumors had significantly higher OncotypeDx scores than PR+ tumors
(24.7 8.53 vs 17.3 7.38; Mean SD; p <0.001), predicting a greater 10-yr risk of distant recurrence. Two hundred and eighty
four pts had HR status reported by OncotypeDx assay. Comparison of IHC and OncotypeDx assay revealed that 20 pts (8.1%)
who were ER+/PR+ by IHC had ER+/PR- tumors by RT-PCR. Of the ER+/PR- cases by IHC, 12 (31.6%) were ER+/PR+ and 2
(5.3%) pts were ER-/PR- by RT-PCR (p <0.001, Table 2). Independent evaluation for ER and PR expression showed a
concordance of 99.3% for ER and 88.7% for PR between the two methods.
Conclusion
Our study shows that ER+/PR- BC tumors are associated with a significantly higher OncotypeDx scores; this interprets into a
higher risk of recurrence. Our data also show that there was a high concordance of 99.3% for ER between IHC and RT-PCR
analyses whereas the concordance for PR was much lower at 88.7%.
Patient characteristics
Variables
ER+/PR+ N=303 (%)
ER+/PR- N=47 (%)
58
59
207 (68.3)
39 (83)
PM
96 (31.7)
8 (17)
301
47
Mean SD
17.3 7.38
24.7 8.53
IDC
248 (82.4)
44 (93.6)
ILC
50 (16.6)
3 (6.4)
Age
Mean
M status
OncotypeDx score
Histology
Stage
T1a
2 (0.7)
2 (4.2)
T1b
49 (16.2)
9 (19.1)
T1c
174 (57.5)
25 (53.3)
T2
74 (24.5)
11 (23.4)
T3
2 (0.7)
Lumpectomy
198 (65.6)
30 (63.8)
Mastectomy
102 (33.8)
17 (36.1)
Yes
93 (30.4)
32 (68.1)
No
210 (69.6)
15 (31.9)
AI
167 (55.4)
30 (65.2)
Tam
89 (29.2)
9 (19.5)
Sequential AI & Tam
36 (10.9)
4 (8.8)
Surgery
Chemotherapy
Endocrine tx
N= number, SD= standard deviation, M= menopausal, PM= premenopausal, IDC= invasive ductal carcinoma, ILC= invasive
lobular carcinoma, AI= aromatase inhibitor, Tam= tamoxifen, = statistically significant
Comparison between IHC and RT-PCR

IHC
RT-PCR
ER+/PR+ N(%)
ER+/PR- N(%)
P value
ER+/PR+
226 (91.9)
12 (31.6)
< 0.001
ER+/PR-
20 (8.1)
24 (63.2)
ER-/PR-
2 (5.3)

Title: Routine histopathological variables predict oncotype DX risk categories
Hidetaka Kawabata1, Keiichi Kinowaki2, Nobuko Tamura1, Yoko Kobayashi1, Masami Kadowaki1, Daishu Miura1, Toshimi Takano3,
Akihiko Shimomura4, Tsuguo Iwatani5, Yuko Kitawaki2 and Takeshi Fujii2. 1Toranomon Hospital, Tokyo, Japan; 2Toranomon
Hospital; 3Toranomon Hospital; 4National Cancer Center Hospital, Tokyo, Japan and 5BioMedical Research Centre, University
College London, London, United Kingdom.
Body: Background: Oncotype DX(ODX) is a clinically validated, commercially available multi-gene assay that provides
prognostic and predictive information in estrogen receptor(ER)-positive, Her2-negative breast cancer. Because health care
system does not cover this expensive assay in many countries, physicians and researchers try to use routine histopathological
variables predict ODX Recurrence Score (RS). Hyunseok Kim et.al presented at ASCO that Estimated Recurrence Score (ERS)
model could identify those patients most likely to benefit from including ODX RS results in therapeutic decision-making (J Clin
Oncol 32:5s, 2014 suppl; abstr 559).
Methods: We retrospectively reviewed histopathological slides between July 2007 and April 2014 from Toranomon Hospital
patients (n=149) with early stage ER-positive, Her2-negative breast cancer, for whom ODX was ordered. We developed an
original linear regression model using routine histopathological markers (Allred Score for progesterone receptor (PR), nuclear
grade(NG), and Ki67) to calculate an Estimated Recurrence Score (ERS), and correlated it with the observed ODX RS assay
result. ERS=20.46+0.298xKi671.48xPR+1.41xNG predicts the ODX RS with an R of 0.48. In order to internally validate this
model in the Toranomon Hospital cohort in Japan, we adapted Hyunseoks algorithm.
Results: 132 patients had an observed RS 25 (89%) and 17 patients had an observed RS above 25 (11%). When the ERS was
< 21 (n=124), we accurately classified 97% of them (120) who were found to have a low risk (observed RS 25). Similarly, 83%
of those (5/6) with an ERS > 30 fell into a high risk category with observed RS > 25.
Table1 (TAILORx risk categories)
Toranomon cohort
ODX 25
ODX RS>25
Total
ERS < 21
120
124
ERS(21-30)
11
19
ERS(> 30)
Total
132
17
149
We also presented the comparison of ERS and observed RS based on original risk categories.
Table2 (Original risk categories )
Toranomon cohort
ODX RS<18
ODX RS 18-30
ODX RS>30
Total
ERS(<18)
66
33
99
ERS(18-30)
17
25
44
ERS(>30)
Total
83
59
149
Conclusions: We developed an ERS model to find those patients most likely to benefit from ODX RS results in clinical practice
setting. Although further external and prospective validation is mandatory, preliminary result supports the usefulness of this ERS
model.

Title: No Show

Title: Validation of the CancerMath prognostic tool for breast cancer in Southeast Asia
Nur Aishah Taib4, Hui Miao1, Helena M Verkooijen1,2, Mikael Hartman1,3, Hoong-Seam Wong5, Cheng-Har Yip4, Ern-Yu Tan6,
Soo-Chin Lee7 and Nirmala Bhoo-Pathy5. 1Saw Swee Hock School of Public Health, National University of Singapore, Singapore;
2
University Medical Center, Utrecht, Netherlands; 3National University Hospital System, Singapore, Singapore; 4University of
Malaya, Kuala Lumpur; 5National Clinical Research Centre, Kuala Lumpur, Malaysia; 6Tan Tock Seng Hospital, Singapore,
Singapore and 7National University Hospital System, Singapore, Singapore.
Body: Background: CancerMath is a set of web-based prognostic tools for non-metastatic breast cancer patients. It predicts
probability of nodal and nipple involvement as well as overall and cancer-specific survival with and without adjuvant therapy at
any of the 15 years after diagnosis. But it has not been externally validated outside United States. This study assessed its
performance in a Southeast Asian setting.
Methods: Using Singapore Malaysia Hospital Based Breast Cancer Registry, clinical information were retrieved from 7064 Stage I
to stage III breast cancer patients who were diagnosed between 1990 and 2011 and underwent surgery. Probability of positive
nodes and overall survival at any of the 15 years after diagnosis were computed using the CancerMath algorithm. Predicted and
observed outcomes were compared. Discriminative ability was evaluated by area under a receiver operating characteristic curve
(AUC). And calibration was assessed by plotting observed survival against predicted survival for each decile of the predicted
survival and a 45 degree line was added to illustrated perfect agreement.
Results: Nodal status calculator predicted 40.6% of patients to be node positive which was similar to the observed 43.6%. The
AUC was 0.71 (95% CI, 0.70-0.72). For outcome calculator, the observed and predicted overall survivals was 83.4% vs 87.3% at
5 year after diagnosis and 70.4% vs 75.3% at 10 year after diagnosis. The difference appeared smaller for patients from
Singapore (5-year and 10-year predicted-observed= 2.5% and 0% respectively) comparing to patients from Malaysia (5-year and
10-year predicted-observed= 5.8% and 8.0% respectively), as well as for patients diagnosed in more recent years. The AUC for
5-year and 10-year overall survival was 0.77 (95% CI: 0.75-0.79) and 0.74 (95% CI: 0.71-0.76). For cases with relatively good
and moderate prognosis, CancerMath predictions were similar to observed outcome as the plotted dots in the calibration plot
were close to the 45 degree reference line from the fourth decile onwards. For therapy calculator, overestimation of survival
persisted for most demographic, pathologic and treatment subgroups, especially for women younger than 40 years, with larger
and high-grade tumor and with more lymph nodes involved. The AUC for 5-year overall survival is 0.73 (95% CI: 0.70-0.77). And
with the exception of two groups, all plotted dots were below the 45 degree reference line in the calibration plot .
Conclusion: The discriminative performance and calibration of CancerMath was fair in this validation study. The greater
discordance observed in Malaysian patients might be explained by different life expectancy, adverse tumor characteristics,
compliance to treatment and lifestyle after diagnosis between Singapore, Malaysia and United States. The results suggested that
direct application of CancerMath should only be limited to certain subgroups in Southeast Asia.

Title: The association of serum albumin and glycated hemoglobin with all cause mortality in patients with breast cancer
Vinay N Minocha1, Kavita Pal1 and Robert Zaiden1. 1University of Florida College of Medicine, Jacksonville, FL.
Body: Background: There has been increasing data within the oncology literature to support an inverse relationship between
serum albumin levels and survival with breast cancer. Various studies have also indicated that diabetes is associated with a
higher mortality in patients with breast cancer. Given the need to develop objective ways to identify patients with a poorer than
expected outcome, the aim of this study is to determine whether serum albumin and glycated hemoglobin at diagnosis bears any
relationship with all cause mortality in patients with all stages of breast cancer.
Methods: We performed a retrospective study of patients diagnosed with breast cancer of all stages at University of Florida
Health Center, Jacksonville between August 2007 and October 2013. The data for this study was obtained from a combination of
the hospital tumor registry and the electronic medical record. We collected data including age, sex, stage of cancer, histological
type, tumour size and grade, lymph node status, estrogen (ER), progesterone (PR) and HER2/Neu receptor status. We collected
baseline serum albumin levels and glycated hemoglobin (HbA1C) values recorded within 3 months of diagnosis of breast cancer.
Wilcoxons rank sum test was used to test whether cancer stage differed between survival outcomes. Spearman correlation was
used to assess the relationship between cancer stage and serum albumin and HbA1C at time of diagnosis. Logistic regression
was used to model the impact of HbA1c, serum albumin and age at time of diagnosis on the probability of survival.
Results: The number of patients with breast cancer included in our study was 294. Of these patients, 219 patients had serum
albumin values available at diagnosis and 69 patients had glycated hemoglobin values at diagnosis. The correlation between
serum albumin and stage is statistically significant (P < 0.0001).
Correlation between HbA1C, Serum albumin, and Stage
Spearman Correlation Coefficients\ P-Value\ Number of Observations
Serum Albumin
Stage
-0.378
<0.0001
217
HbA1C
0.004
0.9731
69
The negative correlation means that higher levels of serum albumin are associated with lower stages. Serum albumin at
diagnosis had a significant effect on the probability of survival (p=0.0005). For every unit increase in serum albumin, the odds of
survival increase (OR=4.938, 95% CI: 2.023, 12.050). Alternatively, for each unit decrease the odds of dying increase by about
five fold. Hypoalbuminemia has a significant effect on survival (p=0.0020). When serum albumin is >3.5, the odds of death
decrease (OR=0.176 95% CI: 0.058, 0.528). Neither HbA1c nor age had a significant effect on survival (P=0.7635, 0.0858
respectively).
Odds Ratios for probability of survival
95% CI
OR
Lower Limit
Upper Limit
p-value
Serum Albumin
4.938
2.023
12.050
0.0005
HbA1C
1.092
0.614
1.942
0.7635
Age
0.974
0.946
1.004
0.0858
Conclusion: In the present study, a low serum albumin level at the time of diagnosis, either as a continuous or categoric variable,
was significantly associated with poorer survival in patients with breast cancer. Furthermore, a lower serum albumin is associated
with a higher stage of breast cancer. In contrast, glycated hemoglobin (HbA1C) at the time of diagnosis did not have a significant
effect on survival.

Title: Prognostic models in male breast cancer
Carmen C vanderPol1, Miangela M Lacle1, Arjen J Witkamp1, Robert Kornegoor2, Miao Hui3, Christine Bouchardy4, Elsken
vanderWall1, Helena M Verkooijen1 and Paul J vanDiest1. 1University Medical Center, Utrecht, Netherlands; 2Gelre Ziekenhuizen,
Apeldoorn, Gelderland, Netherlands; 3Saw Swee Hock School of Public Health, National University of Singapore, Singapore and
4
Institute for Social and Preventive Medicine, Geneva University, Geneva, Switzerland.
Body: Male breast cancer (MBC) is a rare disease and its treatment is largely extrapolated from its female counterpart.
Accurate prognostication is essential for advising on adjuvant systemic treatment and informing patients. Several predictive
models are available for female breast cancer (FBC) including, subsequently, the Morphometric Prognostic Index (MPI),
Nottingham Prognostic Index (NPI), Adjuvant! and Predict. The aim of this study was therefore to compare the prognostic
performance of these models in a group of 166 early MBC patients.
The MPI describes a "good"- (MPI<0,60) and a "poor" prognostic group (MPI>=0,60) by using a formula with mitotic activity index,
tumour size and lymph node status. The NPI is calculated by a formula including size, number of lymph nodes with metastases
and tumour grade and divides patients into three groups; "good-" (NPI <= 3,4), "intermediate-" (3,4<NPI>=5,4) and "poor"
prognosis (NPI>5,4). For the programs Adjuvant!Online and Predict, similar groups with "good"-, "intermediate"- and "poor"
prognosis were defined by using tertiles. The prognostic performance of each test was studied by using the logrank-test and
comparison between the models was done by using C-statistics.
The mean age was 66,4 years old and the median survival was 4,6 years with a mean of 5,8 years overall survival.
Survival of the highest predicted group was higher (MPI: 87%, NPI: 90%, Adjuvant!Online: 91% and Predict: 88%) than for the
moderate groups (NPI: 76%, Adjuvant!Online: 77% and Predict: 75%) and lowest for the poor predicted groups (MPI: 51%, NPI:
43%, Adjuvant!Online: 45% and Predict: 42%), with p-values that were highly statistically significant.
In terms of discrimination, all models were moderately able to discriminate between good and poor survivors (C-statistics; MPI;
0,674, NPI; 0,678, Adjuvant!Online; 0,717 and Predict; 0,711).
In conclusion, the MPI, NPI, Adjuvant! and Predict, prognostic models that were originally developed and validated for FBC,
perform fairly well for MBC. These models may therefore help in MBC prognostication and decisions on adjuvant systemic
therapy.

Title: No Show

Title: No Show

Title: Epithelial-mesenchymal transition phenotype in triple negative breast cancer: ZEB1 as a potential biomarker for poor
clinical outcome
Min Hye Jang1, Hyun Jeong Kim1, Eun Joo Kim1, Hee Jin Lee3 and So Yeon Park1,2. 1Seoul National University Bundang Hospital,
Seongnam, Korea; 2Seoul National University College of Medicine, Seoul, Korea and 3Asan Medical Center, Seoul, Korea.
Body: Background: Triple negative breast cancer (TNBC) is a heterogeneous group of disease. TNBC is closely related to
epithelial-mesenchymal transition (EMT) and breast cancer stem cell (BCSC) phenotype. Recent studies have shown that TNBC
can be classified into six subtypes including basal-like, mesenchymallike and mesenchymal stem-like. However, clinical
significance of EMT phenotype in TNBC is not clear.
Methods: We performed immunohistochemical analyses of EMT markers (expression of vimentin, smooth muscle actin,
osteonectin and N-cadherin; loss of E-cadherin), BCSC markers (CD44+/CD24- and ALDH1) and EMT inducers (CD146 and
ZEB1) in 173 TNBCs using tissue microarrays, and correlated their expressions with clinicopathologic features of the tumor.
Results: Expression of vimentin, CD44+/CD24- and CD146 was significantly higher in basal-like TNBCs (TNBC with expression
of CK5/6 and/or EGFR) than in non-basal-like TNBCs. Expression of EMT markers was commonly correlated with high histologic
grade, CD44+/CD24- phenotype and metaplastic carcinoma. CD146 expression was related to the expression of EMT markers
and CD44+/CD24- phenotype. ZEB1 expression showed an association with the expression of smooth muscle actin, but not with
BCSC markers. And it was closely correlated with high histologic grade and metaplastic carcinoma. In survival analyses, although
expression of EMT and BCSC makers was not associated with the survival of the patients, ZEB1 expression was found to be an
independent prognostic factor for poor disease-free survival of the patients.
Conclusion: EMT phenotype can be a signature of certain subgroup of TNBC. Especially, ZEB1 expression can be used as a
potential biomarker to define a subgroup of TNBC associated with poor clinical outcome.

Title: No Show

Title: Pretreatment neutrophil/lymphocyte ratio and overall survival in African American and white breast cancer patients
Joseph C Rimando1, Jeff Campbell1, Jae Hee Kim1 and Sangmi Kim1. 1Georgia Regents University Augusta, Augusta, GA.
Body: Previous studies have shown that the pretreatment neutrophil/lymphocyte ratio (NLR) is an independent predictor of
mortality in breast cancer patients. Our aim was to further study the relationship between pretreatment NLR and overall survival in
African American and white breast cancer patients treated at an academic cancer center. Electronic medical records were
reviewed for 589 patients treated between 2002 and 2011, and pretreatment NLR data, determined at an average of 12 days prior
to the initiation of cancer treatment, were available from 217 African American and 218 white patients. Other clinical and patient
data were obtained from the hospital tumor registry, with annual follow-up for vital status. There were a total of 102 deaths over a
mean follow-up of 59 months. For data analysis, patients were divided into quartiles based on their NLR (Q1: <1.54; Q2:
1.542.0; Q3: 2.02.79; and Q4: 2.79). Patients in the highest quartile of pretreatment NLR showed an increase in overall
mortality compared to those in the lowest quartile, with a hazard ratio (HR) of 2.4 (p<0.005). After adjustment for age, tumor
stage, and grade, the pretreatment NLR remained significantly associated with overall mortality (HR=2.1, p=0.03). Further
adjustment for race, body mass index (BMI) and smoking increased the effect size (HR=3.2, p<0.01), but only half patients
remained in the analysis due to incomplete data on BMI and smoking. In our study, race, but neither BMI nor smoking, was
associated with pretreatment NLR values: African American patients had higher pretreatment NLR values than white patients
regardless of stage. In a stratified analysis by race, HRs for overall mortality associated with the highest quartile of pretreatment
NLR versus the remaining three quartiles combined were 2.5 (p<0.01) among African Americans and 1.6 (p=0.24) among white
patients, although the interaction between race and pretreatment NLR was not statistically significant. Our data confirm previous
findings that pretreatment NLR, a marker of systemic inflammatory response, is unfavorably associated with overall survival in
breast cancer patients, and also suggest that the association may be more evident among African American patients whose
levels were generally higher than white breast cancer patients. Future studies are warranted to investigate the prognostic utility of
pretreatment NLR in different racial groups and whether a heightened systemic inflammatory response is an underlying
contributor to racial difference in breast cancer outcomes.

Title: Pathological response and its impact on survival with neoadjuvant chemotherapy according to intrinsic subtype in locally
advanced breast cancers in North India: Is it different from the West?
Sushma Agrawal1, Punita Lal1 and Shaleen Kumar1. 1Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, Uttar
Pradesh, India.
Body: Background: Breast cancer patients commonly present in locally advanced stage (LABC) in our country .We propose to
correlate the pathological response to neoadjuvant chemotherapy (NACT) and its impact on survival based on the intrinsic
subtype of breast cancer in our population. Materials and methods: Consecutive patients of LABC who underwent NACT
(taxane and or anthracyclines based )followed by definitive surgery and radiotherapy during the period January 2007 to
December 2012 were grouped on the basis of intrinsic subtypes of tumor (Luminal A, Luminal B, Her-2 Type, Basal). The
pathological response to NACT in tumour as well as axillary nodes [complete response (pCR), partial response (pPR)] was
correlated with the disease free survival (DFS) and overall survival (OS)at 5 years in the four intrinsic subtypes using Kaplan
Meier Analysis. Results: 208 patients were the subject of this analysis. The median age of patients was 46 years (range 24-81
years), 46% were premenopausal and 54% postmenopausal, 42% right sided and 58% left sided. The clinical prechemotherapy
status of tumour and node at presentation was 15% T2, 40% T3, 45% T4 (9% T4a, 35% T4b, 1% T4c) 8% N0, 42% N1, 41% N2,
9% N3.The intrinsic subtype of our population at presentation was Luminal A (16%), Luminal B (23%), Her-2 Type (23%), Basal
(37%).The overall pCR rate to NACT in tumour and in node was 31% and 45%.The pCR rate in tumour according to intrinsic
subtype was 26%, 23%, 39% and 31.5% in Luminal A, Luminal B,Her-2 Type and Basal type respectively. The pCR rate in node
was 26%, 38%, 41.6% and 59% in Luminal A, Luminal B,Her-2 Type and Basal type respectively. At a median followup of 34
months (range 6-84 mo)the 5 year DFS and OS was significantly higher in patients achieving pCR tumour or pCR node in Her-2
type and Basal subtype (Table1) .Conclusions: The pCR rate to NACT in tumour or node seems to be considerably higher in our
population in Her-2 and basal subtypes than that reported in the western literature. pCR (tumour as well as node) as a surrogate
for both DFS and OS at 5 years in Her-2 and basal subtypes of breast cancer has been validated.
pCR rate and Impact of pCR (Tumour and node) according to intrinsic subtype on 5 yr DFS and OS
Intrinsic
Subtype
DFS (% at 5yr)(pCR
pCR(Tumour)(%) pCR(Node)(%) tumor vs pPR
tumor)(pvalue)
DFS (% at 5yr)(pCR
node vs pPR
node)(pvalue)
OS (% at
OS (% at 5yr)(pCR
5yr)(pCR Node vs
Tumour vs pPR
pPR node)(p
tumor)(p value)
value)
Luminal A 26
26
100 vs 67 (0.08)
88 vs 70 (0.49)
100 vs 72 (0.1)
85 vs 75 (0.7)
Luminal B 23
38
68 vs 56 (0.9)
87 vs 32 (0.11)
80 vs 57 (0.9)
85 vs 35 (0.35)
Her-2
Type
39
41.6
90 vs 44 (0.003)
83 vs 45 (0.009)
87 vs 59 (0.02)
90 vs 66 (0.03)
Basal
31.5
59
85 vs 50 (0.004)
46 vs 50 (0.26)
83 vs 50 (0.02)
975 vs 60 (0.19)
pCR(pathological CR),pPR(pathological PR)

Title: Ki-67 and p53 are useful factors to predict long term survival in low-risk luminal A breast cancer patients
Ha Woo Yi1, Se Kyung Lee1, Soo Youn Bae1, Jun Ho Lee1, Hyun-Chul Lee1, Won Ho Kil1, Jeong Eon Lee1, Seok Won Kim1 and
Seok Jin Nam1. 1Samsung Medical Center, Seoul, Korea.
Body: Overexpression of p53 is the most frequent genetic alteration in breast cancer. Recently, many studies have shown that
the expression of mutant p53 differs for each subtype of breast cancer and is associated with different prognoses. In this study,
we aimed to determine the suitable cut-off value to predict the clinical outcome of p53 overexpression and its usefulness as a
prognostic factor in each subtype of breast cancer, especially in luminal A breast cancer. Approval was granted by the
Institutional Review Board of Samsung Medical Center. We analyzed a total of 7739 patients who were surgically treated for
invasive breast cancer at Samsung Medical Center between Dec 1995 and Apr 2013. Luminal A subtype was defined as ER&PR
+ and HER2- and was further subclassified according to Ki-67 and p53 expression as follows: luminal A (Ki-67-,p53-), luminal A
(Ki-67+, p53-), luminal A (Ki-67 -, p53+) and luminal A (Ki-67+, p53+). Low-risk luminal A subtype was defined as negative for
both Ki-67 and p53 (luminal A (ki-67-, p53-)), and others subtypes were considered to be high-risk luminal A breast cancer. A
cut-off value of 10% for p53 was a good predictor of clinical outcome in all patients and luminal A breast cancer patients. The
prognostic role of p53 overexpression for OS and DFS was only significant in luminal A subtype. The combination of p53 and
Ki-67 has been shown to have the best predictive power as calculated by the area under curve (AUC), especially for long-term
overall survival.
Table 1. Factors associated with overall survival in luminal A subtype breast cancer patients (N=3918).
Univariable
Factors
Multivariable*
HR
P-value
95% CI for HR
HR
P-value
95% CI for HR
TM
- (ref.)
- (ref.)
- (ref.)
BCS
0.443
0.0318
0.211-0.932
No
- (ref.)
- (ref.)
- (ref.)
Yes
4.783
<0.0001
2.163-10.578
Positive
2.726
0.3269
0.367-20.244
Negative
- (ref.)
- (ref.)
- (ref.)
1&2
- (ref.)
- (ref.)
- (ref.)
3.793
0.0004
1.806-7.969
- (ref.)
- (ref.)
- (ref.)
- (ref.)
- (ref.)
- (ref.)
2.283
0.1697
0.703-7.416
1.784
0.3380
0.546-5.833
12.967
<0.0001
4.303-39.083
9.912
<0.0001
3.265-30.096
Positive
4.294
0.0032
1.631-11.303
2.587
0.0603
0.960-6.971
Negative
- (ref.)
- (ref.)
- (ref.)
- (ref.)
- (ref.)
- (ref.)
Positive
6.107
<0.0001
2.905-12.837
4.494
<0.0001
2.105-9.594
Breast surgery
Presence of LVI
RM
NG
AJCC Stage
Ki-67
p53
Negative
- (ref.)
- (ref.)
- (ref.)
- (ref.)
- (ref.)
- (ref.)
BCS, breast-conserving surgery; LVI, lymphovascular invasion; NG, nuclear grade; RM, resection margin; TM, total mastectomy;
NG, LVI and total mastectomy were not significant in multivariable Cox analysis;* Cox-proportional hazard regression model.
In this study, we have shown that overexpression of p53 and Ki-67 could be used to discriminate low-risk luminal A subtype in
breast cancer. Therefore, using the combination of p53 and Ki-67 expression in discriminating low-risk luminal A breast cancer
may improve the prognostic power and provide the greatest clinical utility.

Title: Characterizing the clinical presentation of individuals with pathogenic variants in a breast/ovarian cancer gene panel
Erica Vaccari1, Lauren Yackowski1, Melanie Hussong1, Patricia Murphy1, Maria L Cremona1, Jessica Booker1 and Kathleen
Hruska1. 1GeneDx, Gaithersburg, MD.
Body: Background: Hereditary breast and ovarian cancer (HBOC) is a common indication for referral to cancer genetic
counselors. Next generation sequencing panels allow for the efficient evaluation of many genes associated with increased risk of
these cancers. The purpose of this study is to compare clinical histories of those with pathogenic variants in high risk versus
low/moderate risk genes in order to determine which patients might benefit from more extensive testing by a panel approach.
Methods: We queried the results of patients tested at GeneDx for a panel of 21 genes causing increased breast and/or ovarian
cancer risk. Data regarding personal and family history of cancer provided on the test requisition forms were analyzed and
classified according NCCN guidelines for testing criteria for HBOC syndrome.
Results: Of 1709 individuals referred for testing, 146 (8.5%) tested positive for a pathogenic variant. Of these, 33% percent were
found to carry a pathogenic BRCA1/2 variant while 67% tested positive for a pathogenic variant in a gene other than BRCA1/2
(CHEK2: 17%; ATM: 12%; PALB2: 10%; BRIP1: 7%; BARD1: 3%; PTEN: 3%; each of FANCC, MSH2, MSH6, NBN, PMS2,
RAD51C, RAD51D: 2%; MLH1: 1%). Eighty-six percent of these individuals were affected with cancer. In the probands with a
pathogenic BRCA1/2 variant, 66% were diagnosed with breast cancer and 22% with ovarian cancer compared to the probands
with a pathogenic variant in a gene other than BRCA1/2 of whom 84% had a history of breast cancer and 17% had ovarian
cancer. The highest number of breast cancer diagnoses were found, in decreasing order, in association with pathogenic variants
in BRCA1 (20), CHEK2 (20), ATM (17), PALB2 (14), BRCA2 (13), and BRIP1 (12). The greatest number of ovarian cancers were
identified, in decreasing order, in association with pathogenic variants in BRCA2 (8), BRCA1 (3), CHEK2 (3), and ATM (3).
Furthermore, all of the individuals with pathogenic variants met NCCN guidelines for HBOC.
Conclusion: A high yield of pathogenic variants were found in genes other than BRCA1/2. Data analysis also shows that
individuals with a pathogenic variant in genes other than BRCA1/2 did not have a notably less severe clinical history than those
pathogenic variants in BRCA1/2. As all individuals tested meet NCCN guidelines for HBOC testing, panel testing should be
considered in this population.

Title: Spectrum of mutations identified in a 25-gene hereditary cancer panel for patients with breast cancer
Lavania Sharma1, Kelsey Moyes1, John Abernethy1, Heidi McCoy1, Jennifer Saam1, Michelle Landon1 and Richard Wenstrup1.
1
Myriad Genetic Laboratories, Inc, Salt Lake City, UT.
Body: Introduction: With the advancements of next generation sequencing, patients with a personal and/or family history of
cancer that may not be suggestive of one cancer syndrome may be offered testing for mutations in multiple cancer-predisposing
genes simultaneously. The focus of this analysis was to determine the spectrum of gene mutations observed in patients with a
personal history of breast cancer.
Methods: A commercial diagnostic laboratory database was queried for patients with a personal diagnosis of breast cancer who
underwent a 25-gene hereditary cancer panel from September 4, 2013 through April 17, 2014. The panel includes highly
penetrant cancer predisposing genes BRCA1, BRCA2, TP53, PTEN, MLH1, MSH2, MSH6, PMS2, EPCAM, APC, BMPR1A,
CDH1, CDKN2A, MUTYH, SMAD4, STK11 and moderately penetrant genes CHEK2, PALB2, ATM, NBN, BARD1, BRIP1, CDK4,
RAD51C and RAD51D. Sequencing and large rearrangement was performed for all the genes in the panel. All patient data
regarding clinical history was obtained by health care provider report on the test requisition forms.
Results: A total of 3584 patients with a personal history of breast cancer were identified. Of these, 13.8% met the NCCN
guidelines for genetic testing for both Hereditary Breast and Ovarian Cancer syndrome (HBOC) and Lynch syndrome (LS), while
80.3% met criteria for only HBOC and 1.2% met only for LS. 10.4% of females and 18.9% of males with breast cancer were
positive for at least 1 deleterious or suspected deleterious mutation, of which mutations in BRCA1 and BRCA2 comprised 40.2%.
In this cohort, 59.8% of the mutations were detected in other genes (see table below). Of the patients who did not meet either
testing criteria, mutations were found in BRCA2 (2), APC (2), BARD1 (1), CHEK2 (1), MSH2 (1), NBN (1), and TP53 (1).
Interestingly, we also identified 15 patients with two mutations in our cohort. BRCA1 or BRCA2 accounted for one of the
mutations in 9 out of 15 patients and 6 patients had two mutations in other genes.
Gene
% of Total Mutations
BRCA1
21.2%
BRCA2
19.0%
CHEK2
11.3%
ATM
9.7%
PALB2
7.2%
NBN
6.6%
APC
5.6%
BARD1
3.3%
PMS2
2.6%
BRIP1
2.3%
MSH6
2.0%
TP53
1.5%
MSH2
1.3%
RAD51C
1.3%
CDH1
1.0%
RAD51D
1.0%
CDKN2A
1.0%
PTEN
1.0%
MLH1
0.5%
Biallelic MUTYH
0.3%
SMAD4
0.3%
Conclusions: Testing patients using a 25-gene panel identified 234 mutations outside of BRCA 1 and BRCA 2 (157). That is a
149% increase in mutations identified over BRCA 1 and BRCA 2 testing alone. Mutations in moderately penetrant breast cancer
genes (including CHEK2, ATM, PALB2 and NBN) comprised 34.8% of total mutations and 6.4% of mutations were in LS genes.
Additionally, panel testing identified mutations in more than one gene in 4.0% of patients, which would not have been identified by
single-syndrome testing. Panel testing provides a broader understanding of hereditary cancer in breast cancer patients both by
identifying mutations in more genes and identifying patients with mutation in more than one gene. This approach can provide
more guidance both for management of the patient and the patient s family members.

Title: Triple-negative breast cancer: Frequency of inherited mutations in breast cancer susceptibility genes
Fergus J Couch1, Steven N Hart1, Priyanka Sharma2, Amanda Ewart Toland3, Penelope Miron4, Janet E Olson1, Andrew Godwin2,
Vernon S Pankratz1, Curtis Olswold1, Seth Slettedahl1, Lucia Guidugli1, Matthias W Beckmann5, Brigitte Rack6, Arif B Ekici7, Irene
Konstantopoulou8, Florentia Fostira8, George Fountzilas9, Liisa M Pelttari10, Song Yao11, Judy Garber4, Angela Cox12, Hiltrud
Brauch13, Christine Ambrosone11, Heli Nevanlinna10, Drakoulis Yannoukakos9, Susan L Slager1, Celine M Vachon1, Diana M
Eccles14 and Peter A Fasching5. 1Mayo Clinic, Rochester, MN; 2University of Kansas Medical Center, Kansas City, MO; 3Ohio
State University, Columbus, OH; 4Dana-Farber Cancer Institute, Boston, MA; 5University Hospital, Erlangen, Germany;
6
Ludwig-Maximilians-University, Munich, Germany; 7Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, Germany;
8
National Centre for Scientific Research "Demokritos", Athens, Greece; 9Aristotle University of Thessaloniki School of Medicine,
Thessaloniki, Greece; 10University of Helsinki, Helsinki, Finland; 11Roswell Park Cancer Institute, Buffalo, NY; 12University of
Sheffield, Sheffield, United Kingdom; 13Dr. Margarete Fischer-Bosch-Institute of Clinical Pharmacology, Stuttgart, Germany and
14
University of Southampton, Southampton, United Kingdom.
Body: Background: Guidelines recommend germline mutation testing of breast cancer predisposition genes in triple negative
(TN) breast cancer cases with a family history of breast or ovarian cancer or when diagnosed under age 60. However, the
prevalence of mutations in these genes among TN cases unselected for family history of breast or ovarian cancer is not known.
Methods: To assess the frequency of mutations in 16 predisposition genes in TN cases we screened a large cohort of TN patients
(n=1824) unselected for family history of breast or ovarian cancer from 12 centers and 824 study matched unaffected controls for
mutations using a panel-based sequencing approach.
Results: Deleterious mutations were identified in 15% of TN patients: 8.5% had BRCA1, 2.7% had BRCA2, and 3.6% had
mutations in 12 other genes. Mutations in non-BRCA1/2 genes encoding proteins implicated in homologous recombination repair
of DNA double strand breaks were detected at the same frequency as in breast cancer families. TN cases with mutations had
high-grade tumors and were diagnosed at an earlier age than non-mutated cases. However, 10% of TN cases diagnosed at 60
years and 5% with no family history of cancer were also found to carry mutations. Inactivating mutations in non-BRCA1/2
predisposition genes were associated with moderate to high risks of TN breast cancer.
Conclusions: National Comprehensive Cancer Network (NCCN) guidelines support clinical genetic testing of breast cancer
predisposition genes in 95% of TN breast cancer patients carrying mutations in susceptibility genes. In contrast, National Institute
of Health and Care Excellence (NICE) guidelines in the U.K. do not support genetic testing of a substantial proportion of TN
patients with predisposing alleles. Frequency tables for inherited mutations in known predisposition genes based on age of
diagnosis and family history of cancer will allow for selection of TN patients most likely to carry mutations in the predisposition
genes.

Title: Clinical evaluation of multigene testing for hereditary breast and ovarian cancer
Leif Ellisen1, Allison Kurian2, Stephen Lincoln3, Andrea Desmond1, Meredith Mills2, Kristen Shannon1, Michelle Gabree1, Michael
Anderson3, Yuya Kobayashi3, Federico Monzon3 and James Ford2. 1Massachusetts General Hospital, Boston, MA; 2Stanford
University, Stanford, CA and 3Invitae, San Francisco, CA.
Body: Background: Advances in DNA sequencing technology have fueled the development of multigene panels for hereditary
cancer testing. While such assays are potentially both practical and affordable for routine clinical genetic testing, there remains
uncertainty regarding their proper application and interpretation. Among the unanswered questions are which patients are
appropriate candidates for such expanded testing; what is the likelihood that finding deleterious variants will alter risk assessment
and management, and what is the prevalence of variants of unknown significance (VUS) that may create uncertainty for providers
and anxiety for patients.
Methods: 821 patients who met NCCN guidelines for BRCA1/2 testing for hereditary breast/ovarian cancer (HBOC) were
prospectively recruited at two major academic medical centers. These patients received traditional BRCA1/2 tests as part of their
clinical care and also were later tested for a panel of 29 known cancer risk genes. This panel also re-tested BRCA1/2. Both
sequence changes and deletions/duplications were reported, and the panel-testing laboratory was blind to the earlier results.
Panel testing results were validated by comparison with the earlier data or by independent testing using established technologies.
Family history information was collected directly by genetic counselors.
Results: 13.8% of the patients carried BRCA1 or BRCA2 mutations, with >99% concordance between the traditional and panel
results for these two genes. 53 (7.6%) of the BRCA1/2-negative patients carried mutations in other cancer risk genes. Some of
these findings confer a high risk for breast/ovarian cancer (e.g. TP53), while others (i.e. CHEK2, BRIP1, PALB2, RAD51C,
CDKN2A, ATM, and NBN) confer a moderate increase in risk. 10 of these 55 patients were positive for genes involved in Lynch
syndrome (MLH1, MSH2, MSH6, and PMS2), although breast and ovarian cancer are not uniformly accepted as part of this
syndrome. 40% of these patients were heterozygous carriers of pathogenic variants in MUTYH, which have a less clear impact on
breast/ovarian cancer, although this rate (3%) is higher than the expected carrier frequency in this population.
About 60% of patients had one or more VUS in the 29 genes. The rate of VUS was highly gene dependent, with ATM, APC, and
PTCH1 providing the largest number. A small but significant subset of patients did not have personal or family histories consistent
with the classical presentation of their identified mutation. Nevertheless, a majority of the non-BRCA1/2 findings would have
prompted consideration of a management change for the tested patient even considering the personal and family cancer history.
Conclusions: Multigene panel testing for hereditary cancer risk assessment increased the yield of findings with potential clinical
impact for almost 8% of patients. This is consistent with prior data from our laboratory and others. This study, carried out in a
uniform clinical practice setting and with data collected directly by health care providers, provides a representative view of the
benefit from such testing in an unselected patient population.

Title: Individuals with more than one pathogenic variant: Rationale for considering multi-gene panel testing for cancer
susceptibility
Rachel Nusbaum1, Lisa Susswein1, Kathleen Hruska1, Melanie Hussong1, Windy Berkofsky-Fessler1, Mingjuan Liao1, Erica
Rinella1, Nina Sanapareddy1, Joaquin Villar1, Haiyan Wan1, Zhixiong Xu1, Rebecca Y Bassett2, Elisabeth McKeen3, Constance
Murphy3, Deborah Pencarinha4, Jessica Booker1, Maria L Cremona1, Patricia Murphy1 and Rachel T Klein1. 1GeneDx,
Gaithersburg, MD; 2NYU Langone Medical Center, New York, NY; 3Jupiter Medical Center, Jupiter, FL and 4Wellmont Cancer
Institute, Kingsport, TN.
Body: Introduction: Expansion of genetic testing technologies has brought multi-gene panels for cancer susceptibility into the
clinic; however, the clinical utility of these next-generation sequencing (NGS) panels is largely unknown.
Hypothesis: We hypothesized that the use of multi-gene panels would yield a significant number of cases with more than one
pathogenic variant.
Methods: We retrospectively queried oncology tests reported at GeneDx from August 2013 to April 2014 for individuals with more
than one pathogenic variant. Next, we calculated the proportion of individuals with more than one pathogenic variant among all
positive reports, excluding familial and stand-alone BRCA1/2 tests. We then extracted personal and family histories, including
available segregation data, to categorize the phenotypes.
Results: Of 406 unique, unrelated individuals with pathogenic or likely pathogenic findings, 11 (2.7%) had more than one
pathogenic variant. This total includes nine individuals with a mutation in more than one gene, as well as two individuals with two
mutations in trans in the same gene. Ten of these 11 individuals were identified by multi-gene panel tests, one individual by
step-wise (tiered) testing. Seven of 11 individuals were positive for a pathogenic variant in a traditional, highly-penetrant cancer
susceptibility gene and another pathogenic variant in a gene with moderate cancer susceptibility, such as CHEK2 and ATM.
Three of 11 probands had more than one primary tumor. Several of the families were significant for bilineal cancer risk.
Conclusions: Our data suggest that a traditional single gene approach to cancer testing may fail to identify all pathogenic variants
related to the clinical presentation. The identification of a second risk factor for inherited susceptibility to cancer allows for
appropriate testing and management considerations for family members. In conclusion, we provide early evidence for the
consideration of multi-gene panel testing in a clinical oncology setting.

Title: Results of next-generation sequencing panels in a large community-based hereditary cancer risk program
J L Blum1, C A Garby1, A E Simmons1, S A Walker1 and L E Panos2. 1Baylor Charles A. Sammons Cancer Center, Baylor
University Medical Center, Dallas, TX and 2Ambrey Genetic Laboratories, Aliso Viejo, CA.
Body: Background: Next-generation sequencing (NGS) allows for broader germline genetic testing for hereditary cancers. Since
the Supreme Court decision of AMP v. Myriad on June 13, 2013, hereditary cancer multi-gene panels can now include BRCA1
and BRCA2, making these panels an option for first-tier testing. However, questions have been raised about the clinical utility and
implications of extended panels for medical management given the inclusion of unknown to moderate penetrant genes.
Methods: We reviewed all patients who underwent multi-gene panel testing from July 1, 2013 through May 23, 2014. The
indications for testing included personal and/or family history of breast or ovarian cancer. The panels were ordered in a single
genetic counseling clinic within a large community-based cancer center.
Results: A total of 136 patients underwent panel testing via a single commercial laboratory. We identified 12 (8.8%) patients who
were positive for a pathogenic or likely pathogenic mutation in a cancer susceptibility gene; 4 had prior negative BRCA1 and
BRCA2 sequencing and deletion/duplication testing. These positive results included 4 BRCA2 mutations, 2 TP53 mutations, 1
CDH1 mutation, 2 ATM mutations, and 1 patient each with a CHEK2, NBN, or PALB2 mutation. Of the patients found to have a
positive test result, 100% met the National Comprehensive Cancer Network (NCCN) guidelines for Hereditary Breast and Ovarian
Cancer (HBOC) genetic testing. The CDH1 mutation carrier did not meet NCCN guidelines for hereditary diffuse gastric cancer
testing and only one of the TP53 mutation carriers met NCCN guidelines for Li-Fraumeni syndrome. Within our cohort (136), 21
(15.4%) patients had a total of 25 variants of uncertain significance (VUS) and 103 (75.7%) patients had negative test results.
Conclusion: Testing through NGS panels identified 7/12 (58%) patients with a mutation which led to changes in current medical
management and 3/7 (43%) had a mutation in a gene other than BRCA1 or BRCA2. Our findings suggest that there is clinical
utility of NGS panels for use in this patient population despite the inclusion of unknown to moderate penetrant genes and a higher
rate of VUS than single gene testing.

Title: Technical evaluation of multigene testing for hereditary breast and ovarian cancer
Stephen Lincoln1, Allison Kurian2, Andrea Desmond3, Geoffrey Nilsen1, Kevin Jacobs1, Shan Yang1, Reece Hart1, Federico
Monzon1, Leif Ellisen3 and James Ford2. 1Invitae, San Francisco, CA; 2Stanford University, Stanford, CA and 3Massachusetts
Body: Introduction: Next Generation Sequencing, or NGS technology is gaining acceptance in diagnostic laboratories for
multigene panel testing. However questions remain about the sensitivity, specificity and clinical implications of this new
technology and the expanded testing it enables. Moreover, questions are raised as to whether new laboratories using these
methods, typically without the benefit of large historical proprietary databases, can provide similar assessments of pathogenicity
as do the previously established providers. Expanding on our recently published work (Kurian et al., JCO 2014) we considered
whether NGS testing in an independent laboratory can replace traditional BRCA1/2 tests in patients indicated for hereditary
breast/ovarian cancer testing.
Methods: We recruited over 800 patients who were indicated for BRCA1/BRCA2 testing under clinical management guidelines,
and collected over 200 additional samples to increase the power of this ongoing study. All were tested using an NGS-based
multigene panel which reported both DNA sequence and copy-number alterations for BRCA1/2 and 27 other known cancer risk
genes. Traditional genetic testing results using established technologies were also available for comparison. In this report we
focus on the results for BRCA1/2 and 27 other known cancer risk genes.
Results: Sensitivity was high: 261 alterations (196 pathogenic and 65 others) were reported in the traditional genetic data, and all
were detected by NGS when the corresponding test was available. In this set are 141 alterations considered technically
challenging for NGS: insertions, deletions, and complex sequence changes, as well as very large (chromosome) and small (single
exon) copy number changes. Specificity was also high: all NGS variants for which we sought confirmation using independent
methods (n>2000) were confirmed, including 51 alterations not previously reported. Determination of pathogenicity was also
highly concordant: in all but 2 cases positive reports agreed, and in these 2 cases data were available in the literature to support
pathogenicity, although not at a level which meets recent guidelines from the American College of Medical Genetics (ACMG). It is
not clear from the diagnostic reports exactly what evidence supported pathogenicity in the traditional data for these 2 cases.
Rates of Variants of Unknown Significance (VUS) in BRCA1/2 were somewhat different: about 7% of cases in the NGS data vs.
about 4% of fully tested cases in the traditional data had an uncertain report. The root of this difference is also unclear as details
are not provided in the traditional reports.
Conclusions: NGS can be a viable replacement for traditional genetic testing for hereditary breast and ovarian cancer.
Interpretation concordance is high but fully evaluating the details of this is hampered by the limited reporting of proprietary data by
some established laboratories. Recent efforts to establish large public databases of genetic information (particularly ClinVar) will
promote greater transparency and accountability and thus can help improve access to high quality care for hereditary conditions.
Note: All of the variants in this study and their interpretations will be released to public databases by the time of the meeting.

Title: Utilizing next generation sequencing technologies for hereditary breast cancer risk assessments in a private oncology
practice
Barbara Hamlington1, Scot Sedlacek1, Lucy Langer2, Brittney Goetsch1, Katie Lemas1 and Sami Diab1. 1Rocky Mountain Cancer
Centers, Greenwood Village, CO and 2Compass Oncology, Portland, OR.
Body: Background: The recent introduction of clinical Next Generation Sequencing (NGS) technologies has transformed the
ability of health care providers to provide personalized hereditary breast cancer risk assessments to patients and their family
members. However, along with the plethora of data derived from NGS technologies come challenges regarding how to interpret
this data in a clinically meaning way to guide medical management decisions.
Methods: A retrospective analysis was performed on a population of 242 patients with a personal or family history of breast
and/or ovarian cancer and completed NGS genetic testing between March 2012 and May 2014 at Ambry Genetics. Data points
included age, gender, primary diagnosis, type of NGS genetic test and result. Patients with prior genetic testing were not
excluded from the analysis.
Results: Of the 242 patients who completed NGS genetic testing 199 (82%) had a personal history of cancer and 43 (18%) had a
family history of cancer. Ten (5%) clinically significant deleterious mutations were found in the affected population; 3 MUTYH
mutations were consider not clinically significant and excluded from the analysis. Four (40%) were CHEK2 mutations, 2(20%)
BRCA1 mutations, 1(10%) NBN mutation, 1(10%) ATM mutation, 1(10%) MSH6 mutation and 1(10%) PTEN mutation. One
patient with a history of ovarian cancer had two deleterious mutations in the BRCA1 and NBN genes, and three affected patients
had a deleterious mutation in addition to a variant of uncertain significance (VUS); a patient with synchronous breast and ovarian
cancers had a MSH6 mutation and a BRCA2 VUS, a patient with breast cancer had a CHEK2 mutation and a CHEK2 VUS, and a
patient with breast cancer had a CHEK2 mutation and a MSH6 VUS. A total of 37 (19%) VUS were found in the affected
population, including one patient with a history of breast cancer and a VUS in both ATM and MRE11A. A single (2%) clinically
significant BRCA1 deleterious mutation and 6(14%) VUS were detected in the unaffected population.
Conclusions: NGS technologies allow for health care providers to complete more comprehensive hereditary breast cancer risk
assessments for patients and their families. However, the high VUS rate (14-19%) and the possibility for multiple mutations and/or
variants in a single patient create unique challenges when interpreting NGS genetic test results. Furthermore, the paucity of
prospective studies and consensus guidelines limits the ability of the health care provider to interpret NGS genetic test results in a
clinically meaningful way to guide patient management. This is highlighted by the case example above regarding the patient with
synchronous breast and ovarian cancers who was found to have a MSH6 mutation and a BRCA2 VUS. Given that the patient
does not meet Amsterdam criteria, the MSH6 mutations would have been missed had it not been for the NGS technology utilized
for this patient. However, challenges still exist about how to interpret this data for both the patient and her family in a clinically
meaningful way given that the family does not look like a classic Lynch Syndrome family.

Title: Low-level constitutional mosaicism of a de novo BRCA1 gene mutation
Eitan Friedman1, Noa Efrat2, Lior Soussan-Gutman3, Addie Dvir3, Yulia Kaplan3, Tali Ekstein4, Keith Nykamp4, Martin Powers4,
Marina Rabideau4 and Scott Topper4. 1Oncogenetics Unit, Chaim Sheba Medical Center, Tel-Hashomer, Israel; 2Kaplan Medical
Center, Rehovot, Israel; 3Oncotest-Teva, Teva Pharmeceuticals Industries, Petach Tikva, Israel and 4Invitae, San Francisco, CA.
Body: Background:
Germline mutations in the BRCA1 and BRCA2 genes detected in some high-risk breast/ovarian families are used to estimate
cancer risk, plan cancer early detection schemes, and make decisions about risk reducing surgeries. The gold standard for
detecting BRCA sequence changes has long been Sanger sequencing, but recently next-generation sequencing (NGS)
technologies have emerged as an accurate and efficient alternative, with improved sensitivity for detection of mosaic events. Here
we report the detection of low-level constitutional germline mosaicism (5%) for a de novo pathogenic BRCA1 mutation detected
using deep sequencing of three different non-cancerous tissues; and a corresponding high-level detection (50%) in cancerous
breast tissue. This is the first reported case of multiple tissue constitutional mosaicism in BRCA1 at this level of detection.
Patient and Methods:
The patient is a woman of mixed Jewish Ashkenazi - Bulgarian heritage, diagnosed with a large (8x10 cm), triple negative,
high-grade invasive breast cancer in the right breast at age 43 years. Her cancer family history includes a daughter with acute
lymphatic leukemia at age 18 months, a brother with a Central Nervous System (CNS) tumor at age 45 years, a father with a
malignant CNS tumor at age 58 years, a maternal grandfather with a malignant tumor in his 70's, and two of this maternal
grandfather's sisters who were diagnosed with breast cancer. The patient underwent neoadjuvant chemotherapy followed by
bilateral mastectomy (therapeutic mastectomy for the right breast and a contralateral risk-reducing mastectomy). The patient
underwent germline testing using NGS, sequencing of 29 hereditary cancer genes in DNA extracted from blood using Invitaes
Hereditary Cancer Panel. This testing was followed up with additional NGS testing of buccal and contralateral healthy breast
tissue. The patient also underwent somatic mutation testing on her breast tumor.
Results:
NGS of DNA extracted from blood identified a pathogenic BRCA1 mutation, c.1953dupG (p.Lys652GlufX21), in 5% of reads
(X450 coverage). No other pathogenic mutations were detected in other genotyped cancer susceptibility genes. Analysis of
buccal tissue and normal breast tissue removed at initial surgery also identified this mutation in 5% of reads, and confirm that
this individual is a constitutional mosaic for this mutation. Genetic analyses were subsequently performed on the breast cancer
tissue using Genome Health platform (Foundation One) and the same mutation, c.1943dupG was detected in 47% of molecules.
Sanger sequencing of BRCA1 and BRCA2 in DNA extracted from peripheral blood had not detected this mutation. Analysis of the
maternal DNA did not reveal this mutation, and analysis of the father was not possible.
Conclusion:
This is the first reported case of a de novo constitutional mosaicism in BRCA1 at this level of detection and confirmed across
multiple tissue types, and highlights the need to perform deep sequencing in individuals clinically suspected of having cancer
predisposition, prior to considering risk-reducing surgery.

Title: Uptake and outcomes of multiplex testing for breast cancer susceptibility
Angela R Bradbury1, Linda Patrick-Miller2, Brian L Egleston3, Amanda Brandt1, Jessica Long1, Jacquelyn Powers1, Jill Stopfer1,
Laura DiGiovanni1, Jamie Brower1 and Susan M Domchek1. 1University of Pennsylvania, Philadelphia, PA; 2University of Chicago,
Chicago, IL and 3Fox Chase Cancer Center, Philadelphia, PA.
Body: Background: New counseling models for multiplex genetic testing for breast cancer susceptibility are needed. Further, the
risks, benefits and utilities of multiplex genetic panels are unknown.
Purpose: To obtain stakeholder feedback on an innovative tiered-binned model for pretest counseling and informed consent for
multiplex testing and to evaluate the uptake of, cognitive and affective responses to and perceived utility of panel testing.
Methods: Patients previously BRCA1/2- or BRCA1/2 untested completed in-person pre-test (V1) and post-test counseling (V2)
and surveys regarding the novel counseling model and evaluating cognitive and affective responses to, and perceived utility of
the 26 gene Myriad MyRisk panel for cancer susceptibility.
Results: 49 patients (62% of eligible) enrolled and completed V1. 38% of decliners were not interested in panel testing.
BRCA1/2- were more likely to proceed with MyRisk (89%) than BRCA1/2 untested (48%, p<0.01). Although not statistically
significant, those who declined panel testing after V1 had higher anxiety, depression and cancer worry, but no difference in
knowledge. They also had lower perceived utility (p=0.005). Most patients would not change anything about their V1 (86%) or V2
(91%) counseling. Surveys suggest that patients value the engagement, personalization, organization and visual aids of the novel
counseling model. Potential refinements include enhancing assessments of informational overload, confusion and psychosocial
needs, particularly with uncertain results. As shown in Table 1, event anxiety, depression, uncertainty and cancer worry did not
change, while general anxiety decreased. Knowledge increased and perceived utility and satisfaction decreased significantly.
Exploratory analyses by results to date (positive = 5; VUS = 9; negative =22), suggest no difference in uncertainty by test result.
Patients with a positive result might experience greater event anxiety and have less decline in perceived utility (p=0.02) than
those with a negative or VUS result.
Table 1
General Anxiety
General Depression
Event Anxiety
Cancer Worry
Knowledge (K) Total
K-Inheritance
K-Benefits
K-Limitations
Baseline
Post V1
Post V2
Mean (SD)
Mean (SD)
Mean (SD)
6.8 (3.9)*
6.1 (4.0)*
7.2 (3.5)*
6.1 (3.6)*
2.6 (3.0)
2.3 (2.6)
2.6 (2.9)
2.3 (2.6)
37.1 (9.6)
37.7 (9.5)
37.0 (8.9)
37.3 (8.5)
18.3 (15.7)
16.9 (14.1)
18.4 (15.4)
15.7 (14.1)
61.8 (6.1)**
63.9 (6.4)**
62.1 (6.7)**
64.1 (6.8)**
29.5 (3.2)
30.0 (3.2)
29.7 (3.5)
29.8 (3.3)
12.0 (1.4)
12.3 (1.8)
12.0 (1.4)
12.4 (1.7)
20.3 (3.2)**
21.6 (2.8)**
20.4 (3.4)**
21.9 (3.0)**
5.8 (4.2)*
2.9 (3.6)
37.3 (9.4)
6.6 (14.7)
66.3 (6.9)**
30.3 (3.7)
12.4 (1.9)
23.6 (2.6)**
Satisfaction
42.8 (3.8)
42.9 (3.6)*
Uncertainty
Perceived Utility
7.5 (4.3)
6.9 (4.6)
7.7 (4.0)
6.5 (4.5)
41.4 (2.6)*
6.7 (4.6)
37.2 (7.9)
37.7 (7.0)*
33.8 (8.6)*
*p,0.05, **p<0.001
Conclusion: With a tiered-binned counseling model, patients experience increased knowledge. Uptake of panel testing varies by
prior testing and potentially by patient affective factors. Most patients do not experience negative psychological responses,
although this may vary by test result. Declines in satisfaction and perceived utility may also vary by test result and may reflect the
current unclear utility and uncertainty of multiplex testing.

Title: Patient perceptions of the impact of genetic testing for breast cancer risk on health insurance
Erin Hofstatter1 and Anees Chagpar1. 1Yale School of Medicine, New Haven, CT.
Body: Introduction: Genetic testing for breast cancer risk is increasingly available, and patients perceptions regarding how
results may affect their health insurance may affect their likelihood to opt for testing. We sought to determine the changes in
patient perceptions regarding the impact of genetic testing on health insurance over a five year period from 2005 to 2010.
Notably, during this time period two significant U.S. federal health care laws were enacted. First, the Genetic Information
Nondiscrimination Act (GINA) passed in 2008, protecting against the use of genetic information for health insurance coverage.
Second, the Patient Protection and Affordable Care Act (PPACA) passed in 2010, representing a major regulatory overhaul of the
U.S. health care system, promising universal access to health care coverage.
Methods: The National Health Interview Survey is a population-based face-to-face survey conducted annually by the CDC. In
2005 and in 2010, a cancer supplement was fielded which asked respondents who had undergone genetic testing whether they
felt this had, or would, affect their health insurance. We evaluated the cohort of participants who had undergone genetic testing
for breast cancer, and used SUDAAN statistical software to evaluate changes in the proportion who felt genetic testing would
adversely affect their health insurance, and factors associated with this perception.
Results: In 2005, 48 respondents had undergone genetic testing for breast cancer, representing 333,544 people in the
population. Of these, 16.8% felt that genetic testing had, or would, affect their health insurance. In 2010, 16 respondents had
undergone genetic testing for breast cancer, representing 93,301 people in the population. Of these, 17.4% felt that genetic
testing had, or would affect their health insurance. In 2005, this perception varied based on insurance status, education, and
region; but in 2010, perceptions of the impact of genetic testing on health insurance were unaffected by these factors.
Patient Perception of Genetic Testing Effect on Health Insurance
2005
%
Insurance
2010
p-value
0.006
0.276
Uninsured
11.8
22.8
Medicaid
59.5
Medicare
28.7
7.0
Private
15.7
18.8
Other
35.4
Education
0.031
0.463
< Grade 12
8.7
7.9
High School/GED
17.2
Some college
19.0
Associates
34.5
19.5
Bachelors
34.2
5.6
Masters/Prof/Doctorate
38.3
42.5
Region
p-value
0.077
0.122
Northeast
35.3
30.3
Midwest
17.0
6.1
South
3.2
29.1
West
5.8
10.7
Age
0.449
0.463
<55
26.2
14.0
>55
12.2
21.1
Conclusion: More people undergoing genetic testing for breast cancer risk in 2010 were concerned that this would adversely
affect their health insurance than in 2005, and there was less variation in this perception based on insurance, education and
region than in the earlier period. We conclude that, despite the passage of significant U.S. legislation including GINA and PPACA,
nearly 1 in 5 people undergoing genetic testing remained concerned about potential risk to their health insurance coverage.
Accurate and early identification of those people carrying a cancer genetic mutation is crucial to cancer prevention and treatment
efforts; our results suggest a dire need for improved education regarding federal protections for genetic testing so that any
barriers to seeking genetic counseling are eliminated.

Title: Next generation sequencing-based analysis of BRCA1 and BRCA2 genes: Applicability for fast diagnostics of large
samples
Sun-Young Kong1, Eunhae Cho2, Junnam Lee2, Myong Cheol Lim1, Jahyun Jang2, Boyoung Park1, Kyong-Ah Yoon1, Young-Ho
Kim1 and Eun Sook Lee1. 1Research Institute, Hospital, & National Cancer Control Institute, National Cancer Center, Goyang,
Korea and 2Green Cross Genome, Yongin, Korea.
Body: Purpose
BRCA1&2 gene mutation test for hereditary breast ovarian syndrome (HBOC) has been conducted mostly by Sanger sequencing.
Currently, the next generation sequencing (NGS) is rapidly incorporated to the fields of cancer research and clinical diagnostics.
Here we evaluated NGS-based results of BRCA gene analysis and compared with the results of Sanger sequencing for future
diagnostic applications.
Methods
The patients (n=100) who have genetic counseling and decided to perform BRCA test were included. All coding regions of
BRCA1 and BRCA2 analyzed by both of Sanger sequencing and NGS using Access Array BRCA1, BRCA2 and TP53 kit
(Fluidigm, USA). Access array kit was designed to multiplex 48 samples simultaneously for 184 amplicons and the average
sequencing depth per base was 6,500X using MiSeq sequencer (Illumina, USA). We developed analysis pipelines to avoid false
negative results and detect all pathogenic mutations. The reads were aligned to a reference genome (NCBI human genome
assembly build 37) using the BWA-MEM, then all candidate variants called with minimum filtering parameter.
Results
Total of 765 variants were detected by NGS and among them 616 variants (frameshift:22, nonsense:8, splicing:3, missense:290,
and synonymous variants:293) were identical with the results from Sanger sequencing. When we evaluated the results of Sanger
sequencing as standard methods, the mean allele frequency showed difference as 41.7% and 12.0% for true positive
heterozygous variants (616) and false positive variants (149), respectively.
Conclusions
There was no false negative of pathogenic mutations from NGS. The BRCA mutation detection using NGS represented potential
applicability in clinical diagnosis.

Title: Experience of pre-operative genetic testing on surgical decision making in newly diagnosed breast cancer patients
Siddhartha Yadav1, Otavio Pereira-Rodrigues1, Lindsay Dohany2, Heidi Dreyfuss2, Jennifer Fulbright2, Ashley Reeves2 and Dana
Zakalik2. 1Beaumont Health System, Royal Oak, MI and 2James and Nancy Grosfeld Cancer Genetics Center, Beaumont Health
System, Royal Oak, MI.
Body: Introduction:
BRCA 1 and 2 testing has been widely incorporated into clinical care for women at risk for hereditary breast cancer. Knowledge of
BRCA mutation status prior to surgery may influence decision regarding type of surgery. This study analyzes the experience with
pre-operative BRCA mutation testing in patients with suspected hereditary predisposition to breast cancer.
Methodology:
Records of 150 patients referred to Cancer Genetics between November 01, 2013 and April 30, 2014 for pre-operative genetic
testing were analyzed. This cohort consisted of patients with newly diagnosed breast cancer who met current genetic testing
criteria. Patients were excluded if their surgical records were not available or they had not yet completed surgery. A total of 80
patients who completed genetic evaluation and definitive surgery were evaluated. Data on demographic characteristics, tumor
pathology, BRCA mutation status and surgical management was collected on all evaluable patients.
Results:
Records of 80 patients who underwent pre-operative genetic testing for hereditary breast cancer risk were evaluated. The median
age at diagnosis was 51.5 years. Median time from biopsy to initiation of genetics referral was 10 days. From that point, the
median time to cancer genetics appointment was 3 days. Median time from initial genetics visit to definitive surgery was 24 days.
7 (9%) patients underwent surgery within 10 days of the genetics appointment, and prior to receiving the results of the genetic
tests.
Of the 80 patients, 5 (6%) tested positive for a BRCA mutation, 3(4%) had a BRCA variant of unknown significance(VUS), and the
rest tested negative. 4 of the 5 BRCA mutation carriers underwent bilateral mastectomy, as did 2 of the 3 patients with a BRCA
VUS. Of the 72 BRCA -ve patients, 22 (30%) underwent bilateral mastectomies, 42 (58%) underwent partial mastectomy, 7 (10%)
patients underwent unilateral mastectomy, and 1 underwent bilateral partial mastectomy.
There was no significant difference on univariate analysis in the age, histopathology (grade, receptor status, lymph node status
and margins) in the patients who underwent bilateral mastectomy and those who underwent breast conservation. Mean tumor
size was 24.5 mm in the bilateral mastectomy group, compared to 15.8 mm in the unilateral mastectomy group(p=NS). The
bilateral mastectomy group had a greater number of close relatives with breast cancer reflecting a more significant family history.
Conclusions:
This study demonstrates the feasibility and successful implementation of preoperative genetic testing in newly diagnosed breast
cancer patients. The majority of mutation carriers underwent bilateral mastectomies as did a significant proportion of BRCA -ve
patients. Larger tumor size and a more significant family history of breast cancer appeared to be associated with the decision to
pursue bilateral mastectomy. Further studies are needed to better characterize the impact of preoperative genetic testing in newly
diagnosed breast cancer patients.

Title: Distress as a measure of the psychological impact after disclosure of a BRCA1/2 positive test result
Joana Parreira1, Susana Esteves1, Fatima Vaz1, Carla Simes1, Paula Rodrigues1, Ana Luis1, Ana Clara1, Sandra Bento1 and
Maria Jesus Moura1. 1Instituto Portugues Oncologia Lisboa Francisco Gentil, EPE, Lisboa, Portugal.
Body: Introduction and objectives- A positive result after BRCA1/2 screening can represent a difficult psychological experience.
Previous studies have shown that the psychological outcomes following BRCA1/2 testing vary according to the previous individual
and family experiences of each person. Objectives of this study were to measure the distress caused by the disclosure of a
positive BRCA1/2 test result and to analyse the degree of BRCA1/2 carriers retention of information, transmitted at the post-test
counselling interview.
Material and Methods-This is a prospective study. All consecutive individuals with a BRCA1/2 positive test were invited to
participate, after the post-test counselling visit. Participation involved signing an informed consent form and agreeing to a
structured post-test phone interview, one week and one month after disclosure of the test result. Phone interviews were done by
nurses trained by the Psychological Unit of our centre. Measure instruments: 1) Emotional thermometer (ET) - analogical scale
ranging from 0 (no distress) to 10 (maximum distress), measures distress during the previous week 2) Distress questionnaire
(DQ)- 13 items to evaluate depression, anxiety and loss of emotional control, with a global score ranging from 3 (no distress) and
45 (maximum distress). 3) Knowledge of disease status and understanding of the individualized risk management plan- additional
4 items included at the end of DQ. Subgroup analysis was performed according to age, sex, previous cancer diagnosis and
offspring existence using Wilcoxon rank sum test with continuity correction.
Results-From 177 eligible carriers, 28 were not included (14 for logistical reasons; 2 deaths; 1 refused; 3 progressive
symptomatic disease ). A total of 149 carriers were included: 120 women (81%) and 29 men (19%); median age 43 yrs (21-74);
67 (45%) with a previous cancer diagnosis and 82 (55%) healthy at risk; 42 (28%) had no offspring and 102 (68%) were
professionally active. The mean distress scores were 3.07 (SD 2.72) and 20.13 (SD 7.88) for ET and DQ instruments,
respectively. Using the NCCN (2013) guidelines for ET classification, we found that 95 (64%) of our carriers did not have clinically
significant distress. For the DQ (using a cut-off <20) this proportion decreased to 54% (81 pts).
Subgroup analysis: A statistically significant difference was found with both ET and DQ for higher distress levels in women than
men (p=0.006 and p<0.001, respectively). There was no difference in either measure for age ( 50yrs vs > than 50yrs), previous
cancer diagnosis or with vs no offspring. Levels of knowledge and understanding of individual risk management were high
(average 18.7; maximum 20) and no correlation was found with distress levels. Twenty-eight (19%) carriers were found in need of
specialized psychological/psychiatric support and were appropriately referred.
Conclusions-In our BRCA1/2 carrier population clinically significant distress was not frequent and only 19% needed specialized
psychological/psychiatric support. Distress was higher in women than in men. Retention of information given during counselling
was high, and there was no correlation between information retention and distress levels.

Title: Are we appropriately referring and testing breast cancer women for genetic mutations?
Pankhoori Saraf1, Dipen Patel1 and Alice J Cohen1. 1Newark Beth Israel Medical Center, Newark, NJ.
Body: Background Annually it is estimated that Hereditary Breast and Ovarian Syndrome (HBOC) accounts for 5 - 6% of breast
cancer in the US. The majority of HBOC are associated with gene mutations in BRCA 1 & 2; other mutations include p53, PTEN,
CH1, STK11. These genes are tumor suppressor genes and play a role in the maintainence of genomic integrity. Women with
BRCA mutations have a life time risk of developing breast cancer (50 -85%) and ovarian cancer (15-40%). Other cancers have
been associated with these mutations as well. The estimated frequency of a mutation in the BRCA gene is 1/800-1/1000. As a
primary oncologist it is our responsibility to screen breast cancer patients who may harbour such deleterious mutation and offer
appropriate screening, counselling, testing and management. As tests are now available without trained genetic counsellors, the
question remains are all appropriate patients being offered genetic testing to assist in treatment and does onsite genetic
counselling improve testing rates.
Purpose of study To evaluate the difference in identification of high risk breast cancer patients and the number of women who
undergo genetic testing with or without an onsite genetic counsellor.
Methods A retrospective chart review was performed of all newly diagnosed breast cancer patients from March 2012 February
2014. Year 1 was without a genetic counsellor, March 2012 - February 2013 (Gr.1) and year 2 with an onsite genetic counsellor,
March 2013 February 2014 (Gr.2). Information collected included age, stage of breast cancer, receptor type, reason for referral
or genetic testing based on NCCN criteria for HBOC testing and results of tests.
Results 135 new breast cancer patients were identified and 125 were evaluable. In Gr 1, 27/72 (37.5%) met criteria for genetic
testing. 15 were offered testing (55.5%). 5/15 (33.3%) completed testing with no patient positive for BRCA mutation; 10 patients
were non compliant with recommended testing; 3 patients were tested outside of guidelines with negative results.
In Gr 2, 22/53 (41.5%) met criteria for genetic testing. 16 were offered testing (72.3%); 8/16 (50%) completed testing with 4
positive for a BRCA mutation; 8 patients were not tested (6 patients non compliant with recommended testing, 1 due to insurance
issuses and 1 refused).
See Table 1.
Table 1
Groups
Total number of patients
Genetic test criteria met
Offered testing
Actual testing
Gr 1
72
27 (37.5%)
15 (55.5%)
5 (33.3%)
Gr 2
53
22 (41.5%)
16 (72.3%)
8 (50%)
The most common reason for genetic testing was patients with primary breast cancer age < 45 years. Patients who met criteria
but were not offered testing were those with age < 45 years and triple negative breast cancer. 5 patients in Gr 1 were seen by the
genetic counsellor post treatment and all 5 were tested.
Conclusion A greater number of women with newly diagnosed breast cancer were identified and offered genetic testing with an
onsite genetic counsellor. The precentage of individuals tested increased with an onsite counsellor.

Title: Teasing out the PALB2 phenotype
Emily K Dalton1, Rachel McFarland1, Holly Laduca1, Shuwei Li1 and Chia-Ling Gau1. 1Ambry Genetics, Aliso Viejo, CA.
Body: Background: Biallelic mutations in PALB2 (Partner and Localizer of BRCA2) are known to cause Fanconi Anemia Type N.
Multiple reports have demonstrated an increased risk for cancer in individuals heterozygous for PALB2 mutations. For example, a
recent study by Antoniou et al reported a 33-58% lifetime risk for breast cancer in PALB2 mutation carriers, with 30% of carriers
reporting triple negative breast cancer (TNBC). Other studies have suggested associations between PALB2 heterozygosity and
pancreatic cancer, ovarian cancer, male breast cancer, and prostate cancer as well. We aimed to better define PALB2
phenotypes by assessing clinical history of TNBC, pancreatic, ovarian, and prostate cancers amongst PALB2 mutation carriers
identified via multigene panel testing.
Methods: We reviewed clinical histories of 11,007 individuals who underwent PALB2 sequence and deletion/duplication analysis
as part of a multigene hereditary cancer panel. Descriptive statistics were utilized for clinical histories of PALB2 carriers, and chi
square analysis was used to compare clinical histories of PALB2 mutation carriers to mutation-negative controls. Individuals with
mutations in other cancer susceptibility genes were excluded from analysis.
Results: A total of 98 PALB2 mutation carriers identified among 9610 individuals were included in our analysis. The majority of
mutation carriers were Caucasian (80%) and female (92.8%). All identified mutations were truncating (nonsense, frameshift, or
gross deletions). No pathogenic missense mutations were identified in this cohort. 77.6% (n=76) of mutation carriers had breast
cancer, diagnosed at a mean age of 48. Hormone receptor status was available for 48 mutation carriers and 2469 controls. 37.5%
(18/48) of breast cancers in mutation carriers were reported as triple negative, compared to 17.1% (423/2469) of breast cancers
in controls (OR: 2.9 ; p= 0.0002). 7.8% (n= 8) of PALB2 mutation carriers had ovarian cancer. There was no significant difference
in the incidence of ovarian cancer between PALB2 mutation carriers and controls (OR: 0.65 ; p= 0.25). Additionally, mutation
carriers were significantly less likely to have a family history of ovarian cancer than controls (OR: 0.5; p= 0.02). 5.9% (n=6) of
mutation carriers had pancreatic cancer, diagnosed at a mean age of 57.8, compared to 61 for controls. PALB2 mutation carriers
were 1.3 times more likely to have personal and/or family history of pancreatic cancer, although this was not statistically
significant (p= 0.22). Similarly, PALB2 mutation carriers were 1.5 times more likely to have a family history of prostate cancer,
although this was not statistically significant (p= 0.09).
Conclusions: Our data supports existing literature associating PALB2 mutations with TNBC. We did not observe significant
associations between PALB2 carrier status and a clinical history of pancreatic, prostate, or ovarian cancers. However, this data
should be interpreted with caution, as it is possible that unidentified genetic factors contributed to the clinical history of cancer in
our mutation-negative controls. Investigation of PALB2-associated cancer risks in an unselected prospective cohort would help to
further elucidate the PALB2 phenotype.

Title: Trancriptome sequencing of the hisologicaly normal breast epithelium of BRCA mutation carriers
Dadrie Baptiste1, MiRan Choi2, Zhiping Wang1, Yunlong Liu1, Milan Radovich1 and Susan E Clare2. 1Indiana University School of
Medicine, Indianapolis, IN and 2Feinberg School of Medicine, Northwestern University, Chicago, IL.
Body: Background: Meaningful progress in the prevention of breast cancer is unlikely until the risk of the development of breast
cancer is translated into specific, quantifiable molecular alterations. Among those women at the greatest risk are BRCA mutations
carriers: 55 to 65% of women who inherit a deleterious BRCA1 mutation and approximately 45 % who inherit a deleterious
BRCA2 mutation will develop breast cancer by age 70 years. The purpose of this study was to identify the earliest transcriptional
alterations present by examining the histologically normal breast epithelia of BRCA mutation carriers.
Methods: Epithelia were microdissected and the RNA isolated from the histologically normal breast of 16 frozen tissue cores
from known BRCA mutation carriers, who were donors to the Susan G. Komen for the Cure Tissue Bank at the IU Simon Cancer
Center. RNA-sequencing was carried out using the Life Technologies SOLiD XL 5500 Platform. RPKM gene expression values
from the BRCA specimens and from Komen normal breast epithelium (controls; Pardo et al, Breast Cancer Research, 2014) were
merged, quantile normalized, and batch effect corrected. Normalization and differential gene expression was performed using
EdgeR v2.11
Results: 6583 RNAs were differentially expressed with the false discovery rate set at 0.1.
1. DNA damage is signaled by a tripartite system that includes kinases, phosphatases and proteins with modular domains, e.g.,
the BRCA 1 C-terminal (BRCT), that recognize phosphorylated linear motifs in other proteins. Several BRCT domains recognize
motifs phosphorylated by kinases that are activated by DNA damage. To identify the proteins active in DNA damage signaling,
Monteiro and colleagues generated a protein-protein interaction map for seven proteins with tandem BRCT (Cell Signaling, 2012).
A significant proportion of the genes with decreased expression in our data set encode proteins that have been identified in this
network.
2. Two of the top biological processes affected by the BRCA mutations in the histologically normal breast tissue are translation,
and cellular protein metabolism.
3. There was significantly less expression of genes associated with other hereditary cancer syndromes inducing the genes
responsible for Lynch Syndrome: MHL1, MHS2, MSH6 and PMS2; and the multiple endocrine neoplasia genes RET and MEN1.
4. High grade serous carcinoma of the ovary, the histologic subtype associated with BRCA mutation, may originate in the distal
Fallopian tube. The list of differentially expressed genes from microdissected breast epithelia was compared with that derived
from gene array profiling of microdissected Fallopian tube epithelium from histologically normal BRCA1 mutation carriers and
controls. 795 genes were common to both gene sets, of which 354 were regulated similarly, i.e., increased or decreased.
MetaCore data analysis revealed CREB1 and c-Myc as the most important regulatory factors in the similar dataset.
Conclusions: There are significant transcription alterations in the histologically normal breast tissue of BRCA mutation carriers.
These data will have to be corroborated at the protein level and functional level. Once substantiated, they have potential to stratify
risk and to serve as targets for prevention.

Title: BRCA1/2 founder mutations among high-risk HBOC patients from Northeast of Brazil
Ivana L Nascimento1,2, Gabriela ES Felix1, Camila A Sandes1, Taisa B Machado-Lopes1, Thais F Bomfim1, Roberto Meyer1,
Maura Romeo1, Betania Toralles1 and Kiyoko A Sandes1. 1Instituto de Ciencias da Saude-UFBA, Salvador, Bahia, Brazil and
2
Nucleo de Oncologia da Bahia, Salvador, Bahia, Brazil.
Body: Background: Genetic susceptibility to Hereditary Breast and Ovarian Cancer (HBOC) syndrome could be test by the
screening of BRCA1 and BRCA2 genes. Thus, knowing that each population has its degree of heterogeneity and the Brazilian
population is one of the most heterogeneous of the world, we aimed to verify the frequency of BRCA1/2 founder mutations in
high-risk HBOC patients from Bahia, the biggest State of the Northeast of Brazil.
Methods: It was analyzed the DNA of 99 unrelated high-risk HBOC patients, considering the criteria of NCCN v.1.2010. These
patients were tested for the next founder mutations: BRCA1 c.211A>G (Galician), BRCA1 943ins10 (African), BRCA1 3450del4
(Hispanic) and BRCA2 c.156_157insAlu (Portuguese) by sequencing or PCR/RFLP. All positive results were confirmed by two
sequencing reactions. The genetic ancestry was analyzed with a panel of 9 ancestry informative markers (AIMs). The clinical and
epidemiological data were collected during the genetic counseling and were analyzed in Epi Info 7, while the frequency of the
AIMs was analyzed in GENEPOP, and in ADMIX95 and STRUCTURE the genetic admixture was estimated.
Results: All patients had personal and/or familial history of breast and/or ovarian cancer. Most of the patients self-reported as
mulatto (62.63%), followed by black (21.21%), white (10.10%) and others (6.06%). The African, Amerindian and European
ancestry contribution estimated were 35.33%, 11.31% and 53.35% respectively (r2 = 0.998). Three mutations were detected:
BRCA1 3450del4 in five patients (5.05%), BRCA1 c.211A>G in two patients (2.02%), and BRCA1 943ins10 in one patient
(1.01%). All patients with these mutations were from State of Bahia and self-reported as mulatto or white.
Conclusion: Although the Portuguese settled Brazil, here it was observed that this population seems to have high Spanish
ancestry contribution, due to the high frequency of the BRCA1 3450del4 and c.211A>G (7.07%). It is also interestingly the
frequency of BRCA1 943ins10 (1.01%), because the State of Bahia has the higher rate of African-descendants in Brazil and most
of the patients exhibited these physical traits (mulatto and black, 83.84%). Thus, high-risk HBOC patients from that region of
Brazil could be screening first for the BRCA1 3450del4, BRCA1 943ins10, and BRCA1 c.211A>G, where they have founder
effect.

Title: Whole exome sequencing reveals greater intratumor genetic heterogeneity in primary breast cancer arising in African
American compared to Caucasian women
Tanya E Keenan1, Edmund A Mroz1, James W Rocco1, Leif W Ellisen1, Beverly Moy1 and Aditya Bardia1. 1Massachusetts General
Hospital, Boston, MA.
Body: Introduction: Multiple studies have reported that African American women have higher mortality from breast cancer than
Caucasian women. While the higher mortality could be due to several factors including access to care, relatively little is known
about tumor genetic differences that may contribute to this disparity. A particularly uncharted area of investigation is intratumor
genetic heterogeneity arising from various subclones within a tumor. The primary objective of this study was to characterize the
differences in mutation patterns and evaluate intratumor heterogeneity in primary breast cancer between African American and
Caucasian women.
Methods: Somatic mutations based on whole exome sequencing of primary breast cancers were analyzed using data from The
Cancer Genome Atlas (TCGA). Individuals with races that were Native American, Asian, or not reported were omitted. Logistic
regression analyses were conducted to compare the odds ratio (OR) with 95% confidence interval (CI) of non-silent gene
mutations by race. Total non-silent gene mutations and genetic heterogeneity, measured by the mutant-allele tumor heterogeneity
(MATH) algorithm (Mroz et al. Cancer 2013), were compared by race with Mann-Whitney two-sample tests and linear regression
analyses, respectively. All regression analyses were adjusted for age and stage. A p value of 0.05 was considered statistically
significant.
Results: The analytical dataset comprised non-silent somatic mutations in unique genes of primary breast cancers from 799
women (689 Caucasian; 110 African American). TP53 mutations were significantly more prevalent in African Americans than
Caucasians (OR 1.7; CI 1.1-2.7, p = 0.01). However, there was no difference in TP53 mutation prevalence after further
adjustment for triple negative status (OR 1.2; CI 0.7-2.0, p = 0.53). Total non-silent gene mutations per patient were significantly
greater in African Americans compared to Caucasians (Table, p = 0.01). Similarly, MATH was significantly greater in African
Americans than Caucasians (Table), specifically 5.3 units (CI 2.6-7.9, p <0.001) greater after adjustment for age and stage.
Further adjustment for triple negative status showed that MATH was still 4.2 units (CI 1.6-6.9, p = 0.002) greater in African
Americans than Caucasians.
Somatic mutations and genetic heterogeneity by race
Total non-silent mutations per patient (median, interquartile range)

Intratumor genetic heterogeneity (mean, 95% CI)
Caucasians
African Americans
33 (21-60)
39.5 (26-74)
39.0 (38.0-40.0)
43.9 (41.3-46.5)
CI, confidence interval.

Conclusions: Primary breast cancers in African Americans exhibit significantly greater total non-silent somatic mutations and
intratumor genetic heterogeneity but not higher frequency of TP53 mutations, suggesting African Americans have genetically
more complex tumors as compared to Caucasian women. Additional research is needed to determine if the observed tumor
genetic differences could contribute to the known racial disparity in breast cancer mortality.

Title: Household net worth is associated with racial disparities in hormonal therapy adherence among women with early stage
breast cancer
Dawn L Hershman1,2,3, Jennifer Tsui2, Jason D Wright1,3, Ellie J Coromilas1, Wei Yann Tsai2 and Alfred I Neugut1,2,3. 1Columbia
University Medical Center, New York, NY; 2Mailman School of Public Health, New York, NY and 3New York Presbyterian Hospital,
New York, NY.
Body: Background: Non-adherence to adjuvant hormonal therapy is common and has been associated with both increased
copayment amount and black race. Studies suggest controlling for wealth may eliminate racial disparities for a variety of medical
condition. We investigated the impact of personal finance on disparities in adherence rates.
Patients and Methods: Using de-identified, integrated pharmacy and medical claims data from the Optum insurance claims
database, we identified women >50 years old on hormonal therapy for early breast cancer with at least 2 mail order prescription
refills between 1/1/07 and 12/31/11. Variables evaluated included demographic and clinical information, annual household
income, estimated household net worth (<$250,000, $250,000-750,000, >$750,000), insurance type, and monthly copayment
amounts (<$10, $10-20, >$20). Non-adherence was defined as a medication possession ratio <80% of eligible days during a
2-year period. Logistic regression analyses were conducted by sequentially including sociodemographic variables, copayment
amount, and each of the economic variables. Due to the strong correlation between net worth and income, and variance of
income with age, net worth was included in multivariate models; an additional analysis was done stratifying patients by net worth.
Results: We identified 15,522 subjects who initiated hormonal therapy; 25% were non-adherent during the study period.
Adherence was 67% with net worth <$25,000 and 81% with net worth >$750,000 (p<0.001). In a univariate analysis, adherence
was negatively associated with black race (OR 0.76, p<0.001), advanced age, comorbidity, extent of surgery, Medicare
insurance, and higher copayment. Adherence was positively associated with higher household income (OR 1.3, p<0.001)
compared to lowest income and with medium net worth (OR=1.26, p<0.001) and higher net worth (OR 1.5, p<0.001) compared to
the lowest net worth group. The negative association of black race and adherence (OR 0.76) was reduced by the sequential
addition of sociodemographic variables (OR 0.78, p<0.001), copayment (OR 0.80, p=0.004), and net worth (0.87, p=0.08). In the
analysis stratified by net worth, black race was no longer associated with decreased adherence. In patients within the high net
worth category, high copayment amount was also no longer associated with decreased adherence (OR 1.0, p=0.77).
Conclusions: We have shown that financial factors, and in particular net worth, partially explain the lower hormone therapy
adherence rate in black women compared to white women. These results suggest economic factors may contribute significantly
to disparities in the quality of breast cancer care.

Title: United States breast cancer mortality trends in young women according to race
Foluso O Ademuyiwa1, Feng Gao1, Lin Hao1, Daniel Morgensztern1, Rebecca L Aft1 and Cynthia X Ma1. 1Washington University
School of Medicine, St Louis, MO.
Body: Purpose
Studies have shown that there is a negative prognostic impact of young age at diagnosis on clinical outcome in women with
breast cancer (BC). We sought to determine if there is a differential effect of race in this high-risk population of women and
examined mortality trends according to race and diagnosis age.
Methods
Using the Surveillance, Epidemiology, and End Results (SEER) program 1990-2009, women diagnosed with invasive BC under
the age 50 were identified. Clinicopathologic characteristics, overall survival, and BC specific survival rates were compared
between racial groups. After adjustments for stage, ER, PR, histology, age, diagnosis year, tumor size, nodal status, and grade,
multivariate logistic regression analyses determined the risk-adjusted likelihood of survival for whites and blacks. Annual hazard
rates (HR) of BC deaths after diagnosis according to race and calendar period, and adjusted relative hazards of death for white
and black women stratified by age at diagnosis were computed.
Results
Overall, 162,976 women were identified, including 126,573 whites (77.7%), 20,405 blacks (12.5%), and 15,998 of other races
(9.8%). At a median follow-up of 85 months, five year disease specific survival rates were 90.1% for whites, and 79.3% for blacks.
During the study period, the annual HR for death in whites decreased by 26% at 5 years after diagnosis, in contrast to the rates in
blacks women decreasing by only 19%. With 1990 as referent, adjusted relative hazards for death in 2005 for white and black
women <40 years were 0.55 (95% CI 0.46-0.66) and 0.68 (95% CI 0.49-0.93) respectively. In whites and blacks 40, 2005
hazards were 0.53 (95% CI 0.47-0.60) and 0.78 (95% CI 0.61-0.99).
Conclusion
Among young women diagnosed with BC, blacks have a worse outcome than whites. Mortality declines in both racial groups have
been observed over time, although more rapid gains have occurred in whites. Emphasis should be placed on improving outcomes
for young BC patients.

Title: Clues to the abundance of triple negative breast cancer in women of African descent
Candice AM Sauder1, Jillian E Koziel2, Mi Ran Choi3, Gang Feng3, Brittney-Shea Herbert2 and Susan E Clare3. 1University of
Texas MD Anderson Cancer Center, Houston, TX; 2Indiana University School of Medicine, Indianapolis, IN and 3Feinberg School
of Medicine, Northwestern University, Chicago, IL.
Body: Background: Triple negative breast cancer (TNBC) is over-represented in indigenous African women and in women of
African descent. The majority of women diagnosed with breast cancer in Nigeria, Uganda and Kenya have TNBC. Dr. Lisa
Newman and colleagues reported TNBC rates of 83.3% in Ghana, 41.4% among African-Americans (AA) and 15.4% for
Caucasian-Americans (C)(Stark et al, Cancer, 2010). The abundance of TNBCs in both African and African-American women has
yet to be explained and the incidence rates suggest that there may be a genetic predisposition to this particularly aggressive form
of breast cancer.
Lipocalin 2 (LCN2), also known as NGAL, is a small-secreted glycoprotein. It chelates iron making this element unavailable to
infectious agents that require it for growth including salmonella and tuberculosis bacilli, and Plasmodium parasites. Inactivation of
the tumor suppressor gene "hypermethylated in cancer 1" (HIC1) results in upregulation of LCN2 at the mRNA and protein levels.
Recently, it was shown that HIC1 is silenced in TNBC when but not in the luminal and HER2 subtypes (Cheng et al, Cancer
Research, 2013). The purpose of this study was to determine if the expression of either of these two genes, LCN2 and HIC1,
differs in breast epithelia as a function of race.
Methods: Multiple epithelial cell lines from normal, healthy breast tissue donated to the Susan G. Komen for the Cure Tissue
Bank at the IU Simon Cancer Center were established (Sauder et al, BMC Cell Biology, 2014). Six epithelial cell lines from AA
and six from C donors were matched for age, menstrual status and Gail risk score. RNA and DNA were extracted from the
cultured cells. Genome-wide gene expression was assayed using the IlluminaHumanWG-6 v3 Expression Bead Chip. Differential
gene expression was determined using PADE (PAirwise Differential Expression; Expression Analysis, Inc). SNPs were identified
using the Illumina Human 1M-Duo BeadChip.
Results: The expression of LCN2 in the normal breast epithelia of AA donors was 20-fold that in C donors (p=0.01). The
expression of HIC1 was 1.94-fold less in AA but this did not reach statistical significance (p=0.29). Analysis of HIC1 DNA data
was complicated by the small number of samples and the paucity of probes within HIC1 on the Human 1M-Duo BeadChip. The
search, therefore, was extended 10,000 bases upstream. There appears to be a heterozygous deletion in two of the AA and one
of the C donors in this region of the genome. We then searched the ENSEMBLE database to determine if there are any SNPs in
HIC1 or its promoter region which differ by Race; only rs8065350 did so. This SNP is in the promoter region of this gene and
occurs within the ENCODE Chip-Seq. tracks of REST/NRSF and Elf1.
Conclusions: LCN2 expression may provide a survival advantage to individuals residing in regions of the world where
iron-requiring infectious agents are endemic. Iron and folic acid supplementation in a clinical trial in East Africa resulted in
significantly more deaths and hospitalizations from infectious diseases. LCN2 expression may provide innate immunity while at
the same time predisposing women to TNBC in an as of yet undetermined manner. It is interesting to note that LCN2 expression
is increased in prostate carcinoma in AA men when compared to C.

Title: Characteristics of benign breast disease and subsequent risk of breast cancer differ by age among African Americans
Michele L Cote1, Julie J Ruterbusch1, Sudeshna Bandyopadhyay2, Quratulain Ahmed2, Barra Alosh2, Eman Abdulfatah2, Haitham
Arabi2 and Rouba Lynn Ali-Fehmi2. 1Wayne State University, Karmanos Cancer Institute, Detroit, MI and 2Wayne State University,
Detroit, MI.
Body: Introduction: A history of benign breast disease (BBD) is common and certain established pathologic features are
associated with increased breast cancer risk. These associations have been reported primarily from studies of white women,
where incidence of BBD peaks in the 4th or 5th decade of life. Previous work in an African American (AA) cohort of women with
BBD showed AA women were younger at their first BBD diagnosis. Thus investigating whether different features of benign lesions
may be associated with age and/or subsequent breast cancer risk in this population is warranted.
Methods: Benign breast biopsies from 1,867 AA women with BBD diagnosed from 1997-2003 were microscopically reviewed for
15 benign features (apocrine metaplasia (AM), ductal hyperplasia (DH), atypical ductal hyperplasia (ADH), lobular hyperplasia
(LH), calcifications (Calc), cysts, duct ectasisa (DE), fibroadenoma (FA), fibrosis, intraductal papilloma (IDP), radial scar (RS),
sclerosing adenosis (SA), columnar cell alterations (CC), mucocele-like tumors (MLT), and atropy), and followed for subsequent
breast cancer in metropolitan Detroit, Michigan. Data from 439 women under 40 and 1,428 women 40 and older at BBD diagnosis
were available for analysis, with a mean follow-up time of 14 years. Differences between age categories for BBD features were
compared using chi-square tests, and risk of breast cancer was estimated with odds ratios (OR) and 95% confidence intervals
(95% CI) calculated from logistic regression analysis.
Results: Women 40 and over were more likely to be diagnosed with nearly all of the benign characteristics compared to younger
women, including: AM, DH, Calc, cysts, CC, and ADH (all p-values <0.001). Younger women were more likely to present with FA
(56% to 43%, p-value<0.001) and no atrophy (44% versus 24%, p-value<0.001) compared to older women. Risk of subsequent
breast cancer was associated with cysts (OR=3.95, 95% CI: 1.09, 14.29) in the younger age group, but not the older (OR=1.25,
95% CI: 0.76, 2.03). CC were also associated with breast cancer risk in the young (OR=5.35, 95% CI: 1.45, 19.73) but not the
older women (OR=1.44, 95% CI: 0.82, 2.50). RS were associated with increased risk in both groups, but only statistically
significant for the older women (OR=3.60, 95% CI: 1.54, 8.39), and not the younger women (OR=6.64, (95% CI: 0.73, 60.67).
Similarly, risk of cancer was associated with a diagnosis of ADH in both groups, but only statistically significant among older
women (OR=3.02, 95% CI: 1.34, 6.79), and not younger women (OR=6.75, 95% CI: 0.71, 64.43).
Conclusions: Characteristics of BBD differ by age, with more women over the age of 40 being diagnosed with various conditions.
Risk of subsequent cancer also varies, although RS, CC and ADH appear to increase risk in both age groups.

Title: Understanding disparities in breast cancer care in Memphis, Tennessee
Elena M Paulus1, Frances E Pritchard1, Simonne S Nouer2, Elizabeth A Tolley2, Brandon S Boyd1, Jesse T Davidson1, Gitonga
Munene1 and Martin D Fleming1. 1University of Tennessee Health Science Center, Memphis, TN and 2University of Tennessee
Health Science Center, Memphis, TN.
Body: Background: Recent literature highlights the troubling racial divide in breast cancer mortality that continues to widen in
most major cities across the country. Although significant progress has been made in improving overall breast cancer survival,
disparities among racial, ethnic, and underserved groups still exist. Previous studies examine the breast cancer mortality disparity
in the 50 largest U.S. cities, and Memphis demonstrates the largest breast cancer mortality disparity for African Americans (AA).
The goal of this investigation is to quantify racial disparities in the context of breast cancer treatment in order to reduce disparities
in recurrence and mortality for breast cancer in the city of Memphis, Tennessee.
Methods: Patients with a biopsy- proven diagnosis of breast cancer over a 10 year period ending December 31, 2012 were
obtained from the tumor registry of a university hospital system. Females of Caucasian and African-American race were included,
while males, patients less than eighteen years of age, and patients with unknown histology, clinical stage, or type of surgery were
excluded. Primary outcomes measured included overall survival and recurrence. Secondary outcomes examined were stage at
diagnosis by race and time from diagnosis to surgery.
Results: 3072 breast cancer patients were reviewed (41% AA). AA patients were more likely to have advanced (Stages II, III, or
IV) clinical stage of breast cancer at diagnosis versus Caucasian patients. Of the 113 recurrences, 62% occurred in AA. Of the
676 deaths, 54% occurred in AA. After adjusting for race and clinical stage of breast cancer, AA breast cancer patients had a 2.0
higher odds of recurrence when compared to Caucasian breast cancer patients (95% CI 1.4, 3.0). AA breast cancer patients were
1.5 more likely to die compared to Caucasian breast cancer patients (95% CI: 1.2, 1.8), after adjusting for race, age at diagnosis,
clinical stage of breast cancer, ER, PR, and HER2 status, and recurrence. AA women with stages 0, I, II, and III breast cancer all
had a statistically significant longer median time from diagnosis to surgery (TDS) than Caucasian women.
Conclusions: African-American patients were more likely to have advanced clinical stages of breast cancer at diagnosis versus
Caucasian patients on a citywide level in Memphis. African-American breast cancer patients have a higher odds of recurrence
and mortality when compared to Caucasian breast cancer patients, after adjusting for appropriate demographic and clinical
attributes. Several factors have been suggested for the disparities including racial differences in access to and utilization of
screening and treatment, risk factors distributed by race and socioeconomic status (SES), biological differences such as tumor
aggressiveness, and cultural factors. More work is needed to develop, evaluate, and disseminate interventions to decrease
inequities in timeliness of care for breast cancer patients.

Title: Race as an independent factor affecting post-mastectomy reconstruction in Asian women
Carmen F Fong1, Alyssa Gillego1, Theresa Shao1, Erika Reategui1, Catherine Campo1, Sarah Cate1, Christopher Mills1, Mark L
Smith1, Gina Aharonoff1 and Susan K Boolbol1. 1Mount Sinai Beth Israel Medical Center, New York, NY.
Body: BACKGROUND:
Racial disparities exist in many areas of breast cancer treatment. Multiple factors influence whether Asian women with breast
cancer undergo immediate reconstruction after mastectomy. This study aims to evaluate breast reconstruction trends at a
comprehensive cancer center and to determine whether race is an independent predictor of breast reconstruction.
METHODS:
Using an IRB-approved, prospectively maintained database, post-mastectomy reconstruction rates were determined for 2003 to
2013. This database was compiled from three university-affiliated hospitals serving a diverse urban population. A total of 5379
patients were identified who were treated for breast cancer during the 10-year period. The odds-ratio for immediate breast
reconstruction was compared among different races. The data was examined to identify factors influencing the decision for
post-mastectomy reconstruction, including age, stage of presentation, marital status, and family history of breast cancer.
RESULTS:
Thirty percent (n=1614) of women treated for breast cancer underwent mastectomy, while seventy percent (n=3765) received
breast-conserving therapy. In the mastectomy group, a unilateral procedure was performed in 93.1% (n=1503) of women and a
bilateral procedure in 6.9% (n=111). The immediate reconstruction rate after mastectomy was 70% (n=1130), with only 30%
(n=484) of women not undergoing reconstruction. Of the women undergoing mastectomy, 58.5% were white, 14.9% were black,
13.1% were Hispanic, 7.5% were Asian and 6% of women did not report race. Of the women who underwent breast
reconstruction after mastectomy, 60.5% were white women, 13.4% were black women, 13% were Hispanic women, 6.9% were
Asian women and 6.2% of women did not report race. The immediate reconstruction rate by race was 72.2% for white women,
63.2% for black women, 69.8% of Hispanic for women, and 64.8% for Asian women. The remainder did not receive breast
reconstruction. The unadjusted odds ratio (OR) for immediate breast reconstruction, including both unilateral and bilateral cases,
for black versus white women was 0.62 (95% confidence interval 0.46-0.84; P=0.001). The OR for breast reconstruction for
Hispanic versus white women was 0.88 (95% confidence interval; 0.63-1.22; .P=0.45). The OR for unilateral breast reconstruction
for Asian versus white women was 0.66 (95% confidence interval 0.44-0.99; P=0.04.)
CONCLUSION:
Reconstruction rates vary by race, with Asian women being less likely to undergo immediate breast reconstruction after
mastectomy. Compared to white women, Asian and black women were both significantly less likely to have immediate breast
reconstruction following mastectomy. Many variables may contribute to this disparity in breast cancer care. Language and cultural
beliefs may be unique factors that influence Asian womens decision for post-mastectomy reconstruction and warrant further
study.

Title: The effect of socioeconomic status and race on breast cancer tumour biology and stage at diagnosis
Priya Sharma1, Dale F Kraemer1, Sarah M Osian1, Carmen Smotherman1, Mia Vincent1 and Laila Samiian1. 1University of Florida
COM Jacksonville, Jacksonville, FL.
Body: Introduction: Mortality from breast cancer has declined over the last 2 decades. However, not all racial groups have
benefited equally. African American women continue to die from breast cancer at higher rates than do white women. While racial
disparities exist in breast cancer outcomes, they may largely be explained by socioeconomic factors. The effect of poverty on
more aggressive breast cancer has not been well studied. We evaluated the association of race, insurance, and age on breast
cancer stage and tumour biology within an urban inner city safety net hospital.
Methods: A retrospective review of a prospective breast cancer database was used to identify 535 women with stage 0 to IV
breast cancer seen at UF Health Jacksonville from the period January 2009-March 2013. Age, race and insurance status at the
time of diagnosis were used as covariates for defining disparities in stage at diagnosis and tumour biology. Tumour profile was
defined as four groups: luminal A (ER+, PR+, HER-2 neg), luminal B (ER+, PR neg, HER-2 neg), HER-2/neu positive, and triple
negative. Insurance categories were defined as Commercial, Medicare, Medicaid/Charity.
Results: There was equal racial distribution between African American (47.3%) and white women (47.5%) with 5.2 % other
races. A relatively large number of patients had Medicaid/Charity coverage (37.1%), followed by Medicare at 32.7 %, and 30.3 %
had commercial insurance. The mean age was 58.8 (SD=12.9). There was no significant association between race and stage at
diagnosis within our patient population (p=0.869). However, women with Medicaid/Charity coverage were diagnosed at more
advanced stage compared to women with other insurnace (adjusted p-value <=0.011). Stage III disease for Medicaid/Charity was
23.1% vs 13.4% for commercial & 12.7% for Medicare coverage. Luminal A was the most prevalent tumour biology overall (58%),
but triple negative tumours were more frequent in the black population (26.6%) vs white (13.1%) vs other races (17.9%). More
aggressive tumour biology was found in black women compared to white women (adjusted p-value=.0007), and those with
Medicaid/Charity coverage compared to other insurance (adjusted p-value <.0001). For example, 28.1% of Medicaid/Charity
patients had triple negative tumours vs 14.8% of commercial and 14.9% of Medicare population. Younger women were diagnosed
at later stages (r=-0.196, p<.0001) and had more severe tumour subtypes (r=0.235, p<.0001).
Conclusion: UF Health Jacksonville encounters a uniquely equal racial distribution of women with breast cancer. Within our
population, women with lower socioeconomic status based on insurance type present with more advanced stage at diagnosis
independent of race. Black race and Medicaid/Charity coverage was significantly associated with more aggressive tumour
biology. These results suggest that socioeconomic factors may have a significant influence on the breast cancer disease process
and may contribute to racial disparities in breast cancer outcomes.

Title: Breast cancer histolological subtypes by race/ethnicity
Carol Parise1 and Vincent Caggiano1. 1Sutter Institute for Medical Research, Sacramento, CA.
Body: BACKGROUND
It is generally accepted that Asian/Pacific Islanders with breast cancer have the same or better survival than whites. Infiltrating
duct carcinoma (IDC) is the most common histologic subtype of breast cancer but little is known about whether this subtype
varies by race/ethnicity. The purpose of this study was to determine if there was a difference in incidence of IDC among different
race/ethnicities.
METHODS
We identified 197,570 cases of AJCC stages 1-4 first primary female invasive breast cancer from the California Cancer registry
2000-2011. The race/ethnicity distribution was as follows: Whites n=130,634; Blacks n=11,927; Hispanics n=32,291; Asian/Pacific
Islanders (API) n=22,027; and American Indians n= 691. Histology was classified using ICDO-3. The distribution of infiltrating
ductal, infiltrating lobular, mixed ductal/lobular, mucinous, tubular, and medullary cancer was examined using contingency tables
and Pearson Residuals. Odds ratios were computed for infiltrating ductal versus all other histologies for each race/ethnicity using
whites as the comparator with logistic regression analysis adjusting for age, stage, socioeconomic status, grade, and
ER/PR/HER2. The stage X race interaction was tested.
RESULTS
Whites had fewer than expected cases of IDC (z = -8.5) and APIs had more than expected (z = 9.2). Blacks (z = 5.6) and
Hispanics (z = 6.1) had a higher number of IDC cases than expected and also had a higher than expected number of cases of
medullary (black: z=12.8; Hispanic: z =10.7). The proportion of American Indians with each of the subtypes was similar to what
would be expected.
The stage X race interaction was statistically significant so analyses were conducted separately for each stage. APIs had an
increased risk of IDC in all stages except for stage 4. Hispanics had an increased risk in stage 3 but a decreased risk in stage 4.
American Indians had the same risk as whites for all stages. Blacks had decreased odds of IDC in stage 1 but increased odds in
stage 3. (Table 1)
Table 1. Odds ratios and 95% confidence intervals of infiltrating duct carcinoma from the California Cancer Registry
2000-2011.
White
Black
Hispanic
Asian/Pacific Islander
American Indian
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
OR (95% CI)
Stage 1
1.00
0.90 (0.81, 0.99)
0.98 (0.92, 1.05)
1.20 (1.12,1.29)
1.31 (0.89, 1.92)
Stage 2
1.00
0.99 (0.90, 1.09)
1.06 (1.00, 1.14)
1.39 (1.29, 1.50)
0.73 (.051, 1.02)
Stage 3
1.00
1.27 (1.10, 1.48)
1.24 (1.12, 1.37)
1.74 (1.52, 1.99)
1.02 (0.59, 1.76)
Stage 4
1.00
1.02 (0.79, 1.31)
0.78 (0.64, 0.94)
1.21 (0.94, 1.56)
0.48 (0.19, 1.19)
Confidence intervals that include 1.00 indicate that the odds of IDC were no different than the reference category
CONCLUSION
The incidence of IDC is not the same for all race/ethnicities and depends on the stage of disease. APIs are at an increased risk
for IDC in all except for the highest stage of disease.

Title: National trend in African American women breast cancer research productivity from 1992-2012
Ralph T Wynn1, Daniel S Chow1, Victoria L Mango1, Lauren C Friedlander1 and Richard Ha1. 1Columbia University Medical
Body: INTRODUCTION: Despite extensive progress in breast cancer prevention, diagnosis and treatment over the past 20
years, African American women suffer disproportionate breast cancer morbidity and mortality. In 1998, breast cancer deaths
among African American women were 28% higher than in white women and the five-year survival rate for African American
women was 71% compared with 86% for white women. The purpose of this bibliometric study is to assess our national research
efforts to combat this racial disparity by analyzing breast cancer research productivity in the United States focusing on African
American women from 1992 to 2012.
METHODS: This retrospective bibliometric analysis of public data was exempt from Institutional Review Board approval. Articles
with "Breast Neoplasm" as a major medical subject heading (MeSH) term published between 1992 and 2012 were identified in
the National Library of Medicine MEDLINE database. In addition, articles with "African Continental Ancestry Group" which
included African American as a major MeSH term were identified. Country of origin, methodology, journal name, first author
specialty and funding sources were recorded. Growth in number of publications was analyzed using linear and nonlinear
regression statistical analysis.
RESULTS: A total of 113,721 journal articles were identified with "Breast Neoplasm" as a major MeSH term worldwide, of which
34,155 (30.0%) were published from the United States. Among United States publications, 668 (2.0%) were specific to African
ancestral populations. From 1992 to 2012, both African ancestral and non-African ancestral specific articles displayed linear
growth patterns (p < 0.0001). National Institute of Health (NIH) funded studies displayed an exponential growth pattern (p <
0.0001) for African ancestral specific articles and displayed a linear growth pattern (p < 0.0001) for non-African ancestral articles.
The largest specialty contributor of African ancestral specific articles was Epidemiology and Public Health (33.2%), followed by
Medicine (internal medicine, family medicine, obstetrics & gynecology, nursing) (24.7%), Basic Sciences (17.4%), Surgery (15%),
Medical and Radiation Oncology (4.7%), Pathology (2.9%) and Radiology (2.1%). African ancestral specific articles were most
frequently published in Cancer (10.8%).
CONCLUSION: Among breast cancer publications from the United States, only 2% of the articles were specific to African
ancestral population, which is concerning given recent data indicating the persistent ethnic disparity in survival. However, the
trend in research productivity in this population is encouraging, largely due to the exponential growth of NIH-funded studies
specific for African American women in the past 20 years. NIH-funded research accounts for 52% of all published African
American specific breast cancer studies as compared to 40.5% of studies that are non-specific to this population. Ultimately,
improvements in breast cancer incidence, mortality, and survival rates in African American women will undoubtedly result from
quality research and should continue to be a priority in our nations breast cancer research agenda.

Title: Influence of socio-economic deprivation score on outcomes for patients diagnosed with early breast cancer over a 9 year
follow up at Kings College Hospital, UK
Olga Oikonomidou1, Mercy Ofuya1 and Anne S Rigg1. 1Kings College Hospital, London, United Kingdom.
Body: Background: Socio-economic deprivation may have an influence on the outcome for patients diagnosed with breast
cancer. The aim of this study was to investigate differences in survival outcome from breast cancer in women from different
socio-economic backgrounds and to identify underlying factors that may contribute to any variation.
Material and methods: This is a retrospective analysis of data from female patients diagnosed with early breast cancer at Kings
College Hospital, a teaching hospital in London, UK between 2004 and 2012. For this study, mean, standard deviation, range and
50th percentile were calculated for the variables investigated. Incidence of death was estimated for deprivation quintile, ethnicity
and year of diagnosis. The Cox proportional hazards model was used to estimate the relationship between survival and the
several variables (deprivation, income score and ethnicity). The survival rates between levels of deprivation were compared using
Kaplan Meier survival curve and log-rank test.
Results: A total of 330 women with a diagnosis of early breast cancer were identified. Analysis of data on all patients (n=330)
showed that the median survival time at which 50% of the patients survive is 4018 days (95% CI: 2556, 5479). Deprivation
quintile ranged from levels 1 to 5 with level 1 defined as affluent and level 5 being most deprived. These scores were reported
for a total of 80 patients. Forty of these were alive and 40 had died from metastatic breast cancer. The two groups were matched
for age, year of diagnosis and stage of disease. The highest incidence of death (19/33; 58%) was in level 5 of the deprivation
quintile. There was no difference in patients survival rates between the 5 levels of the deprivation quintile. There was no
difference between survival rates in white ethnic group compared with the black (2=0.03, p=0.86). There was no evidence of
significance in relationship between hazard of death and deprivation in black and white ethnic groups. Similarly, there was no
evidence of significance in relationship between hazard of death and income score in black and white ethnic groups. Adjusting for
ethnicity also showed that deprivation and income score had no effect on hazard of death.
Conclusion: According to our study higher degree of socio-economic deprivation is associated with higher mortality in breast
cancer. Previous studies from different hospitals in UK have shown that black women have a significantly poorer outcome that
white patients despite equal access to healthcare. Our study has shown that there is no correlation between ethnicity (white
ethnic and black groups) and survival rates, hazard of death or deprivation in patients treated in Kings College Hospital.

Title: Distinct early proliferation response to neoadjuvant anti-HER2 antibody drug conjugate +/- endocrine therapy in early breast
cancer in the WSG ADAPT HER2+/HR+ trial
Oleg Gluz1,2, Ulrike Nitz1,2, Kuemmel Sherko3, Kraemer Stefan4, Michael Braun5, Claudia Schumacher6, Bahriye Aktas7, Helmut
Forstbauer8, Toralf Reimer9, Peter Fasching10, Jochem Potenberg11, Daniel Hofmann1, Ronald E Kates1, Rachel Wuerstlein1,12,
Matthias Christgen13, Hans H Kreipe13 and Nadia Harbeck1,12. 1West German Study Group, Moenchengladbach, Germany; 2Ev.
Hospital Bethesda, Breast Center Niederrhein, Moenchengladbach, Germany; 3Clinics Essen-Mitte, Clinics for Senology/Breast
Center, Essen, Germany; 4University Clinics Cologne, Breast Center, Cologne, Germany; 5Rotkreuz Clinics Munich, Women
Clinics, Munich, Germany; 6St Elisabeth Hospital, Cologne, Germany; 7University Clincs Essen, Women Clinics, Essen, Germany;
8
Practice Network Troisdorf, Troisdorf, Germany; 9Clinics Suedstadt, Rostock, Germany; 10University Clinics Erlangen, Women
Clinics, Erlangen, Germany; 11Ev. Waldkrankenhaus, Berlin, Germany; 12University Hospital Munich, Breast Center, CCC of LMU,
Munich, Germany and 13Medical School Hannover, Institute of Pathology, Hannover, Germany.
Body: Background: HER2+, hormone receptor-positive (HR+) breast cancer is a distinct subtype associated with good prognosis
but poor response to standard chemotherapy + anti-HER2 (single or dual blockade). Substantial overtreatment by
poly-chemotherapy combined with anti-HER2 therapy is suspected in this subtype. Yet, the efficacy of combining endocrine
therapy (ET) with anti-HER2 therapy or novel antibody drug conjugates like T DM1 without systemic chemotherapy remains
unclear.
Methods: ADAPT HER2+/HR+ is a phase II, randomized, neoadjuvant, 3-arm trial (12 weeks) in patients with cT1c-cT3, cN0/+
HER2+, ER+ and/or PR+ early BC. Arm A: T-DM1 (3.6 mg/kg) alone; Arm B: T DM1 + ET (premenopausal: tamoxifen,
postmenopausal: AI); Arm C: trastuzumab + ET. Postoperative chemotherapy is recommended together with completion of one
year of trastuzumab. Initial and serial core biopsies were obtained prior to therapy and after 3 weeks. First translational analysis
of the trial run-in phase (n=130) focuses on dynamics of HER2, Ki67, ER and PR.
Results: 162 tumors, HR+ and HER2+ by local pathology, were screened; n=130 were HR+ and HER2+ by central pathology and
randomized at 40 trial sites in Germany between 11/2012 and 03/2014 (Arm A/B/C: 37/49/44).
Median age was 49 years; 60% were cT2-3, 32% cN+, 75% central G3; median baseline Ki67 was 30%; 49 patients were treated
by TAM and 44 postmenopausal patients by AI in ET containing arms.
Three-week core biopsies were available in 117 patients (arm A/B/C: n=33/43/41), n=99 with invasive tumor tissue (61/76 (80%)
in T-DM1 containing arms and 38/41(93%) in trastuzumab + ET arm). Three-week Ki67 could only be analyzed in n=73 (53% of
patients in T-DM1-containing arms, 81% in the T+ET arm) due to a lacking amount of cells for counting (<500).
Median fractional decrease in proliferation (Ki67) after 3 weeks of therapy was 40% in the T-DM1 + ET arm (B) as compared to
14% and 25% in the T-DM1 (A) and T+ET (C) arms, respectively. Among postmenopausal patients, the contrast (52% (B) vs. 0%
(A) and 28% (B)) was significant. Mean PR expression change (absolute, measured in %) was -15 in the B Arm vs. 7 and -7 in
the A and C arms, respectively (p=0.04) particularly in postmenopausal women ( 25 vs. +13 and -10, respectively, p=0.04).
Baseline ER expression was positively associated with early proliferation response (fractional Ki67 decrease), e.g., by logistic
regression using 30% decrease as response criterion. Data on the impact of gene mutations and further molecular markers on
proliferation response will be available for presentation at the meeting.
Conclusions: In the unique neoadjuvant ADAPT HER2+/HR+ trial, the combination of T-DM1 with ET (particularly AI) seems to be
associated with a strong early proliferation response and PR expression drop in anti-HER2 antibody (drug conjugate) +/- ET,
without systemic pre-operative chemotherapy. Interim analysis is scheduled after 130 completely treated patients. The high
percentage of non-invasive tissue biopsied after 3 weeks in the T-DM1 arms is intriguing, as it would be consistent with higher
pathological complete response rates.

Title: Clinical, biochemical, and genomic predictors of trastuzumab-related cardiotoxicity: Results of CATS
Shom Goel1, Hao Guo1, William Barry1, Bronwyn Murray2, Jodi Lynch3,4, Patricia Bastick3, Lorraine Chantrill5, Belinda Kiely5,
Richard Bell6, Ehtesham Abdi7, Josie Rutovitz8, Ray Asghari9, Anne Sullivan10, Michelle Harrison2, Maija Kohonen-Corish11 and
Jane Beith2. 1Dana-Farber Cancer Institute, Boston, MA; 2Chris O'Brien Lifehouse, Sydney, NSW, Australia; 3St George Hospital,
Sydney, NSW, Australia; 4Sutherland Hospital, Syndey, NSW, Australia; 5Macarthur Cancer Therapy Centre, Sydney, NSW,
Australia; 6Andrew Love Cancer Centre, Geelong, VIC, Australia; 7Tweed Hospital, Tweed Heads, NSW, Australia; 8Northern
Haematology and Oncology Group, Sydney, NSW, Australia; 9Bankstown Cancer Centre, Sydney, NSW, Australia; 10Concord
Repatriation General Hospital, Sydney, NSW, Australia and 11Garvan Institute, Sydney, NSW, Australia.
Body: Background: Adjuvant therapy for HER2-positive breast cancer (HBC) often comprises anthracyclines (A) followed by
trastuzumab (T). Trastuzumab-related cardiotoxicity (TRC) is Ts primary toxicity, and led 15% of patients in NSABP-B31 to stop
T early. Earlier detection of TRC might enable timely therapeutic intervention, lowering the number of patients ceasing T
prematurely. Small, retrospective, unvalidated studies have proposed various markers as predictors for TRC. The Cardiotoxicity
of Adjuvant Trastuzumab Study (CATS) aims to assess a panel of clinical, biochemical, and genomic/immunologic markers as
predictors for TRC.
Methods: HBC patients with a left ventricular ejection fraction (LVEF) 50% scheduled to receive adjuvant A followed by 12
months T were eligible. Serum N-terminal pro-B type natriuretic peptide (NT pro-BNP) and troponin I were measured centrally at
baseline (T1), at completion of A (T2), and after 3 and 6 months of T. LVEF was measured at T1, T2, and then 3-monthly until
completion of T. Germline SNPs in ERBB2 (V655I), FCGR2A (H131R), and FCGR3A (V158F) were assessed. TRC was defined
as: cardiac death, NYHA class 3/4 heart failure, grade 3/4 arrhythmia/ischemia, drop in LVEF >15% from baseline, or drop in
LVEF of >10% to <55%. 1 patient with LVEF <50% at T2 was excluded. Uni- and multivariate logistic regression were used to
determine features associated with TRC. Serum markers were added to the model using forward selection.
Results: 17 centers enrolled 222 patients (217 evaluable). Patient characteristics are shown.
Baseline characteristics
Age [median (range), yrs]
52 (28-77)
Baseline LVEF [median (range)]
64% (50-82%)
Left-sided tumors [n(%)]
105 (48%)
History of hypertension [n(%)]
48 (22%)
Doxorubicin [n(%), median total dose]
80 (37%), 242mg/m2
Epirubicin [n(%), median total dose]
137 (63%), 300mg/m2
Concurrent taxane with T [n(%)]
217 (100%)
TRC occurred in 33 patients (15.2%). In multivariate models, lower baseline LVEF (OR 3.23 for a 5% lower LVEF, [95% CI
1.96-5.33], p<0.0001) and greater absolute decline in LVEF from T1 to T2 (OR 3.25 for each 5% drop in LVEF, [95%CI
1.83-5.79], p<0.0001] were each independent predictors of TRC. NT pro-BNP increased from T1 to T2 (median =1.31pmol/L,
p<0.0001) and then fell to baseline levels. Troponin I increased from T1 to T2 in 64.5% of patients (p<0.0001) and remained
above baseline 6 months post-A (p<0.0001). NT pro-BNP and troponin I at T1 were not predictive for TRC. There were trends for
the absolute increase in NT-pro BNP and troponin I from T1 (pre-A) to T2 (post-A) to be associated with risk of TRC (p=0.08 and
0.09) in univariate analyses. Germline SNPs were not predictive of TRC.
Conclusion: This is the largest prospective study analyzing predictors for TRC, and the first studying immune-related SNPs in this
context. Consistent with the literature, lower LVEF at baseline increases risk of TRC. Previously unreported, the absolute
decrease in LVEF after A is also predictive for TRC. The persistent elevation of troponin I 6 months after completing A
demonstrates chronic cardiac stress during therapy. Ongoing modeling analyses (to be presented at the meeting) will determine
whether longitudinal changes in serum biomarkers can predict subsequent development of TRC.

Title: Activating mutations in ERBB2/HER2 as found by FoundationOneTM represent potential therapeutic targets in breast
cancer
Gary A Palmer1, Jeffrey S Ross1,2, Kai Wang1, Garrett M Frampton1, Siraj M Ali1, Norma Palma1, Deborah Morosini1, Vincent A
Miller1, Roman Yelensky1, Doron Lipson1, Philip J Stephens1 and Juliann Chmielecki1. 1Foundation Medicine, Cambridge, MA and
2
Albany Medical College, Albany, NY.
Body: Background: Targeted ERBB2/HER2 inhibitors are FDA-approved for the treatment of breast and gastric cancers.
Immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) to test for overexpression and amplification of
ERBB2/HER2, respectively, are performed as part of routine clinical care. Recently, activating mutations in ERBB2 have been
reported and may confer sensitivity to targeted agents. Testing for these mutations is not routine, and testing for amplifications is
not done outside of approved indications. We explored the complete spectrum of activating ERBB2 mutations and amplifications
across a collection of 7,300 solid tumor specimens, including a large number of breast cancer specimens, to determine a)how
many breast cancer patients could benefit from HER2-targeted therapy due to activating mutations in ERBB2 and b)how
widespread ERBB2 alterations are across the solid tumor spectrum.
Methods: Extracted DNA from clinical tumor samples underwent comprehensive genomic profiling using FoundationOne . 3769
exons of 236 cancer-related genes and 47 introns from 19 genes that are frequently rearranged in cancer were fully sequenced to
high, uniform coverage and results were analyzed for base substitutions, insertions and deletions, amplifications/deletions, and
rearrangements.
Results: Known oncogenic ERBB2 alterations were identified in approximately 6% of all solid tumors across 27 different
histologies. Of all the ERBB2 alterations, activating mutations in ERBB2 were identified in 131 samples and amplifications were
observed in 246 samples. Two samples harbored an ERBB2 rearrangement. Ten samples harbored multiple ERBB2 mutations,
yet mutations and amplifications were mutually exclusive in 91% of mutated cases. ERBB2 amplification in breast and gastric
cancers accounted for only 30% of these alterations. Breast cancer accounted for 37% of all the ERBB2 alterations detected, and
included primarily amplifications. 25% of the alterations found in breast cancers were activating base substitutions. Standard tests
for overexpression or amplification of ERBB2 would fail to detect these potentially treatable mutations. Non-amplification ERBB2
alterations were enriched in cases of relapsed lobular breast cancer. Multiple breast cancer patients with non-amplification
ERBB2 alterations have responded to combinations of anti-HER2 targeted therapies.
Conclusions: Comprehensive genomic profiling through FoundationOne identifies 25% more breast cancers that may be
susceptible to HER2 targeted therapy due to the presence of activating mutations in ERBB2. Also, many more solid tumor cases
could potentially benefit as well from ERBB2 testing. The current policy of testing only breast and gastric tumors for only HER2
amplifications is greatly limiting the potential value of the ERBB2/HER2 inhibitors.

Title: Effect of pre-treatment with cyclophosphamide on MM-302 (HER2-targeted liposomal doxorubicin) deposition in
HER2-positive metastatic breast cancer patients assessed by 64Cu-MM-302 PET/CT
Kathy Miller1, Patricia M LoRusso2, Pamela Munster3, Ian Krop4, Cynthia Ma5, Helen Lee6, Joe Reynolds6, Karen Campbell6, Victor
Moyo6, Bart Hendriks6, Thomas Wickham6, Barry A Siegel5 and Anthony F Shields2. 1Indiana University Melvin and Bren Simon
Cancer Center, Indianapolis, IN; 2Karmanos Cancer Institute, Detroit, MI; 3Helen Diller Family Comprehensive Cancer Center,
San Francisco, CA; 4Dana-Farber Cancer Institute, Boston, MA; 5Washington University School of Medicine, St Louis, MO and
6
Merrimack Pharmaceuticals, Inc, Cambridge, MA.
Body: Background: Liposomal encapsulation of doxorubicin has addressed the cardiotoxicity of free doxorubicin, but has not
achieved an improvement in anti-tumor activity in metastatic breast cancer. MM-302 is a HER2-targeted liposomal doxorubicin,
specifically designed to target tumor cells overexpressing HER2 and minimize uptake into normal cells such as cardiomyocytes,
which express low levels of HER2. MM-302 and MM-302 plus trastuzumab are being studied in patients as part of an ongoing
Phase 1 clinical trial. Published reports indicate that deposition into solid tumors is a rate-limiting step in liposome-mediated
delivery of drug to tumor cells. In order to understand drug deposition into solid tumors, we have radiolabeled MM-302 with 64Cu
for imaging by PET/CT. Further, preclinical work has demonstrated that pretreatment with cyclophosphamide has the ability to
improve liposomal drug delivery by making the tumor microenvironment permissive for deposition and retention of liposomes.
This Phase 1 study evaluates the delivery of 64Cu-MM-302 to solid tumors and the ability of cyclophosphamide pretreatment to
increase delivery of 64Cu-MM-302 to tumors.
Methods: Patients aged 18 years with histologically confirmed HER2-positive advanced breast cancer that has progressed or
recurred on standard therapy or for which no standard therapy exists, who have adequate performance status, bone marrow
reserve and organ function, were eligible for the study. Patients received 30 mg/m2 of MM-302 plus 6 mg/kg trastuzumab, q3w
and 400 MBq (10.8 mCi; 3-5 mg/m2, doxorubicin basis) of 64Cu-MM-302 (cycle 1 only) with or without pretreatment of 450 mg/m2
cyclophosphamide. Patients underwent PET/CT on the day of administration and on day 2, day 3 or both. The primary goal of this
analysis was to study delivery of 64Cu-MM-302 and the effect of cyclophosphamide pretreatment. Other endpoints being studied
include safety, dosimetry, and treatment response.
Results: Combination of MM-302 with cyclophosphamide was well tolerated and no issues were reported related to 64Cu-MM-302
administration or imaging. Uptake of 64Cu-MM-302 in tumor lesions increased over time with significant deposition at 24 and 48 h
while activity in the blood decreased over time. Median lesion deposition of 64Cu-MM-302 (as % i.d./kg) in the patients with
cyclophosphamide pretreatment was higher than in those without pretreatment: 6.0 (n=5 patients/20 lesions) vs. 4.4 (n=7
patients/36 lesions) at 24 h and 7.6 (n=4 patients/16 lesions) vs. 4.0 (n=4 patients/17 lesions) at 48 h.
Conclusions: PET/CT imaged 64Cu-MM-302 deposition into diverse tumor lesions, including liver, brain and bone metastases. Our
preliminary data thus far suggest that cyclophosphamide pretreatment increases delivery of 64Cu-MM-302 to patient tumors, as
predicted by preclinical models. Correlation between 64Cu-MM-302 tumor deposition and lesion/patient response is currently
being investigated.

Title: Effects on outcome of concomitant neoadjuvant chemotherapy-trastuzumab compared with sequential neoadjuvant
chemotherapy followed by post-operative trastuzumab
Carlo Palmieri1, Iain RJ Macpherson2, Kelvin Yan3, Felipe Ades Moraes4, Pippa Riddle5, Riz Ahmed5, Waheeda Owadally6,
Barabara Stanley2, Deep Shah3, Ondrej Gojis3, Adam Januszewski2, Conrad Lewanski7, Rebecca Asher8, Daniel Lythgoe8,
Evandro De Azambuja4, Mark Beresford6 and Sacha J Howell9. 1Clatterbridge Cancer Centre NHS Foundation Trust, Bebbington,
Merseyside, United Kingdom; 2Beatson West of Scotland Cancer Centre, Glasgow, Lanarkshire, United Kingdom; 3Imperial
College Healthcare NHS Trust, London, United Kingdom; 4Institut Jules Bordet, Universit Libre de Bruxelles, Brussels, Belgium;
5
West Middlesex University Hospital, London, Middlesex, United Kingdom; 6Royal United Hospital, Bath, Somerset, United
Kingdom; 7Ealing Hospital NHS Trust, London, Middlesex, United Kingdom; 8Cancer Research UK Liverpool Cancer Trials Unit,
Liverpool, Merseyside, United Kingdom and 9Christie NHS Foundation Trust, Manchester, United Kingdom.
Body: Background:
Neoadjuvant chemotherapy delivered with trastuzumab (NCT) has been shown to increase the rates of pathological complete
response (pCR) compared to neoadjuvant chemotherapy (NC) alone in women with HER2 positive breast cancer (BC). pCR in
this setting has been associated with improved event free survival (EFS). However, no study has yet investigated the effect on
outcomes of NCT compared to NC followed by adjuvant trastuzumab initiated postoperatively (NCAT). This study sought to
investigate the impact on breast cancer outcomes of concomitant (NCT) versus sequential (NCAT) treatment in HER2 positive
early breast cancer.
Methods:
Women with HER2 positive invasive breast cancers treated with neoadjuvant chemotherapy between 2006-2010 were identified
at each of 7 European institutions and the case notes reviewed. Preoperative clinical, radiological and pathological details,
treatment details and pathology results following breast surgery were reviewed. To be defined as NCT at least one dose of
trastuzumab needed to be given preoperatively. pCR was defined as absence of invasive disease in breast and lymph nodes.
Multivariable Cox regression and logistic regression were used to Survival outcomes for event-free survival (EFS) were calculated
by log rank analysis model the influence of a number of factors on event-free survival (EFS) and pCR respectively.
Results:
236 patients were identified; 138 (58%) received NCAT & 98 (42%) received NCT. The median follow up for the whole group was
53.7 months (IQR 41.7-68.8), 61.5 months (IQR 50.3-78.5) for NCAT group and 44.8 months (range 37-53.9) for NCT group. The
5-year EFS for NCAT vs NCT was 59.3% (95% CI: 49.8-67.6) and 69.6% (95% CI: 51.5-82.0) respectively. The unadjusted
hazard ratio (HR) for EFS with NCT compared with the NCAT was 0.63 (95% CI 0.371.08; p=0.091). NCT significantly increased
the odds of having pCR relative to NC (OR: 4.39 (2.18-8.86); p<0.001), and pCR was associated with a significantly improved
EFS, with an unadjusted HR of 0.23 (95% CI 0.080.64; p=0.002). Multivariable analysis revealed that treatment group, tumour
size and ER status were significantly associated with EFS. NCT was associated with a reduced risk of relapse relative to NCAT
(HR 0.48, 95% CI 0.26-0.89). In ER negative tumours NCT was significantly associated with a reduced risk of an event relative to
NCAT (HR:0.25; 95% CI, 0.10-0.62), this was not observed for ER positive tumours (HR: 1.07; 95% CI, 0.46-2.52).
Conclusion
Concomitant as compared to sequential trastuzumab is associated with improved outcomes in the neoadjvuant setting for women
with ER negative tumours. These data further support the need for the early introduction of targeted combination therapy in
women with ER negative/HER2 positive BC.

Title: Correlation between ERBB2 mRNA levels, HER2-dependence and susceptibility to trastuzumab in human breast cancer
Elda Tagliabue1, Viola Regondi2, Loris De Cecco3, Serenella M Pupa4, Manuela Campiglio5, Maria Luisa Carcangiu6, Sylvie
Mnard7 and Tiziana Triulzi8. 1Molecular Targeting Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 2Molecular
Targeting Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 3Functional Genomics Core Facility, Fondazione
IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 4Molecular Targeting Unit, Fondazione IRCCS Istituto Nazionale dei Tumori,
Milan, Italy; 5Molecular Targeting Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 6Anatomic Pathology A Unit,
Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 7Molecular Targeting Unit, Fondazione IRCCS Istituto Nazionale
dei Tumori, Milan, Italy and 8Molecular Targeting Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy.
Body: While results thus far demonstrate the clinical benefit of trastuzumab in breast cancer, some patients do not respond to
this reagent. Identifying a robust clinical or molecular predictor of adjuvant trastuzumab benefit has proven challenging and even
the most obvious candidate biomarker for a predictor of trastuzumab benefit, HER2 expression as assessed by
immunohistochemistry (IHC) and/or fluorescence in situ hybridization (FISH), has proven to be surprisingly ambiguous in realizing
benefit from the antibody. Results of a recent study that profiled expression of select genes in archived formalin-fixed,
paraffin-embedded (FFPE) tumor blocks from a randomized study of breast cancer patients treated with adjuvant trastuzumab
support a relationship between ERBB2/ESR1 mRNA levels and trastuzumab sensitivity. Based on that observation, we conducted
whole-genome profiling by DASL technology of archival FFPE tumor blocks from 53 HercepTest 3+/2+ FISH-positive patients
treated with adjuvant trastuzumab in our Institute; 308 genes were significantly associated with relapse-free survival by Cox
proportional hazard model (<0.005; permutation test p<0.01). We then developed a relapse risk 41-gene classifier (TRAR) able
to discriminate cases mainly according to ERBB2 and ESR1 mRNA expression levels and to predict tumor relapse in both
trastuzumab adjuvant and neoadjuvant settings. No correlation was found between ERBB2 mRNA and HER2 protein expression
(the latter assessed by IHC on FFPE tumor sections using serial dilutions of anti-HER2 antibody), nor did HER2 protein
expression correlate with relapse risk in these tumors. Analysis of HER2-amplified breast carcinoma cell lines revealed higher
levels of ERBB2 mRNA in oncogene-addicted than in non-addicted cells in correlation with HER2 activation assed by Western
blot but not with total HER2 protein amount. Moreover, ERBB2 mRNA expression levels in HER2-amplified breast carcinoma
cells correlated with trastuzumab-mediated cellular cytotoxicity (ADCC) in vitro (r=0.96, p=0.038), whereas no correlation was
found between ADCC and membrane-associated HER2.
Together, our findings strongly suggest that ERBB2 mRNA, but not protein levels, mirror HER2 activity and thus oncogene
addiction of tumor cells and susceptibility to trastuzumab in amplified HER2-positive tumors.
Supported by AIRC.

Title: HER2 positive (HER2/CEP17 ratio2.0) invasive mammary carcinomas with average <4.0 HER2 and <2.0 CEP17
signals/cell: Clinicopathologic features, correlation with HER2 immunohistochemistry and response to neoadjuvant chemotherapy
Chad A Livasy1, Kimberly Limentani2, Benjamin C Calhoun3 and Steven Limentani4. 1Levine Cancer Institute, Charlotte, NC;
2
Cleveland Clinic, Cleveland, OH; 3Thomas Jefferson University Hospital, Philadelphia, PA and 4Levine Cancer Institute,
Charlotte, NC.
Body: Background: ASCO/CAP guidelines for the determination of HER2 amplification have recently been revised in an attempt
to clarify which patients will benefit from HER2 directed therapy. Patients with HER2/CEP17 ratios 2.0 were eligible for the first
generation trials of adjuvant trastuzumab regardless of the average number of HER2 signals/cell. Array CGH and FISH studies
have since demonstrated the presence of complex segmental aneusomy of chromosome 17 in a subset of breast tumors. These
focal copy number gains and losses can skew the HER2/CEP17 ratio and result in discordant results with HER2
immunohistochemistry (IHC). There is limited data on the clinicopathologic features, HER2 protein expression and response to
therapy for tumors showing a HER2/CEP17 ratio 2.0 and average <4.0 HER2 signals/cell.
Methods: All cases with HER2/CEP17 ratio 2.0 and average <4 HER2 signals/cell were identified from our database from
2009-2013. HER2 IHC was performed and scored on all identified cases. Tumor grade, histologic subtype, hormone receptor
status and menopausal status were recorded. For patients receiving neoadjuvant chemotherapy, response was measured using
residual cancer burden (RCB) class and yAJCC pathologic stage.
Results: 70 (1.5%) of 4659 FISH cases met criteria for inclusion in the study. 22 (31%) of the 70 patients were premenopausal.
All tumors were ductal (no special type) with the following features features: low-grade 14%, intermediate-grade 40%, high-grade
45%, estrogen receptor positive 79% and progesterone receptor positive 65%. HER2 protein evaluation demonstrated 52 (74%)
negative (0-1+), 16 (23%) equivocal (2+) and 2 (3%) positive (3+) results. Most tumors (83%) showed 1+ or 2+ staining. 12
patients received neoadjuvant chemotherapy (see table). Six of these patients did not receive trastuzumab as part of their therapy
due to congestive heart failure (n=1) or uncertainty about the patient's HER2 status (n=5).
Neoadjuvant chemotherapy patient summary
Patient # (regimen)
Ratio
HER2 copy#
CEP17 copy#
HER2 IHC
ER/PR
yAJCC
RCB class
Patient 1 (AC+TH)
2.2
3.9
1.8
1+
+/+
ypT0 N0(i+)
Patient 2 (TCH)
2.2
3.8
1.7
1+
+/-
ypT0 N0
Patient 3 (TCH)
2.2
3.8
1.7
2+
+/+
ypTis N0
Patient 4 (AC+TH)
2.3
3.0
1.3
+/+
ypT2 N2a
Patient 5 (TCH)
2.2
3.7
1.7
1+
-/-
ypT1c N0
Patient 6 (TCH)
2.2
3.9
1.8
2+
-/-
ypTis N0
Patient 7 (TC)
2.0
3.4
1.7
-/-
ypT0 N0
Patient 8 (AC+T)
2.2
3.6
1.7
1+
+/-
ypT1c N0
Patient 9 (TC)
2.1
3.9
1.9
1+
-/-
ypT0 N0
Patient 10 (AC)
2.5
3.4
1.4
1+
+/+
ypT1b N1a
Patient 11 (TAC)
2.4
3.6
1.5
2+
-/-
ypT3 N1mi
Patient 12 (AC)
2.0
3.3
1.6
1+
-/-
ypT1c N1a
Congestive heart failure

Conclusions: HER2 FISH positive invasive mammary carcinomas with a HER2/CEP17 ratio 2.0 and average <4.0 HER2
signals/cell are rare (1.5% in this study) and frequently show 1+ or 2+ staining by IHC. Treatment decisions for these patients
were impacted by the uncertainty of HER2 status under the previous guidelines. In this small data set, these tumors
demonstrated a good response to neoadjuvant trastuzumab-based chemotherapy and further support the recent modifications of
ASCO/CAP guidelines to determine HER2 amplification.

Title: A phase 1b study of ONT-380, an oral HER2-specific inhibitor, combined with ado-trastuzumab emtansine (T-DM1), in
HER2+ metastatic breast cancer (MBC)
Virginia F Borges1, Erika Hamilton2,3, Denise A Yardley2,3, Jorge Chaves4, Nathalie Aucoin5, Cristiano Ferrario6, Luke Walker7 and
Ian Krop8. 1University of Colorado Cancer Center, Aurora, CO; 2Sarah Cannon Research Institute, Nashville, TN; 3Tennessee
Oncology, PLLC, Nashville, TN; 4Northwest Medical Specialties, Tacoma, WA; 5Hopital Cite-de-la-Sante, Laval, QC, Canada;
6
Jewish General Hospital, Montreal, QC, Canada; 7Oncothyreon Inc, Seattle, WA and 8Dana-Farber Cancer Institute, Boston, MA.
Body: Background: ONT-380 (also known as ARRY-380) is a potent, selective small molecule inhibitor of HER2 with 500-fold
selectivity compared to EGFR. Preclinical studies have demonstrated synergistic activity with ONT-380 and chemotherapy or
trastuzumab, as well as superior activity compared to lapatinib and neratinib in models of HER2+ CNS metastases. In a Phase 1
single agent study in HER2+ MBC, ONT-380 was well tolerated and provided clinical benefit with minimal EGFR-type toxicities,
with an MTD of 600 mg BID using an API-in-capsule formulation. Based on the potential for dual blockade of HER2 to lead to
clinical benefit, ONT-380 is being evaluated in combination with T-DM1 in patients previously treated with a taxane and
trastuzumab for metastatic disease.
Methods: This 3+3 dose escalation study evaluates escalating doses of ONT-380 in a new tablet formulation combined with
T-DM1 at 3.6 mg/kg IV once every 21 days. Prior treatment with trastuzumab and a taxane are required. Prior lapatinib or
neratinib therapy and asymptomatic brain metastases (treated or untreated) are allowed. Previous T-DM1 is not permitted.
Normal left ventricular ejection fraction and anthracycline exposure 360 mg/m2 is required. Study assessments include safety,
ONT-380 and T-DM1 PK, tumor response by RECIST 1.1, and CNS response by both modified RECIST and volumetric criteria.
Dose escalation will be followed by enrollment of expansion cohorts in patients with and without CNS disease.
Results: As of 21 May 2014, 7 patients have been treated with ONT-380 at 300 mg BID for 15 cycles. One dose limiting toxicity
(DLT) of Grade 3 ALT/AST elevation was seen in the ONT-380 300 mg BID cohort, requiring a dose reduction for both agents
with subsequent cohort expansion to 6 patients. No further DLTs have been seen in this cohort, and no other dose reductions
have been required. Most toxicities have been Grade 1 or 2, with the most common regardless of attribution being nausea,
fatigue, diarrhea, and thrombocytopenia. Two Grade 3 AEs have been reported, including the DLT of ALT/AST elevation, and one
event of thrombocytopenia, considered related to T-DM1 but not ONT-380. There has been no Grade 3 diarrhea and no SAEs. In
the four patients evaluable for response to date, best response has been 1PR, 2 SD, and 1 PD. Three patients with prior CNS
radiation have had continued reduction in CNS lesions on study. Initial PK data indicate greater ONT-380 exposure is achieved
with the new tablet formulation compared to the earlier capsule formulation with no evidence of drug interaction with T-DM1.
Conclusions: Treatment with ONT-380 and T-DM1 has been associated with an acceptable safety profile, with only one DLT and
minimal Grade 3 toxicity, including no Grade 3 diarrhea or rash. Early evidence of disease control has been seen, including in
patients with CNS metastases. Dose escalation continues, and updated results will be presented for additional cohorts.

Title: Phase 1 study of trastuzumab emtansine in HER2-positive metastatic breast cancer patients with normal or reduced
hepatic function
Chunze Li1, Priya Agarwal1, Susan Dent2, Anthony Goncalves3, Joo-Hee Yi1, Alexander Strasak4, Marjorie Green1, Sandhya
Girish1 and Pat LoRusso5. 1Genentech, San Francisco, CA; 2Ottawa Hospital Cancer Centre, Ottawa, ON, Canada; 3Institut
Paoli-Calmettes, Marseille, France; 4F. Hoffmann-La Roche Ltd and 5Karmanos Cancer Institute, Detroit, MI.
Body: Introduction
Trastuzumab emtansine (T-DM1) is an antibodydrug conjugate comprising trastuzumab, a stable linker, and the microtubule
inhibitor DM1. In phase 3 studies of HER2-positive metastatic breast cancer (MBC), T-DM1 significantly increased
progression-free survival (EMILIA and TH3RESA) and overall survival (EMILIA) vs. control regimens. Few patients (24%)
treated with T-DM1 experience grade 3 increases in transaminases. Currently, there are no data on the pharmacokinetics (PK)
of T-DM1 in patients with hepatic impairment. This international, multicenter, open-label, parallel group, phase 1 PK study
(BO25499/NCT01513083) is designed to assess the PK of T-DM1 and relevant analytes in MBC patients with normal hepatic
function and mild or moderate hepatic impairment; safety and efficacy will also be evaluated.
Methods
To obtain 8 evaluable patients, up to 10 patients each with HER2-positive MBC and ECOG performance status of 02 were
enrolled in 1 of 3 independent cohorts based on hepatic function per Child-Pugh criteria: normal hepatic function, mild hepatic
impairment (Child-Pugh A), and moderate hepatic impairment (Child-Pugh B). Patients with severe hepatic impairment (Child
Pugh C) were ineligible. Patients received 3 cycles of T-DM1 3.6 mg/kg every 3 weeks. After 3 cycles, patients could continue to
receive T-DM1 until disease progression, unmanageable toxicity, or study termination in the present study, or enroll in an
extension study (BO25430/TDM4529g). PK samples were collected during cycle 1 (days 1 [predose, 30 m and 4 h postinfusion],
2, 3, 4, 8, 11, 15, and 18); cycle 2 (day 1 [predose, 30 m postinfusion]); and cycle 3 (days 1 [predose, 30 m postinfusion], 8, 15,
and 22). T-DM1, total trastuzumab, DM1, MCC-DM1, and Lys-MCC-DM1 were measured using validated assays. Adverse events
were graded per NCI CTCAE, v4.03. All analyses are descriptive. The clinical cutoff date for this interim analysis was January 30,
2014.
Results
PK data were fully evaluable for 10 out of 10 patients each in the normal and mild cohorts and for 6 out of 7 patients in the
moderate cohort. Compared with the normal cohort, T-DM1 clearance at cycle 1 was 1.9- and 3.3-fold faster in the mild and
moderate cohorts, respectively. The trend of faster clearance was less apparent for cycle 3 after repeated dosing, with similar
T-DM1 exposures across the 3 cohorts. Plasma concentrations of DM1 and DM1-containing catabolites were largely comparable
across the 3 cohorts. No new safety signals were seen relative to the known safety profile of T-DM1. Updated safety data will be
presented.
Conclusions
There is a trend for faster clearance of T-DM1 at cycle 1 in patients with mild and moderate hepatic impairment vs. those with
normal hepatic function, which can be partly explained by demographic and pathophysiological covariates such as tumor burden,
albumin, and body weight. The studys small sample size could also partly explain the variability. Work to better understand the
mechanisms for the observed differences in clearance is ongoing. No increase in the systemic concentration of DM1 was
observed in patients with mild or moderate hepatic impairment vs. those with normal hepatic function. No additional safety
concerns were observed.

Title: Initial therapy among patients newly diagnosed with operable early stage human epidermal growth factor receptor
2-overexpressed (HER2+) breast cancer in the US: A real-world retrospective study
Stacey DaCosta Byfield2, Philip O Buck1, Cori Blauer-Peterson2 and Sara A Poston1. 1GSK, Philadelphia, PA and 2Optum, Eden
Prairie, MN.
Body: BACKGROUND: Prognosis and appropriate treatment of breast cancer patients (pts) is influenced by tumor molecular
characteristics. However, few existing retrospective studies have investigated the treatment patterns and outcomes of breast
cancer pts by tumor biomarkers. The objective of this study was to assess the current real-world treatment patterns associated
with resected non-metastatic HER2+ breast cancer in the US.
METHODS: This was a retrospective study of physician-reported clinical information (including date and stage at diagnosis,
HER2 status and hormone receptor (HR) status) for commercially insured breast cancer pts from the Oncology Management
Registry linked with medical and pharmacy claims from a large, national health plan in the US from 01/2008 to 8/2013. The date
of initial diagnosis was the index date. The inclusion criteria were: adult pts (18 years old), enrolled in the health plan for 6
months after the index date, diagnosed with HER2+ Stage I-III disease, known HR status, received breast cancer specific surgery
(mastectomy or lumpectomy) and anti-cancer systemic therapy (ACST) and/or radiation within 6 months of index date. Pts with
other primary cancers during the study period were excluded. The initial phase of care included initial therapy (surgery and ACST
and/or radiation) until 30 days after the last therapy received (surgery, ACST or radiation) prior to a 90-day gap in treatment.
Treatment patterns during the initial phase of care by HR status were examined.
RESULTS: Among 915 pts who met all study criteria, 662 (72%) and 253 (28%) were HR+ and HR-, respectively. Mean age was
52 years (standard deviation=9) and was not significantly different by HR cohort. Approximately 82% (n=749) were diagnosed
with Stage I/II disease. Most pts (80%, n=732) received adjuvant therapy only, 19% (n=177) received both neo-adjuvant and
adjuvant therapy, and <1% (n=6) were observed to have neo-adjuvant therapy only. Among pts who received neo-adjuvant
therapy, mean time from diagnosis to ACST was 21 days (median=21). Among pts who received only adjuvant therapy, mean
time from diagnosis to initial breast cancer specific surgery was 24 days (median=20). Overall, 72% of pts received HER2
targeted therapy (69% HR+, 80% HR-; p<0.01) during their initial phase of care. During neo-adjuvant therapy, 72% of pts
received trastuzumab (67% HR+, 81% HR-; p<0.05). During adjuvant therapy, 72% of pts received trastuzumab (69% HR+, 81%
HR-; p<0.05). The most common neo-adjuvant regimen regardless of HR status was carboplatin+docetaxel+trastuzumab (>40%
pts). The most common regimens during the adjuvant therapy period were carboplatin+docetaxel+trastuzumab with or without
hormone therapy (30% of pts).
CONCLUSION: In this real-world population of commercially insured breast cancer pts treated for operable, early stage HER2+
disease in the US, 28% of pts did not receive targeted therapy. More pts with HR- status received targeted therapy than those
with HR+ status. Further studies are warranted to examine whether pts that have not received targeted therapy are eligible and
would benefit from an HER2 targeted approach.

Title: Advanced HER2 positive breast cancer treated with trastuzumab: Is combination with chemotherapy always needed?
Randomized phase III trial SAKK 22/99
Olivia Pagani1, Dirk Klingbiel2, Thomas Ruhstaller3, Franco Nol4, Serenella Eppenberger5, Christian Oehlschlegel3, Jrg
Bernhard6, Peter Brauchli2, Dagmar Hess3, Christoph Mamot7, Elisabetta Munzone4, Bernhard Pestalozzi8, Manuela Rabaglio9,
Karin Ribi6, Christoph Rochlitz5, Karin Rothgiesser2, Beat Thrlimann3, Roger von Moos10, Khalil Zaman11 and Aron Goldhirsch4.
1
Institute of Oncology of Southern Switzerland, Bellinzona, Ticino, Switzerland; 2Swiss Group for Clinical Cancer Research
Coordinating Center, Bern, Switzerland; 3Kantonsspital, San Gallen, Switzerland; 4European Institute of Oncology, Milan, Italy;
5
University Hospital, Basel, Switzerland; 6International Breast Cancer Study Group, Bern, Switzerland; 7Kantonsspital, Aarau,
Switzerland; 8University Hospital, Zurich, Switzerland; 9Inselspital, Bern, Switzerland; 10Kantonsspital, Chur, Switzerland and
11
University Hospital, Lausanne, Switzerland.
Body: <Background
In advanced HER2+ breast cancer the impact of combining Trastuzumab (T) and chemotherapy (chemo) versus T alone followed
by the addition of chemo at disease progression has not been properly studied.
Study design
The trial compared efficacy, toxicity and quality of life of sequential administration of T followed, at progression, by combination
with chemo (T>TChemo) versus the upfront combination of T and chemo (TChemo) in patients with HER2+ advanced breast
cancer.
Materials and methods
Eligibility: measurable/evaluable HER2+ advanced disease; 2 previous chemo; ECOG performance status <=1. Stratification:
degree of HER2 overexpression, estrogen (ER) receptor status, 1st-line vs 2nd/3rd-line therapy, previous anthracyclines,
institution. Primary endpoint: time to progression on combined TChemo (TTP-TChemo). Secondary endpoints: response rate,
time to 1st progression and to treatment failure, overall survival, toxicity. Substudies: quality of life, predictive value of serum
HER2 levels, association of HER2 immunoprofiles with outcome.
The estimated median TTP-TChemo in the control arm (TChemo) was 5-6 months. A 3 months increase in the experimental
T>TChemo arm was considered meaningful.
The chemo backbone was at investigators choice (taxanes, vinorelbine, cisplatin) and could be stopped after 6 cycles in
responding patients. T was continued until progression. Treatment after progression under TChemo was by investigators
decision.
Patients characteristics
From Sept 1999Jan 2013, 175 patients were enrolled. The trial was stopped prematurely due to insufficient accrual. Baseline
characteristics were well balanced between arms: median age 55 years (3279), ER and/or progesterone receptor positive 63%,
2 disease sites 91%, dominant bone 36% or dominant visceral disease 66%, 1stline therapy 72%.
Results
At the cutoff date (May 2014) 173 patients were evaluable: median follow up 77.7 months, 29 patients (17%) censored when
receiving, at chemo stop, off-protocol treatment before progression (maintenance metronomic chemotherapy or endocrine
therapy), 11 patients (6%) had no event at the end of follow-up.
TTP-TChemo was longer than expected in both arms (12.7 months T>TChemo, 10.3 months TChemo) and not significantly
different (HR=0.7; 95% CI, 0.51.0; p=0.08). In the T>TChemo arm, median TTP before introduction of chemo was 3.7 months
(95% CI 2.35.1). Overall survival was not significantly different, 35.6 months versus 36.3 months (HR=0.9; 95% CI, 0.61.3;
p=0.50). Toxicity was mainly chemo related, consistent with the chosen regimen. Cardiac toxicity was mild (no grade 4, 1 cardiac
failure NYHA III in the T>TChemo arm). No treatment-related death was reported.
Conclusions
The sequential administration of T and chemo showed a non-significant trend to longer TTP-TChemo compared to upfront
combination therapy: it allows to delay chemo use and its toxicity and seems a reasonable approach. TTP-TChemo was better
than projected in both arms.
The sequential strategy with double anti-HER2 targeting (T/Pertuzumab) is now under evaluation in 1st-line patients in the SAKK
22/10 trial.

Title: Results from a randomized phase II study of the Dutch Breast Cancer Research Group (BOOG 2008-03): Concomitant
trastuzumab, bevacizumab and paclitaxel (HAT) versus trastuzumab and bevacizumab, followed by trastuzumab, bevacizumab
and paclitaxel (HA-HAT) at progression as first-line treatment for patients with Her2 positive metastatic breast cancer: The
HAT-Study
Jan C Drooger1, Harm van Tinteren2, Steffen M de Groot3, Albert J ten Tije4, Hiltje de Graaf5, Johanneke EA Portielje6, Agnes
Jager1, Aafke H Honkoop7, Sabine C Linn2, Judith R Kroep8, Frans LG Erdkamp9, Paul Hamberg10, Joan B Heijns4, A Elise van
Leeuwen-Stok11 and Stefan Sleijfer1. 1Erasmus MC Cancer Institute and Cancer Genomics Netherlands, Rotterdam, Netherlands;
2
Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands; 3Comprehensive Cancer Centre the Netherlands,
Amsterdam, Noord-Holland, Netherlands; 4Amphia Hospital, Breda, Netherlands; 5Medisch Centrum Leeuwarden, Leeuwarden,
Netherlands; 6Haga Hospital, Hague, Netherlands; 7Isala Klinieken, Zwolle, Netherlands; 8Leiden University Medical Center,
Leiden, Netherlands; 9Orbis Medisch Centrum, Sittard, Netherlands; 10Sint Franciscus Gasthuis, Rotterdam, Netherlands and
11
Dutch Breast Cancer Research Group, Amsterdam, Noord-Holland, Netherlands.
Body: Background
Until recently, trastuzumab (H) combined with a taxane was standard first line systemic therapy for patients with Her2 positive
metastatic breast cancer (MBC) who are not candidates for endocrine therapy. Clinical studies have shown that bevacizumab (A)
enhances the activity of weekly paclitaxel (T), while preclinical data suggest that A can also augment anti-tumor activity of H. Here
we report outcome data of a randomized phase II study in Her2 positive MBC on the combination of HAT versus sequential
treatment starting with a chemotherapy-free approach with H and A (HA), followed by adding T at progression (HA-HAT).
Methods
Patients with Her2 positive MBC, eligible for first-line systemic therapy, were randomized (1:1) between HAT (3-weekly H 6 mg/kg
(first dose 8/mg/kg) plus A 15 mg/kg both until progression and T 90mg/m2, day 1,8,15 every 4 weeks for a maximum of 6 cycles)
and HA-HAT (doses HA were the same as in HAT with T for a maximum of 6 cycles added to HA at progression). Primary
endpoint was progression-free survival rate at 1 year after randomization (PFR-1yr). A regimen yielding a PFR-1yr of around 40%
was decided to warrant further exploration (p1). A Fleming one-stage design was, therefore, applied to both arms with p2=20%,
alpha=0.05 and beta=0.10. Secondary endpoints included PFS, defined as the time from randomization to documented disease
progression (PD) or death from any cause after patients had received HAT. In addition for the HA-HAT group, a PFS1 and PFS2
was established. PFS1 was defined as the time from randomization to PD or death from any cause, PFS2 as the time from
starting treatment with HAT to PD or death from any cause.
Results
Between April 2009 and September 2013, 84 patients were randomized, 39 to HAT and 45 to HA-HAT. Baseline characteristics
were similar for both arms: mean age 55 years (range 29-80 years), prior adjuvant chemotherapy in 35% and hormone receptor
positive disease in 63%. The primary endpoint was met in both arms. At a median follow-up of 22 months, 26 patients in the HAT
arm had progressed of whom 14 had died. The median PFS for the HAT arm was 19.4 months (95% confidence interval (CI)
14.6-25.2). In the HA-HAT arm 23 patients progressed under HA of whom 3 still have no 2nd progression after adding T. Eleven
patients of the HA-HAT arm died of whom 2 due to toxicity (1 sigmoid perforation, 1 pneumonia). The median PFS1 was 14.9
months (95% CI 10.9-NA) The median PFS2 was 8.2 months (95% CI 5.9-NA). Overall PFS in the HA-HAT arm was 22.6 months
(95% CI 15.4-NA). Grade 3 or higher toxicity was 3 times more often seen during treatment including T. More detailed information
regarding toxicity will be shown at the meeting.
Conclusion
Both HAT and HA-HAT are active regimens in Her2 positive MBC. In particular the sequential approach with starting monoclonal
antibodies only, yielding a median PFS of more than 1 year and a relatively mild toxicity profile, and adding chemotherapy at
progression is worthwhile to be further explored.
Financial support Roche Netherlands.

Title: CDK8 inhibition potentiates anti-ER and anti-HER2 therapies in breast cancer
Martina S McDermott1, Chang-uk Lim1, Mengqian Chen1, Alexander Chumanevich1, James F Catroppo2, Balazs Gyorffy3, David
Oliver1, Igor B Roninson1 and Eugenia V Broude1. 1South Carolina College of Pharmacy, University of South Carolina, Columbia,
SC; 2Microbiology and Immunology, University of South Carolina School of Medicine, Columbia, SC and 3Research Laboratory for
Pediatrics and Nephrology, Hungarian Academy of Sciences, Budapest, Hungary.
Body: CDK8, along with its paralog CDK19, is a cyclin dependent kinase which, in contrast to other members of the CDK family
does not regulate cell cycle progression. CDK8 acts as a pleiotropic transcription regulator potentiating the induction of
transcription by several transcription factors. Immunohistochemical staining of breast tissue arrays and bioinformatics analysis of
gene expression microarray data of breast cancer patients revealed that CDK8 is overexpressed in breast cancer and that higher
CDK8 expression correlates with the failure of systemic therapy. Small-molecule selective inhibitors of CDK8 and CDK19
(Senexin A and Senexin B) inhibited the mitogenic effects of estrogen and estrogen-dependent transcription in estrogen receptor
(ER)+ breast cancer cell lines. CDK8/19 inhibitors had a cytostatic effect on different ER+ cell lines, and this growth inhibition was
synergistic with the effect of the anti-estrogen fulvestrant, particularly in ER+ cell lines resistant to estrogen deprivation. Some of
the ER+ cell lines sensitive to CDK8/19 inhibition also express HER2, and therefore we tested CDK8/19 inhibitors in combination
with the HER2 and EGFR tyrosine kinase inhibitor lapatinib and an anti-HER2 monoclonal antibody, a biosimilar of trastuzumab.
CDK8/19 inhibition produced a synergistic decrease in cell growth with both HER2 inhibitors; this effect was especially
pronounced with a trastuzumab biosimilar. Surprisingly, the synergistic effect with HER2 inhibitors was observed in both ER+
HER2+ and ER-HER2+ cell lines, suggesting an effect on a HER2-complementing molecular target other than ER. Interestingly,
CDK8/19 inhibition also synergized with trastuzumab biosimilar in a breast cancer cell line that exhibits innate resistance to
trastuzumab, suggesting that CDK8/19 inhibition can overcome trastuzumab resistance in breast cancer. These results suggest
that combining anti-estrogen and anti-CDK8 therapy may be more effective than conventional hormone therapy for ER positive
breast cancer and that combining anti-HER2 and anti-CDK8 therapy is a rational potential treatment for HER2+ breast cancer,
regardless of ER status or sensitivity to trastuzumab.

Title: Preclinical data of SYD985 support the clinical investigation of this novel anti-HER2 antibody-drug conjugate in breast
cancer patients with low levels of HER2 expression
Gijs Verheijden1, Patrick Beusker1, Ruud Ubink1, Miranda van der Lee1, Patrick Groothuis1, Peter Goedings1, David Egging1,
Marco Timmers1 and Wim Dokter1. 1Synthon Biopharmaceuticals BV, Nijmegen, Netherlands.
Body: SYD985 is a novel anti-HER2 antibody-drug conjugate (ADC) in development for breast cancer. The ADC consists of three
parts: i) the monoclonal antibody trastuzumab, ii) a linker that can be cleaved by tumor-resident proteases at the dipeptide
valine-citrulline (vc) motif, and iii) the prodrug seco-DUocarmycin-hydroxyBenzamide-Azaindole. The average drug to antibody
ratio (DAR) of this ADC, also named trastuzumab vc-seco-DUBA, is 2.8. After antibody-mediated binding of SYD985 to its
molecular target HER2 on the surface of tumor cells, the ADC is internalized. Subsequently, the active duocarmycin is released
and binds to DNA in the minor groove, followed by alkylation of the DNA and killing of the tumor cells.
In vitro and in vivo experiments were performed to (directly) compare various pharmacodynamic and pharmacokinetic properties
of SYD985 with those of the most progressed (marketing-approved) HER2-targeting ADC, i.e. ado-trastuzumab emtansine
(T-DM1). T-DM1 is currently indicated for second line treatment of breast cancer patients overexpressing HER2 tumor tissue.
In in vitro experiments, SYD985 shows potencies similar to T-DM1 in cell lines expressing HER2 at high levels (IHC HER2 3+),
whereas SYD985 is significantly more potent compared to T-DM1 in cell lines expressing low HER2 levels (IHC HER2 2+ or 1+).
In line with these results, SYD985 has superior efficacy compared to T-DM1 in xenograft models in which patient-derived breast
cancer tumor tissue with low HER2 levels (IHC HER2 2+ or 1+/FISH-) have been used.
Pharmacokinetic experiments (in vitro and in vivo) have revealed that whereas in mice SYD985 has limited stability, in
cynomolgus monkey SYD985 is very stable. Excellent stability in vitro is also observed in human plasma. Moreover, after i.v.
administration of SYD985 in the cynomolgus monkey the amount of free toxin (DUBA) detected in the monkey plasma is
many-fold lower than levels reported for the toxin (DM1) released from T-DM1.
Resuming, the preclinical data of SYD985 support clinical studies to investigate whether SYD985 has benefit in cancer patients
with moderate or even low HER2 levels of the tumor tissue.

Title: No Show

Title: No Show

Title: A novel targeted engineered toxin body for treatment of HER2 positive breast cancer
Erin K Willert1, Sangeetha Rajagopalan1, Garrett L Robinson1, Brigitte Brieschke1, Jennifer Erdman1, William Null1 and Jack P
Higgins1. 1Molecular Templates, Georgetown, TX.
Body: The HER2 receptor is overexpressed in 20-30% of breast cancers. HER2 overexpression in breast cancer has been
successfully targeted by both antibody and antibody-drug conjugate (ADC) approaches. The recently approved
trastuzumab-emtansine (T-DM1, Kadcyla) ADC has shown promising clinical results in patients refractory to trastuzumab
(Herceptin). Because DM1 is a chemotherapeutic and refractory breast cancer patients have typically progressed through several
rounds of chemotherapy, we are developing a new targeted therapy approach utilizing a novel mechanism of action. MT-2H74 is
an engineered toxin body comprised of the trastuzumab single chain variable fragment (scFv) and a modified
ribosome-inactivating protein derived from Shiga-like toxin 1 A (SLT-1A). We have proprietarily modified SLT-1A for reduced
immunogenic potential and increased stability. MT-2H74 selectively binds HER2-expressing breast cancer cells, and exerts a
potent HER2-specific cytotoxic effect on several tested cell lines. MT-2H74 demonstrates effective cell kill at picomolar
concentrations on cell lines expressing high levels of HER2 and is not cytotoxic to HER2 negative control cell lines. Multi-drug
resistance transporter 1 (MDR1) has been reported to confer resistance to chemotherapeutic and ADC treatments, such as direct
T-DM1 cell kill; however, MT-2H74 displays cytotoxic activity against HER2 transfected NCI/ADR-RES cells known to express
MDR1. Murine models are under investigation to determine therapeutic efficacy in vivo. MT-2H74 is a promising HER2-targeted
therapeutic agent against breast carcinomas with a unique mechanism of action and is currently under further development.


Title: A phase 1b study of ONT-380, an oral HER2-specific inhibitor, combined with capecitabine and/or trastuzumab, in HER2+
Erika Hamilton1,2, Denise A Yardley1,2, Gabriel Hortobagyi3, Luke Walker4, Virginia F Borges5 and Stacy Moulder3. 1Sarah Cannon
Research Institute, Nashville, TN; 2Tennessee Oncology, PLLC, Nashville, TN; 3MD Anderson Cancer Center, Houston, TX;
4
Oncothyreon Inc, Seattle, WA and 5University of Colorado Cancer Center, Aurora, CO.
Body: Background: ONT-380 (also known as ARRY-380) is a potent, selective small molecule inhibitor of HER2 with 500-fold
selectivity compared to EGFR. Preclinical studies have demonstrated synergistic activity with ONT-380 plus chemotherapy or
trastuzumab, as well as superior activity compared to lapatinib and neratinib in models of HER2+ CNS metastases. In a Phase 1
single agent study in HER2+ MBC, ONT-380 was well tolerated and provided clinical benefit with minimal EGFR-type toxicities,
with an MTD of 600 mg BID using an API-in-capsule formulation. Based on the potential for dual blockade of HER2 plus a
cytotoxic agent to lead to clinical benefit, ONT-380 is being evaluated in combination with capecitabine and/or trastuzumab in pts
with HER2+ MBC previously treated with trastuzumab and ado-trastuzumab emtansine (T-DM1).
Methods: This parallel 3+3 dose escalation study administers ONT-380 in a new tablet formulation with either capecitabine (1000
mg/mg2 PO BID for 14 days of a 21 day cycle) (Combo 1), trastuzumab (8 mg/kg IV loading dose; then 6 mg/kg IV once every 21
days) (Combo 2), or with both capecitabine and trastuzumab (Combo 3). Prior treatment with trastuzumab and T-DM1 for
metastatic disease is required. Prior therapy with lapatinib or neratinib and asymptomatic brain metastases (treated or untreated)
are allowed. Study assessments include safety, ONT-380 and capecitabine PK, tumor response by RECIST 1.1, and CNS
response by both modified RECIST and volumetric criteria. Dose escalation will be followed by enrollment of expansion cohorts in
patients with and without CNS disease.
Results: As of 21 May 2014, 8 pts have been treated with ONT-380 at 300 mg BID in Combo 1 (n=4) for 27 cycles and Combo 2
(n=4) for 12 cycles. All pts are still active. No dose limiting toxicities (DLT) have been seen for either combination, and most
toxicities have been Grade 1 or 2. The most common toxicities regardless of attribution have been nausea, diarrhea, dyspepsia,
headache, and palmar-plantar erythrodysaesthesia (PPE). There has been no Grade 3 diarrhea. Two Grade 3 events have been
reported, both in Combo 1: PPE and ALT/AST elevation. There have been no SAEs. ONT-380 was dose reduced in one pt
(Combo 1) for persistent Grade 2 diarrhea as well as in the pt with Grade 3 ALT/AST. Capecitabine was dose reduced in three
pts: diarrhea (n=2), Grade 3 PPE (n=1), and Grade 3 ALT/AST (n=1). Best response seen to date in 3 pts evaluable for response
has been stable disease. Initial PK data indicate greater drug exposure is achieved with the new tablet formulation of ONT-380
compared to the earlier capsule formulation with no evidence of drug interaction with capecitabine.
Summary: ONT-380 in combination with either capecitabine or trastuzumab has been associated with an acceptable safety
profile, with no DLTs seen with either combination. No Grade 3 diarrhea has been observed, including pts receiving both
ONT-380 and capecitabine. Early evidence of disease control has been seen. Dose escalation continues, and evaluation of the
three drug combinations is planned. Updated results will be presented for additional cohorts.

Title: Insulin-like growth factor receptor-1 (IGF-1R) expression is highly correlated with HER2 amplification on circulating
epithelial tumor cells (CETCs) in breast cancer - this may be the reason for resistance to trastuzumab
Dorothea Zimon1, Monika Pizon1, Ulrich Pachmann1 and Katharina Pachmann1. 1Transfusion Center, Bayreuth, Germany.
Body: Background: HER2 status has been the most extensively studied biomarker in circulating epithelial tumor cells of breast
cancer patients. HER2 status can change during recurrence of disease and HER2 overexpression is a mandatory requirement to
administer anti-HER2-directed drugs. Nevertheless, 70% of patients with HER2-positive breast cancers develop intrinsic or
secondary resistance to trastuzumab. This resistance has been associated with the activation of an alternative signalling pathway
such as the insulin-like growth factor (IGF) pathway. Therefore we investigated the expression of IGF-IR on the CETCs in addition
to HER2 amplification in breast cancer patients to identify patients who might benefit from a combined targeted therapy against
HER2 and IGF-IR.
Methods: CETCs were determined from blood of 30 breast cancer patients. The number of vital CETCs and the expression of
IGF-IR were investigated using the maintrac approach. Fluorescence in situ hybridisation was used for analysis of HER2
amplification in CETCs.
Results: CETCs could be detected in all breast cancer patients. The number of CETCs ranged from 4 to 163 in 100 l of cell
suspension. IGF-IR expression on the surface of CETCs was detected in all patients. Setting a cut-off 30% positive cells as HER2
positive in 75% of patients HER2 positive CETCs were observed irrespective of the status in the primary tumor. In contrast, only
6% of patients changed their HER2 status from positive tissue to negative CETCs. A statistically high correlation was found
between the percentage of IGF-IR positive and HER2 positive CETCs. A statistically significant association was found between
IGF-IR expression or HER2 amplification and ER/PR receptor status. The higher frequency of HER2 amplified circulating tumor
cells might be due to the difference in preparation steps between tissue and blood borne cells and one of the reasons for
trastuzumab resistance and for the response of some HER2 negative patients to trastuzumab. Re-evaluation of HER2 status in
CETCs could be a valid strategy with potential clinical applications.
Conclusion:
Our results demonstrate a parallel expression of IGF-IR and HER2 amplification in CETCs. IGF-IR may be involved in the
development of resistance to trastuzumab and may be an important potential therapeutic target in breast cancers with HER2
positive circulating tumor cells. Combining targeting of IGF-IR and HER2 may be a rational approach to improve response to
trastuzumab in the sub-group of CETCs that express both, HER2 and IGF-IR.

Title: Skin infections associated with the addition of pertuzumab to trastuzumab-containing chemotherapy
Joanne E Mortimer1, Yuan Yuan1, Daphne Stewart1, Samuel Chung1, Laura Kruper1, Louise C Wong1, Mary Mendelsohn1,
Carolyn Behrendt1, Sanjeet Dadwal1 and Bernard Tegtmeier1. 1City of Hope, Duarte, CA.
Body: Objectives: The addition of pertuzumab, to trastuzumab-based chemotherapy is currently considered first-line therapy for
locally advanced and metastatic disease HER2 positive breast cancer and has also been suggested for use in the adjuvant
setting. Over the past 12 months, we have observed an increase in the incidence of severe skin infections in patients receiving
chemotherapy with pertuzumab and trastuzumab. We report the natural history of what we believe is a previously unrecognized
toxicity of these regimens. Methods: Shortly after the FDA approval of pertuzumab, our clinical team appreciated an increase in
invasive skin infections. We discussed this concern and identified new cases at our weekly research meeting, keeping a log of
cases as they were identified. Infection control reviewed the individual patient and hospital data during this time period. Results:
Eleven women were identified to have severe skin and/or nail infections; 6 after cycle 1; 2 after cycle 2, 1 after cycle 3 and 2 after
cycle 6. The median age was 51 (Range 46-64); 9 received pertuzumab, trastuzumab, carboplatin, and docetaxel (PTCH) and 2
pertuzumab, trastuzumab, and docetaxel. Folliculitis of the scalp, abdomen, and/or buttocks were observed in 4 patients.
Abscesses were observed in 5 patients, 4 of whom required incision and drainage. Severe paronychial infections involving one to
16 digits were observed in 3 (including one who also had folliculitis). 1 pt required surgical removal of 2 nails. Quantitative
immunoglobulins were found to be low in 2 of 8 women tested; 1 patient had a total protein of 4.7 but did not have an assessment
of quantitative immunoglobulins. All patients were initially treated with oral antibiotics, but 3 required hospitalization. Cultures were
obtained on 6 patients, Staph aureus was identified in 2 and methicillin resistant Staph (MRSA) in 4. All patients resolved their
infections and 9 of 10 were able to complete six cycles of chemotherapy. Infection control could identify no increase in Staph
infections at our institution. Patients were treated at different locations and received different lot numbers of drug.
Conclusions: We believe this is the first report of a substantial incidence of invasive skin and nail infections with the addition of
pertuzumab to trastuzumab-based regimens not reported in the product label. Low levels of quantitative immunoglobulin in select
patients suggest a possible mechanism.

Title: The ASCO/CAP guideline update for HER2 testing increases the number of breast cancer patients eligible for
HER2-targeted therapy
Juliana Giacomazzi1,2, Carolina R Hartmann3, Diego D Paskulin2, Luis Fernando Rivero3, Maira Caleffi4, Alessandro C
Pasqualotto1,2, Marcia S Graudenz3 and Daniela D Rosa4. 1Universidade Federal de Ciencias da Saude de Porto Alegre
(UFCSPA), Porto Alegre, Rio Grande do Sul, Brazil; 2Irmandade da Santa Casa de Misericordia de Porto Alegre (ISCMPA), Porto
Alegre, Rio Grande do Sul, Brazil; 3Instituto de Patologia, Porto Alegre, Rio Grande do Sul, Brazil and 4Hospital Moinhos de
Vento, Porto Alegre, Rio Grande do Sul, Brazil.
Body: Background: The American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) published, in
Nov/2013, the interpretive guidelines for HER2 testing of breast cancer patients. This aimed to improve the accuracy of HER2
testing and its utility as a predictive marker in invasive breast cancer. The former version of these criteria was written in 2007.
Objectives: to compare the HER2 immunohistochemical (IHC) analysis using the 2007 versus 2013 algorithms in a cohort of
breast cancer cases diagnosed in a single institution in Southern Brazil. The cases were previously classified as HER2 1+ or 2+,
using the 2007 criteria. Methods: the sample included 100 invasive breast cancer cases. The HercepTest (Dako, Denmark) was
used for determination of HER2 expression. The HER2 testing was analyzed independently by two pathologists. The FISH
analysis was done using a HER2/D17Z1 probe set. Preliminary Results: The HER2 IHC interpretation changed in 11/69 (15.9%)
cases: 8.7% negative or equivocal cases by the 2007 guidelines were positive by the 2013 classification and 7.2% of HER2 1+
cases became equivocal (p<0.001; x2 test). The FISH analyses are ongoing. Conclusion: The 2013 ASCO/CAP guidelines
resulted in less negative cases and in more equivocal (requiring reflex testing) and positive tests. Applying the ASCO/CAP 2013
guidelines resulted in a significantly increase of breast cancer patients eligible for HER2-targeted therapies.

Title: Validation of a software based clinical decision support system for breast cancer treatment in a tertiary care cancer center
in India
Nita Nair1, Sudeep Gupta1, Naresh Ramarajan2, Gitika Srivastava2, Vani Parmar1, Anusheel Munshi3, Shraddha Vanmali1,
Vaibhav Vanmali1, Rohini Hawaldar1 and Rajendra A Badwe1. 1Tata Memorial Centre, Mumbai, Maharashtra, India; 2Navya
Network, Cambridge, MA and 3Fortis Memorial Research Hospital, Gurgaon, Haryana, India.
Body: Introduction: Access to expert, evidence based clinical decision making is crucial in maximizing the outcome of women
with breast cancer, but is a scarce resource, especially in developing countries. The Navya Expert System is a patented, software
based clinical decision support system that exhaustively searches and assimilates relevant medical literature and guidelines to
make specific therapeutic recommendations for individual patients based on their clinical data. This study is a retrospective
validation of Navya Expert System's output against tumor board decisions of a multidisciplinary group of expert breast cancer
clinicians working in a tertiary care oncology center in India.
Methods: Women with non-metastatic breast cancer who had already completed their loco-regional and systemic therapy based
on the recommendations of the tumor board were included in the study. The protocol specified clinical and pathology data of
these women were retrospectively abstracted from their case charts and processed through the Navya Expert System. The output
was classified into major (neo-adjuvant chemotherapy versus upfront surgery and need for adjuvant chemotherapy, endocrine
therapy and radiation therapy, respectively) and minor (breast conservation versus mastectomy, taxane versus non-taxane
adjuvant chemotherapy and need for nodal radiation therapy) therapeutic decisions. Decisions discordant between the tumor
board and the Navya Expert System were adjudicated by an expert panel of breast cancer clinicians from the same institution.
Navya Expert System decisions were classified as discordant with appropriate clinical practice if they were in disagreement with
both the tumor board and expert panel. All other Navya Expert System decisions were classified as concordant. The primary
outcome of the study was concordance between the Navya Expert System and the tumor board or expert panel for major and
minor therapeutic decisions.
Results: A total of 76 patients involving 224 major and 224 minor therapeutic decisions were included in the study. Navya Expert
System's output was concordant with the tumor board or expert review in 224/224 major decisions (100%, 95% CI 99.6%-100%)
and 221/224 minor decisions (98.6%, 95% CI 97.1%-100%). Navya Expert System's output was concordant with the tumor board
alone in 210/224 (93.75%, 95% CI 90.6%-96.9%) major decisions and 160/224 (71.4%, 95% CI 65.5%-77.3%) minor decisions.
Most common reasons for discordance were non-prescription of HER2 targeted therapy by the tumor board due to financial
constraints and non-use of nodal radiation for 1-3 node positive patients. Of the 64/224 Navya Expert System decisions
discordant with the tumor board, only 3 were finally deemed discordant after review by the expert panel.
Conclusions: Navya Expert System treatment recommendations, only requiring the input of commonly available clinical data, are
highly concordant with those of a tumor board comprised of breast cancer experts with high level expertise. If these results can be
prospectively validated, Navya Expert System has the potential to increase global access to evidence based clinical decision
making in breast cancer.

Title: Attitudes of medical oncologists towards research breast biopsies in patients with newly diagnosed stage I-III breast cancer
not enrolled in a clinical trial
Davinia SE Seah1, Sarah Scott2, Hao Guo1, Julie Najita1, Ruth Lederman1, Elizabeth Frank1, Jessica Sohl1, Zsofia Stadler5, Stuart
G Silverman3, Jeffrey Peppercorn4, Eric P Winer1, Steve E Come2 and Nancy U Lin1. 1Dana-Farber Cancer Institute, Boston, MA;
2
Beth Israel Deaconess Medical Center, Boston, MA; 3Brigham and Women's Hospital, Boston, MA; 4Duke University Medical
Center, Durham, NC and 5Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background:
Patients (pts) with breast cancer treated with neo-adjuvant therapy on clinical trials are often asked to consent to pre-treatment
and on-treatment research biopsies. There is increasing interest in obtaining tissue samples at similar time points in pts treated
with neo-adjuvant therapy outside of clinical trials. However, medical oncologists (MOs) attitudes towards approaching pts about
research biopsies in this setting are unknown.
Methods:
Three hundred and nine academic breast MOs identified from websites of the National Cancer Institute (NCI) designated cancer
centers were asked to complete a survey either by paper or online. Eligible MOs (MOs who saw breast cancer pts and who saw
pts >4 hours/week.) were asked to predict what proportion of their pts with newly diagnosed, non-metastatic breast cancer would
consent to research purposes only biopsies (RPOBs) i.e., biopsies with no clinical benefit to pt. Median values are reported.
Two-sided Fishers exact test was used to compare categorical variables using a level of .05.
Results:
Of 221 (101F,85M, 5 unknown) MOs who completed the survey, 30 MOs were ineligible (response rate=221/309,72%). Median
age was 50 (Range 33-80). Median years of oncology experience was 15 (Range 1-45). MOs predicted that 14%, 63% and 21%
of their pts would definitely/probably, maybe, probably not/definitely not consent to a RPOB of the breast.
Forty-one percent, 34%, 19%, 3% of MOs were very comfortable, somewhat comfortable, somewhat uncomfortable, and very
uncomfortable asking pts to consent to RPOBs respectively. The only factor associated with increased comfort discussing an
RPOB was MOs years in practice. MOs with fewer years (<15 years) in practice were more comfortable in asking pts to
participate in RPOBs compared to MOs with more years in practice (>15 years) (Adjusted RR=1.2, p =0.02). Gender, number of
pts enrolled onto clinical trials, and MOs with pts who had research biopsies in the last 3 months was not associated with
increased comfort.
MOs who were more comfortable in approaching pts for RPOBs were associated with estimating a larger proportion of their pts as
willing to undergo RPOB. For example, nearly one third of MOs who were very comfortable with approaching pts for RPOBs
estimated that greater than 50% of their pts would consent to research biopsies. In contrast, nearly all the MOs who were very
uncomfortable with approaching pts for RPOBs estimated that less than 25% of their pts would consent to research biopsies.
The 3 most common reasons why MOs were reluctant to consent pts for a RPOB include pain/discomfort of a biopsy (59%), risk
of a biopsy procedure complication, (44%), and inconvenience to the pt (33%).
Conclusions:
Academic breast MOs predicted that fewer than 1 in 5 women with newly diagnosed, non-metastatic breast cancer would
definitely or probably agree to a request for an RPOB outside of the context of a therapeutic trial, and approximately one-quarter
of MOs expressed discomfort in approaching pts for such procedures. Our results have important implications regarding the
feasibility of such research efforts, and identify potential barriers to target for intervention.

Title: Do participants in adjuvant breast cancer trials reflect the breast cancer patient population?
Alastair M Thompson1,2, Ruth Dryden2, Colin McCowan2, Alison Harrow2 and Shaun Treweek2. 1MD Anderson Cancer Center,
Houston, TX and 2University of Dundee, Dundee, United Kingdom.
Body: Background: To be clinically relevant, randomised clinical trials need to involve participants who reflect those in the
community likely to receive the treatments under evaluation. However, as few as 5% of patients with chronic medical illnesses
(asthma, diabetes) are documented as meeting therapeutic trial entry criteria. This study examined a population based cohort of
women diagnosed with early breast cancer to compare the eligibility criteria of 12 influential adjuvant endocrine therapy,
chemotherapy or radiotherapy trials with clinical eligibility and actual treatment at a single regional cancer centre.
Methods: 421 Phase III trials of adjuvant endocrine therapy, chemotherapy or radiotherapy treatment for early breast cancer
referenced in 5 recent national guidelines were identified and shortlisted by breast cancer specialists of multiple disciplines at a
national meeting to identify the twelve most influential trials: 5 endocrine (ATAC, BIG 1-98, IES, MA17, TEAM), 2 chemotherapy
(NSABP B-28, TACT) and 5 radiotherapy trials (Whelan et al JNCI 2002, EORTC Bartelink NEJM 2001, START A, START B,
TARGIT-A). Eligibility criteria were extracted from protocols and subsequently applied to a 16-year (1993 2008)
community-based cohort of 4811 women who had received a diagnosis of breast cancer, with the proportion who met each
criterion calculated. HER2 therapy trials were excluded since the cohort did not have HER2 testing prior to 2000. Clinical
guideline-based criteria for the adjuvant treatments described within each trial were also applied to the cohort. Finally, the
proportion of women actually prescribed adjuvant endocrine therapy meeting the relevant trial eligibility criteria, taking account of
drug availability, was calculated using linkage to community pharmacy records.
Results: Of 4811 women in the cohort, 3535 (73%) were eligible for at least one trial. Eligibility ranged from 34% (1653 women)
for the chemotherapy trials, through 47% (2247) for at least one endocrine therapy trial, to 71% (3419) for a radiotherapy trial.
The proportion of women considered clinically eligible for the same treatment, was greater (MA17, Whelan et al JNCI 2002,
TACT), similar (BIG1-98, IES, NSABP B-28) or less than (ATAC, EORTC Bartelink NEJM 2001, START A and B trials, TARGIT
A) trial eligibility criteria, with no consistent pattern of exclusion. It was rare for more than 45% of women of the community cohort
to be eligible for a given trial. The proportion of women who actually received an endocrine therapy in clinical practice, but would
not have been eligible for the trial evaluating that therapy (ATAC, BIG 1-98, IES and MA17), ranged from 44% to 83%.
Interpretation: Trial entry criteria for early breast cancer are reassuringly more inclusive for adjuvant breast cancer treatments
than for chronic medical conditions. However, the application of trial evidence in the clinic may be at variance with the trials and,
for adjuvant endocrine therapy at least, a substantial proportion of women receive therapy who would not have been eligible for
the relevant trial.

Title: Relationship between type of therapeutic intervention and funding source in randomized clinical trials (RCTs) in breast
cancer
Artur Katz1, Fernando Santini1, Fabio Y Moraes1, Andrea K Shimada1, Caroline Chaul1, Manuel C Abraao1 and Everardo D Saad2.
1
Hospital Sirio Libanes, So Paulo, Brazil and 2Dendrix Research, Brazil.
Body: Background: Pharmaceutical companies play an important role in new drug development and approval in an environment
where performing RCTs has become increasingly costly and complex. However, myriad questions about the impact of clinical,
surgical and radiotherapy interventions equally require adequate hypothesis testing in RCTs. A potential lack of funding for RCTs
not directly related to new drug development might lead to important gaps in clinical knowledge.
Methods: We searched PubMed for all RCTs published between 01/2009 and 12/2013 in breast cancer. All articles published in
this period were manually screened for eligibility. We included only RCTs with clinical endpoints such as TTP, PFS, OS and
response rate. Two investigators independently selected phase 2 and phase 3 RCTs with at least 50 patients published in
English. We also searched National Cancer Institute (NCI) data for all active, therapeutic phase 3 trials that are currently enrolling
patients (as of 05/05/2014). We classified eligible trials according to the type of intervention (drugs, radiotherapy or surgery) and
the stated funding source (industry versus nonprofit).
Results: We retrieved 1,676 PubMed studies of which 247 (15%) were eligible. 218 (88%) of the RCTs evaluated drugs, 14 (6%)
radiotherapy and 15 (6%) surgery. 183/247 (74%) RCTs were funded entirely or partially by industry (pharmaceutical and device
companies) and 64/247 (26%) by nonprofit organizations (government, academic centers or foundations). There was a significant
association between source of funding and type of intervention: 183/218 RCTs (83%) evaluating drugs were funded by industry,
in comparison to none of surgical and radiotherapy RCTs (P<0.0001). From the NCI data we retrieved 144 RCTs. 116 (81%) of
the RCTs studies drugs, 20 (14%) radiotherapy and 7 (5%) surgery. 55/116 (47%) trials evaluating drugs were funded by industry
in comparison to none of surgical and radiotherapy trials. Eighty eight trials were not funded by industry. Of those, 23(26%) were
performed in China, 16(18%) in the USA, 10 (11%) in France, 7 (8%) in India, 6 (7%) in the UK and the remaining 26 studies
were performed in 16 other countries
Conclusions: The vast majority of RCTs in oncology relates to drug development and is being funded by industry, while 100% of
RCTs evaluating surgical or radiotherapy related questions are not industry funded. This scenario seems not be changing over
the last years given that the comparison between published trials and active trials shows the same scenario. Even though new
drug development is of paramount importance, the extent to which clinically relevant issues are not being properly addressed by
RCTs, at least in part due to lack of funding, should be considered and further evaluated. The oncologic community, as well as
academic and nonprofit organizations, including governments, need to work together to forge new and alternative forms of
research funding in order to allow us to answer critical clinical questions that we, as oncologists, face in our daily practice.

Title: IBCSG BIG 1-98 study: The long-term follow-up experience
Anita Giobbie-Hurder1, Beat Thrlimann1, Bent Ejlertsen1, Patrick Neven1, Robert E Coleman1, Ian Smith1, Andrew M Wardley1,
Istvn Lng1, Marco Colleoni1, Marc Debled1, John F Forbes1, Karen N Price1, Meredith M Regan1, Manuela Rabaglio1, Aron
Goldhirsch1, Alan S Coates1 and Richard D Gelber1. 1BIG 1-98 Collaborative Group and International Breast Cancer Study Group.
Body: Background
Industry-sponsored clinical trials often have duration of patient follow-up that is defined according to regulatory requirements.
However, in diseases such as endocrine-responsive, early breast cancer, recurrences occur after protocol follow-up, and
monitoring of long-term toxicity is important. It is challenging to continue patient follow-up after industry sponsorship ends.
Transferring responsibility for additional follow-up to the participating academic centers is required. One such example is the
long-term follow-up (LTFU) of patients in the Breast International Group (BIG) 1-98 Trial. We present the procedures and current
status of the BIG 1-98 LTFU protocol.
Methods
In 2010, the BIG 1-98 trial embarked on a new LTFU protocol to gather data on patient outcomes for an additional five years after
study completion (2011-2015). Industry sponsorship ceased at the end of 2010. The LTFU study is designed as an observational,
non-interventional study to continue the collection of simplified and updated data on survival, disease status, and long-term
adverse events from centers participating in the 4-arm option. The International Breast Cancer Study Group (IBCSG) is
sponsoring BIG 1-98 LTFU, and per case reimbursement is available.
Results
The potential BIG 1-98 LTFU cohort consists of the 148 academic medical centers that participated in the 4-arm option with a
maximum of 6843 patients enrolled to the parent study. In May 2014, approximately 3 years after initiation of the LTFU protocol,
96 centers had agreed to participate, of which 67 sites had activated the protocol and submitted LTFU data; 31 additional centers
were not participating, and the status of 21 centers was unknown.
Participation Status
Number of Centers
Patients Enrolled in BIG 1-98
Closed
17
Not Participating
28
643
No response/Unknown
21
850
Yes, participating
96
5333
Activated
67
4215
Not Activated
29
1118
Totals
148
6843
Because the original BIG 1-98 informed consent indicated life-long follow-up, only three countries required patient re-consent in
order to participate. At least one LTFU data submission has occurred for 73% of patients participating in the LTFU (May 2014).
Conclusion
Long-term follow-up for a large-scale clinical trial is feasible, but challenging. The methods used for BIG 1-98 LTFU will be
described and the status will be updated at the meeting.

Title: Pretargeted immuno-PET with an anti-carcinoembryonic antigen (CEA) bispecific antibody (BsMAb) and a 68Ga-labeled
hapten-peptide compared to conventional imaging and FDG-PET in metastatic breast cancer patients (BC): First results
Mario Campone1,3, Aurore Rauscher2,3, Alain Faivre-Chauvet3,4, Thomas Carlier2,4, Ludovic Ferrer2,5, Pierre Baumgartner2, David M
Goldenberg6,7,8, Robert M Sharkey8, Jacques Barbet2,9, Franoise Kraeber-Bodr2,3,4 and Caroline Rousseau2,3. 1ICO Cancer
Center, Oncology Unit, Saint Herblain, France; 2ICO Cancer Center, Nuclear Medicine Unit, Saint Herblain, France; 3CRCNA,
Inserm U892, CNRS UMR 6299, Nantes, France; 4University Hospital, Nuclear Medicine Unit, Nantes, France; 5ICO Cancer
Center, Physics Unit, Saint Herblain, France; 6Garden State Cancer Center, Center for Molecular Medicine and Immunology,
Morris Plains, NJ; 7IBC Pharmaceuticals, Inc, Morris Plains, NJ; 8Immunomedics, Inc, Morris Plains, NJ and 9GIP Arronax, Saint
Herblain, France.
Body: Objectives: Different imaging methods are available to detect BC recurrence, but several new noninvasive antibody
imaging methods targeting membranous IGF-1R expression1, HER2/neu2 or CD1383 have been tested in BC pre-clinical trials.
Today, a new generation anti-CEA x anti-HSG humanized trivalent TF2 BsMAb and 68Ga-IMP288 HSG peptide is available with
good features for immuno-PET in preclinical studies4. This study aimed to compare the sensitivity of anti-CEA immuno-PET/CT
using pretargeted 68Ga-IMP288 to morphological imaging and FDG-PET/CT in metastatic BC patients.
Methods: Ten patients, enrolled in an optimization immuno-PET study underwent whole-body immuno-PET/CT recorded 1h and
2h after injection of 150 MBq of 68Ga-IMP288 pretargeted by 120 nmol of TF2 injected 24h to 30h before, in addition to
thoracic-abdominal-pelvic CT and FDG-PET/CT. Bone (n=5) and brain MRI (n=2) were also performed in some cases to confirm
abnormalities detected by other modalities. The gold standard was determined by follow-up and a lesion detected by at least 2
imaging modalities was considered as positive.
Results: Median CA15-3 was 264.1 kUI/L (31.6 to 2448) and median CEA was 48.25 g/L (9.5 to 1359.0). A total of 537 lesions
were detected by immuno-PET/CT, 247 by CT, 160 by bone MRI, and 428 by FDG-PET/CT. To date, 524 lesions were confirmed
as pathologic by the gold standard: 17 in nodes, 1 in lung, 83 in liver, 418 in bone, 1 in skin, and 4 in brain. Overall sensitivity of
immuno-PET was 92.8%, with 100% sensitivity for bone, liver, skin, and brain, 92.8% for nodes, and 28.6% for lung. Overall
sensitivity of CT and FDG-PET/CT were 74% and 95.2%, respectively. CT and FDG-PET/CT had 54.5 and 100% sensitivity for
nodes, 90 and 100 % for liver, 100 and 85.7% for lung, and 68% and 94% for bone, respectively. Bone MRI had 92.3% sensitivity.
Brain lesions were only detected by immuno-PET/CT and confirmed by MRI. Median tumor SUVpeak on immuno-PET at 1h and
FDG-PET/CT were 5.05 (3.52 to 24.55) and 3.3 (0.47-10.95), respectively. A 2h, median tumor SUVpeak on immuno-PET was
5.23 (3.09-34.27), 5/9 patients showing an increased tumor uptake between 1 and 2h, with no lesion being detected only at 2h.
Conclusion: These results demonstrate the high accuracy of anti-CEA pretargeted immuno-PET/CT for staging BC patients,
especially for bone, liver and brain evaluation. Immuno-PET allowed detection of bone lesions in eras not explored by MRI.
1. Heskamp S. J Nucl Med 2010.
2. Dijkers ECF. J Nucl Med 2009.
3. Rousseau C. EJNMMI Res. 2011.4. Schoffelen, Mol Cancer Ther, 2010.

Title: Evaluation of apoptosis in breast cancer using the novel PET probe [18F]ICMT-11 in patients treated with neoadjuvant FEC
chemotherapy: Initial assessment of optimum imaging time and relation to caspase-3 immunostaining
Shairoz Merchant1, Eric O Aboagye1, Adrian Lim1, Kasia Kozlowski1, Naina Patel1, Jennifer Steel1, Susan Cleator1, Sami
Shousha1, Vidhya Varghese1, Raoul C Coombes1 and Laura Kenny1. 1Imperial College London, London, United Kingdom.
Body: Background:
[18F]ICMT-11 is an isatin analogue which has been developed by our group as a novel PET radiotracer for studies of apoptosis in
vivo. Preclinical studies have demonstrated subnanomolar affinity to caspase 3, and validated the potential for imaging apoptosis
in xenograft models. A first-in-man study showed that the agent was well tolerated with acceptable dosimetry. This is the first
study of this agent to measure the effect of chemotherapy on radiotracer uptake in patients. As apoptosis is a dynamic process,
one of the main objectives of the study was to determine the optimal time-point for imaging post-chemotherapy and compare the
results with immunohistochemistry assessments at the same time-points. The study was approved by a regional ethics committee
and ARSAC.
Methods:
7 patients with breast tumour lesions measuring 15mm or more, due to undergo neoadjuvant chemotherapy with FEC (5FU,
Epirubicin, Cyclophosphamide) had dynamic PET scans for 66mins 30seconds following intravenous injection of [18F]ICMT-11
with a mean activity of 340.82 Mbq20.76 and Specific Activity range of 447.014-5128.34 Gbq/mol prior to chemotherapy and
24h-2 weeks post-chemotherapy. A breast biopsy was also obtained within a few hours of the 2nd PET scan to correlate
apoptosis in the breast tissue utilising TUNEL and Caspase 3 staining by immunohistochemistry. Volumes of interest were drawn
manually and analyzed using Analyze software.
Results:
The scans were well tolerated in all patients. Uptake of [18F]ICMT-11 was demonstrated in all tumor lesions. The tumours studied
included ER positive and PR positive, HER2 positive and triple negative patients. The first cohort patients were imaged
pre-chemotherapy and 24-48h post chemotherapy. Tumour to Breast ratio (TBR) showed an increase from 1.420.21(pre) to
1.710.33 (post). Tumour to muscle ratio (TMR) was not increased, 1.520.30(pre) and 1.220.09 (post). In addition, an increase
the SUV was noted in the lymph nodes of patients, at both 24 and 48h. (SUVav 0.390.02 (pre), 0.450.03 (post), and SUVmax
0.87 0.02(pre) ,1.220.12 (post). The lymph nodes were however not sampled for immunohistochemistry. A further cohort of
patients had the follow-up scan 2 weeks post chemotherapy, TBR and TMR were both increased in this cohort 1.500.22 (pre),
2.520.48 (post),and 1.820.10 (pre), 2.080.0.04 (post), respectively. Caspase and TUNEL labelling with immunohistochemistry
also showed increased in apoptosis in the breast biopsies at 2 weeks compared to baseline in keeping with the PET data.
Conclusion:
These preliminary data suggest that [18F]ICMT-11 is a promising marker for chemotherapy induced apoptosis in vivo, and
correlates with findings in tumor biospsies using TUNEL and immunostaining for caspase 3. Further work is underway to study a
larger cohort of patients, and identify the optimal PET pharmacokinetic parameter to describe [18F]ICMT-11 uptake and retention.

Title: Opto-acoustic nomograms for improving breast cancer diagnosis
Philip Lavin1 and Thomas Stavros2. 1Boston Biostatistics Research Foundation, Framingham, MA and 2Seno Medical Instruments,
San Antonio, TX.
Body: Background/Objective:
Diagnostic specificity remains disappointingly low for ultrasound-based methodologies optimized to achieve near 100%
sensitivity. Opto-acoustic (OA) imaging is a fusion of real time co-registered, interleaved laser optic and ultrasound imaging that
shows fused functional findings (relative de-oxygenation of hemoglobin) and morphologic information (tumor angiogenesis) within
and around breast masses using a hand-held duplex OA probe. We assess the improvement in sensitivities and specificities of
nomograms (N) based on regression models to classify (CL) benign (B) vs. malignant (M) and to project probability of malignancy
(POM) based on five feature-based metrics scored by 21 independent readers (IRs) and an expert reader. We examine OA
feature separation and nomogram performance.
Methods:
Masses from a series of 100 subjects with 80 biopsies (42 B, 38 M) were blindly evaluated prior to both core biopsy and excision.
The OA algorithm was locked. Three internal OA findings (density of vascularity [V], relative blood oxygen saturation [O], and
hemoglobin [H]) and two external OA findings (boundary zone [Z] and peri-tumoral radiating vessels [R]) were assigned 0-5/6
ordinal scores. Feature distributions were compared using a two-sided Kruskal-Wallis test.
Results:
There were consistently significant differences between the feature distributions for benign vs. malignant: density of vascularity
(14/21 IRs), relative blood oxygen saturation (11/21 IRs), hemoglobin (16/21 IRs), boundary zone (all IRs), and peri-tumoral
radiating vessels (all IRs) always with lower scores for benign vs. malignant. The mean IR sensitivities were near 100% for all
IRs. The mean specificities from the nomograms were 49% (POM), 43% (CL), and 48% (averaged).
Conclusions:
The study results indicate that OA findings can be independently and quickly mastered by practicing IRs to consistently
differentiate masses. Nomograms offer further confidence to enhance decision making to differentiate. If confirmed in a larger
series, OA findings might be useful in differentiation and thus sparing biopsies in addition to more customized surgeries.

Title: In vivo lymphatic imaging of a human inflammatory breast cancer model
Sunkuk Kwon1, Germaine Agollah1, Grace Wu1 and Eva M Sevick-Muraca1. 1University of Texas Health Science Center, Houston,
TX.
Body: Background: Inflammatory breast cancer (IBC) is the most aggressive but poorly understood type of breast cancer. IBC
constitutes only appoximately 5% of all newly diagnosed breast cancers, yet responsible for 8-10% of breast cancer-related
deaths. Unfortunately, there are no definitive molecular or pathological, early diagnostic criteria for IBC. IBC grows in nests or
sheets, spreading outward into the skin and eventually distant sites through dermal lymphatic vessels; therefore, one of the key
pathological characteristics of IBC is dermal lymphatic invasion by tumor emboli, which can lead to obstruction of the lymphatic
drainage possibly causing the clinical inflammatory features of diffuse erythema, rapid breast enlargement and edema as
indicated by peau dorange (orange peel) appearance covering at least a third of the breast surface. However, relatively little is
known about the role of lymphatic function in IBC growth and metastasis. The purpose of this study was to non-invasively and
longitudinally image changes of lymphatic drainage patterns in mice bearing the triple negative, human IBC SUM149 cells, which
were stably transfected with iRFP gene reporter (iRFP-SUM149) in order to better monitor tumor growth and metastasis.
Methods: Eight to ten weeks old female non-obese diabetic/severe combined immunodeficiency (NOD/SCID) mice were housed
and fed special sterilized pelleted food and sterilized water. Two weeks after feeding special diets, mice were subcutaneously
injected with iRFP-SUM149 cells in the left hindlimb or orthotopically in the left inguinal mammary fat pad (MFP). Fluorescence
images were acquired immediately after and for up to 10 mins after i.d. injection of indocyanine green (ICG) using a custom-built
NIRF imaging system. To achieve a greater magnification, a macrolens was used.
Results: Our NIRF imaging data showed gradual changes of lymphatic vessels around the tumor. In mice with an orthotopic
tumor, ICG accumulation was observed in the distal peritumoral region at up to 3 weeks p.i., indicative of the start of tumor
obstruction of normal lymphatic drainage. As tumors grew, the greater extent of lymph flow, but not all, was obstructed by a
tumor, resulting in dermal backflow and thus staining more lymphatic capillaries and eventually rerouting of lymphatic drainage
was detected due to complete obstruction of lymph flow by a tumor. Similar to an orthotopic model, mice with a s.c. tumor also
showed altered lymphatic drainage patterns during tumor growth. Interestingly, extravasation of ICG into iRFP-SUM149 was
detected at 3 weeks p.i., due to leaky lymphatic vessels. A similar lymphatic phenotype has also been observed in IBC patients.
Previously, we have imaged atypical tortuous lymphatics and altered lymphatic drainage in the arm and affected breast of a
patient with IBC using clinical NIRF imaging developed in our team to image as deep as 3-5 cm with a microdose of ICG.
Conclusions: Longitudinal, non-invasive imaging of lymphatic functional and architectural changes provides a new method to
dynamically monitor lymphatic response to IBC growth and metastasis.

Title: Early optical tomography changes correlate with residual cancer burden scores in women receiving neoadjuvant
chemotherapy
Emerson A Lim1,2, Jacqueline E Gunther3, Molly Flexman4, Hyun K Kim3, Hanina Hibshoosh2,5, Kevin Kalinsky1,2, Katherine
Crew1,2,6, Matthew Maurer1,2, Sheldon Feldman2,7, Bret Taback2,7, Preya Ananthakrishnan2,7, Margaret Chen2,7, Susan Refice2,
Andreas Hielscher3,8,9 and Dawn L Hershman1,2,6. 1Internal Medicine Columbia Univeristy Medical Center, New York, NY;
2
Herbert Irving Comprehensive Cancer Center, New York, NY; 3BioMedical Engineering Columbia University, New York, NY;
4
Philips Corporation, Tarrytown, NY; 5Pathology Columiba University, New York, NY; 6Mailman School of Public Health, New
York, NY; 7Columbia University Medical Center, New York, NY; 8Electrical Engineering Columbia University, New York, NY
and 9Radiology Columbia University Medical Center, New York, NY.
Body: Background: The Residual Cancer Burden (RCB) score predicts survival in patients (pts) with breast cancer (BC) treated
with neoadjuvant chemotherapy (NACT). Predicting tumor response early during NACT may allow for treatment optimization.
Diffuse optical tomography (DOT) is an imaging modality that measures the distribution of water (H2O), oxy- (HbO), and
deoxy-hemoglobin (Hb) concentrations as a surrogate for vascularity and architecture. We hypothesize that the 2-week change in
DOT parameters will correlate with the RCB score. We also explored the association between DOT parameters and tissue
biomarkers: Ki-67 change and microvessel density (MVD).
Methods: Women with stage II-IIIc invasive BC scheduled to receive NACT with 12 cycles of a weekly taxane followed by 4 cycles
of doxorubicin with cyclophosphamide were enrolled. Treatment with biologic therapies was allowed. DOT assessments were
made before NACT and after 2 weeks on treatment. DOT data were reconstructed into 3D images of the tumor region, from which
HbO, Hb, and H2O concentrations were extracted. Final pathology specimens were scored for the RCB index (continuous), RCB
class (0, 1, 2, 3), and a dichotomized RCB score (RCB class 0 or 1: responders; RCB class 2 or 3: non-responders). Ki-67 was
measured on baseline tumor biopsies and surgical specimens. MVD was assessed on baseline tumor biopsies. Correlation
analysis, ANOVA testing, and two sample t-tests were used to evaluate the relationship between the 2-week changes in DOT
parameters and the RCB score and Ki-67 change. Correlation was assessed between MVD and baseline DOT measures.
Results: Since July 2011, we have recruited 28 pts of a total planned accrual of 40. 25 pts have had surgery and complete data
are available for 23. Of the 23 pts, 6 had a pCR (RCB 0), 2 had RCB 1, 10 had RCB 2, and 5 had RCB 3. The Pearson
correlations (r) between the 2-week change in HbO, Hb, and H2O with the continuous RCB index were 0.65 (p=0.002), 0.70
(p=0.0006), and 0.70 (p=0.0006), respectively. There was a significant difference in the 2-week Hb change for pts with RCB 0
compared to pts with RCB 1, 2, or 3. There were significant differences in the 2-week change in H2O and HbO for pts with RCB 0
compared to pts with RCB 2. There were also significant differences between DOT parameters by the dichotomized RCB score
(table 1). There was an association in Ki-67 change and 2-week H2O change (r=0.43 p=0.059). A subset of 15 pts had MVD
assessments, but these did not correlate with baseline DOT parameters (r 0.18, p>0.5).
2-week DOT % change by dichotomized RCB score
HbO
Hb
H2O
Responder (RCB 0/1)
-31.0%
-22.3%
-5.5%
Non-Responder (RCB 2/3)
-4.4%
-2.3%
-0.3%
P value
0.02
0.01
0.015
Conclusions: Two-week DOT change is an early predictor of response to NACT as measured by the RCB score. We found
significant associations between the RCB index with 2-week changes in HbO, Hb, and H2O. Significantly different changes in
DOT parameters were associated with the other RCB classifications. Ki-67 changes and baseline MVD were not statistically
significantly associated with DOT parameters. We are analyzing static and dynamic DOT data on the remaining pts. Additional pts
are being recruited to evaluate DOTs predictive ability by tumor subtype.

Title: Characterization of metastatic breast cancer lesions with ferumoxytol MRI and treatment response to MM-398,
nanoliposomal irinotecan (nal-IRI)
Jasgit C Sachdev2, Ramesh K Ramanathan2, Natarajan Raghunand3, Jaeyeon Kim1, Stephan G Klinz1, Eliel Bayever1, Jonathan
B Fitzgerald1 and Ronald L Korn4. 1Merrimack Pharmaceuticals, Inc, Cambridge, MA; 2Virginia G. Piper Cancer Center, Scottsdale
Healthcare, Scottsdale, AZ; 3Translational Cancer Imaging, Arizona Cancer Center, Tucson, AZ and 4Imaging Endpoints,
Scottsdale, AZ.
Body: Introduction
Irinotecan has known activity in metastatic breast cancer (MBC). MM-398, nanoliposomal irinotecan (nal-IRI), is designed to
exploit leaky tumor vasculature for enhanced drug delivery to tumors. Tumor deposition of nal-IRI and subsequent conversion to
SN-38 in both neoplastic cells and tumor associated macrophages (TAM) may positively correlate with activity. Predictive
biomarkers to measure tumor deposition could identify patients likely to benefit from nal-IRI. Ferumoxytol (FMX), an iron-oxide
superparamagnetic nanoparticle with MRI contrast properties, is taken up by TAMs with similar distribution patterns to nal-IRI in
preclinical models. Our previous work has shown the feasibility of quantitative FMX MRI (Fe-MRI) of tumor lesions, and we
developed a quantitative mechanistic PK model of FMX deposition (AACR 2014, abstract #CT224). Here we report nal-IRI activity
and FMX levels in MBC patients on the study.
Patients (n=15) with refractory solid tumors and at least two metastatic lesions >2 cm accessible for percutaneous biopsy were
enrolled in a Phase 1 study. Fe-MRI scans were performed using T2* iron sensitive sequences prior to and following FMX
infusion (1 h, 24 h, 72 h). T2* signal was used to calculate FMX levels in total lesions by comparison to a standard curve.
Comparison of quantified FMX lesion uptake with a mechanistic PK model previously indicated that tissue permeability to FMX
contributed to early Fe-MRI signals at 1 h and 24 h, while FMX binding contributed at 72 h. Patients then received nal-IRI (80
mg/m2 q2w) until progression. Core biopsies were obtained 72 h after both FMX and nal-IRI infusions. RECIST evaluation was
done by CT every 8 weeks.
Results
FMX was well tolerated, and adverse events to nal-IRI were consistent with previous studies. Three of the 13 patients receiving
nal-IRI had ER/PR+ MBC (median # of prior Rx: 8 compared to 4 for all study patients). Thirteen liver lesions (4-5/pt) were
evaluated by FMX-MRI and CT for these 3 patients. Average lesion size: 26.911.2 mm diameter and 8.111.3 cm3 (median 4.7
cm3). Time on treatment for the 3 patients was 57, 126 and 256 days (study median 57 days). Best overall response was 1 stable
disease (SD) and 1 partial response (PR) in these 3 patients. The patient with a PR had an average lesion size reduction of
44.5%, while the patient with SD had an average lesion size increase of 12.5% at final evaluation. Lesions that shrank after
nal-IRI showed higher early levels of FMX compared to the study median (median 39.6 vs. 32.6 mcg/mL at 1 h; median 37.7 vs.
34.5 mcg/mL at 24 h). This relationship between lesion response and FMX levels was consistent with the lesion behavior in the
full data set (n=31 lesions/9 patients across 7 indications) of the study.
Conclusions
Clinical activity of nal-IRI was observed in a subset of heavily treated ER/PR+ MBC patients. The relationship between FMX
levels in tumor lesions and nal-IRI activity suggests that lesion permeability to FMX may be a useful biomarker for nal-IRI
deposition and tumor response in MBC and potentially other indications. A multi-institution expansion of this study in
HER2-negative MBC is planned to confirm these findings.

Title: Fidelity of FDG-PET in breast cancer: Reproducibility at multiple sites
Hannah M Linden1, Jennifer Specht1, Lanell Peterson1, Brenda Kurland2, Andrew Shields1, Darrin Byrd1, Alena Novakova1,
Rebecca Christofel1, Mark Muzi1, David Mankoff3 and Kinahan Paul1. 1University of Washington, Seattle, WA; 2University of
Pittsburgh, Pittsburgh, PA and 3University of Pennsylvania, Philadelphia, PA.
Body: Introduction
Breast cancer is a common, treatable malignancy, with frequent metastasis to bone. Patients with bone-dominant disease are
often excluded from clinical trials due to a lack of RECIST "measurable" disease. FDG-PET can help quantitatively measure
multiple tumor sites, and assay disease activity in bone. However, the reproducibility of FDG-PET at multiple sites is unproven.
We sought to pilot alignment of clinical protocols prior to imaging, phantom testing and then repeat patient scans, in patients
undergoing clinical FDG-PET to test the fidelity of quantitative FDG-PET imaging.
Methods
After determination and alignment of clinical protocols, and serial successful phantom imaging, ten female patients with
metastatic breast cancer underwent paired FDG-PET/CT test-retest studies with no more than 15 days between scans and
without interim change in treatment. Seven patients were studied in the same scanner and 3 patients were studied in 2 different
scanners in our clinical network. Each PET/CT scanners quantitative performance was monitored with NIST-traceable reference
sources to ensure proper calibration. Images were interpreted and SUV metrics were estimated at a central lab. Linear mixed
models with a random intercept were fitted to compare test-retest differences in multiple lesions per patient.
Results
SUVmax was assessed in a total of 68 lesions (52 bone, 16 other sites). Average SUVmax ranged from 2.3 to 18.2 (meanSD =
5.72.6) per patient. The median SUVmax difference was 0.25 (5%) for 35 lesions imaged twice in the same scanner, and was
0.01 (0.2%) for 33 lesions imaged in two different scanners. In a linear mixed-effects model with random patient effects, there was
no difference in average percentage SUV difference for the same scanner versus different scanners (p=0.70). In the same model,
the absolute percentage difference in SUVmax for bone lesions was estimated as 6 percentage points lower than for other sites
(p=0.002, 95% confidence interval 2%-10%).
Conclusions
If PET/CT systems are carefully calibrated, and imaging protocols are consistent, then variability associated with FDG SUVmax
between scans is similar to prior test-retest studies. Bone lesions appear to have tighter reproducibility than soft tissue lesions.
Clinical trials that utilize quantitative PET/CT imaging throughout a network of calibrated PET/CT scanners could increase patient
recruitment and improve confidence in trial results. Accrual is ongoing and results will be updated.
Research support
Supported by NIH grant U01-CA148131 and NCI-SAIC Contract 24XS036-004.

Title: Single dose acute toxicity and long-term biodistribution of perfluoropentane loaded iron doped silica nanoshells
Sarah L Blair1, Alexander Liberman1, Robert Viveros1, Jacqueline Corbeil1, Christopher V Barback1, Robert F Mattrey1, William C
Trogler1 and Andrew C Kummel1. 1University of California, San Diego, La Jolla, CA.
Body: Background: Our lab has been focusing on developing a better method of localizing non-palpable breast cancers without
wire or seed localization. Perfluoropentane (PFP) loaded Fe-SiO2 nanoshells have been developed as a color Doppler ultrasound
contrast imaging agent which can act as small volume (100 ul) injectable stationary guide-marker for breast tumor resection.
Preliminary experiments have demonstrated that the nanoshells can provide robust contrast for periods extending past 10 days in
vivo in Py8119 epithelial breast tumor bearing mice with no adverse affect to the mice. Short-term biodistribution over 72 hours of
nanoshells using In111 labeled nanoshells demonstrated with gamma scintigraphy that intravenously dosed particles primarily
accumulate in the liver but some radioactive signal can be seen in the bladder. The long imaging lifetime of these nanoshells
necessitates the need to study long-term toxicity and biodistribution.
Materials and Methods: Fe-SiO2 nanoshells and Pure SiO2 nanoshells where synthesized via sol-gel method on polystyrene
templates and then calcined to yield 500 nm hollow rigid nanoshells which were then filled with vaporized perfluoropentane. 100
ul of nanoshells at 4 mg/ml of the Fe-SiO2 nanoshells and at a dose of 2 mg/ml of pure SiO2 nanoshells were injected IV into
healthy 8-week old Swiss white mice. The difference in mass dose was due to make the particle count between the two doses
equivalent. Blood was collected weekly for serum chemistry and hematology. After 10 weeks mice were sacrificed, H&E was
performed on organs of interest as well as inductively coupled plasma optical emission spectroscopy (ICP-OES) for trace silicon
determination for long-term biodistribution.
Results: No significant effect due to the administration has yet been observed on the health of brain, lung, heart, kidney, liver,
spleen or muscle tissue examined from these animals at a dose 4 mg/ml 100 ul of the Fe-SiO2 nanoshells and at a dose of 2
mg/ml of pure SiO2 nanoshells. Mouse weight steadily increased from 25.8 2 grams to 30.7 2.6 grams over the course of 10
weeks. Creatinine levels were detected at 0.2 0.14 mg/dl indicating healthy renal function. Serum glutamic pyruvic
transaminase (SGPT) was used as a measure of liver health, and SGPT values for both control (55.81 6.31 U/L) and nanoshell
injected mice (47.74 11.04 U/L) are approximately the same over the course of 10 weeks indicating good liver health. Silicon
content in mouse organs diminished over the course of 10 weeks by ICP-OES in both the Fe-SiO2 and pure SiO2 nanoshells.
Conclusions: No indication of toxicity was observed from a 400 ug systemically administered dose of Fe-SiO2 nanoshells.
Furthermore, the reduction of silicon content in the organs over the course of 10 weeks suggests a possible excretion pathway for
silica or solid nanoparticulate materials. The efficacy in long term ultrasound contrast and high margin of safety indicates that this
particle formulation is ready for phase 1 clinical trial in humans as a future method to localize nonpalpable breast cancers.

Title: Improving specificity and refining diagnostic accuracy of MRI in breast cancer with dedicated breast PET (dbPET)
Ines Dominguez1, Michel Herranz1, Sze Yiun Teo2, Elena Brozos1, Carmela Rodriguez1, Jasper Chaal3, Juan Cueva1, Jose
Ramon Antnez1, Gabriel Gonzalez Pavn4, Marielle Fortier1, Rafael Lpez1 and Alvaro Ruibal1. 1Complejo Hospitalario
Universitario de Santiago de Compostela, Santiago de Compostela, Galicia, Spain; 2Kerbang Kerbau (KK) Women and Childrens
Hospital, Singapore, Singapore; 3Clinical Imaging Research Centre, Singapore, Singapore and 4Oncovision, Valencia,
Comunidad Valenciana, Spain.
Body: BACKGROUND.
Continued progress in the control of breast cancer will require sustained efforts to provide high-quality screening, diagnosis, and
treatment to all segments of the population. MRI has the advantages of providing a 3D view of the breast with high sensitivity,
using non-ionizing radiation. However, MRI has significant limitations including its moderate specificity that, in combination with
high sensitivity, often leads to unnecessary biopsies. Recently, the MAMmography with Molecular Imaging (MAMMI) dedicated
breast PET (dbPET) has emerged as an additional imaging tool for breast cancer diagnosis, clarification of complex lesions and
therapy follow-up. It is well known that 18F-FDG-PET has high specificity by assessing metabolic activity, potentially reducing the
number of false positive findings. To compare FDG-dbPET with the conventional magnetic resonance imaging (MRI) on breast
lesion characterization, we analyzed sensitivity, specificity, Positive Predictive Value (PPV) and Negative Predictive Value (NPV)
of 37 cases of patients with BC using both dbPET and MRI.
METHODS.
Thirty-seven women with known or suspected breast carcinoma (41 lesions: 36 invasive carcinomas, 2 noninvasive carcinomas,
1 case of ductal hyperplasia, and 2 benign lesions) were enrolled in this study. Both a prone dbPET Mammi scan, and routine
breast MRI scans were performed. A joint reading of MRI and PET scans side-by-side by a nuclear medicine physician and a
radiologist was performed. Sensitivity and specificity of MRI and dbPET scans were calculated on the basis of post-surgical
pathology reports. Breast MRI examinations were performed in a 1.5-T or 3-T commercial imager (Siemens Medical Solutions)
with the use of a dedicated breast coil. The imaging sequence included a sagittal T1-weighted localizing sequence followed by a
sagittal T2- weighted sequence. A T1-weighted 3D, fat-suppressed fast spoiled-gradient-echo sequence was then performed with
an injection of 0.1 mmol per kilogram body weight of gadolinium dimeglumine (Magnevist; Schering). A prone position
high-resolution dedicated breast PET was performed 60 min after administration of 90-120 MBq of 18F-FDG.
RESULTS.
A total of 41 lesions were assessed. Lesion size range was 0.2 to 7.6 cm. In lesion-by-lesion analysis, sensitivity and specificity of
MRI alone were 91% and 54%, respectively; while lesion-based sensitivity of dbPET was 93% and breast-based specificity was
100%. The positive predictive value and the negative predictive value for MRI alone were 69% and 85%, respectively; and for
dbPET were 100% and 89%, respectively. In a significant number of cases, dbPET helped to clarify or disprove positive findings
by MRI, and in four cases helped to define new positives that had gone unnoticed at MRI,
CONCLUSION:
Dedicated breast PET scans increase the specificity of MRI. False positives, one of the most challenging aspects of MRI in breast
lesions, are reduced. The results of the current study show that FDG-dbPET is more effective than MRI in detecting true breast
cancer positives. Its functional information may improve the likelihood of a successful excision, reduce costs from additional
procedures, and minimize discomfort and anxiety for the patient.

Title: In vitro and in vivo imaging of Her2 with cyclic peptides derived from directed evolution
Lindsay E Kelderhouse1, Amanda Hardy2, Yong Pan1, Patrea Rhea1, Argentina Ornelas1, Seth Gammon1, Peiying Yang1, Stephen
Fiacco2 and Steven W Millward1. 1MD Anderson Cancer Center, Houston, TX and 2EvoRx Technologies, Pasadena, CA.
Body: The implementation of personalized medicine for Her2-positive breast cancer treatment (Trastuzumab and Pertuzumab)
has made a dramatic effect on overall patient outcome. Unfortunately, there are currently no FDA-approved imaging agents to
monitor Her2-positive breast cancer leaving physicians to rely on invasive biopsies and anatomical imaging to monitor treatment
with Her2-targeted therapeutics. There is also increasing evidence that women initially diagnosed with Her2-negative breast can
present with Her2-positive disease upon recurrence. Noninvasive, whole-body visualization of Her2 would identify Her2-recurrent
disease and serve as a powerful tool to monitor the effectiveness of new and emerging Her2-targeted therapeutics (e.g. TDM-1).
While many Her2-antibody based imaging agents have been used in preclinical applications, their long circulating half-lives, high
liver uptake, and poor synthetic accessibility pose a challenges for clinical translation. Here we describe the use of highly stable
cyclic peptides derived from biological display as imaging agents for Her2-positive breast cancer. These peptides have
antibody-like high affinity (<10 nM) for Her2-positive tumors and their size (<2 kDa) allows for higher rates of clearance, lower
background signal, and consequently higher image sensitivity. Anti-Her2 peptides were pre-optimized for affinity, protease
resistance and bio-stability prior to labeling with the Cy7.5 near-infrared dye. Her2-specificity was determined by in vitro assays
with Her2-postive (BT474 and SKOV3), HER2-normal (MCF7) and Her2-negative (MDA-MB-231) cell lines. Sequences that
showed superior affinity and selectivity were used to acquire fluorescent images of ectopic and orthotopic murine models of
Her2-positive breast cancer. Our preliminary imaging data suggests that this scaffold has excellent translational potential for
targeted molecular imaging of Her2-postive breast cancer.

Title: PEM tests will have a high accuracy detecting breast cancer, if an individualized dosage of F-18-FDG is used
Frank HH Mller1, Hentschel Michael2, Anton G Mller1 and Jamshid Farahati3. 1Radiology & Nuclear Medicine, Ludwigshafen,
Rheinland-Pfalz, Germany; 2Inselspital, Bern, Switzerland and 3Clinic for Radiology & Nuclear Medicine, Elisabeth-Hospital,
Dorsten, Nordrhein-Westphalen, Germany.
Body: Introduction: The diagnostic performance of the first European PEM (PET mammography) is based on individualized
dosage of FDG evaluated in 108 females with 166 suspected breast lesions.
Methods: PEM imaging of the breast was performed in 108 consecutive females with 166 suspicious breast lesions or known
breast cancer (BC) 90 min after i.v. application of 3.5 MBq/kg F-18-FDG per kg body weight. The maximum PEM uptake value
(PUV) was derived from a ROI around the target lesion and was correlated with a corresponding non-target ROI in the contra
lateral healthy breast to determine the target/non-target ratio (PUV-ratio). Images were analyzed by 2 experienced readers
independently as compared to histopathology in all cases. The between group analyses for all malignant, benign and
corresponding non-target lesions were calculated by paired Student t-Test. The mean target/non-target ratio in patients with BC
compared to healthy patients was calculated by independent Student t-Test. Significance level was considered at p value <0.05.
Receiver operating characteristic (ROC) analyses were employed to determine associations with PUV and PUV-ratio.
Results: A total of 27 out of 166 (16.2%) lesions were malignant. Mean of PUV was estimated to be 3.92.5 in malignant lesions
and 1.20.4 for the contra-lateral healthy breast (p<0.001). The mean PUV-ratio in patients with BC of 3.41.5 was significantly
higher as compared to benign lesions 1.20.3 (p<0.001). The area under the ROC curve was 0.997 (0.000-1.000) for PUV and
0.986 (0.965-1.000) for PUV-ratio.
PEM was true-positive in 27 cases of cancers and false positive in 3 cases (papilloma, mastopathia bds.) considering a PUVmax
> 1,9, resulting in sensitivity of 100%, specificity of 97%, positive predictive value of 90%, and a negative predictive value of
100%.
Conclusion:
The different European approach in PEM using personalized FDG-dosage enables the comparability of FDG-metabolism in
patients, resulting in a high diagnostic accuracy of PEM tests.
Literature:
1. Hentschel M. et al. Can body volume be determined by PET? Eur J Nucl Med Mol Imaging 32:564568 (2005)
2. Berg W., et al. Comparative Effectiveness of PEM and MRI for Presurgical Planning of the Ipsilateral Breast in Women with
Breast Cancer. Radiology 1:258 (2011)
3. Narayanan D, Madsen KS, Kalinyak JE, Berg WA. Am J Roentgenol. 196:971-81(2011)
4. Kalinyak, JE et al, PET guided breast biopsy. Breast J. 17:143-151(2011)
5. Caldarella C. et al., Diagnostic Performance of Dedicated Positron Emission Mammography Using Fluorine-18Fluorodeoxyglucose in Women With Suspicious Breast Lesions: A Meta-analysis. Clinical Breast Cancer: in Press.

Title: Breast specific gamma imaging (BSGI) and breast magnetic resonance imaging (MRI): Comparison of sensitivity and
specificity in women prior to breast biopsy with BIRADS 4 or 5 finding on mammography in a community setting
Alison K Conlin1, Nicole Moxon1, Helena Hoen1, Christina Gougoutas-Fox1, Maureen O Baxter1, Amy Weinstein1, Maritza Martel1,
Tracy L Kelly1 and Walter J Urba1. 1Providence Cancer Center, Portland, OR.
Body: Background: Diagnostic imaging following a new diagnosis of breast cancer remains an active area of research balancing
value and outcomes. Decisions about surgical options and neo-adjuvant therapy depend greatly on the accuracy of these
pre-operative assessments. BMRI use has increased tenfold from 2000 to 2011 (Stout et al, JAMA 2013) and estimation of
sensitivity has been high but specificity has varied between 30-80%(Bluemke et al, JAMA 2004). BSGI is a novel molecular
imaging technique that uses a gamma camera to track the uptake of a radio tracer (technitium Tc99m sestamibi) by breast cancer
cells and has been used interchangeably with BMRI without rigorous evidence of equivalency (Khalkhauli, et al, J Nuc Med
2000).The majority of research into the sensitivity and specificity of these tests has been retrospective, only on women with
known cancer, and potentially biased by post-biopsy changes to breast tissue.
Methods: Therefore we performed a prospective study employing both techniques to image women with BIRADS 4 or 5 lesions
on diagnostic mammogram prior to their planned breast biopsy. The BSGI and BMRI were reviewed by one of three dedicated
breast radiologists and the pathology was reviewed on the biopsy or any additional biopsy/excision by one pathologist. We
compared the BSGI and BMRI against the final pathology for sensitivity and specificity. In addition, we surveyed the women for
quality of life measures 3 months later.
Results: Between January 2012 and April 2014 we enrolled 74 women (ages 30-80) at 2 NAPBC accredited breast centers
located in a community based setting in Portland, OR. The initial diagnostic mammographic studies resulted in 23 women (32%)
with BIRADS 4A, 27 (37%) with BIRADS 4B, 8 women with 4C (11%) while 8 women (11%) had BIRADS 5 lesions prompting
biopsy. All women were biopsied and 27 (37%) were found to have an invasive or in situ cancer while 5 (7%) had atypical
hyperplasia or LCIS found. Sixteen women had additional biopsies performed, outside of the planned area, as a result of BMRI or
BSGI, 11 (69%) were based on BMRI findings and 5 (31%) were areas seen on both BMRI and BSGI. In these additional biopsies
5 were in situ or invasive cancer and 2 were contra-lateral cancers, the rest were benign tissue. The sensitivity of BMRI was
84.0% and BSGI was 74.1%. The specificity was found to be 57.8% and 80.4% respectively. One patient withdrew and 3 women
did not complete BMRI due to claustrophobia or body habitus.
Quality of life data is still being analyzed.
Conclusions: We report here the sensitivity and specificity of BMRI compared with BSGI in women with BIRADS 4 and 5 breast
lesions on diagnostic mammography. Importantly imaging was done before biopsy and therefore not biased by any effect from
that procedure. In this study we find BMRI appears to have better sensitivity but lower specificity than BSGI. We also observed
that the use of BMRI and/or BSGI prompted 16 extra biopsies of which less than half were additional or contra-lateral cancer. The
incorporation of these tests into the evaluation of suspected cancer should consider these findings as well as cost and quality of
life.

Title: The spectroscopic feature of breast cancer
Hiroyuki Ogura1, Nobuko Yoshizawa2, Kenji Yoshimoto3, Hatsuko Nasu2, Yumiko Taki1, Youko Hosokawa1, Ryouichi Matsunuma1,
Yoshimi Ide1, Etsuko Yamaki3, Toshihiko Suzuki3, Motoki Oda3, Yukio Ueda3, Yutaka Yamashita3 and Harumi Sakahara2. 1Breast
Surgery, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan; 2Central Research Laboratory, Hamamatsu
Photonics K.K., Hamamatsu, Shizuoka, Japan and 3Radiology, Hamamatsu University School of Medicine, Hamamatsu,
Shizuoka, Japan.
Body: Objectives: To examine optical properties of breast cancer by time-resolved spectroscopy.
Materials and Methods: We irradiated a pulsed laser of 760, 800, and 830 nm wave-length lights at multiple sites of both breasts
including the site just above the cancer and detected the light transmitted through the breast with TRS-20SH (Hamamatsu
Photonics K.K.). Absorption coefficient (a), reduced scattering coefficient (s), total hemoglobin (tHb), and oxygen saturation
(SO2) of the breast were calculated by photon diffusion equation. The clinical trial started in January 2007. A total of one hundred
fifty-two breast cancer patients participated in the trial and written informed consent were obtained from all of the patients.
Results: The absorption coefficient (a) in 760, 800 and 830 nm wave-length of breast cancer tissue was significantly high
compared with contra-lateral normal breast (760nm:cancer;0.078 normal;0.063 p<0.001. 800nm cancer:0.071, normal;0.050,
p<0.001. 830nm: cancer;0.084, normal;0.063, p<0.001).
The absorption coefficient and reduced scattering coefficient between breast cancer tissue and contra-lateral normal breast
a (/cm)
s'(/cm)
cancer
normal breast
cancer
normal breast
760nm
0.078
0.063
9.58
9.71
800nm
0.071
0.05
9.23
9.4
830nm
0.084
0.063
9.07
9.22
There was no difference in reduced scattering coefficient (s) between breast cancer tissue and contra-lateral normal
breast(760nm: cancer;9.58, normal;9.71, 800nm: cancer:9.23, normal;9.40. 830nm; cancer;9.07, normal;9.22). The tHb of breast
cancer tissue was significantly high, compared with normal breast (cancer:32.314.6, normal breast;22.08.6, p=0.001). There
was no difference in oxygen saturation (SO2) between breast cancer tissue and contra-lateral normal breast (cancer:73.24.3,
normal breast;73.65.9, p=0.31).
The tHb and SO2 between breast cancer tissue and contra-lateral normal breast
cancer
normal breast
p-value
tHb(M)
32.314.6
22.08.6
p<0.001
SO2(%)
73.24.3
73.65.9
p=0.32
Conclusion: Absorption coefficient (a) and tHb increased in breast cancer, whereas reduced scattering coefficient (s) and
oxygen saturation did not.

Title: Visualisation of histologic proven breast cancer on the MAMMI-PET: A dedicated PET for hanging breast imaging
Suzana C Teixeira1, Jos Ferrer Rebolleda2, Bas B Koolen1, Ral Snches Jurado2, Marcel P Stokkel1, Mara del Puig Czar
Santiago2, Emiel J Th Rutgers1 and Renato A Valds-Olmos1. 1Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital,
Amsterdam, Noord-Holland, Netherlands and 2ERESA, General University Hospital of Valencia, Valencia, Spain.
Body: Aim: The MAMMIPET, a highresolution fullring system for dedicated hanging breast imaging was originally developed
in the context of a EUfounded project to improve the detection of breast cancer. The aim of the present study was to evaluate
the performance of the MAMMIPET device in patients with at least one histologic confirmed primary breast cancer lesion (or
index lesion), scanned in two European centres. All patients were included in the study after being scheduled to receive
preoperative chemotherapy (NAC) or radiotherapy.
Material and methods:
From March 2011 to March 2014, we included 234 female patients (mean age 52 y, range 2482y) with histologically confirmed
breast cancer. All patients were scanned with the MAMMIPET (Oncovision, Valencia, Spain) after giving informed consent.
Scans were acquired 110 min after a dose of a mean dose of 197.12 MBq18FFDG. In both centers the acquisitions, the
reconstruction of the images and the data collection were performed using similar standardized methods. We tested the relation
between visualization of the primary tumor and possible additional lesions on the MAMMIPET as well as the influence of various
variables; including age, weight, breast cancer subtypes and receptor status, breast length, maximal tumor diameter and affected
breast quadrants.
Results: A total of 236 breasts were imaged and 211 (98.4%) of the index lesions (diameter 5170 mm, mean 32 mm) were
located within the MAMMIPET scanning range. Of all index lesions within the scanning range 1.4% was not FDG avid on the
MAMMIPET images. Lesions that were FDG-avid were either clearly (86.3%) or moderately (12.3%) visible. The overall
MAMMIPET sensitivity increased from 88.6% to 98.6% after exclusion of lesions outside the scanning range. No significant
differences in lesion visibility were found due to breast cancer subtypes or breast quadrant location. Of the 35 index lesions
touching the pectoral muscle 62.9% reached into the scanning range. A total of 41 additional FDG-avid lesions were detected, not
categorized as an index tumor.
Conclusions:
The MAMMIPET missed only a small percentage of malignant lesions located within the scanning range of the device. Lesions
near the pectoral muscle were the subgroup less often visualized. No significant influence on the visualization of the FDG avid
lesions was seen due to tumor subgroups, hormone receptor status, and breast quadrant location or tumor size.

Title: The uptake of 18F-fluorodeoxyglucose (18F-FDG) by normal breast tissues as measured by a dedicated breast positron
emission tomography (PET) scanner: A preliminary study
Sze Yiun Teo1, Jasper Chaal2, Jung Ah Lee1 and Michel Herranz3. 1KK Women's and Children's Hospital, Singapore; 2Agency for
Science, Technology and Research, Singapore and 3Complexo Hospitalario Universitario Santiago de Compostela, Spain.
Body: Introduction
Breast positron emission tomography (PET ) is a molecular imaging method which identifies breast cancers with high sensitivity
and specificity. Dedicated breast PET scanners have an improved spatial resolution compared to whole body PET scanners, and
are able to differentiate between glandular and fat tissues within the breast. The aim of this study is to determine how normal
breast tissues take up 18F-FDG with time.
Method
In this IRB-approved prospective study, patients with a newly diagnosed invasive breast cancer who desired breast conservation
surgery were recruited. Participants underwent bilateral breast PET scans using a two-ring dedicated breast PET scanner
(MAMMI breast PET, Oncovision, GEM Imaging S.A.) after receiving a standard 18F-FDG dose of 5 mCu. After an uptake phase
of 60 minutes, all patients were scanned at 3 time points: at 60 minutes, 90 minutes and 120 minutes post-injection.
Reconstructed images obtained on the MAMMI breast PET scanner were then read and processed using the OsiriX software. For
each scan performed at each time point, a series of continuous axial images through the central portion of the breast containing
the nipple as a landmark was identified. Two elliptical 1cm2 regions of interest (ROI) were drawn, one on the central portion of the
glandular parenchyma and a second on the fat. This was propagated through the included slices, after ensuring that the ROI lay
entirely within the desired component of the breast in the same region for each breast. In the ipsilateral breast, the quadrant
opposite the index tumor was used to avoid contamination. The mean SUV values for breast parenchyma and breast fat were
measured, compared to the mammographic density of the breast tissues and tabulated against time.
Results
9 patients participated in the study, yielding a total of 54 breast PET scans for analysis (bilateral scans for each patient at 3 time
points). The mean SUV measurements for breast parenchyma ranged from 0.7 to 2.8 and the mean SUV measurements for
breast fat ranged from 0.1 to 0.8. The mean SUV of breast parenchyma was approximately 3 to 4 times that of the mean SUV of
breast fat at each time point. There was a gradual increase in the mean SUV of breast parenchyma and breast fat with time,
although the increase appeared more pronounced with breast parenchyma. Patients with dense breast tissues tended to have a
higher parenchymal mean SUV value.
Conclusion
Normal breast parenchyma demonstrates a greater uptake of 18F-FDG compared to breast fat tissues at a ratio of approximately
3-4 to 1. There is accumulation of 18F-FDG with time in both parenchymal and fatty tissues within the breast, more so with the
former. Knowledge of the background SUV measurement of parenchymal tissues is important as this affects the SUV
measurement of the underlying breast cancer. Increasing accumulation of FDG with time may decrease the contrast resolution of
a cancer, especially in patients with dense breast parenchyma.


Title: Comparison of ESR1, PGR, HER2 and KI67 expression by central IHC and MammaTyper RT-qPCR kit in the FinHer trial
Ralph M Wirtz1, Pirkko-Liisa Kellokumpu-Lehtinen2, Jorma Isola3, Vesa Kataja4, Petri Bono5, Taina Turpeenniemi-Hujanen6,
Sirkuu Jyrkki7, Harri Sitho8, Sebastian Eidt9, Ugur Sahin10 and Heikki Joensuu5. 1STRATIFYER Molecular Pathology GmbH,
Cologne, Germany; 2Tampere University Hospital, Tampere, Finland; 3Laboratory of Cancer Biology, Institute of Medical
Technology, Tampere, Finland; 4Cancer Center, Kuopio University Hospital, Kuopio, Finland; 5Helsinki University Central
Hospital, Helsinki, Finland; 6Oulu University Hospital, Oulu, Finland; 7Turku University Hospital, Turku, Finland; 8Laboratory of
Molecular Oncology, Biomedicum, Helsinki, Finland; 9BioNTech Diagnostics GmbH, Mainz, Germany and 10Institute of Pathology
at the St-Elisabeth-Hospital, Cologne, Germany.
Body: Background: Subtyping of breast cancer has become an integral part of standard evaluation of breast cancer patients1.
However, assaying of ER, PgR and Her-2/neu by immunohistochemistry (IHC) carries an up to 20% risk of erroneous results2,3.
Moreover, reliable assessment of Ki-67 by IHC in grade 2 breast cancer is challenging due to high intra- and interobserver
variations4,5. Interobserver concordance for ESR1, PGR, HER2, and KI67 determination on mRNA level using MammaTyper
reagents were 96.8%, 97.2%, 100%, 97.6%, respectively, based on predefined cutoffs (Laible et al., abstract submitted). Here we
tested the clinical concordance and prognostic value of MammaTyper determinations in the FinHer trial patient population.
Methods: RNA was extracted from formalin-fixed paraffin-embedded (FFPE) breast tumor tissue, and candidate gene expression
was analyzed using the RNXtract IVD and MammaTyper IVD kits (BioNTech Diagnostics GmbH, Mainz) from 791 patients
who participated in the FinHer trial. RNA levels of ESR1, PGR, HER2, and KI67 mRNA expression were measured using
RT-qPCR,normalized according to the 40-DDCT method and compared with IHC or CISH results. Associations with distant
disease-free survival (DDFS) and overall survival (OS) were assessed using the log-rank test.
Results: ESR1, PGR, HER2, and KI67 mRNA levels exhibited strong correlations with the respective clinical assays (each
p-value <0.0001). The concordance rate of the mRNA assay and the clinical asssay was 92% for ESR1, 92% for HER2, 83% for
PGR, and 68% for KI67 RNA. Using predefined cut-off values for ESR1, PGR, and KI67, the mRNA levels were prognostic for
DDFS (p=0.002, p=0.005, and p=0.0005) and OS (p<0.0001, p=0.0001, and p=0.0024, respectively), whereas HER2 mRNA
expression was not (p=0.17 and 0.11, respectively). Unexpectedly, the HER2 mRNA expression distribution of the ER-negative
cancers was bimodal with little overlap between the HER2-low and HER2-high subsets, while in ER-positive cancers HER2
mRNA distribution was almost unimodal and in between the two subpopulations of ER-negative cancers. When different cut-offs
were used for ER-positive and ER-negative cancers, tumor HER2 mRNA was also significantly associated with DDFS (p=0.031)
and OS (p=0.018). Interestingly, 17% of ESR1 mRNA-negative and HER2 mRNA-negative patients exhibited high mRNA
expression of PGR, and such patients had high 5-yr DDFS and OS (>95%).
Conclusions: Determination of ESR1, PGR, HER2, and KI67 without macrodissection of routine whole tissue FFPE specimens
results in highly concordant results when compared to clinical assays. A significant minority of HER2 negative breast cancers
expressed PGR mRNA despite low ESR1 mRNA levels and had superior outcome. HER2 mRNA levels differed substantially
between ER-positive and ER negative tumors. This may explain why HER2 determination using a single cut-off for HER2 mRNA
with the Oncotype DX assay frequently results in a false negative finding6. The MammaTyper-defined ESR1, PGR, HER2, and
KI67 expression showed strong correlations with the corresponding clinical assays and survival.
1) Goldhirsch et al., Annals of Oncology 2013
2) Hammond et al., JCO 2010
3) Wolff et al., JCO 2007
4) Varga et al., PloS 2012
5) Polley et al., JNCI 2013.

Title: Central testing of ER, PR, HER2, and Ki67, routinely analyzed at 27 pathology departments in Sweden Potential
consequences for the choice of adjuvant therapy
Maria Ekholm1,2, Dorthe Grabau2,3, Per-Ola Bendahl2, Gran Elmberger4, Hans Olsson5, Leila Russo6, Guiseppe Viale6 and
Mrten Fern2. 1Ryhov County Hospital, Jnkping, Sweden; 2Division of Oncology and Pathology, Lund University, Lund,
Sweden; 3Skne University Hospital, Lund, Lund, Sweden; 4rebro University Hospital, rebro, Sweden; 5Linkping University
Hospital, Linkping, Sweden and 6European Institute of Oncology, University of Milan, Milan, Italy.
Body: Background
Immunohistochemical (IHC) techniques are still gold standard for measurement of estrogen receptors (ER), progesterone
receptors (PR), the proliferation marker Ki67, and human epidermal growth factor receptor 2 (HER2) (confirmation with in situ
hybridization for 2+) in routine breast cancer pathology. These biomarkers are also included in the St Gallen 2013 four-marker
panel, as a surrogate for the intrinsic subtype classification; Luminal A-like, Luminal B-like (HER2-), Luminal B-like (HER2+),
HER2+ (non-luminal), and Triple negative. Correct analyses are of outmost importance since these biomarkers are underlying
treatment decisions in breast cancer. It is mandatory for departments performing breast cancer pathology to participate in quality
assurance programs.
Aims
In 2013, SweQA-breast cancer (Swedish Quality Assurance) performed a study to investigate the agreement of ER-, PR-, Ki67-,
and HER2-scoring between pathology departments in Sweden and a central reference laboratory (European Institute of Oncology
(EIO), Milan, Italy). To explore the clinical consequences, the aim was also to investigate to what extent the central testing would
potentially have altered the adjuvant medical therapy.
Methods
Twenty-seven of the 31 pathology departments in Sweden participated by collecting the first breast carcinoma diagnosed after the
15th each month during 2012, except for July and December (n=270). Paraffin embedded tumor tissue were collected and sent to
EIO, where new sections, stainings, and evaluations were performed. These results were compared with the results from the
routine analyzes performed in Sweden. Expression scores of ER, PR, Ki67, and HER2 were dichotomized according to
predefined thresholds from the St Gallen consensus statement of 2013 (Goldhirsch et al. Ann Oncol 24(9);2206-13;2013).
Results
The pairwise agreement figures for ER, PR, and Ki67 were; 99% (kappa-value ()=0.95), 95% (=0.84), and 85% (=0.70)
respectively. Eight of the PR discrepant cases were locally positive but centrally negative, whereas six showed the opposite
pattern. In 34 of the 39 discrepant cases for Ki67, one or both assessments were located near cut-off (15-25%). When
categorizing HER2 as 0/1+ vs. 2+/3+, agreement was 86% (=0.65). The HER2-IHC 2+/3+ cases in Sweden were further
analyzed with in situ hybridization as part of clinical routine, and when these results were used to obtain the final HER2 status,
only 1 discrepant case of 265 evaluable cases was observed. When using the St Gallen 2013 surrogate definition for the five
intrinsic subtypes, the overall pairwise agreement between Sweden and Italy, for all tumors with complete data (n=256), was 86%
(=0.78).
Conclusions
The agreement was very good (>0.80) or good ( 0.61-0.80) for all four biomarkers (ER, PR, Ki67, and HER2). Almost 90% of
the discrepant cases for Ki67 were explained by values near cut-off. When strictly applying the St Gallen 2013 surrogate definition
of the five intrinsic subtypes, without regard to TNM stage, the results of the central testing would potentially have altered the
medical adjuvant treatment for 37 of the 256 patients (14%). For 28 of these 37 patients, the discordance was explained by Ki67.

Title: A comparison of the hot spot and the average cancer cell counting methods and the optimal cut-off point of the Ki-67 index
for luminal type breast cancer with or without recurrence A case control study for prognostic factors
Nobuyuki Arima1, Reiki Nishimura1, Tomofumi Osako1, Yasuyuki Nishiyama1, Yasuhiro Okumura1, Masahiro Nakano1, Mamiko
Fujisue1, Rumiko Tashima1 and Yasuo Toyozumi1. 1Kumamoto City Hospital, Kumamoto, Japan.
Body: Background: Uncontrolled proliferation is a key factor of malignant tumors. The Ki-67 index is known to be a significant
prognostic factor in terms of disease-free and overall survival in breast cancer. However, there are several limitations in the use of
this biomarker, the biggest being a difference in opinion among breast cancer specialist on how to create an international
standard for the Ki-67 index. Therefore, the aim of this case control study was to investigate the most suitable area to count and
to determine the optimal cut-off point of the Ki-67 index for a more accurate prognosis.
Thirty recurrent cases (< 5 years after initial treatment) were selected among hormone receptor (HR)-positive/HER2-negative
breast cancer patients. As a control, 90 non-recurrent cases (>5 years after initial treatment) were randomly selected by allotting
3 controls to each recurrent case based on the following predetermined criteria; age, nodal status and tumor size. All patients
were treated with adjuvant endocrine therapy alone. Both the hot spot and the average area of the tumor were evaluated on a
Ki-67 immunostaining slide and then photographs were taken. The Ki-67 index was automatically scored using the CountCell
(Seiko Tech., Fukuoka, Japan). Moreover, the difference in the Ki-67 index values (Ki-67) between the hot spot and the average
area were calculated. The Chi-square and Fishers exact tests were used for inter-group comparison. Paired t-test and Wilcoxons
(non-parametric) test were used to compare the mean value for the Ki-67 index values. Logistic regression analysis was used to
calculate the odds ratio of the Ki-67 index related to recurrence.
Results:
1) A higher Ki-67 index value at the hot spot was more significantly associated with recurrence than the average area,
and the Ki-67 index value of 20% at the hot spot was the most suitable cut-off point for predicting recurrence.
Irrespective of the area counted, the Ki-67 index value was significantly higher in all of the recurrent cases (p<0.0001). The next
step was to determine the most suitable cut-off point for the Ki-67 index. Out of all the cut-off points used in this study, the most
significant difference was found in the values 20, 25 and 30% at only the hot spot (each p<0.0001). Furthermore, logistic
regression analysis demonstrated that the highest odds ratio was 20% at the hot spot.
2. The Ki-67 index value was significantly correlated with recurrence and the Ki-67 index value of 10% was the most
suitable cut-off point.
The Ki-67 index values ranged from 2 to 54% (median: 10%) and significantly correlated with recurrence and the Ki-67 value at
hot spot (p<0.0001). Logistic regression analysis revealed that a Ki-67 index cut-off point of 10%.
Conclusion:
A higher Ki-67 index value at the hot spot was strongly correlated with recurrence, and the most suitable cut-off point was 20%.
Ki-67 index value also significantly correlated with recurrence, and the most suitable cut-off point was 10%. Finally, the hot spot
counting method is strongly related to tumor biology and prognosis in HR-positive/HER2-negative breast cancer.

Title: Tumor heterogeneity impairs robustness of Ki67 scoring in breast cancer
Matthijs V Nijenhuis1, Emilie Groen1, Tim J Dekker2, Caroline A Drukker1, J Sanders1, V T Smit2, S Linn1, E J Rutgers1 and J
Wesseling1. 1Antoni van Leeuwenhoek and 2Leiden University Medical Center, Leiden, Netherlands.
Body: Background
A study by Cheang et al. reported that for ER-positive, HER2-negative breast cancer a Ki67 score 14% distinguishes Luminal B
from Luminal A tumors. The (neo)adjuvant treatment of these Luminal B tumors consists of endocrine treatment and
chemotherapy while chemotherapy can be omitted for Luminal A tumors. For the determination of these subtypes, resection
specimens and particularly core biopsies in the neoadjuvant setting are used. Since Ki67 scoring is advocated by some to be a
marker in deciding if systemic chemotherapy will be given or not, its assessment should be highly reproducible and reliable. I.e,
Ki67 assessment should not vary significantly between different laboratories, but also not within the tumor. To test this, we
analyzed (1) interlaboratory agreement using whole slides and (2) agreement within a particular tumor using tissue microarrays
(TMA) as well as whole slides from surgical specimens.
The first 100 patients from the microarRAy prognoSTics-in-breast-cancER (RASTER) study (n=427) were selected. From each
patient clinicopathological characteristics and tumor blocks were available. From each tumor block 2 whole slides were cut for
staining in laboratory 1 and 2. Also six single cores were taken to construct 2 tissue microarrays (core 1-3 and core 4-6). Intrinsic
subtypes were defined as follows: Luminal A, Luminal B/HER2-, Luminal B/HER2+, HER2-overexpressing, and Basal-like.
Experienced pathologists (EG, JS, VTHMBS, JW) performed the scoring.
Results
There were 99 whole slides suitable for analysis. Substantial agreement ( = 0.723, P<0.001) was demonstrated in discrimination
for Ki67 low and high between whole slides stained in laboratory 1 and laboratory 2 with a discordance rate of 13.1% (Table 1).
Moderate agreement ( = 0.595, = 0.584, P<0.001) was demonstrated in discrimination of Ki67 low and high between whole
slide and core 1-3 and core 4-6 with a discordance rate of 20.3% and 20.8% respectively (Table 2). Substantial agreement ( =
0.666) was demonstrated for Ki67 expression between core 1-3 and core 4-6.
Table 1. Discordance between laboratories for whole slide Ki67 result
Laboratory 2
Laboratory 1
<14%
14%
Discordance rate %
Kappa
Sign.
<14%
56
13.1
0.723
<0.001
14%
13
30
Table 2. Discordance between whole slide and core biopsies 1-3 and 4-6 for Ki67 result
Whole slide
<14%
Core 1-3
Core 4-6
Discordance rate %
14%
<14%
33
14%
30
<14%
31
14%
30
Kappa
Sign.
20.3
0.595
<0.001
20.8
0.584
<0.001
Conclusion
Tumor heterogeneity results in substantial variation of Ki67 scores between TMA cores and whole slides that may result in
considerable differences in distinguishing luminal A from luminal B ER-positive, HER2-negative tumors. This may have far
reaching consequences for the choice for (neo)adjuvant treatment.

Title: Assessment of HER2 status in invasive breast cancer with increased centromere 17 copy number by fluorescence in situ
hybridization
Min Hye Jang1, Eun Joo Kim1, Hyun Jeong Kim1 and So Yeon Park1,2. 1Seoul National University Bundang Hospital, Korea and
2
Seoul National University College of Medicine, Seoul, Korea.
Body: Background: A subset of breast cancers can show increased copy numbers of chromosome 17 centromere (CEP17) by in
situ hybridization (ISH). However, recent studies have revealed that true polysomy 17 is a rare event in breast cancer, and
increased copy number of CEP17 represents amplification or coly number gain in the centeromeric region. In this situation, the
utility of CEP17 in ISH is very limited and thus, alternative methods for precise assessment of HER2 status are necessary. ISH
using probes of other genes on chromosome 17 as additional reference genes has been proposed by 2013 ASCO/CAP guideline
and several previous studies. In this study, we applied this method to breast cancers with increased CEP17 copy number (2.6),
and compared it with conventional method based on ASCO/CAP guideline 2007.
Methods: After reviewing all HER2 fluorescence in situ hybridization (FISH) reports based on HER2/CEP17 ratio, documented
from June 2004 to December 2011 at our institution, we identified 300 cases (29.6%) with 2.6 CEP17 copy number from 1013
breast cancers. We performed FISH with probes for RARA, SMS, TP53 genes on 243 breast cancers with available tissue blacks,
using tissue microarrays. If one or more genes showed < 2.6 copy number, the largest number of them was chosen for alternative
to CEP17 and re-graded the HER2 status based on HER2: alternative gene ratio.
Results: Of 243 breast cancers with 2.6 CEP17 copy number, 2 cases showed 2.6 copy numbers in all RARA, SMS and TP53
genes. Of 151 breast cancers classified as non-amplified based on HER2:CEP17 ratio, 42 (27.8%) were re-graded as amplified
after additional FISH studies, and 7 of 8 cases which had been classified as equivocal were re-graded as amplified. Of the 47
cases with mean HER2 copy number of 4 to 6, 30 (63.8%) were upgraded from non-amplified to amplified, and 5 (10.6%) were
upgraded from equivocal to amplified. Two (1.9%) of 75 cases with < 4 mean HER2 copy number were upgraded from
non-amplified to amplified, and 24 (23.8%) were changed from non-amplified to equivocal. After re-grading, equivocal cases were
significantly increased, compared to the original reports (14.1% vs. 3.3%). And the concordance between FISH and
immunohistochemistry became poorer (kappa value: 0.390 vs. 0.624).
Conclusion: Using additional reference genes can be an effective way for accurate HER2 evaluation in breast cancer with
increased CEP17 copy number. However, it still has some limitations. It can lead to over-grading of HER2 status and increase of
the equivocal cases, when tumor has loss of reference genes. Additional study for searching most stable genes that barely show
copy number alteration is needed to simplify the procedure and to increase accuracy.

Title: Digital quantification of estrogen receptor expression in normal breast in post-menopausal women with breast cancer and
association with tumor subtypes
H Evin Gulbahce1,2, Cindy Blair3, Carol Sweeney4, Rachel Factor1,2 and Mohamed Salama1,2. 1University of Utah School of
Medicine, Salt Lake City, UT; 2ARUP Laboratories, Salt Lake City, UT; 3University of Minnesota, Minneapolis, MN and 4University
of Utah School of Medicine, Salt Lake City, UT.
Body: Background: ER expression in normal breast epithelium (NBR) is higher in women with a history of breast cancer (BC)
compared to controls. In adjuvant setting, metanalysis showed effective Tamoxifen treatment was restricted to ER positive
cancers. However, it is also known prophylactic oophorectomy (a form of estrogen suppression) significantly reduces the
incidence of BC in BRCA1 carriers. This is in contrast with the 80% rate of ER negative tumors in BRCA1 patients. The aim of this
study is to quantify ER expression in NBR away from tumor in women with BC and to correlate it with BC subtypes.
Methods: 204 consecutive patients were identified for whom NBR away from tumor was available. 27 reduction mammoplasty
(RM) tissues from women with no history of BC were used as controls. Tissue microarrays were constructed and slides were
stained with ER and scanned using Aperio XT Scan Scope. Normal terminal duct lobular epithelium was manually circled on
scanned images and annotations were recorded in separate digital layers. ER staining was quantitated in marked areas of the
electronic image using an optimized scoring nuclear IHC algorithm (Aperio technologies, Inc., Vista, CA). Clinical information and
tumor characteristics (menopausal status, ER, HER2 expression, grade, size, number of positive nodes, stage) were recorded
based on pathology reports and tumor registry data.
Results: The mean ER positivity in NBR was 16 12.4 % (range: 0-5-5.7%) for all patients with BC, 20.813.9% for
postmenopausal (n=74 ) and 13.7 10.9% for peri+ premenopausal (n=170) subgroups. ER positivity was higher in patients with
BC compared to those undergoing RM with no history of BC. ER positivity in NBR did not vary by tumor size, positive lymph
nodes status, tumor grade, or stage in post-menopausal, and pre + perimenopausal women. Older age at diagnosis was
significantly associated (p<0.0001) with ER in NBR. In postmenopausal women ER expression in NBR was significantly higher
in patients with ER negative or triple negative tumors.
Estrogen receptor expression in normal terminal duct lobular units of post-menopausal women with breast cancer in relation to
tumor subtypes.
Postmenopausal (N=74)
N
Mean % (SD)
Tumor ER Status
P Value
0.05
Negative
14
27.4 (13.9)
Positive
60
19.3 (13.6)
Tumor HER2 Status
0.34
Negative
57
21.8 (14.9)
Positive
17.7 (5.7)
Tumor Triple Negative
0.05
No
53
19.7 (13.6)
Yes
12
28.5 (14.7)
HER2 status was not known in some cases and was excluded from statistical calculations.
Conclusion: This computer assisted image analysis study confirms ER expression in NBR increases with age and menopausal
status in women with BC. We report, for the first time, a significant association between ER expression in normal breast
epithelium with ER negative and triple negative cancers in post-menopausal women. Our study suggests that ER expression in
normal epithelium may play a role in development of hormone receptor negative breast cancers.

Title: Effect of the new 2013 ASCO / CAP guidelines on HER2 reporting
H Evin Gulbahce1,2, Rachel E Factor1,2, Erinn Downs-Kelly1,2, Margaret Coppin2 and Katherine B Geiersbach1,2. 1University of Utah
School of Medicine, Salt Lake City, UT and 2ARUP Laboratories, Salt Lake City, UT.
Body: Introduction: In 2013 ASCO/CAP published new guidelines for HER2 testing to decrease false negatives. We
retrospectively classified cases submitted to a national reference laboratory for HER2 testing using the new guidelines in order to
1) see the overall effect of the 2013 vs 2007 guidelines and 2) predict shifts in interpretation of different HER2 testing methods in
the future.
M&M: Our laboratory offers HER2 immunohistochemistry (IHC) by HercepTest (Dako) or 4B5 (Ventana) and HER2 dual probe
FISH (Abbott Molecular). HER2 IHC and/or FISH tests performed between 1/2010-8/2013, originally scored with the 2007
guidelines, were reclassified with the 2013 guidelines. For IHC, the guideline scores of 0, 1+, 2+ and 3+ were recorded along with
intensity (weak, moderate, strong) and % circumferential staining. For FISH, the HER2 and CEP 17 copy number per cell, and
HER2:CEP 17 ratio were recorded.
Results: 2358 samples submitted for HER2 FISH were available for review. Using the 2007 guidelines, 246 (10.4%) were
amplified (Amp), 62 (2.7%) equivocal (Eq), and 2050 (86.9%) were non-amplified (NA). 29/62 (46.8%) FISH Eq cases had
HER2:CEP17 ratio between 2.0-2.2, 1 (1.6%) had HER2 copy number >6 per cell. Therefore 30/2358 (1.3%) of all FISH tests
previously classified as Eq would be reclassified as Amp following new guidelines increasing the overall FISH Amp rate by 12.5%
to 11.7%. Over the same time period, 4043 HER2 IHC tests were performed using old guidelines 1097/4043 (27.1%) of which
were Eq (2+). 731/1097 (66.6%) had reflex FISH requested on 2+ IHC cases. 24/731 (3.3%) did not have an iterpretable FISH
result. Using the 2007 guidelines, 49 (6.9%) of the 2+ cases were FISH Amp, 23 (3.3%) remained Eq, 635 (89.8%) were NA.
Using the new guidelines, an additional 8 (1.1%) of these 707 reflex FISH would be classified as Amp (7 cases due to
HER2:CEP17 ratio 2.0 to 2.2 and one case for HER2 copy number >6). Also 8 NA cases would be reclassified as Eq due to a
HER2 copy number between 4 and 5.9. With the new guidelines, the overall FISH Amp rate identified from reflex 2+ IHC is 8.1%.
507/707 (71.7%) of HER2 Eq (2+) cases reflexed to FISH had 10-30% circumferential staining (8 strong; 136 moderate; 363
weak). HER2 Amp rates for these are shown in Table 1.
HER2 Amplification Status of IHC Equivocal (2+) Cases by Percent Membrane Staining
IHC 2+
FISH Amplified (2013 Guidelines) (%)
2+ Cases (All)
57/707 (8.1%)
2+ Cases, 10-30% Membrane Staining
37/507 (7.3%)
-------Strong
-------1/8 (12.5%)
-------Moderate
-------16/136 (11.8%)
-------Weak
-------20/363 (5.5%)
Conclusion: 2013 guidelines will identify a higher % (1.1% for FISH) of eligible patients for targeted therapy. In this series, 2+ IHC
cases with reflex FISH testing also had higher Amp rates by 2013 guidelines (8.1% vs 6.9%). Evaluating these rates over time will
help assess the effect of the change in the scoring criteria. Amp rates for Eq (2+) IHC cases with 10-30% strong and moderate
membrane staining are similar, although only the former group is classified as positive (3+) following new guidelines and would
not be retested by FISH. Cases with less than 30% membrane staining may represent a biological spectrum that is not well
represented by current IHC testing guidelines, and FISH confirmation on these cases may be justified.

Title: Incidence of misattributed specimen provenance among surgical breast biopsies
Arthur G Lerner1, Arla L Bush2, Andrew S Kenler3, David D Dorfman4, Travis A Morgan2, Beth Boyd1, William E Burak1 and
Richard E Fine1. 1Advanced Breast Technologies Consulting LLC, Palm City, FL; 2Strand Diagnostics LLC, Indianapolis, IN;
3
Bridgeport Hospital, Yale New Haven Health, Bridgeport, CT and 4Zwanger-Pesiri Radiology, Lindenhurst, NY.
Body: Background:
The medical literature reports on the diagnostic challenges posed by tissue contamination and transposition among surgical
biopsy specimens. These specimen provenance complications (SPCs) can lead to a misdiagnosis of cancer, resulting in
unnecessary surgery and a potential delayed diagnosis. The histopathology process involves many manual steps during which
specimens must be estranged from their identification, and provenance errors are often invisible absent DNA analysis. Prostate
biopsy is the setting in which specimen provenance has been studied, with complication rates reported in over 0.9% of positive
diagnoses despite best efforts to minimize errors. Because the processing workflow is virtually identical for histopathology
specimens of all types, it is expected that error rates among breast biopsy specimens are similar to prostate.
Methods:
We analyzed over 4200 patients diagnosed with breast cancer between February 2011 and April 2014. All biopsies were
collected using a uniform best-practice protocol including forensic chain of custody principles, bar-coding of specimen containers,
and collection of the patients reference DNA sample via buccal swab. After a pathologic diagnosis of breast cancer a portion of
the diagnostic specimen was forwarded to an independent DNA laboratory (Strand Diagnostics, Indianapolis, IN) where genetic
STR profiles were compared to the patients reference DNA to rule out the presence of undetected SPCs prior to therapy.
Results:
3,545 breast cancer cases from 7 practices contributing 100 or more cases each were examined (Fig.1). DNA testing revealed
occult provenance complications in 16 cases (0.45%), of which 6 (0.17%) were a complete transposition with another patient and
10 (0.28%) reflected contamination of the specimen by tissue from one or more unidentified individuals. Four (57%) of the
practices experienced at least one provenance error during the study period, with the highest error rate being 1.41% at one
practice. Pathology was performed by 14 different laboratories, 6 (43%) of which were implicated in occult SPCs (Fig. 2). Finally,
patients seen by 8 (13%) of the 61 physicians performing surgical biopsies in the cohort were subjects of occult specimen
provenance errors.
Conclusions:
These data suggest that the incidence of SPCs among breast biopsies is comparable to that reported for prostate biopsies. The
errors are distributed broadly across laboratories, practices, and physicians. Due to the potential clinical consequences for
patients with undetected SPCs, and the medical malpractice implications, further study of the nature and economics of
provenance complications in the breast biopsy setting is warranted.
Fig 1
Type I
Type II
Cases
Errors
Errors
SPC Rate
Practice A
2,205
0.54%
Practice B
617
0.00%
Practice C
173
0.00%
Practice D
162
Practice E
142
Practice F
128
Practice G
118
Total
3,545
1
2
0.62%
1.41%
0.78%
0.00%
10
0.45%
Figure 1. SPCs by Practice

Type I=Transpositions,
Type II=Contaminations
Fig 2
Type I
Type II
Cases
Errors
Errors
SPC Rate
Lab A
1,435
0.42%
Lab B
967
0.31%
Lab C
430
Lab D
260
Lab E
182
Lab F
84
Lab G
78
Lab H
71
Lab I
20
Lab J
Lab K
0.00%
Lab L
0.00%
Lab M
0.00%
Lab N
0.00%
Total
3,545
Figure 2. SPCs by Pathology Laboratory

Type I=Transpositions,
Type II=Contaminations
0.00%
2
0.77%
0.00%
1
1.19%
0.00%
2.82%
0.00%
2
10
22.22%
0.45%

Title: Ki67-labelling index of luminal breast cancers: Agreement of core biopsy and surgical specimen results
Cornelia M Focke1 and Thomas Decker1. 1Dietrich Bonhoeffer Medical Center, Neubrandenburg, Germany.
Body: Objective: To highlight the agreement between Ki67-Labelling indices (LIs) of core biopsies (CB) and related surgical
specimens (SSP) in luminal breast cancers (BCs) adopting the recommendations of the St Gallen Consensus Conference 2013
and the International Ki67 In Breast Cancer Working Group.
Methods: We investigated CBs and the related SSPs of 220 luminal Her2 negative BCs. To assess the Ki67-LI a total of 510
tumour cells including hot spot, cold spot and an area of intermediate proliferation was counted by one observer. Every positive
nucleus was counted, independent of the staining intensity. A Ki67-LI cut-off <20% as recommended by the St Gallen Consensus
was used to define luminal A BCs. Agreement between CB and SSP, rates of under- and overestimation, and positive predictive
values (PPV) of CB based subtyping (luminal A vs. B) were calculated.
Results: In SSP 140/220 BCs (64%) were luminal A, 80 (36%) were luminal B. The agreement between CB and OP was 77%
(169/220). Overestimation of proliferation in CB occurred in 10 (4%) and underestimation in 41 (19%) cases. The PPVs of CB
based Ki67-LI for luminal A and luminal B were 0.76 (130/171) and 0.79 (39/49), respectively.
Conclusion: The agreement of CB and SSP based subtyping of luminal BCs is generally high. However, underestimation of
proliferation in CB may result in subtype misclassification in almost one fifth of luminal Her2 negative luminal tumours.

Title: Factors influencing agreement of Ki67 labeling index between core needle biopsy and surgical resection specimens
Zsuzsanna Bago-Horvath1, Fabian Roessler1, Philipp Wimmer1, Martina Mittlboeck2, Nicolas Kozakowski1, Katja Pinker-Domenig3,
Rupert Bartsch4, Peter Dubsky5, Martin Filipits6 and Margaretha Rudas1. 1Clinical Institute of Pathology, Comprehensive Cancer
Center, Medical University of Vienna, Vienna, Austria; 2Informatics and Intelligent Systems, Section for Clinical Biometrics,
Medical University of Vienna, Vienna, Austria; 3Division of Molecular and Gender Imaging, Medical University of Vienna, Vienna,
Austria; 4Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria; 5Comprehensive Cancer Center, Medical
University Vienna, Vienna, Austria and 6Institute of Cancer Research, Comprehensive Cancer Center, Medical Univeristy of
Vienna, Vienna, Austria.
Body: Reliable determination of Ki67 labeling index (Ki67 LI) on core needle biopsy specimens (CNB) is essential for determining
breast cancer intrinsic subtype (IST), preoperative treatment decisions and short-term treatment response during neoadjuvant
therapy. However, analyses investigating robustness of Ki67 LI assessment upon repeated tumor biopsies are scarce and results
of studies investigating agreement of KI67 LI between CNB and surgical resection (SR) specimens are conflicting. In the present
study, we analysed the role of clinical and pathological factors in influencing concordance of Ki67 LI between CNB and SR
specimens.
502 matched pairs of CNB and SR specimens of patients with invasive ER-positive, HER2-negative breast cancer were included
in our study. Ki67 LI was determined according to recent recommendations. Luminal B disease was defined by Ki67 LI > 20%
according to SR. Ki67 LI values were considered concordant by a divergence of 10% points. Agreement was calculated by
Cohens kappa. Associations with clinicopathological factors were analyzed by Chi square test and logistic regression. Factors
investigated included age, menopausal status, CNB method, BIRADS assessment category of imaging abnormality, time between
CNB and SR, extent of surgery, histological grade (including individual grading score components according to Elston and Ellis),
tumor size, lymph node stage, estrogen- (ER), progesterone- (PR) and HER2-receptor status.
A cutoff value of 20% for KI67 LI in SR demonstrated a sensitivity of 90% and a specificity of 60% for identifying luminal B breast
cancer in CNB. Upon investigating measurement agreement, we found substantial agreement of Ki67 LI between CNB and SR
specimens with a weighted kappa value of 0.837. IST assesment in CNB and SR showed only moderate concordance with a
kappa value of 0.587. Concordant diagnosis of IST in CNB and SR was significantly associated with PR expression and
histological grade (p>0.05). Agreement of Ki67 LI was higher in tumors expressing low and high levels of PR compared to tumors
with intermediate PR score. 27% and 22% of low and intermediate grade breast cancers showed discordant IST in CNB and SR,
respectively. In contrast, only 3% of high grade breast cancers differed regarding IST upon repeated measurements. Interestingly,
concordance of IST was significantly associated with all separate grading score components in CNB samples such as extent of
glandular differentiation (p=0.015), nuclear polymorphism (p=0.005) and number of mitotic figures (p>0.001).
We conclude that agreement of breast cancer IST according to Ki67 LI between CNB and SR specimens is significantly
influenced by histological tumor grade and PR status. These factors, including all three grading score components are likely to
mirror tumor heterogeneity that might compromise obtaining a CNB sample representative of the entire tumor. Our results cast a
doubt upon robustness of single CNB-driven measurements of prognostic indicators and outcome predictors in estrogen-receptor
positive breast cancer of low or intermediate histological grade. Whether molecular testing of CNB specimens improves
classification of luminal breast cancer and helps resolve diagnostic disparities remains to be determined.

Title: Triple negative breast cancer: The role of classic histological and prognostic factors on disease free survival
Ahmed Elkhanany1, Vera J Suman1, Victoria Cafourek1, Judith A Gilbert1, James N Ingle1, Fergus Couch1, Daniel W Vissscher1
and Matthew P Goetz1. 1Mayo Clinic, Rochester, MN.
Body: Background: Triple negative breast cancer (TNBC) represents 15% of all breast cancers, and is characterized by an
aggressive clinical course. While efforts are ongoing to characterize the molecular basis for variation in TNBC, there are
conflicting data regarding the impact of prognostic factors such as age and Ki-67 typically used for treatment decisions in other
breast cancer subtypes. We identified a retrospective cohort of women with ER- breast cancer with long term follow-up, and
performed central confirmation of ER, PR, HER2, Ki-67 and histological classification to assess the association of Ki-67 and
histologic subtype on clinical outcome in TNBC pateints (pts).
Methods: 9,836 women who underwent breast cancer surgery at Mayo Clinic Rochester from 1985-2005 were identified. Pts were
excluded due to: ER+ disease (7363); prior cancer (553); non-invasive disease (465); bilateral breast cancer (167); metastatic
disease [at diagnosis or within 60 days of surgery (110)]; ER- disease treated with neoadjuvant chemotherapy (121) or adjuvant
hormonal therapy (112); or tumor block exhausted (94). For all others, centralized ER/PR/HER2 identified the following for
exclusion, >1% ER staining (n=76), > 1% PR (n=14) and HER2+ (n=144) by IHC (3+) or FISH amplification. 225 cases are still
undergoing review. For the centrally confirmed pts with TNBC (n=392), Ki-67 and histological characterization (WHO subtypes)
were assessed.
Results: Patient characteristics are listed in the table. The median age for all TNBCs was 45 (range 29-88) and nearly all tumors
were grade 3 (91%) with high proliferation (63% with Ki-67 > 30%). The median follow-up was >10 years, where 139 pts had a
disease event: progression (86) or second primary (53) and 53 died without a disease event. The 10 yr DFS rate was 55.8%
(95%CI: 48.8-63.9%) among the 238 pts who did not receive adjuvant chemotherapy (AdjCT) and 58.9% (95% CI: 48.9-70.9%)
among the 102 pts administered AdjCT. In pts without adjCT, DFS was found to differ with respect to number of positive LNs (0
vs. 1-4 vs. 4+; log rank p=0.003) and medullary histology (log rank p=0.015) but not with age (< 50 vs. 50), tumor size (< 2.0 vs.
2-5 vs. 5.0 cm) or Ki-67 (0-15% vs. 15.1-30% vs. >30 %).
Conclusions: Our findings confirm the poor prognosis of TNBC. The medullary histological subtype is associated with significantly
better prognosis than other TNBC subtypes (10 year DFS rate of 78.5%; 95%CI: 59.1-99.9% in pts not treated with adjCT). After
central confirmation, neither age nor Ki-67 provide additional prognostic information.
Patient characteristics
Median Age (range)
45
(29-88)
Apocrine
28
7.1
Carcinoma, anaplastic or undiffrentiated
93
23.7
Central scar/necrotizing
34
8.7
Clear Cell
29
7.4
Infiltrating ductal ca
71
18.1
Large Cell neuroendocrine
14
3.6
Medullary and atypical medullary
43
11.0
Metaplastic ca
24
6.1
Trabecular ca
33
8.4
other histologies
23
5.9
Histology
Grade
1
0.8
31
7.9
358
91.3
0.1-2 cm
196
50
> 2.0 cm
195
49.6
not stated
0.3
251
64
1-3
82
20.9
4-9
33
8.4
10+
22
5.6
not evaluated
1.0
0-15
77
19.6
15.1-30
65
16.6
>30%
247
63.3
not done
0.5
Yes
102
26.0
No
238
60.7
unknown
52
13.3
Tumor Size
Nodal status
Ki-67
Adjuvant Chemotherapy

Title: Concordance between semiquantitative immunohistochemical assay and oncotype DX RT-PCR assay for estrogen and
progesterone receptors
Patricia Novas1, Elena Galve1, Sara Fernndez1, Maria ngeles Sala1, Alberto Arvalo1, Juan Arango1, Borja Lpez de San
Vicente1, Seila Fernndez1, Ane Zumrraga1 and Purificacin Martnez del Prado1. 1Basurto University Hospital, Bilbao, Vizcaya,
Spain.
Body: Introduction: Estrogen receptor (ER) and progesterone receptor (PR) expression are generally determined by
semiquantitative immunohistochemistry (IHC) as recommended in the clinical practice guidelines of the American Society of
Clinical Oncologists. This method may be subject to variability because of differences in fixation, ER antibody clones, as well as
to immunostain interpretation and use of arbitrary cut-off points. Materials and methods: We identified 72 breast cancer cases at
Basurto University Hospital from 1 September 2012 to 31 May 2014 that have been analyzed by Oncotype DX with reporting on
ER/PR expression levels by RT-PCR (Genomic Health Inc, Redwood City, CA). According to the companys report, a tumor is
considered as ER positive with expression units of >=6.5 and PR positive with expression units of >=5.5. Oncotype DX is a
commercial assay that predicts tumor recurrence in node-negative ER-positive breast cancers. It is a reverse
transcription-polymerase chain reaction (RT-PCR) based assay that analyzes the expression of 21 genes (16 cancer-related and
5 control genes) to provide a distant disease recurrence score ranging from 0 to 100. At our institution, hormone receptors by IHC
were analyzed on corresponding surgery breast specimens at the time of initial diagnosis. They were evaluated by IHC on
formalin-fixed paraffin-embedded tissue, between 8-72 hours. ER was assessed using Roche antibody clone SP1 and PR was
assessed using antibody clone 1E2. Percentage of positive nuclei was determined by visual microscopic estimation: <1 negative
and 1 positive. We compared our semiquantitative method of reporting ER and PR status based on IHC to the RT-PCR hormone
receptor results from Genomic Health. Results: There was 100% concordance between IHC and the RT-PCR assay for estrogen
receptor status. PR findings between IHC and Oncotype Dx revealed a lower concordance (82%) than for ER. RT-PCR was
negative in 10 (14%) cases in which IHC was positive. Pearson correlation coefficient for PR was 0.313. Although our results
demonstrated high concordance between IHQ and Oncotype Dx for ER, our data showed poor concordance for PR.

Title: Triple negative breast cancer, the impact of isotype-specific progesterone receptor antibodies on the diagnosis results
Jacques Bonneterre1, Jacques Bosq2, Charline Alleaume3, Erard Gilles4, Philippe Jamme1 and Alexander Zukiwski5. 1Centre
Oscar Lambret, Lille, France; 2Gustave Roussy Institute, Villejuif, France; 3Biodoxis, Romainville, France; 4Invivis
Pharmaceuticals, Bridgewater, NJ and 5Arno Therapeutics, Flemmington, NJ.
Body: Background: Given the poor therapeutic outcomes for triple negative breast cancer (TNBC), diagnostic accuracy is vital. In
routine IHC testing, the progesterone receptor (PR) is determined using bispecific antibodies (Ab) that recognize epitopes
common to PRA and PRB. PRA and PRB expression can be imbalanced in BC. Relative expression of the PR isotypes appears
to be prognostic in BC evaluated with a bispecific Ab (Hopp 2004). Tumors can express PRA or PRB on different cells in the
same tumor (Mote 2008), which differs from those expressing ER (Zukiwski 2013). A true TN phenotype might escape detection
with the use of one single Ab to detect both PRA and B epitopes, depending on sensitivity/specificity. This study evaluated the
use of two isotype-specific PR Abs to fully characterize the PR status.
Methods: 83 dual ER and HER2 negative archived BC specimens with clinical data were obtained from the Oscar Lambret
Cancer Center, Lille, FR. IHC was performed using anti-PRA, anti-PRB, and the bispecific anti-PR Pg636 antibodies. PR tumor
positivity was explored using 2 cut-offs, 1% or 5% stained tumor cells. PR positive tumors were defined as either PRA or PRB
positive.
Results: For PR positive tumors with 1% positive cells, average PRA positivity was 41%, PRB was 38% (PRA vs PRB p = NS),
and PRAB 3% (PRA vs PRAB = 0.001, PRB vs PR AB p = 0.0001). Using 1% as positivity cut off PRA and PRB were discordant
in 14% of the cases, PRA and PR AB in 12%. PR B and PR AB were discordant in 2%. Discordance between PR positivity (either
A or B) and PRAB positivity was 7% with no PR negative PRAB positive tumors i.e. no positivity was missed using two PR A and
B antibodies while all 7% cases were missed by the PRAB antibody. Using 5% as a cut off, the discordance rate was 8% between
PRA and PRB, 25% between PRA and PRAB and no PRA negative and PRAB positive case were found, and 26% between PRB
and PRAB with no PRB negative PRAB positive cases either. PR positivity (A or B) was missed in 30% of the cases with a PRAB.
Patients with tumors identified as PR positive ( 5%) using the isotype specific antibodies and PR negative with the PRAB
antibody have a better prognosis (DFS).
Conclusion: In TNBC, there is a different staining pattern when using isotype-specific vs bispecific anti-PR anti-bodies. The
average percent of positive tumors cells is substantially reduced when using a bispecific Ab as compared to isotype-specific AB.
This translates into potential false negative PRAB testing which varies from 7% to 30% depending of the cut off criteria. TNBC
reclassified by the use of isotype-specific anti-PR antibodies may be appropriate for investigation with anti-progestins.

Title: Breast cancer biology varies dramatically by method of detection and may account for overdiagnosis
Brandon Hayse1, Prathima Kanumuri1, Brigid K Killelea1, Nina R Horowitz1, Anees B Chagpar1 and Donald R Lannin1. 1Yale
University School of Medicine, New Haven, CT.
Body: Background: Epidemiological and clinical trial data suggest that 20%-30% of breast cancers diagnosed by screening
represent overdiagnosis; that is cancers that, if left untreated, would never present symptomatically nor cause any harm to the
patient later in life. The biology and presentation of these overdiagnosed cancers, however, is not well understood.
Methods: A retrospective review was performed of a prospectively collected database of breast cancers diagnosed at a tertiary
academic medical center from 2004-2013. The mode of initial presentation was categorized into five separate groups according to
the abnormality that first precipitated a breast workup: screening mammogram, screening MRI, screening ultrasound,
self-detected masses, and physician detected masses. The relationship between tumor characteristics and mode of initial
presentation was evaluated using bivariate analysis and multivariate logistic regression.
Results: The table shows data for a total of 2,714 cases. As expected, screen-detected cancers were significantly smaller than
cancers found by the patient or the physician, and included a higher percent of T1 cancers, whereas palpable cancers had a
larger percent of T2 and T3 cancers (p< 0.001). Also not surprisingly, screening modalities detected a higher rate of DCIS
compared to symptomatic presentation (p < 0.001). However in addition to a simple stage shift, screen-detected cancers also had
a higher proportion of luminal and low-grade cancers, whereas symptomatic cancers had a higher incidence of high-grade and
triple-negative cancers (p < 0.001 for each). In a multivariate logistic regression model adjusted for age, race, and tumor size,
cancers detected by screening had a lower odds of being triple-negative (OR 0.51, 95% CI 0.35-0.75), and a lower odds of being
high-grade (OR 0.43, 95% CI 0.29-0.64) compared to cancers found by the patient.
Conclusion: In addition to a stage shift, screening detects cancers with a much more indolent biology and this may account for
the observed rate of overdiagnosis. With the increasing use of MRI and ultrasound for screening, the rate of overdiagnosis is
likely to increase further.
Tumor Characteristics by Method of Detection
Screening
mammogram
Screening MRI
Screening
ultrasound
33% (556/1669)
36% (36/101) 13% (6/46)
Self-detected
mass
Physician physical
exam
p value
4% (33/786)
4% (5/112)
<0.001
IN SITU
% DCIS
INVASIVE
Molecular type
<0.001
ER/PR+, Her2- 79% (718/907)
85% (45/53)
85% (33/39)
61% (393/641)
75% (63/84)
ER/PR+,
Her2+
7% (63/907)
2% (1/53)
5% (2/39)
11% (67/641)
7% (6/84)
ER/PR-, Her2+ 5% (43/907)
4% (2/53)
6% (40/641)
6% (5/84)
ER/PR-, Her2- 9% (83/907)
9% (5/53)
10% (4/39)
22% (141/641)
12% (10/84)
Grade
<0.001
28% (280/983)
45% (25/56)
42% (15/36)
12% (73/633)
20% (17/87)
53% (517/983)
46% (26/56)
42% (15/36)
49% (313/633)
49% (43/87)
19% (186/983)
9% (5/56)
16% (6/36)
39% (247/633)
31% (27/87)
Tumor size
<0.001
T1
78% (874/1113)
81% (53/65)
77% (31/40)
38% (284/753)
49% (52/107)
T2
18% (205/1113)
14% (9/65)
20% (8/40)
47% (354/753)
41% (44/107)
T3
3% (30/1113)
5% (3/65)
3% (1/40)
12% (89/753)
6% (6/107)
T4
1% (4/1113)
3% (26/753)
5% (5/107)

Title: What is the optimal diagnostic approach for small breast lesions? Fine needle aspiration cytology vs. Vacuum-assisted core
needle biopsy
Satoko Nakano1, Masahiko Otsuka1, Akemi Mibu2 and Toshinori Oinuma3. 1Kawaguchi Municipal Medical Center, Kawaguchi,
Saitama, Japan; 2Kawaguchi Municipal Medical Center, Kawaguchi, Saitama, Japan and 3Kawaguchi Municipal Medical Center,
Kawaguchi, Saitama, Japan.
Body: [Background] Although advanced imaging technology enables us to detect small breast lesions, it remains a challenge
whether it is a benign or malignant tumor with imaging findings alone. A definitive diagnosis is required with cytological or
pathological diagnosis under ultrasonography guidance. Excessive examinations and malpractice are also concerned with
increased examinations. The purpose of the study was to identify the optimal diagnostic approach for small breast
lesions.[Patients and Methods] We reviewed a total of 1532 cases of ultrasonography-guided vacuum-assisted core needle
biopsy (VAB) performed at a single institution between June 1996 and December 2013. There were 519 small breast lesions (274
non-palpable lesions and 245 non-mass lesions). Ultrasonographic examinations were performed on a LOGIC 500 (GE
Healthcare, Waukesha, WI, USA) using an 11 MHz linear transducer before November 2011, and on Aplio MX (TOSHIBA,
Minato, Tokyo, Japan) using an 8 MHz linear transducer since then. We performed VAB under ultrasonography guidance using
11-gauge probes (Mammotome Biopsy system; Biopsys Medical Inc., Irvine, CA, USA) for diagnosis and 8-gauge probes for
excision of the lesion.[Results] The mean age of the 519 patients was 52.7 years. Ultrasonography-guided fine needle aspiration
cytology (FNAC) was performed before VAB in 269 cases (51.8%). The FNAC revealed 76 malignant, 23 suspected malignant,
92 indeterminate, 47 benign or normal, and 31 inadequate cases. Of these FNAC results, the final pathological diagnosis was
benign in 2 of 76 malignant and 6 of 21 suspected malignant cases, and malignant in 4 of 47 benign cases. Accordingly, the true
positive, true negative, false positive, and false negative rates for the 146 cases (excluding the indeterminate and inadequate
cases) by FNAC were 95.8%, 84.3%, 15.7%, and 4.2%, respectively. The pathological results of the VAB specimens were
malignant in 229 and benign in 290 cases. Of the 290 cases classified as benign by VAB, we performed post-VAB excisional
biopsy in 20 cases. The reasons for this second pathological examination were malignant or suspicious findings for malignancy
by FNAC (8 cases), inconsistent imaging and FNAC findings (5 cases), and others (7 cases). Excisional biopsy revealed
malignancy in 3 of these 20 cases. Furthermore, we re-performed VAB in 9 out of the original 290 cases, and these were all
diagnosed benign as with the first VAB. Excision of the lesion was the primary reason for the second VAB (6 cases), followed by
inconsistencies with the imaging findings (2 cases), and suspected malignancy by FNAC (1 case). The true positive, true
negative, false positive and false negative rates of VAB were 98.7%, 100%, 0%, and 1.3%, respectively. The mean follow-up
duration was 43 months. [Conclusion] To prevent the excessive examinations, FNAC should be excluded from the initial
diagnostic approach for small breast lesions. VAB is a highly reliable technique as the initial diagnosis for small breast lesions
with high true positive and true negative rates and a very low false positive rate. The optimal strategy for diagnostic procedures
should be adopted in consideration of reducing stress and anxiety in patients and costs.

Title: No Show

Title: The role of lung biopsy in the management of lung nodules in breast cancer patients
Kazuo Matsuura1, Takayuki Kadoya2, Norio Masumoto2, Hideo Shigematsu2, Akiko Emi2, Keiko Kajitani2, Morihito Okada2,
Tsuyoshi Kataoka2, Rumi Haruta2, Koji Arihiro2, Midori Noma1 and Toshiyuki Itamoto1. 1Hiroshima Prefectural Hospital, Hiroshima,
Japan and 2Hiroshima University Hospital, Hiroshima, Japan.
Body: Background: A biopsy of lung nodules in patients, who had received previous surgery for breast cancer, can be performed
with three aims: to confirm that the lesion is lung metastasis, to confirm the diagnosis of other diseases including primary lung
cancer, and to reassess tumors characteristics. Discordance in estrogen receptor (ER), progesterone receptor (PR) and human
epidermal growth factor receptor 2 (HER2) status between primary breast cancer and metastatic lesions has been reported. The
aim of this study was to assess the role of lung biopsy in the diagnosis and to determine the changes in hormonal receptor and
HER2 status of the metastatic lesions.
Methods: A total of 38 consecutive patients who underwent surgery in 31 or transbronchial lung biopsy (TBLB) in 7 for lung
nodules between 1997 and 2014 after curative operation for breast cancer were reviewed.
Results: Eighteen patients (47%) had a solitary lung nodule. The pathologic diagnoses of lung nodules were lung metastases of
breast cancer in 20 patients, primary lung tumor in 14 (Adenocarcinoma in 10; Large cell carcinoma in 2; Small cell carcinoma
and Carcinoid tumor in 1 each), and other diagnoses in 4 (Inflammation and organizing pneumonia in 2 each). Median follow up
duration were 118.8 months in metastatic breast cancer patients and 105.3 months in other histology patients (p=0.392). The
average disease-free interval from the surgery for primary breast cancer were 68.6 months in metastatic breast cancer patients
and 66.3 months in other histology patients (p=0.897). The 10-year survival rate after the surgery for primary breast cancer was
significantly longer in other histology patients (92.3%), including primary lung cancer patients than in metastatic breast cancer
patients (55.1%) (p=0.0308). In 20 cases of metastatic breast cancer, rates of discordance were 5%, 10% and 10% for ER gain,
PgR gain and HER2 loss, respectively. Sixteen of patients maintained the same tumor phenotype, whereas 4 changed during
progression. Especially, 2 cases of ER, PgR gain could receive endocrine therapy instead of chemotherapy.
Conclusion: As lung nodules that appear in breast cancer patients are not always lung metastases, the pathologic diagnosis
should be confirmed, and surgery is an option for the pathologic confirmation. Furthermore, discordance in biomarker status
between primary breast cancer and the lung metastasis occurred in 20% of cases. It is necessary for clinicians to check
biomarker status in recurrent breast cancer patients as it may assist a shift in the treatment plan.

Title: Provista-001 a multi-center prospective study of protein signature used in the differentiation of benign breast lesions from
invasive breast cancer in women under the age of 50 with a BI-RADS 3 or 4
David E Reese1, Michael Silver1, Susan Yeh2, Sherri Borman2 and Henderson C Meredith2. 1Provista Diagnostics, New York, NY
and 2Provista Diagnostics, Scottsdale, AZ.
Body: Background: Over-diagnosis of breast lesions represents a significant problem in detection and screening of breast
cancer, especially in women under the age of 50. Despite this issue, few new approaches have been developed to augment
standard of care in the more precise detection of breast cancer. The combination of breast imaging, which provides anatomical
evidence, with a robust protein signature that would detect biochemical cues of breast cancer offers an attractive approach to this
problem. We have recently identified a protein signature composed of immune-regulatory cytokines, growth factors and tumor
derived autoantibodies. Here, we test the hypothesis that a protein signature, combined with standard of care can greatly
increase the precision of breast cancer diagnosis in women under the age of 50 in a randomized and blinded study.
Methods: Provista-001 enrolled 351 patients from 10 sites across the US and will follow patients for 6 months following the first
blood draw under IRB approval. Patients were consented after first assessment of a BIRADS 3 or 4 and considered eligible if they
were under the age of 50, no history of cancer, no prior breast biopsy, and were assessed as BIRADS 3 or 4 within 21 days of
blood draw. Upon enrollment, patients were randomized to either training or validation groups. Clinical truth was set at equal to or
greater than 80% sensitivity/specificity. Serum protein biomarkers and autoantibodies identified in prior proteomic screens were
measured in serum samples collected prior to biopsy. Individual biomarker (22 serum protein and autoantibodies) concentrations,
together with specific patient data were evaluated using various logistic regression models developed from prior retrospective
studies. A training set, comprised of 200 patients, was used to develop and refine new models, which were then validated in the
remaining 151 subjects. Clinical findings were compared to biopsy (largely BIRADS 4) or will be followed for 6 months and
re-assessed (BIRADS 3).
Results: The novel algorithm utilizing patient data, serum protein and autoantibody concentrations combined with regression
models was able to distinguish benign from breast cancer lesions in a statistically significant manner. Importantly, the serum
protein biomarkers alone were unable to adequately distinguish benign lesions, consistent with prior work. However, the addition
of tumor autoantibodies markedly increased both the sensitivity and specificity of the assay in this group of women. The use of
the algorithm in conjunction with imaging was more accurate than imaging alone in this population.
Conclusions: Our current findings suggest that when used in combination, the protein signature developed here and breast
imaging provides a more precise detection methodology than either alone. This is particularly important in women under the age
of 50 where a low prevalence of disease makes detection difficult. The follow-up data at six months (BIRADS 3) will yield
additional data in this understudied group of women. Additional studies utilizing the protein signature with women over the age of
50 are currently underway.

Title: Breast cancer related immune suppression in the sentinel lymph node can be effectively countered by combined CpG-B
administration and JAK2/STAT3 inhibition
Kim M van Pul1, Ronald JCLM Vuylsteke2, Sinad M Lougheed1, Lisette EA te Velde3, Petrousjka van den Tol3, Emiel J Th
Rutgers4, Hein BAC Stockmann2 and Tanja D de Gruijl1. 1VU Medical Center, Amsterdam, Noord-Holland, Netherlands;
2
Kennemer Gasthuis, Haarlem, Netherlands; 3VU Medical Center, Amsterdam, Noord-Holland, Netherlands and 4Netherlands
Cancer Institute, Amsterdam, Noord-Holland, Netherlands.
Body: Background
Increasing evidence suggests that immune regulated pathways influence both breast cancer (BrC) development and response to
(neo) adjuvant chemotherapeutic treatment. The sentinel lymph node (SLN) is the first site of BrC induced suppression of immune
effector subsets, rendering them unable to mount an effective anti-tumor response. Since detailed knowledge of the functional
status of these immune effector subsets is lacking, we compared the immune status of BrC SLN with healthy lymph nodes (HLN)
using viable cell samples and correlated our findings to several clinico-pathological characteristics. Additionally we investigated if
ex vivo conditioning with the Toll-like receptor-9 ligand CpG-B with or without simultaneous inhibition of JAK2/STAT3 signaling (a
downstream signaling pathway implicated as the master switch in tumor induced immune suppression) could overcome the
observed immune suppression in BrC SLN.
Methods
Viable SLN cells were obtained from 40 clinically node negative BrC patients. Axillary HLN were obtained from 17 BRCA-1 or -2
patients undergoing a prophylactic mastectomy. Frequencies and activation state of dendritic cell (DC) subsets and regulatory T
cells (Treg) were determined by extensive multi-color flowcytometric analyses. Additionally, SLN cells from 12 BrC patients were
cultured in absence and presence of CpG-B (PF-3512676) and the combination of CpG-B with a JAK2-STAT3 inhibitor (AG-490).
Results
Of the 40 BrC SLN 11 contained metastasis. We found clear evidence of BrC-related immune suppression, as significantly higher
Treg frequencies were observed in metastasis negative BrC SLN as compared to HLN. These frequencies further increased in
metastasis positive BrC SLN. Activation state (by expression of the co-stimulatory molecules CD40, CD83 and CD86) of lymph
node resident (LNR)-DC subsets (both CD11c+ myeloid DC and CD123+BDCA2+ plasmacytoid DC), but not of CD1a+ migratory
subsets, was significantly lower in BrC SLN as compared to HLN. Additionally, in BrC SLN, activation state of these LNR-DC
subsets was further reduced upon metastatic involvement and in non-luminal BrC subtypes (triple negative and HER-2+) as
compared to luminal BrC subtypes (ER and/or PR positive). We previously showed in early-stage melanoma patients that local
treatment with CpG-B led to the preferential activation of LNR-DC subsets. Similarly, ex vivo targeting of BrC SLN with CpG and
CpG + JAK2/STAT3 inhibition resulted primarily in significantly enhanced activation of LNR-DC subsets, with superior effects of
combined CpG and JAK2/STAT3 targeting evidenced by significantly higher CD83 expression.
Conclusion
BrC induced immune suppression in SLN seems to be primarily mediated by hampered activation of LNR-DC subsets, which was
most effectively overcome (ex-vivo) with the immune activating compound CpG-B in combination with a small-molecule inhibitor
of the immune suppressive JAK2/STAT3 signaling pathway (AG490). This combined immune targeting strategy might be of
clinical benefit in enhancing effectiveness of conventional (neo)adjuvant chemotherapeutic treatment, especially in -more immune
suppressed- non-luminal BrC subtypes with known poor prognosis.

Title: Hyaluronan (HA) depletion sensitizes HAhigh tumors to antibody-dependent cell-mediated cytotoxicity
Netai C Singha1, Tara Nekoroski1, Chunmei Zhao1, Rebecca Symons1, Ping Jiang1, Gregory Frost2, Zhongdong Huang1 and H
Michael Shepard1. 1Halozyme Therapeutics, San Diego, CA and 2Intrexon Corporation, San Diego, CA.
Body: Therapeutic efficacy of monoclonal antibodies (MAbs) against solid tumor targets, like trastuzumab (anti-HER2) and
cetuximab (anti-EGFR), have been used successfully to treat cancer, despite the many physical barriers impeding their access to
the malignant cell surface1. For example, only about 50% of HER23+ patients have a durable response to therapy with
trastuzumab2. Extracellular matrix (ECM)-mediated inhibition may be among the mechanisms of resistance to MAb therapy of
solid tumors. Aberrant accumulation of hyaluronan (HA), a major component of the ECM in many tumors, is associated with poor
prognosis and treatment-resistance in multiple malignancies3-5. We investigated HA-dependent pericellular matrix-mediated
inhibition to ADCC in HAhigh human cancer cells in vitro and in vivo. We observed high levels of tumor associated HA (HA3+) in
>50% of HER23+ breast adenocarcinoma and 40% of EGFR+ head and neck squamous cell carcinoma (HNSCC) primary
tumors. Human hyaluronan synthase 2 (HAS2)-overexpressing breast cancer cells formed an HAhigh pericellular matrix, which
inhibited both natural killer (NK) cell access to tumor cells and ADCC in vitro. Hyaluronan depletion by PEGPH20, a pegylated
recombinant human PH20 hyaluronidase currently in clinical study for pancreatic cancer, increased NK cell access to
HAS2-overexpressing breast cancer cells and greatly enhanced trastuzumab- or cetuximab-dependent ADCC. Trastuzumab and
NK cell accessibility to HAS2-overexpressing tumors was enhanced following HA-depletion by PEGPH20. In an in vivo
ADCC-based efficacy study, PEGPH20 treatment in combination with trastuzumab and NK cells enhanced tumor growth
inhibition. This work describes a novel tumor microenvironment (TME)-dependent mechanism of inherent resistance to
therapeutic antibody-mediated ADCC in vitro and in vivo, and furthermore shows that ADCC can be enhanced by hyaluronan
depletion. These results may help to explain as to why tumors with high levels of HA are more aggressive, and suggest potential
benefits of PEGPH20-mediated HA depletion in combination with therapeutic antibodies like trastuzumab or cetuximab in the
treatment of HAhigh solid tumors.
References:
1. J. Christiansen, A. K. Rajasekaran, Biological impediments to monoclonal antibodybased cancer immunotherapy. Mol Cancer
Ther. 3, 1493-1501 (2004).
2. H. M. Shepard, C. M. Brdlik, H. Schreiber, Signal integration: a framework for understanding the efficacy of therapeutics
targeting the human EGFR family. J. Clin. Invest. 118, 3574-3581 (2008).
3. A. Kultti, X. Li, P. Jiang, C. B. Thompson, G. I Frost, H. M. Shepard Therapeutic Targeting of Hyaluronan in the Tumor Stroma.
Cancers 4, 873-903 (2012).
4. R.K. Jain, Normalizing tumor microenvironment to treat cancer: bench to bedside to biomarkers. J Clin Oncol. 31, 2205-18
(2013).
5. R. K. Boregowda, H. N. Appaiah, M. Siddaiah, S. B. Kumarswamy, S. Sunila, K. N. Thimmaiah, K. Mortha, B. Toole, S. D.
Banerjee, Expression of hyaluronan in human tumor progression. J. Carcinog. 5, 2 (2006).

Title: Deconvoluting immune cell populations using in silico flow cytometry with CIBERSORT: Association with neoadjuvant
therapy response and genomic instability in TNBC
Shaveta Vinayak1, Aaron Newman2, Sylvia Adams3, Anosheh Afghahi2, Kristin C Jensen2, Sunil S Badve4, James M Ford2, Ash A
Alizadeh2 and Melinda L Telli2. 1University Hospitals Seidman Cancer Center/ Case Western Reserve University School of
Medicine, Cleveland, OH; 2Stanford University School of Medicine, Stanford, CA; 3New York University Langone Medical Center,
New York, NY and 4Indiana University School of Medicine, Indianapolis, IN.
Body: Background: Increased tumor infiltrating lymphocytes (TILs) are prognostic and predictive of therapy response in TNBC.
CIBERSORT, a highly novel in silico flow cytometry gene expression-based method, can assess the overall immune content and
relative levels of distinct leukocyte subsets in tumors.
Methods: We applied CIBERSORT to PrECOG 0105, a neoadjuvant trial of carboplatin, gemcitabine and iniparib for patients with
clinical stage I-IIIA TN or BRCA1/2 mutation-associated BC. H&E tumor sections from pre-therapy biopsies were evaluated for
density of stromal (sTILs) and intratumoral (iTILs) lymphocytes. Pathologic response was assessed at definitive surgery. Overall
immune content and fractions of 23 distinct leukocyte subsets were derived from Affymetrix U133 plus 2.0 arrays, using
CIBERSORT. Using an input matrix of reference gene expression signatures of 23 purified leukocytes, this methodology can infer
an unknown fraction of each cell type from tissue expression profiling, yielding a p value for deconvolution. All patients had
BRCA1/2 genotyping and levels of tumor genomic instability were assessed by the homologous recombination deficiency (HRD)
assay.
Results: 57 pts had gene expression and TILs assessment. TILs, sTILs, and iTILs were significantly associated with a measure of
absolute immune content determined by CIBERSORT, termed an immune score (R 0.73, p<0.0001; R 0.70, p<0.0001; R 0.57,
p<0.0001 respectively). Both iTILs and the CIBERSORT immune score were significantly associated with pathologic complete
response (pCR; ypT0/is N0) in independent models. Specific leukocyte subsets significantly associated with pCR, included
activated memory CD4 T cells, CD8 T cells, and M1 macrophages (all p<0.05). Regarding genomic instability measures,
BRCA1/2 germline mutation status was not associated with TILs or immune score, whereas a high HRD assay score was
significantly associated with the CIBERSORT immune score (p=0.038). Individual leukocyte subsets significantly associated with
a high HRD assay score, included plasma cells, activated memory CD4 T cells, M1 macrophages, and unstimulated mast cells
(all p<0.05).
Conclusions: Neoadjuvant platinum-based therapy response significantly associates with both iTILs and CIBERSORT immune
score. A measure of global genomic instability (HRD score) significantly associates with immune score alone. Enumeration of 23
leukocyte subsets in therapy-nave TNBC by CIBERSORT revealed three distinct cell types that significantly predict pCR, two of
which also associate with genomic instability. These results suggest an intimate interplay between genomic instability and
immune infiltration, potentially shaping adaptive anti-tumor humoral immune responses, and thereby affecting neoadjuvant
response in TNBC.


Title: Co-stimulation through 4-1BB/CD137 improves expansion and function of tumor-infiltrating T lymphocytes from primary and
metastatic triple-negative breast cancer and inflammatory breast cancer
Michiko Harao1, Hui Gao1, Jie Qing Chen1, Elizabeth A Mittendorf1, Gildy V Babiera1, Sarah M DeSnyder1, Korrene F Rockwood1,
Savitri Krishnamurthy1, Huiming Sun1, Jie S Willey1, Naoto T Ueno1, James M Reuben1 and Laszlo G Radvanyi1,2. 1University of
Texas MD Anderson Cancer Center, Houston, TX and 2Lion Biotechnologies, Woodland Hills, CA.
Body: Background: Increased CD8+ tumor-infiltrating lymphocytes (TIL) is associated with improved prognosis in triple-negative
breast cancer (TNBC) suggesting that T-cell responses at the tumor site can be harnessed for autologous T-cell therapy using
TIL expanded ex vivo. Although TIL therapy has been developed for solid tumors such as melanoma, cervical, and ovarian
cancer. Moreover, methods facilitating CD8+ TIL expansion from TNBC are desirable given their cytotoxic potential against tumor
cells. One approach to address this need is to provide agonist signals through the 4-1BB/CD137 pathway during TIL expansion
selectively costimulating CD8+ T-cell activation. In this study, we established a method of expanding TIL from surgical specimens
and core biopsies from primary TNBC patients and compared the phenotype and function of these TIL to lymphocytes from
peripheral blood.
Patients and methods: Eight primary human TNBC tumor samples were obtained by surgical resection or core biopsy after
neo-adjuvant chemotherapy. Small (4-6 mm2) tumor fragments were cultured for 28 days in 24-well plates in medium containing
3000 IU/ml IL-2 alone or in combination with 10 g/ml agonistic anti-4-1BB IgG4 (BMS663513) added at the start of culture.
Viable cell numbers and the expression of CD3, CD8, CD4, CD27, CD28, CD56, CD16, Granzyme B, and Perforin were
determined by flow cytometry on day 28 after culture. Cytotoxic function of the TIL was evaluated by measuring Caspase 3
cleavage in target cells. Blood samples collected at surgery were immunophenotyped for subsets of T cells and NK cells, and
assessed for ability of T cells to synthesize Th1/Th2 cytokines following activation through the T-cell receptor (TCR), and potential
of NK cells to kill K562 targets.
Results: TIL were successfully expanded from tumor fragments in 6/8 of the cases, with addition of anti-4-1BB greatly increasing
the percentage and yield of CD8+CD3+ T cells. CD8+ TIL isolated from cultures receiving 4-1BB costimulation however had
decreased CD27 and CD28 expression together with increased cytotoxic T-cell activity. Gene expression analysis also found that
TIL from these 4-1BB costimulated cultures had a more differentiated CD8+ T-cell gene profile. Peripheral blood CD3+, CD8+,
and CD4+ T cells were lower than those of healthy controls, but TCR-activated cytokine synthesis was not significantly different.
Peripheral blood NK cells expressed normal Granzyme A/B and Perforin levels and exhibited normal IFN- secretion and
CD107a-release following exposure to IL-12/IL-18 or with K562 targets.
Conclusions: TIL can be reproducibly expanded from TNBC tumors with IL-2 after neo-adjuvant therapy, with CD8+ TIL
outgrowth and effector activity increased by provision of 4-1BB/CD137 costimulatory signals during culture initiation. These CD8+
TIL however had a more differentiated phenotype with higher cytotoxic activity. Peripheral blood T-cell and NK cell function was
comparable to those of healthy donors. Our results support further development of an autologous TIL expansion protocol after
neo-adjuvant therapy for use in an adoptive cell therapy approach to treat TNBC recurrence or metastasis.

Title: Oral immunomodulatory agents prevent tumor growth and increase tumor CD8 T cell infiltrate; bexarotene further improves
tumor response to conventional chemotherapy in breast tumors
Sasha E Stanton1, Ekram Gad1, Edmond Marzbani1, Lauren Rastetter1 and Mary L Disis1. 1University of Washington, Seattle, WA.
Body: The tumor immune environment is important in breast cancer with greater than 50% immune infiltrate (LPBC) prior to
neoadjuvant chemotherapy predicting improved pathologic complete response, and LPBC and CD8 infiltrate prior to adjuvant
therapy predicting improved survival. Unfortunately, the majority of breast cancers do not have LPBC or robust CD8+ infiltrate.
Evidence has emerged that conventional chemotherapy can increase CD8+ T cells therefore discovering ways to boost this
response should further enhance the anti-tumor function. Three oral agents have modest anti-tumor function: metformin (oral
biguanide), bexarotene (retinoic receptor agonist), and celecoxib (COX2 inhibitor). In vitro data suggest a role for these agents in
increasing Th1 immunity: metformin was shown to increase MHC class I expression on tumor cells, bexarotene was shown to
decrease CD8+ T cell apoptosis, and celecoxib has been shown to decrease MDSC cells. The goal of this study was to
demonstrate whether addition of these oral agents to conventional chemotherapy enhanced the anti-tumor function of
chemotherapy, possibly by modifying the immune environment in the transgenic mouse mammary tumor model TgMMTV-neu
(genetically similar to luminal breast cancer).
Two active chemotherapies in human breast cancer, doxorubicin and paclitaxel, inhibited tumor growth and increased CD8+ T
cell tumor infiltrate in transgenic mice. Treatment of mice with 100 mm3 tumors with doxorubicin (5 mg/kg weekly for four weeks)
showed a 32% increase in CD8+ T cells and 85% decrease in tumor growth as compared to control mice (p=0.0001) and
treatment with paclitaxel (10 mg/kg weekly for four weeks) showed a 40% increase in CD8+ tumor infiltrate (p=0.0068) and 60%
decrease in tumor growth as compared to control treated mice (p=0.0026). A third chemotherapy cyclophosphamide (100 mg/kg
weekly for four weeks) increased CD8+ tumor infiltrate by 45% (p=0.011) but did not show a significant decrease in tumor volume
(p=0.57). Metformin and bexarotene also demonstrated increased CD8+ tumor infiltration and decreased breast tumor growth but
celecoxib did not. Metformin treated mice had a 46% increase in CD8+ tumor infiltrate (p=0.001) and a 52% decrease in mean
tumor volume (p=0.011) as compared to controls (75 mg/m2 of metformin for four weeks). Bexarotene treated mice had 44%
increase in CD8+ tumor infiltrate (p=0.05) and 60% decrease in mean tumor growth (p=0.03) compared to control (treated with 50
mg/m2 bexarotene for four weeks). Of all three oral therapies, bexarotene was most effective inhibiting spontaneous tumor
growth in the mice. Furthermore, addition of bexarotene to chemotherapy was superior to chemotherapy alone. Adding
bexarotene to weekly doxorubicin decreased tumor growth by 98% (p=0.008 compared to doxorubicin alone), adding bexarotene
to weekly paclitaxel decreased tumor growth by 86% (p=0.02 compared to paclitaxel alone), and adding bexarotene to weekly
cyclophosphamide inhibited tumor growth by 82% (p=0.04 compared to cyclophosphamide alone). These results suggest that
addition of bexarotene, a well-tolerated oral agent, to chemotherapy may improve tumor response in breast cancer.

Title: Multispectral imaging allows visualization and quantification of multiple immunologic cell types in breast tumor tissues
Elizabeth A Mittendorf1, Chichung Wang2, Peng Yau1, Kristin Roman2, Yun Wu1, Gheath Alatrash1 and Clifford Hoyt2. 1University
of Texas MD Anderson Cancer Center, Houston, TX and 2Perkin Elmer, Hopkinton, MA.
Body: Background: Multiple studies have identified tumor infiltrating lymphocytes (TIL) in breast cancer and shown them to have
prognostic and predictive significance. TIL are comprised of CD8+ and CD4+ T cells, regulatory T cells (Treg) and B cells. To
date, the majority of studies have identified TIL using H&E staining. Studies utilizing IHC to stain for TIL generally have been
limited to evaluating a single immune cell phenotype such as CD8+ T cells, while studies evaluating multiple immune cell types
have generated single cell suspensions from tumors for analysis by flow cytometry to determine relative percentages of specific
immune cell populations. Although these studies have confirmed that immune infiltrates in tumors are heterogeneous, they have
not addressed the spatial relationship between tumor and immune cells, which could be critical to anti-tumor immune effects. This
study was undertaken to demonstrate the feasibility of multispectral imaging and multiplexed immunofluorescence to visualize
and quantify specific immune infiltrates in the tumor and surrounding stroma on single FFPE tissue sections.
Methods: Single FFPE slides from 9 HER2+ breast cancer patients receiving neoadjuvant chemotherapy were stained with
primary antibodies targeting cytokeratin, CD8, CD4, FoxP3 (Treg), CD20 (B-cells) and PD-L1 (T cell inhibitory molecule).
Tyramide signal amplification was used to improve signal and specificity, as well as to reduce background. The Vectra
multispectral imaging instrument was used for image acquisition and InForm software was used to quantitate the density (number
of cells per square millimeter) of specific cell types in the tumor and in the surrounding stroma.
Results: Multispectral imaging successfully captured and quantified multiple immune cell types in all tissues (images to be shown
at time of presentation). CD8+ and CD4+ T cell densities as well as PD-L1 expression in the tumor and surrounding stroma are
shown.
Summary of immune cell infiltrate
Within Tumor
Within Stroma
Sample
CD8 density
CD4 density
PDL1 H-score
CD8 density
CD4 density
PDL1 H-score
50
63
1301
207
34
30
591
892
38
122
60
2211
1331
10
305
3833
106
34
808
131
75
10
1189
1591
110
177
546
21
585
1143
17
3918
68
121
859
161
74
4834
2570
196
For patients not achieving a pathologic complete response (pCR), the density of both the CD8 (p=.03) and CD4 (p=.05) infiltrates
in the stroma were significantly greater than in the tumor. For patients achieving a pCR, there was no significant difference in the
densities of stromal and intratumoral CD8 (p=.11) or CD4 (p=.75) infiltrates suggesting that T cell infiltration into the tumor from
the stroma is critical.
Conclusion: Multispectral imaging allows different immune cell phenotypes to be visualized and quantified simultaneously in the
same tissue section enabling further study of the relationships and distribution of these cells within the tumor and tumor
microenvironment, and their spatial distribution and proximity to the tumor cells. This technology will enable improved
understanding of the immune infiltrate in breast tumors thereby facilitating the rational design and use of immunotherapeutic
agents in combination with standard systemic therapies.

Title: Expression of novel immunotherapeutic targets in luminal breast cancer patients
Gargi D Basu1, Anatole Ghazalpour1, Randal Vader1, Sandeep Reddy1, Karen Anderson3, Ann McCullough2 and Barbara Pockaj2.
1
Caris Life Sciences, Phoenix, AZ; 2Mayo Clinic, Scottsdale, AZ and 3Arizona State University, Tempe, AZ.
Body: Background: The development of novel chemotherapeutic agents has significantly improved the prognosis and survival of
patients with breast cancer. ER positive or luminal tumors represent around two thirds of all breast cancers and these cancers are
comprised of different histologies including differing gene expression and mutational profiles. This study examined biomarkers
involved in immune evasion including PD-L1 and its association with other biological pathways as potential treatment options for
luminal breast cancer patients.
Methods: We analyzed 1311 breast samples using a multiplatform approach including whole genome mRNA expression
(HumanHT-12 v4 BeadChip Illumina Inc., San Diego, CA), protein expression (immunohistochemistry), gene copy number
changes (in situ hybridization) and gene sequencing and an additional 304 breast samples were tested for PD-1 and PD-L1 by
IHC. The mRNA expression data was based on whole tumor and represents cell type heterogeneity. Heat map analysis was done
to look at differential gene expression between the PD-L1 high vs low luminal breast cancers.
Results: Based on expression of ER, PR and HER2 by IHC, we subdivided the data into sub- cohorts. Elevated mRNA
expression of immune markers including PD-L1, CTLA4, B7H-3, and IDO1 was noted in the ER+HER2- luminal population
including ER+ PR- and ER+PR+ cohort. Positive correlation was found between PD-L1 and other immune regulators including
CTLA-4, B7-H3 and IDO1 (Spearman correlations of 0.49, 0.36 and 0.46). Protein expression of PD-L1 was found to be specific
to the HER2 negative cohort with no expression in the HER2 positive cohort regardless of ER status. Within the ER+HER2cohort, PD-L1 expression was 5% {ER+PR+ was 5% (4/81) and the ER+PR- was 6% (2/34)}. In contrast, PD-L1 expression was
higher in the triple negative cohort, 17% (13/75). The expression of PD-1 on the other hand was present throughout the different
cohorts. PD-1 expression ranged from 43% in the ER+HER2- cohort {41% (14/34) in ER+PR-HER2-; 43% (35/81) in
ER+PR+HER2-} to 33% in the ER+HER2+ cohort {44% (4/9) in ER+PR-Her2+; and 17% (1/6) in ER+PR+HER2+}. In the
ER-HER2+ cohort PD-L1 expression was 75%{6/8 in ER-PR-HER2+ and 0/0 in ER-PR+HER+} and 63%(47/75) in the triple
negative cohort. Pathway analysis of the PD-L1 negative vs positive luminal population identified 127 genes involved in various
cancer pathways including EGFR and VEGF signaling network (P=7.47e-09).
Conclusions: The expression of immune regulatory targets in the breast cancer population suggests that immune- targeted
therapies with anti PD-1/PD-L1, CTLA4, B7-H3 and IDO1 may be effective especially in the luminal cohort. Our study further
shows that ER+HER2- breast cancer patients may be better candidates for immunotherapy with checkpoint inhibitors as
compared to ER+HER2+ due to lack of PD-L1 expression in the HER2 positive cohort. Further validation of our findings is
ongoing.

Title: Characterization of neutrophil elastase receptor in breast cancer: Implication for immunotherapy
Celine Kerros1, Satyendra C Tripathi2, Anne V Philips3, Gheath Al-Atrash1, Kathryn E Ruisaard1, Karen C Dwyer1, Elizabeth A
Mittendorf3, Samir M Hanash2 and Jeffrey J Molldrem1. 1University of Texas MD Anderson Cancer Center, Houston, TX;
2
University of Texas MD Anderson Cancer Center, Houston, TX and 3University of Texas MD Anderson Cancer Center, Houston,
TX.
Body: Neutrophil Elastase (NE), a serine protease released by tumor-associated neutrophils in the tumor microenvironment
induces invasion and metastasis. We have demonstrated that NE is not endogenously expressed in breast cancer (BrCA), but is
taken up by BrCA cells. NE uptake results in the expression of the HLA-A2-bound peptides CCNE1 and PR1, derived from cyclin
E and NE, respectively, on the surface of triple-negative (TN) BrCA cells (MDA-MB-231). Expression of these peptide/HLA-A2
molecules induces BrCA cell susceptibility to cytolysis by CCNE1- and PR1-specific cytotoxic T lymphocytes (CTLs), and to 8F4,
a monoclonal antibody that binds specifically to the PR1/HLA-A2 complex. We hypothesize that NE uptake is a receptor-mediated
process that results in cross-presentation of NE-derived peptides on HLA molecules of BrCA cells.
Here, we found that NE uptake is specific, time- and dose-dependent, and saturable, suggesting a receptor-mediated uptake
mechanism. We showed that MDA-MB-231 did not take up cathepsin G, a related serine protease, suggesting specificity of
receptor uptake. NE internalization was partially blocked by chlorpromazine and by wortmannin, suggesting clathrin-dependent
uptake and PI3Kinase-dependence, respectively. Confocal microscopy showed that NE was colocalized with the early endosome
marker, EEA-1, as early as 10 min after uptake. Flow cytometry indicated that surface-bound NE on MDA-MB-231 cells
decreased after 5 minutes, and both flow cytometry and Western blot showed a simultaneous decrease of phospho-Erk and loss
of IRS-1 signaling. Inhibition of NE enzyme activity by elafin or PMSF potentiated NE uptake in BrCA cells, thus enzyme activity is
not required for uptake. Conclusion: The results support a novel mechanism of rapid receptor-mediated uptake of soluble
exogenous NE by TN BrCA. NE uptake is efficient, PI3 kinase-dependent, sensitive to clathrin inhibition, and associated with
down-regulation of Erk phosphorylation and IRS-1. In addition, uptake does not require NE enzyme activity. Following uptake, NE
colocalizes to an early endosomal compartment, an organelle associated with peptide loading of MHC-I molecules for expression
of the cell surface. We previously showed that the NE-derived peptide PR1 is cross-presented on MDA-MB-231 cells after NE
uptake, leading to susceptibility to immunotherapies that target PR1/HLA-A2. Taken together, our results demonstrate
receptor-mediated NE uptake, which is a potentially novel paradigm that could vastly expand the number of tumor-associated
antigens that could be targeted on BrCA. An understanding of the mechanism that mediates NE uptake will help us develop
strategies to increase expression of immunogenic peptides on cancer cells and to guide the design of targeted immunotherapies
against CCNE1 and PR1 as new clinical therapies for BrCA.

Title: Characterization of the immune infiltrate in HER2+ breast cancer
Shridar Ganesan2, Gabriela Alexe1, Kim M Hirshfield2, Ming Yao2, Gyan Bhanot2 and Toppmeyer Deborah2. 1Broad Institute of
MIT and Harvard, Boston, MA and 2Rutgers Cancer Institute of New Jersey, New Brunswick, NJ.
Body: A subset of HER2+ breast cancer is characterized by high expression of lymphocyte-associated genes and the presence
of a robust, histologically apparent, lymphocytic infiltrate. The presence of an immune signature and lymphocytic infiltration is
associated with improved outcome, and may be a marker of improved response to trastuzumab-based therapy. The relationship
of the immune signature in these cancers to the biology and functional status of the lymphocyte subsets present in the tumor
microenvironment remain unclear. To better characterize the immune microenvironment of HER2+ cancers, we have analyzed
paired RNA sequencing data and histologic image data from The Cancer Genome Atlas. We have confirmed the presence of a
strong immune signature in a subset of HER2+ breast cancers that is correlated with the presence of a robust lymphocytic
infiltrate. Immuno-phenotyping of the lymphocytic infiltrate was performed on an independent set of HER2+ breast cancers from
the Rutgers Cancer Institute of New Jersey. CD8+ T-cells and macrophages are a significant component of the immune infiltrate
in these cancers. Interestingly, CD20+ B-cells were also present in the tumor stroma as large aggregates whose organization is
suggestive of the presence of ectopic germinal centers. The relationship of lymphocyte subsets present in HER2+ breast cancer,
immunoglobulin subtypes expressed and correlation with immune checkpoint protein expression will be presented. The
composition of the immune infiltrate and its organization into ectopic germinal centers may give new insight into the role of the
immune microenvironment in the development of HER2+ breast cancer and the response to trastuzumab.

Title: No Show

Title: Targeting of phosphatidylserine by monoclonal antibodies enhances activity of immune checkpoint inhibitors in breast
tumors
Bruce Freimark1, Jian Gong1, Van Nguyen1, Shen Yin1, Rich Archer1 and Jeff Hutchins1. 1Peregrine Pharmaceuticals, Tustin, CA.
Body: Phosphatidylserine (PS) is a phospholipid normally residing in the inner leaflet of the plasma membrane and becomes
exposed on tumor vascular endothelial cells (ECs) and tumor cells. PS exposure becomes enhanced in response to
chemotherapy, irradiation, and oxidative stresses in the tumor microenvironment. PS exposure in tumors promotes an
immunosuppressive microenvironment which includes the recruitment of myeloid derived suppressor cells (MDSCs), immature
dendritic cells, and M2-like macrophages as well as the production of anti-inflammatory cytokines. Binding of PS targeting
antibodies on tumor endothelial cells, tumor cells and their secreted microparticles triggers an Fc-FcR mediated pro-inflammatory
cellular and cytokine response that reverses the immunosuppressive PS meditated checkpoint, thereby enhancing anti-tumor
immunity. A chimeric PS-targeting antibody, bavituximab, is being used in combination with chemotherapy to treat patients with
solid tumors in multiple late-stage clinical trials. Using breast tumors in immune competent mice, we demonstrate PS targeting
antibodies enhance the anti-tumor activity of anti-CTLA-4 and anti-PD-1 antibodies in the presence and absence of conventional
chemotharapy. Tumor growth inhibition correlates with infiltration of immune cells in tumors and induction of adaptive immunity.
The combination of these mechanisms promotes strong, localized, anti-tumor responses without the side-effects of systemic
immune activation.

Title: Toll like receptor-9 and CD73 may act on the same pathway to induce immunosuppression in triple negative breast cancer
cells
Katri S Selander1, Johanna M Tuomela2, Mikko Mella3, Joonas Kauppila3, Jouko Sandholm2, Gennady Yegutkin2, Peeter
Karihtala3, Arja Jukkola-Vuorinen3, Kirsi-Maria Haapasaari3, Katri S Vuopala4 and Kevin W Harris1. 1University of Alabama,
Birmingham, AL; 2University of Turku, Finland; 3University of Oulu, Finland and 4Lapland Central Hospital, Rovaniemi, Finland.
Body: Toll like-receptor-9 (TLR9) is an innate immune system DNA-receptor which is also widely expressed in breast cancer cell
lines and in clinical breast cancer specimens. Although TLR9 ligands (such as bacterial DNA and synthetic, CpG-sequence
containing oligonucleotides) induce TLR9-mediated invasion in breast cancer cells in vitro, the contribution of this protein to
breast cancer pathophysiology remains unclear. We showed previously that tumor TLR9 expression is a highly significant
prognostic factor in breast cancer, but only in tumors that are triple negative (TNBC). Specifically, low tumor TLR9 expression in
TNBC is associated with a significantly shortened disease-specific survival. Our published, preclinical results further suggest that
tumor TLR9 expression may be an important determinant of tumor immunophenotype and that patients with low-TLR9 TNBC may
not gain the immunogenic benefit from chemotherapy. A possible mechanistic explanation for the change in the
immunophenotype of low TLR9-TNBC tumors involves CD73, which is a 5ectonucleotidase that converts extracellular adenosine
monophosphate (AMP) into the highly immunosuppressive adenosine. CD73 is also expressed in breast cancer cells and in
clinical specimens. High CD73 expression has been associated with poor prognosis, but similar with TLR9 expression, the
prognostic significance was specific only for TNBC. There are no previous reports on TLR9 regulation of CD73 in any cancer.
Interestingly, however, TLR9 deficiency was shown to promote CD73 expression of T-cells in a mouse model of diabetes. To
begin to test the hypothesis that TLR9 affects CD73 expression in TNBC, we studied CD73 mRNA and protein expression in
control and TLR9 siRNA TNBC cells in vitro. We discovered that TLR9 siRNA TNBC cells express significantly higher CD73
mRNA concentrations than control siRNA cells specifically in hypoxia. Similar findings were observed with immunofluorescence.
Our results suggest that the lack or TLR9 in TNBC may allow high CD73 expression and through this mechanism result in an
immunosuppressive phenotype of the tumors. The immunosuppression then would explain the poor prognosis associated with
low TLR9-TNBC tumors. We are currently studying this with clinical TNBC specimens.

Title: Decreased functions of natural killer cells in peripheral blood of advanced triple negative breast cancer patients
In Hae Park1,2, Sun-Young Kong3,4, Joo Hyun Kang2, Hye Jin Mo2, Tae Sik Kim4, Keun Seok Lee1 and Jungsil Ro1,2. 1Center for
Breast Cancer, National Cancer Center; 2Breast & Endocrine Cancer Branch of Research Institute, National Cancer Center;
3
Hospital, National Cancer Center and 4Translational Epidemiology Branch, Research Institute, National Cancer Center.
Body: Background : Natural killer (NK) cells are a major player in innate immune response. Two distinct NK cell subsets with
different functions as the strongest cytotoxic activity (CD56dimCD16+) or cytokine production (CD56brightCD16+/-) have been
known. In this study, we evaluated the phenotypic and functional difference of NK cells in peripheral blood (pB) from the patients
who had advanced triple negative (TN) breast cancer compared to healthy controls.
Methods : We enrolled advanced triple negative (TN) breast cancer patients treated at the National Cancer Center, Korea
between March 2012 and March 2014. Healthy controls who visited for health screening were recruited at the same time period
and peripheral blood (pB) samples were collected. In patient groups, pB was sampled either prior to the initiation or during the
courses of chemotherapy. Then, patients were retrospectively classified into two groups as follows; non-responders (the group A)
and responders (the group B) to the treatment. NK cells were isolated from peripheral blood mononuclear cells (PBMC) and
analyzed based on their expression of CD56 and CD16. We looked at cytotoxic activity of pB NK cells against K562 cell line and
measured CD107a expression as a parameter of NK cell activation.
Results : The absolute numbers of lymphocytes and NK cells per cubic millimeter of pB were similar among healthy controls
(N=24), the group A (N=16), and the group B (N=21). There was no significant difference in the proportion of pB NK cell subsets
among three groups except CD56brightCD16+/- subset which was significantly higher in the group B compared to healthy
controls (P=0.0014). Even in the presence of similar proportions of NK subsets, the activities of CD56dimCD16+ NK cells
measured by CD107a expression were significantly lower both in the group A (P=0.0023) and in the group B (P=0.0127)
compared with those of control group. The activities of CD56dimCD16+ NK cells were not different between group A and B. In the
case of CD56brightCD16+/- NK cells, CD107a positivity was significantly lower in the group A compared with controls (P=0.0068)
and group B (P=0.0465).
Conclusions : Our data suggested that cytolytic functions of NK cells were significantly decreased in advanced TN breast cancer
patients compared with those of healthy controls despite a similar proportion of NK cell subsets in pB.

Title: No Show

Title: No Show

Title: Targeting the mevalonate pathway to overcome acquired anti-HER2 treatment resistance
Huizhong Hu1, Lukas M Simon1, Agostina Nardone1, Chad A Shaw1, Gary C Chamness1, Laura M Heiser2, Nicholas Wang2, C
Kent Osborne1 and Rachel Schiff1. 1Baylor College of Medicine, Houston, TX and 2Oregon Health & Science University, Portland,
OR.
Body: Background: Compelling preclinical and clinical evidence suggests that a more complete blockade of the HER receptor
layer and its signaling, by combining anti-HER2 drugs, such as Trastuzumab (T) and Lapatinib (L), is highly effective. However,
resistance is still common and remains a challenge. To understand resistance mechanisms and further to identify novel
therapeutic strategies, we established a broad panel of L, T, and L+T resistant cell line models. Initial mRNA expression profiling
identified upregulation or restoration of the mevalonate (MVA) pathway in some models where HER signaling is completely and
sustainably blocked. The MVA pathway is commonly considered as a biosynthetic process primarily for cholesterol and
isoprenoid intermediates, particularly farnesyl and geranylgeranyl pyrophosphates (FPP and GGPP, respectively). Statins,
widely-used cholesterol-lowering drugs, block this pathway via inhibition of the rate-limiting enzyme, HMG-CoA reductase. While
accumulating evidence also suggests a role of the MVA pathway in tumor initiation and progression, its role in anti-HER2
resistance remains elusive.
Methods: SKBR3, AU565, and UACC812 parental HER2+ cells and their T, L, and L+T resistant (TR, LR, and LTR respectively)
derivatives were used in this study. Cell growth after treatment with statins in the presence or absence of MVA, cholesterol,
squalene, FPP, or GGPP was measured by methylene blue staining. Apoptosis was determined by Annexin V staining and the
protein level of cleaved PARP. Parallel analysis of molecular signaling was done by western blotting.
Results: Blocking the MVA pathway with lipophilic statins, simvastatin or atorvastatin, led to a marked growth inhibition or
apoptosis in LR/LTR models, in which the HER signaling remains sustainably inhibited, while cognate parental cells and TR cells,
in which HER is (re)activated, were only slightly inhibited. Interestingly, only lipophilic statins (which can be taken up by cancer
cells), but not hydrophilic statins such as pravastatin (whose primary target is liver cells), conveyed the inhibitory effect.
Prevention of statin-induced apoptosis by adding exogenous MVA indicated that the cell death caused by statin treatment was via
its specific blockade of the MVA pathway. Cholesterol or its precursor squalene could not rescue growth inhibition. In contrast,
both FPP and GGPP reversed the growth inhibition or apoptosis in SKBR3 and AU565 LR/LTR models, while in the
UACC812LTR model only GGPP rescued. Interestingly, mTOR was identified as the downstream signaling target of the MVA
pathway in SKBR3 and AU565LTR models, while in the UACC812LTR model, the growth inhibition by statin was due to
substantial estrogen receptor (ER) protein reduction.
Conclusion: The MVA pathway plays a key role as an escape pathway by activating alternative signaling, including mTOR and
ER pathways, in acquired resistance to potent HER2 inhibition in a cholesterol-independent but FPP/GGPP-dependent manner.
Targeting the MVA pathway or its downstream effectors could provide a novel therapeutic strategy to overcome anti-HER2
resistance.

Title: Prognostic impact of single-nucleotide polymorphisms (SNPs) in or near the ZNF423 and CTSO genes in estrogen receptor
(ER)-positive breast cancer patients receiving adjuvant endocrine therapy
Yukari Hato1, Yumi Endo1, Nobuyasu Yoshimoto1, Tomoko Asano1, Mina Yamaguchi1, Satoru Takahashi1 and Tatsuya Toyama1.
1
Nagoya City University Graduate School of Medical Sciences.
Body: Background: Selective estrogen receptor modulators (SERMs) can reduce the occurrence of breast cancer in high-risk
women by 50%. Recently, a genome-wide association study identified SNPs in or near the ZNF423 (rs8060157) and CTSO
(rs10030044) genes that were associated with breast cancer risk during SERM therapy and these SNPs were reported to be
involved in estrogen-dependent induction of BRCA1 expression (Ingle JN. et al. Cancer Discovery 2013).
Materials and methods: A total of 588 breast carcinomas collected between 1983 and 2003 were available for polymorphism
assay. TaqMan pre-designed SNP genotyping assays for ZNF423 rs8060157 and CTSO rs10030044 were used. We investigated
whether these SNPs are associated with prognosis in breast cancer patients. The effects of several variables on survival were
tested by Cox proportional hazards regression analysis.
Results: Estrogen receptor (ER)-positive breast cancer patients receiving adjuvant endocrine therapy with the genotype GG at
CTSO rs10030044 showed significantly shorter disease-free survival (DFS) and overall survival (OS) (P = 0.0024 and P = 0.0003,
respectively). On the other hand, this genotype were not associated with prognosis in ER-negative breast cancer patients.
Multivariate Cox regression analysis revealed that the GG genotype at CTSO rs10030044 was an independent poor prognostic
factor in ER-positive breast cancer patients receiving adjuvant endocrine therapy (OS: RR = 1.86; 95%CI, 1.18 to 2.85).
Univariate and multivariate Cox regression analysis of factors associated with overall survival
tumor size
Nodal status
Grade
ER atatus
PR status
HER2 status
rs10030044
Variables
n (%)
2cm
210 (36)
>2cm
376 (64)
Negative
369 (63)
Positive
216 (37)
1,2
426 (72)
123 (21)
Positive
440 (75)
Negative
145 (25)
Positive
377 (64)
Negative
208 (35)
Positive
55 (9)
Negative
324 (55)
TT+GT
479 (81)
GG
109 (19)
Univariate (P value)
Multivariate (P value)
Multivariate (RR :95% CI)

1 (reference)
<.0001
<.0001
2.268 (1.615-3.181)
1 (reference)
<.0001
<.0001
3.117 (1.984-4.954)
1 (reference)
0.0001
0.970
0.994 (0.731-1.356)
1 (reference)
<.0001
0.069
1.666 (0.962-2.893)
1 (reference)
0.0005
0.224
1.399 (0.812-2.380)
1 (reference)
<.0001
0.0008
0.420 (0.261-0.690)
1 (reference)
0.0282
0.0082
1,860 (1.181-2.853)
The SNP, ZNF423 rs8060157, was not associated with prognosis in this study.
Conclusion: We show that the genotype GG at CTSO rs10030044 is an independent factor indicating poor prognosis in
ER-positive breast cancer patients receiving adjuvant endocrine therapy.

Title: Clonal evolution of the HER2 L755S mutation leads to acquired HER-targeted therapy resistance that can be reversed by
the irreversible HER1/2 inhibitor afatinib
Xiaowei Xu1, Agostina Nardone1, Huizhong Hu1, Lanfang Qin1, Sarmistha Nanda1, Laura M Heiser2, Nicholas Wang2, Kyle R
Covington1, Edward S Chen1, Alexander Renwick1, Tao Wang1, Carmine De Angelis1, Alejandro Contreras1, Carolina Gutierrez1,
Suzanne AW Fuqua1, Gary C Chamness1, Chad Shaw1, David A Wheeler1, Joe W Gray2, Susan G Hilsenbeck1, Mothaffar F
Rimawi1, C Kent Osborne1 and Rachel Schiff1. 1Baylor College of Medicine, Houston, TX and 2Oregon Health & Science
University, Portland, OR.
Body: Background: Targeting HER2 with lapatinib (L), trastuzumab (T), or the LT combination, is effective in HER2+ breast
cancer (BC), but acquired resistance commonly occurs. In our 12-week neoadjuvant trial (TBCRC006) of LT without
chemotherapy in HER2+ BC, the overall pathologic complete response rate (pCR) was 27%. To investigate resistance
mechanisms our lab developed 10 HER2+ BC cell lines resistant (R) to these drugs (LR/TR/LTR). To discover potential predictive
markers/therapeutic targets to circumvent resistance, we completed genomic profiling of the cell line panel and a subset of
pre-treatment baseline specimens from TBCRC006.
Methods: Parental (P) lines and LR/TR/LTR derivatives of 9 HER2+ BC cell line models were profiled with whole exome and RNA
sequencing. Mutations detected in R lines but not in same-model P lines were identified. cDNAs were assessed by targeted
Sanger sequencing. Single cells of the BT474AZ-LR line were cloned and their cDNAs were sequenced. Mutant-specific Q-PCR
was designed to sensitively quantify mutations. Whole exome sequencing (minimum depth 100X) of 17 baseline tumor/normal
pairs from TBCRC006 were performed on Illumina HiSeq.
Results: We found and validated the HER2 L755S mutation in the BT474ATCC-LTR line and the BT474AZ-LR line (30% of
DNA/RNA/cDNA in BT474AZ-LR), in which the HER pathway was reactivated to cause resistance. Overexpression of this
mutation was previously shown to induce L resistance in HER2-negative BC cell lines, suggesting a role as an acquired L/LT
resistance driver in HER2+ BC. Sanger sequencing of BT474AZ-LR single cell clones found the HER2 L755S mutation in every
clone but only in 30% of the HER2 copies. Using sensitive mutant-specific Q-PCR, we found statistically higher levels of HER2
L755S expression in BT474ATCC-P and BT474AZ-P compared to parentals of other HER2+ BC cell lines
(UACC812/AU565/SKBR3/SUM190). These data suggest that this mutation exists subclonally within BT474 parental lines and
was selected to become the more dominant population in the two resistant lines. The HER1/2 irreversible tyrosine kinase inhibitor
(TKI) afatinib (Afa) robustly inhibited growth of both BT474ATCC-LTR/AZ-LR cells (IC50: Afa 0.02M vs. L 3 M). Western blots
confirmed inhibition of the HER and downstream Akt and MAPK signaling in the LR cells by Afa. Sequencing of TBCRC006
baseline samples found the HER2 L755S mutation in 1/17 subjects. This patient did not achieve pCR after neoadjuvant LT
treatment. The variant was present in 2% of the reads, indicating it as a subclonal event in this patients baseline tumor.
Conclusion: Acquired resistance in two of our BT474 LR/LTR lines is due to selection of HER2 L755S subclones present in the
parental cell population. The higher HER2 L755S levels detected in BT474 parentals compared with other HER2+ BC parental
lines, and detection of its subclonal presence in a pre-treatment HER2+ BC patient, suggest that sensitive mutation detection
methods will be needed to identify patients with potentially actionable HER family mutations in primary tumor. Treating this patient
group with an irreversible TKI like Afa may prevent resistance and improve clinical outcome of this subset of HER2+ BC.

Title: Acquired resistance to everolimus occurs independently of mTORC1 inhibition in preclinical in vivo models of ER+ breast
cancer
Neil A O Brien1, Dylan Conklin1, Michaela Ching1, Luo Tong1, Raul Ayala1, Shawnt Issakhanian1, Sara A Hurvitz1, Emmanuelle
di Tomaso2, Ronald Linnartz3, Richard S Finn1, Samit Hirawat2 and Dennis J Slamon1. 1Division of Hematology/Oncology, David
Geffen School of Medicine at UCLA, Los Angeles, CA; 2Novartis Pharmaceuticals Corporation, Cambridge, MA and 3Novartis
Pharmaceuticals Corporation, East Hanover, NJ.
Body: Background: The addition of the selective mTORC1 inhibitor everolimus (RAD001/Afinitor), to exemestane significantly
prolongs progression free survival in patients with advanced hormone receptor (ER+) positive breast cancer. However, despite
these improved outcomes, the majority of patients that achieve an initial clinical benefit will eventually go on to develop
progressive disease. In order to prevent or delay the onset of this acquired resistance, it is essential to identify the mechanisms
by which continued proliferation is being driven in these resistant tumors. In this study, we used reverse phase protein array
(RPPA) technology to compare cell signaling in ER+ breast cancer cell line xenografts that were responsive to everolimus versus
those that had acquired resistance to everolimus.
Materials and Methods: Xenograft models were established from three ER+ breast cancer cell lines containing different molecular
alterations that confer activated PI3K/mTOR signaling; MCF7 and KPL-1 (both PIK3CA mutant) and ZR75-1 (PTEN-null). Mice
were treated daily with either vehicle or 10 mg/kg everolimus until progression. Snap frozen tissue samples (n = 3) were collected
from vehicle control tumors, from everolimus resistant tumors and from everolimus responsive tumors. Whole cell lysates were
prepared from tumor tissues and processed for RPPA analysis. A customized panel of 27 known cancer/PI3K pathway
associated signaling proteins were selected for the RPPA analysis.
Results: In the vehicle controls, significantly higher levels of pAKT, pPRAS40 and p4EBP1 were detected in the PTEN-null
ZR75-1 tumors compared to the PIK3CA mutant tumors, indicating potential variations in intensity of PI3K/AKT pathway signaling
depending on the initiating molecular alteration. Complete inhibition of tumor proliferation was observed in response to everolimus
in all 3 xenograft models for up to 7-weeks of treatment, until tumors showed signs of progression. Significantly lower S6
phosphorylation was observed in tumors responding to everolimus, which was accompanied by feedback activation of pAKT.
Significant inhibition of S6 phosphorylation was also observed in each of the ZR75-1 and MCF7 tumors that were progressing on
everolimus, indicating that although the tumor cells have acquired resistance to the anti-proliferative activity of everolimus,
mTORC1 signaling is still being successfully blocked. Feedback activation of pAKT was observed in the ZR75-1 resistant model,
however no feedback activation of AKT occurred in the MCF7 resistant model. The KPL-1 model showed the expected loss of S6
pathway inhibition upon resistance. No significant signaling changes were observed in any of the other signaling proteins
measured, including ER, HER2, EGFR, MEK and ERK. Ongoing studies to measure the mRNA expression changes in these
tumors using mRNA microarrays may identify signaling pathways that are driving the proliferation of these resistant tumors.
Discussion: These preclinical data show that multiple pathways of resistance can develop in response to long-term everolimus
treatment and resistance can occur despite continued inhibition of PI3K/mTOR signaling. Data from mRNA microarray
experiments will be presented at the time of the meeting.

Title: Low molecular weight cyclin E regulates response to aromatase inhibitors in post-menopausal breast cancer patients
Iman Doostan1, Stacy L Moulder2, Kelly K Hunt3 and Khandan Keyomarsi1. 1University of Texas, Graduate School of BioMedical
Sciences, MD Anderson Cancer Center, Houston, TX; 2University of Texas MD Anderson Cancer Center, Houston, TX and 3MD
Anderson Cancer Center, Houston, TX.
Body: Almost seventy percent of all breast cancer patients have estrogen receptor (ER) positive tumors requiring hormonal
therapy. Aromatase inhibitors (AIs) are considered as the first line hormonal therapy for ER+ post-menopausal patients. However,
resistance to these drugs remains a major challenge in clinic and the biology of such resistance is not clear.
Previous studies have shown that cyclin E pathway, a key regulator in the G1 to S transition of cell cycle, is deregulated in breast
cancer. Full-length cyclin E is abnormally cleaved into low molecular weight isoforms (LMW-E) that renders patients to poor
survival. Here we hypothesize that cyclin E deregulation can confer resistance to AIs. To address this, we engineered aromatase
overexpressing MCF7 cells to overexpress LMW-E under doxycycline inducible promoter. Full-length cyclin E, GFP and empty
vector transfected cells were also generated and used as controls. Our results indicated that AIs inhibited proliferation by
arresting the cells at G1 phase of the cell cycle. However, LMW-E expression significantly enhanced proliferation of the cells
when treated with AIs. In addition, LMW-E bypassed G1 arrest following AI treatment. At the molecular level, AIs decreased
CDK2, pCDK2, and Rb levels and attenuated Rb phosphorylation. However, these effects were completely rescued only when
LMW-E was expressed. Moreover, using an in vitro kinase assay we indicated that AIs decreased CDK2 kinase activity while
LMW-E expression reversed this effect by increasing CDK2 enzymatic activity. Taken together, these results suggest that LMW-E
inactivates Rb protein as a tumor suppressor and renders the cells to bypass G1 checkpoint following AI treatment. In addition,
this study provides early evidence that CDK2 inhibitors could be beneficial in combination with AIs for LMW-E expressing tumors.
Currently we are investigating whether LMW-E can bypass the activity of AIs using inducible breast cancer cell line xenograft. We
are also examining the correlation between cyclin E status and response to treatment in a cohort of patients who received AIs in
the neo-adjuvant setting.

Title: Proteomics studies reveal important pathway and phosphoprotein changes contributing to cancer stem cell properties and
drug resistance of triple negative breast cancer cells cultured in attached and suspension conditions
Xinyu Deng1, Joe Capri2, Huan Ming Hsu1, Morris Kohanfars1, William Luo1, Puneet Souda2, Julian P Whitelegge2 and Helena R
Chang1. 1Gonda/UCLA Breast Cancer Research Laboratory and the Revlon/UCLA Breast Center, Los Angeles, CA and 2Pasarow
Mass Spectrometry Laboratory, NPI-Semel Institute, University of California, Los Angeles, CA.
Body: Breast tumors without ER, PR and HER2 expression are referred to as triple-negative breast cancer (TNBC) and have
been associated with a higher rate of recurrence and distant metastasis compared to other types of breast cancers. Because they
lack a clear therapeutic target, chemotherapy is the only systemic treatment option for TNBC patients. Previous works suggested
that the chemotherapy-resistant residual tumor cells may contribute to the high rate of recurrence and metastasis of TNBC.
Recently cancer stem cells and circulating tumor cells have been reported to have a close relationship with cancer metastasis.
Culturing tumor cells in suspension conditions was also reported to be an efficacious way to enrich cancer stem cells and to
mimic the living conditions for circulating tumor cells. In the current study, four TNBC cell lines HCC1937, HCC1187,
MDA-MB-468 and MDA-MB-231 were cultured in both attached and non-attached suspension conditions. HCC1937 and
MDA-MB-231 cells formed tumorspheres in suspension condition and became more resistant to docetaxel comparing to those in
the attached condition. Although MDA-MB-468 cells were unable to form tumorspheres in suspension conditions they formed
clusters of aggregated cells that were also more resistant to docetaxel in suspension. HCC1187 cells could not form either
tumorspheres or clusters in suspension conditions and were found to be more sensitive to docetaxel. Toward uncovering the
signaling changes in these cell lines in different conditions, we performed a proteomics study to compare the phosphoproteomes
of these cells in both culturing conditions. Totally 2,027 phosphorylated protein groups with no decoy, 5,474 unique
phosphopeptides and 5,711 unique phosphosites were identified by a LTQ-Orbitrap LC-MS/MS system. Pathway analysis
showed that MAPK signaling pathway, Focal adhesion pathway, p53 signaling pathway and MicroRNA pathways were
dramatically changed in HCC1937, MDA-MB-231 and MDA-MB-468 cells, suggesting that these pathways may contribute to the
stem cell properties and drug resistance of these cells. To further find out why HCC1187 behaved differently from the three other
cell lines, we analyzed the phosphoprotein changes between the two groups in both culturing conditions. Interestingly, in the
suspension condition, 17 phosphoproteins down-regulated in the three cell lines were found to be up-regulated in HCC1187 and
one phosphoprotein up-regulated in the three cell lines was found to be down-regulated in HCC1187. More detailed studies of
these functional changes to the phosphoproteome may further elucidate the roles of cancer stem cells and circulating tumor cells
in drug resistance and relapse in TNBC.

Title: Elucidating molecular resistance to trastuzumab using next generation sequencing in isogenic cell models
Abde M Abukhdeir1, Matthew Najor1, Sanja Turturro1, Melissa R Pergande1, Jeffrey A Borgia1, Hanif G Khalak2 and Melody
Cobleigh1. 1Rush University Medical Center, Chicago, IL and 2Weill Cornell Medical College, Doha, Qatar.
Body: A minority of all breast cancers will express increased levels of the ERBB2 protein. They are eligible for
trastuzumab-based therapy. Some will respond, but all will progress. Thus, the problem of resistance to trastuzumab has
generated an urgent need to determine the underlying mechanisms of that resistance.
Cancer is a genetic disease and the mechanism of trastuzumab resistance is likely also genetic in nature. However, the
significant genomic heterogeneity between and within patient tumors greatly complicates the identification of a genetic
mechanism of resistance for trastuzumab. In order to overcome some of these challenges, we looked to an isogenic model of
trastuzumab resistance. We acquired the trastuzumab-sensitive breast cancer cell line, BT474 and two clones of this cell line that
were conditioned to exhibit trastuzumab resistance.
To investigate a possible genetic mechanism of trastuzumab-resistance, we performed whole exome sequencing using Ampliseq
chemistry on the Ion Torrent platform from Life Technologies and paired-end RNA-sequencing on the Illumina HiSeq platform.
Next-generation sequencing data was bioinformatically analyzed using tools that allowed us to filter relevant variants based on
statistical and functional significance. Variants of interest were those that that arose during drug treatment, which were identified
as those in each of the resistant clones, which were novel compared to the parent clone. Proteins from whole cell lysates were
resolved in two dimensions using 3-10 nonlinear strips for isoelectric focusing followed by resolution via 4-20% SDS-PAGE.
Proteins were visualized via Gelcode blue and cored with a biopsy punch, trypsinized, and submitted for protein ID on an LTQ XL
mass spectrometer. We performed functional validation of genetic alterations through the use of somatic cell gene targeting of an
ERBB2-expressing clone of the MCF-10A cell lines, a non-tumorigenic model of breast cancer, which is sensitive to trastuzumab.
Exome sequencing initially yielded more than 10,000 unique DNA variants across the three clones, which after bioinformatic
analysis resulted in 1000 variants of interest. Two-dimensional gel electrophoresis revealed 25-30 differentially expressed
proteins per sample. Correlation between the sequencing and proteomics data provided us with a candidate gene list of less than
100 genes and several cellular pathways related to growth signaling and immunity. Genetic alterations were tested for their ability
to cause trastuzumab resistance in MCF-10A clones.
We describe herein a detailed molecular analysis for a model of trastuzumab resistance. Validated genetic alterations will be
investigated in a unique collection of archival specimens, which we hope will open the path towards the development of novel
agents to augment the effects of trastuzumab.

Title: Sensitivity to c-Met inhibition is increased in dasatinib resistant TNBC cells
Patricia Gaule1, Brendan Corkery1, John Crown2, Micheal J Duffy2 and Norma O' Donovan1. 1National Institute for Cellular
Biotechnology, Dublin City University, Dublin, Ireland and 2St Vincent's University Hospital, Dublin, Ireland.
Body: Pre-clinical models TNBC cells have demonstrated sensitivity to the multi-targeted Src kinase inhibitor dasatinib, however
clinical trials with single agent dasatinib showed limited efficacy in unselected populations. Trials of dasatinib in combination with
chemotherapy are ongoing. In order to study potential mechanisms of resistance to dasatinib in TNBC we established a cell line
model of acquired dasatinib resistance (231-DasB).
The dasatinib resistant cell line (231-DasB) was developed by constant exposure to incrementally increasing concentrations of
dasatinib, from 200 nM to 500 nM over a period of 13 weeks.Cell proliferation was measured by acid phosphatase assay after 5
day treatment with SRC inhibitors (dasatinib, PD180970), EGFR inhibitors (gefitinib and neratinib), chemotherapy drugs
(carboplatin, docetaxel and doxorubicin) and a c-Met inhibitor (CpdA, Amgen). P values were calculated using the Students
T-Test (2 tailed with unequal variance). Expression and phosphorylation of c-Met and Src was measured by immunoblotting and
multiplex magnetic bead assays carried out on a MAGPIX Instrument.
Following approximately 3 months exposure to dasatinib, 231-DasB cells were resistant to dasatinib with IC50 > 5M compared
to 0.04 0.001 M in MDA-MB-231. 231-DasB cells also showed resistance (2.2-fold) to the Src kinase inhibitor PD180970 [table
1]. 231-DasB cells showed small but statistically significantly increases in sensitivity to docetaxel and doxorubicin.
Table 1: Sensitivity to chemotherapy and targeted therapies in the MDA-MB-231 and 231-DasB cells.
Drug
MDA-MB-231
231DasB
p value
PD180970 IC50 (M)
0.40 0.04
0.87 0.07
0.003
CpdA IC50 (M)
>10
2.1 0.1
0.0001
Docetaxel IC50 (nM)
1.9 0.1
1.2 0.1
0.003
Doxorubicin IC50 (nM)
138.6 1.0
97.2 6.6
0.007
Carboplatin IC50 (M)
13.2 0.8
11.5 1.3
0.147
Gefitinib IC50 (M)
23.1 2.2
20.9 2.1
0.289
Neratinib (% growth @ 10M)
94.5 2.0
95.9 7.6
0.834
No significant change in sensitivity to carboplatin or to the EGFR inhibitors gefitinib and neratinib was observed. However, the
231-DasB cells demonstrated a significant increase in sensitivity to the c-Met inhibitor, CpDA, with an IC50 value of 2.1 0.1 M
compared to an IC50 greater than 10 M in the parental MDA-MB-231 cells. Treatment of MDA-MB-231 cells with dasatinib (100
nM) blocked phosphorylation of Src kinase. In contrast, dasatinib treatment (100 nM) did not decrease p-Src levels in the
231-DasB cells. p-Met levels were significantly increased in 231-DasB cells relative to MDA-MB-231. Treatment with 2 M CpdA
decreased p-Met and p-Src in both 231-DasB and MDA-MB-231 cells. Other key receptor tyrosine kinases (EGFR, HER2, HER3,
HER4, IGFIR and IR) show no significant changes in phosphorylation in 231-DasB cells compared to MDA-MB-231.
Constitutive activation of p-Src through increased c-Met signalling may be a potential mechanism of resistance and suggests that
combined treatment with dasatinib and a c-Met inhibitor may block the development of acquired resistance.

Title: Autophagy in three dimensional cultures provides survival advantage against trastuzumab in HER2+ mammary
adenocarcinoma cells
Cristina E Rodrguez1, Sara Reidel1, Elisa D Bal de Kier Joff1, Mara A Jasnis1 and Gabriel L Fiszman1. 1Institute of Oncology
Angel H. Roffo, Ciudad Autnoma de Buenos Aires, Ciudad de Buenos Aires, Argentina.
Body: HER2 is overexpressed in 20-25% invasive breast tumors, and correlates with low free disease survival. Trastuzumab
(Tz), monoclonal antibody anti HER2, is used to treat HER2+ tumors; however more than half of them are resistant or acquire
resistance during treatment. Autophagy has been proposed as a tumoral escape mechanism. Multicellular tumor spheroids
(MCS) are a model of cell growth in 3D that mimics the structure of in vivo avascular tumors. We have previously demonstrated
that MCS present different subpopulations, with a gradient of proliferative, quiescent and apoptotic cells towards the center of the
spheroid and these characteristics makes it more resistant to Tz than monolayers. The aim of this study was to analyze the role of
autophagy in MCS growth and its relevance on the resistance of breast cancer cells to Tz. MCS of overexpressing HER2+ human
mammary tumor cells (BT474 cell line) were cultured as cell suspensions on agar, one per well, and experiments were conducted
when MCS reached 550 um initial size. When the autophagy marker LC3 was analyzed in MCS by Western blot, both LC3-I and
LC3-II were significantly up-regulated compared with cells cultured as monolayers. The functional autophagic flux was confirmed
by immunoblots of LC3 and p62 (SQSTM1/sequestosome1) in cells treated with Bafilomycin A1 (5 nM). MCS were fixed and
included in paraffin to analyze the expression of LC3 and observed a differential localization of autophagic cells, increasing
towards the center of the spheroid, correlating with the hypoxic population previously described. Uppon Tz adittion at a
concentration of 50 ug/ml, a higher and uniform expression of LC3 was found in all the living cells. These observations were
further supported by the finding that p62 was down-regulated in Tz treated spheroids in opposition to controls (commercial IgG).
In 2D, Tz (1 ug/ml) also exerted LC3-II conversion and increase in autophagosomes formation. Autophagy inhibition by
3-methiladenine (3-MA) used at 1 mM, in combination with Tz decreased two fold vs Tz alone in monolayers (49% vs 76% cell
viability respectively, p<0.05). In 3D, the reduction in size induced by the autophagy inhibition plus Tz was 13% vs 10% Tz alone.
3-MA alone did not elicited citotoxicity in 2D or 3D. To investigate the link between apoptosis and autophagy, we exposure
monolayers and MCS to Tz during 6h with 1h of pre-incubation with 3-MA and analyze by Annexin V/propidim iodide. In
monolayers, the combination of 3-MA with Tz enhanced Tz sensitivity since they induced an increase in total cell apoptosis 35%
compared to Tz alone (15% vs 11% early + 21% vs 13% late apoptosis, p<0.05). Surprisingly in 3D, Tz had an opposite effect
and decreased by 30% late apoptosis; however, this Tz-protection against apoptosis was fully reversed by the inhibition of
autophagy. We conclude that Tz exerted a differential effect in BT474 breast cancer cells cultured as MCS, inhibiting apoptosis
and generating a smaller spheroid composed only of remaining Tz-resistant living cells which are autophagic addicted. We
propose that this model could be useful to study the mechanisms involved in resistance to Tz.

Title: Leptin peptide receptor antagonist linked to nanoparticles: A novel adjuvant therapy for triple negative breast cancer
Tia L Harmon1, Adriana Harbuzariu1, Courtney D Dill1, Lily Yang2 and Ruben R Gonzalez-Perez1. 1Morehouse School of Medicine,
Atlanta, GA and 2Emory University, Atlanta, GA.
Body: Background: Obesity and high leptin levels are strongly associated with breast cancer, relapse, drug resistance, and poor
patient outcomes. Over expression of leptin and its receptor, Ob-R, induce cell proliferation, angiogenesis, and metastasis in
Triple Negative Breast Cancer (TNBC). This aggressive form of the disease has no targeted therapy and chemotherapeutics
show several undesirable side effects. We have created a Leptin Peptide Receptor Antagonist, LPrA2, which has been shown to
effectively prevent leptin signaling. LPrA2 was coupled to iron oxide nanoparticles (IONPs) and used to determine its potential
use as an adjuvant to chemotherapeutics.
Methods: IONPs, bound to LPrA2, were confirmed by Western Blot. TNBC cells were then treated with IONP-LPrA2 plus
Cisplatin, Doxorubicin, Paclitaxel, Cyclophosphamide, and Sunitinib. Subsequently, the TNBC cells were analyzed for
proliferation, cell cycle progression, and apoptosis with the Cellometer Vision Image Cytometer .
Results: IONP-LPrA2 was found to cause a greater decrease in DNA synthesis during the S phase of the cell cycle in TNBC cells
than LPrA2 alone. Additionally, IONP-LPrA2 when combined with chemotherapeutics or anti-angiogenic drugs showed synergistic
effects on cell proliferation and apoptosis.
Conclusion: These findings indicate that IONP-LPrA2 may be useful in the prevention and treatment of TNBC. Further
IONP-LPrA2 treatment may increase the efficiency of chemotherapeutics. These results could be particularly relevant for obese
patients, whom show high incidence and the poorest outcome of TNBC.
Acknowledgements: This work was partially supported by the National Institutes of Health and National Cancer Institute Grant
1SC1CA138658-05 and U54 CA118638, and DOD Idea Award BC 123427 to RRGP; and facilities, and support services at
Morehouse School of Medicine (NIH RR03034 and 1C06 RR18386) and NIH/NCRR grant 1G12RR026250-03.

Title: Growth hormone receptor silencing sensitizes triple negative breast cancer cells to chemotherapy
Arunkumar Arumugam1, Ramadevi Subramani1, Sushmita Nandy1, Rebecca Lopez-Valdez1 and Rajkumar Lakshmanaswamy1.
1
Texas Tech University Health Sciences Center, El Paso, TX.
Body: Triple negative breast cancer (TNBC) is an aggressive form of breast cancer, which, accounts for approximately, 15% of
breast cancers diagnosed in women. TNBCs lack expression of estrogen receptor, progesterone receptor and human epidermal
growth factor receptor type 2 (HER2). So, there are no targeted therapies available for TNBCs and chemotherapy is the main
choice for the treatment. Although the tumors respond initially, large numbers of patients develop recurrence due to
chemoresistance. Growth hormone receptor (GHR) is a class I cytokine receptor, which plays vital role in the development of
chemoresistance. Studies have shown that silencing of GHR sensitizes breast cancer cells to chemotherapeutic drugs. In this
study we investigated the impact of GHR silencing in TNBC cells and further analyzed the possible mechanisms associated with
GHR silencing induced sensitization. GHR was silenced using small interfering RNAs in metastatic breast cancer cells MDA MB
231 and MDA MB 468. Molecular analyses were performed to determine apoptosis, cytotoxicity, colony formation, invasion and
migration in GHR knockdown cells. Silencing of GHR induced cytotoxicity and apoptosis in TNBC cells. Migratory and invasive
potentials were drastically reduced in GHR silenced cells. Moreover, epithelial to mesenchymal transition markers were
significantly down regulated by GHR siRNA treatment. GHR targeting significantly increased the efficiency of docetaxel against
TNBC cells. Inhibition of GHR also inhibited the expression of BCRP, which is frequently associated with the development of
chemoresistance in breast cancer. Further, treatment with GH induced the overexpression of drug transporter proteins involved in
chemoresistance. Inhibition of GHR in breast cancer cells reverted the expression of these proteins and sensitized the cells to
docetaxel. These findings support the hypothesis that targeting GHR could have a potential new therapeutic approach to
overcome chemoresistance in TNBCs.

Title: No Show

Title: Obesity confers chemotherapy resistance in triple negative breast cancer cells
Victoria R Lehrmann1, Laura W Bowers1, Andrew J Brenner2 and Linda A deGraffenried1. 1University of Texas, Austin, TX and
2
University of Texas Health Science Center, San Antonio, TX.
Body: Introduction: Obesity is associated with a more aggressive breast cancer and a worse outcome following chemotherapy
treatment. Triple negative breast cancer is a highly aggressive subtype of breast cancer characterized by the absence of
estrogen, progesterone, and human epidermal growth factor-2 (HER2) receptors. This subtype has also been correlated with a
worse prognosis. This study aims to investigate the impact of obesity on triple negative breast cancer cells response to
chemotherapy treatment and elucidate potential therapeutic options to improve response.
Methods: Sera was collected from breast cancer patients at the Cancer Therapy and Research Center at The University of Texas
Health Science Center at San Antonio (UTHSCSA) and pooled according to body mass index (BMI) category (Control:18.5 to
24.9 kg/m2; Obese:30 kg/m2). MDA-MB-231 cells, a triple negative breast cancer cell line, were grown in serum-free media
supplemented with 2% obese or control patient sera to create an in vitro model of obesity. The effects of docetaxel, a
chemotherapeutic agent, on cell viability in the presence of obese versus control sera were examined by MTT. Modulation of
B-cell lymphoma 2 (Bcl-2), cyclooxygenase-2 (COX-2), and mammalian target of rapamycin (mTOR) were evaluated as possible
mechanisms for obesity-induced chemotherapy resistance.
Results: The obese sera significantly reduced sensitivity of the MDA-MB-231 cells to docetaxel. In fact, there was no variance in
the viability level of cells grown in obese patient sera for 96 hours with and without docetaxel treatment. The mechanism for
resistance does not appear to involve Bcl-2 or COX-2, previously implicated in docetaxel resistance. Intriguingly, suppression of
mTOR did provide significant benefit, in agreement with our previous in vivo studies investigating obesity and aromatase inhibitor
(AI) response. The role of the mTOR pathway in modulating therapeutic response is still being investigated.
Conclusion: Exposure to obesity-associated circulating factors confers chemotherapy resistance in triple negative breast cancer
cells. Further research will be conducted to determine the mechanism by which obesity promotes chemotherapy resistance and
whether modulation of the mTOR pathway will provide significant benefit.

Title: Homologous recombination deficiency (HRD) score predicts response to standard neoadjuvant chemotherapy in patients
with triple negative or BRCA1/2 mutation-associated breast cancer
Melinda L Telli1, William Audeh2, Kirstin C Jensen1, Shikha Bose2, Kirsten Timms3, Alexander Gutin3, Victor Abkevich3, Jerry
Lanchbury3, Chris Neff3, Elisha Hughes3, Zaina Sangale3, Joshua Jones4, Richard Wenstrup4, Anne-Renee Hartman4, Pei-Jen
Chang1, Shaveta Vinayak5 and James M Ford1. 1Standford University School of Medicine, Stanford, CA; 2Samuel Oschin Cancer
Institute at Cedars, Sinai Medical Center, Los Angeles, CA; 3Myriad Genetic Laboratories, Inc, Salt Lake City, UT; 4Myriad
Genetic Laboratories, Inc, Salt Lake City, UT and 5Case Western Reserve University School of Medicine, Cleveland, OH.
Body: Background: Defects in the homologous recombination (HR) DNA repair pathway sensitize tumors to therapeutics that
target this pathway. A significant proportion of triple negative breast cancers (TNBC) carry HR defects. Recently three
DNA-based measures (LOH, Abkevich et al.; TAI, Birkbak et al; LST, Popova et al.) have been developed and shown to be highly
associated with BRCA1/2 mutation status and sensitivity to platinum-based chemotherapy in TNBC. Standard chemotherapy
consists of some combination of an anthracycline and cyclophosphamide, with or without a taxane. This study assesses the
association of LOH, TAI, LST and the sum of these measures, the HRD Score, with response to standard neoadjuvant
chemotherapy in patients with TNBC.
Methods: Tumor samples were retrospectively obtained from 45 TNBC patients and 2 BRCA1/2 mutation positive hormone
receptor-positive/HER2-negative breast cancer patients who received anthracycline-based neoadjuvant chemotherapy at
Stanford University Medical Center or Cedars-Sinai Medical Center under IRB approved protocols. Measures of LOH, TAI, LST
were obtained; tumor BRCA1/2 mutation analysis was conducted; and BRCA1 promoter methylation status was determined.
Response was categorized by the residual cancer burden (RCB) score with responders defined as RCB 0 or 1, and
non-responders as RCB 2 or 3. Data were also analyzed using the outcome of pathologic complete response (RCB 0). BRCA1/2
deficiency was defined as either BRCA1/2 germline or somatic mutant, or BRCA1 methylated with loss of the second allele in the
tumor confirmed by LOH at the affected gene. Associations of LOH, TAI, LST, HRD Score, BRCA1/2 deficiency and mutation
status with response to neoadjuvant therapy were evaluated with univariate logistic regression models.
Results: All three metrics showed significant association with response (LOH: p=0.0018; TAI: p=0.0057; LST: p=0.0043) and
pCR (LOH: p=0.0061; TAI: p=0.0044; LST: p=0.047). The sum of the scores, HRD Score, was also significantly associated with
response (p=0.0021) and pCR (0.011). Neither BRCA1/2 mutation status (p=0.78 and p=0.31 respectively) nor BRCA1/2
deficiency (p=0.12 and p=0.34 respectively) were significantly associated with response or pCR in this cohort. When the analysis
was restricted to BRCA1/2 intact samples (n=29) all scores were still significantly associated with pCR (LOH: p=0.019; TAI:
p=0.0046; LST: p=0.013; HRD: p=0.0067), while only the sum was significant for response (p=0.042).
Conclusions:All three measures of HR deficiency, LOH, TAI and LST and the sum of these three metrics, the HRD Score, are
significantly associated with response to standard neoadjuvant chemotherapy in TNBC. This observation is consistent with the
mechanisms of action of doxorubicin and cyclophosphamide as DNA damaging agents, though we have yet to examine cohorts
of patients treated with non-anthracycline-containing standard chemotherapy regimens, such as docetaxel and
cyclophosphamide. The HRD Score could be used clinically to identify patients with increased sensitivity to DNA damaging
therapeutics.


Title: Combination of the PARP inhibitor E7449 with eribulin +/- carboplatin in preclinical models of triple negative breast cancer
Sharon McGonigle1, Jiayi Wu1, Donna Kolber-Simonds1, Natalie C Twine1, Jue-lon Shie1, Noel Taylor1, Sergei Agoulnik1, Zoltan
Dezso1, Shannon McGrath1, Mark Matijevic1, Shanqin Xu1, Galina Kuznetsov1, Mary Woodall-Jappe1 and Kenichi Nomoto1. 1Eisai
Inc, Andover, MA.
Body: Introduction: In a small neoadjuvant study in patients with triple negative breast cancer (TNBC) the combination of
eribulin plus carboplatin was effective, with a pathologic complete response rate of 43% following 4 cycles of treatment.
Significant numbers of sporadic TNBC tumors are deficient in DNA repair capacity and share clinical and pathological features
with hereditary BRCA1 mutant disease. PARP inhibitors have demonstrated synthetic lethality in cancer cells with defective DNA
repair and have therapeutic potential for TNBC. In this study we describe the combination of PARP inhibitor E7449 with eribulin
+/- carboplatin in preclinical models of TNBC.
Methods: E7449, an orally available PARP inhibitor, was administered in combination with eribulin +/- carboplatin to 4 s.c.
xenograft models of TNBC: MDA-MB-436 (BRCA1 mutant, PTEN deficient), MDA-MB-468 (BRCA wild type, PTEN deficient),
HCC1806 and MDA-MB-231 (BRCA and PTEN wild type).
Results and Discussion: Addition of E7449 to eribulin significantly delayed tumor progression in PTEN deficient MDA-MB-468
xenografts. In the BRCA1 mutant and PTEN deficient MDA-MB-436 xenograft model, combination of E7449 with eribulin
enhanced antitumor activity versus eribulin alone. Similar potentiation was observed for carboplatin upon combination with
E7449. Treatment of MDA-MB-436 xenografts with the triple combination of E7449 + eribulin + carboplatin was more efficacious
than any double combination and was well tolerated at the doses examined. In contrast, no significant combination activity was
observed for E7449 plus eribulin in the BRCA and PTEN wild type xenografts HCC1806 and MDA-MB-231, and similarly no
potentiation of carboplatin was observed in an MDA-MB-231 xenograft. Notably, combination activity was observed in the BRCA1
mutant (MDA-MB-436) and PTEN deficient (MDA-MB-436 and MDA-MB-468) xenografts and not in the BRCA and PTEN
wild-type models (HCC1806 and MDA-MB-231). Data from ongoing studies to evaluate the combination activity of E7449 +
eribulin in patient-derived xenograft (PDx) models of TNBC will be presented at the meeting.
Potential biomarkers of sensitivity to the combination are under investigation in both cell line xenograft and PDx models and will
be described.
Conclusion: The addition of E7449 to eribulin +/- carboplatin increased antitumor activity in a subset of TNBC models. Biomarker
studies aimed at a better understanding of the underlying cause of sensitivity are underway. The preclinical data support
assessment of E7449 + eribulin + carboplatin combination therapy in the current phase I/II clinical trial.

Title: The PARP inhibitor niraparib demonstrated activity in patient-derived triple-negative breast cancer xenograft models with
high homologous recombination deficiency (HRD) score
Yan Wang1, Stefano Cairo2, Olivier Deas2, Anne-Renee Hartman3, Joshua Jones3, Alexander Gutin3, Jerry Lanchbury3, Zaina
Sangale3, Cara Solimeno3, Jean-Gabriel Judde2, Kirsten Timms3 and Keith Wilcoxen1. 1Tesaro, Waltham, MA; 2Xentech, Evry,
France and 3Myriad Genetic Laboratories, Inc, Salt Lake City, UT.
Body: Triple negative breast cancer (TNBC), which comprises 15% of all breast cancers, has a poor prognosis and currently
lacks effective treatment. TNBCs are highly proliferative, genomically unstable and share molecular characteristics with that of
BRCA1/2 mutation driven breast cancer. Poly(ADP-ribose) polymerase-1 (PARP) is a key DNA repair enzyme that mediates
single strand break (SSB) repair through the base excision repair (BER) pathway. PARP inhibitors have been demonstrated to
selectively kill tumor cells that harbor BRCA1 and BRCA2 mutations. In addition, pre-clinical and preliminary clinical data suggest
that PARP inhibitors are selectively cytotoxic for tumors with homologous recombination repair deficiency caused by dysfunction
of genes other than BRCA1 or BRCA2.
Niraparib is a potent, orally active PARP inhibitor that is being evaluated in Phase 3 clinical studies for ovarian cancer and BRCA
related breast cancer. Previously, we demonstrated that a subset of basal breast cancer (BBC) patient-derived xenograft (PDX)
models responded robustly to single agent niraparib treatment. To understand the selectivity observed, the samples from a
collection of 37 BBC PDX models have been subjected to homologous recombination deficiency (HRD) analysis. HRD analysis is
a DNA-based assay that is capable of detecting homologous recombination deficiency independent of its etiology. Genome-wide
SNP data was generated from a custom Agilent SureSelect XT capture followed by sequencing on an Illumina HiSeq2500. SNP
data was analyzed using three algorithms (LOH, TAI and LST scores), and the final HRD score is the sum of the LOH+TAI+LAST
scores.
Niraparibs antitumor activity was investigated in patient derived BBC models with various HRD scores. The correlation between
niraparib efficacy, HRD score and BRCA deficiency will be discussed.

Title: miRNAs associated with DNA repair capacity in Puerto Rican women with breast cancer
Jaime Matta1, Clara Isaza1, Carmen Ortiz1, Erick Suarez2 and Luisa Morales1. 1Ponce School of Mediicne and Health Sciences,
Ponce, Puerto Rico and 2University of Puerto Rico, San Juan, Puerto Rico.
Body: BACKGROUND: MicroRNAs (miRNA) are short non-protein-coding RNAs that regulate gene expression at the
post-transcriptional level via binding to 3-untranslated regions of protein-coding transcripts. Some miRNAs have been used as
diagnostic, prognostic and therapeutic markers of breast cancer (BC). It is well established that dysregulation of DNA repair
capacity (DRC) is an important risk factor of BC. However, there is little published information as to what specific miRNAs are
associated with DRC in women with BC. OBJECTIVE: The main objective of this study was to identify candidate miRNAs
associated with dysregulation of DRC in women with BC. METHODS: Plasma samples from 30 BC cases and 30 controls
selected based on their DRC levels (low, high) using a proprietary algorithm. Samples were analyzed for miRNA expression
utilizing protocols from Applied Biosystems (Life Technologies). The miRNA expression profiling was performed utilizing the
RT-PCR TaqMan Array Human MicroRNA A Cards v 2.0 (Applied Biosystems) containing 383 miRNA probes. Single-stranded
cDNA was synthesized from 200 ng of total RNA in 8 Multiplex RT primer pool reactions containing stem-looped RT primers that
were specific to mature miRNAs. U6 snRNA-001973 was selected for normalization based on our own experimental validations.
To quantify the association of the miRNA expression the fold change () was estimated for every detector with the p-values
calculated using the t-test. RESULTS: Candidate miRNAs that showed a statistically significant expression were: miR-146,
miR-34a, miR-221, Let-7b, miR-193b, miR-132, miR-192, miR-21, miR-197, miR-24, miR-26b, miR-29c were identified based on
a false discovery rate of 4%. The results showed that twelve miRNAs differentially expressed in patients with BC. Candidate
miRNAs have been reported associated with the expression of twenty seven DNA repair genes. Two of these genes are part of
the NER pathway which has been identified by previous studies as important in BC. CONCLUSION: Our preliminary data
suggests that differential expression of specific miRNAs might be associated with dysregulation of DRC in BC. The molecular
mechanisms by which miRNAs regulate DNA repair genes remain to be elucidated. However, our results lend further promise to
the concept of miRNAs as a tool to study the regulation of DRC. We see potentialfuture applications in prognosis and therapy of
women with BC. Supported by grants S06 GM008239-20 and 1SCA157250 from the NCI Center to Reduce Health Disparities
and NIH-MBRS Program (NIGMS) and NIH-NIGMS #GM082406 (CO).

Title: Expression of APOBEC3B in primary breast cancer of Japanese women
Eriko Tokunaga1,2, Nami Yamashita1, Kimihiro Tanaka1, Yuka Inoue1, Hiroshi Saeki1, Eiji Oki1, Hiroyuki Kitao3 and Yoshihiko
Maehara1. 1Kyushu University, Fukuoka, Japan; 2Kyushu University, Fukuoka, Japan and 3Kyushu University, Fukuoka, Japan.
Body: Background: Human cancer genomes contain tens of thousands of mutations. APOBEC (apolipoprotein B mRNA editing
enzyme, catalytic polypeptide-like) family of cytidine deaminases normally function in innate immune responses that protects
against retrovirus and retrotransposon propagation. However, these enzymes can also deaminate cytosines in the host genome
and generate C to T mutations. APOBEC3B is overexpressed in several human cancer types, and this overexpression correlates
with the presence of the APOBEC3B mutation signature. Recent studies have demonstrated APOBEC3B as a source of
mutations in various malignancies including breast cancers. However, the relationships between the expression of APOBEC3B in
breast cancer and the clinicopathological features have not been fully elucidated.
Aims: To investigate the expression of APOBEC3B mRNA in primary breast cancers and to evaluate the relationships between
the APOBEC3B mRNA expression and the clinicopathological characteristics and prognosis in the primary breast cancer of
Japanese women.
Methods: Specimens were obtained from 305 patients with primary breast cancers who underwent surgery without neoadjuvant
systemic therapy. APOBEC3B mRNA expression was analysed using quantitative reverse transcription-PCR (qRT-PCR). The
APOBEC3B expression level was normalized to that of the constitutive housekeeping gene TATA binding protein (TBP). Four
breast cancer subtypes were determined by the immunohistochemical analysis of ER, PR and HER2; hormone receptor (HR; ER
and/or PR)+/HER2-, HR+/HER2+, HR-/HER2+(HER2) and triple negative (TN).
Results: Expression of APOBEC3B mRNA was detected in 277 tumors, while it was not detected in 28 tumors. The APOBEC3B
expression was significantly higher in ER-negative, PR-negative, high grade tumors. The APOBEC3B expression was positively
correlated with Ki67 index and highest in TN and lowest in HR+/HER2- subtype. There were no correlations between the
APOBEC3B expression and age, tumor size, lymph node metastasis and the stage. The APOBEC3B expression was not
statistically different between invasive ductal carcinoma and DCIS, suggesting that the APOBEC3B expression is related to the
carcinogenesis of breast cancer. High expression of APOBEC3B was significantly associated with the poor recurrence free
survival in all cases and ER-positive cases; however, APOBEC3B expression was not related to the prognosis in ER-negative
tumors.
Conclusions: The high APOBEC3B expression was related to the aggressive phenotype of breast cancer and the poor prognosis.

Title: A novel role for breast cancer associated protein 2 (BCA2) in regulation replication-stress mediated DNA damage
responses
Yuan-Hao Lee1, Kaushlendra Tripathi1, David Clark1 and Komaraiah Palle1. 1Mitchell Cancer Institute, University of South
Alabama, Mobile, AL.
Body: Breast cancer associated gene 2 (BCA2) has been originally identified from invasive breast cancer cells and shown to be
overexpressed in over 50% of invasive breast cancers. Its expression is known to be highly associated with estrogen
receptor-alpha (ER-) status and promote cell proliferation and invasive properties. Importantly, expression of BCA2 is minimal or
undetectable in most normal cells and tissues, which makes it as a valuable biomarker for ER- positive breast cancers and a
potential therapeutic target. BCA2 protein is a RING and ZINC-finger domain containing E3 ubiquitin ligase that has been shown
to auto-ubiquitylate and interact with several proteins including Rab7, tetherin, ubiquitin, Ubc9 and p21, which are involved in
different cellular processes. However, most of these studies have been focused on tumor progression, migration and invasive
properties and almost no information on its role in carcinogenesis. Since many RING and ZINC-finger domain containing ubiquitin
ligases are implicated in oncogenic signaling and DNA damage responses (DDR), in this study we examined the role of BCA2 in
regulation of spontaneous and chemotherapeutics induced DDR in different breast cancer cell lines (ER- positive versus triple
negative). Interestingly, siRNA mediated down regulation of BCA2 induced spontaneous DDR, such as activation of replication
checkpoint, slow cell cycle progression and double strand breaks (H2AX foci). Exposure of BCA2 knockdown cells to DNA
topoisomerase inhibitors (camptothecin and etoposide) potentiated DDR induced by these drugs. However, the molecular basis
for this enhanced DDR is yet to be determined. Consistent with the previous studies, BCA2 knockdown attenuated cell
proliferation, and compromised migration and invasion properties of these cells. Moreover, this novel role for BCA2 in DDR
strongly suggests its status may also influence tumor response to chemo and radiation therapies and in carcinogenesis process.
Further evaluation of breast cancer cells responses to different chemotherapeutic agents revealed distinct cellular responses
based on the status of BCA2 and ER- status. Taken together, our studies implicate a novel role for BCA2 in regulation of DDR
and its influence on tumor response to chemotherapy. Additionally, we aim to present E3 ligase dependent and independent roles
of BCA2 in these processes and tumor responses to different therapeutic agents.
Acknowledgment: This is work is supported by Abraham Mitchell Cancer Research Scholar Endowment grant.

Title: Granulin, a novel STAT3-interacting protein, promotes breast cancer tumorigenicity
Jennifer E Yeh1, Simion Kreimer2, Sarah R Walker1,3, Andrea Richardson4, Alexander R Ivanov2 and David A Frank1,3.
1
Dana-Farber Cancer Institute, Boston, MA; 2Barnett Institute of Chemical and Biological Analysis, Northeastern University,
Boston, MA; 3Brigham and Women's Hospital and Harvard Medical School, Boston, MA and 4Brigham and Women's Hospital and
Harvard Medical School, Boston, MA.
Body: Since the neoplastic phenotype of a cell is largely driven by its gene expression patterns, increasing attention is focused
on transcription factors that regulate critical mediators of tumor formation and metastatic progression like the oncogenic
transcription factor, signal transducer and activator of transcription 3 (STAT3). Whereas normal cells have transient activation of
STAT3 due to tight control by negative regulators, cancer cells frequently have inappropriate constitutive activation of STAT3
which drives increased expression of genes involved in tumorigenesis. However, little is known about proteins that interact with
STAT3 to modulate its function. To identify novel STAT3-interacting proteins, we performed liquid chromatography tandem mass
spectrometry-based profiling of STAT3-containing complexes immunoprecipitated from the triple-negative breast cancer cell lines
MDA-MB-468 and SUM159PT, which have constitutively active STAT3. We identified granulin (GRN) as a novel
STAT3-interacting protein and validated the STAT3-GRN interaction in breast cancer cells by co-immunoprecipitation. To
investigate the functional effect of GRN on STAT3 activity, we silenced GRN using small interfering RNA. We found that GRN
was necessary for constitutive and maximal cytokine-induced STAT3 transcriptional activity in breast cancer cells. GRN
modulated cytokine-induced STAT3 function by enhancing STAT3 DNA binding and increasing the time-integrated amount of
STAT3 activation and nuclear translocation. Silencing GRN mirrored the effect of silencing STAT3 on reducing the viability,
clonogenesis, and migratory capacity of triple-negative breast cancer cells. Furthermore, GRN mRNA levels were significantly
and positively correlated with STAT3 gene expression signatures indicative of STAT3 activation as well as with reduced overall
survival in breast cancer patients. These studies used a proteomics approach to identify GRN as a novel STAT3 interacting
protein that may serve as an important prognostic biomarker and potential therapeutic target in breast cancer.

Title: No Show


Title: Characterization of a new canine inflammatory mammary cancer (IMC) cell line (IPC-366)
Sara Caceres1, Laura Pea2, Lara C Alvarez de la Cerda3, Maria J Illera1, Richard Larson3, Hui Gao4, Bisrat G Debeb3, Wendy A
Woodward3, James M Reuben4 and Juan C Illera1. 1Facultad de Veterinaria, UCM, Madrid, Spain; 2Surgery and Pathology,
Facultad de Veterinaria, UCM, Madrid, Spain; 3University of Texas MD Anderson Cancer Center, Houston, TX and 4University of
Texas MD Anderson Cancer Center, Houston, TX.
Body: Spontaneous canine inflammatory mammary cancer (IMC) shares epidemiologic, histopathological and clinical
characteristics with the disease in humans and has been proposed as a natural model for human inflammatory breast cancer
(IBC). Few cell lines are available to study IBC, such us SUM 149, SUM 190 and MDA-IBC3. The aim of this study was to
characterize a new cell line from IMC (IPC-366) for the comparative study of both IMC and IBC. Tumors cells from a female dog
with clinical IMC were collected. The pathological diagnosis of IMC was confirmed at the Veterinary Pathology Service of UCMVeterinary Clinical Hospital. The cells were grown under adherent conditions in DMEM/F12 with 5% fetal bovine serum and 1%
antibiotics (streptomycin and penicillin). The growth and mammospheres production capability, and cytological, ultrastructural and
immunohistochemical (IHC) characteristics of IPC-366 were evaluated. Tumorigenicity and metastatic potential of IPC-366 were
also assessed by inoculating the cells on the mammary fat pad of 18 female Balb/SCID mice and the development of tumor was
monitored by imaging and luciferase assay. Microscopic examination of tumor revealed an epithelial morphology with marked
anysocytosis. Doubling time of the tumor cells was approximately 24 h. Under non-adherent conditions, IPC-366 cells formed
mammospheres in approximately 3 days. Cytological and histological examination of smears and ultrathin sections by electron
microscopy revealed that IPC-366 is formed by highly malignant large round or polygonal cells characterized by marked atypia
and prominent nucleoli and frequent multinucleated cells. Some cells had cytoplasmic empty spaces covered by cytoplasmic
membrane resembling capillary endothelial cells, a phenomenon described as vasculogenic mimicry. IHC characterization of
IPC-366 was basal-like: epithelial cells (AE1/AE3+, CK14+, vimentin+, actin-, p63-, ER-, PR-, HER-2 (DAKO, HER2 P4 ),
overexpressed COX-2 and high Ki-67 proliferation index (87.15 %). Imaging and luciferase assay revealed that at 3 weeks after
inoculating the IPC-366 cells, a tumor mass was found in 50 % of mice. At 8 weeks after inoculation metastases in lung, liver and
lymph nodes were found. Xenograph tumors maintained the original IHC characteristics of the female dog tumor. In summary, the
cell line IPC-366 is a fast growing malignant triple negative cell line model of inflammatory mammary carcinoma that can be used
for the comparative study of both IMC and IBC.

Title: Regulation of system Xc- by signal transducer and activator of transcription 3 and 5 in human breast cancer cells
Katja Linher-Melville1, Jennifer Fazzari1, Patrick Gunning2 and Gurmit Singh1. 1McMaster University, Hamilton, ON, Canada and
2
University of Toronto-Mississauga, Mississauga, ON, Canada.
Body: In order to survive and proliferate, cancer cells adapt to high levels of oxidative stress by countering the accumulation of
reactive oxygen species (ROS) with an increased production of intracellular antioxidant molecules such as glutathione. The cell
surface transport system Xc- is a cystine/glutamate antiporter that exports glutamate while importing cystine, thereby mediating
levels of cysteine required for glutathione synthesis and the maintenance of cellular redox balance. Transcription factors that
regulate key antioxidant defense mechanisms, including system Xc-, may therefore be of therapeutic interest. Recently, signal
transducer and activator of transcription (STAT) proteins have emerged as potential targets for the development of novel
anti-cancer therapies. In particular, inhibitors of STAT3 and STAT5 may become clinically relevant as anti-cancer agents for
breast and brain cancer, as well as leukemia. Interestingly, suppression of STAT3 has been linked with increases in ROS and the
induction of apoptosis. Upon activation by phosphorylation, dimerization, and nuclear translocation, STAT proteins
transcriptionally regulate diverse target genes by binding to promoter regions containing gamma-activated site (GAS) motifs. The
human xCT (SLC7A11) gene encodes the functional subunit of system Xc-. We provide evidence that expression of xCT is
regulated by STAT3, and potentially also STAT5, affecting antiporter function in both MCF-7 and MDA-MB-231 human breast
cancer cells. Computational analysis of the xCT promoter region revealed the presence of a distal GAS site. Its truncation
significantly increased luciferase activity in a reporter assay, with similar increases obtained after treating cells transfected with
the full-length xCT promoter construct with various STAT3/5 pharmacologic inhibitors. Knock-down of STAT3 or STAT5A using
specific siRNAs produced similar results, suggesting that these STAT proteins act in a transcriptionally repressive manner. We
also demonstrated binding of STAT3 and STAT5A to the xCT promoter in MDA-MB-231 cells, which was disrupted by
preincubating the cells with specific inhibitors. xCT mRNA and protein levels increased significantly following treatment with
STAT3/5 inhibitors. Pharmacologically suppressing STAT3/5 activation also significantly increased glutamate release and total
levels of intracellular glutathione. We hypothesize that blocking STAT3/5-mediated signaling induces an adaptive, compensatory
mechanism that protects breast cancer cells from ROS by up-regulating Xc- antiporter expression and function. Our findings
suggest that targeting system Xc- may synergize with STAT3/5 inhibitors, heightening their therapeutic anti-cancer effects,
particularly in conjunction with traditional chemotherapy treatments.
This work is supported by the Canadian Breast Cancer Foundation (CBCF).

Title: Exploring the involvement of TTK kinase in centrosome amplification and Her2+ breast cancer
Jamie L King1 and Harold I Saavedra1. 1Emory University, Atlanta, GA.
Body: The centrosome is the cellular organelle responsible for accurate chromosome segregation. In normal cellular function,
regulation of the centrosome duplication cycle in concert with the cell cycle is necessary for accurate passage of genetic
information. However, when modulators of the cell and/or centrosome duplication cycles are deregulated, this process can result
in centrosome amplification (CA, the acquisition of more than two centrosomes), leading to improper segregation of
chromosomes and genomic instability. CA generates low-level aneuploidy (which is tolerated) and polyploidy (selected against
when checkpoints are present). An integral goal in the field of centrosome biology is to characterize how alterations in modulators
of CA influence disease development. In cancer, CA is associated with aggressive tumor types and metastasis and is likely to
affect response to treatment. However, the full role of CA in tumorigenesis is poorly understood. Specifically in breast cancer, CA
is observed in pre-cancerous lesions, which suggests CA in conjunction with genomic instability is an early contributor to
tumorigenesis. In an effort to study mechanisms associated with CA, we are investigating the function of CA in in vitro and in vivo
breast cancer models. Our lab and others have detected that TTK (MPS1) kinase, a proposed modulator of centrosome
duplication, is overexpressed in Her2+ breast cancer cell lines at the mRNA and protein level compared to non-transformed
mammary epithelial cells. We hypothesize that overexpression of TTK leads to CA and that inhibiting TTK expression in cancer
cells will prevent active generation of CA and CIN (chromosome instability), leading to suppressed tumorigenic properties and
further cancer evolution. Preliminary data shows that transient knockdown of TTK via siRNA attenuates the degree of CA in
Her2+ breast cancer cells. Additional preliminary data shows that stable abrogation of TTK via shRNA can inhibit the proliferation
of Her2+ breast cancer cells but does not alter expression of some members in the intrinsic apoptotic pathway. Ongoing studies
will address the impact of altering TTK expression on CA, CIN, downstream centrosomal duplication signaling and tumorigenic
properties of breast cancer cells. The results of these studies will address the mechanistic complexities, such as what
centrosomal signaling pathways associated with TTK are underlying CA in breast cancer. We believe this work will help discern
how deregulated centrosome regulatory molecules influence mammary tumorigenesis and/or responses to treatment.

Title: Combination of IGF1R/FAK inhibition and PI3K/mTOR inhibition in triple-negative breast cancers
LaTonia D Taliaferro-Smith1, Tongrui Liu1, Tiffanie Y Alcaide1, Tanisha Z McGlothen1 and Ruth M O'Regan1. 1Winship Cancer
Institute of Emory University, Atlanta, GA.
Body: Introduction: Triple-negative breast cancers (TNBCs) account for 15%-20% of all breast cancers with limited treatment
options and poor prognosis. The poor outcomes seen with TNBCs are in part due to a lack of viable therapeutic targets.
Overexpression of insulin-like growth factor 1 receptors (IGF1R) and focal adhesion kinase (FAK) are closely associated with
invasive breast carcinomas. However, in our previous study, we found that the combined use of IGF1R inhibitor and FAK inhibitor
had produced limited effects on TNBC cells inhibition. NVP-BEZ235 is a potent PI3K/mTOR dual inhibitor and has been shown to
be effective in TNBC cell lines, especially for the mesenchymal-like and luminal-androgen receptor subtypes. Unlike rapamycin,
which produces a feedback activation of Akt, NVP-BEZ235 alone successfully blocks the Akt activation and effectively inhibits the
cells proliferation. Our hypothesis is that the combined inhibition of IGF1R/FAK and PI3K/mTOR produces greater suppression on
TNBC cell growth. Methods: We examined the effects of NVP-TAE226, the dual inhibitor of IGF1R and FAK, in combination with
NVP-BEZ235 on human MDA-MB-231 and BT549 TNBC cell lines. SRB cell survival assays were performed following
NVP-TAE226 or NVP-BEZ235 treatments alone and NVP-TAE226 in combination with NVP-BEZ235. Western blotting was used
to detect expression and phosphorylation of down-stream signaling proteins and epithelial to mesenchymal transition
(EMT)-related markers. Matrigel invasion chamber assay was performed to evaluate the TNBC cells invasion patterns under
treatments of either NVP-TAE226 or NVP-BEZ235 alone or the two drugs in combination. Spheroid migration assay will also be
used to assess combination effect on the metastatic nature of TNBC cells. Results: The combined IGF1R/FAK inhibition with
NVP-TAE226 and PI3K/mTOR inhibition with NVP-BEZ235 resulted in significantly greater cytotoxicity than either single agent
alone in MDA-MB231 and BT549 cell lines (P<0.05). The combination of IGF1R/FAK and PI3K/mTOR inhibition suppressed the
PI3K/Akt and MEK/ERK signaling cascades, reduced FAK and ZEB1 activity and significantly decreased the cell invasion for
TNBC cell lines (p<0.05). Our data indicated that the combination treatment targeting both PI3K/Akt pathway and EMT related
protein molecules (IGF1R/FAK) lead to greater cytotoxic effect and suppression of EMT and invasion. Conclusion: These results
suggest that the combined inhibition of IGF1R/FAK and PI3K/mTOR may be an effective strategy for TNBC and warrant further
investigation in in vivo animal studies.

Title: Identification and characterization of a new TIMP-1 binding protein
Jos M Moreira1, Mikkel Heberg1, Ulrik Ulrik Lademann1, Birgitte Viuff1, Lena V Jensen1, Jan Stenvang1, Sune B Nygrd1, Maj S
rum-Madsen1, Mette V Vistesen1, Anja T Fuglsang2, Siqi Liu4 and Nils Brnner1. 1Section for Molecular Disease Biology and
Sino-Danish Breast Cancer Research Centre, University of Copenhagen; 2University of Copenhagen and 3Proteomic Analysis
Group, Beijing Institute of Genomics.
Body: Tissue inhibitor of metalloproteinases 1 (Timp-1), is one of the four known endogenous inhibitors of matrix
metalloproteinases (MMPs); in recent years, Timp-1 has become increasingly recognized as a multifunctional protein that,
independently of its MMP inhibitory activity, is able to regulate core cellular processes such as cell proliferation and apoptosis.
Consistent with this pro-survival function, Timp-1 expression was shown to be able to protect cancer cells from epirubicin or
paclitaxel-mediated cytotoxicity. Consistent with this effect, clinical studies have shown high Timp-1 tumor levels to be predictive
of resistance to adjuvant anthracycline-based chemotherapy in metastatic breast cancer patients. In spite of abundant evidence
directly involving Timp-1 in regulation of cell growth and apoptosis, the downstream mechanisms of Timp-1mediated cell
signaling underlying these effects, and its biological consequences, have remained unclear. In order to address this issue, we
aimed to identify cellular binding partners for Timp-1, which may be able to induce signaling. Therefore, we performed yeast two
hybrid screening using a mammary gland cDNA library. We report here the identification of a novel Timp-1 interactor, CD74.
CD74, also known as MHC class II invariant chain (li), was mainly thought to function as an MCH class II chaperone promoting
the exit of MHC class II molecules from the endoplasmic reticulum (ER). However, a fraction of cellular CD74 has been found to
traffic to the plasma membrane where it functions as an accessory-signaling molecule, being quickly recycled back into the
endosomal pathway.
The interaction between TIMP-1 and CD74 was confirmed by co-immunoprecipitation studies in the triple negative breast cancer
cell line MDA-MB-231, and we showed that CD74 is necessary for Timp-1 cellular internalization and Timp-1-mediated activation
of Akt signaling. To determine the applicability of our findings to a broader context, we analyzed a breast cancer patient cohort for
expression of CD74, CD63 and Timp-1 by immunohistochemistry (IHC) and in situ hybridization (ISH). We found that cancer cells
which were negative for TIMP-1 mRNA but positive for Timp-1 protein, indicating an active transport of Timp-1 into the cells.
These results raise the possibility that CD74 may be a useful target for effecting Timp-1 mediated chemoresistance.

Title: Adenanthin, a new peroxiredoxin inhibitor, induces a switch between estrogen receptor alpha-mediated and Src/Akt-driven
signaling in breast cancer cells
Malgorzata Bajor1, Agata O Zych1, Patrick C O'Leary2,3, Anna Czekalska1, William M Gallagher3, Jakub Golab1 and Radoslaw
Zagozdzon1. 1Center of Biostructure Research, Medical University of Warsaw, Warsaw, Poland; 2Helen Diller Family
Comprehensive Cancer Center, University of California, San Francisco, CA and 3Cancer Biology & Therapeutics Laboratory, UCD
Conway Institute, UCD School of Biomolecular and BioMedical Science, Dublin, Ireland.
Body: Increasing evidence indicates that oxidative stress is involved in the progression of estrogen receptor (ER)-positive breast
cancer. A moderate increase in cellular oxidants contributes to the genomic instability and to the change in cellular growth
pattern, which in turn can facilitate progressive transformation of normal cells into cancer cells. Accordingly, the oxidative
stress-related gene expression signature has been suggested to correlate with therapy resistance and poorer outcome in breast
cancer. Therefore, it is crucial to determine the antioxidant defense mechanisms that are utilized by breast cancer cells to
regulate oxidative stress.
Peroxiredoxin 1 (PRDX1) is one of the most prevalent hydrogen peroxide scavenging enzymes in mammalian cells. Our recent
studies indicated that PRDX1 is an independent biomarker of favorable prognosis in ER-positive breast cancer. Our results
indicate the mechanistic link between PRDX1 and ER in breast cancer and suggest a role for PRDX1 in mammary
carcinogenesis. We provide a molecular explanation for this phenomenon in the current project.
To evaluate the importance of PRDX1 activity in ER-positive breast cancer, we have used adenanthin, a newly described
PRDX1/2 inhibitor. In our studies, we have shown that adenanthin strongly inhibits metabolism of exogenous hydrogen peroxide
by breast cancer cells. This phenomenon is accompanied by a shift from H2O2-degrading PRDX1 dimers into enzymatically
inactive monomers and by a dramatic decrease of ER protein presence in the cells. Moreover, we have observed that
incubation of ER-positive breast cancer cells with adenanthin leads to a marked increase in phosphorylation status of proteins
associated with Src-Akt-driven signaling in breast cancer. Thus, our results suggest that PRDX1 can play an important role in
controlling the switch between estrogen receptor- and growth factor-driven signaling in breast cancer.
In summary, in our studies we describe for the first time molecular consequences of rapid dysfunction of PRDX-related system in
ER-positive breast cancer. The deeper knowledge on the mechanisms of PRDX1 functioning can change our understanding of
the events leading to the progression of ER-positive breast cancer and provide new opportunities for pharmacological
interventions in this disease, especially in the context of recent observations connecting the oxidative stress and resistance to
endocrine therapy.

Title: Tyrosine phosphorylation of p27kip1 regulates the activity of cyclin D-cdk4 complexes in breast cancer
Priyank Patel1, Elina Shetyn1, Cindy Gomez1, Susan RS Gottesman2, Angela Tyner3, Benedikt Asbach4, Ralf Wagner4 and Stacy
W Blain1. 1SUNY Downstate Medical Center, Brooklyn, NY; 2SUNY Downstate Medical Center, Brooklyn, NY; 3University of Illinois
at Chicago, Chicago, IL and 4University of Regensburg, Insitute of Medical Microbiology and Hygiene, Regensburg, Germany.
Body: The oncogenes Cyclin D and cdk4 are overexpressed in breast cancer, but the levels of these proteins are not always
accurate indicators of oncogenic activity because p27Kip1 is required to assemble this otherwise unstable dimer. However, p27s
association activates or alternatively inhibits cyclin D-cdk4, serving as a bona fide ON/OFF "switch." Tyrosine (Y) phosphorylation
of residues Y88/89 in p27 displaces its C-terminus from the cdk4 active site, permitting both ATP binding and CAK
phosphorylation of cdk4s T loop. This model leads to the following hypothesis: modulation of p27 pY controls cdk4 activity, which
in turn regulates efficient cell cycle passsage, and in breast cancer where cdk4 activity is deregulated, p27 may be constitutively
switched ON. Deregulated Src Family Kinase (SFK) signaling in cancer may increase p27 pY, constitutively activating oncogenic
cdk4, causing continuous cell cycling. Using our p27 pY phosphospecific antibody, we have shown in primary tumors, that p27 pY
is not detected in benign tissue regions, but is detected in grade 1 and progressively higher grade tumors, suggesting that p27 pY
may be a marker for increased oncogenic cdk4 activity and cdk4 inhibitor sensitivity. We identified an SH3 recruitment domain
within p27 that controls p27 pY, and in turn controls cdk4 activity. Blocking the SH3:p27 interaction with small peptides prevents
p27 pY and cdk4 activity in vitro and in vivo. Using a phage-ELISA assay, we identified PTK6/Brk (Protein Tyrosine Kinase
6/Breast Tumor Kinase) that functions as a high-affinity kinase, able to phosphorylate p27 in vitro and associate with
phosphorylated p27 in vivo. Overexpression of PTK6 in vivo increases p27 pY and increases resistance to specific cdk4 inhibition
by the chemical inhibitor, PD0332991. An ALTernatively spliced form of PTK6 (ALT), which contains the SH3 domain, specifically
associates with p27 in cells arrested by contact or serum-starvation, blocking pY and acting as an endogenous inhibitor of cdk4.
As PTK6/Brk is overexpressed in more than 60% of human breast carcinomas, our data suggest that PTK6/Brk overexpression
facilitates cell cycle progression by increasing cdk4 activity through direct p27 Y phosphorylation. As PD0332991 moves into the
clinic, p27 pY could serve as a marker to identify tumors sensitive to cdk4 inhibition, while blocking the PTK6:p27 interaction with
small molecules represents a novel therapeutic option to inhibit cdk4 activation.

Title: Inhibition of CDK4/6 induces senescence and autophagy in ER positive breast cancers
Smruthi Vijayaraghavan1,2 and Khandan Keyomarsi1,2. 1University of Texas MD Anderson Cancer Center, Houston, TX and
2
University of Texas Graduate School of BioMedical Sciences, Houston, TX.
Body: Deregulation of the cell cycle machinery is a hallmark of breast cancer, facilitating aberrant proliferation that fuels
tumorigenesis and disease progression. This makes the cell cycle, particularly the cyclin dependent kinases (CDKs) an ideal
choice for drug targeting in these tumors. Palbociclib or PD0332991, a potent CDK4/6 inhibitor is a known anti-proliferative agent
that induces G1 arrest and prevents tumor growth in several cancers including ER+ breast cancer. This drug has shown
tremendous success in Phase II clinical trials in ER+ breast cancers and is presently undergoing phase III trials in combination
with letrazole. However, little is know about its precise mechanism of action and modes of resistance in ER+ breast cancers. This
is critical in understanding the biology of treatment response, drug resistance and combination strategies, and this project aims at
addressing these gaps in knowledge.
We have used a series of ER+ breast cancer cell lines to examine the mechanism of action of Palbociclib and identify the nodes
that could mediate resistance to this agent. Our results have revealed that continuous treatment of ER+ breast cancer cells with
Palbociclib results in a G1 arrest with concomitant downregulation of pRb. Treated cells undergo senescence and autophagy, but
not apoptosis. Further, we have shown that the induction of quiescence or senescence and autophagy occurs in a time and dose
dependent manner.
Since this agent is known to be a specific CDK4/6 inhibitor, we next downregulated these kinases in our model system and
subjected them to Palbociclib treatment. Interestingly, downregulation of CDK4 or CDK6 did not significantly alter the sensitivity of
these cells to the anti-tumor effects of Palbociclib. Next we asked if depletion of Rb can render the cells resistant to Palbociclib.
Results revealed that the complete depletion of Rb was sufficient to reduce the sensitivity of ER+ cells to Palbociclib by 4-6 fold.
However, the very steep pattern of the growth response curves suggest that in the absence of Rb, Palbociclib can effectively
inhibit another target, albeit, at a higher drug concentration.
Lastly, to interrogate if deregulation of the G1/S checkpoint can render cells more resistant to this inhibitor, we examined if
overexpression of the low molecular weight isoforms of Cyclin-E (LMW-E), which by themselves are sufficient to induce an
oncogenic phenotype through constitutive phosphorylation of Rb, can mediate resistance in these cells. These results, which
were similar to those from the Rb knockout studies, revealed that LMW-E overexpression can reduce the sensitivity of ER+ breast
cancer cells to Palbociclib by 5-7 fold.
Collectively, our studies suggest that inhibition of G1/S transition is not the sole mode of action of Palbociclib and that there is
another unidentified target, which is specifically inhibited by this agent, when the G1/S check point is compromised. Since the
deregulation of the G1/S checkpoint is likely to occur clinically in patients treated with Palbociclib, identification of this second
target will be instrumental in efforts to overcome resistance and identify biomarkers of sensitivity to this agent in breast cancer
patients.

Title: Nottingham prognostic index plus (NPI+): Validation of the modern clinical decision making tool in breast cancer
Andrew R Green1, Daniel Soria2, Jacqueline Stephen3, Desmond G Powe4, Christopher C Nolan1, Ian Kunkler5, Jeremy Thomas5,
Gill Kerr5, Wilma Jack5, David Camreron5, Tammy Piper5, Graham R Ball6, Jonathan M Garibaldi2, Emad A Rakha1,4, John MS
Bartlett5,7 and Ian O Ellis1,4. 1School of Medicine, University of Nottingham, Nottingham, United Kingdom; 2School of Computer
Science, University of Nottingham, Nottingham, United Kingdom; 3College of Medicine and Veterinary Medicine, University of
Edinburgh, Edinburgh, United Kingdom; 4Cellular Pathology, Nottingham University Hospitals NHS Trust, Nottingham, United
Kingdom; 5Edinburgh Cancer Research Centre, University of Edinburgh, Edinburgh, United Kingdom; 6School of Science and
Technology, Nottingham Trent University, Nottingham, Canada and 7Ontario Institute for Cancer Research, Toronto, Canada.
Body: Introduction
Current management of breast cancer (BC) relies on risk stratification based on well-defined clinicopathologic factors. The
Nottingham Prognostic Index Plus (NPI+) is based on the assessment of biological class combined with established
clinicopathologic prognostic variables providing improved patient outcome stratification superior to the traditional NPI. This study
aimed to validate the NPI+ in an independent series of BC.
Methods
A Validation series of 469 primary early-stage BC cases treated in Edinburgh, UK were matched for size, stage and grade to
cases from Nottingham, UK used to develop the NPI+ (Training series). Adjuvant therapy was similar in both series except that
143 Edinburgh cases received endocrine therapy whilst the matched Nottingham cases had no adjuvant therapy. However, there
was no significant difference in 10 year BC specific survival (BCSS) between the Training and Validation series.
Cases, prepared as TMAs, were immunohistochemically assessed for Cytokeratin (Ck)5/6, Ck18, EGFR, Estrogen Receptor
(ER), Progesterone Receptor (PgR), HER2, HER3, HER4, Mucin 1 and p53 expression. NPI+ biological class based on the
expression of the 10 biomarkers was determined. Subsequent NPI+ prognostic scores were assigned using individual algorithms
for each biological class developed using the Training series incorporating clinicopathologic parameters: positive nodes (including
nodal stage), tumour size, tumour grade (including mitotic index) and PgR. NPI+ biological classes, prognostic scores and
prognostic groups were compared between the Validation and Training series and their role in prediction of patient outcome. A
p-value of <0.01 was considered significant.
Results
As anticipated, there was a comparable distribution of NPI+ biological classes between Training and Validation series: Luminal A,
n=143 (31%) vs n=115 (25%); Luminal N, n=99 (21%) vs n=89 (19%); Luminal B, n=75 (16%) vs n=85 (18%); Basal p53 altered,
n=54 (12%) vs n=72 (15%); Basal p53 normal, n=37 (8%) vs n=53 (11%); HER2+/ER+, n=31 (7%) vs 18 (4%); HER2+/ER-, n=30
(6%) vs n=37 (8%; 2=13.792, p=0.032). BCSS was analogous between the Validation and Training series in each of the NPI+
biological classes except Luminal B (p=0.042). Similar BCSS was observed in the NPI+ Biological classes of the Training versus
Validation series when taking into consideration adjuvant treatment modalities.
The mean NPI+ score was similar between the Validation and Training series (2.30 vs 1.89, Pearsons Regression p=0.079). The
NPI+ prognostic groups significantly predicted patient outcome in each molecular class (BCSS, p<0.0001) in the Validation series
irrespective of adjuvant treatment. Comparing the BCSS in each of the NPI+ prognostic groups demonstrated there were no
significant differences between patient outcome in each of the NPI+ prognostic groups between the Validation and Training
series.
Conclusion
This study validates the NPI+ in an independent series of primary BC confirming its reproducibility. The NPI+ provides improved
individualised clinical decision making for breast cancer for both prediction of clinical outcome and relevant therapeutic options.
Acknowledgements
Funded by the MRC
References
Rakha EA et al Br J Cancer. 2014 110:1688-97.

Title: The CYP19 RS4646 polymorphism is related to the prognosis of stage III and operable stage III breast cancers
Xiying Shao1, Xiaojia Wang1, Yangbing Zheng1, Ping Huang1 and Jian Huang1. 1Zhejiang Cancer Hospital, Hangzhou, Zhejiang,
China.
Body: Background: Aromatase, encoded by the CYP19 gene, catalyzes the final step of the conversion of androgens to
estrogens. In premenopausal women, estrogen is mainly produced by the ovary, while in postmenopausal women, aromatization
of androgen in extragonadal tissue, for example, adipose tissue, is the main source. Given the critical role of CYP19 in estrogen
synthesis and the association of clinical outcome with rs4646 polymorphism, we investigated the clinical relevance of CYP19
rs4646 genotypes in early breast cancer.
Methods: Genotype for CYP19 rs4646 variants was performed on 406 Chinese women with early breast cancer. Associations
were examined between rs4646 genotypes with histopathological characteristics and disease-free survival.
Results: In patients younger than 50 years, women who are homozygous for the minor allele (AA) have a longer disease-free
survival (DFS)compared with those carrying the major allele(CC or AC) (87 months versus 48.7 months; HR 0.560; 95CI
0.318-0.985; P =0.041). This differences was further demonstrated by a multivariate analysis (HR 0.456; 95CI 0.249-0.836;
P=0.011). Conversely, the same variants(AA) were found to be associated with a poorer DFS in women with age above 50 years
(AA versus AC or CC: 38.9months versus 79.7 months; HR 2.758; 95% CI: 1.432-5.313; P=0.002). Furthermore, the differences
was proved by the COX proportional hazards model (HR 2.983; 95CI 1.494-5.955; P=0.002).
Conclusions: The present study indicates that CYP19 rs4646 polymorphism is associated with DFS in early breast cancer and
that the prognosis index of the homozygous for the minor allele (AA) may depend on circulating estrogens levels due to the
different age. The founding is novel, if confirmed in a larger prospective independent cohort, rs4646 genotypes may provide
useful information for routine management in breast cancer.

Title: Expression of the C9Orf72 long-isoform in cancer tissues prognosticates disease-free and breast cancer-specific survival
Nabila Chaher1, Esha Madan2, Clifford Qualls3, Melanie Royce4, Periannan Kuppusamy2, Rajan Gogna2 and Hugo Arias-Pulido2.
1
Centre Pierre et Marie Curie, Algiers, Algeria; 2Geisel School of Medicine, Dartmouth Medical College, Hanover, NH; 3University
of New Mexico, Albuquerque, NM and 4University of New Mexico Cancer Center, Albuquerque, NM.
Body: Background: An expansion of the GGGGCC hexanucleotide repeat in C9Orf72 gene promoter results in reduced mRNA
expression of its long-isoform, and has been associated with amyotrophic lateral sclerosis, and frontotemporal dementia.
Over-expression of C9Orf72 in cultured neuronal cells resulted in increased cell death, but its role in oncogenesis and,
specifically, in brain metastasis is unknown.
Materials and Methods: C9Orf72 mRNA levels were measured by qPCR in 200 normal breast, 217 (cohort 1) and 100 (cohort 2)
breast tumor samples, and correlated with clinico-pathological variables, breast cancer-specific (BCSS) and disease-free survival
(DFS). MCF-7 tumor xenografts models were used to observe the effect of C9Orf72 on tumor metastasis.
Results: C9Orf72 mRNA levels were statistically higher in tumor than in normal breast tissues (P<0.0001). In cohort 1, comprised
of 117 inflammatory and 100 locally-advanced breast cancers, C9Orf72 levels were (1) low in IBC than in LABC (P=0.79); (2)
statistically significant lower in cohort 1 than in cohort 2 (P<0.001); and (3) associated with local and distant recurrences
(P<0.001) in both cohorts. Kaplan-Meier analysis showed that low C9Orf72 levels were associated with both worse DFS and
BCSS in the two cohorts (P<0.0001 for both). Cox proportional hazards model determined that ER status (Hazard Ratio [HR] =
0.65; 95%CI=0.44-0.98; p=0.04), lymph node ratios (HR=1.84; 95%CI=1.19-2.84;p=0.006), and low C9Orf72 levels (HR=37.4;
95%CI=17.08-81.72; p<0.0001) were prognosticators of DFS; and low C9Orf72 levels (HR=7.02; 95%CI=4.1-12.1; p<0.0001)
were associated with worse BCSS in the training-cohort. In the validation cohort, low C9Orf72 levels were associated with worse
DFS (HR=291.1; 95%CI=16.6-5121.1; p<0.0001). Tumor grade (HR=2.5; 95%CI=1.59-3.8; p<0.0001), and low C9Orf72 levels
(HR=6.9; 95%CI=3.6-13.2; p<0.0001) were independent prognostic factors of worse BCSS. Overexpression of C9Orf72 in a
MCF-7 orthotopic mouse xenograft also resulted in significant tumor regression and reduced metastatic events to inguinal, axillary
lymph nodes, pancreas, liver and kidney.
Discussion: C9Orf72 low levels were highly specific and sensitive prognostic factors of worse DFS and BCSS in two
demographically distinct cohorts of breast cancer patients, and increased C9Orf72 expression resulted in reduced metastatic
events in a mouse xenograft model. The molecular mechanisms of C9Orf72-induced cell death are unknown and warrant deep
investigation because of its possible association with brain metastasis. C9Orf72 has the potential to emerge as a very attractive
prognostic biomarker and potential therapeutic candidate in breast cancers.
Senior authors: RG, HAP. This work was partially funded by GlaxoSmithKline ERI grant.

Title: Reduction in raf kinase inhibitor protein predicts poor outcomes and correlates with promoter hypermethylation as well as
matrix metalloproteinases expression in patients with breast cancer
Ji Shin Lee1, Min Ho Park2 and Jung Hang Yoon2. 1Chonnam National University Medical School and 2Chonnam National
University Medical School.
Body: Background: Invasion and metastasis are the direct causes of mortality in breast cancer patients. Reduction in raf kinase
inhibitor protein (RKIP) expression has been shown to be an indicator of metastatic spread in numerous cancers. However, the
role of RKIP alteration in the successive steps of breast carcinogenesis and its association with outcome variables has not been
well established in breast carcinoma.
Materials and Methods: To elucidate the role of RKIP alteration in the successive steps of breast carcinogenesis and its
association with outcome variables, immunohistochemical staining with anti-RKIP antibody was performed in a total of 324
patients with 26 normal breasts, 25 usual ductal hyperplasia, 76 ductal carcinoma in situ (DCIS), and 198 invasive breast
carcinoma (IBC) using tissue microarray. In addition, we studied the promoter hypermethylation of RKIP as a mechanism for loss
of RKIP expression in IBC. To investigate the potential involvement of RKIP in the modulation of matrix metalloproteinases
(MMPs) in breast cancer, we also performed immunohistochemical staining for MMP-1, -2, -9, and -13 in IBC.
Results: RKIP expression appeared to decrease progressively along the continuum of neoplastic changes from normal breast
epithelium to IBC (P < 0.001). Reduced RKIP expression in IBC was significantly higher than in DCIS (P < 0.05). Reduced RKIP
expression was significantly associated with the metastatic relapse (P < 0.001). The patients with reduced RKIP expression had a
significantly poorer prognosis for disease-free and overall survival than those with normal expression (P < 0.001 and P < 0.001,
respectively). Reduced RKIP expression was one of the statistically significant independent risk factors for disease-free survival
(P = 0.003). Reduced RKIP expression in IBC was significantly correlated with promoter hypermethylation (P < 0.05). MMP-2 and
-9 expressions in IBC with reduced RKIP expression were significantly higher than that in IBC with normal RKIP expression (P <
0.05 and P < 0.01, respectively)).
Conclusion: Our results suggest that tumor progression in breast epithelium is accompanied by reduced RKIP expression.
Reduced RKIP expression may serve as a new parameter for the prognostic prediction in patients with IBC. In IBC, RKIP
promoter hyermethylation may be involved in RKIP gene inactivation and RKIP can control tumor cell invasion and metastasis
through regulation of MMPs, particularly MMP-2 and -9.

Title: Prediction of bone metastases of breast cancer using combined markers of bone metabolism and inflammation
Arisa Nishimukai1, Naoya Shibata2, Wataru Kikuchi2, Hiroki Hutawatari2, Hideki Ishihara2 and Yasuo Miyoshi1. 1Breast and
Endocrine Surgery, Hyogo College of Medicine, Nishinomiya, Hyogo, Japan and 2Nittobo Medical Co., Ltd, Koriyama, Hukushima,
Japan.
Body: Introduction
Bone metastases in breast cancer impair a patient's QOL because of skeletal-related events such as bone pain, fractures, spinal
cord compression, and hypercalcemia. It might be important to predict bone metastases and initiate adequate treatment early in
the disease process. Screening and diagnosis of bone metastases are performed using serum markers and imaging systems
such as CT, MRI, PET, and SPECT in postoperative testing. However, a method for predicting bone metastases for stratifying
patients who require treatment has not been established. Although various markers of bone metabolism have been approved for
monitoring of postoperative bone metastases, these are not considered clinically practical because of their low specificity. We
selected TRACP-5b as a marker of bone metabolism; likelihood of bone metastases, and CRP as a marker of inflammation;
likelihood of distant recurrence. We hypothesized that the combination of these two markers of different aspects would provide an
accurate prediction of bone recurrence.
Three hundred forty-nine breast cancer patients who underwent surgery in our hospital between August 5, 2010, and October 31,
2013, were enrolled in this study. Their serum levels of TRACP-5b and CRP were measured in a blinded manner at the R & D
laboratory of Nittobo Medical Co., Ltd. Eighty-one patients were excluded (78 cases; neoadjubant chemotherapy administration, 3
cases; T4), and the data from the remaining 268 patients were included in the statistical analysis. The cutoff values were
380mU/dL for TRACP-5b and 0.016 mg/dL for CRP. Patients with both values above the cutoff value were classified as +/+, and
they were compared with the other patients. The odds ratio between +/+ and the others were calculated using MedCalc statistical
software.
Results
Patients stratified into four classes according to the value of TRACP-5b and CRP: +/+ (n=60), +/- (n=49), -/+ (n=76) and -/(n=83), (+ means above the cutoff value). Eight of the 268 patients had relapsed metastases: three in the bone only, one in the
bone and lung, three in lymph nodes only, and one in the lung only). The Incidence of bone metastases was 5 %(3/60) in the +/+
patients and 0.5 %(1/208) in the others. The incidence was significantly higher in the +/+ patients than in the others(odds ratio:
10.9, 95% CI 1.11 to 106.74, p= 0.040). When the other relapses not including bone metastases were included in the analysis, no
significant difference was observed between the two groups (odds ratio: 0.4, 95% CI 0.02 to 7.07, P=0.513). TRACP-5b
concentration alone could not classify the patients into two groups according to significantly different incidences of bone
metastases(odds ratio: 4.5, 95% CI 0.46 to 43.57, P=0.197).
Conclusion
The results presented here show that the prediction of bone metastases by the combination of TRACP-5b and CRP
concentrations is clinically relevant. We plan to increase the number of patients to provide sufficient statistical power to confirm
this diagnostic potential.

Title: Decrease of tumor F3 expression after neoadjuvant chemotherapy associates to lower survival in breast cancer
Gins Luengo-Gil1, Enrique Gonzlez-Billalabeitia1, Roco Gonzlez-Conejero1, Asuncin Chaves-Benito1, Elena
Garca-Martnez1, Gloria Soler-Snchez1, Elisa Garca-Garre1, Lorena Velzquez1, Vicente Vicente1 and Francisco Ayala de la
Pea1. 1University Hospital Morales Meseguer, Murcia, Spain.
Body: Background:
Pathologic complete response (pCR) is the main prognostic factor after neoadjuvant chemotherapy (nCT) for breast cancer (BC).
However, in cases without pCR, additional prognostic biomarkers are needed for subsequent prognostic and therapeutic
stratification of patients. Tissue factor (F3) is the protease initiator of blood coagulation cascade and is expressed in solid tumors
including BC. F3 oncogenic functions derive both from coagulation activation and from its cytoplasmic domain, although they are
not well known yet. Prognostic impact of F3 circulating levels and tumor expression seems to be variable, and no studies
evaluating F3 expression changes in the residual tumor after nCT are available. Our aim was to evaluate chemotherapy-mediated
changes in F3 mRNA levels and their prognostic value in BC treated with nCT.
Methods:
RNA was isolated from FFPE samples of pre- and post-CT tumors. Post-CT F3 levels were analyzed only in patients without
primary tumor pCR. Quantification of F3 was performed by RT-qPCR. F3 expression was categorized as no expression or
above/below median expression. Change in expression levels (F3), defined as pre-nCT minus post-nCT, was categorized by
value of change (positive vs. negative). Association of F3 levels with clinical and pathological characteristics and analysis of
paired samples was evaluated with non-parametric tests. Kaplan-Meier curves, log-rank test and Cox proportional hazard
regression multivariate models were used for survival analysis. To externally validate our results, we also studied the correlation
between F3 and the rest of the set of genes in the TCGA database. The best 150 directly and 150 inversely correlated genes (r
>0.3 and <-0.3 respectively) were selected and functional prediction was performed using Genemania software for both groups.
Results:
We included 108 consecutive women with invasive BC, mostly with stages IIB or IIIA-C; Her2+: 25.0%, triple negative: 22.2%.
After nCT including anthracyclines and taxanes, pCR rate was 19.4%. nCT significantly increased F3 expression (p<0.000001).
Pre-CT F3 levels were not associated with prognostic or predictive variables in our series. Loss or low post-CT F3 levels were
associated with poor prognosis only in the univariate analysis. However, a nCT-induced decrease in F3 expression had a
negative impact on overall survival in both univariate and multivariate analysis including cN (p=0.001 and 0.013 respectively).
Both in our series and in external databases, F3 mRNA levels have an inverse correlation with proliferative genes, and nCT
enhances these correlations (pre-CT: MYBL2: r=-0.358, p=0.001; MKI67: r=-0.267, p=0.019; post-CT: MYBL2: r=-0.495, p<0.001;
MKI67: r=-0.498, p<0.001). A functional strong inverse correlation between F3 and mitotic functions was also confirmed in TCGA
database.
Conclusion:
Our data demonstrate that nCT consistently increases tumor F3 expression. However, those cases with decreased expression of
F3 after chemotherapy show poor overall survival rates. Functional analysis in our series and in public databases demonstrate
that decreased F3 mRNA expression correlates with an increased mitotic activity, suggesting that low F3 mRNA levels could be a
marker of active and resistant to treatment tumors.

Title: Clinical relevance of TP53 mutations and genomic instability in node positive breast cancer
Vassiliki Kotoula1, Sotiris Lakis1, Zoi Alexopoulou2, Elpida Charalambous1, Kyriaki Papadopoulou1, Aggeliki Lyberopoulou1,
Eleftheria Tsolaki1, Eleni Timotheadou1, Dimitrios Pectasides1, Flora Zagouri1, George Pentheroudakis1, Angelos Koutras1, Helen
Gogas1, Konstantine T Kalogeras1, Ralph M Wirtz3 and George Fountzilas1. 1Hellenic Cooperative Oncology Group (HeCOG),
Athens, Greece; 2Health Data Specialists Ltd, Athens, Greece and 3STRATIFYER Molecular Pathology GmbH, Cologne,
Germany.
Body: Background - Aim: Malignant tumors are currently understood as genetic mosaics. For breast cancer (BC) patient
assessment, the challenge remains to use the increasingly available tumor genomic data along with standard clinicopathological
parameters.
Methods: Histologically reviewed, paraffin tumor tissue DNA samples (N=1771), from patients who had received
anthracycline-based adjuvant chemotherapy in the frame of 3 randomized trials by HeCOG (HE10/00, pre-trastuzumab [T];
HE10/05 & HE10/08, post-T era), were investigated with targeted massively parallel sequencing (tMPS) for variants in 58 genes
previously implicated in BC (custom panel, Ion Torrent systems). Upon multiple stringent quality filters, pathogenic coding
mutations (MUT) and allelic (im)balance (AI, for 5 or more imbalanced SNPs; AI-high: at least 3 AI loci per sample) were
evaluable in 1388 cases (78.4%). IHC4 (HER2 ASCO 2013 guidelines, Ki67 20% cutoff) was used for BC subtyping (luminal A,
luminal B, luminal HER2, HER2-enriched and TNBC).
Results: In total, 1264 tumors carried 1 or more MUT and 788 were positive for AI. No two tumors had the same genetic profile.
Top 5 MUT genes (>5% of all cases) were TP53, PIK3CA, CDH1, GATA3 and MLL3; top 5 genes with >50% frequency within the
tumor (possible drivers) were TP53, PIK3CA, GATA3, MAP3K1 and PTEN. Top imbalanced (unstable) loci included 5p15.33
(TERT), 7p12 (EGFR), 17p13.1 (TP53), 17q21 (BRCA1), 10p15 (GATA3) and 10q23.3 (PTEN). TP53 MUT were more prominent
in HER2-enriched and TNBC (p<0.0001) and were strongly associated with AI (p<0.0001). Tumor genomic characteristics were
modelled with 3-year disease-free survival (DFS) as end-point for the evaluation of tumor aggressiveness; patients without
infiltrated lymph nodes (LN) were not evaluated in this analysis. In all 3 cohorts, 1-3 positive LN were favourable and >3 LN were
unfavourable prognosticators. Among patients with HER2-negative tumors (evaluable N=851), those with >3 positive LN and
AI-low tumors fared similarly to patients with 1-3 LN; those with AI-high tumors fared significantly worse (log-rank, p<0.0001).
DFS for AI-low TNBC was similar to luminal A and luminal B tumors, while AI-high TNBC fared significantly worse (p=0.0007).
Again in HER2-negative patients, TP53 MUT were associated with worse DFS irrespectively of subtype (p<0.0001). Among
patients with HER2-positive tumors (evaluable N=246), DFS for T-treated patients (N=100) without TP53 MUT was similar to
those who did not receive T, while T-treated patients with TP53 MUT fared significantly better (p=0.011); AI-high was an adverse
prognosticator only for patients with >3 positive LN who did not receive T, while this feature did not interfere with outcome in
T-treated patients or in non-T-treated patients with 1-3 LN (p<0.0001).
Conclusions: This prospective-retrospective translational research study, one of the largest presenting tMPS data from paraffin
tissues, confirms the extensive genetic diversity of BC at the individual tumor level. Genomic (in)stability, as assessed with AI and
TP53 MUT status, may help in further defining prognosis in BC patients with early aggressive disease. TP53 MUT are indirectly
shown to predict for T-specific benefit, which merits validation in larger cohorts.

Title: A copy number aberration driven endocrine response gene signature stratifies risk in estrogen receptor positive breast
cancer
Jingqin Luo1, Li-Wei Chang1, Yu Tao1, Jeremy Hoog1, Samuel Leung2, Torsten O Nielsen2 and Matthew J Ellis1. 1Washington
University School of Medicine, St Louis, MO and 2University of British Columbia, Vancouver, BC, Canada.
Body: Background: Many prognostic gene signatures have been developed for estrogen receptor positive (ER+) breast cancer
(BC); however, most have been solely based on mRNA expression data without integrated information on underlying primary
drivers such as genomic aberrations. We therefore coupled gene expression and copy number aberration (CNA) in an attempt to
improve upon prognostic signatures for ER+ BC.
Methods: mRNA expression based discovery was conducted between 172/59 ER+ BC with low/high Ki67 levels after neoadjuvant
aromatase inhibition and significant genes (significance analysis of microarray, q-value less than 0.05) were screened by
correlation (Mann-Whitney-Wilcoxon test P less than 0.05) with CNA using Agilent comparative genomic hybridization array.
Further interrogation on prognosis of relapse-free survival (RFS) by univariate survival analysis (P less than 0.05) in patients
treated with adjuvant endocrine monotherapy from a public data set produced a Copy Number Aberration Driven Endocrine
Response (CADER) signature consisting of treatment sensitive/resistant genes. MetaCore (GeneGo Inc) pathway analysis was
conducted for enriched pathways. We subsequently applied Nanostring nCounter technology to formalin fixed archival tumor RNA
from 620 ER+ adjuvant tamoxifen treated BC (UBC TAM-series) for CADER gene profiling. Patients in multiple independent
public data sets and the TAM-series were classified into treatment sensitive defined by up-regulated sensitive gene centroid and
down-regulated resistant gene centroid by the median cutoffs, treatment resistant defined with the reverse pattern and
indeterminate otherwise. CADER risk stratifications were associated with patient survival outcomes in public cohorts and the
TAM-series. The Kaplan-Meier (KM) analysis and Cox models were used for survival analysis. Published PAM50 intrinsic
subtypes and subtype based risk of relapse (ROR-S) assignments were used (Nielsen CCR 16:5222, 2010).
Results: A 54-gene CADER signature, 27 resistant/27 sensitive genes, was derived. Pathway analysis indicated that CADER was
enriched with sensitive genes of cell survival functions while resistant genes were largely drivers of cell cycle progression.
CADER stratifications were significantly prognostic of relapse free survival (RFS) in all public cohorts (log rank test P=0.05 for all)
and in the UBC TMA-series (P=0.0001, BC specific survival and RFS). CADER showed an additional value (likelihood ratio test
P=0.05) in all cohorts when both standard clinical variables and ROR-S were incorporated in multivariate Cox models. CADER
were highly concordant with intrinsic subtypes and ROR-S (p=0.0001) in all data sets. However, CADER may stratify risk within
ROR-S medium risk patients (P=0.002 METABRIC; P=0.003 and 0.036 in TAM-series for BC specific survival and RFS).
Conclusions: We have developed a signature that is prognostic of long-term survival in postmenopausal BC, further splits risk
within ROR-S medium risk group and identifies some highly resistant BC in presence of ROR-S and clinical variables (see Ellis et.
al. abstract for evaluation of a CADER single sample predictor in the MA12 Phase 3 clinical trial).

Title: Audit of the accuracy of immunohistochemical (IHC) testing of HER2 negative status of breast cancer in the United
Kingdom
Bharat Jasani1, Fiona Campbell2, Phillapa Jones3, Jane Gilbert5, James Dowd5, Keith Miller3, Merdol Ibrahim4, Ian Ellis6, Emma
Hurley7, Mary Falzon8, Barrett-Lee Peter9 and Jane Starczynski5. 1Cardiff University School of Medicine, Cardiff, United Kingdom;
2
Cardiff & Vale LHB, Cardiff, United Kingdom; 3University College, Advanced Diagnostics, London, United Kingdom; 4UKNEQAS
for IHC & ISH, London, United Kingdom; 5Heart of England NHS Foundation Trust, Birmingham, West Midlands, United Kingdom;
6
Nottingham University Hospitals, Nottingham, Nottingham, United Kingdom; 7Source BioScience, Nottingham, Nottingham,
United Kingdom; 8University College London Hospitals NHS Foundation Trust, London, United Kingdom and 9Velindre Cancer
Centre, Cardiff, United Kingdom.
Body: Background: The analysis of the level and distribution of HER2 protein expressed by cancer cells (HER2 status) is of
great clinical value in the management of breast cancer patients both for the determination of the prognosis of disease and for
identification of those patients who are eligible for anti-HER2 therapy. Accurate assessment of the HER2 status is essential for
identifying patients which will benefit from HER2 targeted therapy. HER2 status in the UK is established using a two tier strategy
with IHC as the initial test and subsequent reflex of equivocal results to in situ hybridization (ISH). IHC staining of the HER2
protein is graded as 0, 1+, 2+ or 3+ dependent upon the intensity of staining, cellular localisation and the percentage
of cells positive in accordance with CAP/ASCO and UK guidelines. HER2 3+ cases are considered as positive, with HER2 2+
cases (equivocal) retested by ISH to ascertain the gene amplification status. Cases that are scored as 0 and 1+ by IHC have no
additional testing and are classed as negative. The literature indicates that a subset of these IHC negative cases show HER2
gene amplification by FISH (range 1.1-11.5%). The aim of this audit is to evaluate the discordance rate of HER2 IHC negative,
FISH positive breast cancer in the UK, with a secondary objective to resolve if this is related to the choice of antibody and assay
platform used.
Materials and methods: This audit selected a total of 600 sequential cases reported as HER2 negative on IHC, from three UK
reference centres receiving cases from 29 different hospitals. The cases were given a unique identifying number and
annonymised. Each of the three centres used a different IHC method for frontline HER2 testing with centre one using
HercepTestTM (DAKO), centre two Pathway 4B5 (Roche), and centre three, Oracle (Leica Microsystems). HER2 gene
amplification status was determined using dual colour FISH analysis, PathVysion (ABBOTT) fluorescence ISH (FISH)
in a single centre to provide standardised methodology and assessment. HER2 was classed as amplified when the HER2/CEP 17
ratio was two or greater in accordance with UK guidelines. All cases which showed discordance between IHC and FISH were
re-tested with each of the HER2 IHC platforms to discover whether these are truly discordant results or if the discrepancy is a
consequence of the choice of antibody.
Results: 16/600 (2.8%) unequivocal HER2 gene amplification (mean ratio >2.0) whilst 8/600 (1.2%) had borderline
amplification status(mean ratio = or <2.0). The overall assay specific discordance rates were 3.0% (HercepTest), 2.5%
(4B5) and 3.0% (Oracle), respectively.
Conclusion: The observed level of discordance is well within the range of discordance rates reported by previous
studies. The discrepancies could be due to inadequate quality fixation and/or inadequate sensitivity of the assay
platforms used, or under scoring. A detailed analysis of possible assay related source of discrepancy is currently
underway by repeating the analyses of the 24 discordant cases using like for like three assay platforms at an
independent expert centre.


Title: TRAIL receptor agonists target basal B triple negative breast Cancer (TNBC) that expresses vimentin and axl
Ciara C O'Sullivan1, Laleh Amiri-Kordestani1, Sara Sinclair2, Kathryn J Chavez1, Jennifer L Dine1, Brandon Stone1, Stephen M
Hewitt1, Seth M Steinberg1, Sandra M Swain3 and Stanley Lipkowitz1. 1National Cancer Institute, Bethesda, MD; 2Eastern Maine
Medical Center, Bangor, ME and 3Washington Cancer Institute, Washington, DC.
Body: Background:
Tumor Necrosis Factor Related Ligand (TRAIL) triggers apoptosis by binding to cell surface receptors. We previously showed that
TRAIL receptor agonists preferentially kill TNBC cells with mesenchymal features (Basal B cell lines) through activation of TRAIL
Receptor 2 (TRAIL-R2). We used preclinical models to identify predictive biomarkers of TRAIL sensitivity and tested expression of
these markers in TNBC patient (pt) samples.
Methods:
We tested sensitivity to the TRAIL-R2 specific agonist drozitumab in vitro using cell viability and caspase assays. We examined
expression of the mesenchymal proteins Axl and vimentin(vim) in breast cancer cell lines including TNBC by immunoblotting and
using commercially available cDNA microarray data sets. Next, we performed an exploratory analysis on IHC tumor expression of
vim and Axl in 53 African American pts with TNBC diagnosed between February 2003 and February 2009. In a retrospective
cohort study, overall survival (OS) was calculated from date of surgery until date of death or last follow up. Disease-free survival
(DFS) was calculated from the date that the pt was identified as being disease free until date of recurrence or date of last followed
without recurrence. The significance of the difference among Kaplan-Meier curves was determined by a log-rank test. Axl, vim,
and age at diagnosis values were divided approximately into quartiles based on data from all available pts before being used in
actuarial analyses. A Cox proportional hazards analysis was also performed to determine if Axl or vim retained prognostic value
after adjusting for other factors jointly associated with outcome. All p-values were two-tailed.
Results:
As previously demonstrated with TRAIL, drozitumab selectively killed Basal B TNBC cell lines. Gene analysis and protein
expression demonstrated that vim and Axl were selectively expressed in drozitumab sensitive Basal B cells. Analysis of cDNA
microarray data sets showed that approximately 40% of TNBC express high levels of both Axl and vim. IHC confirmed that
expression of Axl and vim seen on cDNA microarray was in TNBC tumor cells. In an exploratory analysis of the relationship of vim
and Axl expression to OS and DFS, Axl, vim, stage, and response to neoadjuvant chemotherapy were factors found to be
potentially associated with OS in univariate analyses while Axl, vim, age and stage were associated with at least trends towards
significance with respect to DFS in univariate analyses. Cox models showed that higher vim levels (p=0.08) and stage I and II
disease (p=0.024) were potentially associated jointly with OS, while higher Axl levels (p=0.05), age (p=0.016) and stage I and II
disease (p=0.0007) were jointly associated with DFS.
Conclusions:
Preclinical data suggest that expression of vim and Axl can identify TRAIL Receptor agonist sensitive TNBC cells. Based on
microarray and IHC, a subset of TNBC tumors express these markers in tumor cells. In our exploratory analysis with limited pts,
greater vim and Axl expression were associated with a trend towards better DFS and OS. Vim and Axl may be useful predictive
biomarkers for identifying TNBC pts in whom to test TRAIL receptor agonists.
Research funding: Safeway Foundation and National Cancer Institute.

Title: Risk of recurrence estimates with IHC4 are tolerant of variations in staining and scoring
Andrew Dodson1, Lila Zabaglo2, Belinda Yeo1, Keith Miller3, Ian Smith1 and Mitch Dowsett1. 1Royal Marsden Hospital, London,
United Kingdom; 2Institute of Cancer Research, London, United Kingdom and 3UK NEQAS for ICC & ISH, London, United
Kingdom.
Body: Aims
1. To determine the degree to which alterations in immunohistochemical (IHC) staining for ER, PgR, and Ki67 affect IHC4 scores
and distant recurrence estimates.
2. To examine the level of scoring precision needed for accurate use of the IHC4 score.
Background
The IHC4+clinical (C) score combines assessment of protein expression levels of ER, PgR, HER2 and Ki67 with
clinicopathological parameters to identify breast cancer patients at low, intermediate or high risk of distant disease recurrence so
aiding treatment decision-making (Cuzick J, et al. JCO 2011;29:4273).
The score is used in clinical decision making in our hospital (RMH). Our published studies have shown it provides information that
improves decision-making on adjuvant chemotherapy in >65% of patients (Barton S, et al. BrJCancer 2012;106:1760; Yeo BJ, et
al. AnnalsOncol 2014;25suppl_1:i2).
Despite its low cost and wide availability reported use of IHC4+C in other institutions remains limited (Lakhanpal R, et al. JCO
2014;32:abstr2549); one explanation is the perception that IHC-based methods and the assessment of them lack precision,
reproducibility and portability. We have examined the effect of decentralized testing and easily reproduced estimate-based
scoring methods on the IHC4 score to determine its suitability for wider adoption.
Methods
A TMA was constructed from 30 breast cancer cases representative of those for which IHC4+C is requested at RMH. Sections
were distributed to three centers that undertake diagnostic breast cancer IHC and that use reagents and platforms from Dako,
Leica or Ventana who provide most IHC for oncology globally. Centers carried-out staining using their standard procedures and
returned slides for central assessment. Results were compared against those obtained at RMH using standardized methods
previously described (Barton S, et al. BrJCancer 2012;106:1760), and were used to calculate IHC4+C and 10-year distant
recurrence probability (%DRprob) scores. In parallel the TMA slide stained at RMH for ER was scored by a variety of simplified
non-counting based methods. Results were compared with those produced by counted H-Score when used to calculate IHC4+C
and %DRprob.
Results
There was a high-degree of correlation between individual IHC results produced by all three external centers and those of RMH,
and of the IHC4 and %DRprob scores derived from them (Tables 1 & 2).
Scoring method for ER could be adapted to require less precision without significantly affecting IHC4+C and %DRprob results
(correlation coefficients range: 0.90 - 0.98).
IHC4
RMH
Dako
Dako
0.92
Leica
0.89
0.95
Ventana
0.93
0.96
%DRprob
Leica
RMH
Dako
Leica
0.97
0.95
Table 1. Correlation coefficients for all pair-wise comparisons
0.98
0.98
0.98
0.98
0.98
Mean
Median difference
RMH
Dako
RMH
9.9%
Dako
8.4%
1.0%
Leica
8.9%
0.9%
0.5%
Ventana
9.1%
0.6%
0.5%
Leica
0.6%
Table 2. Showing for each center, the mean %DRprob score and median value obtained when the absolute difference was
calculated between matched scores
Conclusion
The IHC4+C algorithm is tolerant of variation in staining and ER-scoring method used. Although additional comparative studies
are required to confirm them, these data support the use of IHC4+C in routine clinical practice outside its institute of origin.

Title: The LiquidBiopsy in metastatic breast cancer (MBC): A novel diagnostic platform for next generation sequencing (NGS) of
circulating tumor cells (CTCs)
William Strauss1, Jessamine Winer-Jones1, Laura Austin2, Paul W Dempsey1 and Massimo Cristofanilli2. 1Cynvenio Biosystems,
Inc, Westlake Village, CA and 2Thomas Jefferson University, Kimmel Cancer Center, Philadelphia, PA.
Body: Background: Effective treatments for advanced breast cancer are limited leaving palliative care as the only option for
patients. The development of effective treatments has been challenging in part due to the biologic heterogeneity of the disease.
The challenges to improve patient outcomes are represented by the availability of diagnostic technologies allowing real-time
access to the genetic drivers of disease in each individual patients and the ability to match such information with novel targeted
therapies. Blood based tests have a long history of providing real-time evidence in medicine. Thus blood derived templates could
fill this role in genetic diagnosis. To perform, any technology for molecular analysis of tumor cells from blood must overcome
limiting amounts of template and high signal to noise. Current efforts are directed to develop blood based tests for circulating
biomarkers that can describe the molecular changes in breast cancer. We investigated the ability to productively sequence tumor
cells recovered from whole blood.
Methods: EpCAM expressing Circulating tumor cells (CTCs) were recovered from 49 patients with metastatic breast cancer on
the LiquidBiopsy platform (Cynvenio Biosystems, Inc). CTC events were characterized using a nuclear stain (DAPI),
Cytokeratin expression and absence of CD45. DNA template recovered from this population was case control sequenced with the
Life Technologies AmpliSeq targeted resequencing panel on an Ion Torrent Platform using a CLIA validated sequencing pipeline
and a sensitivity of 1%. Somatic single nucleotide variants (SNV) present in CTC but not the germline were identified. Critically,
the false positive score for 31 normal donors thus evaluated was zero.
Results: Circulating target cells were detected in all 49 patients with metastatic BC (CTC Median 28, range 2 - 4098). Upon
resequencing, shared germline polymorphisms were observed in CTC and germline populations. In contrast, SNV were detected
in EpCAM enriched populations but not germline in 32% of samples (allele frequency 1.0%-49.6% with 2417-22068 x coverage).
Enumeration of cytokeratin expressing cells was predictive of neither the presence nor the allele frequency of detectable SNV.
Interestingly mutations in the gene for PIK3ca accounted for 20% of the mutations observed including variants at K111E, E542K,
E545K, and H1047L. Mutations were also observed in TP53 (8%), and MET, FGFR2, SMAD4. Twenty samples had tissue biopsy
sequence data available and of those, 11 contained mutations that were evaluable in the CTC. All but one mutation were absent
in the evaluable tissue biopsy
Conclusions: CTC recovered by EpCAM selection directly from blood can be sequenced by NGS. The predictive value of the CK+
phenotype as a surrogate marker for mutation bearing cells is not supported by these data. Even with a relatively narrow 50 gene
target platform, cancer relevant SNVs were detected in >30% of the patients, making targeted NGS of CTCs a promising
approach for biomarker discovery and validation in MBC.

Title: No Show

Title: Correlation between germline and tumor CYP450 2D6 gene polymorphisms
Mehdi Dehghani1, Anneliese O Gonzalez2, Neda Hashemi-Sadraei2, Songlin Zhang2 and Kevin P Rosenblatt1,2. 1CompanionDx
Reference lab, Houston, TX and 2University of Texas Health Science Center, Houston, TX.
Body: Tamoxifen is used for the treatment and prevention of ER (estrogen receptor)-positive BrCa (Breast Cancer). Several
studies have shown that genetic variations in the CYP2D6 gene and/or drug-dependent inhibition of CYP2D6 enzyme activity
(drug-drug interactions or DDIs) are associated with significant reductions in the circulating levels of endoxifen the active
tamoxifen metabolite. These studies demonstrated that changes in CYP2D6 metabolism, as predicted from CYP2D6 genotyping,
affect efficacy. Recent negative results with regards to CYP2D6 genotyping from two large studies suggest that testing for
CYP2D6 status has no practical clinical value. Tumor DNA was used for genotyping in a large fraction of the BrCa patients in
these studies. Genetic bias may result when CYP2D6 genotyping is carried out on the tumor (somatic) genome rather than the
host genome (germline DNA) since it is the host genome that determines CYP450 enzyme activity within the liver and GI tract.
Also expanded CYP2D6 polymorphism coverage, using a more comprehensive genotyping panel, may improve risk stratification
when using CYP2D6 genotyping as a prognosticator in BrCa patients treated with tamoxifen. We hypothesize that BrCa tissues
will harbor numerous mutations, due to a mutator phenotype inherent in most cancers, including within the CYP450
family of genes and, specifically, within the CYP2D6 gene. We expect that comparisons between germline DNA isolated
from peripheral blood, and mutated DNA isolated from BrCa (somatic DNA) specimens within the same patient will
reveal extensive differences in CYP2D6 genotypes.
The aim of this study, is to extensively genotype 70 BrCa patients using a retrospective cohort with matched blood and tumor
tissue from an existing biobank at UTHealth. We plan to perform genotype to phenotype conversions on each patient, for both
germline and somatic DNA. We will look for discrepancies in genotypes and phenotypes and determine the magnitude of genetic
bias possible in such cohorts.
Method: DNA samples were extracted from matched archived tumor cells, dissected by laser microdissection microscopy and
blood. Genotyping was performed by clinically validated Taqman discrimination assays on the most common alleles for
CYP2D6. We also studied the genotype status of these paired samples for CYP2C9, VKORC1, Factor II, Factor V, MTHFR,
CYP3A4 and CYP3A5 genes. CYP2D6 gene copy number and gene rearrangement with CYP2D7 pseudogene were also
assessed by Taqman copy number assays at 3 three different sites within gene. Genotype-phenotype conversion was performed
using an in-house developed, clinically validated genotype-phenotype translator package. Genotype agreement was assessed
between the two DNA sources.
Results : Noticeable non-concordant results between DNA from breast tumors and blood were observed in the genotyping of
polymorphisms in the CYP2D6 gene. However, strong agreement between DNA from breast tumors and blood was detected in
the genotyping of polymorphisms in all other studied genes.
These results suggest that previous publications refuting the association between CYP2D6 genetic polymorphisms and tamoxifen
efficacy could have reached an inaccurate conclusion due to genetic bias and study design. Further research in this biomarker
area is needed.

Title: Effect of prolonged cold ischemic time on immunohistochemical testing of estrogen receptor, progesterone receptor and
HER2 expression in breast cancer
Tae-Kyung Yoo1, Hyeong-Gon Moon1, Jisun Kim2, Jun Woo Lee3, Min Kyoon Kim1, Eunshin Lee1, Jongjin Kim1, Wonshik Han1,
In-Ae Park4 and Dong-Young Noh1. 1Seoul National University College of Medicine, Seoul, Korea; 2University of Ulsan College of
Medicine, Asan Medical Center, Seoul, Korea; 3Ewha Womans University School of Medicine, Seoul, Korea and 4Seoul National
University College of Medicine, Seoul, Korea.
Body: Background: Estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2)
are the most important predictive and prognostic biomarkers in breast cancer. The American Society of Clinical Oncology/College
of American Pathologists recommends that the time from tumor removal to fixation (cold ischemic time) should be kept within 1
hour. Through this study we mean to review the actual cold ischemic time in real practice and analyze whether delayed formalin
fixation effects immunohistochemical (IHC) testing results.
Methods: Patients, who received surgery for invasive or in situ breast cancer in Seoul National University Hospital, Seoul, Korea
between February and December 2013, were retrospectively reviewed. Cold ischemic time was calculated by extracting the time
of formalin fixation from the time when surgery ended. All patients were equally divided into two groups (short ischemic group,
long ischemic group) according to median cold ischemic time. Chi-square test was done for ER and PR positive/negative (0%
negative, 1% positive) and student t-test was done for ER and PR percentage. Also 2 test was done for HER2 positive/negative
and scoring system ranging from 0 to 3+.
Results: A total of 615 patients were included in this study. The median cold ischemic time was 2h 43min 4sec (range 6min 36sec
84h 26min 20sec). Only 48 patients had a cold ischemic time shorter than 1 hour. No association between ER, PR expression
and cold ischemic time was found in the 2 test (p=0.581, p=0.954) and student t-test (p=0.648, p=0.978). As for HER2
expression, in the long ischemic group, there were significantly more patients with positive immunohistochemical testing results
(2 test, p=0.016), and significantly higher grades in HER2 scoring system (2 test, p=0.022). Compared to IHC results, FISH
testing for HER2 amplification showed no significant difference (2 test, n=145, p=0.500). This tendency was persisted when
patients were divided into four groups by 25 quartile of cold ischemic time.
Conclusions: Our findings show that the actual cold ischemic time in practice is longer than recommended guidelines. Despite
that, ER, PR expression was not associated with cold ischemic time. As for HER2 expression, longer cold ischemic time was
associated with more HER2 positive and higher HER2 score. But this tendency was not showed in FISH testing for HER2
amplification.

Title: Comparison of the parallel quantitative expression profiling in paired FF and FFPE samples with IHC4 by using two
prespecified prognostic models
Xiaoyan Huang1, Genhong Di1, Xiaojing Xu2, Bingyu Sun2, Rui Bi3, Wentao Yang3, Rui Li2, Guangyu Liu1, Jiong Wu1, Zhenzhou
Shen1 and Zhimin Shao1. 1Fudan University Shanghai Cancer Center, and Shanghai Medical College, Fudan University,
Shanghai, China; 2Shanghai Biotechnology Corporation, National Engineering Center for Biochip at Shanghai, Shanghai, China
and 3Fudan University Shanghai Cancer Center, and Shanghai Medical College, Fudan University, Shanghai, China.
Body: Introduction: Breast cancer is one of the leading malignancies and causes of cancer death for women now in China
especially in the developed cities. Early diagnosis and adjuvant therapy have helped to improve the survival of the patients. The
technique limitation slowed the pace of development of gene expression profiling. QuantiGene Plex(QGP) assay of bDNA
technique without requiring RNA extraction, cDNA synthesis or PCR amplification which is suitable for the FFPE samples
analysis. QGP could quantify the expression of RNA in FFPE tissues. Our primary purpose was to verify the value of the QGP
platform in conjunction with multi-analysis in paired FFPE and FF tissues. The second purpose is focusing the key genes (ESR1,
PGR, ERBB2 and MKI67) expression signature in early breast cancer patients (T1-2N0-1M0), and compare the gene expression
profiling with the immunohistochemistry(IHC) corresponding panels by using two prespecified models-power and linear, based on
the different combinations of the variables (FF4_P, FF4_L, FFPE4_P, FFPE_L, IHC4_P, IHC4_L).
Methods: We retrospectively selected 103 paired archival tumor blocks and fresh frozen samples consecutively from the fresh
tissue bank from the year 2006 to now. All the tissues are without neoadjuvant treatment. The tumor specimens with local
recurrence or distance metastasis had the priority. The QGP data were harvested from a paralled bDNA assay with multiplex
capability of the Luminex platform by coupling xMAP fluorescent beads. We compared the QGP result of FFPE samples with FF
and IHC results head-to-head. We used automatic IHC staining instrument (BenchMark XT) to retest all the Ki67 staining. We
divide the total sample into training and test groups. According to the data from the training group to create the two models, and
through the cross-validation to build a model based on a subset of the provided data, then calculate their average value to create
the final model. And then through the test group on the construction of group training model validation.
Results: The storage time of the FF and FFPE samples used in our study ranged from 20.5 to 96.6 months. There is no different
expression in each storage times intergroup and intra-group. FFPE and FF of QGP assay (single gene markers ESR1, PGR,
HER2, and MKI67) correlated with the IHC result. The two models created the different scores. Both the power and linear models
scores of FF4, FFPE4 and IHC4 are the prognostic factors in univariate and multivariate analysis of disease free survival with
significant statistic(p<0.01), but in the bootstrapping resampling the FFPE4_L was borderline (p=0.068). The AUC of ROC in the
test group is from 0.760-0.820 of each scores, and there is no statistic difference between each AUC.
Conclusion:The multiplex bDNA assay is a reliable and accuracy technique, and could bridge the gap among the FF tissue to
FFPE. The both models of QGP could be complement of the immunohistochemistry panels and provide more reproducible
results. The parallel multiplex QGP could leading to a great potential for translational cancer research for future study and resolve
the complex pathological interpreting.


Title: Low influence of tumor cell content on mRNA expression levels of ESR, PGR, HER2 and KI67 when performing the
MammaTyper RT-PCR kit
Ralph M Wirtz1, Tilman Rau2, Mark Laible3, Kornelia Schlombs3, Sotirios Lakis1, David Wachter2, Ugur Sahin3 and Arndt
Hartmann2. 1STRATIFYER Molecular Pathology GmbH, Cologne, Germany; 2Institute for Pathology, University Clinic Erlangen,
Erlangen, Germany and 3BioNTech Diagnostics GmbH, Mainz, Germany.
Body: Background
Breast cancer patient management relies on approximations of molecular subtypes by immunohistochemical staining (IHC) of ER,
PgR, Her2/neu and Ki-67. However, routine application of IHC is subject to important pre-analytical, analytical and interpretational
variations which result in significant inaccuracy (Hammond et al. 2010, Polley et al, 2013). In contrast, quantification of biomarker
RNA expression by RT-qPCR using the MammaTyper RTqPCR kit displayed a high concordance of single marker results of
100 % HER2, 96.8 % ESR1, 97.2 % PGR and 97.6 % KI67 based on predefined cutoffs (see Laible et al, abstract submitted).
However, varying tumour cell content (TCC) could possibly affect the validity of quantitative assessment of ESR1, PGR, HER2
and KI67. Herein we aimed to investigate the performance of MammaTyper RT-qPCR kit under extreme scenarios of TCC
enrichment.
Ten extreme cases with low TCC (10 - 30%) and enriched in DCIS (15 70%) were selected. Two RNA samples were prepared
for each case using the RNXtract IVD kit. One sample contained at most 20% TCC whereas its pair contained > 80% after
macrodissection by an experienced technician. ESR1, PGR, HER2 and KI67 RNA were determined using the MammaTyper
RT-qPCR kit on a Roche Light Cycler. Differences in DDCT values and coefficient of variation were used to analyze differences
between non-macrodissected and macrodissected paired samples. IHC served as a reference for biomarker status evaluation.
Results
Despite the varying TCC content of invasive carcinomas the median 40-DDCT Differences of the mRNA Expression of HER2,
ESR1, PGR and HER2 between non-macrodissected and macrodissected samples were 0.34, 0.46, 0.27 and 0.41, respectively.
When previously established clinical cut-offs for biomarker positivity were considered, only a single case for KI67 appeared to be
affected by low TCC (negation). The concordance with IHC data was 88.9% for ESR1, 100% for PR, 88.9 for HER2 and 44.5%
for KI67 in this series.
Relative mRNA expression differences between non-macrodissected and macrodissected tumor specimen
Sample DCIS
ID
content
Invasive
Carcinoma
content
Difference in
Macrodissected ESR1 mRNA
[DDCT]
Difference in
Difference in PGR
HER2 mRNA
mRNA [DDCT]
[DDCT]
Difference in KI67
mRNA [DDCT]
10%
30%
>80%
0,65
0,26
0,65
0,55
70%
10%
>80%
0,43
0,35
0,38
-0,19
40%
10%
>80%
-0,44
-0,54
-0,17
0,25
30%
5%
>80%
0,46
-0,98
0,3
0,59
10%
50%
>80%
1,32
0,95
0,7
1,16
20%
10%
>80%
0,16
0,32
-0,14
0,58
25%
10%
>80%
1,69
1,32
1,26
-0,56
70%
15%
>80%
0,16
-0,42
-0,07
-0,31
40%
10%
>80%
0,46
-0,13
0,47
1,09
10
40%
30%
>80%
0,52
0,27
0,28
0,28
Conclusion
The performance of the MammaTyper diagnostic assay does not appear to be affected by fluctuations in the TCC of the original
FFPE specimens under the presence of increased amounts of DCIS. Similar findings have been previously reported in a research
setting (Kotoula, Virchows Arch 2013). Effective RNA extraction and optimal PCR output normalization provide sufficient
robustness for tolerating up to 8-fold TCC specimen changes, independent of the presence of DCIS. Our analysis suggests that
extra time spent on macro-dissection of specimens for routine RT-qPCR assays could be avoided with safety when using the
MammaTyper RT-qPCR kit.

Title: Centrosome amplification score: A quantifiable cancer cell trait with putative risk-predictive value in breast cancer
Vaishali Pannu1, Padmashree CG Rida1, Sergey Klimov1, Karuna Mittal1, Guilherme Cantuaria2, Michelle Reid3 and Ritu Aneja1.
1
Georgia State University, Atlanta, Gabon; 2Northside Hospital Cancer Institute, Atlanta, GA and 3Emory University Hospital,
Atlanta, GA.
Body: Breast tumors harbor extensive intratumoral heterogeneity (ITH), both within primary and metastatic lesions. The
generation of this genetic diversity relies on chromosomal instability (CIN), a dynamic and complex multilayered phenotype. CIN
comprises of an increased propensity to missegregate chromosomes during mitosis and ostensibly can be regarded as a survival
state adapted to aneuploidy, frequent aberrant mitosis and a sustained reshuffling of the genome. Centrosome amplification (CA),
a well-established cancer cell-specific trait is known to compromise mitotic fidelity resulting in CIN. Essentially, CA assists cancer
cells in concocting an array of diverse clones that drives tumor evolution by providing basic infrastructure for ITH. On this note,
logical reasoning and rational thinking led us to hypothesize that CA, a cell-biological cancer cell selective feature, may be
profoundly crucial and serve a causal role in driving ITH associated with tumor progression. While some studies suggest that CA
is an indicator of aggressiveness, others report that extent of CA increases with grade. No study yet has ever quantified CA or
emphatically demonstrated the extent of CA during the course of malignant evolution from a well differentiated to a
poorly-differentiated tumor. Here we have developed first-of-a-kind novel method to quantitate the degree, extent and type of CA
(both numeral and structural) within tumor samples to evaluate the trend in incidence and severity in breast tumors across
grades. Tissue specimens from core biopsies of 200 breast tumors were immunostained for centrosomes and nuclei. Employing
immunofluorescence confocal imaging, a stack of optical sections was acquired to capture all centrosomes and nuclei within 10
regions of interest (ROIs) per sample. Centrosomes were categorized as (i) individually-distinguishable centrosomes (iCTRs) or
(ii) as megacentrosomes (mCTRs) comprised of several tightly clustered centrosomes whose precise number cannot be
determined. For each ROI, number of nuclei as well as numbers and volumes of all iCTRs and mCTRs were determined and a
cumulative Centrosome Amplification Score (CAS) was obtained for each ROI as CAStotal = CASi+CASm. Low grade (Grade I,
n=75) tumors exhibited significantly higher CASi(3.9 vs 2.3), CASm (9.5 vs 5.4) and CAStotal (12.8 vs 8.05) values than high
grade (Grade II and III, n=125) ones, which does not support the previously held notion that CA increases during disease
progression. This postulation is additionally supported by the observation that low grade tumors that exhibit lymph node infiltration
and metastasis (n=30) had higher CASm (7.1 vs 9.8) and CAStotal (9.5 vs 11.5) (reverse these numbers) values as compared to
the non-invasive ones (n=50). In conclusion, our innovative method to quantitate CA in tumor samples establishes CA as a
"quantifiable cell biological property", that can potentially predict the risk of a low grade tumor being or becoming an aggressive
and invasive one.

Title: Empowering the Nottingham Grading System: An integrated Ki67-mitosis classifier yields a better patient stratification tool
Vaishali Pannu1, Padmashree CG Rida1, Sergey Klimov1, Nikita Wright1, Guilherme Cantuaria2, Michelle Reid3 and Ritu Aneja1.
1
Georgia State University, Atlanta, Gabon; 2Northside Hospital Cancer Institute, Atlanta, GA and 3Emory University Hospital,
Atlanta, GA.
Body: Therapeutic decision-making for personalized management of breast cancer relies on patient stratification based on the
risk conferred by clinicopathologic factors, such as stage, Ki67 Index (KI) and tumor histological grade. The most widely used
histologic grading system for breast cancer, the Nottingham Grading System (NGS) provides prognostic information about breast
tumor samples by combining analysis of the extent of glandular differentiation, nuclear pleomorphism and mitotic activity present
in the tumor sample. In the NGS, the mitotic Index (MI) is defined as the number of mitotic cells per ten high-power fields. KI is a
universal prognostic indicator but is not part of the NGS. Although the NGS has been widely used owing to its reproducibility and
significant prognostic value, its accuracy in predicting disease prognosis and aggressiveness is limited. Our earlier work has
demonstrated that by rationally integrating KI and MI into a new metric called the Ki67-Adjusted Mitotic Score (KAMS), which is a
measure of the proportion of mitotic cells amongst Ki67-positive cells, we are able to glean a new layer of prognostic information
about metastatic risk. We found that for Nottingham Grade II and III patients, high KAMS values predicted relatively better overall
survival (OS) than low KAMS values.
We therefore asked if the incorporation of a KAMS-based classifier subsequent to conventional Nottingham classification, would
improve patient stratification to enable their funneling towards more optimal treatment choices. Ideal KAMS thresholds were
established by analyzing the KAMS-stratified survival groups and selecting the thresholds which created the widest survival
stratification that was additionally confirmed to be statistically significant via a Log-Rank test. For Nottingham Grade II and III
patients, an above-threshold KAMS value was deemed as low-risk and a below-threshold KAMS value was deemed as high-risk
for the purpose of grade adjustment. Grade adjustments were based solely on the patients KAMS values. Thus the adjusted
Nottingham Grade I consisted of the original Nottingham Grade I patients along with KAMS determined low-risk original
Nottingham Grade II patients. Adjusted Nottingham Grade II patients were composed of KAMS determined high-risk patients
originally in Nottingham Grade II along with KAMS determined low-risk original Nottingham Grade III patients. Finally, the
adjusted Nottingham Grade III was made up exclusively of high-risk patients from the original Nottingham Grade III.
We found that the adjusted system represents a wider separation between OS curves with adjusted Grade I (n = 774) OS being
95.48%, adjusted grade II (n = 727) OS at 87.62%, and adjusted Grade III (n = 110) having an OS of 78.18%. Overall survival
groups between adjusted grade I and II and I and III (p < 0.001) showed statistical significance. We also found that the hazard
ratios for all Nottingham grades are significantly better after adjustment (2.875 versus 2.034 for Grade II and 5.115 versus 3.456
for Grade III) indicating that the KAMS can function as an effective risk-stratification biomarker and that incorporation of a
KAMS-based classifier enhances patient stratification.

Title: Evaluation of RT-qPCR and luminex-based methodolgies in HER2 breast cancer testing
Elizabeth N Kornaga1, John B McIntyre1,2, Alexander C Klimowicz3, Natalia Guggisberg1, Don G Morris1,4 and Anthony M
Magliocco5. 1Translational Research Laboratories, Alberta Health Services, Calgary, AB, Canada; 2University of Calgary, Calgary,
AB, Canada; 3Boehringer Ingelheim Pharmaceuticals Inc, Ridgefield, CT; 4University of Calgary, Calgary, AB, Canada and
5
Anatomic Pathology, H. Lee Moffitt Cancer Center & Research Institute, Tampa, FL.
Body: *Co-First Authors
Background: Currently, patients diagnosed with breast carcinoma undergo HER2 testing to direct clinical treatment decisions. At
present, immunohistochemical (IHC) and in-situ hybridization methodologies are employed in the clinical setting to ascertain
HER2 status. While these tests represent the current standard, their interpretation and variability of results with respect to HER2
prognostic and predictive value remains an outstanding issue.
Aim: In this comparative study we assessed the utility of testing HER2 gene expression by RT-qPCR and the Luminex
Quantigene Plex 2.0 (Affymetrix) methodology against the clinically accepted IHC assay.
Methods: Local cases from 2008-2010 that were clinically evaluated for HER2 were identified and underwent further pathologist
review. In cases where there was sufficient tumor, formalin fixed paraffin embedded samples were retrieved. A total of 207 cases
were identified which met selection criteria. Tumour sections were stained for HER2 and scored 0-3, following ASCO/CAP
guidelines. For molecular assessment total RNA was extracted from tumours and those samples with sufficient RNA yield and
quality were assessed for Her2 transcript level expression by RT-qPCR (n=129) and Luminex Quantigene Plex 2.0 assays
(n=166).
Results: Results for RT-qPCR are relative to two normal breast calibrator samples and reported as the mean relative
quantification (RQ) value. For HER2 IHC negative cases (0/1+), the mean score was 0.13 (0.004-1.84, SD 0.22); equivocal
cases (2+), mean score was 0.19 (0.007-0.72, SD 0.18); and positive cases (3+), mean score was 1.51 (0.03-6.78, SD 1.61).
Students t-test was performed to compare the means between groups and results are as follows: negative vs. equivocal p=0.17;
equivocal vs. positive p=0.0002; and negative vs. positive p<0.0001. Luminex methodology is reported as the normalized mean
fluorescence intensity (nMFI) for each group. For HER2 IHC negative cases (0/1+), the mean score was 0.28 (0.01-1.19, SD
0.23); equivocal cases (2+), mean score was 0.42 (0.10-1.23, SD 0.27); and positive cases (3+), mean score was 4.74
(0.1-9.09, SD 2.5). Students t-test to compare the means between groups was again utilized, and results are as follows:
negative vs. equivocal p=0.0036; equivocal vs. positive p< 0.0001; and negative vs. positive p< 0.0001.Conclusions: Results
demonstrate that both RT-qPCR and Luminex Quantigene Plex 2.0 methods are able to discern strong positive HER2 cases
(IHC 3+) from negative HER2 (IHC 0/1+). For cases with moderate IHC staining (2+), the Luminex-based assay was found to
perform better than RT-qPCR. For each reporting group, the range of HER2 gene expression values were observed to overlap;
thus no distinct cut point could be assigned. While the results from this pilot study are promising, the adoption of molecular
methods for HER2 diagnostic testing will require further rigorous investigation before any clinical considerations can be made.

Title: Evaluation of PgR status by immunohistochemistry may be inferior to PgR results by Oncotype DX for assessing the
recurrence risk in ER+/HER2- breast cancer with low or intermediate tumor proliferation
Peter Sinn1, Mark Kriegsmann1, Ute Felten2, Jrgen Wacker2, Andreas Schneeweiss3 and Zsuzsanna Varga4. 1University of
Heidelberg, Heidelberg, Germany; 2Frst Stirum Klinik, Bruchsal, Germany; 3National Center for Tumor Diseases, Heidelberg,
Germany and 4Institut fr Klinische Pathologie, Universittsspital Zrich, Zrich, Switzerland.
Body: Background: Loss of progesterone receptor expression is known to be a predictor of Luminal-B subtype, and is of
independent prognostic significance in luminal type breast cancer. However, the threshold of PgR status in IHC has been subject
to debate, and the PgR status as reported by IHC not infrequently is at variance with the PgR result as reported by the Oncotype
DX (ODX) assay.
Methods: We retrospectively reviewed tumor IHC and ODX data from resection specimens with ER+/HER2- breast cancer, and
low to intermediate Ki-67 proliferation rates (Ki-67 < 30%). A total of 74 specimens were re-analyzed, all from resection
specimens that had been selected for ODX assay for the purpose of planning of adjuvant chemotherapy, from two institutions
(Univ. Heidelberg, Germany, and Univ. Zurich, Switzerland). PgR expression of < 10% using IHC (clone PgR636, Ventana and
Dako systems) or PgR values of < 5,5 in ODX assay were considered negative.
Results: All cases were positive for ER by IHC and ODX, and 20% were negative for PgR by ODX assay. This compared to only
8% of negativity by IHC, but all except one case negative for PgR by IHC were also correctly identified as PgR negative by ODX.
Median ODX recurrence score (RS) was 35 for PgR negativity by ODX assay, as compared to 30 for negative PgR status by IHC.
For positive PgR status, median RS was calculated as 15 vs. 17 by ODX vs. IHC. Risk categories by ODX were high (RS >= 31)
in 9 cases with negative PgR status by ODX, but only 3 of these high risk cases were negative for PgR by IHC. Median Ki-67
values were 22.5 for PgR negative cases by ODX as compared to 15 for PgR negative cases by IHC. Similar results were
obtained when a higher PgR threshold of 20% was applied to IHC results.
Conclusions: In luminal type breast cancers, the evaluation of PgR status by IHC may underestimate the proportion of PgR
negative cases and did not identify most cases of tumors with a high risk of recurrence. This may be caused by an overly high
sensitivity of PgR status in routine automated immunohistochemistry.

Title: The chromosomal genomic landscape and targetable co-amplified genes in HER2 positive breast cancer patients who
relapsed on an adjuvant trastuzumab chemotherapy trial
Shelly Gunn1, Chris McCaskill1, Linda Daley1, Agnes Puskas1, Lina Asmar2, Yunfei Wang2 and Stephen Jones1. 1MolecularHealth,
Woodlands, TX and 2McKesson Specialty Health, Woodlands, TX.
Body: Background
Early stage HER2 amplified breast cancer has a generally favorable prognosis with over 95% of patients showing 2-year disease
free survival (DFS) when treated with adjuvant trastuzumab. However, a subset of these tumors are refractory to treatment and
present a challenge for the oncologist, particularly when clinical and histologic parameters, including the patients nodal and
hormonal status, are indicative of lower-risk HER2 positive disease. In this study we describe the genomic landscape of three
clinically lower-risk patients with HER2 amplified tumors who relapsed on adjuvant docetaxel and cyclophosphamide plus 1 year
of trastuzumab in a phase 2 study (Jones et al, Lancet Oncology 14: 1121, 2013). All patients tumors showed high-level HER2
amplification ratios by FISH (8.51-14.46) and were analyzed in parallel with a fourth clinically matched 2-year DFS patients tumor
from the same trial with high HER2 gene amplification (FISH ratio 11.38).
Methods
Primary tumor genomic DNA analysis was performed from archival tissue by next generation sequencing (NGS) on the Illumina
HiSeq 2500 platform in a CLIA certified laboratory. Tumors were screened for point mutations and copy number alterations
(CNAs) by NGS using a targeted-whole exome 613 gene panel. CNAs detected by NGS were confirmed on a DNA microarray
featuring high-density probe coverage of the same 613 genes on the targeted panel. CGH chromosome ratio plots were overlaid
with algorithmically derived NGS copy number data to generate a map of the chromosomal genomic landscape for each patients
tumor.
Results
High-level HER2 gene amplification status was confirmed, and co-amplified chromosome 17 genomic regions were detected in all
three relapsed patients tumors. High-level HER2 amplification was also confirmed in the non-relapsed patients tumor but
co-amplified regions were not detected on chromosome 17 or elsewhere in this patients tumor genome. Two of the relapsed ER
negative tumors shared focal high-level CCNE1 gene amplifications on chromosome 19. High-level MAP3K3 gene amplification
on chromosome 17 was detected in the one ER positive tumor from a relapsed patient. Focal PIK3CA gene amplifications were
not identified in any of these tumors, but two tumors (one from the relapse group and one non-relapse) were positive for activating
H1047R mutations.
Conclusions
Combined NGS and CGH analysis of HER2 positive early stage breast cancer can be performed in the clinical laboratory to
reveal the tumors chromosomal genomic landscape. Combined with other test results, this tumor map can help identify patients
at high-risk for relapse and reveal alternative predictive biomarkers of therapeutic response.

Title: No Show

Title: Correlation of the effects of potential and known chemopreventive agents on proliferation rates in normal mammary glands
and mammary cancers with their chemopreventive efficacy
Brandy M Heckman-Stoddard1, Clinton J Grubbs2, Fariba Moeinpour2, Vernon E Steele1 and Ronald A Lubet1. 1National Cancer
Institute, Rockville, MD and 2University of Alabama, Birmingham, AL.
Body: Core needle biopsy, fine needle aspiration, or imaging (e.g., mammography) are currently used to examine
chemopreventive agent efficacy in Phase II breast cancer trials. However, biomarker endpoints, such as proliferation rates using
Ki67 in normal or "at risk" breast tissue, have not been formally validated relative to cancer outcome. The aim of this study was to
validate Ki-67 in "normal" mammary tissue from the methylnitrosourea (MNU) rat mammary tumor model with mammary cancer
multiplicity within the same animal as surrogate biomarker for agent efficacy. Multiple studies were performed in female
Sprague-Dawley rats to correlate this proliferation biomarker in mammary tissue after two weeks of chemoprevention agent
treatment with mammary cancer incidence and multiplicity at the end of the study. In brief, MNU was given at 50 days of age, one
week later administration of the agent was started, and after two weeks a biopsy of the mammary gland was taken. Treatment
with the agents continued for approximately 150 days, at which time the study was terminated and mammary tumor multiplicity
was measured. Changes in proliferation and mammary tumor multiplicity were compared to a control group of untreated group of
animals within the same experiment using a two-sided Student t-test.
[table1]
Agent
Normal Mammary Gland Proliferation Index
Final Mammary Cancer Multiplicity
Vorozole (1.25 mg/kg BW/day)
88%
90%
Lipitor (125 mg/kg diet)
38%
16%
Targretin (150 mg/kg diet)
90%
92%
Naproxen (200 mg/kg diet)
6%
45%
Iressa (10 mg/kg BW/day)
52%
93%
Tamoxifen (3.3 mg/kg diet)
77%
100%
Metformin (150 mg/kg BW/day)
25%
71%
In general, highly effective agents to preventing cancers (e.g., tamoxifen) also prevented normal mammary gland proliferation
after only two weeks of treatment, while inactive agents (e.g., naproxen) had minimal effects on normal gland proliferation. The
effects of the agents on established mammary cancers (in which the agents were given for 7 days to rats bearing small
MNU-induced mammary cancers) showed similar correlations. Additional biomarkers, as well as the proliferation change and
efficacy of these agents in rats fed a high-fat (Western) diet, will also be presented. In conclusion, determining the effect of a
potential chemopreventive agent on cell proliferation following short-term treatment appears to be an effective method for
predicting its efficacy in preventing mammary cancers. These data further confirm that Ki67 measurements are useful in Phase II
prevention trials as a biomarker of agent efficacy.


Title: Development and implementation of a breast cancer risk identification and reduction program in a large health care system
Maureen A MacSweeney1, Helen Roorda1, Richard Lippert2, Ivy Guardolia1, Nefertari Burrell1, Luis DeJesus1, Augustus J
Scarletto1 and David A Decker1. 1Florida Hospital Cancer Institute, Orlando, FL and 2i/o Trak, Orlando, FL.
Body: Background: Multiple national and international organizations (USPSTF, ASCO, NCCN, and NICE) recommend identifying
patients at an increased risk of breast cancer and counseling these patients concerning risk reduction strategies. Primary care
providers have not complied with these guidelines.To fill this void, we initiated a program which identifies risk and offers risk
reduction strategies.This abstract describes the development and preliminary outcomes of this program.
Material and Methods: Florida Hospital has 10 separate mammography center sites. Screening mammograms were performed on
48,917 women the year prior to the start of the program. The program was created to target this population. Data was collected
between 4/1/2013 and 3/31/2014. Each of the 10 sites were incrementally added throughout one year. Women at the time of
screening mammogram were offered participation in the program by the mammography technician. Mammography technicians
were educated and trained. A tablet was developed with our IT department. Women that consented answered 8 questions from
the NCI modification of Gail risk, 1 question addressing chest wall radiation, and 5 questions modified from the NCCN Guidelines
BR/OV -1 to identify hereditary risk possibly missed by the modified Gail. The results, demographics and referring provider
information was collected in a database. Patients and referring provider received letters tailored to their individual risks, by
Mammologix, a mammography informatics company. High risk was defined as a modified Gail score of 1.7% for a 5 years,
>17% lifetime risk, or a positive response to the 5 additional questions. Patients at risk were offered a risk and reduction
consultation at our Breast Health Center. Consults were provided by a genetic counselor and/or ARNP. The risk and benefits of
intervention were discussed. The referring provider was included in any discussion concerning risk reducing surgery, medication,
and follow-up. Patients were referred back to their primary provider for continuing follow-up.
Results: 15,165 women were given a risk questionnaire at the time of screening mammography. 2752 patients opted out. 3243
were high risk by the modified Gail method.1329 were high risk by both the modified Gail and our modified NCCN BC/OV-1
genetic guidelines questions. 14 were high risk due to chest wall radiation. Of the 4044 women at risk, 141 opted to be counseled
(3.88%). 29 chose to begin prevention medication. 16 chose counseling regarding their hereditary risk.
FHCI Breast Health Center Data
Q1
Q2
Q3
Q4
Pts Screened
1025
1434
5251
7344
Pts at risk by Modified Gail
196
290
1166
1569
Pts at risk by additional questions
51
75
444
355
Pts seen in Center by ARNP
15
15
17
80
Pts seen in Center by Genetic Counselor
Pts treated for Breast Cancer Prevention
10
Conclusion: The program provides a comprehensive breast cancer identification and reduction option to a large number of
women. It also provides the primary provider a service and resources regarding breast cancer risk identification and prevention.

Title: Inhibition of alternative NF-B activity prevents the expansion of genetically unstable progenitor cells in BRCA1-deficient
mammary glands
Andrea Sau1, Rosanna Lau2, Emma Nolan3, Peter A Crooks4, Jane E Visvader3 and Christine MA Pratt1. 1University of Ottawa,
Ottawa, ON, Canada; 2UT MD Anderson Cancer Center, Houston, TX; 3Walter and Eliza Hall Institute of Medical Research,
Parkville, Victoria, Australia and 4University of Arkansas for Medical Sciences, Little Rock, AR.
Body: Understanding the biological mechanisms underlying the initiation and progression of breast cancer is an important step
for prevention and treatment. Individuals with mutations in the breast cancer-associated gene 1 (BRCA1) have a lifetime
increased risk of up to 85% of developing breast cancer. BRCA1 participates in DNA damage repair and cell cycle checkpoint
control, serving as a tumor suppressor gene to maintain global genomic stability. However, BRCA1 has also been shown to play
a key role in the expansion of mammary stem/progenitor cells, which are the targets for carcinogenesis in individuals who have
undergone loss of heterozygosity (LOH) for BRCA1.
We have used in vitro and in vivo models to demonstrate that BRCA1 loss or mutation induces alternative NF-B pathway
activation and leads to the expansion of a genetically unstable progenitor cell population in the mammary gland.
Our data showed that BRCA1 loss or mutation is responsible for activation of the alternative NF-B pathway evidenced by IB
kinase- (IKK) phosphorylation, processing of p100 to p52 and p52/RelB nuclear localization. Remarkably, we found that RelB
and p100/p52 were highly expressed in 20-50% of the lobular structures in histologically normal breast tissue obtained from
human BRCA1 mutation carriers.
After DNA damage, ATM directs the nuclear export of NEMO (NF-B essential modifier) that in turn activates the alternative
NF-B pathway. We found high levels of phospho-ATM in breast tissue from human BRCA1-mutation carriers and in progenitor
cells from BRCA1-knockout mouse mammary glands. Moreover, co-immunoprecipitation studies showed increased
ATM/NEMO-containing complexes in BRCA1-deficient cells. Progesterone injections into MMTV-cre;BRCA1f/f mice induced an
increase in Ki-67 and phospho-H2AX levels, demonstrating a role for progesterone in DNA damage amplification in mouse
mammary glands.
Mammary epithelial progenitor cells obtained from BRCA1-mutated carriers as well as BRCA1-/- mouse glands can form acini in
Matrigel in a progesterone-independent manner. We found that in vivo inhibition of IKK/ using the IKK inhibitor
dimethylaminoparthenolide (DMAPT) completely blocked mammary acini formation in Matrigel, requiring several estrus cycles
post-injection to recover. Importantly, knockdown studies showed that p100/p52 was necessary for progenitor cell proliferation in
Matrigel. Consistent with the continued dependence on alternative NF-B activity in tumors derived from BRCA1-deficient
progenitor cells, human breast cancer xenografts derived from BRCA1-mutation carriers infected with lenti-shp100/p52 showed a
significant growth inhibition.
Overall, these results suggest an exciting new approach for the prevention of breast cancer in patients wherein intermittent or
cyclic therapy using DMAPT (generic name LC-1) has the potential to mitigate breast cancer risk in BRCA1 mutation carriers
through acute reduction of aberrant progenitor activity. As such, NF-B-directed chemopreventive therapies may provide
promising alternatives to prophylactic mastectomy in this high risk patient population.

Title: Pregnancy-induced epigenetic changes in the insulin-like growth factor signaling pathway
Tiffany A Katz1, Serena G Liao1, Thushangi Pathiraja3, Robert K Dearth2, George C Tseng1, Steffi Oesterreich1 and Adrian V Lee1.
1
University of Pittsburgh, Pittsburgh, PA; 2University of Texas, Pan American, Edinburg, TX and 3Genome Institute of Singapore,
Singapore, Singapore.
Body: Prevention will prove to be the single most effective way of eradicating breast cancer. Currently, the most effective natural
breast cancer prevention is an early first full term pregnancy. While it is not feasible to use pregnancy to protect women from
breast cancer, understanding how the protective effect is elicited will inform the development of new prevention strategies.
Women who were pregnant in their twenties are protected thirty to forty years later creating a complicated mechanism to tease
out experimentally. In order to understand the long-term protection we have investigated epigenetics, specifically DNA
methylation, which is known to be stable over long periods of time. A cohort (Parous) of female FVB mice were bred, gave birth,
and pups were weaned. A control group (Nulliparpous) never saw male mice. Mammary glands were harvested immediately or 6
months after involution. These two time points allowed us to identify changes in DNA methylation that occurred in response to
pregnancy, and additionally, changes that lasted long after parturition. DNA was isolated, and genome-wide DNA methylation was
assessed using bisulfite-conversion and SureSelect Methyl-Seq. Bismark v0.7.12 was used for alignment of pair-end reads,
followed by the R package "methylKit" for quality control and data analysis. A mapping efficiency of 50%68.1% was achieved
with 89,512,619 base pairs covered. CpG Pearson correlation plots and PCA analysis showed global similarity between samples.
We then conducted a logistic regression to ascertain parity-induced differentially methylated regions and identified 153 and 236
persistently hypomethylated and hypermethylated genes, respectively. Among the differentially methylated genes were many
signaling molecules involved in growth factor signal transduction, including insulin-like growth factor 1 and 2 receptors (IGF1R
and IGF2R). It has previously been shown that circulating IGF1 levels are reduced in parous women, and similarly the growth
hormone/IGF axis is altered in rodent models. Collectively, these findings suggest that the IGF pathway is regulated at multiple
levels during pregnancy, and that its modification might be critical in the protective role of pregnancy. We are currently following
up on these data, including protein analysis of the IGF pathway members and downstream signaling molecules in human
specimens. Finally, we are expanding our analysis to additional genes and pathways epigenetically altered by pregnancy, with the
ultimate goal to develop new prevention strategies.

Title: Understanding predictive values of short-term morphologic assays of cancer chemoprevention for efficacy in animal
mammary gland tumor assays
Barbara K Dunn1, Vernon E Steele1, Carol F Topp2, Richard M Fagerstrom1 and Barnett S Kramer1. 1National Cancer Institute,
Rockville, Md and 2CCS Associates, McLean, VA.
Body: Background: The predictive value of chemopreventive agent efficacy in morphologic (in vitro/in vivo) assays for efficacy in
animal (mouse and rat) in vivo tumor assays is not well characterized. Over a 25-year period, the Chemopreventive Agent
Development Research Group in the U.S. NCIs Division of Cancer Prevention has tested approximately 800 agents for potential
chemopreventive activity. The current project focuses on a subset of 146 that were tested in both morphologic and mammary
gland tumor assays in order to gain a deeper understanding of the relevant predictive value.
Materials and Methods: The early stages of the testing pathway involve two critical steps: (1) in vitro/in vivo morphologic assays
and, for agents successful in these, (2) testing for tumor prevention (measured in terms of tumor incidence and multiplicity
reduction) in animal tumor assays. The ultimate goal is to test agents that successfully decrease tumor incidence and multiplicity
in animal tumor assays in humans. In the current project we evaluated the predictive values of the earlier-stage (morphologic)
assays for efficacy in the later-stage (animal tumor, specifically mammary tumor) assays. Statistical modeling to determine how
well the six most commonly used morphologic assays predicted efficacy of the 146 tested agents in mouse and rat mammary
gland tumor assays was carried out by multimodel inference applied to ordinal logistic regression.
Results: The ability of these six morphologic assays to predict tumor outcomes was evaluated in the mouse and rat mammary
gland cancer assays. Based on this statistical modeling, each morphologic assay was assigned a value describing how strongly it
predicted outcomes in the mammary gland tumor assays. Selected morphologic assays (the mouse mammary organ culture
(MMOC) and human foreskin epithelial cell (HFE) morphologic assays) in combination give a predictive value that meaningfully
forecasts results for chemopreventive efficacy in the mouse and rat mammary gland tumor assays.
Conclusions: These predictive models can be used to guide our future decision-making with respect to agent selection as well as
morphologic and animal tumor assay use. Our future work is focused on deepening our understanding of our Predictive Value
approach by: (1) identifying the mammary gland tumor assays that best reflect anti-tumor efficacy in animals; (2) teasing apart
those classes of agents that exhibit the highest predictive values overall; and (3) examining which agent classes show the highest
efficacy in specific mammary gland tumor assays.

Title: Predicting the effect of tamoxifen on the breast: Change in measures of breast density, serum markers and SNPs
Anthony Howell1, Sue Astley1, Elaine Harkness1, Julia Wiseman1, Jill Fox1, Paula Stavrinos1, Mary Wilson1, Yit Lim1, Valerie
Reece1, Ursula Beetles1, Anil Jain1, Jamie Sergeant2, Jack Cuzick1, Ruth Warren2 and Gareth Evans1. 1Nightingale Centre and
Genesis Breast Cancer Prevention Centre, Manchester, Greater Manchester, United Kingdom and 2Centre for Cancer
Prevention, London, Greater London, United Kingdom.
Body: Body: Introduction In the IBIS I Prevention Trial we demonstrated that tamoxifen prevented breast cancer in women with
greater than 9 % absolute reduction in visually assessed mammographic density over a 12-18 month period as assessed by an
expert radiologist, RW (1). In this study we investigated: whether RW obtained similar results in a different group of women
treated with tamoxifen; how other expert mammographic film readers with experience of density assessment (radiologists,
advanced practitioner radiographers and a breast physician) estimated mammographic change; how visual change relates to
change in automated measures of volumetric density ; and whether we can predict change by serum markers or SNPs. Methods
105 women aged 33 to 46 at greater than 1 in 6 lifetime breast cancer risk completed one years treatment with tamoxifen. RW
assessed mammographic percentage density in increments of 5% for baseline and one year mammograms. Two or three of a
pool of eight readers estimated % density using a visual analogue scale (VAS). Percent change was also estimated using a
computer-assisted thresholding technique (Cumulus). Change in volume of dense tissue (Quantra & Volpara) was measured.
Changes in lipids, IGF1, insulin and relevant SNPs were measured to determine whether they predicted density change. Results.
Estimates of change obtained by RW were broadly consistent with those from our previous study, as were those from Cumulus,
but could not be replicated using estimates from the pooled results of VAS readers. Tamoxifen was associated with marked
reductions of dense volume of the breast (Table 1) However the different methods were at best only moderately correlated with
RW ( r less than 0.6). Change in serum triglycerides significantly predicted density change as measured by RW but none of the
other serum measures or SNPs did.
Table 1 Numbers of women and percentage change in density measures after one year of tamoxifen
% Density
change
RW
%change
Other radiologists
%change
Cumulus
%change
% Dense volume
change^
Quantra
%change
Volpara
%change
40+
40+
15
30+
30+
14
10
20+
12
13
20+
20
20
10+
35
30
10+
17
26
0+
49
81
43
0+
16
19
0-
18
0-
16
15
Number
105
105
98
98
99
_____
RW = Ruth Warren. CUMULUS was measured by JS, a trained operator of the software. * Percentage point change eg if density
changed from 40% to 30% this was regarded as a 10% change.^ Reduction in dense volume expressed as a percentage of
baseline dense volume. We used Quantra version 2.0 and Volpara version 1.4.5
Conclusions Whilst RW remained consistent with previous density estimation it was not possible to use a pool of other expert
readers to predict change. Cumulus gave similar results to RW but is difficult to use in practice. There was good agreement
between the two objective volumetric measures used (r=0.5) and since these are automated they may be suitable for clinical
practice. However their relationship to the long term breast cancer preventative effect of tamoxifen needs to be established.
Reference 1 Tamoxifen-induced reduction in mammographic density and breast cancer risk reduction: a nested case-control
study. Cuzick J, Warwick J, Pinney E, Duffy JW, Cawthorn S, Howell A, Forbes JF, Warren RM. J Natl Cancer Inst. 2011 May 4;
103 (9): 744-52.

Title: Metformin decreases Ki67 in HER2+ve ductal carcinoma in situ in a window of opportunity trial
Bernardo Bonanni2, Andrea DeCensi1, Aliana Guerrieri-Gonzaga2, Giancarlo Pruneri2, Matteo Puntoni1, Massimiliano Cazzaniga2,
Andrea Vingiani2, Davide Serrano2, Oreste Gentilini2, Harriet Johansson2, Valentina Aristarco2, Matteo Lazzeroni2, Clara
Varicchio2, Marilena Petrera1 and Giuseppe Viale2. 1E.O. Ospedali Galliera, Genoa, Italy and 2European Institute of Oncology,
Milan, Italy.
Body: Background : In a presurgical trial in 200 non-diabetic women with breast cancer, we previously showed a heterogeneous
effect of metformin on the Ki67, with a decreased proliferation in women with insulin resistance (HOMA>2.8) and an opposite
effect in women with HOMA 2.8 (Bonanni et al. JCO 30:2593, 2012). Here we determined the effect of metformin on noninvasive
proliferative disorders.
Methods: Patients with operable breast cancer were randomly assigned to metformin, 850 mg or placebo once daily on days 1-3
followed by two 850 mg tablets after dinner on days 4 to 28. A total of 3-5 specimens of adjacent (1 cm from tumor) and distant
(>1 cm from tumor) tissue were obtained from the surgical specimens to assess systematically the prevalence of LCIS and DCIS
adjacent to invasive cancer and of distant ductal hyperplasia (DH) in normal tissue. We also determined the effect of metformin
on Ki67 in these lesions overall and by molecular subtype. All HER2 2+ DCIS by IHC were assessed by FISH.
Results: Overall, the prevalence of LCIS, DCIS and DH was 4.5% (9/200), 66% (132/200) and 35% (69/200), respectively. The
Ki67 was positively associated with DCIS grade (p-trend<.001). The median levels of Ki67 by treatment arm in different
premalignant groups is summarized below
Median (IQR) Ki67 level (%) in premalignant disorders by treatment arm
Premalignant group
Premalignant subgroup
Metformin arm
Placebo arm
p-value*
LCIS (n=9)
15 (5-15)
5 (4-6)
0.1
DH (n=69)
3 (1-4)
3 (1-4)
0.5
All DCIS (n=132)
12 (8-20)
10 (7-24)
0.9
DCIS grade 1/2 (n=108)
10 (7-16)
10 (6-17)
0.9
DCIS grade 3 (n=24)
33 (25-55)
40 (32-40)
0.2
HER2+ve (n=22)
22 (11-32)
35 (30-40)
0.06
HER2-ve (n=58)
16 (10-20)
17 (8-26)
0.7
ER+ve/HER2+ve (n=15)
12 (7-18)
32 (27-42)
0.004
ER+ve/HER2-ve (n=53)
16 (10-20)
15 (8-22)
0.8
PR+ve/HER2+ve (n=12)
18 (12-18)
32 (24-44)
0.02
PR+ve/HER2-ve (n=48)
16 (10-20)
12 (8-20)
0.6
p-interaction
0.2
0.04
0.001
0.05
*Wilcoxon rank-sum test; p-interaction between treatment and DCIS grade or HER2 status from a linear regression model,
adjusted for age and BMI.
The effect of metformin on DCIS was different by HER2 status (p-interaction=.04) and, among this molecular subtype, by ER and
PR status (p-interaction=.001 and .02, respectively). In HER2+ve DCIS, metformin decreased Ki67 by 40% overall (p=.06), by
over 60% in ER+ve/HER2+ve DCIS (p=.004). and by 45% in PR+ve/HER2+ve DCIS (p=.02) There was no effect of metformin in
HER2-ve DCIS, regardless of ER or PR status. Metformin did not affect Ki67 in DH overall, but showed a trend towards a
decrease in women with abdominal adiposity (p-interaction=.05).
Conclusions: Window of opportunity pre-surgical models provide insight into a drugs preventive potential by targeting
intraepithelial proliferations adjacent to invasive cancer. The model shows a high prevalence of preinvasive lesions (field
cancerization) in apparently normal breast tissue adjacent to breast cancer. Metformin selectively decreased Ki67 in HER2+ve
DCIS, particularly in the ER+ve subgroup, in line with a selective inhibitory effect on the HER2 pathway observed both in in vitro
and in vivo preclinical models. Our results provide the background for a phase III trial of metformin in HER2+ve DCIS
(ISRCTN16493703, Supported by AIRC, LILT and Health Ministry 2009-RF-153222).

Title: Initial report of a randomized trial of letrozole in high risk women taking hormone replacement therapy
Carol J Fabian1, Bruce F Kimler1, Jennifer L Nydegger1, Trina Metheny1, Carola M Zalles2, Brian K Petroff1, Hung-wen Yeh1 and
Michael D Alvarado3. 1University of Kansas Medical Center, Kansas City, KS; 2Mercy Hospital, Miami, FL and 3University of
California, San Francisco, CA.
Body: Background:
On the basis of positive results in a pilot study, we initiated a randomized, double-blind, placebo-controlled trial of the aromatase
inhibitor letrozole in post-menopausal women at high risk for development of breast cancer who were taking hormone
replacement therapy (HRT). The objective was to determine if risk biomarkers for breast cancer in benign breast tissue sampled
by random peri-areolar aspiration (RPFNA) could be favorably modulated.
Methods
Women who exhibited cytologic hyperplasia +/- atypia and Ki-67 immunocytochemistry staining 1.5% on screening RPFNA were
eligible to be randomized 1:1 between placebo and letrozole (2.5. mg daily) for six months, followed by repeat RPFNA. Women
were then given the option to receive open-label letrozole for a second six months, and a third RPFNA. The primary analysis was
a difference between the two groups for the change in Ki-67 between baseline and 6 months. The initial accrual goal was 108
subjects, with the expectation of 96 subjects evaluable for the baseline6 months comparison.
Results
55 subjects were enrolled between March 2007 and March 2014, when accrual was closed. From the time of our successful pilot
study to present, there had been a steady decline in the use of HRT by women in our high risk cohort, both in frequency of
women using as well as the type and strength of HRT. The result was fewer potential subjects for screening and fewer still that
satisfied the 1.5% Ki-67 criterion. Thus, the trial was closed early. Of 55 enrolled subjects, two dropped out prior to 6 months; 52
completed 6 months and provided evaluable RPFNA specimens for analysis, with one subject scheduled for repeat aspiration in
September. Six subjects went off study between 6 and 12 months; 42 have completed the entire 12 month schedule, and 5 are
still on trial. At baseline, 18 women displayed hyperplasia (Masood score 13-15) and 37 had hyperplasia with atypia (Masood
score 14-17). Median Ki-67 was 3.0%, with a range from 1.6 15.4%. For 52 comparisons between baseline and 6 months, 8
women had no change by Masood score, 8 had an increase and 36 exhibited a decrease, i.e., less abnormality. Two women had
no change in Ki-67 staining, 13 exhibited increased Ki-67 and 37 showed a decrease; median at 6 months 1.7%, median change
-1.4%. For 42 comparisons between baseline and 12 months, 11 women had no change by Massod score, 9 increased, and 22
decreased. One woman had no change in Ki-67 staining, 10 exhibited increased Ki-67 and 31 showed a decrease. The
decreases in Masood score and Ki-67 between baseline and 12 months (when all subjects had received letrozole, either for 6 or
12 months) were statistically significant (p<0.005, Wilcoxon signed rank test). When 6-month data are available for the final
subject, the randomization will be unblinded by the statistician and the primary study question will be addressed.
Conclusion
While pending final analysis of the primary (blinded) endpoint, preliminary analysis indicates favorable modulation of
cytomorphology and proliferation by the aromatase inhibitor letrozole in high risk post-menopausal women taking hormone
replacement therapy.
Funding: NIH RO1 CA122577; Novartis Pharmaceuticals Corp.

Title: De-escalating doses of letrozole in post menopausal women at high risk for breast cancer
Ana Maria Lopez1, Hsiao Hui Sherry Chow1, Denise Frank1, Sandhya Puthi2, Judy Boughey2, Paul Hsu3, Jose Guillen3, Marjorie
Perloff4, Michelle Ley2 and Julie E Lang5. 1University of Arizona Cancer Center, Tucson, AZ; 2Mayo Clinic, Rochester, MN;
3
University of Arizona Cancer Center, Tucson, AZ; 4National Cancer Institute, Bethesda, MD and 5Keck School of Medicine,
University of Southern California, Los Angeles, CA.
Body: Background: Although breast cancer (BC) may occur at any age, its prevalence is greater postmenopause. Greater than
75 % of postmenopausal BCs are hormone dependent. Aromatase inhibitors suppress postmenopausal estrogen biosynthesis.
Letrozole has demonstrated efficacy against BC in the adjuvant and metastatic setting at the treatment dose of 2.5 mg daily. Its
potential role in BC prevention has been inferred from reductions in contralateral BCs. Its side effect profile, similar to other AIs,
includes exacerbation of menopausal symptoms that negatively impact quality of life (QOL) and may result in discontinuation of
the drug. Both anastrazole and exemestane have been demonstrated to reduce BC in high-risk women.
Hypothesis: Lower and intermittent doses of letrozole effectively suppress estrogen in the high-risk postmenopausal woman and
provide a better side effect profile.
Methods: A randomized, double-blind study comparing the impact of varying letrozole doses (2.5 mg daily, 2.5 mg MWF, 1.0 mg
MWF, or 0.25 mg MWF) on estrogen suppression and side effects--lipids, bone resorption, menopause, and QOL-- was
conducted. Participants randomized to intermittent dosing received placebo on nontreatment days.
Results: 112 participants were enrolled at 2 clinical sites. Mean patient age was 62.8 years, and average BMI was 29.8. Analysis
of available data after 24 weeks of therapy revealed statistically significant increase in triglycerides (N=94): 114.6748.39 to
125.7954.31 (p<0.01); vasomotor symptoms (N=95): 2.251.26 to 2.741.67 (p<0.01); and C-telopeptide (N=75): 0.390.23 to
0.550.28 (p<0.0001). Statistically significant decrease in estradiol (N=68): 5.575.19 to 1.261.41 (p<0.0001) and estrone
(N=68): 22.3913.02 to 1.642.66 (p<0.0001) were observed. No differences in QOL (SF 36) were noted after letrozole
treatment. P-values were derived from paired t-test (signed rank test) for the difference between baseline and after 24 weeks of
study drug.
Conclusions: De-escalating doses of letrozole suppress postmenopausal estrogen effectively and result in statistically significant
increases in triglycerides, C-telopeptide and vasomotor symptoms without impact on QOL. Presentation will include unblinded
intervention arm outcomes.

Title: Breast cancer (BC) following prophylactic mastectomy (PM), a clinical entity: Presentation, management, and outcomes
Robert W Mutter1, Tanya L Hoskin2, Marlene H Frost3, Joanne L Johnson4, Lynn C Hartmann5 and Judy C Boughey6. 1Mayo
Clinic, Rochester, MN.
Body: Objective: Contralateral PM (CPM) and Bilateral PM (BPM) markedly decrease, but do not completely eliminate the
possibility of development of a new BC on the side of the PM. Given the relative infrequency of its occurrence, little is known
about the clinical characteristics, presentation, and management of patients who develop BC after PM. Our aim was to review our
institutional experience of BC occurring after PM.
Methods: Between 1960 and 1993, 1,065 women underwent BPM and 1,643 women with unilateral BC treated with therapeutic
mastectomy underwent a CPM. Medical records were reviewed and study-specific questionnaires were sent to all women at 10
years and 20 years after PM. BC after PM included locoregional invasive BC or DCIS on the side of the PM.
Results: Thirteen patients who underwent BPM developed BC after PM. Twelve patients who underwent CPM developed a
subsequent BC on the side of the CPM. The median follow-up time from PM was 22 years (range 3-34). Detailed clinical
characteristics of BC after PM are shown in table 1. Presentations included: disease limited to the axilla without evidence of a
local primary 4 (16%); synchronous local and axillary disease 1 (4%); synchronous local disease and distant metastases 4 (16%);
clinically isolated local disease 17 (68%).
Of the 17 patients with isolated local disease, 11 (65%) underwent a completion/redo mastectomy, local excision of the tumor was
performed in 5 (29%), and surgical management was unknown in 1 (6%). Ten of 17 (59%) underwent axillary lymph node
dissection, 1 (6%) underwent sentinel lymph node biopsy, 1 did not undergo axillary staging, and axillary management was
unknown in 5 (29%). Median tumor size was 0.9 cm (range 0.3-3.5) and only 1 of 17 (6%) patients was confirmed to have
pathologic nodal involvement. Twelve of 17 (71%) received some type of adjuvant therapy: chemotherapy and/or endocrine
therapy 3 (18%); radiotherapy 2 (12%); both 5 (29%); none 5 (29%). With a median follow-up of 7 years since diagnosis of local
BC after PM, there has been one isolated local recurrence and 2 distant recurrences as first event.
Conclusion: BC can occur after PM. With rising rates of PM, understanding management of BC after PM is important. Most
common presentation is local disease and can be managed with resection with consideration of adjuvant therapy.
Multidisciplinary management of these cases is needed.
Characteristics of BC Following PM
All Patients (n=25)
BPM Cohort (n=13)
CPM Cohort (n=12)
Median age at diagnosis of BC after PM
56 (range 38-81)
58 (range 38-71)
54 (range 39-81)
Median time to development of BC after PM (years)
7 (range 1-25)
6 (range 2-25)
8 (range 1-21)
Self-detected abnormality
23 (92%)
12 (92%)
11 (92%)
Screening mammogram
1 (4%)
2 (16%)
Not known
1 (4%)
1 (8%)
Local disease only
17 (68%)
10 (77%)
7 (58%)
-Sub-areolar
7 (41%)
6 (40%)
1 (8%)
-UOQ/axillary tail
2 (12%)
2 (16%)
-Lower/inframmamary crease
2 (12%)
1 (10%)
1(8%)
Chest wall or unspecified
6 (35%)
3 (30%)
3 (25%)
Local & regional (axillary) disease
1 (4%
1 (8%)
Axillary BC without evidence of local primary
4 (16%)
1 (8%)
3 (25%)
Synchronous local and distant disease
3 (12%)
2 (15%)
1 (8%)
Presentation

Title: Adolescent and young adult breast cancer in Texas: Risk factors for delayed diagnosis, including distance to clinical trials
Deborah Vollmer Dahlke1 and Ellyn Cohen2. 1TX A&M School of Public Health, College Station, TX and 2University of California,
San Francisco Medical Center, San Francisco, CA.
Body: Background:Breast cancer is the most commonly diagnosed cancer among adolescent and young (AYA) adult females,
ages 15-39. Approximately 14% of all AYA cancers diagnosed in females are breast cancer. AYAs represent 7% of all female
breast cancer diagnoses. AYAs diagnosed with breast cancer have larger proportions of cancers with lower estrogen receptor
positivity and over expression of HER2, triple negative subtypes, and cancers associated with familial history. These factors
suggest the need for increased surveillance of young women diagnosed with breast cancer and strong consideration for their
enrollment into clinical trials.
Objective: To analyze geographic and socioeconomic factors that may increase risk and disparities in access to clinical trials for
Texas AYA breast cancer survivors.
Research Questions; 1) What are the factors, including access to clinical trials, that affect Texas AYA breast cancer patients
being diagnosed at later versus earlier stages? 2) What are the associations between various demographic and diagnostic risk
factors and AYAs breast cancer survivors distance to clinical trials?
Data Sources, Population and Methods: Data sources include SEER 18 and Texas Limited Use data bases for incidence
analyses. Data on breast cancer clinical trials was provided by www.BreastCancerTrials.org. Texas AYA breast cancer population
data for 4153 women diagnosed from 2005 to 2009, was provided by the Texas Cancer Registry under IRB protocol 13-022.
Methods for this study included use of SEER Stat for incidence analysis, ESRI ArcGIS 10.1 mapping and ESRI GIS Network
Analysis tools to determine individual patient locations and distance to trials. Stata statistical software (12.1) was used to for
bivariate and logistic regression modeling and analyses.
Selected Results:
1) In a comparison of U.S. SEER 18 and Texas Limited Use data, for the overall female AYA breast cancer populations and for
nearly all of the racial and ethnic subgroups, Texas' age adjusted incidence rates for this age group (15-39) are higher than the
U.S. incidence rates.
2) With insured patients as the reference, uninsured self pay patients were 70% (p-value< 0.000, CI 1.31, 2.20) more likely to be
diagnosed at later stages;
3) Being of Hispanic/Latina ethnicity was associated with at 36% greater risk of late stage diagnosis (p-value <0.000, CI
1.16,1.59)
4) Black African American AYAs were 31% more likely to be diagnosed at a later stage (p-value 0.004, CI 1.08,1.58)
5) Using <45 miles distance to clinical trials as a reference, AYA breast cancer patients living 45-100 miles from an appropriate
trial were 102% more likely to be diagnosed at later stages (p-value< 0.000, CI 1.65,2.48)
6) AYA patients living furthest from trials, over 200 miles, were 49% more likely to be diagnosed at later versus earlier stages of
breast cancer (p-value 0.020, CI 1.06, 2.09).
Discussion: This is the first study to consider geographic, demographic and diagnostic factors, including distance to appropriate
trials, in exploring early versus late staging of breast cancer among AYAs. The findings and discussion of results have both
practice and policy implications related to access to care and clinical trials for AYA breast cancer survivors.

Title: Association between breastfeeding and breast cancer risk by receptor status: A meta-analysis
Marisa Weiss1,2, Ying Liu3, Paolo Boffetta4, Graham Colditz3, Ahmedin Jemel5 and Farhad Islami5. 1Breastcancer.org, Ardmore,
PA; 2Lankenau Medical Center, Wynnewood, PA; 3Washington University School of Medicine, St Louis, MO; 4Icahn School of
Medicine at Mount Sina, New York, NY and 5American Cancer Society, Atlanta, GA.
Body: Background:
The rising incidence of breast cancer is mainly due to changes in reproductive, lifestyle, and environmental factors, not inherited
genetic mutations. Many risk factors can be modified, offering important opportunities for prevention.
In the era of personalized care, treatment is subtype dependent. While most prevention strategies are not subtype specific, we
wanted to see if breastfeeding confers the same protection based on subtype, especially against poorer prognostic subtypes.
Methodology:
Relevant articles from casecontrol or prospective studies were identified by searching the PubMed and Scopus databases
through 2014 and reference lists of relevant articles. Two researchers independently did the search and evaluated the articles.
The summary risk estimates and 95% confidence intervals were calculated using random effects models (DerSimonianLaird
method) for the association between breastfeeding and breast cancer by receptor status. The reference category in most of the
studies was never breastfeeding, but in a few studies this also included women who breastfed for a short time.
Results:
This metaanalysis of 27 articles (19 from casecontrol, 8 from prospective cohort studies) published between 1983 and 2014
included 36,881 women with invasive breast cancer from four continents. Ever breastfeeding was inversely associated with breast
cancer risk in hormone receptornegative subtypes, even in cohort studies and when the results were adjusted for age, body
mass index (BMI), parity, and family history of breast cancer (RR= 0.88; 95% CI 0.74 1.06 for nonluminal cancers). The
inverse association between ever breastfeeding and triplenegative subtype was slightly stronger, but this was based on a limited
number of cohort studies. In contrast, there was no association between breastfeeding and luminal subtypes in cohort studies.
Association between ever breastfeeding and breast cancer risk by receptor status
Receptor status
No of OR (95% CI) I2
P for
studies
statistics heterogeneity
(%)
Nonluminal (ER,
PR)
Cohort
0.84 (0.72
0.97)
50
0.06
Cohort, adjusted*
0.88 (0.74
1.06)
42
0.18
Casecontrol
13
0.76 (0.67
0.86)
59
0.004
Cohort
0.74 (0.62
0.88)
0.46
Cohort, adjusted*
0.81 (0.62
1.04)
Casecontrol
0.73 (0.64
0.84)
12
0.34
Triplenegative
Luminal (ER+
and/or PR+)
Cohort
0.98 (0.89
1.07)
75
<0.001
Cohort, adjusted*
1.04 (0.98
1.10)
0.75
Casecontrol
18
0.82 (0.76 69
<0.001
0.89)
, negative; +, positive; ER, estrogen receptor; PR, progesterone receptor
I2 statistics show % of total variation across studies that is due to heterogeneity rather than random variation (chance).
Arbitrarily,I2 percentages may be interpreted as follows: 25%, low; 50%, moderate; and 75, high heterogeneity.
*Results adjusted at least for age, body mass index, parity, and family history of breast cancer.
Conclusion:
Breastfeeding is a powerful strategy to reduce the risk of several aggressive breast cancer subtypes, with a relative risk reduction
of approximately 10% to 20%, depending on receptor status. To maximize breastfeeding use for the longterm health of mothers
and babies, it is important to remove barriers in the home, community, and workplace as well as provide targeted education and
support before and after delivery.

Title: The beliefs, knowledge, understanding, attitudes and treatment access to breast cancer amongst rural women in Northern
Nigeria
Malcolm O Tagbarha1. 1University of Abuja, Abuja, Federal Capital Territory, Nigeria.
Body: Goal: The goal of this study is to ascertain the beliefs, knowledge, understanding, attitudes and treatment access to breast
cancer among rural women in Nigeria.
Background: Breast cancer has become a popular topic in recent years with several thousands of women diagnosed to be
positive every year. The availability of care/treatment upon early detection is key to survival.
Methods: An interview guide was designed specifically for this study in which 200 rural women in Northern Nigeria, age 45 and
over took part in. It contained questions about beliefs, knowledge, understanding and attitudes about Breast Self-Examination
(BSE), Clinical Breast Examination (CBE) and mammogram. In addition, questions assessing the variables of the Health Belief
Model and health motivations also were included. The data were obtained during face-to-face interviews in the primary language
of the participating woman. The interviews were transcribed and translated into English.
Results: Out of the 200 women who participated, only 1% two (2) of the participants practiced BSE monthly, 8% had undergone
at least one CBE during their lives, and 91% had never had a mammogram. There were little or no access to treatment even at
early detection in these rural areas causing thereby vulnerability to loss of life. Majority of these rural women (95%) said they
knew little or nothing about breast cancer. While 15% of the women said detecting cancer early was important, only 3% reported
that cancer could be cured. Age, education, or mother tongue showed no statistically significant relationship with the breast health
practice scores. However, proficiency with the English language (p = 0.009) and number of years exposed to awareness and
education (p = 0.009) had a significant relationship with the breast health practice scores. The significant explanatory factor for
the variable breast health practices was a cue to action (p = 0.009).
Conclusions: The level of awareness and treatment access to breast cancer amongst Northern Nigerias rural women is
extremely low thereby making them not to engage in screening and/or detection practices. This alarming situation calls for urgent
intervention of medical/health organizations to provide immediate breast cancer awareness, screening and care so as to reduce
incidences or threat at early detection.

Title: Docosahexanoic acids modulation of survival and invasion is associated with altered CCL20/CCR6 chemokine levels and
signaling in hyperplastic, DCIS and metastatic breast cancer cell lines
Ching Hui Chen1, Laura Garcia1, Carol Fabian2, Stephen Hursting3 and Linda deGraffenried1. 1University of Texas, Austin, TX;
2
University of Kansas Cancer Center, Kansas City, KS and 3University of North Carolina, Chapel Hill, NC.
Body: Introduction: High dietary intake of docosahexanoic acid is associated with a lower risk of breast cancer and reduced
metastasis. Epidemiological and preclinical studies suggest that the regulation of monocyte recruitment may play an important
role in lowering breast cancer risk and decreasing breast cancer metastasis. In addition to promoting the recruitment of
pro-inflammatory leukocytes, the CCL20/CCR6 chemokine axis has been implicated in promoting breast cancer cell migration
and invasion. We hypothesize that one mechanism by which DHA suppresses breast cancer progression and metastasis is
through the suppression of CCL20/CCR6 signaling.
Methods: The 21PT, 21NT and 21MT-1 cell lines have been previously described as reflecting the characteristics of hyperplastic,
in situ and metastatic breast cells, respectively. We measured the impact of physiological DHA concentrations on cell survival and
cell proliferation using colony formation and MTT assays respectively. Invasion of 21PT, 21NT and 21MT-1 cells were evaluated
using invasion chambers. Changes in CCL20 and CCR6 expression were measured using qPCR. The activity and expression
levels of JNK, ERK1/2 and c-Jun, downstream modulators of the CCL20/CCR6 axis, were evaluated using Western blot analyses.
Results: Following a 24 hr exposure to 20 uM DHA and 5-7 days of recovery, colony counts of all three cell lines were significantly
suppressed, with 21NT cells experiencing the largest percent reduction. Invasion capacity of 21PT, 21NT and 21MT-1 accurately
mirrored the stages of breast cancer they represent. Treatment with DHA reduced the invasion capacity of 21MT-1 to the levels of
its hyperplastic counterpart. CCL20 mRNA levels were reduced when cells were exposed to DHA for 48 hrs. Western blot
analyses suggest that activation of ERK1/2 and JNK signaling may be critical in orchestrating DHA-associated reduction of
invasion.
Conclusions: With the support from epidemiological and preclinical studies, the use of omega-3 based preventive regimen may
prove useful for reducing breast cancer risk and breast cancer metastasis. In vitro studies using premalignant breast cancer can
provide invaluable insights to the molecular mechanisms accountable for preventive properties of omega-3 fatty acids.
DHA-associated regulation of CCL20/CCR6 signal transduction may be an important preventive mechanism and future studies
will warrant deepened understanding of how nutraceutical compounds can prevent breast cancer.

Title: Efficacy of tabebuia avellandae extract on a cell culture model for triple negative breast cancer
Nitin T Telang1, Hareesh B Nair2 and George YC Wong3. 1Palindrome Liaisons Consultants, Montvale, NJ; 2Texas BioMedical
Research Institute, San Antonio, TX and 3American Foundation for Chinese Medicine, New York, NY.
Body: Background: The triple negative breast cancer (TNBC), a molecular subtype of clinical breast cancer consisting of
epithelial cells that lack estrogen receptor- (ER), progesterone receptor (PR) and human epidermal growth factor receptor-2
(HER-2) expression, is non-responsive to either endocrine therapy or HER-2 targeted therapy. Chemotherapy is frequently
associated with long-term systemic toxicity, acquired tumor resistance and resultant compromised treatment efficacy. These
aspects emphasize a need to identify efficacious non-toxic agents for secondary prevention/therapy of TNBC. Non-fractionated
aqueous extract from the inner bark of the Tabebuia avellandae (TA) tree found in the Amazon rainforest, available as a dietary
supplement under the name of Taheebo or Pau darco, has documented efficacy in a cell culture model for the Luminal A breast
cancer subtype, as well as for several other cancers. Present study examines the inhibitory effects of the TA extract, and
identifies possible mechanistic targets for its efficacy in a cell culture model for TNBC.
Nutritional Supplement, Experimental Model and Biomarkers: Lyophilized powder of non-fractionated TA extract from Taheebo
Japan, Osaka, Japan, provides the source material for the study. The ER-/PR-/HER-2- MDA-MB-231 cell line represents the cell
culture model for TNBC. Anchorage dependent growth, cell cycle progression, status of cell cycle regulatory proteins and
anchorage independent colony formation, represent the quantitative biomarkers for efficacy.
Results: Relative to the non-tumorigenic 184-B5 human mammary epithelial cells, the tumor derived MDA-MB-231 cells exhibited
decreased population doubling time, increased saturation density, decreased G1: S+G2/M ratio and increased S+G2/M: Sub G0
ratio, indicating loss of homeostatic growth control. Additionally, unlike 184-B5 cells, MDA-MB-231 cells exhibited increased
anchorage independent growth in vitro and tumor development in vivo, indicating enhanced cancer risk. Treatment of
MDA-MB-231 cells with TA resulted in a substantial dose dependent cytostatic growth arrest (IC50:1.0%; IC90: 2.5%). Cell cycle
analysis of TA treated cells revealed G1 arrest, leading to a progressive dose dependent increase in the G1: S+G2/M ratio.
Mechanistically, TA decreased Cyclin D1 expression and attenuated RB phosphorylation, predicting Cyclin D-CDK4-pRB pathway
as a molecular target for efficacy. Furthermore, TA effectively inhibited anchorage independent growth in a dose dependent
manner.
Conclusions: Present data demonstrated pronounced efficacy of TA as a naturally occurring nutritional substance in a cell culture
model for TNBC, and validated TA as a promising non-toxic natural agent for secondary prevention/therapy of clinical TNBC.

Title: Evaluation of sustained antiemetic efficacy over repeated cycles of anthracycline-cyclophosphamide (AC)-based
chemotherapy: A phase 3 study of NEPA, a fixed-dose combination of netupitant and palonosetron for prevention of
chemotherapy-induced nausea and vomiting (CINV)
Matti Aapro1, Hope Rugo2 and Giada Rizzi3. 1Institut Multidisciplinaire d'Oncologie, Clinique de Genolier, Genolier, Switzerland;
2
University of California, San Francisco Comprehensive Cancer Center, San Francisco, CA and 3Helsinn Healthcare, Lugano,
Switzerland.
Body: Background: International antiemetic guidelines recommend co-administration of an NK1 receptor antagonist (RA) and a
5-HT3 RA in breast cancer (BC) patients receiving anthracycline cyclophosphamide (AC) chemotherapy as this population is at
increased risk of developing CINV. NEPA, a novel, oral fixed-dose combination of a new NK1 receptor antagonist (RA), netupitant
(NETU 300 mg), and the 5-HT3 RA, palonosetron (PALO 0.50 mg), was previously reported to be superior to oral PALO after a
single cycle (Aapro et al, Annals of Oncology 2014) and multiple cycles (Aapro et al, ASCO 2014) of chemotherapy. This posthoc
analysis evaluates sustained efficacy over multiple cycles when censoring patients who experienced CINV in the previous cycle.
Methods: This was a multinational, randomized, double-blind, parallel group study evaluating the efficacy/safety of single oral
doses of NEPA versus oral PALO in chemotherapy-nave patients receiving multiple cycles of anthracycline-based
chemotherapy. All patients also received oral dexamethasone (DEX) 12 mg (NEPA) or 20 mg (PALO), only on Day 1. Overall
(0-120 h) complete response (CR: no emesis, no rescue medication) was the efficacy endpoint evaluated. The analysis of
sustained CR evaluates the probability that patients would remain complete responders over 4 cycles by censoring continuing
patients who failed to have a CR in the prior cycle. A Kaplan Meier method and log-rank test comparing NEPA with oral PALO
were utilized.
Results: 1455 patients were randomized; 1286 participated in the multiple cycle extension after cycle 1. Treatment groups were
comparable with 98% females and 97% with BC; the mean age was 54. The percentage of patients who experienced a CR in
cycle 1 and who sustained a CR over cycles 2-4 was greater for NEPA than for oral PALO (p <0.0001, log rank test). The table
shows the percent of patients with continuing CR over time; N = the number of patients at risk.
Time since first chemotherapy
NEPA + DEX
Oral PALO + DEX
Cycle 1
74.3% (N = 724)
66.6% (N = 725)
Cycle 2
68.5% (N = 485)
57.1% (N = 434)
Cycle 3
65.7% (N = 423)
52.7% (N = 348)
Cycle 4
63.6% (N = 375)
50.6% (N = 300)
Conclusions: This multiple cycle analysis indicates that NEPA, a novel, fixed-dose antiemetic combination, more effectively
demonstrates sustained control of CINV over multiple cycles than oral PALO. As females with breast cancer represent a
particularly challenging population in terms of emesis control, it is especially crucial that antiemetic recommendations are followed
to allow these patients to maintain their quality of life and continue their treatment plan over multiple cycles of chemotherapy.
NEPA offers effective guideline-recommended prophylaxis in a convenient single dose.

Title: Safety of letrozole-gonadotropin controlled ovarian stimulation protocol in women with breast cancer undergoing fertility
preservation before or after tumor resection via embryo or oocyte cryopreservation: A prospective cohort study
Kutluk Oktay1,2, Jayeon Kim1,2, Giuliano Bedoschi1,2 and Volkan Turan1,2. 1New York Medical College, New York, NY and
2
Innovation Institute for Fertility Preservation and In Vitro Fertilization, New York, NY.
Body: Purpose: We have previously described the concurrent use of aromatase inhibitors to reduce estrogen exposure in women
with breast cancer undergoing controlled ovarian stimulation (COS) with gonadotropins for fertility preservation (FP) via oocyte or
embryo cryopreservation. To purpose of this study was to investigate the impact of this letrozole-gonadotropin COS protocol on
survival in women who underwent fertility preservation before or after breast surgery.
Patients and Methods: A total of 364 women with stage 3 breast cancer, who pursued FP consultation or FP treatments at our
institution were prospectively evaluated. Of those, 146 elected to undergo COS with letrozole and gonadotropins for FP (120 prior
to chemotherapy and 26 after chemotherapy). The remaining 218 patients elected to not to undergo a fertility-preserving
procedure and served as controls.
Result(s): Demographic information and tumor characteristics at enrollment were similar between patients who pursued COS with
letrozole and gonadotropins (COS group) and control groups.
Table 1. Demographics and tumor characteristics of patients who pursued fertility preservation vs. controls
Treatment Group (n=120)
Control Group (n=218)
P-value
Age at FP consultation (years)
35.24.5
37.05.1
0.03
Age at cancer diagnosis (years)
34.84.5
34.94.7
0.88
BMI (kg/m2)
22.83.7
23.13.9
0.56
Node involvement (%)
34
48
0.02
1.91.9
0.62
<2 67
66
0.59
2-5 31
30
>5 2
Tumor size (cm)

MeanSD 1.8 1.2
Lymphovascular space invasion (%)
31
37
0.44
1-2 40
39
1.00
3 57
59
Estrogen receptor positive (%)
82
77
0.27
HER-2/neu positive (%)
35
33
0.78
Adjuvant tamoxifen use (%)
89
87
1.00
Length of follow-up (years, MeanSD)
5.02.1
6.93.6
<0.001
Histologic grade
The median follow-up after diagnosis was 4.9 years in COS and 6.2 years in the control group. In the COS group, the hazard ratio
(HR) for recurrence after IVF was 0.77 (95% CI: 0.28, 2.13) and the survival was not compromised compared with controls
(P=0.61). In the COS group, survival was not different between patients with ER-positive and ER-negative breast cancer (P=0.75)
and between patients who underwent COS before and after tumor resection (P=0.56). The survival was also not different
between patients who pursued COS before and after chemotherapy (P=0.57).
Conclusion(s): Here we presented the largest prospective data with longest follow up on the safety of ovarian stimulation in
women with breast cancer. COS with letrozole and gonadotropins for FP is unlikely to cause substantially increased recurrence
risk in breast cancer, even in patients who have not yet undergone breast surgery. Larger studies are needed to confirm the
findings from the subgroup analysis.
Support: Supported by NIH RO1 HD053112.

Title: Baseline joint pain predicts severity of subsequent joint symptoms in women initiating aromatase inhibitors for early stage
breast cancer
Lea Baer1, Katherine D Crew1, Danielle Awad1, Kevin Kalinsky1, Matt Maurer1 and Dawn L Hershman1. 1Columbia University
Medical Center, New York, NY.
Body: Background: Aromatase inhibitors (AIs) are the standard adjuvant treatment of hormone-sensitive breast cancer (BC) in
postmenopausal women. However, these therapies are associated with musculoskeletal complaints which may lead to
non-adherence and early discontinuation. The aim of this study was to characterize the natural history of the AI-induced
arthralgia.
Methods: Postmenopausal women with stage I-III BC were prospectively enrolled at the onset of adjuvant AI therapy. Subjects
completed the Brief Pain Inventory (BPI) questionnaire at baseline, 3, 6, 9, and 12 months. BPI worst pain scores were
categorized as 0-3, 4-6 and 7-8. Multiple logistic regression analysis was used to evaluate the association between baseline
factors and having a 2-point worsening in BPI worst pain score. A Cox-proportional hazards model was used to evaluate factors
that influenced the time to first 2-point worsening in pain score. Clinically significant change was defined as 2-point increase
from baseline. Those with a baseline BPI worst pain score of 9 were removed from analysis.
Results: Among 180 consented subjects, 1 was found to be ineligible due to being perimenopausal, 42 subjects were lost to
follow up, 17 subjects came off their AI and 8 subjects dropped out. Mean age was 61; 60% were white, 32% Black, 10%, Asian
and 31% were Hispanic; 86% started on anastrazole; 24% had a change in AI. At baseline, 76 women had a BPI worst pain score
between 0-3; 28 between 4-6; and 18 had 7+. Seventy subjects (64%) experienced at least a 2-point worsening of BPI from
baseline and women who developed a 2-point worsening had a lower mean baseline BPI (2.57 vs. 3.75) compared to those who
did not. Those experiencing an initial worsening in the first 6 months of therapy improved over time; while the patients
experiencing BPI worsening at 9 months, continued to have progressive increase in BPI. In a multiple logistic regression model
adjusting for AI switching, BMI, age, baseline score, prior chemotherapy, osteoarthritis, and prior hormone replacement therapy,
those who had a baseline worst pain score between 4-6 (p=0.04) or 7+ (p=0.005) were less likely to develop a 2-point worsening
in BPI at 12 months from baseline. Similarly, those with a baseline worst pain categorization of 7+ had a hazard ratio of 0.34 for
time to 2-point worsening.
Conclusions: Women with low baseline BPI score are more likely to develop worsening BPI score over time. Women who develop
symptoms later in the course of their therapy are at greatest risk for persistent symptoms. This has implications for studies
evaluating interventions for prevention of AI arthralgias.
BPI
mean
score
Patients with >2

Patients with >2
Patients with >2
Patients with >2
Patients with >2
point increase in BPI point increase in BPI point increase in BPI point increase in BPI point increase in BPI
at any time
at 3 months
at 6 months
at 9 months
at 12 months
Baseline
score
3.75
2.57*
2.59
2.30
2.11*
1.85*
Three
month
score
4.25
4.52
6.08*
4.68
4.13
3.87
Six month
score
4.50
4.83*
5.32*
6.37*
5.07
4.22
Nine month
4.00
score
5.10
5.11
5.13
5.75*
5.04
Twelve
month
score
5.37*
5.48
4.86
5.66
6.21*
4.25
*p value <0.05 compared to baseline

Title: Prognostic understanding and associations with mood and quality of life in patients with metastatic breast cancer
Amanda Parkes1, Jennifer A Shin1, Helen Knight1, Stephen M Schleicher2, Areej El-Jawahri1, Lara Traeger1 and Jennifer S
Temel1. 1Massachusetts General Hospital, Boston, MA and 2Brigham and Women's Hospital, Boston, MA.
Body: Background:
Data suggest that patients with metastatic cancer who understand their prognosis are more likely to prefer and receive care
concordant with their wishes. Despite the importance of prognostic information for patients decision-making, there are no current
data describing prognostic understanding in patients with metastatic breast cancer (MBC). The aims of this study were to
describe prognostic understanding in patients with MBC and to explore its associations with mood, distress, and quality of life
(QOL).
Methods:
We conducted a cross-sectional study of 50 patients who were receiving first- or second-line chemotherapy for MBC. Participants
completed a series of questionnaires. We used a 13-item questionnaire to assess patients perceptions of their prognosis and
goal of therapy. We evaluated mood, level of distress, and QOL using the Hospital Anxiety and Depression Scale (HADS), the
Distress Thermometer (DT), and the Functional Assessment of Cancer Therapy-Breast (FACT-B), respectively.
Results:
The majority of patients (92%) reported it was at least somewhat important to know details about their prognosis. 19/47 (40%)
patients reported that the primary goal of treatment was to cure their cancer, and 21/46 (46%) patients reported that the chance of
cure was at least somewhat likely (25% chance of cure). 24/49 (49%) patients viewed themselves as terminally ill. There was a
high prevalence of psychological morbidity in our patient cohort. 24/48 (50%) patients screened positive for distress (distress
thermometer 4), 17/50 (34%) patients reported significant anxiety symptoms (HADS-Anxiety 8), and 11/50 (22%) patients
reported significant depression symptoms (HADS-Depression 8). Distress, depression, and anxiety were each associated with
lower QOL scores (93.0 vs. 114.0, p<0.001; 87.9 vs. 107.8, p=0.002; and 85.5 vs. 112.8, p<0.001, respectively). Patients who
acknowledged their illness as terminal reported higher depression than those who did not perceive themselves as terminally ill
(M=6.1 vs. 2.5, p=0.0006).
Conclusion:
Although the majority of patients with MBC receiving first- or second-line chemotherapy feel it is important to know detailed
information about their prognosis, many incorrectly perceive that their cancer is curable. Accurate prognostic understanding was
associated with increased depression symptoms. This study highlights the need to develop interventions to enhance patients
prognostic understanding while providing adequate psychosocial support.

Title: Evaluation of the effect of low level laser therapy on oral mucositis in breast cancer patients: A retrospective analysis
Jeroen Mebis1,2,3, Sandrine Censabella1, Annelies Maes1,3, Leen No1,3 and Paul Bulens1,3. 1Jessa Hospital, Hasselt, Belgium;
2
Hasselt University, Diepenbeek, Belgium and 3Limburg Oncology Centre, Hasselt, Belgium.
Body: Background
The goal of this retrospective study was to investigate the effectiveness of low level laser therapy (LLLT) in managing
chemotherapy-induced oral mucositis (OM) in breast cancer patients.
Methods
Breast cancer patients treated with chemotherapy at the Jessa Hospital (Hasselt, Belgium) and having received LLLT for OM
were retrospectively selected from our database, provided sufficient data with regard to OM was available.
LLLT treatment was provided using an AsGA diode laser ( = 665 nm; output power: 100mW) combined with an infrared laser
(continuous emission, output power: 500mW), delivered by an optical fiber with a diameter of 600m. The energy delivered was 4
J per point of application. Treatment was applied to a maximum of seven sites (tongue, palate, tonsil, left and right inside of the
cheek, floor of the mouth, and lips), depending on the location of OM. Patients received treatment two times a week until healing
of each lesion.
Endpoints were the number of treated areas and the severity of OM at the start and the end of LLLT, graded by trained nurses
according to the WHO scale (0 = no change; 1 = soreness, erythema; 2 = erythema, ulcers, can eat solids; 3 = ulcers, requires
liquid diet only; 4 = oral alimentation not possible; if more than one location was treated, the highest grade was taken into
account). An OM score was calculated for each patient by summing the WHO grades of all treated areas. Finally, where
available, pain scores (obtained through a visual analogue scale ranging from 0, no pain, to 10, worst possible pain) were taken
into account.
Results
Data from 93 patients with stage 0-IV breast cancer were included in these analyses. Mean age was 55.37 years (standart
deviation [SD] = 9.72, median = 56). Most of the patients received anthracycline-based chemotherapy (65%). At the start of LLLT,
mean time since start of chemotherapy was 48.92 days (SD = 39.43, median = 39). The median duration of LLLT was 2 weeks.
OM outcomes at the start and the end of LLLT are presented in Table 1. At the end of LLLT, the number of areas that had to be
treated significantly decreased. More importantly, there was a significant improvement in the severity of OM (highest WHO grade
and OM score) and in pain. This improvement was also observed when patients were categorized according to their status at the
end of LLLT (for each OM outcome: worsened, unchanged, or improved).
Table 1. Patients' status at the start and the end of Low Level Laser Therapy (LLLT)
Outcome
Start LLLT
End LLLT
N improved
Mean number of treated areas
3.89
2.16*
66 (71%)
Mean OM score
6.60
2.78*
75 (80.6%)
Mean pain scorea
5.14
1.64*
20 (90.9%)
11 (11.8%)
60 (64.5%)*
60 (64.5%)
N (%) WHO grade 1

a
Pain scores were available for 22 (of the 93) patients. * p < 0.0001 (t-test or chi-square, as appropriate).
Conclusion
This retrospective analysis showed that LLLT, a standard management strategy for OM in head and neck cancer, significantly
reduced the severity of chemotherapy-induced oral mucositis and relieved pain in patients with breast cancer. This is the first
study in this population. Further research, preferably high-quality randomized controlled trials, is warranted to better investigate its
usefulness in this population.

Title: Open-label randomized parallel controlled study comparing bone mineral density between alendronate plus alfacalcidol
combination and single administration of alfacalcidol in postmenopausal women receiving aromatase inhibitor as adjuvant therapy
Mitsue Saito1 and Joe Matsuoka1. 1Juntendo University, Bunkyo-ku, Tokyo, Japan.
Body: <Background>
One of the most worrisome side effects of endocrine therapy is loss of bone mineral density. Optimized bone therapy is thus
warranted.
<Objectives>
The primary endpoint of this study was to assess the difference in bone mineral density between aromatase inhibitor-treated
breast cancer patients receiving alfacalcidol alone versus also prescribed alendronate.
Secondary endpoints were measured levels of surrogate markers for bone health and adverse events associated with
bone-preserving therapies.
<Material and method>
Postmenopausal breast cancer patients (stage I-III) receiving any form of aromatase inhibitor (anastrozole (ANA), exemestane
(EXE) or letrozole (LTZ)) whose bone mineral density as measured by DEXA (Dual-energy X-ray absorptiometry) was lower than
the adult mean and from whom written informed consent had been obtained between March in 2008 and September in 2010 were
studied. The study period was 2 years after enrollment. This study was approved by our institutional review board.
Patients were randomized into two arms stratified by age (<70 or not), use of an aromatase inhibitor (ANA, EXE or LTZ) and T
score on DEXA (<-1.0 or not). Patients enrolled in arm A were treated with 35mg of oral alendronate weekly and 1 g of
alfacalcidol daily. Patients in arm D were given 1 g of alfacalcidol daily.
Patients underwent DEXA (L2,3,4) every 6 months and blood tests for 1CTP (carboxyterminal telopeptide of type I collagen), BAP
(bone alkaline phosphatase) and urine testing for NTX (type I collagen cross-linked N-telopeptide) every 3 months. Adverse
events were monitored by physicians every 3 months.
<Results>
We enrolled 58 patients. Nine out of the 29 patients in arm D dropped out due to adverse events caused by alfacalcidol (2),
caused by the aromatase inhibitor (2), metastasis (3) and severe bone loss (1). Six out of the 29 patients in arm A dropped out
due to adverse events caused by alendronate plus alfacalcidol (1), caused by the aromatase inhibitor (2), metastasis (2) and
financial difficulties (1). Improvement from the DEXA baseline in arm A was significantly better (p<0.0001) than that in arm D by
analysis of covariance. NTX and BAP Improved significantly in arm A (p<0.0001), but 1CTP did not (p=0.0382).
<Considerations>
Strict indications for and durations of bone therapy require further investigation. The dosage of alendronate was lower in this
study than the international recommendation because the dosage approved by the Japanese government is half that in other
countries and this applies to all the oral bisphosphonates because of the ethnic differences in absorption rate demonstrated in
phase II studies.
<Conclusions>
Co-administration of alendronate and alfacalcidol contributed to preserving bone mineral density during adjuvant aromatase
inhibitor treatment without producing severe adverse events. NTX and BAP are possible surrogate markers for bone therapy.

Title: Association of patient preference for adjuvant chemotherapy (chemo) at baseline (BL) with toxicity, mental health, function,
quality of life (QoL) and survival in older women with early stage breast cancer (ESBC) [CALGB 49907 Alliance]
Ajeet Gajra1, Linda McCall2, Hyman B Muss3, Harvey J Cohen2, Aminah Jatoi4, Karla V Ballman4, Ann H Partridge5, Linda Sutton2,
Barbara A Parker6, Gustav Magrinat7, Jaqueline M Lafky4 and Arti Hurria8. 1Upstate Medical University, Syracuse, NY; 2Duke
University Medical Center, Durham, NC; 3University of North Carolina Cancer Center, Chapel Hill, NC; 4Mayo Clinic, Rochester,
MN; 5Dana-Farber Cancer Institute, Boston, MA; 6UC San Diego, Moores Cancer Center, LaJolla, CA; 7Cone Health Cancer
Center, Greensboro, NC and 8City of Hope Comprehensive Cancer Center, Duarte, CA.
Body: Background: In CALGB-49907 (NEJM 2009;360:2055), older patients (pts) with ESBC were randomized to standard
adjuvant chemo (AC or CMF versus capecitabine). The objective of this secondary QoL analysis is to assess if pts BL chemo
preference (CP, defined as high or low), is associated with the following during and after completion of chemo: self and
professional-reported toxicity, changes in mental health, function, QoL, recurrence-free (RFS) and overall (OS) survival.
Patients and Methods: Of 633 trial pts 350 participated in the QoL substudy; 145/350 pts completed the BL assessment regarding
CP. CP was measured by asking the amount of benefit women would require to choose adjuvant chemo in a hypothetical
situation, irrespective of chemo agents. Women who chose chemo for an increase in OS of 12 months (mo) were designated as
high chemo preference (HCP) and those who chose >12 mo were designated low chemo preference (LCP). CP associations
were evaluated with: BL perception of self-health and perceived QoL on chemo; patient reported outcomes (PROs), changes in
function and QoL (based on EORTC-QLQ-C30); anxiety and depression (Hospital Anxiety and Depression Scale); observed
grade 3-5 adverse events (AEs) by NCI common toxicity criteria (CTC v2.0). Pts were assessed at midtreatment and at 1, 12, 18
and 24 mo post-treatment. Chi-square tests, t-tests, and Cox models were used for categorical, continuous, and time-to-event
variables, respectively.
Results: The demographic and tumor characteristics of women (median age 71) who provided CP at BL were not different from
women in the QoL subset or from non-QoL pts. 68/145 (47%) women had a HCP. CP groups did not differ based on age, surgery
type, tumor and nodal stage, hormone receptor status, performance status, chemo assignment, education, marital or employment
status except the LCP group had a higher proportion of white women (95% vs. 78%, p=0.004).
At BL, there were no differences in perception of self-health based on CP but women with LCP predicted QoL on chemo to be
worse than women with HCP (p=0.006) and reported greater nausea/vomiting. Mid-treatment, LCP pts reported worse
nausea/vomiting, financial worries, and cancer symptoms. Post-treatment, LCP pts had worse constipation (at 1 mo) and financial
worries (at 24 mo).
There were no differences based on CP for dyspnea, pain, fatigue, insomnia, anxiety or depression at any timepoint.
Mid-treatment, LCP women reported lower QoL and worse social, emotional and physical function compared to HCP women.
These scores were not significantly different after treatment completion. LCP women had significantly higher rates of grade 3-5
AEs (53 vs. 34%, p=0.02) during treatment but these did not persist post-therapy. CP was not significantly associated with OS
(HR =0.75, p=0.36) or RFS (HR=0.94, p=0.84).
Conclusions: LCP at BL was associated with lower QoL, worse physical symptoms, AEs and function mid-therapy, but not mental
health. Mid-therapy declines in women with LCP largely reversed post-therapy. This information may be useful for oncology
professionals to counsel older ESBC pts with LCP receiving adjuvant chemo.

Title: Exercise intervention to run away from breast cancer treatment side effects: An integrative approach
Soraya Casla1, Sara Lpez-Tarruella2, Yolanda Jerez2, Ivn Mrquez-Rodas2, Ricardo Cubedo3, Isabel Calvo4, Ana Martinez1,
Sara Cano1, Rubn Barakat1 and Miguel Martn2. 1Technical University of Madrid, Faculty of Physical Activity and Sport
Science-INEF, Madrid, Spain; 2Servicio Oncologa Mdica. Instituto de Investigacin Sanitaria Gregorio Maran, Madrid, Spain;
3
Hospital Universitario Puerta de Hierro, Majadahonda, Madrid, Spain and 4Hospital Madrid Norte-Sanchinarro, Madrid, Spain.
Body: Background. Some studies have shown that exercise increases breast cancer (BC) patients quality of life (QoL) during
and after treatments. The aim of this study was to investigate the effect of an specific exercise intervention in QoL and in the
exercise levels of these patients.
Methods. A randomized controlled trial evaluated an intervention (IG) vs a control group (CG) in early stage BC patients who
recently finished chemo & radio therapies. Intervention consists on controlled group classes combining aerobic and resistance
activities. CG maintained their previous lifestyle. QoL, leisure-time exercise levels (LTEL), chest (CMS) and legs maximal strength
(LMS), physical capacity (PC) and some psychological variables, were assessed at baseline and after 3 months in both groups.
Women who participated in the IG were followed up (FU) after 6 months.
Statistical analysis (SA) were performed using unpaired t-test on continuous variables comparing CG vs IG, considering possible
confounders. Pearson correlation analyses were employed to examine possible correlations. 6 months FU data were analyzed
comparing baseline and after intervention results using non-parametric test (n=13). SA was performed with SPSS v18 software.
95% CI was calculated and statistical significance level of p < 0.05 was used.
Results. 59 women (median 48.978.35 years old) completed this study. No differences between groups were observed at
baseline. Average attendance rate to the program was 89%. There was a significant rise in LTEL (t123=16.33; p=0.0001) and in
QoL (t123=2.88; p=0.005) comparing IG vs CG. Results also showed a correlation between QoL and LTEL in patients of IG only
(r=0.22; p=0.013). Significant differences between groups were observed in both physical and psychological variablesafter
training program.
Table 1. Results summary comparing intervention and control group.
Intervention Group (n= 27)
MeanSD
Control Group (n=32)

MeanSD
Follow-Up (n=13)
MeanSD
QoL*+
112.8817.74
103.1119.26
110.9614.40
LTEL*+
45.1114.61
14.874.78
34.5619.51
Body Fat Mass
34.636.7
36.477.47
35.744.56
Muscle mass
41.33.16
38.788.51
41.482.01
Strength Index*+
2.60.83
2.120.69
2.320.58
CMS*+
45.8911.07
31.479.87
53.7612.42
LMS*+
93.0727.3
69.0024.94
114.6524.29
PC*+
32.584.96
27.083.73
32.117.10
FACT-F *+
135.9418.20
124.0024.20
138.2417.49
SF-36 Physical Dimension*+
48.494.83
45.615.82
49.752.75
SF-36 Psychological
Dimension*+
43.008.11
37.0312.58
44.337.33
Depression*+
6.837.83
12.5510.77
5.885.19
VARIABLES
*Significant differences were found between control and intervention group. + Significant differences were maintained between
baseline and follow-up assessments.
No differences taking into account confounders were observed.Variable changes were maintained in FU participants assessed
after 6 months. Correlation between QoL and LTEL (r=0.52; p=0.008) and significant weight loss (22=6.08; p=0.048) were
observed.
Conclusion. These results suggest that a specifically designed BC exercise program increases LTEL, which correlates to a better
QoL. This may reduce psychological and physical side effects of systemic treatment in patients with early BC that have recently
finished treatments, even producing lifestyle changes in BC patients that could be long lasting.

Title: National survey of chemotherapy-induced appearance issues in breast cancer patients
Takanori Watanabe1, Hiroshi Yagata2, Mitsue Saito3, Hiroko Okada4, Tomoko Takayama5, Hirohisa Imai6, Yuko Yoshida3, Nao
Tamai4, Keiko Nozawa7, Tamiko Yajima8 and Kojiro Shimozuma9. 1National Hospital Organization Sendai Medical Center, Sendai,
Japan; 2St Luke's International Hospital, Tokyo, Japan; 3Juntendo University Hospital, Tokyo, Japan; 4University of Tokyo, Tokyo,
Japan; 5National Cancer Center, Tokyo, Japan; 6National Institute of Public Health, Wako, Japan; 7National Cancer Center
Hospital, Tokyo, Japan; 8NPO Japan Clinical Research Support Unit, Tokyo, Japan and 9Ritsumeikan University, Kusatsu, Japan.
Body: Background: Many breast cancer patients suffer hair loss due to chemotherapy, and not only scalp hair loss, but also
eyebrow loss, eyelash loss and nail changes induced by chemotherapy are traumatic for patients. These side effects diminish
self-esteem and greatly reduce quality of life. However, there has been little research in this field until now. To clarify the actual
situation concerning appearance issues in breast cancer patients who received adjuvant chemotherapy, and to consider a
support system for these patients, we conducted a questionnaire survey.
Methods: Disease-free breast cancer patients who have received adjuvant chemotherapy containing anthracycline and/or taxane
within 5 years were recruited from 47 hospitals or clinics in Japan from April to October 2013. The patients participating in this
survey completed a 65-question questionnaire concerning appearance issues (48) and their perception of physical and
non-physical side effects (17). The drugs administered and treatment period were filled out by their doctors beforehand. The
completed questionnaires were mailed directly to the data center by the patients.
Results: A total of 1511 patients returned the questionnaire to the data center with a response rate of 82% (1511/1853). Since 33
patients did not meet the entry criteria, the questionnaires returned by 1478 patients were analyzed in this survey. The mean age
was 54.7 years (+-10.4, range 17-79). The distribution of the patients by time from the end of chemotherapy to this survey was as
follows: < 1 year: 28%; 1 to 2 years: 24%; 2 to 3 years: 19%; 3 to 4 years: 15%; 4 to 5 years: 13%. In this survey, the side effect
that most patients (92%) considered traumatic was hair loss. The second most traumatic side effect was fatigue (83%), while the
7th place was taken by nail changes (72%) and nausea/vomiting was in the 10th place (56%). During chemotherapy, scalp hair
loss occurred in 98% of patients. Eyebrows fell out in 90% and complete eyebrow loss occurred in 36%. Eyelashes fell out in 88%
and complete eyelash loss occurred in 37%. Fingernail changes occurred in 77% and toenail changes in 62%. In 60-70%, scalp
hair, eyebrow and eyelashes recovered to the original appearance by 1 to 1.5 years after chemotherapy, but in 3-7%, scalp and
face hair loss did not recover at all by 1 to 1.5 years. This proportion remained almost the same for 1.5 to 5 years too. During or
after chemotherapy, 84% of patients used wigs. This decreased to 47% by 1 year after chemotherapy and 15.2% by 1.5 years.
However 10% of patients were still using a wig 4 to 5 years after chemotherapy. Approximately 30% of the patients had trouble
using and selecting a wig. In 51% of the patients, sufficient information on scalp hair loss was obtained. However, sufficient
information on eyebrow loss, eyelash loss and nail changes was only obtained from 28%, 25% and 31%, respectively.
Conclusions: Our survey demonstrated the outline of hair loss and appearance issues in breast cancer patients who received
chemotherapy. Hair loss is the most distressing and occasionally long-lasting side effect. Lack of information is a serious problem.
These facts suggested a need for long-time and careful support of these patients.

Title: A combination natural product therapy attenuates common side effects associated with chemotherapy
Coy D Heldermon1, Ben G Griffith2,3, Matt Bloom1, Jennifer Owen4, Regina J Martuscello2,3, Yosef Schwartz2,3, Brent A Reynolds2,3
and Loic P Deleyrolle2,3. 1University of Florida, Gainesville, FL; 2University of Florida, Gainesville, FL; 3Preston A. Wells, Jr. Center
for Brain Tumor Therapy and McKnight Brain Institute, University of Florida College of Medicine, Gainesville, FL and 4College of
Veterinary Medicine, University of Florida, Gainesville, FL.
Body: Background: The taxanes and platinum agents provide substantial improvement in the treatment of breast and other
cancers. Neuropathy is often the dose limiting toxicity of these agents and can devastatingly affect the patient due to diminished
fine motor skills and pain in the hands and feet that diminishes the ability to exercise and interact in normal life. Clinically
significant neuropathy occurs in 40% of patients and >10% will persist past a year causing permanent effects on quality of life.
Perhaps more devastating, patients often have to discontinue effective treatment due to the development of these symptoms.
Additionally, anemia contributes to dose reductions and delays that compromise therapy and cause significant fatigue that
diminishes quality of life.
Interventions that could reduce these symptoms would provide a substantial improvement in the ability to care for these patients.
Toward this end we have developed a non-toxic dietary approach that incorporates supplementation with several natural
products. Together, this therapeutic approach, called CS.001 is able to alleviate many of the negative side effects of breast
cancer chemotherapy.
Methods: C57BL/6 mice were treated with Paclitaxel weekly or Oxaliplatin three times weekly. Animals were tested for
neuropathic pain using a cold sensitivity test [acetone test] and sensitivity to mechanical stimulus [Von Frey Test]. A CBC and
CMP were performed to assess systemic toxicity. Animals were provided with a nutritional complete diet that limited their
carbohydrates to 10% of total caloric input and were additionally supplemented with the following natural products: [1] Medium
chain triglycerides [30g/kg], [2] Curcumin [1200mg/kg], [3]EGCG [1200mg/kg] and [4]broccoli sprout powder [20g/kg]. Animals
were placed on the CS.001 diet 1 week before beginning chemotherapy.
Results: Paclitaxel & Oxaliplatin treatment resulted in a statistically significantly increase in sensitivity to cold stimulus [Acetone
test, p<0.001] that was reduced to control levels in the CS.001 treated animals. Mechanosensitivity was increased with Oxaliplatin
& reduced with Paclitaxel treatment, and addition of CS.001 resulted in a statistically significant attenuation of the chemotherapy
treatment [Paclitaxel vs Paclitaxel+CS.001, p<0.005 & Oxaliplatin vs Oxaliplatin+CS.001, p<0.05]. Additionally, reductions in
RBC, Hemoglobin and Hematocrit with Oxaliplatin treatment [p<0.001] were significantly attenuated [p<0.05] with CS.001
treatment.
Conclusion: Dietary intervention combined with a supplementation of 4 natural products was able to attenuate chemotherapy
induced anemia and neuropathy following chemotherapy treatment in mice. This combination is a promising quality of life
intervention to evaluate in cancer patients receiving taxanes or platinums.

Title: The distress screening tool: Initial experience with electronically curated patient reported measures
Shelley Hwang1, Steve Power1, Ilona Stashko1, Rachel Blitzblau1, Rachel Greenup1, Janet Horton1, Kellly Westbrook1, Kimberly
Blackwell1, Heather Sperling1, Jeffrey Peppercorn1, Gretchen Kimmick1 and Kelly Marcom1. 1Duke University, Durham, NC.
Body: Background: In June 2013, our health system transitioned to an electronic medical record (EMR) which included
collecting patient quality of life data at each clinic visit. We used the NCCN distress thermometer (DT), a short, simple to use,
self-report measure which uses a 10-point scale from 0 (no distress) to 10 (extreme distress) as well as an associated problem
checklist which queries the source(s) of their distress. Among our breast cancer clinic population, we studied the severity and
sources of distress as well as whether the DT score was associated with stage at diagnosis and time interval since diagnosis.
Methods: Between October 1, 2013 and April 30, 2014, starting 3 months after implementation of a comprehensive EMR, all
patients seen at our tertiary breast cancer clinic were asked to complete the DT survey at each clinic visit. DT data were collected
and entered into the EMR at point of care. The DT tool was correlated with demographic and tumor information from our
prospectively curated electronic datamart.
Results: We collected 7276 DT surveys from 3267 unique patients over seven months. Median age of the cohort was 60 years;
73% were white and 21% were black. Among those with available staging data and a diagnosis of breast cancer, stage
distribution was 10% stage 0, 34% stage I, 37% stage II, 15% stage III and 4% stage IV. The median reported distress score was
1.0 (range 0-10) with score distribution shown in Figure 1. The most commonly reported source of stress was fatigue (8.0%)
followed by pain (6.8%). For new patient appointments the most commonly reported sources were worry (9.5%) followed by
nervousness (8.0%). There was no significant correlation between overall distress score and stage at diagnosis. Among patients
who were seen more than once during the study interval, the DT score changed for 33.7% of patients. The lowest distress scores
were reported among women >3 years from initial diagnosis.
Conclusions: The transition to an integrated EMR system has allowed collection of analyzable patient reported data to inform
medical and psychosocial intervention. Structured data collection at point of care allows for efficient identification of and
management for the major sources of distress among patients during breast cancer treatment and survivorship.

Title: Clinico-biological characteristics of patients surviving more than two years with metastatic breast cancer (MBC): Results of
a transversal national multicentric survey
Delphine Loirat1, Camille Tlemsani2, Jennifer Arrondeau2, Audrey Bellesoeur2, Christophe Le Tourneau1 and Benoit Rousseau3.
1
Institut Curie, Paris, France and 2AERIO, Paris, France.
Body: Introduction:
MBC cancer patients may have prolonged survival, and MBC cancers can be considered as a chronic disease. The main goal of
the present study was to describe the clinico-biological features of patients surviving more than two years with MBC.
Method: During 4 months, we conducted a national multicentric survey about patients aged 18 in metastatic setting (all solid
tumors) for more than 24 months. Clinico-biological data from 200 patients were collected in 39 French centers. Preliminary
results of MBC patients (N=88, 87 women/one man) are presented.
Results: Most of them were ductal carcinoma (88%), expressing hormonal receptor HR (77%). 43% of tumors overexpressed
HER2 (HER2+ tumors: 43%; Triple negative tumors: 6%). Median age at MBC diagnosis was 53 years [29-85]. 18% had
metastatic disease at diagnosis and 82 % were localized with a disease-free survival of 65 months [4-312]. Median time of MBC
disease was 4,5 years [2-20]. At the time of MBC diagnosis, 64% of patients were not single; 55% were working while 30% were
retired. At data collection, 89% of non-single patients were not separated, and 43 % of working patients at diagnosis were still
working.
Mean number of treatment lines in advanced disease was 4.7 [1-13]. 89% of MBC patients received at least one chemotherapy,
68% hormonotherapy, 70% targeted therapy and 38% had been included in at least one clinical trial. 97% of patients had a local
treatment of their primary tumor. Concerning metastasis, 23% had a surgical treatment and 40% radiotherapy treatment. 80% of
patients remain with PS of 0 or 1. Only 9% of patients were followed by a palliative care team, 24% by a psychologist and 23% by
a nutritionist.
Conclusion: Our preliminary results of suggest that an important proportion of MBC cancer patients who live more than 2 years
are young, have been treated with chemotherapy, hormonal and targeted therapy, have participated to clinical trials and still have
good performance status. No change in marital status was observed. Half of working patients at MBC diagnosis continue to work.
Few of them received palliative care. This study may help to better describe long-term survivors with MBC, and socio-medical
burden as cancer became a chronic disease.

Title: Restoration by compression therapy of skin blood perfusion levels decreased during breast cancer chemotherapy,
alleviating peripheral neuropathy
Tsuyoshi Ohno1, Takashi Mine2, Hiroki Yoshioka2, Mikiko Kosaka2, Kazuhiro Matsuda2, Maiko de Kerckhove2 and Charles de
Kerckhove2. 1Nagasaki Prefecture Shimabara Hospital, Shimabara, Nagasaki, Japan and 2Nagasaki Harbor Medical Center City
Hospital, Nagasaki, Japan.
Body: <Background> Nanoparticle albumin-bound paclitaxel (nab-PTX) has become a key drug used in chemotherapy for breast
cancer, but it often causes adverse effects such as peripheral neuropathy (PN). No effective prophylactic management has so far
been established. We have applied a "3S" approach to prevent and treat PN based on two concepts: compression therapy using
stockings and sleeves, and medication therapy using selected prophylactic medications. We previously reported better CTCAE
v4.0 PN grades and notably superior nab-PTX dose maintenance in a 3S group compared to a control group. However, little is
known about the effects of compression therapy on a patients level of skin blood perfusion; it is also unknown whether any such
effects might vary by 1) PN grade or 2) the number of nab-PTX cycles. <Patients and Methods> To establish whether a
compression therapy-skin perfusion relationship exists, the skin perfusion of the lower limbs was measured before and after
stocking use in a 3S prophylactic treatment group for nab-PTX therapy (n=44), and in a control group of healthy volunteers
(n=50). The skin perfusion was measured using a laser Doppler blood flow meter with an integrated probe (NL-101 Nahri Nexis
Japan). To find how compression therapy affects skin perfusion by PN grade, the 3S group was subdivided into three PN grade
subgroups (n = 12 for Grade 0, n = 20 for Grade 1, and n = 12 for Grades 2 and 3 combined). To find how compression therapy
affects skin perfusion by number of nab-PTX treatment cycles undergone by patients, the 3S group was subdivided into the
following three subgroups: 1 to 5 cycles of nab-PTX treatment (first period group; 1P, n=18), 6 to 10 cycles (second period group;
2P, n=16), and more than 10 cycles (third period group; 3P, n=12). <Results> In the control group of healthy volunteers, stocking
use tended to increase the median skin perfusion level (mL/min/100g) from 10.9 3.8 to 11.8 4.3 (p=0.06). Interestingly, the
median skin perfusion level for the 3S group as a whole significantly increased from 8.7 3.3 before stocking use to 11.3 3.8
after stocking use (p< 0.001). When examining the effects of stocking use on skin perfusion by PN grade, the increase in median
skin perfusion level was significant in the Grade 0 subgroup (8.2 2.7 to 12.4 4.5, p=0.005), in the Grade 1 subgroup (8.3 3.8
to 10.4 4.0, p=0.0499), and in the Grades 2 and 3 subgroup (7.9 2.3 to 10.6 2.4, p=0.005). When examining the effects of
stocking use on skin perfusion by number of treatment cycles, the increase in median skin perfusion level was significant in the
1P subgroup (7.9 3.0 to 11.4 4.8, p=0.007) and in the 2P (7.9 2.9 to 10.8 2.9, p=0.005), but not in the 3P (9.3 3.5 to
10.8 3.2, p=0.15). <Conclusion> This study demonstrated that the skin perfusion of the lower limbs is decreased following
nab-PTX chemotherapy, and also that the skin perfusion is improved by compression therapy. Compression therapy appeared to
successfully restore the skin perfusion levels across every grade of chemotherapy-induced PN. Therefore, our 3S approach is
suitable for alleviating CIPN by proactively maintaining skin perfusion from the beginning of nab-PTX therapy.

Title: The effect of introducing an enhanced recovery programme on improving compliance with arm exercises post breast cancer
surgery and reducing morbidity
Alistair RM Macey1, Jayne I McGivern1, Juliette Murray1 and Alison K Lannigan1. 1Wishaw General Hospital, Wishaw, Lanarkshire,
United Kingdom.
Body: Introduction
The introduction of an enhanced recovery programme for breast cancer surgery in our institution has provided an opportunity to
redesign the service and the way in which patients receive pre-operative information. We have encouraged pre-operative
consultations with physiotherapists to demonstrate post-operative arm exercises to see if this can improve compliance with
exercises and reduce morbidity.
Aim
To audit patient reported upper limb symptoms after breast cancer surgery both before and after the introduction of an enhanced
recovery programme with more detailed patient information and input from physiotherapy.
Methods
An enhanced recovery programme was first introduced in our institution in Sep 2012. Between December 2012 and December
2013, 80 patients having breast cancer surgery in Wishaw were sent a detailed questionnaire asking them about all aspects of
their treatment. Two patients were excluded from the questionnaire one because of learning difficulties and the other because of
acute psychiatric illness. There was an 89% response rate with 71 responses having been received to date. A previous similar
questionnaire was sent to patients at 2 and 4 weeks post surgery between November 2011 and April 2012 before the introduction
of the enhanced recovery programme. On this occasion there was a 73% response rate with 63 responses having been received
from 86 sent. The two groups were compared.
94% of patients reported having received advice about physiotherapy and those who had a session with a physiotherapist
increased from 16 to 39% with the introduction of the enhanced recovery programme. 80% of patients received a photographic
exercise leaflet in addition to verbal and DVD information on arm exercises.
Postoperatively in the enhanced recovery group, 39% of patients reported some numbness in their ipsilateral upper arm (57% in
the group who had undergone axillary node clearance compared with 28% in the sentinel node group). 11% of patients reported
persistent ongoing pain after surgery and 9% felt that discomfort affected their daily activities. This contrasts with the previous
cohort where 43% of patients reported reduced arm function and almost all of these patients (93%) felt that this impacted on their
daily activities. As expected performance across a range of arm and shoulder functions was worse in for axillary node clearance
than sentinel node biopsy for both groups.
ConclusionIntroducing an enhanced recovery programme changed the way in which information was provided to patients about
post-operative arm exercises and reduced self reported arm morbidity.

Title: Survival duration and quality of life expectations in patients with metastatic breast cancer: The role of treating oncologists in
influencing patients expectations of therapy A questionnaire-based study
Anis Toumeh1, Ria Kundu1 and Iman Mohamed1. 1University of Toledo, Toledo, OH.
Body: Introduction
Breast cancer is the most common female cancer in the United States. Median survival for patients with metastatic disease is 18
to 24 months with some patients enjoying long term survival1,2. Multiple studies have documented that cancer patients tend to
overestimate survival duration and that their understanding of their prognosis is insufficient. This is due in part to vague doctorpatient communication, Lack of information concerning the effect of alternative therapies on overall outcomes and Lack of patient
understanding of the likely outcomes of their disease3,4.
Objectives:
To investigate the expectations of patients with metastatic breast cancer in regards to survival duration and quality of life from
different treatments, and identify oncologists role in influencing these expectations.
Methods:
Electronic charts were used to select women diagnosed with metastatic breast cancer between the age of 35 and 80 treated at
the cancer center at the University of Toledo medical center. A detailed questionnaire evaluated the level of comfort between
patients and oncologist/s regarding survival duration and quality of life. The questionnaire also evaluated patients expectations
from treatments they received, are receiving.
Results:
40 patients were identified. 26 out of which completed the survey. The majority of patients ( 65-77 % ) expected different
treatments for their metastatic disease to prolong their survival for more than 5 years. 57 to 66% of our patients stated that their
expectations regarding QOL outcomes from different treatments changed after discussion with their oncologist. 7% of our patients
stated that their oncologist does not spend enough time explaining what to expect from the treatment in regards to quality of life
outcomes.
Conclusion:
Our study highlights the importance of communication between oncologists and metastatic breast cancer patients and the
influential role they play in patients expectations regarding survival and quality of life. It is also important to recognize patients
concerns and spend enough time explaining expected outcomes from different treatments. Although this could be challenging in
busy practices, short interval follow up visits and re-addressing those concerns along with involving the palliative care team early
could be a potential steps for improvement.
References:
1. Greenberg PA. J Clin Oncol 14 (8): 2197-205, 1996.
2. Jane C et al. JAMA.1998;279:1709-1714.
3. Siminoff LA et al. J Clin Oncol. 1989;7:1192-1200.
4. Bernheim JL et al. Monogr Ser Eur Organ Res Treatment Cancer. 1987;17:285- 295.

Title: Breast cancer survivor advocacy at the University of Wisconsin Breast Center
Jennifer A Mirrielees1, Lee G Wilke1, Kayla R Breckheimer1, Teresa A White1, Deborah A DeNure1, Michelle M Schroeder1 and
Amye J Tevaarwerk1. 1University of Wisconsin, Madison, WI.
Body: Peer-to-peer support programs provide unique psychosocial and educational support for breast cancer patients. With the
support of funding from the South Central Wisconsin Affiliate of Susan G. Komen for the Cure, we developed a Patient-Survivor
Advocacy (PSA) program to complement University of Wisconsin Breast Center (UWBC) patient navigator to facilitate
peer-to-peer support between those who have completed primary breast cancer treatment and those newly diagnosed patients.
We evaluated the feasibility and utility of this peer-to-peer support program for both patients and advocates.
Methods:
We recruited advocates from the pool of women previously treated for breast cancer at UWBC. PSAs completed 3 training
sessions over a 5 month period, including a volunteer orientation and the After Breast Cancer Diagnosis (ABCD) program.
Training topics included a patient survivor advocacy orientation, guidance for communication with patient matches, and
documentation and reporting of patient contact to the UWBC patient navigator. Following training, PSAs were matched to patients
based on age, type of breast cancer and other life factors identified as important by each referee. PSAs contacted matches via
phone and documented communication frequency and content. PSAs and patients then completed surveys describing their
experience and satisfaction with the PSA program. Survey questions were tailored to the program strengths, deficits, and areas
for improvement. PSAs were also surveyed regarding curriculum and the training process.
Results:
Between 11/2012 and 4/2014, 14 PSAs were recruited and trained, and 40 patients were referred to the program. Half of patient
referrals (20) were from UWBC physicians, while 18 were from UWBC nursing staff, in addition to 9 self-referrals (this includes
overlapping referrals from multiple providers). Six decided not to participate, and one was transferred to hospice before being
matched, while 8 patients were referred to the ABCD program due to lack of suitable UWBC PSA matches. Twenty six patients
were successfully matched to active PSAs. The number of patients matched to each PSA ranged from 1 to 8; an average of 3
phone and 4 email contacts were made by each PSA-patient dyad. Communication topics included the effect of diagnosis and
treatment on mood and mental health, physical health, daily life, and interpersonal relationships, as well as additional community
resources for support. Patients and PSAs expressed positive impressions of the program: feeling supported by UWBC staff,
experiencing a sense of accomplishment (PSAs) or direct benefit (patients). Responses suggested new topics for PSA training
and cancer education, including surgical interventions, radiology, chemotherapy, integrative medicine, medical updates, and new
topics in social support.
Conclusions:
The first year of the UWBC PSA program saw a successful peer-to-peer psychosocial support infrastructure for newly diagnosed
breast cancer patients. PSA and patient survey responses provided useful feedback critical to development of the program.
Future goals for this program include increased patient utilization, provision of additional education topics and materials during
PSA training, and documentation of the impact of PSA program support on the UWBC nursing burden.

Title: National survey of long-term recovery from chemotherapy-induced hair loss in patients with breast cancer
Hiroshi Yagata1, Takanori Watanabe2, Hiroko Okada3, Mitsue Saito4, Tomoko Takayama5, Hirohisa Imai6, Yuko Yoshida4, Nao
Tamai3, Keiko Nozawa7, Tamiko Yajima8 and Kojiro Shimozuma9. 1St Luke's International Hospital, Tokyo, Japan; 2National
Hospital Organization Sendai Medical Center, Sendai, Japan; 3University of Tokyo, Tokyo, Japan; 4Juntendo University Hospital,
Tokyo, Japan; 5National Cancer Center, Tokyo, Japan; 6National Institute of Public Health, Wako, Japan; 7National Cancer Center
Hospital, Tokyo, Japan; 8NPO Japan Clinical Research Support Unit, Tokyo, Japan and 9Ritsumeikan University, Kusatsu, Japan.
Body: Background: Altered appearance due to chemotherapy is a very distressing adverse event and can remain unrecovered
for a long time after chemotherapy. To clarify the current status of appearance change and its support systems, we conducted a
national questionnaire survey of patients with breast cancer who had received chemotherapy in Japan. Here, we report on the
long-term recovery of scalp hair loss during and after chemotherapy.
Patients and methods: A questionnaire was distributed to patients in hospitals throughout Japan between April and October 2013.
The questionnaire was regarding the current status of the patients appearance issues (scalp hair, eyebrows, eyelashes, nails,
skin) related to chemotherapy and its support systems, including chemotherapy regimens received, endocrine therapy received,
and duration after chemotherapy. Eligible patients were women with breast cancer without any recurrence who had received
adjuvant or neoadjuvant chemotherapy containing anthracycline (A) and/or taxanes (paclitaxel, P; docetaxel, D) and who were
within 5 years from the last chemotherapy treatment. The physicians of each hospital asked their patients to fill out the
questionnaire and mail it directly to the data center. The scalp hair status was analyzed in a cross-sectional manner according to
the duration from chemotherapy.
Results: The questionnaires were returned from 1511 patients in 47 hospitals (response rate, 82%; 1511/1853). Thirty-three
patients were excluded, mainly because >5 years had passed since chemotherapy. In total, 1478 questionnaires were ultimately
analyzed. The median age was 50 (range, 1779) years. The distribution of patients according to time from the last chemotherapy
treatment was as follows: <1 year, 28%; 12 years, 24%; 23 years, 19%; 34 years, 15%; and 45 years, 13%. During
chemotherapy, scalp hair loss occurred in 98.4% of the patients, and 94% experienced >80% hair loss. Hair growth began during
chemotherapy in 13.1% of patients and after chemotherapy in 80.3% (6.6% left the question unanswered). Within 6 months from
the start of hair growth, 65% of patients felt a change in hair thickness, while 82% felt it was becoming thin. Of the patients, 70%
felt a change in quality, while 48% felt that it had become unruly; 44% felt a color change, while 80% felt that they were growing
more gray hair. Of the patients who answered the questions, >80% hair volume recovery was seen in 52.7% of patients within 1
year; in 63.5%, in 13 years; and in 61.7%, even after 3 years. After 3 years, volume recovery was seen in 67.8% of patients after
an A+Pcontaining regimen; in 43.4%, after A+D; in 63.5%, after D; and in 88.9%, after A. Patients who had received A+P, D,
and A+D had significantly less volume recovery than patients who had received A (P<0.001 for all).
Conclusions: Almost all patients with breast cancer experienced severe hair loss during standard chemotherapy, but a recovery
trend was noted after chemotherapy. However, hair remained unrecovered to various degrees in a significant number of patients
even 35 years after chemotherapy, especially in those who had received taxane-containing regimens. We should consider the
support needs of patients who experience chemotherapy-induced hair loss.

Title: Hemoglobin levels and quality of life in patients with breast cancer and symptomatic chemotherapy-induced anemia
enrolled in the eAQUA study
Mario Airoldi1, Dominique Spaeth2, Joan Van den Bosch3, Charalambos Christofyllakis4, Laura Belton5, Chet Bohac6, Jan-Henrik
Terwey7 and Giuseppe Tonini8. 1Azienda Ospedaliera Universitaria S. Giovanni Battista Le Molinette, Torino, Italy; 2Centre
d'Oncologie de Gentilly, Nancy, France; 3Albert Schweitzer Hospital Location Dordwijk, Dordrecht, Netherlands; 4401 Military
Hospital, Athens, Greece; 5LB Biostatistics, London, United Kingdom; 6Amgen Inc, Thousand Oaks, CA; 7Amgen (Europe) GmbH,
Zug, Switzerland and 8Universit Campus Bio-Medico, Roma, Italy.
Body: Background: Fatigue associated with chemotherapy-induced anemia (CIA) is common in patients with breast cancer, and
can have adverse effects on quality of life (QoL). Erythropoiesis-stimulating agents (ESAs) reduce the need for transfusions and
may improve QoL in patients with symptomatic CIA. Information on hemoglobin (Hb) levels and effects of fatigue on QoL in
patients with breast cancer and CIA in real-world clinical practice is limited.
Methods: The Electronic Assessment of Quality of Life in Patients With Symptomatic CIA (eAQUA) study evaluated
improvements in QoL for patients with CIA receiving ESAs who had an increase in Hb of 1 g/dL by week 9. This phase 4,
international, longitudinal, prospective, observational study enrolled patients with solid tumors who received chemotherapy and
had symptomatic anemia. Patients received ESA therapy for up to 13 weeks based on European indication. The primary outcome
was the proportion of patients with increase in Hb 1 g/dL and improvement in fatigue-related QoL based on the Functional
Assessment of Cancer Therapy-Fatigue (FACT-F; scale = 0 to 52 with lower scores indicating worse fatigue) subscale scores and
fatigue Visual Analog Scale (VAS; scale = 0 to 100 with higher scores indicating worse fatigue) from baseline to week 9. FACT-F
change scores were anchored to VAS change scores to determine the minimally important difference (MID) for improvements in
QoL. Patients with a FACT-F change score that was the MID were considered to have an improvement in QoL. For Hb and QoL
outcomes, week 9 data were those assessed closest to on-treatment day 57 (after initiation of ESA) and within on-treatment days
43 to 70 inclusive, to account for different ESA dosing schedules and the observational nature of the study. Secondary outcomes
included rates of red blood cell (RBC) transfusions or iron supplementation during the study.
Results: Of 1262 patients enrolled in eAQUA, 289 had breast cancer and were included in the full analysis set (FAS; had at least
one ESA dose); of these, 152 patients were eligible to be included in the primary analysis set (PAS; had QoL and Hb data
available at baseline and week 9). At baseline, mean (standard deviation [SD]) Hb was 9.4 (0.6) g/dL; mean (SD) FACT-F and
VAS scores were 27.1 (10.5) and 52.7 (22.8), respectively, in the FAS. Mean (SD) change from baseline at week 9 was 1.3 (1.3)
g/dL for Hb; 4.1 (10.8) score change for FACT-F; and 4.7 (25.9) score change for VAS in the FAS. A total of 54 (18.7%) patients
in the FAS required an RBC transfusion and 79 (27.3%) received iron supplementation. At week 9, 77 of 152 patients in the PAS
had achieved improvement in fatigue-related QoL (50.7%; 95% confidence interval [CI] = 42.7%, 58.6%); 93 patients had
increased Hb 1 g/dL (61.2%; 95% CI = 53.4%, 68.9%); and 59 patients (38.8%; 95% CI = 31.1%, 46.6%) had achieved both
improvement in fatigue-related QoL and increased Hb 1 g/dL.
Conclusions: In this exploratory subgroup analysis, patients with breast cancer and symptomatic CIA treated with ESAs achieved
clinically meaningful improvements in fatigue-related QoL and Hb levels.

Title: Prophylaxis of chemotherapy-induced febrile neutropenia with biosimilar filgrastim: Description of patients, treatment
patterns and outcomes in the MONITOR-GCSF study in the breast cancer cohort
Pere Gascn1, Matti Aapro2, Heinz Ludwig3, Mario Boccadoro4, Carsten Bokemeyer5, Matthew Turner6, Michael Muenzberg6, Ivo
Abraham7,8, Kris Denhaerynck7 and Karen MacDonald7. 1University of Barcelona, Barcelona, Spain; 2Clinique de Genolier,
Genolier, Switzerland; 3Wilhelminenspital, Wien, Austria; 4Azienda Ospedaliero Universitaria S. Giovanni Battista di Torino,
Torino, Italy; 5Universitaetsklinikum Hamburg Eppendorf, Hamburg, Germany; 6Sandoz Biopharmaceuticals, Holzkirchen,
Germany; 7Matrix45, Tucson, AZ and 8University of Arizona, Tucson, AZ.
Body: Introduction: MONITOR-GCSF is a European prospective observational study of practice patterns and outcomes of
patients treated with Sandoz filgrastim (EP-2006) in the prophylaxis of chemotherapy-induced febrile neutropenia (CIN/FN).
Objectives: To describe patients, treatment patterns of EP-2006, and outcomes in the breast cancer cohort of the
MONITOR-GCSF study.
Methods: Prospective observational study following 466 evaluable patients from 23 centers in Europe for up to 6 cycles within a
single chemotherapy line including a total of 2714 cycles.
Results: Median age was 56y (range 25-91); all but 3 patients were female. Table 1 presents chemotoxicity in terms of % FN risk
and prophylaxis type. GCSF was correctly initiated as either primary or secondary prophylaxis per EORTC guideline
recommendations (considering chemotherapy-related FN risk and patient-related factors) in 62% of patients. Eleven percent were
undertreated, i.e., secondary prophylaxis when primary was indicatedeither when CIN/FN risk >20% or when CIN/FN risk was
10-20% in combination with patient-related risk factors. Twenty-seven percent were overtreated, i.e., primary or secondary when
not indicatedeither primary prophylaxis in <10% risk of CIN/FN or in 10-20% risk of CIN/FN in the absence of risk factors, or
secondary prophylaxis in <10% risk of CIN/FN in absence of prior CIN/FN. EP-2006 was started on average 2.82.6 days after
chemotherapy was initiated and given for 4.92.1 days. Dosing was 45% at 30MIU/day, 55% at 48MIU/day. CIN (any grade)
occurred in 12.0% of all cycles and 32.8% of patients had one or more episodes of CIN. 19.5% of patients had at least one
episode of Grade 3 or 4 CIN of which 6.2% were febrile. CIN/FN-related hospitalizations were experienced by 5.2% of patients.
CIN/FN-related chemotherapy disturbances (dose reduction, delay or cancellation) occurred in 9.4%.
Conclusions: Variation in treatment with biosimilar GCSF in breast cancer patients is evident in terms of decision to treat with
primary prophylaxis relative to guideline recommendations as well as day of initiation, duration and dose, yet incidence of CIN/FN
and related events is low. Forthcoming analyses will determine whether variability in treatment is associated with differential
outcomes.
Table 1
Chemotherapy toxicity (risk of FN)
Phrophylaxis
Prophylaxis decision relative to guidelines
>20%
250
53.8
10-20%
170
36.5
<10%
45
9.7
Primary
372
79.8
Secondary
94
20.2
Undertreated
50
10.8
Correct
289
62.1
Overtreated
126
27.1

Title: What we know and what we must do: A metastatic breast cancer alliance quality of life landscape analysis
Musa Mayer1, Katherine Crawford-Gray2, Shirley Mertz3, Ginny Knackmuhs3 and Marc Hurlbert2. 1AdvancedBC.org, New York,
NY; 2Avon Foundation for Women, New York, NY and 3Metastatic Breast Cancer Network, New York, NY.
Body: The Metastatic Breast Cancer Alliance (MBCA)[1] is comprised of non-profit advocacy, funding and industry organizations
and individuals who seek to transform and improve the lives of women and men living with metastatic breast cancer. Stage IV or
metastatic breast cancer is different from early breast cancer. The disease is not curable. People with MBC are always in
treatment, switching treatment regimens as their disease progresses. Metastasis is the cause of virtually all breast cancer deaths,
and nearly 40,000 die annually of MBC. Because their time is limited, patients daily experience a host of psychosocial and quality
of life issues.
Objective: To review prior literature and patient survey reports related to quality of life needs for patients with MBC, interview key
experts in the field, and to assess the extent to which non-profit organizations, clinical providers and others are meeting those
needs.
Methods: We conducted (1) a literature review of >140 recent articles and studies in psychosocial research as well as survey
findings from over 6,000 patients living with MBC; (2) a desk research analysis of MBCA members efforts in patient advocacy,
research, policy, education and support, and public awareness; analysis of websites (n=24) and print materials (n=27); (3)
interviews with MBC Alliance members about their information and services for MBC patients (n=16); and (4) an online survey of
hospital-based patient navigation programs (n=31) and telephone helplines (n=8) provided by breast cancer and all-cancer
organizations.
Results: We found inconsistent and incomplete development of patient education materials about metastatic disease and
treatment options. Health care teams are often not taking time to educate patients on treatment options, or to routinely assess
and treat their side effects and symptoms. Resources, staffing and time are not available to meet patients quality of life and
psychosocial needs. Anxiety and depression in MBC patients remain untreated in many cases. Palliative care is misunderstood
and still associated with end of life care by both doctors and patients. Yet, quality of life for MBC patients can be improved with
increased access to palliative care earlier in the diagnosis of the disease.
Conclusion: Psychosocial research and patient surveys identify the information and service needs that would improve the quality
of life for MBC patients; however, our study finds that those needs are rarely met. Advocacy groups and health care providers
need to act strategically to put into place programs and support services that address the psychosocial and quality of life needs of
patients living with MBC.
[1] https://www.mbcalliance.org/

Title: SurvivorLink: Evolution of 1:1 peer support to connect young women with breast cancer
Jean Rowe1, Michelle Esser1, Megan McCann1 and Stacy Lewis1. 1Young Survival Coalition, New York, NY.
Body: Background
There are 250,000 breast cancer survivors living in the United States today who were diagnosed under the age of 40. Compared
to older women, young women (YW) generally face more aggressive cancers, lower survival rates, an increased risk of metastatic
recurrence, and a higher rate of anxiety and depression.
YW also face unique issues as a result of their diagnosis. They are more likely to be single and dating, starting a career, raising
young children, or possibly starting a family. Cancer treatments may impact fertility, cause premature menopause, and sexual
dysfunction. Due to these and other concerns, YW diagnosed with breast cancer strongly desire to connect with other young
survivors.
Young Survival Coalition (YSC) is the premier global organization dedicated to the critical issues unique to YW and breast cancer.
It is YSCs goal to ensure that no young woman goes through breast cancer alone.
Methods
In order to connect YW diagnosed with breast cancer with others similarly situated, YSC initiated its SurvivorLink program (SL) in
2006 then known as Point of Contact (POC). YW seeking to connect with another young breast cancer survivor called or emailed
a YSC staff person, who connected her to a trained POC. POCs received training through 2-day in-person sessions held
periodically throughout the country. In 8 years, there were 89 volunteers trained in 7 in-person training sessions. Between 2008
and 2010, 776 YW requested connections through SL. In 2011, usage of SL dropped significantly.
In 2013, YSC re-examined the SL program to determine whether it was a needed resource. If so, YSC sought to determine how
to revitalize and expand it.
Results
During the analysis, a few key issues arose. Despite the stated desire of YW to connect with others, SL was not well-utilized from
2011 on. Further research determined that many young survivors were not aware of SL, even if they knew of YSC. The in-person
trainings were expensive and not all interested volunteers were located in the same geographic area. This limited the ability to
maintain a pool of fresh volunteers with diverse experiences. Data on the program had not been maintained in one central
location. It was difficult to discern who was matched through SL, when, and the result of their interaction.
YSC believes that the program is important to continue. While other organizations offer call-in support or peer matching, none
focus solely on YW with breast cancer nor have a diversity of YW with different diagnoses and experiences who could serve as
potential matches.
Taking these issues into consideration, YSC took steps to bring renewed energy to this program. First, an online training portal for
SL volunteers was developed. Through eight on-line modules, YSC can train more volunteers in a shorter period of time for less
cost. Individuals can complete the training in their own time from the comfort of their own homes. In order to make YW aware of
the program, SL is now regularly advertised on Facebook, the YSC homepage, and elsewhere. Finally, a central database was
initiated to capture information on our trained volunteers, those who call-in, the matches made and the results of their interaction.
This allows YSC to track the use and success of the program.

Title: Evaluation of quality of life in patients with advanced and metastatic breast cancer proposed for palliative chemotherapy
and best supportive care versus best supportive care
Anda-Natalia T Ciuhu1, Gabriela I Rahnea Nita1, Mihaela T Popescu2 and Roxana Andreea D Rahnea Nita3. 1Saint Luke Hospital,
Bucharest, Romania; 2Colentina Clinical Hospital, Bucharest, Romania and 3Romanian Society of Palliatology and Thanatology,
Bucharest, Romania.
Body: Background:
In Romania many patients neglect their symptoms and present to the doctor in advanced stages of cancer. In the case of patients
with breast cancer who present with very advanced locoregional disease (ulcerated or hemorrhagic tumors) and symptomatic
metastatic disease (visceral or skeletal pain, dyspnea), the choice between chemotherapy/hormonal therapy and best supportive
care and only best supportive care became very difficult.
Objective:
Evaluation of quality of life in patients with advanced and metastatic breast cancer proposed for palliative chemotherapy and best
supportive care versus best supportive care
In the last 6 month of the year 2013, 57 patients with advanced and metastatic breast cancer were randomized to either
chemotherapy/hormonal therapy in addiction to best supportive care or to best supportive care. The patients were naive to
chemotherapy, with performance status ECOG between 2 and 3 and had symptomatic disease. The randomization of patients
was made according to their choice. The performance status and the symptoms were evaluated at the baseline and, later, at each
admission in the Oncology Palliative Care Department.
Results:
More patients in the chemotherapy group (92.68%, 38/41) had an improve quality of life compared to those in the best supportive
care group (56.25 %, 9/16). The overall survival was longer in the chemotherapy group.
Conclusions:
The results show that chemotherapy/hormonal therapy improve the quality of life and overall survival in very advanced
locoregional disease and symptomatic metastatic breast cancer. The number of patients who benefit from chemotherapy and best
supportive care is higher versus patients who benefit only from best supportive care.
Key words: chemotherapy/hormonal therapy, best supportive care, breast cancer.

Title: End-of-life outcomes and hospice and palliative care utilization in hospitalized patients with metastatic breast cancer
Amanda Parkes1, Jennifer A Shin1, Helen Knight1, Areej El-Jawahri1, Lara Traeger1 and Jennifer S Temel1. 1Massachusetts
Body: Background:
Hospitalizations in patients with metastatic cancer occur commonly at the end of life but have not been well-described in those
with metastatic breast cancer (MBC). The goal of this study was to describe the reasons for admission, end-of-life outcomes, and
hospice and palliative care utilization in hospitalized patients with MBC.
Methods:
We identified all patients with MBC who had their first hospital admission (index admission) at Massachusetts General Hospital
since their diagnosis of metastatic disease between 1/1/2009 and 12/31/2010 through a centralized clinical data registry. We
collected demographic and clinical information and data on all hospital admissions and utilization of palliative care and hospice
services during a three-year follow-up period.
Results:
We identified 123 patients hospitalized for the first time since their diagnosis of MBC. The median number of hospital admissions
during the three-year follow-up period was 2 (range 1 to 17). Uncontrolled symptoms accounted for half (62/123, 50%) of the
index admissions [20 (16%) with central nervous system systems, 17 (14%) with respiratory symptoms, 10 (8%) with pain, 9 (7%)
with gastrointestinal symptoms, 4 (3%) with failure to thrive, and 2 (2%) with dehydration]. The majority of patients (93/123, 76%)
died during the follow-up period. The median time from first hospitalization to death was 6 months (range 0 to 53). Among
patients who died, 20/93 (22%) died in the hospital and 49/93 (53%) died within 14 days of hospital discharge. Among the
deceased group, the inpatient palliative care team evaluated 53/93 (57%) of patients at least once during an admission, but only
18/93 (19%) patients attended an outpatient palliative care clinic appointment. A minority of patients who died (27/93, 29%) were
referred to hospice upon discharge. An additional 34/93 (37%) patients who died were referred to hospice from the outpatient
setting.
Conclusions:
Hospitalized patients with MBC are commonly admitted for uncontrolled symptoms. They have a poor prognosis, and
approximately half die within two weeks of a hospital admission. However, only a minority receive outpatient palliative care or are
referred to hospice services from the inpatient setting. These findings highlight the need to develop interventions to improve
end-of-life care for patients with MBC who are hospitalized.

Title: Prognostic factors predicting overal survival after radiotherapy for metastatic spinal cord compression in breast cancer
patients treated in the Brazilian public health system
Tomas Reinert1, Ricardo Zylberberg1, Frederico MT Lima1, Christiane S Pinto1 and Alexandre Boukai1. 1Instituto Nacional do
Cncer (INCA), Rio de Janeiro, RJ, Brazil.
Body: Objective: To evaluate potential prognostic factors for survival after radiotherapy of metastatic spinal cord compression
(MSCC) on breast cancer patients treated at the Brazilian public health system.
Methods and materials: We evaluated 51 consecutive patients treated in a single cancer center between May 2011 and May
2012. The following potential prognostic factors were investigated retrospectively: age, performance status, interval between
tumor diagnosis and MSCC, presence of visceral metastases, interval between the development of symptoms and radiotherapy,
site of compression and Tokushashi index (prognostic criteria index that divides patients into 3 groups with different life
expectancy according to their total number of scoring points).
Tokuhashi Score Parameters
General condition (ECOG PS)
No. extraspinal bone metastasis foci
No. metastasis in the vertebral bodies
Metastases to the major internal organs
Primary site of cancer
Spinal cord palsy
Adapted from: Spine 1990; 15: 1110-3.
Results: Patients characteristics are shown in Table 2.
Patients characteristics
Age: median (range)
54y (28-79)
Interval between BC diagnosis and MSCC: median (range)
32mo (0-131).
Days of hospitalization: median (range)
10 (2-67)
Visceral metastases (yes/no)
49%/51%
Site of compression (cervical/thoracic/lumbar)
24%/55%/49%
Spinal cord palsy (none/incomplete/complete)
31%/47%/21%
Tokuhashi score risk (low/intermediate/high)
30%/35%/33%
Inpatient mortality was 18%. Among the 16 patients who had non-ambulatory status at admission only 2 (12%) were able to walk
after treatment. Overall survival was 13,6 months. Overall survival according to the Tokuhashi index was 1,5mo high risk, 13,6mo
intermediate risk and 23,3mo low risk.
The following prognostic factors for inferior OS were identified on univariate analysis: high risk Tokuhashi score - HR 3,7 (p
0,0001); PS4 - HR 3,2 (p 0,0004); Presence of visceral metastasis - HR 4,0 (p 0,0001); interval between symptoms and
radiotherapy >14d - HR 2,5 (p 0,01). On multivariate analysis, a high risk Tokuhashi score was statistically associated with inferior
survival.
Conclusion: Metastatic spinal cord compression is an oncological emergency associated with significant morbidity and mortality.
Life expectancy is a key factor for therapeutic planning. In the limited resource setting of the Brazilian public health system, the
validation of prognostic factors is essential to guide the clinician on referring the patient to a tertiary cancer care center or to
provide palliative care avoiding burden for debilitated patients resulting from multiple daily trips to the radiation oncology
department and painful positioning on the treatment couch .This study validated the Tokuhashi index as a useful prognostic tool in
this population.

Title: The impact of skeletal-related events on pain interference in patients with advanced breast cancer and bone metastases
Lesley Fallowfield1, Donald L Patrick2, Roger Von Moos3, Charles S Cleeland4, Ying Zhou5, Arun Balakumaran5 and Yi Qian5.
1
University of Sussex, Sussex Health Outcomes Research and Education in Cancer (SHORE-C), Brighton, United Kingdom;
2
University of Washington, Seattle, WA; 3Cantonal Hospital Graubnden, Chur, Switzerland; 4University of Texas MD Anderson
Cancer Center, Houston, TX and 5Amgen Inc, Thousand Oaks, CA.
Body: Background: Patients with advanced breast cancer and bone metastases are at an increased risk for experiencing
skeletal-related events (SREs), which include pathological fracture (PF), surgery to bone (SB) radiation to bone (RB), and spinal
cord compression (SCC). The pain of SREs can be severe enough to interfere with daily functioning. Here we evaluated the
impact of SREs on pain interference in patients with advanced breast cancer and bone metastases.
Methods: In a double-blind, double-dummy, placebo-controlled trial, patients were evenly randomized to receive monthly
denosumab 120 mg SC or zoledronic acid 4 mg IV, (adjusted for renal function). Pain interference (overall, emotional well-being,
and physical function) was assessed at baseline and each study visit using the Brief Pain Inventory-Short Form (BPI-SF) with
scores that ranged from 0 (no interference) to 10 (complete interference). A change of 2 points from baseline was considered
clinically meaningful. To evaluate the overall impact of SREs on pain interference, we conducted a post-hoc analysis using a Cox
Proportional Hazards model adjusting for SREs as time-dependent covariates and stratified by treatment and randomized
stratification factors. The impact of on-study SREs was evaluated using patients first on-study SRE, starting 28 days before the
SRE occurrence.
Results: 687 first on-study SREs were reported (450 PF, 201 RB, 20 SB, 16 SCC). SCC, RB, and PF were associated with a
greater risk of a clinically meaningful increase in overall pain interference (Table 1). For the subdomains, RB and SCC were
associated with an increased risk of pain interference with emotional well-being, while PF, RB, and SB were associated with an
increased risk of pain interference with physical function.
Impact of on-study SREs on time to 2-point increase from baseline in pain interference BPI score
PF
RB
SB
SCC
Pain interference - overall (n = 1829)

HR (95% CI)
1.31 (1.05, 1.63) 2.41 (1.80, 3.23) 1.85 (0.68, 5.05) 4.26 (1.38, 13.19)
P value
0.0159
< 0.0001
0.2322
0.0120
Pain interference - emotional well being (n = 1806)

HR (95% CI)
1.20 (0.97, 1.48) 2.25 (1.72, 2.95) 1.11 (0.41, 3.00) 4.74 (2.15, 10.44)
P value
0.1003
< 0.0001
0.8432
0.0001
Pain interference - physical activity (n = 1690)

HR (95% CI)
1.35 (1.09, 1.69) 2.30 (1.70, 3.10) 2.86 (1.09, 7.47) 2.26 (0.45, 11.37)
P value
0.0070
< 0.0001
0.0326
0.3232
Includes patients with baseline BPI score 8; HR = hazard ratio; CI = confidence interval.
Conclusions: In patients with advanced breast cancer, SREs are associated with an increase in pain interference. Effective
treatments that prevent SREs may reduce the burden of pain on patients daily functioning.


Title: Development and validation of the Penn arthralgia aging scale among breast cancer survivors on aromatase inhibitors
Moriah J Brier1, Dianne L Chambless1, Laura Lee2, Angela DeMichele2 and Jun J Mao2. 1University of Pennsylvania, Philadelphia,
PA and 2Perelman School of Medicine at University of Pennsylvania, Philadelphia, PA.
Body: Background: Aromatase inhibitors (AIs) have been shown to decrease the recurrence rate and increase the overall
survival rate of hormone receptor positive breast cancer among post-menopausal women. Unfortunately, arthralgia is a frequently
experienced side effect of AIs, leading some women to discontinue AIs prematurely. Qualitative investigations suggest that joint
pain may cause women on AIs to feel they are aging faster than they should be. Since general perceptions of aging have been
shown to predict important health outcomes, such as adherence and mortality, aging perceptions related to joint pain are worth
further exploration. Objective: The purpose of this study was to develop and validate a measure that captures perceptions of
aging related to joint pain. A psychometrically validated tool will advance our ability to quantify and further understand the
importance of this construct. Method: We developed the eight-item Penn Arthralgia-Aging Scale (PAAS) from interviews with 67
patients on AIs. The scale was pilot-tested, and changes to items were made based on patient feedback, as well as feedback
from oncologists, nurses, and physical therapists. To validate the scale, participants suffering from joint pain were selected from a
larger study examining the genetic determinants of symptom distress and disease outcomes among women on AIs. Five hundred
and fifty-six breast cancer survivors completed the PAAS, as well as the Hospital Anxiety and Depression Scale, the pain
interference and pain intensity subscales of the Brief Pain Inventory, and a demographic questionnaire. Exploratory factor
analysis using oblique rotation was conducted to examine the factor structure of the scale. Convergent validity was assessed by
correlating the PAAS with joint-pain severity. To determine whether the scale provides important information beyond existing
measures, we used hierarchical regressions to calculate whether it predicted incremental variance in anxiety, depression, and
pain interference outcomes. Results: The resulting scale had a one-factor structure (eigenvalue = 6.21), high internal consistency
(Cronbachs alpha = 0.94), and strong convergent validity (Spearman r = .55, p < 0.01 for joint pain). Additionally, the PAAS was
found to explain additional variance in anxiety (7%, p < 0.001) and depression (28%, p < 0.001) after pain severity and age were
controlled. The PAAS also explained additional variance (5%, p < 0.001) in joint pain interference above and beyond the variance
accounted for by anxiety, depression, joint pain severity, and age. Conclusions: These findings suggest that the PAAS is a
reliable and valid tool that captures a meaningful construct of perceptions of aging attributable to arthralgia. With further research,
the PAAS may advance our understanding of breast cancer survivors emotional, behavioral, and clinical outcomes.

Title: Quality of life in patients receiving first-line eribulin mesylate for HER2- locally recurrent or MBC
Lee Schwartzberg1, Kristi McIntyre2, Joyce O Shaughnessy3, Stefan Glck4, Erhan Berrak5, James Song5, David Cox5 and
Linda Vahdat6. 1West Clinic, Memphis, TN; 2Texas Oncology, Dallas Presbyterian Hospital, US Oncology, Dallas, TX; 3Baylor
Charles A. Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, TX; 4Sylvester Comprehensive Cancer Center,
University of Miami Health System, Miami, FL; 5Eisai Inc, Woodcliff Lake, NJ and 6Weill Cornell Medical College, New York, NY.
Body: Introduction: Eribulin mesylate is a nontaxane microtubule inhibitor approved to treat MBC in patients (pts) who previously
received 2 chemotherapeutic regimens for MBC. A phase 2 study of first-line eribulin for HER2-negative (HER2-) MBC showed
an overall response rate of 29%, median 6.8 m progression-free survival, and tolerability consistent with earlier studies. We
present prespecified quality of life (QoL) results for this trial.
Methods: Pts (N=56) received eribulin mesylate 1.4 mg/m2 IV on days 1 and 8 of each 3-wk cycle (median: 7 cycles). QoL was
assessed using the EORTC QoL assessment (QLQ-C30) and a breast-cancer specific questionnaire (QLQ-BR23) pretreatment
(baseline) and on day 1 of every other cycle during treatment. Percentage of pts with at least 10-point change from baseline was
summarized descriptively. Linear mixed-effects models were used to evaluate changes over time and compare responders vs
nonresponders controlling for baseline score and time effect. Time-to-event analysis was performed on time to deterioration,
defined as time from 1st dose to 1st occurrence of worsening in QoL score that reached minimally clinically important difference
(MID; eg, 10 points in global health status in QLQ-C30) from baseline without further improvement of at least MID.
Results: For QLQ-C30 at cycle 6 (n=29), more pts had at least a 10-point improvement from baseline in role, emotional, and
social functioning; fatigue, nausea/vomiting, pain, dyspnea, and insomnia item scores, than had worsening. More pts had
worsening in global health status/QoL, cognitive functioning, and diarrhea (Table).
Category, n (%)
Improved
Stable
Worsened
Global health status/QoL
2 (7)
16 (55)
11 (38)
Physical functioning
9 (31)
13 (45)
7 (24)
Role functioning
14 (48)
10 (35)
5 (17)
Emotional functioning
9 (31)
14 (48)
6 (21)
Cognitive functioning
2 (7)
13 (45)
14 (48)
Social functioning
10 (35)
13 (45)
6 (21)
Fatigue
16 (55)
5 (17)
8 (28)
Nausea and vomiting
8 (28)
16 (55)
5 (17)
Pain
15 (52)
8 (28)
6 (21)
Dyspnea
9 (31)
18 (62)
2 (7)
Insomnia
12 (41)
15 (52)
2 (7)
Appetite loss
7 (24)
16 (55)
6 (21)
Constipation
5 (17)
18 (62)
6 (21)
Diarrhea
2 (7)
21 (72)
6 (21)
Median time to deterioration in global health status/QoL was 5.06 m (responders, 8.54 m; nonresponders, 3.71 m; hazard
ratio=0.60, P=0.22). In linear mixed models, responders (n=16) performed better than nonresponders (n=40) in role functioning
(P=0.011), emotional functioning (P=0.031), fatigue (P=0.007), pain (P=0.047), insomnia (P=0.018), and appetite loss (P=0.032).
Mean symptom scores were significantly correlated with corresponding adverse event rates for nausea and vomiting, dyspnea,
appetite loss, constipation, and diarrhea; Spearman rank correlation coefficients ranged from 0.31 to 0.54. For QLQ-BR23 at
cycle 6, symptom scores were mostly stable; more pts had worsening in body image and systemic therapy side effects than had
improvement and more pts had improvement in breast and arm symptoms than had worsening. Responders also had longer time
to symptom deterioration.
Conclusions: In this study of first-line eribulin treatment for HER2- MBC, a majority of pts had stable or improvement in QoL
scales. Responders to eribulin were more likely than nonresponders to have stable or improved QoL.

Title: Quality of life results from a phase 2, multicenter, single-arm study of eribulin mesylate plus trastuzumab as first-line
therapy for locally recurrent or metastatic HER2+ breast cancer
Lee Schwartzberg1, Sharon Wilks2, Shannon Puhalla3, Joyce O Shaughnessy4, Erhan Berrak5, James Song5, David Cox5 and
Linda Vahdat6. 1West Clinic, Memphis, TN; 2US Oncology-Cancer Care Centers of South Texas, San Antonio, TX; 3University of
Pittsburgh Medical Center, Pittsburgh, PA; 4Baylor Charles A. Sammons Cancer Center, Texas Oncology, Dallas, TX; 5Eisai Inc,
Woodcliff Lake, NJ and 6Weill Cornell Medical College, New York, NY.
Body: Introduction: Eribulin mesylate is a nontaxane microtubule dynamics inhibitor that has showed an overall survival benefit
relative to other commonly used agents in patients with 2 prior MBC therapies. Primary data from a phase 2 trial for first-line
eribulin + trastuzumab [TRAS] in HER2+ patients with MBC showed an objective response rate of 71%, clinical benefit rate of
84.6%, disease control rate of 96.2%, PFS of 11.6 months, and tolerability similar to known profiles for these agents. Here, we
present prespecified QoL, efficacy, and safety/tolerability results.
Methods: Patients received eribulin mesylate 1.4 mg/m2 IV on days 1 and 8 of each 21-day cycle and initial TRAS (8 mg/kg
IV/day 1), followed by 6 mg/kg on day 1 of each subsequent cycle. Response, PFS, QoL as measured by EORTC QoL
assessment tool (QLQ-C30) and QLQ-BR23, and tolerability were assessed. Percentage of patients with at least 10-point
change from baseline was calculated at each visit. Time to deterioration was defined as time from first dose to worsening in QoL
score that reached minimally clinically important difference (MID) (ie, 10 points in global health status [GHS] in QLQ-C30) without
further improvement of at least MID; this was estimated overall and by response status.
Results: At cycle 6 (n=44; completion rate=84.6% of 52 patients enrolled), more patients fell in the stable category (within +/-10
points change from baseline), except for pain (47.7% with improvement), cognitive functioning (45.5% worsening), fatigue and
systemic therapy side effects (50% worsening for each), and arm symptoms (47.7% improvement) (Table). Median times to
deterioration for GHS/QoL were 7.6 months overall (n=51), and 7.6 and 7.0 months for responders (n=36) and nonresponders
(n=15), respectively (HR 0.73; 95% CI 0.32, 1.68; P=0.446). Mean symptom scores in EORTC QLQ-C30 were significantly
correlated with corresponding AE rates for fatigue (r=0.31), nausea/vomiting (r=0.50), pain (r=0.41), dyspnea (r=0.49), insomnia
(r=0.35), constipation (r=0.30), and diarrhea (0.40; P0.03 for all comparisons). The most common treatment-related AEs (all
grade incidence 25%) were alopecia (88.5%), fatigue (69.2%), peripheral neuropathy (69.2%), neutropenia (59.6%), nausea
(46.2%), diarrhea (32.7%), anemia (25%), constipation (25%), and decreased appetite (25%).
Table. Qol EORTC QLQ-C30 Scores: Change from Baseline to Cycle 6 (n=44)
Functional or
Category, n (%)
Symptom Scale
Improved
Stable
Worsened
GHS/Qol
13 (29.5)
23 (52.3)
8 (18.2)
Physical
10 (22.7)
26 (59.1)
8 (18.2)
Role
13 (29.5)
20 (45.5)
11 (25.0)
Emotional
17 (38.6)
18 (40.9)
9 (20.5)
Cognitive
6 (13.6)
18 (40.90
20 (45.5)
Social
12 (27.3)
18 (40.9)
14 (31.8)
Fatigue
14 (31.8)
8 (18.2)
22 (50.0)
Nausea/vomiting
9 (20.5)
24 (54.5)
11 (25.0)
Pain
21 (47.7)
14 (31.8)
9 (20.5)
Dyspnea
12 (27.3)
19 (43.2)
13 (29.5)
Conclusions: Given the improvements in pain and in arm and breast symptoms, long median time to deterioration in
functioning/symptom scales in this analysis, and the tumor response rates and safety profile in the primary analysis, combination
eribulin/TRAS may be an acceptable treatment option for locally recurrent or HER2+ MBC and merits further study in larger
clinical trials.

Title: Long term hair loss in patients with early breast cancer receiving docetaxel chemotherapy
Nicola J Thorp1, Felicity Swift1, Donna Arundell1 and Helen Wong1. 1Clatterbridge Cancer Centre, Wirral, Merseyside, United
Kingdom.
Body: Background
There is increasing recognition that a small number of patients receiving docetaxel-containing regimes for early breast cancer
(EBC), experience permanent alopecia. However, there is little data to inform discussions with patients regarding this serious late
side effect. The aim of this study was determine the incidence, the site, the extent, and duration of the hair loss.
Methods
A postal questionnaire was sent (in October 2013) to patients who had received docetaxel during 2010, in the neo/adjuvant
settings for EBC at our regional cancer centre. This comprised questions relating to scalp hair loss (using the Ludwig scale to
provide a pictorial description of the pattern of hair loss), hair loss to other parts of the body, hair products used, and any
comments that the respondents wished to add about their experience of hair loss. Univariate and multivariate analyses were
undertaken to determine any other risk factors for persistent alopecia.
Results
134 of 189 (71%) questionnaires were returned. Of those responding, 72 patients were pre-, 10 were peri- and 52
post-menopausal. 26 patients were taking anastrazole, 14 letrozole, 74 tamoxifen and 20 no adjuvant hormones. Of the
respondents, 99 (74.4%) patients had no significant scalp hair loss, and 21 (15.8%) had significant scalp hair loss. 13 (9.8%) of
patients gave equivocal responses and 1 patient did not answer the scalp hair loss question. 16 patients in the study were using
products such as wigs and hair extensions. 5 patients reported no regrowth of eyebrows, 2 patients reported no eyelash regrowth,
6 no regrowth of nostril hair and 14 no regrowth to other parts such as legs. Univariate and multivariate analyses showed no
significant associations with other patient and treatment characteristics (eg adjuvant endocrine therapy). Patients observations
regarding the social and emotional consequences of permanent hair loss confirmed a significant impact on quality of life.
Conclusions
This retrospective questionnaire study confirms that long term significant scalp alopecia (here lasting for up to 3.5 years following
completion of chemotherapy) may affect 10-15% of patients following docetaxel for EBC (taking into consideration a potential bias
for no hair loss in the non- responders). This rate is higher than previous estimates. Long term hair loss to other parts of the body
was also widely reported. This appears to be unrelated to other patient and treatment characteristics. Long term hair loss had a
significant impact on quality of survival. This is an important quality of life issue for patients which merits prospective study to
confirm incidence, to identify effective preventive and management strategies. This risk should be discussed routinely (as part of
the process of informed consent) with all patients embarking upon docetaxel as a component of management of EBC.

Title: It's as simple as ABC: Can we improve the provision of coordinated medical and supportive care to patients with advanced
breast cancer?
Sally Greenberg1, Meron Pitcher1, Bruce Mann2,3, Kathleen Hendry1, Kerry Shanahan2,3, Melanie K Fisher1 and Jung H Foo4.
1
Western Hospital, Melbourne, Victoria, Australia; 2Royal Melbourne Hospital, Melbourne, Victoria, Australia; 3Royal Women's
Hospital, Melbourne, Victoria, Australia and 4Western and Central Melbourne Integrated Cancer Service, Melbourne, Victoria,
Australia.
Body: Back ground: International and local guidelines regarding advanced breast cancer (ABC) management recommend
coordinated care, discussion in multi-disciplinary meetings (MDM), breast care nurse (BCN) involvement and supportive care
screening (SCS). There was concern among clinicians, at three University affiliated hospitals in Melbourne Australian, that these
were not routinely performed for all patients at our hospitals. A WCMICS service improvement grant was used to design a Model
of Care (MOC) aiming to address these deficiencies.
Method: A 6 month retrospective audit of all new ABC patients was conducted at each institution. We reviewed timing and details
of diagnosis, MDM discussions, BCN involvement, SCS and communication with primary care physicians (PCP). Surveys/
interviews of PCP, patients and oncology clinicians were conducted. A best practice MOC was developed, piloted for three
months and evaluated.
Results: The new MOC resulted in an increased proportion of patients discussed in MDM (from 33% to 88%), reviewed by a BCN
(from 30% to 80%), and improved provision (from 61% to 90%) and timing (median 20 down to 16 days) of communication with
PCP. 50% of patients interviewed in both the baseline and subsequent evaluation surveys did not feel their care was well
coordinated between specialists and PCP.
Conclusions: A best practice MOC for patients with newly diagnosed ABC is possible to implement across a number of institutions
with improvements in the provision of coordinated medical and supportive care. Further improvements can be made in
coordination of care between PCP and specialists.

Title: Emotional/psychological characteristics of women with triple-negative breast cancer: Do socioeconomic, demographic, and
provider variables impact emotional change from diagnosis to post-treatment?
Kathleen D Swiger1, Jocelyn A Sendecki2, Janine E Guglielmino3, Hope S Rugo, Carey K Anders4, Susan M Domchek5, Arin
Ahlum Hanson3, Hayley Dinerman6 and Catherine Creme Henry3. 1Consultant, Greensboro, NC; 2Thomas Jefferson University,
Philadelphia, PA; 3Living Beyond Breast Cancer, Haverford, PA; 4University of North Carolina, Chapel Hill, NC; 5University of
Pennsylvania, Philadelphia, PA and 6Triple Negative Breast Cancer Foundation, Norwood, NJ.
Body: Purpose: To determine whether women with triple-negative breast cancer (TNBC) experience greater levels of emotional
concern from diagnosis through post-treatment compared to women with other breast cancer subtypes.
Respondents and Methods: Women diagnosed with breast cancer responded to an 80-question online survey to identify
education, information, and support needs. Respondents self-reported their breast cancer subtype and rated the emotions they
experienced at diagnosis, during, and after treatment on a scale of "none," "low," "moderate," and "high." The responses of 656
women with TNBC (25.1%) were compared to 1,954 non-TNBC women (74.9%). Differences between TNBC and non-TNBC
women were assessed using logistic regression at each time point. Using generalized logistic modeling, differences in score
changes were categorized as decreased, same, or increased in relation to cancer subtype and other covariates.
Results: At all time points in an unadjusted analysis, TNBC women reported more fear, anxiety, and worry than non-TNBC
women, although this finding was only statistically significant for fear and anxiety at diagnosis (high fear: 67% vs. 62%, p=0.046,
high anxiety 68% vs. 64%, p=0.046 respectively). Change in emotions between diagnosis and treatment phase was not
significantly different between cancer types. Between treatment and post-treatment, women with TNBC were significantly less
likely than non-TNBC patients to report a decrease in negative emotion (fear: 58% vs. 66%; anxiety: 54% vs. 65%; worry: 53% vs.
63%, p<0.001 for all). TNBC women with young children were less likely to report a decrease and more likely to report an
increase in worry than non-TNBC women (decrease: 61% vs. 70%; increase 8% vs. 4%, p=0.09). A similar pattern was seen in
TNBC women with income <$50K annually with respect to fear (decrease: 47.3% vs. 68%, increase 11% vs. 6%, p=0.06). Cancer
stage was significantly associated with emotional change. Women with TNBC stage >=2 showed greater increases in negative
emotion and lesser decreases in positive emotion than non-TNBC women with similarly staged cancers (p<0.001). Race/ethnicity,
age, education, children, living situation, or use of a mental health professional did not influence this relationship.
Conclusion: Women with TNBC experience greater fear, anxiety, and worry than women with non-TNBC subtypes at all points
from diagnosis though post-treatment. While women with all breast cancer subtypes report a reduction in negative emotion over
time from treatment to post-treatment, this change is less profound in TNBC women and appears to be driven nearly entirely by
concern about the disease. The marginal effect on change in fear with respect to income may reflect concerns about paying for
care, and increased worry in women with small children may reflect concerns about prognosis. Most strikingly, cancer stage was
the strongest modifier of emotional change: TNBC women at cancer stage >=2 showed the least decline in negative emotion
compared to corresponding non-TNBC women. These data support the development of TNBC-specific interventions focused on
these patients emotional needs during and after treatment.

Title: TransHERA: The cell cycle regulator p27 predicts benefit from trastuzumab treatment in HER2-positive early breast cancer
patients treated within the HERA trial
Martin Filipits1, Michael Gnant1, Urania Dafni2, Varvara Polydoropoulou2, Margaret J Hills3, Brian Leyland-Jones4, Martine
Piccart-Gebhart5 and Mitch Dowsett3. 1Medical University of Vienna, Comprehensive Cancer Center, Austria; 2Frontier Science
Foundation-Hellas, Greece; 3Royal Marsden Hospital, London, United Kingdom; 4Stanford University and Avera Cancer Institute
and 5Jules Bordet Institute, Belgium.
Body: Purpose: Predictive biomarkers may help predicting adjuvant trastuzumab response and thus optimize the treatment of
patients with HER2-positive breast cancer. The aim of the present study was to assess the prognostic/predictive value of various
biomarkers involved in cell cycle regulation or proliferation.
Methods: Expression of p27, cyclin D1, TOP2a, and Ki67 was immunohistochemically determined in tissue micro arrays of
specimens from 862 patients randomized to the trastuzumab (1 or 2 year; N=561) and observation (N=301) arms of the HERA
trial. The primary endpoint of the analysis was disease-free survival (DFS). Biomarker expression status was determined as
continuous variable or by pre-defined categories. The interaction terms between the four biomarkers and treatment were
assessed in multivariate Cox proportional hazards regression models adjusted for variables of clinical interest. Associations were
considered significant only if the false discovery rate (FDR) adjusted p-values remained significant.
Results: Baseline characteristics were well balanced between the two study arms. A total of 249 DFS events (28.9%) were
observed in the TransHERA cohort, with an overall 8-year DFS of 70.5% (95% CI 67.2%-73.5%). None of the four biomarkers
was significantly associated with DFS in the total study population. When biomarkers were categorized according to pre-defined
cut-off levels, only p27 turned out to be highly predictive: Expression data for p27 were available in 753 TransHERA patients. A
highly significant interaction was detected between p27 and treatment when adjusting for clinical parameters and the remaining
three biomarkers (p=0.0039). For patients classified as p27 low (70% p27-positive tumor cells; N=318), a significant treatment
effect was observed, with the hazard of a DFS event being greater for the observation group compared to patients treated with
trastuzumab (HRTrast vs Obs=0.43, 95% CI 0.29-0.64, p<0.001). In contrast, no statistically significant effect of trastuzumab
treatment was detected in the p27 high group (N=435; HRTrast vs Obs=0.97, 95% CI 0.66-1.44, p=0.89), indicating that p27 high
patients derived little or no benefit from trastuzumab treatment. Cyclin D1, TOP2a, and Ki67 used as categorical variables were
not predictive, while cyclin D1 used as continuous variable was predictive of adjuvant trastuzumab benefit.
Conclusion: HER2-positive early breast cancer patients with low p27 expression in their tumors appear to benefit from
trastuzumab treatment, whereas patients with high p27 expression do not.
This study was funded by Roche.

Title: Trastuzumab interruption for treatment-induced cardiotoxicity in HER2 positive early breast cancer
Anthony F Yu1, Nandini U Yadav1, Betty Y Lung1, Anne A Eaton1, Howard T Thaler1, Clifford A Hudis1, Chau T Dang1 and Richard
M Steingart1. 1Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background
Trastuzumab (H) improves outcomes among patients with HER2-positive breast cancer but is associated with a risk of
treatment-induced cardiotoxicity (TIC), especially when administered after an anthracycline (A). H interruption is recommended
for significant asymptomatic or symptomatic declines in left ventricular ejection fraction (LVEF). It is unclear how frequently TIC
leads to H interruption, and the subsequent management is variable in clinical practice.
Methods
Patients (pts) with HER2-postive early breast cancer receiving adjuvant H with chemotherapy between January 2005 and October
2010 were studied (n=608). Tumor characteristics, chemotherapy regimen, cardiac risk factors, LVEF (at baseline and during
treatment), and treatment interruption were obtained from the medical record. We evaluated the incidence, time of occurrence,
management, and associated risk factors of H interruption due to TIC.
Results
Median age was 51 years (range 26-81); 488 (80%) pts had A prior to H administration. H was interrupted in 108 (18%) pts.
Cumulative dose of H was lower among pts in the interrupted group (median 86 vs. 108 mg/kg, p < 0.0001). The most common
reason for interruption was TIC (66 of 108 pts): 20 (30%) had symptomatic congestive heart failure and 46 (70%) had
asymptomatic decline in LVEF. Of the 66 pts, 55 (83%) were referred to a cardiologist and 36 (55%) were prescribed a new
cardiac medication. The mean LVEF at baseline, at time of cardiotoxicity diagnosis, and at follow-up after interruption of H was
63%, 45%, and 55%, respectively. Pts with H interruption for TIC were older (54 vs. 50 years, p=0.014) with lower LVEF before A
(63 vs. 67%, p<0.0001) and before H therapy (62% vs. 67%, p<0.0001), compared to those with continuous H treatment.
Thirty-three of 66 pts were re-treated with H, and 5 pts had a significant recurrent decline in LVEF.
Conclusion
At our institution, interruption of H therapy is common and most often due to TIC, from both A and H, with the majority of pts
receiving A prior to H. Risk factors associated with H interruption were older age and lower LVEF prior to A and H administration.
Cardiac dysfunction improved after interruption of treatment but did not fully recover to baseline. Strategies to prevent
cardiotoxicity and minimize H treatment interruption should be investigated to prevent persistent LV dysfunction in affected pts.



Title: The Promher Study: An observational Italian study on HER2+ve, pT1a-b, pN0, M0 breast cancer (BC) patients (pts)
Stefania Gori1, Monica Turazza1, Simona Duranti1, Elena Fiorio2, Jennifer Foglietta3, Marcella Gulisano4, Ilaria Marcon5, Marta
Gubbiotti6, Maria Giovanna Cavazzini7, Simon Spazzapan8, Valeria De Simone9, Giancarlo Bisagni10, Chiara Saggia11, Luigi
Cavanna12, Emilio Bria13, Laura Iezzi14, Elisabetta Cretella15, Patrizia Vici16, Daniele Santini17, Alessandra Fabi17, Ornella
Garrone18, Antonella Ferro19, Silvana Saracchini20, Lucia Evangelisti21, Sandro Barni22, Lucia Mentuccia23, Lucio Laudadio24,
Alessandro Inno1, Gianluigi Lunardi1, Francesca Coati1 and Luca Boni25. 1Ospedale Sacro Cuore Don Calabria, Negrar
Verona, Italy; 2Ospedale Civile Maggiore AO Universitaria, Verona, Italy; 3Ospedale S.Maria della Misericordia, Perugia, Italy;
4
Ospedale ULSS 6, Vicenza, Italy; 5Ospedale Circolo e Fondazione Macchi, Varese, Italy; 6Ospedale Civile, Macerata, Italy; 7AO
C.Poma, Mantova, Italy; 8Istituto Nazionale Tumori-IRCCS, Aviano (Pordenone), Italy; 9Ospedale San Donato, Arezzo, Italy;
10
IRCCS AO Santa Maria Nuova, Reggio Emilia, Italy; 11Ospedale Maggiore, Novara, Italy; 12Ospedale G. da Saliceto, Piacenza,
Italy; 13AO Universitaria Integrata Policlinico GB Rossi, Verona, Italy; 14P.O.SS.Annunziata, Chieti, Italy; 15Ospedale Azienda
Sanitaria Alto Adige, Bolzano, Italy; 16Istituto Nazionale Tumori Regina Elena, Roma, Italy; 17Universit Campus Bio Medico,
Roma, Italy; 18ASO S.Croce e Carle, Cuneo, Italy; 19Ospedale S.Chiara, Trento, Italy; 20Ospedale S.Maria degli Angeli,
Pordenone, Italy; 21ASL-CN1 Ospedale Civile, Saluzzo-Cuneo, Italy; 22AO Treviglio, Bergamo, Italy; 23Ospedale SS.Trinit, Sora
Frosinone, Italy; 24Ospedale Renzetti, Lanciano Chieti, Italy and 25Istituto Toscano Tumori AOU Careggi, Firenze, Italy.
Body: Background. The management of small ( 1 cm), node-negative, HER2+ve BC is controversial, since data from
randomized clinical trials specifically addressing the benefit of adjuvant systemic treatment with or without Trastuzumab in this
setting are still lacking. The aims of this retrospective study are to assess how pts are managed in routinary clinical practice in
Italy, whether clinical or biological features may influence the choice of adjuvant systemic therapy and if there is any difference in
the outcome between treated and not treated pts.
Patients and methods. Data of 268 consecutive pts who underwent surgery from January 2007 to December 2012 for HER2+ve,
pT1a-b pN0 BC, were collected from 25 Italian centres. Descriptive statistical analyses and multivariate logistic regression models
were used, with the aim of investigating the relationship between the baseline clinical and biological features and the adjuvant
treatment strategy.
Results. Pts characteristics were: median age 57, 69% postmenopausal status, 77% had conservative surgery, 32% pT1a, 68%
pT1b, 48% G3, 66% ER+ve, 75% Ki67 14%. Ninety percent of pts received adjuvant systemic therapy: 19% hormone therapy
(HT) alone, 3% chemotherapy (CT) +/- HT, 64% Trastuzumab + CT +/- HT and 4% Trastuzumab + HT. At the multivariate
analysis, the odds of being treated with adjuvant systemic therapy with or without Trastuzumab, resulted higher in presence of
conservative surgery (p=0.002), pT1b (p<0.001) and positivity of hormone receptors status (p<0.001). Among the patients treated
with adjuvant systemic therapy, the administration of Trastuzumab appeared to be more frequently associated with pT1b
(p=0.010) and negative hormone receptors (p=0.004). After 37 months of median follow-up, local and/or distant recurrence were
4/29 (14%) for pts who did not receive any systemic treatment, 2/59 (4%) for pts receiving systemic treatment without
Trastuzumab and 2/180 (1%) for pts receiving Trastuzumab.
Conclusion. This preliminary analysis shows that in Italy the majority of these pts received systemic adjuvant treatment and about
2/3 were treated with Trastuzumab. Pathological tumor size (pT1b) and negative hormone receptor status represent the main
factors influencing the choice of including Trastuzumab in the adjuvant treatment. Survival data are still not mature to drive
definitive conclusions about outcome.

Title: Trastuzumab can be safely administered concurrently with anthracycline for adjuvant treatment of HER2-positive breast
cancer
Songjie Shen1, Qiang Sun1, Ying Xu1, Yidong Zhou1, Jinghong Guan1, Feng Mao1, Yan Lin1 and Xuejing Wang1. 1Peking Union
Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China.
Body: Background Anthracycline and trastuzumab are the preferential choices in the treatment of HER2-positive breast cancer.
Due to the unacceptably high rates of cardiotoxicity observed in the metastatic breast cancer, concurrent administration of
anthracycline and trastuzumab (A+H) is contraindicated. However, in the neoadjuvant setting, A+H has shown high rates of
pathological complete response and very low cardiotoxicity. So far, all the large adjuvant trials have only evaluated the sequential
strategy of administration anthracycline and trastuzumab (A-H), whereas the safety and efficacy of A+H has never been
evaluated prospectively. Thus, we conducted a prospective study to evaluate the cardiac safety and efficacy of A+H regimen in
the adjuvant treatment of HER2-positive breast cancer.
Methods This is a prospective, randomized and controlled trial. Participants, with HER2-positive, operable breast cancer, but
without previous neoadjuvant treatment, were randomized to receive adjuvant A+H or A-H. If anthracycline was administered
alone or sequentially to taxane, the dose of doxorubicin and epirubicin was 60mg/m2 and 90-100mg/m2, respectively. If
anthracycline was given concurrently with taxane, the dose of doxorubicin and epirubicin was 50mg/m2 and 75mg/m2,
respectively. Trastuzumab was given every 3 weeks (loading dose of 8mg/kg, followed by 6mg/kg) for one year. Left ventricular
ejection fraction (LVEF) was monitored by echocardiogram (ECHO) at baseline (before chemotherapy) and 3, 6, 9, 12 and 24
months after the initial dose of trastuzumab. The primary endpoint was cardiac safety. The second endpoints were disease-free
survival (DFS) and overall survival. ClinicalTrials.gov ID: NCT01413828.
Results Between August 2011 and March 2014, 196 HER2-postive breast cancer patients (98 in the A+H group and 98 in the
A-H group) were enrolled and randomized. Women in the two groups had similar baseline characteristics including age, tumor
stage, hormonal receptor status, chemotherapy regimen, radiation therapy and endocrine therapy. Trastuzumab was
well-tolerated in both groups and the primary cardiac event was asymptomatic decrease in LVEF. In the A+H group, there were
11 (11.2%) patients showed more than 10% but less than 20% reduction in LVEF (NCI-CTC Grade I). In the A-H group, there
were 14 (14.3%) patients showed NCI-CTC Grade I LVEF reduction and 1 (1.0%) patient showed more than 20% reduction in
LVEF (NCI-CTC Grade II). There was no case of congestive heart failure. The difference of the rates of cardiac events between
the two groups was not significant (P=0.400). The mean LVEF of the baseline and 3, 6, 9, 12 and 24 months after initial dose of
trastuzumab also showed no difference between the two groups. Patients in both groups had excellent disease control at a
median follow up of 16 months. There were numerically more DFS events in the A-H group (6/98, 6.1%) than the A+H group
(10/98, 10.2%), but the difference of DFS did not reach the statistical significance (P=0.485). There was no death in both groups.
Conclusions Trastuzumab administered concurrently with anthracycline is a safe adjuvant regimen and might improve the
survival of patients with HER2-positvie breast cancer.

Title: Randomized Phase III trial of afatinib plus vinorelbine versus trastuzumab plus vinorelbine in patients with
HER2-overexpressing metastatic breast cancer who had progressed on one prior trastuzumab treatment: LUX-Breast 1
Nadia Harbeck1, Chiun-Sheng Huang2, Sara Hurvitz3, Dah-Cherng Yeh4, Zhimin Shao5, Seock-Ah Im6, Kyung Hae Jung7, Kunwei
Shen8, Jungsil Ro9, Jacek Jassem10, Qingyuan Zhang11, Young-Hyuck Im12, Marek Wojtukiewicz13, Qiang Sun14, Shin-Cheh
Chen15, Rainer-Georg Goeldner16, Annick Lahogue17, Martina Uttenreuther-Fischer16, Binghe Xu18, Martine Piccart-Gebhart19 and
on Behalf of the LUX-Breast 1 Study Group. 1Brustzentrum Frauenklinik der Universitt Mnchen, Munich, Germany; 2National
Taiwan University Hospital, Taipei, Taiwan; 3David Geffen School of Medicine at UCLA/Translational Research in Oncology, Los
Angeles; 4Taichung Veterans General Hospital, Taichung, Taiwan; 5Fudan University Shanghai Cancer Center, Shanghai, China;
6
Seoul National University Hospital, Seoul, Korea; 7Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea;
8
Comprehensive Breast Health Center, Ruijin Hospital, Shanghai, China; 9National Cancer Center, Goyang, Korea; 10Medical
University of Gdansk, Gdansk, Poland; 11Third Affiliated Hospital of Harbin Medical University, Heilongjiang, China; 12Samsung
Medical Center, Seoul, Korea; 13Comprehensive Cancer Centre, Medical University, Bialystok, Poland; 14Peking Union Medical
College Hospital, Beijing, China; 15Chang Gung Medical Foundation-Linkou Branch, Taoyuan County, Taiwan; 16Boehringer
Ingelheim Pharma GmbH & Co. KG, Biberach and der Riss, Germany; 17SCS Boehringer-Ingelheim Comm.V, Brussels, Belgium;
18
Cancer Institute and Hospital, Chinese Academy of Medical Sciences, Beijing, China; 19Institut Jules Bordet, Brussels, Belgium
and 20none.
Body: Background: Afatinib is an oral, irreversible ErbB family blocker with anti-tumour activity in patients (pts) with
HER2-positive metastatic breast cancer (MBC) after failure on trastuzumab.1 Preclinically, afatinib + vinorelbine (AV) showed an
additive effect; clinically, the AV combination had a manageable safety profile and showed activity in two Phase I trials.2,3 This
randomized, open-label, Phase III trial (LUX-Breast 1) compared AV with trastuzumab + vinorelbine (TV) in pts with
HER2-positive MBC who had progressed on a prior T-based regimen.
Methods: Pts with HER2-positive MBC and failure of one T-based regimen (adjuvant/first-line) were randomized 2:1 to AV (40
mg/day oral + 25 mg/m2/week iv) or TV (2 mg/kg/week iv after 4 mg/kg loading dose + 25 mg/m2/week iv). Treatment continued
until progressive disease (PD) or unacceptable adverse events (AEs). The primary endpoint was progression-free survival (PFS)
by investigator review; secondary endpoints included objective response rate (ORR), overall survival (OS) and safety. Planned
accrual was 780 pts.
Results
Between August 2010 and April 2013, 508 patients were randomized (AV:339, TV:169). Baseline characteristics were balanced in
both arms (mean age 52 yrs, Asian 50.6%, White 41.6%, ER/PR positive 28.7%). 41.1% of pts failed on prior adjuvant and 58.9%
on 1st line T-based treatment. A pre-planned risk/benefit assessment was found unfavorable by the DMC and recruitment was
stopped. Pts ongoing on AV therapy were switched to TV, received A or V monotherapy, or stopped treatment. Primary endpoint
analysis was performed with 307 of the originally 484 planned PFS events (211 [62.2%] AV arm; 96 [56.8%] TV arm). Median
PFS was 5.5 months with AV vs 5.6 months with TV (HR 1.10; 95% CI 0.86, 1.41; P=0.4272). ORR was 46.1% with AV and
47.0% with TV (OR 1.04; 95% CI 0.71, 1.51; P=0.8510). OS analysis was based on 144 (28.4%) OS events (108 [31.9%] in AV
arm; 36 [21.3%] in TV arm). Median OS was 19.6 months with AV and 28.6 months with TV (HR 1.76; 95% CI 1.20, 2.59;
P=0.0036).
The most common drug-related AEs were diarrhea (80.1%), neutropenia (75.1%) and rash (45.1%) with AV, and neutropenia
(78.7%), leukopenia (37.3%) and anemia (27.8%) with TV. Rate of infections (53.0% vs 40.5%) was higher with AV vs TV. More
AV than TV pts discontinued due to AEs (15.4% vs 7.1%). Fatal AEs were reported for 18 (5.3%) in the AV vs 5 (3.0%) pts in the
TV arm, and were mainly associated with PD (9 pts in AV and 1 in TV arm). Three AV pts died due to treatment-related causes
(sepsis/multi-organ failure; septic shock; pulmonary fibrosis).
Conclusions: AV and TV demonstrated similar PFS and ORR, but OS diverged and was shorter for AV compared to TV in pts
with HER2-positive MBC. The safety profile of AV was consistent with the individual monotherapies, but its tolerability compared
unfavorably to TV. Analyses are ongoing to elucidate potential factors (e.g. impact of follow up treatments) contributing to the
diverging PFS and OS outcomes.
1. Lin NU et al. Breast Cancer Res Treat 2012;133:1057-65
2. Bahleda R et al. J Clin Oncol 2011;29; abs 2585

3. Masuda N et al. SABCS 2013 abs P4-16-11.

Title: Everolimus in combination with exemestane in hormone receptor-positive locally advanced or metastatic breast cancer
(BC) patients progressing on prior non-steroidal AI (NSAIs): Ballet study (CRAD001YIC04)
Guy Jerusalem1, Gabriella Mariani2, Eva M Ciruelos3, Miguel Martin4, Vivianne CG Tjan-Heijnen5, Patrick Neven6, Joaquin Gavila
Gregori7, Andrea Michelotti8, Filippo Montemurro9, Istvan Lang10, Josef Mardiak11, Bjoem Naume12, Maura Camozzi13, Katia
Lorizzo13, Dariusz Brenski13 and Pierfranco Conte14. 1CHU Sart Tilman Liege and Liege University, Liege, Belgium; 2Fondazione
IRCCS Instituto Nazionale dei Tumori, Milan, Italy; 3University Hospital 12 de Octubre, Madrid, Spain; 4Hospital General
Universitario Gregorio Maran, Madrid, Spain; 5Maastricht University Medical Centre, Maastricht, Netherlands; 6KU Leuven and
University Hospital Leuven, Leuven, Belgium; 7Fundacion Instituto Valenciano de Oncologia, Valencia, Spain; 8Azienda
Ospedaliera Universitaria Pisana, Santa Chiara Hospital, Pisa, Pisa, Italy; 9Fondazione del Piemonte per l'Oncologia, Institute of
Candiolo Cancer Center (IRCCs), Candiolo (Torino), Italy; 10National Institute of Oncology, Budapest, Hungary; 11National Cancer
Institute, Bratislava, Slovakia (Slovak Republic); 12Oslo University Hospital and Norwa and K.G. Jebsen Center for Breast Cancer
Research, Institute for Clinical Medicine, University of Oslo, Oslo, Norway; 13Novartis Farma S.p.A., Origgio/VA, Italy and 14Istituto
Oncologico Veneto IRCCS and Universit di Padova, Padova, Italy.
Body: Background:
Despite progress, a large number of breast cancer patients experience metastatic relapse and death.
Hyperactivation of the mTOR pathway has been observed in patients (pts) with BC progressing on endocrine therapy. The
BOLERO-2* trial demonstrated significant doubling of PFS obtained via dual blockade with everolimus (EVE) and exemestane
(EXE), versus EXE alone in pts refractory to NSAIs.
Methods:
BALLET is a European multi-center open-label, single-arm, expanded-access study to evaluate the safety of EVE (10 mg/day)
and EXE (25 mg/day) in postmenopausal women with hormone receptor-positive HER2 negative locally advanced or metastatic
BC progressing on prior NSAIs. Study treatment continued until disease progression, unacceptable toxicity, death, drug locally
reimbursed, discontinuation from the study for any other reason or last patient last visit (June 30th, 2014), whichever occurred
first. Here we report an ad hoc-analysis that includes all pts recruited from May 12th 2012 till Dec 31st 2013 with cut-off date
March 17th 2014 (final database will be available on October 31st 2014).
Results:
2.133 pts were recruited in 269 sites across 14 European countries. Baseline characteristics were median age: 64 yrs; PS
(ECOG) 0/1/2: 64%/30%/3%; median time from first diagnosis: 8 yrs; Stage IV pts: 99%. At the data cut off, a total of 1795 pts
(84%) had discontinued the treatment. Reasons for discontinuation were: disease progression (38%), drug reimbursement (35%),
adverse events (15 %), consent withdrawn (4%), death (1.5%) and others (6.5%). EVE and EXE were administered as first line
treatment in 10% of pts, as second line in 23%, as third line in 22% and as fourth line or beyond in 45% of pts. 74% of pts
received more than 1 line of chemotherapy in the metastatic setting. 80% of pts experienced at least one adverse event (AE)
referred by the investigators as related to EVE [45% stomatitis, 7% non-infectious pneumonitis (NIP)]. The most frequent grade
3-4 drug related AEs were stomatitis (8.9%), asthenia (3.2%), GGT increase (2.4%), hyperglycemia (2.4%), and NIP (1.8%).The
median time to onset of stomatitis and NIP was 3-4 weeks and 2-3 months respectively.
Conclusions:
These results confirm that the combination of EVE + EXE is a tolerable treatment in a real world setting even in pts more heavily
pretreated by chemotherapy compared to BOLERO 2. The better understanding of side effects leading to treatment
discontinuation in this large European study where investigators frequently administered the drug for the first time, will allow
defining priority actions for better management of side effects including patient education and early interventions.
Longer follow up with mature data (expected in October 2014) will give additional information on the safety profile, stratified by
line of treatment.
* Everolimus in Postmenopausal Hormone- ReceptorPositive Advanced Breast Cancer. J. Baselga et al, NEJM 2012.


Title: A phase I/II study of neratinib plus temsirolimus in HER2+ metastatic breast cancer reveals ongoing HER2 pathway
dependence in many patients despite several lines of HER2 targeted therapy
Devika Gajria1, Shanu Modi1, Cristina Saura2, Rita Sakr1, Karl Solano1, Helen Won1, Harpreet Pannu1, Sujata Patil1, Diana Lake1,
Tiffany Traina1, Tari King1, Michael Berger1, Jose Baselga1, Neal Rosen1, Cliff Hudis1 and Sarat Chandarlapaty1. 1Memorial Sloan
Kettering Cancer Center, New York, NY and 2Vall d'Hebron University Hospital, Barcelona, Spain.
Body: Background: Targeting a single node in the HER2/PI3K pathway has been associated with benefit but frequently results in
acquired resistance in the metastatic setting. Direct inhibition of HER2 even with combinations may be limited by mutational
activation of the downstream PI3K pathway. Conversely, inhibition downstream results in feedback upregulation of receptor
tyrosine kinase signaling which can diminish efficacy. We hypothesize that dual targeting of HER2 and mTOR will be tolerable
while overcoming these limitations. We conducted a phase I/II trial examining the tolerability and efficacy of temsirolimus (T), an
mTOR inhibitor, and neratinib (N), a HER1/2 kinase inhibitor, in patients with HER2+ metastatic breast cancer (MBC). Tumor
biopsies were obtained to ascertain if PI3K pathway activation is frequent and to identify biomarkers of response.
Methods: The phase I study utilized a 3+3 dose escalation design to determine the maximum tolerated dose (MTD) of T (IV
weekly) with N (fixed dose 240 mg oral daily) in patients with HER2+ trastuzumab-refractory MBC. Loperamide prophylaxis (4 mg
daily) was initiated at Day 1 and then left to patient/physician discretion. The phase II study utilized a Simon two-stage design to
assess the overall response rate by RECIST in HER2+ trastuzumab-refractory MBC. An expansion cohort has been subsequently
initiated to investigate the benefit of more aggressive loperamide prophylaxis (16 mg daily) and temsirolimus dose escalation. All
patients on the initial Ph I/II study underwent biopsy of metastatic disease for biomarker assessment. Activating mutations in the
PI3K pathway were assayed using the Sequenom MassARRAY system or as part of Next Generation Sequencing of panel of
250 cancer related genes along with PTEN immunohistochemistry.
Results: Eight patients enrolled in the phase I trial and the MTD was determined to be T at 8 mg IV weekly with neratinib at 240
mg daily with grade 3 diarrhea as the dose limiting toxicity. The phase II trial enrolled 34 patients. Seventy percent of patients on
the Ph I/II studies had progression of disease on prior lapatinib, pertuzumab, or T-DM1. The most frequent treatment-related
grade 2/3 events at the MTD (n=40) were diarrhea (gr 2 35%/gr 3 25%), mucositis (23%/10%), and leukopenia (28%/5%).
Thirty-five patients treated at the MTD are evaluable for response; 13 patients had PR (9 cPRs) and 2 patients had SD 6
months. Most of the 15 responders had prior T-DM1 (6), pertuzumab (2), or lapatinib (8). Molecular analyses of pretreatment
biopsies is ongoing and thus far has revealed PI3K pathway activation (PIK3CA or AKT mutation or PTEN low) in >70% of the
analyzed tumors (24/33). Responses were frequent in this cohort but not in a group of tumors that had lost HER2 overexpression.
Conclusions: Temsirolimus and neratinib has clinical activity in the setting of HER2+ MBC with mutational activation of PI3K
pathway and among patients exposed to multiple prior HER2 targeted agents. Final results of the efficacy, safety, and biomarker
data will be presented along with preliminary data from the ongoing expansion study.

Title: Combination vorinostat and lapatinib reverses epithelial-mesenchymal transition, inhibits the cancer stem cell population of
HER2+ breast cancer cells and is effective in heavily pretreated advanced tumors
Amanda J Schech1, Saranya Chumsri2, Preeti Shah1, Stephen L Yu1, Nancy S Tait2, Kenneth S Bauer2, Jane C Lewis2, Ting Bao4,
Martin J Edelman3, Katherine H Tkaczuk2 and Angela H Brodie1,2. 1University of Maryland School of Medicine, Baltimore, MD;
2
University of Maryland Greenebaum Cancer Center, Baltimore, MD; 3University of New Mexico Cancer Center, Albuquerque, NM
and 4Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Although strides have been made in the treatment of breast cancer, patients often relapse due to recurrence or
metastasis. The targeting of these cellular processes has become essential in the quest for a cancer cure. Previously, we have
demonstrated the efficacy of histone deacetylase inhibitors (HDACi) in inducing differentiation and inhibiting metastasis in triple
negative breast cancer cells. In addition, we have shown that HER2 regulates the cancer stem cell (CSC) population in AI
resistant cells. Based on these previous results, we hypothesize that the combination of both HDACi and HER2 inhibition may be
synergistic and further reduce CSC. In this study, we treated a panel of HER2+ breast cancer cell lines with the combination of
1M vorinostat (HDACi) and 1M lapatinib (a dual EGFR/HER2 inhibitor) for 72 hours and examined its effects on cell number,
mesenchymal and epithelial markers, migratory potential, and cancer stem cell characteristics. The combination reduced cell
number when compared to either agent alone (p<0.01 compared to vehicle). Cells treated with the combination exhibited a more
epithelial morphology when compared to vehicle treated cells. Furthermore, when compared to vehicle treatment, the combination
downregulated the expression of mesenchymal proteins: Twist (p<0.001), Snail (n.s.), and Vimentin (p<0.001) and concurrently
upregulated epithelial proteins: cytokeratin 18 (p<0.001) and E-cadherin (n.s.), suggesting that the combination may reverse EMT
(epithelial-mesenchymal transition). The combination was also able to inhibit the migratory potential of cells as measured using
xCELLigence. In addition to its effects on EMT and migration, the combination also significantly reduced markers of the CSC
population. Cellular markers of CSC, including mammary stem cell markers CD49f (p<0.01), CD24lo/CD44hi (p<0.01), and
aldehyde dehydrogenase activity (p<0.001) were decreased following treatment with the combination. The self-renewal capability
of cells was also affected by the combination, as evidenced by decreased expression of pluripotency proteins BMI-1 (p<0.05) and
-catenin (n.s.), as well as reduction of both primary (p<0.05) and secondary (p<0.05) mammosphere formation. We further
conducted a phase I/II clinical trial of vorinostat in combination with lapatinib. There were 12 patients enrolled (9 in phase I and 3
in phase II). The treatment was fairly well tolerated with no DLT observed in phase I. PK analysis showed no evidence of
drug-drug interaction. Among 8 patients with HER2-positive (HER2+) breast cancer, the clinical benefit rate (CR, PR, and SD)
was 37.5% (1 PR, 2 SD). Intriguingly, we observed that none of the HER2+ breast cancer patients developed metastasis at a new
site during the treatment. Their disease progression was due to the enlargement of previously existing lesions. Taken together,
these results suggest that the combination of vorinostat and lapatinib may target both metastasis and CSC and that the
combination is well tolerated and effective in patients with advanced HER2+ disease.

Title: 4EVER - Final efficacy analysis of the phase IIIb, multi-center, open label study for postmenopausal women with estrogen
receptor positive locally advanced or metastatic breast cancer (BC) treated with everolimus (EVE) in combination with
exemestane (EXE)
Hans Tesch1, Oliver J Stoetzer2, Thomas Decker3, Christian M Kurbacher4, Romy Neumeister5, Frederik Marm6, Andreas
Schneeweis7, Christoph Mundhenke8, Andrea Distelrath9, Peter A Fasching10, Michael P Lux10, Diana Lueftner11, Peyman Hadji12,
Wolfgang Janni13, Mathias Muth14, Julia Kreuzeder14, Claudia Weiss14 and Diethelm Wallwiener15. 1Oncology Bethanien,
Frankfurt/Main, Germany; 2Haemato-Oncology, Munich, Germany; 3Oncology Ravensburg, Ravensburg, Germany; 4Medical
Centre Bonn Freidensplatz, Bonn, Germany; 5University Hospital Magdeburg, Magdeburg, Germany; 6University Hospital
Heidelberg, Heidelberg, Germany; 7National Center for Tumor Diseases, Heidelberg, Germany; 8University Hospital Kiel, Kiel,
Germany; 9Medical Healthcare Centre East Hessen GmbH, Fulda, Germany; 10University Hospital Erlangen, Erlangen, Germany;
11
Med. Clinic for Hematology, Oncology and Tumor Immunology, Charit Campus Benjamin Franklin, Berlin, Germany; 12Hospital
North West GmbH, Frankfurt/Main, Germany; 13University Hospital Ulm, Ulm, Germany; 14Novartis Pharma GmbH, Nuremberg,
Germany and 15University Hospital Tuebingen, Tuebingen, Germany.
Body: Introduction:
The phase III BOLERO-2 trial showed a significant doubling of PFS benefit with EVE + EXE over EXE alone in postmenopausal
women with hormone receptor positive advanced BC progressing after non-steroidal aromatase inhibitor (NSAI) therapy. The
4EVER trial further evaluated the combination of EVE+EXE with regard to efficacy and safety, quality of life and health resources
utilization in a broader patient population, i.e. without limitations as to the number of previous chemotherapy lines, the time point
of progression after NSAI therapy, and the previous EXE therapy.
Methods:
From May 2012 to November 2012 a total of 299 postmenopausal women with metastatic or locally advanced, hormone receptor
positive, HER2 negative breast cancer, refractory to NSAI were recruited to this phase IIIb study. Here we report the results of the
planned analysis of the primary and secondary endpoints. The primary endpoint was the overall response rate (ORR) at week 24.
The secondary endpoints included: Progression-free survival (PFS), ORR at week 48, overall survival (OS), and quality of life.
This study includes a broad exploratory translational research program e.g. changes in serum bone turnover biomarkers, the
correlation of Interleukin-6 with anxiety and depression, presence and molecular characteristics of circulating tumor cells, the
correlation of response to EXE+EVE with pharmacogenomics.
Results:
Trial database lock will occur in late June 2014, therefore, the final data concerning the primary and secondary efficacy and safety
endpoints will be presented at SACBS 2014.
The preliminary baseline analysis included 299 patients (data cut off 15 Nov 2013):
HR status: ER+/PgR+ 78.1%, ER+/PgR- 20.9%, 0.7% ER-/PgR+, 0.3% ER-/PgR-. Tissue for receptor status analysis: 71.0%
primary tumor, 29.0% metastasis. The mean time since initial diagnosis was 9.6 years, the mean time since first
relapse/metastasis was 4.3 years. The mean time since last relapse/metastasis was 2.8 months. 68.1% of patients had bone
lesions. Last anti-neoplastic therapy had been administered in the adjuvant (23.9%) and metastatic setting (73.0%). 25.9% of
patients had no prior antineoplastic therapy in the metastatic setting, 16.3% had one, 12.2% two and 47.4% three or more prior
therapies.
Conclusion:
The final analysis of the 4EVER study provides more important information on disease patterns and benefits of the combined
treatment with EVE and EXE.

Title: Randomized Phase II trial of afatinib alone or with vinorelbine versus investigators choice of treatment in patients with
HER2-positive breast cancer with progressive brain metastases after trastuzumab and/or lapatinib-based therapy: LUX-Breast 3
Javier Corts1, Veronique Dieras2, Jungsil Ro3, Jrme Barriere4, Thomas Bachelot5, Sara Hurvitz6, Emilie Le Rhun7, Marc Espie8,
Sung-Bae Kim9, Andreas Schneeweiss10, Joo Hyuk Sohn11, Jean-Marc Nabholtz12, Pirkko-Liisa Kellokumpu-Lehtinen13, Julie
Taguchi14, Federico Piacentini15, Eva Ciruelos16, Petri Bono13, Mahmoud Ould-Kaci17, Flavien Roux18 and Heikki Joensuu13.
1
Hospital Universitari Vall d'Hebron, Barcelona, Spain; 2Institut Curie, Paris, France; 3National Cancer Center, Goyang-si, Korea;
4
Centre Antoine Lacassagne, Nice, France; 5Centre Lon Brard, Lyon, France; 6TRIO-US Network, Los Angeles; 7Centre Oscar
Lambret and University Hospital, Lille, France; 8Hpital Saint-Louis, Paris, France; 9Asan Medical Center, Seoul, Korea; 10National
Center for Tumor Diseases, Heidelberg, Germany; 11Severance Hospital, Seoul, Korea; 12Centre Jean Perrin, Clermont-Ferrand,
France; 13Comprehensive Cancer Center, Helsinki, Finland; 14Sansum Clinic, Santa Barbara; 15Universit di Modena e Reggio
Emilia, Modena, Italy; 16Hospital Universitario 12 de Octubre, Madrid, Spain; 17Boehringer Ingelheim, Paris, France and
18
Boehringer Ingelheim, Reims, France.
Body: Background: Over one-third of patients (pts) with HER2-positive (+) advanced breast cancer (BC) develop brain
metastases (BM), which often leads to short survival. Afatinib (A), an irreversible ErbB family blocker, demonstrated activity in pts
with heavily pretreated, HER2+ metastatic BC (MBC) progressing after trastuzumab (T) therapy, with partial responses (PR) in
10% and clinical benefit in 46% of pts (Lin 2012a). We evaluated the activity of A alone or in combination with vinorelbine (V),
versus investigators choice of therapy for MBC (IC), in pts with HER2+ BC with BM after prior T and/or lapatinib (L) therapy.
Methods: Eligible pts had at least one measurable and progressive lesion in the central nervous system (CNS; 10 mm on
magnetic resonance imaging) after prior systemic and/or radiation therapy. Pts were randomized to receive A (40 mg/day oral),
AV (40 mg/day oral + 25 mg/m2/week i.v.) or IC in 3-week cycles. Stratification factors were: ECOG performance status (PS, 01
vs 2), number of BM (3 vs >3) and prior exposure to L (yes/no). The primary endpoint was pt benefit at 12 weeks (i.e. absence
of CNS and extra-CNS disease progression per RECIST 1.1, and no tumor-related worsening of neurological signs/symptoms or
increase in steroid dosage). Secondary endpoints included progression-free survival (PFS), overall survival (OS), objective
response rate (ORR) in CNS/extra-CNS lesions, and safety.
Results: 121 pts were randomized (2 were not treated): median age, 53 years; ECOG PS 01, 83%; >3 BM, 59%; prior L therapy,
78.5%. The IC treatment consisted of: T+ chemotherapy (CT) (22 pts); T+L+CT (3 pts); L+CT (10 pts); L alone (1 pt); or CT alone
(6 pts). Results for efficacy endpoints are shown (Table).
A (n=40)
AV (n=38)
IC (n=43)
Pt benefit rates at 12 weeks, n (%)
12 (30)
13 (34)
18 (42)
Median PFS, weeks
11.9
12.3
18.4
Median OS, weeks
57.7
37.3
52.1
CNS lesions
3 (8)
6 (14)
Extra CNS lesions
3 (8)
2 (5)
CNS lesions
27(68)
27 (71)
31 (72)
Non CNS lesions
17 (43)
19 (50)
26 (61)
ORR (PR), n (%):
Disease control, n (%):
One pt randomized but not treated

In the treated set (n=119), the most frequent treatment-related adverse events (AEs) in the A and AV arms were diarrhea (90%
and 84%) and rash (38% and 54%); neutropenia (51%) was also common in the AV arm. Diarrhea (33%), neutropenia (21%) and
asthenia (21%) were the most frequent related AEs in the IC arm. Grade (G) 3/4 treatment-related AEs were observed in 50%/3%
(A), 57%/24% (AV; G4 AEs were mainly neutropenias) and 14%/7% (IC) of pts; there were no treatment-related G5 events.
Conclusions: Approximately one third of pts with HER2+ MBC benefited from the assigned treatments and two thirds had CNS
lesions controlled per RECIST in each group. Objective response in CNS was infrequent (0 to 14%) with all treatments. Overall,
AEs were manageable in this heavily pretreated pt population.
Lin et al. Breast Cancer Res Treat 2012;133:10571065.

Title: IMMU-132, a potential new antibody-drug conjugate (ADC) for the treatment of triple-negative breast cancer (TNBC):
Preclinical and initial clinical results
David M Goldenberg1,2, Linda T Vahdat3, Alexander N Starodub4, Aditya Bardia5, Ellen Chuang3, Rebecca L Moroose5, Jennifer R
Diamond6, Robert M Sharkey2, Pius P Maliakal2, Steven A Hamburger2 and Allyson J Ocean3. 1Garden State Cancer Center,
Center for Molecular Medicine and Immunology, Morris Plains, NJ; 2Immunomedics, Inc, Morris Plains, NJ; 3Weill Cornell Medical
College, New York, NY; 4Indiana University Health Center for Cancer Care, Goshen, IN; 5Massachusetts General Hospital Cancer
Center, Boston, MA; 6University of Florida Health Cancer Center, Orlando, FL and 7University of Colorado Cancer Center,
Denver, CO.
Body: Despite major advances in breast cancer treatment, TNBC, which comprises approximately 15% of all breast cancer
cases, continues to have a poor prognosis, with an increased risk of recurrence and mortality. IMMU-132 is a new ADC targeting
Trop-2, an antigen found in high prevalence in many epithelial cancers, including TNBC, and conjugated to SN-38, a
topoisomerase inhibitor and active metabolite of irinotecan, at a high drug:antibody ratio (7.6:1). Studies in mice bearing human
pancreatic tumor xenografts (Capan-1) have shown IMMU-132 remains intact in the serum until internalized within the tumor cell
where SN-38 is released (pH dependent), resulting in selective cancer cell death. IMMU-132 was found to deliver >120-fold
higher amount of SN-38 to xenografted tumors than irinotecan. In animals bearing established MDA-MB-468 TNBC xenografts,
IMMU-132 at well-tolerated doses improved responses (more robust regression and delayed progression), compared to
irinotecan at a maximum therapeutic dose regimen or to an irrelevant antibody conjugate control. Animals that progressed after
receiving the control responded to IMMU-132 therapy, even after those tumors had more than doubled from their initial size.
These encouraging pre-clinical data led to the development of the ongoing phase I/II clinical trial in patients (pts) with diverse
relapsed/refractory epithelial cancers, including TNBC (NCT01631552). In dose escalation (8 to 18 mg/kg given on days 1 and 8
of a 3-week treatment regimen), dosing was limited primarily by neutropenia, with IMMU-132 being tolerated best for multiple
cycles at 8 to 10 mg/kg. Since most (26/30) archival tumors from pts with TNBC expressed Trop-2 (30% 2+) and many cancer
cell lines express abundant (i.e., >100,000) copies of Trop-2, no enrichment strategies for TNBC were employed to treat TNBC
pts with IMMU-132. As of June 1, 2014, 10 pts with TNBC have completed their first response assessment by CT, with 6 having
disease shrinkage as their best response; 2 pts had a partial response (-32 and -51% shrinkage of target lesions, with time to
progression of 18+ and 30 wks, respectively), and 4 had stable disease (-3, -14 and -19 and -27% shrinkage for 18+, 14, 45, and
24 wks, respectively), according to RECIST. For all pts with diverse cancers treated to date at the phase II dose levels of 8 and
10 mg/kg, 6/25 (24%) had G3 neutropenia; G4 febrile neutropenia occurred in 1 pt. Alopecia, diarrhea, fatigue, nausea, and
vomiting are the more common related non-hematological toxicities, but most occur at Grade 2. There has been no evidence of
immunogenicity to the antibody or the drug, even over multiple cycles of treatment.
These promising results in pts who failed multiple prior therapies (median, 4; range, 1-8) suggest IMMU-132 is a safe and
potentially effective drug for pts with TNBC, justifying continued evaluation of IMMU-132 in less refractory pts and also in
combination with other agents.

Title: Stage 1 results from MDV3100-11: A 2-stage study of enzalutamide (ENZA), an androgen receptor (AR) inhibitor, in
advanced AR+ triple-negative breast cancer (TNBC)
Tiffany A Traina1, Joyce O'Shaughnessy2, Rita Nanda9, Lee Schwartzberg10, Vandana Abramson11, Javier Cortes3, Amy
Peterson4, Iulia Cristina Tudor4, Martha Blaney4, Joyce L Steinberg5, Catherine Kelly12, Maureen Trudeau13, Ahmad Awada14, Eric
Winer15, Clifford Hudis1, Peter Schmid6 and Denise A Yardley7,8. 1Memorial Sloan Kettering Cancer Center, New York, NY; 2Texas
Oncology, Baylor Charles A Sammons Cancer Center, Dallas, TX; 3Vall d"Hebron Institute of Oncology, Barcelona, Spain;
4
Medivation Inc, San Francisco, CA; 5Astellas Pharma Global, Northbrook, IL; 6Memorial Sloan Kettering Cancer Center, London,
United Kingdom; 7Sarah Cannon Research Institute, Nashville, TN; 8Tennessee Oncology PLLC, Nashville, TN; 9University of
Chicago, Chicago, IL; 10University of Tennessee Health Science Center, Memphis, TN; 11Vanderbilt University, Nashville, TN;
12
Mater Misericordiae University Hospital, Dublin, Ireland; 13Sunnybrook Health Science Centre, Toronto, ON, Canada; 14Jules
Bordet Institute, Brussels, Belgium and 15Harvard University, Cambridge, MA.
Body: Background
TNBC is a heterogeneous disease with many subtypes that share one commonality; "triple-negative" breast tumors lack sufficient
expression of a target (ER, PgR, HER2) associated with a therapeutic agent. AR expression occurs in a subset of TNBC and
could identify patients (pts) that respond to AR inhibition. Preclinically, AR+ TNBC cell lines grow in response to AR stimulation
and this growth is inhibited by ENZA. A phase 2 clinical study investigating bicalutamide in AR+ TNBC demonstrated a 19%
clinical benefit rate at 24 weeks (CBR24) with no objective responses in 24 pts. ENZA is a potent AR inhibitor that improves
overall survival in men with metastatic castration-resistant prostate cancer and is being evaluated in pts with advanced AR+
TNBC.
Methods
MDV3100-11 is an open-label, Simon 2-stage study evaluating ENZA (160 mg daily) in pts whose breast cancer expresses AR
(>0% by IHC) but not ER, PgR or HER2 amplification (NCT01889238). There was no limit to prior therapies; bone-only
non-measurable disease was allowed. Pts with CNS metastases or seizure history were excluded. Tissue was required;
prescreening for AR was allowed. The primary endpoint is CBR at 16 weeks (CBR16), defined as complete response (CR), partial
response (PR), or stable disease (SD) 16 weeks per RECIST 1.1 in Evaluable pts. Evaluable pts were prespecified as those
with tumors expressing 10% AR by central review and who had assessment for response. ITT analyses include all pts.
Secondary endpoints include CBR24, response rates, safety and tolerability. The analysis plan specified progression to Stage 2 if
CBR16 is 3 of 26 Evaluable pts in Stage 1, and the null hypothesis (true CBR16=8%) is rejected if overall CBR16 is 9 in 62
Evaluable pts.
Results
Complete data on all Stage 1 pts (N=42) are reported herein; 16 were not evaluable (10 had AR <10%, 6 had AR 10% but no
response assessment). In the 26 Evaluable pts, median age was 62.5 years, 77% had measurable disease, 69% had 3 sites of
metastases, 62% had visceral involvement, and 42% received ENZA in 3rd line. CBR16 was 42% (11 of 26; 95%CI 24, 62) and
CBR24 was 35% (9 of 26; 95%CI 18, 54), with 1 PR and 1 CR. Related adverse events (AEs) of any grade 10% in the ITT were
fatigue (29%), nausea (26%), decreased appetite (19%), diarrhea (14%), hot flush (12%) and vomiting (10%). Fatigue (7%) was
the only Grade 3 related AE in 5%. Go to Stage 2 criteria were met, and enrollment is complete at 118. As of Sept 2014, 13
additional pts had clinical benefit at week 16 (including 3 PRs); data continue to mature.
Conclusion
The 42% CBR16 observed in Stage 1 alone was sufficiently high to reject the null hypothesis for the whole study. Data beyond
Stage 1 are not mature; however, responses continue to be observed. AEs from ENZA in women with TNBC were generally mild
and consistent with other studies of ENZA. These encouraging results suggest ENZA may provide meaningful benefit to pts with
AR+ TNBC. Ongoing IHC and genomic analyses on 400 collected tissue samples will further inform how best to identify pts most
likely to derive benefit from ENZA.

Title: Tolerability and anti-tumor activity of the oral PI3K inhibitor GDC-0941 in combination with paclitaxel, with and without
bevacizumab or trastuzumab in patients with locally recurrent or metastatic breast cancer
Patrick Schffski1, Sara Cresta2, Ingrid A Mayer3, Hans Wildiers1, Isabelle Rooney4, Doris Apt4, Steve Gendreau4, Kari Morrissey5,
Mark Lackner6, Jill Spoerke4 and Eric Winer7. 1University Hospitals Leuven, Laboratory of Experimental Oncology, Leuven,
Belgium; 2Fondazione IRCCS Instituto Nazionale dei Tumori, Milan, Italy; 3Vanderbilt-Ingram Cancer Center, Nashville, TN;
4
Exploratory Clinical Development, Genentech Inc, South San Francisco, CA; 5Small Molecule Clinical Pharmacology, Genentech
Inc, South San Francisco, CA; 6Oncology Biomarker Development, Genentech Inc, South San Francisco, CA and 7Dana-Farber
Cancer Institute, Boston, MA.
Body: Background:
GDC-0941 is a potent and selective oral inhibitor of class I PI3K kinases that demonstrates broad activity in diverse xenograft
cancer models and increases the anti-tumor activity of taxanes, associated with increased apoptotic cell death, in multiple BC
xenografts.
Methods:
In the Phase Ib dose-escalation study, GDC4629g, GDC-0941 was administered in combination with paclitaxel dosed at 90
mg/m2 intravenous [IV] weekly for 3 out of 4 weeks, bevacizumab at 10 mg/kg IV biweekly, and trastuzumab 24 mg/kg IV
weekly in 28-day cycles. The study was conducted in 2 parts: Part 1 evaluated oral doses of 60 and 100 mg GDC-0941 on a
"21+7" (21 days on/7 days off) dosing schedule in combination with paclitaxel with and without bevacizumab. Part 2 evaluated
GDC-0941 on a "5+2" schedule (5 days on/2 days off in a repeating pattern), at doses of 165 - 330 mg, in combination with
paclitaxel (Arm A), paclitaxel plus bevacizumab (Arm B) or paclitaxel plus trastuzumab (Arm C). The primary objective was to
evaluate safety, tolerability, and pharmacokinetics (PK), and to determine the maximum tolerated dose (MTD) for each
combination. Anti-tumor activity was evaluated by RECIST v1.0 and available archival tumor samples were analyzed to assess
PI3K related biomarkers.
Results:
63 patients were enrolled: 20 in Part 1 and 43 in Part 2 (18 in Arm A, 16 in Arm B, and 9 in Arm C). The most frequent grade 3
drug-related adverse events (AEs) were: Part 1, neutropenia (45%), peripheral neuropathy and nail disorder (15% each); Part 2,
Arm A, neutropenia (38%), pneumonia and rash (11% each); Arm B, neutropenia and rash (25% each); Arm C, rash and
hypertension (11% each). The most common AEs assessed as being related to GDC-0941 were as follows:
Most common AEs related to GDC-0941
Adverse Event
All Grades
Grade 3
DIARRHEA
45 (71.4%)
2 (3.2%)
NAUSEA
33 (52.4%)
NEUTROPENIA
29 (46.0%)
17 (27.0%)
RASH
29 (46.0%)
7 (11.1%)
FATIGUE
28 (44.4%)
STOMATITIS
19 (30.2%)
DECREASED APPETITE
15 (23.8%)
1 (1.6%)
VOMITING
15 (23.8%)
In Part 1, the MTD was not reached, the maximum administered dose (MAD) of GDC-0941 was 100 mg, a dose-limiting AE was
subclavian vein thrombosis. In Part 2, Arm A, the MTD was 250 mg, Grade 3 febrile neutropenia, bacteremia, and rash were
DLTs. In Arm B and Arm C, the MAD was 260 mg, Grade 3 rash was a DLT in both arms. No differences in the PK of either GDC
0941 or paclitaxel were observed when administered in combination compared to historical single-agent data. Best tumor
response by RECIST was as follows: Part 1, 1 (5%) complete (CR) and 4 (21%) partial responses (PR); Arm A, 1 (6%) CR and 3
(17%) PR; Arm B, 7 (47%) PR; Arm C, 1 (11%) PR.
Conclusions:
GDC-0941 was generally well-tolerated in combination with paclitaxel with or without bevacizumab or trastuzumab at dose ranges
of up to 330 mg at the "5+2" dosing schedule. Anti-tumor activity has been observed in combination with paclitaxel with and
without bevacizumab or trastuzumab. Updated biomarker data and associations with clinical outcomes will be presented.

Title: Noninterventional study HELENA Advanced (metastatic or locally recurrent, inoperable) HER2-positive breast cancer:
First-Line trEatmeNt with pertuzumAb after adjuvant trastuzumab therapy
Marc Thill1, Katja Ziegler-Lhr2, Harald Wagner3, Gertrud Helling-Giese4, Jasmin Greinemann5, Otto Schmalhofer6 and Dietmar
Reichert7. 1Agaplesion Markus Krankenhaus, Frankfurt a. M; 2Schwerpunktpraxis fr Gynkologische Onkologie, Kln;
3
Schwerpunktpraxis Hmatologie & Internistische Onkologie, Frth; 4St Elisabeth-Krankenhaus GmbH, Kln; 5iOMEDICO AG,
Freiburg; 6Roche Pharma AG, Grenzach-Wyhlen and 7Gemeinschaftspraxis fr Hmatologie und Onkologie, Westerstede.
Body: Background: The CLEOPATRA trial (808 patients) demonstrated a significant benefit for overall survival (HR=0.66; 95%
CI 0.520.84; p<0.0008) and progression free survival (HR=0.62; 95% CI 0.510.75; p<0.0001) in the first line treatment of
patients with HER2-positive metastatic breast cancer (MBC) receiving pertuzumab in addition to trastuzumab plus docetaxel. An
exploratory subgroup analysis was performed for patients who had received prior neoadjuvant and/or adjuvant trastuzumab
therapy (88 patients). The observed hazard ratios of 0.68 (95% CI 0.301.55) and 0.62 (95% CI 0.351.07; 16.9 months) indicate
a benefit in overall and progression free survival for this subpopulation.
Methods & Aims: HELENA is a multicentre, noninterventional study recruiting patients with advanced HER2-positive breast
cancer, who relapsed after receiving trastuzumab in the adjuvant setting. All patients receive first line treatment with pertuzumab
in combination with trastuzumab and docetaxel in clinical routine in Germany according to local SmPC. The main objective of
HELENA is to confirm the progression free survival of 16.9 months observed in the CLEOPATRA trial in this specific patient
population in a real world setting.
Parameters of interest:
Time-based efficacy parameters of pertuzumab in combination with trastuzumab and docetaxel, in particular PFS in female
patients with prior trastuzumab treatment in the adjuvant setting and in clinically relevant subgroups (hormone receptor-positive
and -negative, visceral and non-visceral metastatic disease, patients < 65 years and 65 years as well as < 75 years and 75
years)
Interval between the last documented dose of the adjuvant trastuzumab treatment and the onset of pertuzuab first-line treatment
Effective treatment duration, reason for treatment discontinuations and treatment modifications
Demographic characteristics and medical history of the patient
Evaluation of potential prognostic variables: ECOG performance status, co-morbidity, tumor stage, type of histological
classification, histologic grading, location of metastases, primary surgery method, residual tumor burden
Extension and localization of metastasis or locally recurrent, inoperable tumor before onset of pertuzuab first-line treatment
Safety of pertuzumab: Incidence, management and outcome of (S)AEs and pregnancies
Outcome of (neo)adjuvant (DFS) trastuzumab therapy
Recruitment: HELENA started recruitment in June 2013 and will enroll 478 patients in 150 sites, being suitable in number, type
and geographic distribution for providing a representative picture of first-line treatment of advanced (metastatic or locally
recurrent, inoperable) HER2-positive breast cancer in Germany.
Results: First interim results of HELENA will be presented at the SABCS 2014 meeting.

Title: Subgroup analysis on efficacy in the routine treatment - Results of the 2nd interim analysis of BRAWO, the
non-interventional trial "Breast Cancer Treatment with Everolimus and Exemestane for HR+ Women"
Christian Jackisch1, Eva-Maria Grischke2, Andreas Schneeweiss3, Thomas Decker4, Christoph Uleer5, Frank Frster6, Oliver
Tom7, Pauline Wimberger8, Christian M Kurbacher9, Bettina Mueller10, Nadia Harbeck11, Christoph Mundhenke12, Sherko
Kuemmel13, Mathias Muth14, Julia Kreuzeder14, Wilhelm Bloch15, Hans Tesch18, Diana Lftner16, Florian Schtz19 and Peter A
Fasching17. 1Sana Klinikum Offenbach, Offenbach, Germany; 2University Tuebingen, Tuebingen, Germany; 3National Center for
Tumor Diseases, University Hospital, Heidelberg, Germany; 4Medical Centre Onkologie Ravensburg, Ravensburg, Germany;
5
Medical Centre Hildesheim, Hildesheim, Germany; 6Poliklinik GmbH Chemnitz, Chemnitz, Germany; 7St Vincentius Kliniken
Karlsruhe, Karlsruhe, Germany; 8Technical University Dresden, Dresden, Germany; 9Medical Centre Bonn, Bonn, Germany;
10
GRN-Klinik Weinheim, Weinheim, Germany; 11Breast Center, Comprehensive Cancer Center, University of Munich, Munich,
Germany; 12University Hospital Schleswig-Holstein, Campus Kiel, Kiel, Germany; 13Kliniken Essen-Mitte Breast Centre, Essen,
Germany; 14Novartis Pharma GmbH, Nuremberg, Germany; 15University Cologne, Cologne, Germany; 16Charit University
Medicine, Campus Benjamin Franklin, Berlin, Germany; 17University Erlangen, Erlangen, Germany; 18Onkologie Bethanien,
Frankfurt, Germany and 19University Heidelberg, Heidelberg, Germany.
Body: Introduction
BRAWO is a German non-interventional study of 3000 patients (pts) with advanced or metastatic, hormone-receptor-positive,
HER2-negative breast cancer treated with everolimus (EVE) and exemestane (EXE). We report results of the 2nd preplanned
interim analysis (IA) with a data cut-off on 8th July 2014.
Methods
BRAWO collects data on the routine clinical treatment with EVE and EXE at about 400 sites. Main objectives are to extend the
knowledge on a) the efficacy in the clinical routine and the impact of physical activity on efficacy and quality of life, b) prophylaxis
and management of stomatitis, and c) the sequence of therapy in the clinical routine. The 2nd IA was defined to take place 12
months after the inclusion of the 500. patient into the documentation.
Results
Efficacy data will be reported for the first 500 documented pts. Apart from data on the total study population we will present the
PFS results for subgroups including pts with or without prior EXE therapy, with or without prior chemotherapy for the advanced
setting, with or without visceral metastasis and with regard to the line of treatment in the advanced setting and the extent of
physical activity. The respective summary of adverse events for these patients will be presented as well as results on treatment
compliance and dosing. Baseline characteristics and medical history as well as insights into treatment sequences before EVE and
EXE will further be shown for all enrolled patients with valid baseline documentation (approx. 1200).
Conclusion
This subgroup analysis will provide insights into the treatment efficacy in the clinical routine and will add to a more comprehensive
understanding of the treatment with EVE/EXE.

Title: Clinical activity of abemaciclib, an oral cell cycle inhibitor, in metastatic breast cancer
Sara M Tolaney1, Lee S Rosen2, Muralidhar Beeram3, Jonathan W Goldman2, Leena Gandhi1, Anthony W Tolcher3, Kyriakos P
Papadopoulos3, Drew W Rasco3, Scott P Myrand4, Palaniappan Kulanthaivel4, Joan M Andrews4, Martin Frenzel4, Damien M
Cronier4, Edward M Chan4, Keith T Flaherty5, Patrick Y Wen1, Geoffrey I Shapiro1 and Amita Patnaik3. 1Dana-Farber Cancer
Institute, Boston, MA; 2University of California, Los Angeles, CA; 3South Texas Accelerated Research Therapeutics, San Antonio,
TX; 4Eli Lilly and Company, Indianapolis, IN and 5Massachusetts General Hospital Cancer Center, Boston, MA.
Body: Background: Abemaciclib, a small molecule inhibitor with selectivity against cyclin-dependent kinases 4 and 6 (CDK4/6),
induces G1 arrest in Rb-proficient human breast cancers. In an early phase clinical trial, the safety and antitumor activity of
abemaciclib (LY2835219) were evaluated in 2 cohorts of patients with metastatic breast cancer (mBC). One cohort evaluated
single-agent abemaciclib in an unselected population of patients with mBC [Part D], while the combination of abemaciclib plus
fulvestrant was evaluated in patients with hormone receptor positive (HR+) mBC [Part G]. We previously reported early results for
these 2 cohorts of patients with mBC treated with either single-agent abemaciclib or the combination of abemaciclib plus
fulvestrant (Patnaik et al, ASCO 2014). In the single-agent cohort, 47 patients with previously treated mBC were enrolled (36
HR+). All patients with >30% tumor reduction had HR+ mBC (13 of 36 patients). In this group of 13 patients with HR+ mBC, 9
patients had confirmed response for an objective response rate of 25%, and 4 patients had unconfirmed response. This study
was ongoing with 14 of 36 HR+ mBC patients on treatment at time of analysis (range 238-471 days). Patients continuing on
single-agent abemaciclib included 4 patients with unconfirmed response and 6 patients with confirmed response. For the
combination of abemaciclib plus fulvestrant, 18 patients with HR+ mBC enrolled and 13 patients (72%) were still on treatment
(range 31-143 days) at the time of analysis.
Methods: In the single-agent cohort, patients with mBC were treated with abemaciclib at 150 or 200mg orally every 12 hours on
a continuous schedule. In the combination cohort, patients with HR+ mBC (n=18) were treated with the combination of
abemaciclib plus fulvestrant. Patients received abemaciclib at 200mg orally every 12 hours on a continuous schedule. Patients
also received fulvestrant at 500mg intramuscularly every month. NCI CTCAE v4.0 was used to grade adverse events (AEs) and
RECIST v1.1 was used to assess tumor response.
Results: In the single-agent cohort, patients began enrolling in May 2012 with the last patient enrolled in March 2013. Patients
had a median of 7 prior systemic therapies and 81% of patients had 2 metastatic sites. In the combination cohort, patients began
enrolling in September 2013 with the last patient enrolled in January 2014. Patients in the combination cohort had a median of 4
prior systemic therapies and 67% of patients had 2 metastatic sites.
An updated analysis will be presented for objective response rate, duration of treatment and clinical benefit rate and will include
an additional 6 months of information for both the single-agent and combination cohorts. New analyses will include time to
response, duration of response, change in tumor size over time, and characteristics of responders. In addition, safety data will
include longer term follow-up through approximately September 2014.
Conclusions: Abemaciclib is an oral cell cycle inhibitor that demonstrates single-agent activity against mBC, especially for HR+
disease. Based on its safety and efficacy profile, abemaciclib warrants further clinical investigation in confirmatory studies, both
as a single agent and in combination with endocrine therapy.

Title: A phase 1b study of trebananib plus paclitaxel and trastuzumab or capecitabine and lapatinib in patients with HER2+ locally
recurrent or metastatic breast cancer
Peter A Kaufman1, Gilles Freyer2, Margaret Kemeny3, Anthony Goncalves4, Guy Jerusalem5, Alison Stopeck6, Nandagopal
Vrindavanam7, Florence Dalenc8, Nuwan Nanayakkara9, Benjamin Wu10, Cheryl A Pickett-Gies11 and Hans Wildiers12. 1Norris
Cotton Cancer Center, Dartmouth-Hitchcock Medical Center, Lebanon, NH; 2Oncologie Medicale, Centre Hospitalier Lyon Sud,
Pierre Benite Cedex, France; 3Queens Cancer Center, Great Neck, NY; 4Institut Paoli-Calmettes, Marseille, France; 5Centre
Hospital Universitaire de Lige (Sart-Tilman), Liege, Belgium; 6University of Arizona, Arizona Cancer Center, Tucson, AZ; 7Signal
Point Clinical Research Center, LLC, Middletown, OH; 8Institut Claudius Regaud, Toulouse, France; 9Quintiles, La Jolla, CA;
10
Amgen Inc, Thousand Oaks, CA; 11Amgen Inc, Thousand Oaks, CA and 12University Hospital, Leuven, Belgium.
Body: Background: The angiopoietin axis is distinct and critical for angiogenesis. Trebananib suppresses tumor angiogenesis
by binding to angiopoietin-1 and -2 (Ang1/2), thereby inhibiting their interaction with the Tie2 receptor. This interim analysis
evaluated the tolerability and efficacy of trebananib plus paclitaxel and trastuzumab or capecitabine and lapatinib in HER2+
locally recurrent or metastatic breast cancer (MBC).
Methods: Patients (pts) in cohorts A1 and A3 (no prior 1st-line trastuzumab, lapatinib, or chemotherapy for MBC) received
trebananib (A1, 10 mg/kg; A3, 30 mg/kg) IV QW plus paclitaxel 80 mg/m2 IV QW and trastuzumab 8 mg/kg loading dose, then 6
mg/kg Q3W. Pts in cohorts B1 and B3 (history of failed 1st-line MBC treatment, no prior lapatinib or capecitabine) received
trebananib (B1, 10 mg/kg; B3, 30 mg/kg) IV QW plus capecitabine 1000 mg/m2 PO Q12 h, days 1-14 Q21D and lapatinib 1250
mg PO QD. Cohorts were expanded to n = 20 if 1 of 6 or 2 of 9 pts had dose-limiting toxicities (DLTs). Cohort B3 was closed
early due to poor enrollment. Endpoints were treatment-emergent adverse events (AEs) and DLTs (primary); and efficacy,
pharmacokinetics (PK), and incidence of anti-trebananib antibodies (secondary).
Results: All pts received 1 dose of trebananib. Across all cohorts, two DLTs occurred (A1: grade 3 transient ischemic attack, n =
1; A3: grade 3 increased gamma-glutamyltransferase, n = 1). Across A1 and A3, AEs >50% were peripheral edema, diarrhea,
alopecia, fatigue, nausea, nail disorder, and rash; AEs grade 3 in >10% of pts were peripheral/sensory neuropathy and dyspnea.
No pt in A1 or A3 died during treatment. Across B1 and B3, AEs >50% were diarrhea, nausea, palmar-plantar erythrodysesthesia
syndrome (PPES), and peripheral edema; AEs grade 3 in >10% of pts were diarrhea, PPES, and neutropenia. One pt in B1 died
from sepsis considered to be possibly related to administration of trebananib and capecitabine. Peripheral edema was a
frequently reported AE for cohorts A1, A3, B1, and B3 (n = 13, 15, 10, 3); most were grade 1 and 2 and appeared to be
manageable. No AEs of gastrointestinal perforation, ascites, or proteinuria were reported. Efficacy endpoints are summarized in
Table 1. No apparent PK drug-drug interaction was observed. No pt developed neutralizing binding antibodies during treatment.
Table 1
1ST LINE-TREATMENT
2ND-LINE TREATMENT
A1 (n = 20)
A3 (n = 20)
B1 (n = 20)
20
17
14
Complete response
Partial response
16
12
Stable disease
Progressive disease
Not assessed
Objective response rate, %
80
88
50
TUMOR RESPONSE
Evaluable pts, n
Best response, n
PROGRESSION-FREE SURVIVAL
Median, months
14.5
18.5
8.6
95% CI
6.920.6
10.421.9
3.714.4
Evaluable pts, n
16
15
Median, months
12.6
16.6
8.3
95% CI
4.320.2
8.2not estimable
DURATION OF RESPONSE
Tumor
response was assessed only in pts with measurable disease at baseline

the small sample size (n = 5)
Efficacy
4.116.8
results for B3 are not reported due to
Conclusion: Interim results from this phase 1b study of pts with HER2+ locally recurrent or MBC suggest that adding trebananib
to paclitaxel and trastuzumab or capecitabine and lapatinib is tolerable and may improve antitumor activity.

Title: Prescribing preferences of US-based medical oncology physicians for patients with hormone receptor positive, HER2
negative metastatic breast cancer following prior endocrine therapies
Mary E Cianfrocca1, Arden D Buettner2, Susan L Britton2, Maria L Lankford2 and Mark R Green2. 1Banner MD Anderson Cancer
Center, Gilbert, AZ and 2Xcenda, Palm Harbor, FL.
Body: Background: Non-steroidal aromatase inhibitors (NSAI) are a standard first-line therapy (Rx) for post-menopausal pts
with hormone receptor positive (HR+), HER2 negative (HER2-) advanced and recurrent metastatic breast cancer (mBC). When
studied in phase III clinical trials, combination hormonal Rx regimens have not consistently produced significant improvements in
time to event measures compared to sequential single agents. However recent phase II and III studies combining hormonal and
targeted agents have shown significant extensions in PFS and potential increases in OS. Due to the emergence of several
targeted agents, standards of care for treatment after failure on a NSAI are evolving.
Methods: Prescribing preferences (PPrefs) of 187 U.S.-based medical oncology physicians (MOPs) were studied using a
validated, proprietary, live, case-based market research tool (Challenging Cases). Data were acquired using blinded,
audience-response iPad technology and acquisition events took place in March and May 2014. A core case was constructed and
PPrefs for three 1st, three 2nd, and one 3rd failure scenario were assessed. Core case: 60-year-old female; History of mild
hypertension well-controlled by medication; diagnosed with stage 2B HR+, HER2-, grade 3, invasive ductal carcinoma; Primary
Rx included lumpectomy+ sentinel node biopsy/axillary dissection, chemotherapy (CT) and radiotherapy (RT) followed by
anastrozole with plans for 5 years (yrs) of Rx. Question posed: What systemic Rx would you recommend now?
Results:
Figure 1: PPrefs for Rx for first recurrent disease during or after 5 yrs of adjuvant anastrozole
Scenario
Time to
Recurrence(REC) Met
sites
Endocrine
Rx
Another
NSAI
Fulvestrant Exemestane
(F)
(EXE)
Targeted
Strategy
CT
Other
EXE +
Everolimus
(EVE)
Single
agent(SA)/
combo
Other/ Clinical Trial

(CLT)/Continue NSAI
and add F
18 mths while on NSAI

S1a,
Non-Visceral (N-VIS) and
N=187
Visceral (VIS)
10%
28%
6%
35%
14%
7%
S1b, 18 mths while on NSAI

N=97 N-VIS, Unproven VIS
7%
46%
5%
28%
7%
6%
S2, 2 yrs post 5 yrs NSAI

N=187 N-VIS
25%
35%
18%
17%
3%
5%
Figure 2: PPrefs for next Rx following adjuvant anastrozole and then letrozole (LET) at first failure.
Scenario
Time to REC after TX with LET
Met sites
S3, N=187 5 mths N-VIS
Endocrine
Rx
TS
CT
Other
EXE
EXE +
EVE
SA/Combo
CLT/RT to rib + no Rx change,

Continue current NSAI + add F
42%
6%
37%
7%
9%
46%
6%
36%
5%
9%
65%
5%
21%
3%
6%
EXE
EXE +
EVE
SA/Combo
Other/CLT
N/O
3%
40%
55%
2%
Time to REC on F
S6, N= 186 6 mths N-VIS, VIS
N/O - not offered
Conclusion:
MOPs PPrefs for management of pts with ER+, HER2- mBC are dictated by time to REC, number of RECs, and
presence/absence of VIS disease. EXE and EVE has substantial traction in all scenarios studied. Since these PPref data were
acquired, OS findings from the BOLERO-2 trial of EXE alone or EXE + EVE have been presented. PPrefs using these case
scenarios will be studied at 2 additional events prior to SABCS, allowing more robust data and insights into the early impact of the
BOLERO-2 survival data on PPrefs to be available at the meeting.

Title: Effect of palbociclib concentration on heart rate-corrected QT interval in patients with cancer
Jenny Zheng1, Michael Amantea1 and Diane Wang1. 1Pfizer, San Diego, CA.
Body: Background: Palbociclib is a selective cyclin-dependent kinase 4/6 inhibitor that blocks G1/S cell cycle progression. The
current population analysis assessed the effect of palbociclib exposure on QT interval and heart rate (via evaluation of RR
interval) in cancer patients.
Methods: Plasma palbociclib concentration (C) and electrocardiogram (ECG) data were pooled from 3 studies of once-daily oral
palbociclib: (Study 1) phase 1 dose-escalation study in advanced cancer (Schedule 1: 25, 50, 75, 100, 125, and 150 mg on a 3
wk on/1 wk off cycle [3/1]; Schedule 2: 100, 125, 200, and 225 mg [2/1]); (Study 2) phase 2 study in mantle cell lymphoma (125
mg [3/1]); (Study 3) phase 1/2 study in advanced breast cancer (Cycle 1: 125 mg [2/1]; Cycle 2+: 125 mg [3/1] + letrozole 2.5 mg
daily). Blood samples were collected predose, after first palbociclib dose, and at steady state, including the anticipated time of
maximal palbociclib concentration. Triplicate (2 min apart) ECGs were obtained at baseline; 3 h postdose (Study 1); predose on
Day (D) 1 and D21 (Study 2); predose on D1 and predose, 2, 4, 8, 24, 48, and 96 h postdose on D14 of Cycle 1, predose and 4 h
postdose on D1 and D14 of Cycle 2, and end of treatment (Study 3, phase 1); and predose on D1 of Cycle 1&3, D14 of Cycle
1&2, and end of treatment (Study 3, phase 2). Baseline singlet ECG data (from time closest to first palbociclib dose) were used to
estimate a study-specific heart rate correction factor [], using a linear mixed effect model of log(QT) vs log(RR/1000) with
intersubject variability only on intercept. The averaged triplicate ECG data with time-matched palbociclib concentration data were
used to explore RRC and corrected QT interval (QTc)C relationships. A linear mixed effects model was used to assess RRC
and QTcC with intersubject variability on both intercept and slope; sex was tested as a covariate on QTc interval intercept.
Results: 184 patients supplied 569 matched pharmacokinetic and ECG assessments. Estimated values for Studies 1, 2, and 3
were 0.367, 0.369, and 0.363, respectively. Compared with QT correction by Fridericia (QTcF) and Bazett (QTcB) methods, the
study-specific correction (QTcS) best minimized the apparent QTRR correlation and therefore was selected for use in
subsequent analyses. Palbociclib had no effect on RR interval in RRC analysis but increased QTcS in a
concentration-dependent manner. The average (90% CI) QTcS increase at the mean and median maximum steady-state
concentrations for patients on palbociclib 125 mg (Cmax,ss = 107 and 112 ng/mL, respectively [Study 3]) were 5.60 (2.488.72) and
5.88 (2.619.16) ms. Sex was not a significant covariate for intercept by analysis of variance. A similar effect of palbociclib on
QTcF was observed.
Conclusion: In a pooled population analysis of patients with cancer, palbociclib had no concentration-dependent effect on heart
rate. There was a slight positive linear relationship between palbociclib concentration and QTcS; however, the upper bound of the
1-sided 90% CI for the increase in QTcS at Cmax,ss did not exceed the threshold of 10 ms. Therefore, QT prolongation is not a
major safety concern for palbociclib at the 125-ng/mL recommended therapeutic dose.

Title: Final results of the phase I "HIT" study: A multicenter phase I-II study evaluating trastuzumab administered by intrathecal
injection for leptomeningeal meningitis of HER2+ metastatic breast cancer (MBC)
Maya Gutierrez1, Emmanuelle Mouret Fourme1, Emilie Le Rhun2, Olivier Tredan3, Veronique Dieras1, Patricia Tresca1, Fawzia
Mefti1, Isabelle Turbiez1, Sophie Taillibert1, Cline Desvignes4 and Gilles Paintaud4. 1Curie, Saint Cloud, Ile de France, France;
2
Oscar Lambret, Lille, Picardie, France; 3Leon Berard, Lyon, France and 4Tours Hospital, Tours, France.
Body: Metastatic breast cancer (MBC) is the leading non hematologic cause of leptomeningeal metastases (LM), associated with
a poor prognosis. Amongst breast cancer (BC) subtypes, HER2+ tumors present a high LM incidence. Intravenous (IV)
trastuzumab (T) has improved survival of MBC patients but better control of central nervous system (CNS) disease is needed.
The main physiopathological hypothesis for intracranial recurrences is the low level of T in cerebrospinal fluid (CSF) due to his
high molecular weight (148kD). Intraventricular (via Ommaya port) or intrathecal (IT) T administration would permit LM
progression control through high therapeutic T concentrations in the CSF. This prospective trial, sponsored by Institut Curie, was
the first Phase I-II study to investigate the safety and efficacy of IT T administration for HER2+ LM in BC. The final results of the
Phase I cohort are reported herein. Methods: LM diagnosis was based on either CSF with HER2+ cytology or clinical/MRI
imagings of meningitis. Adequate organs functions for all patients were required. The T IT (or via Ommaya port) injections were
given on weekly basis during 8 weeks. We used a Fibonacci dose escalation design, 4 dose levels (DL) (30-150mg). The
objectives of this phase I were to investigate the maximum tolerated dose (MTD) and the recommend dose (DR) of T. A
pharmacokinetic (PK) data analysis was performed. We targeted a concentration in the CSF close to the residual serum
concentration (30g/mL) achieved with standard IV schedule.
Results: Starting May 2011, 19 patients were included, with 16 evaluable for toxicity (three had not received the treatment)
treated in 4 DL (30-150mg). Twelve patients were evaluable for PK and 210 samples T concentrations available (103 in the LCR
and 107 in the serum). The MTD was not reached and the treatment appears to have been well tolerated by the patients. Neither
Grade 3 toxicity nor neurological toxicity related to IT T was observed. The T target-concentration in the CSF was reached at
DL4 and the recommended T dose for the Phase II trial is 150 mg. Five patients have experimented a evident clinical benefit and
received more than 8 weekly injections with an average of 23 [12-40]. Extended data on clinical outcome and PK will be
presented. Conclusion: Intraventricular or IT injections appears to be safe at the dose of 150 mg and possibly effective. The
Phase II study will investigate the efficacy of the recommended dose on neurological progression-free survival at 2 months in 19
patients. This trial was performed with Roche funding.


Title: Impact of adding palbociclib to letrozole on pain severity and pain interference with various activities of daily life in patients
with ER+, HER2- metastatic breast cancer as first line treatment
Timothy Bell1, John Paul Crown2, Istvan Lang3, Helen Bhattacharyya1, Giovanni Zanotti1, Sophia Randolph1, Sindy Kim1, Xin
Huang1, Cynthia Huang Bartlett1, Richard Finn4 and Dennis Slamon4. 1Pfizer, New York, NY; 2St Vincent's University Hospital,
Dublin, Ireland; 3Orszagos Onkologiai Intezet, Kemoterapia B, Budapest, Hungary and 4UCLA School of Medicine, Division of
Hematology / Oncology, Los Angeles, CA.
Body: Background
Palbociclib, a selective inhibitor of CDK-4/6, prevents DNA synthesis by blocking cell cycle progression. A randomized Phase 2
study of palbociclib (P) 125 mg QD for 3 weeks followed by 1 week off plus letrozole (L) 2.5 mg QD continuously compared to L
alone was conducted. A secondary objective of the study was to assess the impact of adding P in combination with L compared
to L alone on pain severity and pain interference with various activities of daily life.
Methods
This Phase 2 trial was designed to evaluate P+L in front-line ER+/HER2- metastatic breast cancer (MBC) compared to L alone.
The primary endpoint was investigator assessed progression-free survival (PFS) defined as time from randomization to objective
progression or death. Patient reported outcomes of pain severity (PS) and pain interference (PI) with various activities of daily life
were assessed using the Brief Pain Inventory (BPI) at baseline and day 1 of each cycle thereafter. The PS and PI scores were
summarized by cycle using observed values as well as changes from baseline.
Results
The final analysis of primary endpoint showed a statistically significant improvement in PFS for the P+L arm (20.2 months) vs. L
arm (10.2 months) with hazard ratio (HR)=0.488 (95% CI: 0.319, 0.748) and 1-sided p=0.0004. The most common adverse
events in the P+L arm were neutropenia, leukopenia, fatigue, and anemia.
For the PS scale, patients in the P+L arm showed numerically lower scores and greater reductions from baseline than the L arm,
until the later cycles, when the number of patients had decreased to a small number. The difference between treatment arms in
the mean change from baseline was statistically significant at some of the earlier cycles (Cycles 5, 6, 7, 8, 10, 12; p<0.05; no
adjustments were made for multiplicity) representing a numerically greater decrease in the pain experienced by patients in the
P+L arm compared with the L arm.
For the PI score, observed mean scores for both treatment arms appeared to be stable over time until the later cycles, when the
number of patients had decreased to a small number. Patients in the P+L treatment arm also generally showed a consistently
greater numeric reduction from baseline in pain interference (ie, a decrease in PI on daily activities) until later cycles although
without reaching statistical significance at any cycle.
Conclusions
The addition of palbociclib to letrozole was associated with increased efficacy without negatively impacting pain severity or pain
interference on daily activities.

Title: A phase I study of MK-2206 in combination with lapatinib in patients with advanced solid tumors followed by
dose-expansion in advanced HER2+ breast cancer
Moniba Nazeef1, Amye J Tevaarwerk1, Jens Eickhoff1, Mark E Burkard1, Jennifer Heideman1, Glenn Liu1, Chris Flynn1, Jill M
Kolesar1 and Kari B Wisinski1. 1University of Wisconsin Carbone Cancer Center, Madison, WI.
Body: Background : AKT mediates signaling in the human epidermal growth factor receptor-2 (HER2) pathway. HER2 inhibition
can result in feedback regulation of signaling, leading to high AKT activity. Preclinical studies demonstrate activity of combined
HER2 and AKT inhibition. MK-2206 is an oral selective inhibitor of AKT. This study was designed to determine the maximum
tolerated dose (MTD), dose limiting toxicities (DLTs), adverse events (AEs), clinical activity, pharmacokinetic (PK) parameters
and explore potential biomarkers of the combination of MK-2206 with lapatinib.
Methods : The dose escalation cohort included adult patients (pts) with solid tumors treated with MK-2206 (30-60 mg qod) and
lapatinib (1000-1500 mg qd) continuously. Cycles were 28 days, except cycle 1 (35 days), due to a 1 week MK-2206 lead-in to
evaluate for PK interactions. MK-2206 plasma concentrations were evaluated on day 1, 9 (steady state) and 15 (steady state in
combination with lapatinib). Lapatinib plasma concentrations were evaluated on day 9 (first dose) and day 15 (steady state in
combination with MK-2206). Both used a validated LC-MSMS assay. Pharmacokinetic parameters were calculated using
non-compartmental methods with WinNonLin Phoenix version. The dose expansion cohort included women with advanced
HER2+ breast cancer treated at the MTD. Peripheral blood mononuclear cells (PBMCs) and archived tumor samples were
collected for all pts.
Results : In the dose escalation cohort, 23 pts (median age 59 [range 22-72];15 female:8 male) were enrolled. Cancers were
colorectal (8 pts), lung (4 pts), breast (3 pts) and other (8 pts). 19 evaluable pts were treated a median of 8 weeks (range 3-35).
At dose level four, 1 pt had grade 4 hyponatremia, grade 3 rash and hypocalcemia and 1 pt had intolerable grade 2 mucositis with
delivery of <75% of drug. The MTD was 45mg po qod of MK-2206 with 1500 mg po qd of lapatinib. The most common AEs at
least possibly related to therapy included diarrhea (grade 3-4 in 3 pts; grade 1-2 in 16 pts), nausea (grade 3 in 2 pts; grade 1-2 in
14 pts) and rash (grade 3 in 2 pts; grade 1-2 in 12 patients). PK analyses demonstrated lapatinib half-life was approximately 18
hours on both day 9 and 15. MK-2206 half-life was 65 119 hours on day 1, 104 84 on day 9 and 59.125 on day 15. Dose
adjusted MK-2206 AUC (hr*ng/mL/mg) was 26.6 31 of day 1, 100.6 78 on day 9 and 62.8 58 on day 15. In the dose
expansion cohort, five HER2+ women were enrolled (median age 43 yrs [range 33-56]). The majority were heavily pretreated: all
had prior progression on trastuzumab and 2 with prior lapatinib. They were on study for a median of 8 weeks (range 4-24 weeks).
One pt had stable disease for 6 months and another pt had clinical benefit with non-measurable skin disease and she remains on
study. Exploration of PBMCs for evidence of target inhibition in HER2-PI3K-AKT signaling and tumor tissue for PTEN loss, PI3K
mutations will be presented.
Conclusions : Continuous dosing of MK-2206 in combination with lapatinib is well-tolerated. Preliminary results of
pharmacokinetic analysis indicate a potential for a drug interaction. Clinical benefit was seen in patients with HER2+ breast
cancer.

Title: TRASTYVERE study: A retrospective analysis of HER2-positive metastatic breast cancer (MBC) patients treated in Spain
with lapatinib (L) plus trastuzumab (T)
Joaquin Gavil1, Begoa Bermejo2, lvaro Rodrguez-Lescure3, Juan Lao Romera4, Luis Manso5, Joan Brunet6, Eva Muoz7,
Marta Santisteban8, Csar A Rodrguez9, Ana Santaballa10, Juan de la Haba11, Pedro Snchez-Rovira12, Manuel Ruiz-Borrego13,
Jose ngel Garca-Saenz14, Javier Corts15 and Antonio Llombart15. 1Instituto Valenciano de Oncologa, Valencia, Spain;
2
Hospital Clnico Universitario de Valencia, Valencia, Spain; 3Hospital General Universitario de Elche, Elche, Alicante, Spain;
4
Hospital Universitario Miguel Servet, Zaragoza, Spain; 5Hospital 12 de Octubre, Madrid, Spain; 6Institut Catal d'Oncologia,
Girona, Spain; 7Vall d'Hebron Institute of Oncology (VHIO), Barcelona, Spain; 8Clnica Universidad de Navarra, Pamplona, Spain;
9
Hospital Universitario de Salamanca-IBSAL, Salamanca, Spain; 10Hospital Universitario la Fe de Valencia, Valencia, Spain;
11
Hospital Universitario Reina Sofa de Crdoba, Crdoba, Spain; 12Complejo Hospitalario de Jan, Jan, Spain; 13Hospital
Universitario Virgen del Roco, Sevilla, Spain; 14Hospital Clnico San Carlos, Madrid, Spain and 15Medica Scientia Innovation
Research, Barcelona, Spain.
Body: Background: Vertical dual blockade with L and T in heavily pretreated HER2+ MBC patients has shown consistent survival
gain in a phase III trial (Blackwell KL et al. 2012), justifying an EMA approval for the hormone-negative subgroup. However, there
is very limited information about the futility of the combination in clinical practice, mostly in patients progressing also on prior L
regimens.
Methods: We conducted a retrospective analysis among patients treated in Spain by compassionate uses for the combination of
T-L. The study was approved by the regulatory authorities and ethics committees from the 14 participating centers. Major
inclusion criteria were (1) HER2+ MBC; (2) progression on at least one prior line of T for advanced disease; and (3) T-L treatment
started between JAN/2005 and DEC/2012. Concomitant endocrine therapy for HR+ patients as well as prior exposure to L was
allowed. Chemotherapy combinations were excluded. A total of 111 patients were predefined for the primary outcome: clinical
benefit rate (CBR). Secondary endpoints included time to progression (TTP), overall survival (OS) and toxicity. 114 women were
included and externally monitored.
Results: The median age was 60 years (34 - 89); 64% HR+; 77% visceral disease; 32% CNS disease (37 patients); 47% with 3
organs involved. Mean number of prior T lines 4 (range 0-13); 64% previously treated with L. A total of 40 patients (35%)
achieved a CBR (95%CI 2644%); 6 CR, 19 PR and 15 SD lasting >24 weeks. The median time to progression was 3.8 months
(95%CI 3.35.1) and the median overall survival 21.6 months (95%CI 17.127.3). CBR, median TTP and median OS achieved in
patients with CNS disease were 32.4% (95%CI 17.347.5%), 3.6 (95%CI 2.85.9) and 15.4 (95%CI 10.927.3) months,
respectively.
The CBR was independent of L treatment (41.5% L nave vs. 31.5% L pretreated, p=0.285) and HR status (39% HR- vs. 32.9%
HR+, p=0.509). Patients with <3 metastatic sites showed higher CBR than patients with 3 (45 vs. 24.1%, respectively, p=0.019).
No significant trends were observed in any pre-specified condition for TTP and OS. Grade 3/4 toxicities were reported in 20
patients (17.5%). Only 2 patients report asymptomatic cardiac toxicities.
Conclusions: The combination of T-L seems safe and active in heavily pretreated patients. The combination remains active
among patients progressing on prior L. Future research may focus on the ability of endocrine therapy to increase activity among
HER2+/HR+ patients.
REFERENCES:
1 Blackwell KL, Burstein HJ, Storniolo AM, et al: Overall survival benefit with lapatinib in combination with trastuzumab for
patients with human epidermal growth factor receptor 2-positive metastatic breast cancer: final results from the EGF104900
Study. J Clin Oncol 2012;30(21):2585-2592.

Title: A phase I clinical trial of entinostat and lapatinib in patients with trastuzumab refractory HER2-positive metastatic breast
cancer
Naoto T Ueno1, Summer A Jackson1, Fahad A Faruqi1, Jie Willey1, Ricardo H Alvarez1, Daniel J Booser1, Sharon H Giordano1,
Ana M Gonzalez-Angulo1, James L Murray1, Vincente Valero1 and Jangsoon Lee1. 1MD Anderson, Houston, TX.
Body: Background: Our preclinical data show that entinostat enhances the efficacy of lapatinib in HER2 positive (HER2+) breast
cancer cell lines via FOX-03-mediated Bim1 expression. In-vitro and in- vivo, the combination of entinostat and lapatinib
enhanced apoptosis in lapatinib-resistant cells. We conducted a phase I study with the primary objective to determine the RP2D
of entinostat plus lapatinib in patients with HER2+ metastatic breast cancer with progressive disease (PD) after trastuzumab
treatment.
Methods: This was a single-center, open-label study to evaluate the dose limiting toxicity (DLT) and determine the MTD of
every-other-week entinostat plus daily lapatinib in 28-day cycles. Patients with locally recurrent or metastatic breast cancer in
whom trastuzumab had failed were enrolled. DLT was defined as: any toxicity resulting in 14 or more days of treatment delay,
febrile neutropenia (NTP), grade 4 NTP for over 7 days, any grade 3 or higher non-hematologic toxicity, grade 3 nausea,
vomiting, diarrhea or electrolyte imbalance lasting over 48 hours despite adequate supportive care, or a platelet count less than
10,000. Grading was assigned according to common toxicity criteria (CTC 4). A standard 3+3 dose escalation design was used.
Entinostat was given at 10 mg (level 0), 12 mg (level 1), or 15 mg (level 2) by mouth every 14 days. Lapatinib was given at 1250
mg by mouth daily. Toxicity was evaluated on day 15 and day 28 for C1 and C2, and at the end of each cycle thereafter.
Results: Fifteen patients with HER2+ metastatic breast cancer were enrolled. Median age was 52 years (range 26-69), median of
all prior systemic treatment was 4 (range 1- 12), and median prior trastuzumab-based regimens was 2 (range 1-6). Eight patients
had prior lapatinib exposure. Seven patients had ER+/HER2+ tumors. Median number of cycles completed was 2 (range 1-13).
DLT was observed in 0 of 3 patients at level 0, 1 of 6 at level 1, and 1 of 6 at level 2. 3 patients at level 1 had lapatinib dose
reductions because of grade 3 rash (n=2) or grade 3 dyspnea (n=1) or grade 3 abdominal pain (n=1). 2 patients at level 2 had
entinostat dose reductions because of grade 4 NTP in cycle 1 (n=1) or grade 3 NTP in cycle 5 and grade 4 neutropenia in cycle 6
(n=1). Five patients had SD (defined per 2009 RECIST guidelines): two at dose level 1 for 6 months and 13 months, three at level
2 for 3, 4, and 8 months. Median time to progression was 2 months (range 1-13). The most common treatment related adverse
events were fatigue (n=15), myalgia (n=14), nausea (n=14), diarrhea (n=13), anemia (n=11), and rash (n=11).
Conclusion: MTD was not reached. Cumulative toxicity of entinostat plus lapatinib was fairly well tolerated. The combination
therapy suggests there was clinical activity in at-least 5 patients who had SD in this cohort of heavily pretreated
trastuzumab-resistant MBC. Data from the EG104900 clinical trial showed that lapatinib plus trastuzumab significantly improved
median overall survival compared with lapatinib alone. Therefore, this study was modified after 15 patients to add trastuzumab to
the combination. We are currently conducting a phase 1b trial to determine the MTD of entinostat, trastuzumab and lapatinib.

Title: A phase I clinical trial of ganetespib (heat shock protein 90 inhibitor) in combination with paclitaxel and trastuzumab in
human epidermal growth factor receptor-2 positive (HER2+) metastatic breast cancer
Komal Jhaveri1, Karen Cadoo2, Sarat Chandarlapaty2, Eleonora Teplinsky1, James Speyer1, Gabriella D' Andrea2, Sujata Patil2,
Sofia Haque2, Kent Friedman1, Scott Heese1, Deirdre Neville2, Francisco Esteva1, Clifford Hudis2 and Shanu Modi2. 1Laura and
Isaac Perlmutter Cancer Center at NYU Langone Medical Center, New York, NY and 2Memorial Sloan Kettering Cancer Center,
New York, NY.
Body: Introduction: Targeted therapies in HER2+ metastatic breast cancer (MBC) have significantly improved survival, however
efficacy is limited by development of therapeutic resistance. HSP90 is a molecular chaperone involved in the stability and function
of multiple signaling onco-proteins. HER2 is an acutely sensitive HSP90 client and HSP90 inhibition can overcome trastuzumab
resistance. Ganetespib is a novel, synthetic HSP90 inhibitor with increased potency and tolerability compared with earlier agents.
Our group has conducted a single agent ganetespib trial in unselected patients which showed anti-tumor activity in HER2+ and
triple negative breast cancer. In addition, preclinical data suggests HSP90 inhibition is synergistic with taxanes with potential for
significant clinical activity. Ganetespib has been combined with docetaxel in non-small cell lung cancer, it has not previously been
combined with paclitaxel and trastuzumab. This study will define the maximum tolerated dose (MTD) and recommended phase II
dose (RP2D) of ganetespib when given with paclitaxel and trastuzumab for patients with HER2+ MBC.
Methods: In this 3+3 phase I dose escalation study, patients with trastuzumab resistant HER2+ MBC receive trastuzumab
(4mg/kg loading dose, then 2mg/kg) and paclitaxel weekly (80mg/m2) with ganetespib day 1, 8 ,15 of 28 day cycle. Patients are
required to have prior pertuzumab and T-DM1 (prior pertuzumab and T-DM1 are not mandated if heavily pretreated prior to their
respective FDA approvals). Hormone receptor positive patients are required to have at least one prior line of endocrine therapy.
The single agent dose limiting toxicity (DLT) of ganetespib is diarrhea and therefore patients receive prophylactic anti-motility
agents. The anticipated MTD of ganetespib in this combination has been informed by experience with docetaxel and based on
this only three dose levels of ganetespib are being explored 100mg/m2, 150mg/m2 and a third intermediate cohort of 125mg/m2,
if needed. Secondary endpoints include evaluation of effects of ganetespib on the pharmacokinetics of paclitaxel and preliminary
assessment of efficacy of the combination (scans at 8 weeks and every 12 weeks thereafter, RECIST 1.1).
Results: The first dosing cohort has fully enrolled and there were no significant toxicities or DLTs reported. Median age was 48
years (range 39-49), median prior lines of chemotherapy were 4 (range 3-7) and included prior pertuzumab and T-DM1 in all 3
patients. 5 adverse events have been defined as possibly/probably related to ganetespib grade 2 anemia and leukopenia, grade
1 diarrhea (2 patients), fatigue, and rash. Enrollment to the second and potentially final cohort is underway.
Conclusion: This study will define the RP2D of ganetespib in combination with paclitaxel and trastuzumab. Final safety,
pharmacokinetic and preliminary response data for all patients will be presented. This combination, with a novel anti-HER2 agent,
has encouraging potential for activity in HER2+ breast cancer which is refractory to other HER2 targeting agents.

Title: Phase Ib/II study of LEE011 and BYL719 and letrozole in ER+, HER2 breast cancer: Safety, preliminary efficacy and
molecular analysis
Dejan Juric1, Erika Hamilton2, Laura Garcia Estvez3, Richard H De Boer4, Ingrid Mayer5, Mario Campone6, Shizuka Asano7, Suraj
Bhansali8, Vickie Zhang7, Becker Hewes7 and Pamela Munster9. 1Massachusetts General Hospital, Boston, MA; 2Sarah Cannon
Research Institute, Nashville, TN; 3Hospital de San Chinarro, Madrid, Spain; 4Royal Melbourne Hospital, Victoria, Australia;
5
Vanderbilt-Ingram Cancer Center, Nashville, TN; 6Institut de Cancrologie de l'Ouest/Ren Gauducheau, Nantes, France;
7
Novartis Institutes for BioMedical Research, Cambridge, MA; 8Novartis Pharmaceuticals Corporation, East Hanover, NJ and
9
University of California, San Francisco, CA.
Body: Background: Activation of the PI3K/AKT/mTOR and cyclin DCDK4/6INK4Rb pathways, including through alteration of
PIK3CA and CCND1, has been implicated in resistance to aromatase inhibitors. BYL719 (BYL), an -isoform selective PI3K
inhibitor, demonstrated clinical activity as a single agent and in combination with hormone therapy in pts with advanced HR+
breast cancer (BC), while the CDK4/6 inhibitor LEE011 (LEE) showed evidence of clinical activity as a single agent in pts with
advanced solid tumors and in combination with letrozole (LET) in pts with heavily pretreated advanced ER+ BC. In ER+ BC
models, the triple combination of LEE, BYL, and LET had enhanced activity vs each agent alone. A Ph Ib/II, 3-arm study is
currently investigating the combination of LEE, BYL, and LET in pts with ER+ BC. Here we report on safety, preliminary efficacy,
and molecular analysis from Arm (A)1 (LEE + LET) and A2 (BYL + LET) of the Ph Ib part of the study.
Methods: Postmenopausal pts with advanced ER+, HER2 BC receive daily oral LEE (3-wks-on/1-wk-off; A1) or BYL
(continuous; A2), plus fixed, daily LET (2.5 mg, continuous) in 28-day cycles. Primary objective: determine the MTD and/or RP2D
of each combination. A Bayesian Logistic Regression Model using the escalation with overdose control principle and real-time PK
guide dose escalation. Secondary objectives: safety, PK, and preliminary efficacy. Potential biomarkers that are predictive of
response are also being assessed by next-generation sequencing of tumor samples.
Results: As of March 28, 2014, 10 pts received 600 mg LEE plus LET (A1), and 7 pts 300 mg BYL plus LET (A2). At study entry,
all pts had stage IV disease; number of prior endocrine regimens for advanced disease was: 01 (47%); 23 (35%); 45 (18%);
35% of pts had previously received PI3K/AKT/mTOR pathway inhibitors for advanced disease. Of the 15 pts evaluable for dose
determination (10 in A1 and 5 in A2), 1 dose-limiting toxicity was observed (A1: Grade [G]4 neutropenia; data cut-off: May 15).
Most common (all grade >30%) study drug-related AEs (all grade/G34) were: A1: neutropenia (90%/50%) and nausea
(40%/0%); A2: hyperglycemia (57%/14%), decreased appetite, diarrhea, and nausea (43%/0% each). PK of LEE and BYL on
Days 1 and 21, and LET on Day 1, are comparable with historic single-agent data. PK of LET at steady state is being evaluated.
In A1, of 6 pts with known response, 1 pt had a PR, 2 pts had SD, 1 pt without measurable disease had NCRNPD, and 2 pts had
PD. In A2, of 5 pts with known response, 2 pts had SD, and 3 pts had NCRNPD. Biomarker analysis showed that 2 pts in A2 with
SD who are still on study (1 with 25% tumor shrinkage) had PIK3CA mutations.
Conclusion: LET plus LEE or BYL had an acceptable safety profile and preliminary clinical activity in pts with ER+/HER2
advanced BC. Dose escalation continues, and upon determination of the MTD/RP2D in A1 and A2, enrollment into A3 (LEE +
BYL + LET) will begin to determine the MTD/RP2D. Genetic alterations in PIK3CA were observed in A2. Analysis of baseline
aberrations in the cyclin DCDK4/6INK4Rb pathway in A1 and A3 are ongoing and will be updated. The Ph II part of the trial
will compare LET plus LEE or BYL with the triple combination.

Title: Multi-institutional retrospective analysis of clinical and pathological factors predicting resistance to lapatinib-based therapy
in HER2 positive metastatic breast cancer (HER2+ MBC)
Stefania Gori1, Valentina Rossi2, Monica Turazza1, Elena Fiorio3, Alessandra Fabi4, Giancarlo Bisagni5, Jennifer Foglietta6,
Daniele Santini7, Ida Pavese8, Alberto Zambelli9, Patrizia Vici13, Vita Leonardi10, Sandro Barni11, Silvana Saracchini12, Giuseppe
Bogina14, Simona Duranti1, Alessandro Inno1, Gianluigi Lunardi1 and Filippo Montemurro2. 1Medical Oncology Sacro Cuore
Don Calabria Hospital, Negrar Verona, Italy; 2Medical Oncology Institute for Cancer Research and Treatment IRCCS,
Candiolo Torino, Italy; 3Medical Oncology AO Universitaria Integrata, Verona, Italy; 4Medical Oncology A Istituto
Nazionale Tumori Regina Elena, Roma, Italy; 5Medical Oncology IRCCS AO S.Maria Nuova, Reggio Emilia, Italy; 6Medical
Oncology S.Maria della Misericordia Hospital, Perugia, Italy; 7Medical Oncology Universit Campus Bio-medico, Roma,
Italy; 8Medical Oncology S. Pietro Fatebenefratelli Hospital, Roma, Italy; 9Medical Oncology IRCCS Fondazione S.
Maugeri, Pavia, Italy; 10Medical Oncology ARNAS Civico, Palermo, Italy; 11Medical Oncology Azienda Ospedaliera di
Treviglio, Treviglio Bergamo, Italy; 12Medical Oncology S.Maria degli Angeli Hospital, Pordenone, Italy; 13Medical Oncology
B Istituto Nazionale Tumori Regina Elena, Roma, Italy and 14Pathology Sacro Cuore Don Calabria Hospital, Negrar
Verona, Italy.
Body: Background
The combination of the dual HER1/HER2 inhibitor lapatinib (L) and capecitabine (C) is a therapeutic option for patients (pts) with
HER2+MBC whose disease progresses after treatment with the monoclonal antibody trastuzumab. At present time, no clinical or
pathological factors except HER2 status are clearly recognized as predictors of the activity of LC. We conducted a retrospective
analysis of pts with HER2-positive metastatic breast cancer receiving LC after trastuzumab failure to identify factors associated
with resistance to LC.
Materials and methods
We collected clinical and pathological data from 151 pts with HER2+ MBC receiving LC after failing a prior trastuzumab-based
treatment (either adjuvant or for metastatic disease) treated at 13 Italian Institutions between March 2007 and December 2013.
Time to progression (TTP) and overall survival (OS), calculated by the Kaplan Meier (KM) method, were from LC treatment
beginning to disease progression or to death in the absence of progression (TTP), and to the date of death or to the date of last
follow-up (OS), respectively. LC resistance was defined as TTP from treatment initiation lower or equal to the median TTP for the
overall population. KM curves were compared by the Log-rank test. Logistic regression analysis was used to study predictors of
TTP below the median value for patients receiving LC. Analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago,
IL).
Results
At a median follow-up of 41 months (IQR 23-62), median TTP to LC therapy was 7 months (IQR 5.5-8.5) and median OS was 18
months (IQR 10-28). Fifteen pts were excluded because of short follow-up (i.e. on LC treatment and <7 months of fu). Of the
remaining 136, a total of 74 pts with a PFS7 months were defined LC-resistant (LC-R) and a total of 62 pts were defined
LC-sensitive (LC-S). All clinical and pathological variables analyzed resulted evenly distributed between the two groups, except
best tumor response (CR+PR) to LC, which was higher in patients with LC-S disease (72% vs 29%, p<0.001). Conversely, best
tumor response in LC-R patients showed higher rates of PD (43% vs 2%, p<0.001). Median OS was 14 months (IC 95%
11.4-22.6) and 26 months (IC 95% 22.5-29.5) in LC-R and LC-S pts, respectively (p<0.001). Although we could not find
independent predictors of LC-R, factors indicating failure of the first-line trastuzumab based therapy, as PD as best tumor
response and short duration of first-line trastuzumab, were associated to LC-R.
Conclusions
A short time to progression during capecitabine and lapatinib (LC-R) is associated with reduced OS in patients failing prior
trastuzumab based therapy for HER2+ MBC. Patients who had modest clinical benefit from previous trastuzumab-based therapy
could experience LC-R indicating the possibility of primary resistance to anti HER2-treatment. For these patients, alternative
targeting strategies are urgently needed.

Title: Efficacy and tolerance of everolimus in 123 consecutive very advanced luminal breast cancer patients. A multicenter
retrospective study
Dorothe Chocteau-Bouju1,3, Camille Chakiba2, Laurent Mignot1, Nicolas Madranges2, Jean-Yves Pierga1, Philippe Beuzeboc1,
Nathalie Quenel-Tueux2, Vronique Diras1, Herv Bonnefoi2, Marc Debled2 and Paul H Cottu1. 1Institut Curie, Paris, France;
2
Institut Bergoni, Bordeaux, France and 3CHRU Bretonneau, Tours, France.
Body: Background: The on line publication in December 2011 of the Bolero-2 study (Baselga et al, NEJM) establishing the
added value of everolimus (Eve) in endocrine resistant breast cancer (BC) patients (pts) has triggered its immediate use in
current practice. We evaluated our practice 2 years (y) after the French marketing authorization (July 2012).
Patients and Methods: We retrospectively reviewed the medical charts of 123 consecutively treated pts in two Unicancer
Institutions (Institut Curie & Institut Bergoni). All pts had luminal (ER positive, HER2 negative) BC. Median age at diagnosis was
48 y (29-78) and median delays to first met event and to everolimus therapy were 6.0 y (0-25) and 12.6 y (1.3-34.8), respectively.
PS at inclusion was 0 (47.6%), 1 (50.8%) or 2 (1.6%). Pts had received a median number of 2 lines of chemotherapy (0-8) and of
2 lines of endocrine therapy (0-6) for metastatic disease. Visceral disease was present in 56% of the pts. Pts had either 1
(25.4%), 2 (32.2%) or 3+ (42.4%) involved sites.
Results: Eve based therapy. Initial dose of Eve was 10 mg (76.4%) or 5 mg (23.6%) and was combined with exemestane
(79.9%), anastrozole/letrozole (14.0%) or tamoxifen (6.1%). GnRH agonists were used in 10 premenopausal women. Dose
adjustments of Eve were necessary in 41.7% of the 10 mg pts, and in 33.3% of the 5 mg pts (p=0.426). Overall, grade 2 or grade
3 side effects were experienced by 47.7% and 34% of the pts, respectively. Most frequent side effects were grade 2/3 mucositis
(32.5%/11.2%), grade 1/2 decreased appetite (42.8%/24.3%), grade 1/2 rash (46.7%/22.7%), and grade 2/3 fatigue
(33.3%/7.1%). Grade 2/3 weight loss was observed in 27.2%/7.6% of the pts. Only 5 cases of grade 3 pneumonitis were
recorded. One patient had a grade 4 hypercholesterolemia which quickly resolved after cessation of therapy. No toxicity related
death was observed.
Response and Survival. Out of 116 evaluable/measurable pts, the best observed response was disease improvement (RECIST
and non RECIST objective response) in 47 pts (40.5%), stable disease in 28 pts (24.1%) and progressive disease in 41 pts
(35.4%). From onset of Eve based therapy and after a median follow up of 10 mo, overall survival was 21 mo (0.4-26+), median
progression free survival was 9 mo (0.4-26+), and time to treatment failure was 5.7 mo (0.4-16+). Eve was stopped for
progression, toxicity or both in 64 pts (52%), 36 pts (29.2%) and 8 pts (6.5%) respectively. Multivariate analysis showed that more
than 2 lines of previous chemotherapy was an independent predictor for PFS (HR for progression=2.28 p=0.01), and that 2 or
more involved sites was an independent predictor for OS (HR for death=2.7 p=0.021).
Conclusion: We evaluated a multicenter population in routine practice of very advanced, slowly evolving luminal BC patients,
which appears very close to the Bolero-2 population although more heavily pretreated. Eve based therapy appears feasible with
dose reduction in more than 40% of the population and side effect rates are very similar to those reported in the pivotal Bolero-2
trial. Efficacy is highly encouraging and deserves a further evaluation of everolimus in this population.

Title: IMMU-132, a new antibody-drug conjugate (ADC) against Trop-2, as a novel therapeutic for patients with
relapsed/refractory, metastatic, triple-negative breast cancer (TNBC): Results from Phase I/II clinical trial (NCT01631552)
Aditya Bardia1, Alexander Starodub2, Rebecca L Moroose3, Ingrid A Mayer4, Jennifer R Diamond5, Ellen Chuang6, Serengulam V
Govindan7, Robert M Sharkey7, Pius Maliakal7, William A Wegener7, Steven A Hamburger7, Allyson J Ocean7, David M
Goldenberg7,8 and Linda T Vahdat7. 1Massachusetts General Hospital Cancer Center, Harvard Medical School, Boston, MA;
2
Indiana University Health Center for Cancer Care, Goshen, IN; 3UF Health Cancer Center, Orlando, FL; 4Breast Cancer Program,
Vanderbilt-Ingram Cancer Center, Nashville, TN; 5University of Colorado Cancer Center, Aurora, CO; 6Weill Cornell Medical
College, New York, NY; 7Immunomedics, Inc, Morris Plains, NJ and 8Garden State Cancer Center, Center for Molecular Medicine
and Immunology, Morris Plains, NJ.
Body: Background: TNBC, comprising 15-20% of all invasive breast cancers, represents an aggressive phenotype with high risk
of recurrence and mortality. Trop-2 is a cell-surface glycoprotein expressed on many human carcinomas, including TNBC. High
Trop-2 expression is associated with more aggressive disease and poor prognosis in several cancers, including breast cancer.
We report interim results from a Phase I/II trial evaluating a novel ADC, IMMU-132 (isactuzumab govitecan), comprising a
humanized anti-Trop-2 antibody conjugated to the topoisomerase I inhibitor, SN-38 (active metabolite of irinotecan). The
drug:antibody ratio of 7.6 facilitates the delivery of high-dose chemotherapy preferentially to the tumor cells.
Methods: Patients (pts) with relapsed/refractory metastatic epithelial tumors were enrolled at escalating IMMU-132 doses (8 to
18 mg/kg), given on days 1 and 8 of a 21-day cycle. The Phase II dose at this schedule was 10 mg/kg. CT scans were performed
every 6-8 weeks to assess response using RECIST 1.1. During the dose-escalation portion, evidence of antitumor activity,
including 3 partial responses (TNBC, small-cell lung cancer and colorectal cancer) and many with durable stable disease (SD),
was observed, leading to Phase II expansion.
Results: As of Sept. 25, 2014, a total of 132 pts have been enrolled, including 30 with advanced/metastatic TNBC. Currently
evaluable TNBC pts (N=17) had a median age of 50 (33-77), with a median of 4 prior drug regimens (range 1-8), and 67% having
received prior platinum-containing regimens. In this heavily pre-treated population, there were 4 PRs (25%) and 9 SDs (56.3%)
per RECIST v1.1, representing a disease control (PR+SD > 4 mos) of 53% among evaluable pts with adequate follow-up. A
maximum shrinkage of target lesions of 33%, 44%, 51%, and 60% for pts with PRs, and 14%, 19%, and 27% for 3 pts with SD,
was determined. Biomarker CA15.3 directional changes correlated with RECIST. All but one pathology specimen were Trop-2+ by
immunohistochemistry. HPLC analysis of serum samples found <5% unbound SN-38. The half-life of IMMU-132 was 23 h, which
is similar to the predicted half-life from in vitro serum stability studies. Grade 3/4 toxicities were: neutropenia (G3, 4 pts, 23.5%)
with 1 febrile neutropenia (5.9%), and lymphocytopenia (1 Gr 3, 1 pt, 5.9%). Grade 1/2 events were fatigue (35.3%), diarrhea
(41.2%), and alopecia (29.4%). No pt discontinued therapy due to toxicity.
Conclusions: Based on laboratory and initial clinical results, IMMU-132 is an ADC that selectively delivers a topoisomerase I
inhibitor to cancer cells without the need for enrichment by a companion diagnostic. It is safe, well-tolerated, with preliminary
evidence of encouraging efficacy in heavily-pretreated pts with relapsed/refractory metastatic TNBC. Randomized Phase III and
combination trials are being planned.

Title: PTEN status dictates the roles of PI3K isoforms p110 and p110 in modulation of AKT/mTOR and response to growth
factor signaling in ER+ breast cancer
Lloye M Dillon1, Stephanie J Bouley1 and Todd W Miller1. 1Geisel School of Medicine at Dartmouth and Norris Cotton Cancer
Center, Lebanon, NH.
Body: Class 1A phosphatidylinositol 3-kinases (PI3Ks) regulate cell growth, survival, and metabolism. PI3Ks are heterodimeric
lipid kinases composed of a p85 regulatory subunit and a p110 catalytic subunit (p110, p110, or p110). p110 and p110 play
distinct roles in PI3K signaling in carcinoma cells. p110 is frequently activated by growth factor receptor kinase signaling. In
contrast, p110 was shown to play a role in insulin metabolic action, G protein-coupled receptor (GPCR) and Rac1 signal
transduction, and oncogenic transformation. Cancer cells deficient in PTEN, the tumor suppressor phosphatase that antagonizes
PI3K signaling, are often sensitized to pharmacological inhibitors of p110. As a result, early clinical studies with p110 inhibitors
are often restricted to patients with PTEN-deficient cancers. However, analysis of data from the Genomics of Drug Sensitivity in
Cancer database revealed that genetic lesions in PTEN or PIK3CA (encodes p110) were significantly and independently
associated with increased sensitivity to the p110 inhibitor AZD6284. Among 668 cancer cell lines evaluated, 61 lines had
AZD6482 IC50 values 2 mM, but only 25 lines harbored an alteration in PTEN and/or PIK3CA. Thus, a significant fraction of
cancer cell lines (and tumors) without PTEN/PIK3CA alterations are likely to be sensitive to p110 inhibition.
To explore the role of p110 in PI3K signaling in ER+ PTEN-deficient breast cancer, we treated cells with the p110 inhibitor
GSK2636771 and the p110 inhibitor BYL-719, alone or in combination, and assessed effects on steady state and growth
factor-induced activation of AKT and MEK/ERK activation, and cell growth. p110 inhibition reduced the viability of
PTEN-deficient cells; however, combined inhibition of p110 and p110 was more effective at reducing AKT and ERK
phosphorylation and increasing apoptosis in ER+ PTEN-deficient cells. Furthermore, anti-estrogen treatment potentiated the
anti-proliferative effects of PI3K inhibition. p110 inhibition reduced insulin-like growth factor 1 (IGF-1)-induced pAKT levels, and
delayed AKT phosphorylation in both PTEN-deficient and PTEN-wild-type cells. In contrast, p110 inhibition sensitized both
PTEN-wild-type and PTEN-deficient cells to heregulin stimulation, and promoted PI3K (p85)/HER3 interaction. These results
indicate that p110 inhibition desensitizes cells to IGF-1 stimulation, hypersensitizes cells to heregulin, and modulates
downstream AKT and MEK/ERK activation in response to growth factor receptor activation. Our findings suggest that the
anti-tumor efficacy of p110 inhibitors may be related to growth factor dependence and PTEN status.

Title: Understanding pharmacodynamics and consequences of PI3K inhibition in ER+ breast tumors
Wei Yang1, Jennifer R Bean1, Lloye Dillon1, Laurent Salphati2, Jodie Pang2, Xiaolin Zhang2, Michelle Nannini Pepe2 and Todd W
Miller1. 1Norris Cotton Cancer Center, Geisel School of Medicine at Dartmouth, Lebanon, NH and 2Genentech, San Francisco,
CA.
Body: PI3K inhibitors have shown promise for the treatment of anti-estrogen-resistant breast cancers. Current PI3K inhibitor
treatment regimens incompletely and transiently inhibit the pathway in carcinomas, and are accompanied by adverse effects in
patients. We found that different periods of PI3K inhibition (12, 24, 36 h) potentiated anti-estrogen-induced apoptosis and
inhibition of proliferation to similar extents in cultured ER+ cells. We thus hypothesized that short-term, complete inhibition of PI3K
will have a greater anti-tumor effect and reduced systemic toxicity than chronic, partial inhibition.
Pharmacokinetic analysis of the orally available pan-PI3K inhibitor GDC-0941 at low (100 mg/kg) and high (800 mg/kg) doses in
mice revealed that plasma levels peaked after 15-30 min. (18.6 uM and 20.7 uM, respectively), and decreased to a plateau phase
after 1 h that was maintained for 8 h with low dose (6.8-10.7 uM) and 23 h with high dose (7.9-15 uM). We performed MCF-7
tumor pharmacokinetic analyses with low and high doses, and with 2 low doses administered 12 hours apart. Tumor GDC-0941
levels peaked after 9 h (1.6 uM with low-dose; 16.9 uM with high-dose). The second low dose increased tumor drug
concentrations to 3.2 uM at 9 h after the second dose, compared to 1.6 uM at 9 h after the first dose. After 48 h, tumor drug
concentrations decreased to 0 uM with low dose, and to 4.5 uM with high dose.
Mice bearing MCF-7 tumors were treated with fulvestrant (5 mg/wk). Three days later, GDC-0941 was administered to assess
pharmacodynamic effects. Phospho-AKT and -S6 levels (markers of PI3K and mTORC1 activities, respectively) were maximally
suppressed after 1 h and 3 h of high- and low-dose treatments, respectively, returned to baseline within 16 h after low-dose
treatment, and remained suppressed for 36 h following high-dose treatment. PARP cleavage (marker of apoptosis) occurred
within 1 h and 3 h of high- and low-dose treatments, and increased over time. Re-treatment of mice with low-dose GDC-0941
after 12 h induced continued inhibition of PI3K and mTORC1 for 9-12 h, suggesting that BID low-dose treatment may be sufficient
to continually inhibit PI3K. Comparison of high-dose and low-dose BID tumors showed that these treatments induced similar
amounts of PI3K inhibition and PARP cleavage at 21-24 h.
Mice bearing MCF-7 or fulvestrant-resistant T47D/FR tumors were treated with vehicle, fulvestrant, GDC-0941 (100 mg/kg QD 5
d/wk; 100 mg/kg BID 3 d/wk, 800 mg/kg QW), or combinations of fulvestrant and GDC-0941. Drug combinations induced tumor
regression, fulvestrant did not affect tumor growth, high-dose GDC-0941 QW slowed tumor growth, and low-dose GDC-0941 QD
or BID appreciably inhibited tumor growth. However, there was no significant difference among doses and schedules of
GDC-0941 in the context of a fulvestrant backbone in either tumor model. These data suggest that transient/metronomic (QD,
BID) and chronic/infrequent (QW) PI3K inhibition may provide similar anti-tumor efficacy in combination with an anti-estrogen.
However, these tumor growth data conflict with cell fate data indicating that high-dose GDC-0941 induced much more apoptosis
than low-dose GDC-0941. Ongoing studies will reveal how different schedules of PI3K inhibition shape tumor biology.

Title: Autophagy promotes escape from PI3K inhibition in ER+ breast cancer
Wei Yang1 and Todd W Miller1. 1Norris Cotton Cancer Center, Geisel School of Medicine at Dartmouth, Lebanon, NH.
Body: PI3K inhibitors have shown promise for the treatment of ER+ breast cancers. Autophagy has been suggested to promote
resistance to PI3K inhibitors. Interestingly, cancer cells often exhibit basal levels of autophagy, which is thought to serve as a
mechanism to manage oxidative stress and remove damaged organelles (e.g., mitochondria). Pharmacodynamic analysis of the
orally available pan-PI3K inhibitor GDC-0941 at low (100 mg/kg) and high (800 mg/kg) doses, and 2 low doses administered 12
hours apart (BID), in combination with fulvestrant in mice bearing MCF-7 xenografts revealed that treatment induced
accumulation of LC-3 II (marker of autophagy) after 24-48 h. In ER+ breast cancer cells cultured in hormone-depleted medium,
PI3K inhibition gradually induced apoptosis while increasing LC-3 II. Autophagy promotes glucose metabolism, and PI3K
activation is a major driver of glucose uptake. We thus hypothesized that inhibition of autophagy will potentiate the anti-tumor
effects of PI3K inhibition by suppressing glucose metabolism.
Chloroquine (CQ) is an anti-malarial drug that inhibits autophagy in mammalian cells by an unknown mechanism. CQ inhibited
proliferation and autophagy in ER+ breast cancer cells in vitro. CQ treatment potentiated GDC-0941-induced inhibition of cell
proliferation and promotion of apoptosis in growth conditions (10% FBS) with or without fulvestrant, and in hormone-depleted
conditions (10% DCC-FBS). CQ treatment increased PI3K inhibitor-induced mitochondrial membrane depolarization, suggesting
that this drug combination engages an intrinsic apoptotic pathway. PI3K inhibition also induced autophagy in fulvestrant-resistant
MCF-7/FR and CAMA-1/FR cells. CQ increased GDC-0941-induced apoptosis in MCF7/FR and CAMA-1/FR cells, offering
combined targeting of PI3K and autophagy as a promising therapeutic strategy for the treatment of anti-estrogen resistant breast
cancer.
Mice bearing ZR75-1 xenografts are currently being treated with vehicle, CQ (2 mg/day via drinking water), 800 mg/kg GDC-0941
QW, or the combination. Thus far, single-agents treatments significantly inhibit tumor growth. However, the drug combination has
not shown synergistic effects in vivo. Analysis of different doses and schedules of GDC-0941 plus CQ, with or without
anti-estrogen treatment, is ongoing.


Title: Effect of adjuvant systemic therapy in reducing rates of loco-regional recurrence in early-stage breast cancer: Results from
nine NSABP randomized phase III trials
Eleftherios P Mamounas1, Gong Tang1,3, Qing Liu1,3, Jong-Heyon Jeong1,3, Thomas B Julian1,4, Priya Rastogi1,5, Charles E
Geyer1,6, Sandra M Swain1,7, Soonmyung Paik1,8, D Lawrence Wickerham1,4, Joseph P Costantino1,3 and Norman Wolmark1,4.
1
National Surgical Adjuvant Breast and Bowel Project (NSABP), Pittsburgh, PA; 2UF Cancer Center at Orlando Health; 3University
of Pittsburgh Graduate School of Public Health and School of Biostatistics, PA; 4Allegheny Cancer Center at Allegheny General
Hospital; 5University of Pittsburgh Medical Center, Pittsburgh, PA; 6Virginia Commonwealth University, Massey Cancer Center,
Richmond, VA; 7Washington Cancer Institute, Medstar Washington Hospital, Washington, DC and 8Severance BioMedical
Science Institute and Yonsei University College of Medicine, Korea.
Body: Background
Adjuvant systemic therapy reduces risk of distant recurrence (DR) and breast cancer death. In addition, adjuvant systemic
therapy reduces risk of loco-regional recurrence (LRR). We examined the magnitude of the effect of adjuvant systemic therapy
(tamoxifen, chemotherapy, and chemotherapy + trastuzumab) in reducing incidence rates and cumulative incidence rates of LRR
as first event in nine recent NSABP randomized trials that were conducted from 1981 to 2005 and included a total of 21,815
patients.
Methods
Nine NSABP clinical trials of adjuvant (or neoadjuvant) systemic therapy, in which a reduction in LRR or DR was observed, were
included in the analysis (NSABP B-13, B-14, B-19, B-20, B-21, B-27, B-28, B-30, and B-31). The cumulative incidence rates of
LRR as the first disease-free survival (DFS) event were estimated and compared across treatment arms via log-rank tests. The
sub-distribution proportional hazards models were applied to estimate the reduction in incidence rate of LRR from adjuvant
systemic therapies. The corresponding magnitude of reduction in the incidence rate of any DFS event was estimated from Cox
proportional hazards models.
Results
Across all nine clinical trials, adjuvant systemic therapy resulted in reductions in LRR that were comparable to or greater than the
reductions in DFS events (Table). The observed reductions in LRR with adjuvant chemotherapy were of greater magnitude in
trials of node-negative patients (35-58%) than in trials of node-positive patients (13-15%). Reductions in LRR were of similar
magnitude with adjuvant chemotherapy as with adjuvant tamoxifen. In B-27, the sequential addition of neoadjuvant or adjuvant
docetaxel to neoadjuvant AC reduced LRR rates by 27%. The addition of trastuzumab to adjuvant chemotherapy decreased LRR
rates by 34%.
Conclusions
Rates of LRR have steadily declined over time in NSABP adjuvant clinical trials. This decline can be attributed to improvements in
surgical and radiotherapy techniques but is also the result of the use of increasingly effective adjuvant systemic therapy.
NSABP Trial Population
Treatment
Comparison
HR(95%CI)DFS HR(95%CI)LRR
B-13
(n=1,084)
N(-) / ER(-)
MF v No Adj Rx
0.66 (0.55,0.79) 0.42 (0.29-0.62) 5.9 v 13.5
<0.001
B-14
(n=4,028)
N(-) / ER(+)
TAM v Placebo
0.70 (0.65, 0.77) 0.54 (0.45-0.66) 5.2 v 11.2
<0.001
B-19
(n=1,074)
N(-) / ER(-)
CMF v MF
0.69 (0.57, 0.84) 0.48 (0.31-0.73) 5.3 v 10.0
<0.001
B-20
(n=2,299)
N(-) / ER(+)
CMF/MF+TAM v
TAM
0.76 (0.66, 0.88) 0.57 (0.41-0.78) 3.7 v 7.1
<0.001
TAM v Placebo
0.93 (0.71, 1.21) 0.65 (0.38-1.09) 4.7 v 8.8
0.10
B-21 (n=645) N(-) / 1cm
10-yr Cum Incidence of

LRR(%)
Log rank
p-value
B 27
(n=2,346)
Operable T1-3
N0-1
ACT v AC
(neoadj)
0.92 (0.81, 1.05) 0.73 (0.56-0.94) 9.1 v 12.2
0.015
B-28
(n=3,036)
N(+)
ACP v AC
0.90 (0.81, 1.00) 0.87 (0.69-1.1) 8.6 v 10.0
0.24
B-30
(n=5,240)
N(+) / HER2(-)
ACT v AT/TAC
0.84 (0.76, 0.93) 0.85 (0.65-1.1) 4.6 v 5.4
0.23
B-31
(n=2,063)
N(+) / HER2(+)
ACP+H v ACP 0.59 (0.50, 0.69) 0.66 (0.46-0.94) 5 v 7.4
0.02
N: Node; M: Methotrexate; F: 5-FU; TAM: Tamoxifen; C: Cyclophosphamide; A: doxorubicin; T: Docetaxel; P: Paclitaxel; H:

trastuzumab

Title: Final results of the VitaCal randomized phase III study evaluating the impact of a tailored oral vitamin D supplementation
regimen on serum 25-hydroxyvitamin D levels in early breast cancer patients
William Jacot1, Nelly Firmin1, Lise Roca1, Delphine Topart2, Sophie Gallet1, Anna Durigova1, Simone Mirr1, Stphane Pouderoux1,
Jean-Pierre Bleuse1, Pierre-Jean Lamy1 and Gilles Romieu1. 1ICM Val d'Aurelle Montpellier Cancer Institute, Montpellier,
France and 2Montpellier University Hospital, Montpellier, France.
Body: Purpose: Only a minority of patients with early breast cancer (EBC) treated with adjuvant or neoadjuvant chemotherapy
have sufficient baseline vitamin D. The current recommendations regarding daily vitamin D supplementation appears too low to
correct this deficiency in this population. Optimal vitamin D dosing has yet to be determined in this setting. The current
randomized phase III study address the issue of the effectiveness and safety of a tailored high dose oral vitamin D
supplementation as a means for restoring normal 25-hydroxy vitamin D (25OHD) levels in a large population of
chemotherapy-treated EBC patients.
Methods: Chemotherapy-treated EBC patients were stratified according to the degree of Vitamin D deficiency, time between
chemotherapy initiation and inclusion (0 to 6 months versus 6 to 12 months), hormone receptors status and menopausal status.
Participants were randomly assigned to receive a 6-months conventional (C) vitamin D and calcium supplementation or a
6-months high dose oral vitamin D regimen tailored on the degree of deficiency (T) associated with a conventional calcium
supplementation. Primary endpoint was the efficacy (6-months percentage of 25OHD serum levels normalization) in the T arm
compared with the C arm. Patients without vitamin D normalization from the C arm were allowed to switch to the T arm after 6
months. Statistical analyses were performed on an intent to treat basis.
Results: The trial accrued 215 patients, among which 197 patients presented with vitamin D deficiency, and randomized 195
patients (T, 100; C, 95) from July 2011 to January 2013. The groups were well balanced in regard to the stratification
characteristics, as well as in regard of median weight and neoadjuvant or adjuvant chemotherapy status. Compliance to the daily
oral supplementation was low in both arms, 64% of the patients in both arms taking less than 80% of the planned oral
supplementation dose. Compliance to the tailored high dose vitamin D schedule appeared better (78%). After 6 months of
treatment, at the primary endpoint analysis time, significantly more patients in the T arm presented with normalized serum vitamin
D levels compared to the C arm (30% vs. 12.6%; p=0.003). Vitamino-calcic supplementation was well tolerated, with no
difference in the treatment-related toxicity between the 2 arms. 52 patients without vitamin D normalization from the C arm
switched to the T arm after 6 months. At the 12 months endpoint, 44% of these patients achieved vitamin D normalization.
Conclusion: In this randomized phase III study, a tailored high dose oral vitamin D supplementation allowed a statistically higher
percentage of serum 25OHD levels normalization compared to a conventional regimen, without any increase in side effects, in a
large population of chemotherapy-treated EBC patients. Observance of a daily oral supplementation remains poor in this setting,
advocating for an adaptation of the schedule and dosage of this supplementation in a population of patients subject to
chemotherapy-induced emesis.
Clinical trial number NCT01480869.

Title: Delay in trastuzumab initiation leads to decreased overall survival in patients with HER2+ early stage breast cancer
Christopher M Gallagher1, Kenneth More2, Anthony Masaquel3, Tripthi Kamath3, Annie Guerin4, Raluca Ionescu-Ittu4, Marjolaine
Gauthier-Loiselle4, Roy Nitulescu4, Nicholas Sicignano5, Brian Barnett3 and Eric Wu6. 1Walter Reed National Military Medical
Center, Bethesda, MD; 2Naval Medical Center Portsmouth, Portsmouth, VA; 3Genentech, San Francisco, CA; 4Analysis Group,
Inc, Montreal, QC, Canada; 5Health ResearchTx, LLC, Trevose, PA and 6Analysis Group, Inc, Boston, MA.
Body: Background
Trastuzumab reduces the risk of relapse in women with HER2+ early stage breast cancer. Yet, little information exists on the
timing of trastuzumab initiation and its association with relapse and survival outcomes in these patients. The study aimed to
investigate the impact of delaying the initiation of adjuvant trastuzumab treatment for >6 months on time to relapse, overall
survival, and relapse-free survival among patients with HER2+ early stage breast cancer who did not receive neoadjuvant
therapy.
Methods
Adult women initiating trastuzumab adjuvant therapy within 1 year of breast cancer surgery who did not receive neoadjuvant
therapy were selected from the US Department of Defense health claims database from 01/2003 to 12/2012 (N = 2,749). By
design, participants had to be alive and relapse-free at the time they initiated adjuvant trastuzumab. Patients were classified into
two groups based on the time from breast cancer diagnosis to trastuzumab initiation: 6 months and >6 months. An algorithm
based on secondary neoplasm ICD9 codes along with treatment gaps and initiations was used to identify relapses. Percent
relapses and/or deaths were reported by study groups and compared using 2 tests. The impact of delaying trastuzumab initiation
on time to relapse, overall survival, and relapse-free survival was estimated from Cox regression models adjusted for age, overall
comorbidity profile at the time of the BC diagnosis (Charlson index), type of surgery (breast conserving vs. breast removing), and
radiotherapy (prior to the initiation of trastuzumab). In all three Cox models the follow-up started at adjuvant trastuzumab initiation.
Results
Of 2,749 women who met the selection criteria, 79.3% initiated adjuvant trastuzumab 6 months of diagnosis and 20.7% initiated
adjuvant trastuzumab >6 months after the diagnosis (Table). Patients who delayed the initiation of trastuzumab for >6 months
were younger (57.2% aged <65 years vs. 50.9%, p = .008) and a higher proportion of them received radiotherapy prior to the
initiation of trastuzumab compared to those who initiated trastuzumab earlier (77.2% vs. 53.1%, p < .001). There were no
significant differences between the two groups in overall comorbidity profile and type of surgery. Patients who initiated
trastuzumab >6 months after diagnosis had a higher risk of relapse, death, or relapse/death than those who initiated trastuzumab
6 months of diagnosis in both unadjusted and adjusted analyses (Table).
N events (% events)
Hazard Ratio (95% CI)
6 months group N = >6 months group N

>6 months group vs 6
p-value
2,180
= 569
months group
Relapse outcome
333 (15.2%)
134 (24.3%)
< .001
1.40 (1.14 - 1.72)*
Death outcome (overall survival)
138 (6.3%)
64 (11.6%)
< .001
1.44 (1.06 - 1.96)*
Relapse or death outcome

(Relapse-free survival)
386 (17.6%)
148 (26.8%)
< .001
1.33 (1.09 - 1.61)*
*p-value < .05

Conclusions
The results of this population-based study among patients with HER2+ early stage breast cancer who did not receive neoadjuvant
therapy suggest that delays of over 6 months in the initiation of trastuzumab among HER2+ early stage breast cancer patients
are associated with a higher risk of relapse and shorter overall survival and relapse-free survival.
Disclaimer
Research derived from an IRB approved protocol at Naval Medical Center Portsmouth, VA. The views expressed in this abstract
are those of the authors and do not necessarily reflect the official policy or position of the Department of the Army, Department of
the Navy, Department of Defense or the United States Government. Dr. C.G. and Dr. K.M. are members of the U.S. military. This
work was prepared as part of their official duties. Title 17 U.S.C. 105 provides that 'Copyright protection under this title is not
available for any work of the United States Government.' Title 17 U.S.C. 101 defines a United States Government work as a work
prepared by a military service member or employee of the United States Government as part of that person's official duties.

Title: Clinical intervention trial with physical activity during chemotherapy for primary breast cancer: Different effects of endurance
and resistance training on physical fitness and quality of life
Christoph Mundhenke1, Weisser Burkhardt2, Walter Jonat1, Juliane Drkop1, Lisa Keller1, Arne Falk2 and Thorsten Schmidt1.
1
OB/GYN, Breast Unit, UKSH, University of Kiel, Kiel, Germany and 2Sports Science, University of Kiel, Kiel, Germany.
Body: Aim: Previous findings suggest that physical activity during breast cancer treatment can reduce side effects and improve
clinical outcome. However, physical training is not performed to the same degree as e.g. for coronary heart disease. In part, this
is due to the lack of studies comparing different forms of exercise on their effectiveness, which aggravates the composition of
exercise-guidelines. As a contribution for this goal, this intervention study compares the effects of moderate endurance and
moderate resistance training on physical fitness, fatigue, concentration and the quality of live.
Methods: In a randomized, controlled intervention trial 12-week supervised endurance (ET) or resistance training (RT) were
compared with standard usual care (UC) in patients with primary breast cancer during adjuvant or neoadjuvant chemotherapy. 78
female patients were enrolled. Endpoints were muscular strength (NM), endurance (Watt/kg/bodyweight in kg) and well during
endurance stress test (Borg scala), quality of live (QL) (EORTC QLQ C30+BR23) before and after 12 weeks of treatments. 14 out
of 81 patients dropped out (due to chemotherapy related side effects or withdrawal of consent). 67 patients are fully evaluable.
Results: RT (vs. UC) was superior for improving strength (p: 0.015). A trend towards improvement in strength was observed in the
(ET vs. UC) (p:0.149; n.s.).All groups decrease in the endurance stress test /Watt/kg/bodyweight in kg) after 12 weeks (RT: p: n.s;
ET: p: n.s; UC: p: n.s.), however the maximal endurance lost in Watt was greatest in UC (p: 0.001). The subjective perceived
exertion at 100 watts remained stable in the RT (p: n.s.) and decreased in ET (p:0.02) and in UC (n.s.).
In the RT group quality of life score improved significantly during 12 weeks of intervention (p: 0.011). There is also a trend for
improvement of QL in the ET group (p: >0.05; n.s.). The UC group showed a decreased QL. Conclusion: Important improvements
in strength, endurance and quality of life from exercise training in breast cancer patients receiving chemotherapy are
demonstrated in this trial. RT showed a superior improvement in physical strength, in subjective perceived exertion and quality of
life over ET and UC. The beneficial results suggest that physical intervention (including a resistance intervention) should be
implemented into standard of care during adjuvant chemotherapy for breast cancer. A combined intervention of endurance and
resistance intervention may be optimal and needs to be further prospectively evaluated.

Title: ERALOP study: Post adjuvant FEC - docetaxel chemotherapy for early breast cancer: Hair regrowth in the real life
Hugues Pierre Bourgeois1,2, Aurlie Jamet3, Franoise Grud2, Carole Adounkpe3, Pierre Kerbrat4, Remy Delva5, Hlne Simon6,
Philippe Deguiral7, Bertrand Diquet3, Pascale Lain3 and Anne-Lise Septans5. 1Centre Jean Bernard, Clinique Victor Hugo, Le
Mans, Pays de la Loire, France; 2Observatoire ddi au Cancer Bretagne Pays de la Loire, Angers, Pays de la Loire, France;
3
CHU Angers, Angers, Pays de la Loire, France; 4CRLCC Eugne Marquis, Rennes, Bretagne, France; 5Institut de Cancrologie
de l'Ouest Paul Papin, Angers, Pays de la Loire, France; 6CHRU Brest Morvan, Brest, Bretagne, France and 7Clinique Mutualiste
de l'Estuaire Cit Sanitaire, Saint Nazaire, Pays de la Loire, France.
Body: Background : during 2008 we have collected one hundred observations of persistent significant alopecia (PSA). FEC
100-docetaxel 100 mg/msq regimen was mostly concerned. We therefore decided to evaluate exact incidence of this relevant
side effect through women points of view.
Methods : ERALOP is a retrospective study using a self-questionnaire targeting patients (pts) treated with this sequential
regimen from 2008 to 2009. The primary objective was to estimate the incidence of a PSA at 6 months after last course of
docetaxel with CTCAE 4.0 classification : grade 1 : hair loss of up to 50% not obvious from a distance, a different hair style may
be required to cover the hair loss, grade 2 : hair loss > 50% with a psychosocial impact. The sample size calculation of 635
patients took into account : PSA incidence of 3.2% (TAC regimen), precision of 0.015, risk at 0.05, 20% patients lost for follow
up. ERALOP study was approved by local ethic comitee.
Results: from July 2012 to October 2012, 829 pts received a self-questionnaire. 176 pts (21%) did not answer and were
considered as without PSA. 653 (79%) answers with medical data fully documented were collected. Median age of patients was
56 years. Six months after last docetaxel course, PSA incidence grade 2 : 8.6% (71 pts), grade 1 : 32.6% (271 pts), grade 0 :
56% (466 pts), NA : 2.5% (21 pts). 73% of pts with PSA received hormonotherapy. At the time of the inquiry (median follow up of
3.7 years), PSA incidence grade 2 was 3.5% (29 pts), grade 1 : 30% (248 pts), grade 0 : 63,8% (529 pts), for a global PSA
incidence of 33.4%. Between 6 months after last course of docetaxel and time of the inquiry, it appears slight or total regrowth for
40 pts with PSA grade 2 and 187 with PSA grade 1.Three pts still wore a wig, and many pts had suboptimal regrowth of eyelash
(31%), eyebrow (47%), pubic hair (27%), and nail disorders (27%). A multivariate analysis was performed to look for PSA risk
factors. All the oral treatments, including dexpanthenol, biotin, methionine cysteine and cystin-vitamin B6 proved to have no
efficacy. Treatment by topical minoxidil could have a little efficacy. Impairment of health-related quality of life will be assesed by
Dermatology Life Quality Index (DLQI) and alopecia grade will be evaluated at the end of hormonotherapy.
Conclusions: Physicians and patients should be aware of this new distressing side-effect. This high level of PSA lead us to
conduct ALOPREV trial to investigate, in spite of FEC induced alopecia, the properties of cooling cap prevention trial during
docetaxel infusion. Preliminary results are encouraging and this option could be considered.

Title: Clinical outcomes according to pathological complete response (pCR) and proliferation index of residual tumor (RT) after
neoadjuvant chemotherapy (NC) in invasive breast cancer (IBC)
Antonella Ferro1, Alessia Caldara1, Mariachiara Dipasquale1, Chiara Trentin1, Renza Triolo1, Mattia Barbareschi1, Daniela
Bernardi1, Marco Pellegrini1, Daniela Cazzolli1, Gabriella Berlanda1, Fabio Gasperetti1, Francesca Maines1, Paolina Tuttobene1,
Orazio Caffo1 and Enzo Galligioni1. 1Santa Chiara, Trento, Italy.
Body: BACKGROUND:
IBC is a heterogeneous disease with several subtypes molecularly identified by gene expression profile. Since subtypes defined
by immuhistochemistry (IHC) panel are similar although not identical to molecular subtypes, IHC may represent an easier
alternative to identify them.
PURPUSE:
To assess the clinical outcomes of pts who received NC for IBC and the differences by IHC-related subtypes.
METHODS:
We retrospectively reviewed the clinical records of the pts treated with NC for stage II-III IBC from 2000 to 2013. For each pt we
recorded baseline tumor size, type of NC [which consisted of anthracyclines (A) + taxanes (T) in HER2- and T + trastuzumab (H)
A in HER2+ pts), type of surgery, pathological response (pCR defined as the absence of invasive cells in the breast and the
lymph nodes regardless of DCIS). IHC subtypes were defined according to ER and PgR expression, Ki-67 level, and HER2
status:
Luminal A (LA): ER and PR+, neg HER2 and Ki67< 14% (= 3%)
Luminal B (LB): ER and/or PR+, neg HER2 and Ki6714% (=30%)
Luminal HER2 (LHER2): ER and/or PR+, positive HER2 and any Ki67 (=27%)
HER2 positive (HER2+): neg ER and PR, positive HER2 and any Ki67 (=12%)
Triple negative (TN): neg ER and PR, neg HER2 and any Ki67 (13%)
Unknown subtype in 33 cases (15%)
The loco-regional and distant RFS and OS were evaluated according to pCR.
pCR and survival outcomes were also assessed on the basis of both pre- and post- NC Ki67 levels.
RESULTS:
In the consecutive series of 213 pts who received NC median age was 50 yrs (r. 25-75). The NC consisted of an A+T based
regimen in HER2 negative (145 pts) and of a T+ H with A (31 pts) or without A (34 pts) in HER2+ disease.
Only 14 did not receive surgery: 10 for distant metastases development and 4 because still on NC. Quadrantectomy was
performed in 120 pts (60%). Among all pts, pCR was achieved in 44 pts(22%) with further 4 pts showing a RT 1 mm.
Relationship between pCR and subtypes, ki67 and recurrence rate
LA (%)
LB (%)
LHER2 (%)
HER2+
TN (%)
Median Ki67 (%)
Recurrence Rate (%)
pCR
12.5
42.5
27.5
17.5
48
4.5
No pCR
100
42.3
29.2
8.8
14.6
37
31.5
p Value
<0.001
=0.001
All but 19 HER2+ pts (84) received H obtaining pCR in 38% of cases regardless chemotherapy type (A-based 35% vs Not A38%)
The median follow-up was 45 months (range 1-166 ms).
The 4y-RFS and OS were better in which achieved pCR than those did no (RFS 92 vs to 74%; p=0.0014 and OS 95 vs 78%;
p=0.0074).
Median Ki67 in pretreated core biopsy was 40 compared to 27% in post-NC RT. Patients with high (>30%) post-NC Ki67 levels
showed significantly higher risk for disease relapse (4 y-RFS 60%; p=0.0019) and death (4y OS 71%; p=0.018) compared with
patients with <15% (4y-RFS 93 and OS 88%) or >15-30 Ki67 levels (4y-RFS 83 and OS 82%) .
CONCLUSIONS:
According to literature data, pts achieving pCR after NC showed better RFS and OS compared to no pCR pts. The pCR rate was
significantly higher in aggressive subtypes (HER2 and TN). In HER2 disease, pCR was achieved by using chemo + H,
irrespective of A-addition. Interestingly high pre-NC KI67 levels seem to predict the possibility obtaing pCR, while post-NC Ki67
levels seem to be of prognostic value in pts who do not receive pCR.

Title: Unexpected low incidence of cardiovascular related events in women with primary breast cancer; population derived
retrospective cohort
Antroula Papakonstantinou1, Laila Hubbert2, Rasmus Mikiver3, Jonas Bergh1 and Elham Hedayati1. 1Karolinska Institutet,
Stockholm, Sweden; 2Linkoping University, Linkoping, Sweden and 3Regional Cancer Centre Southeast Health Region,
Linkoping, Sweden.
Body: Background
Many studies and even a meta-analysis of the EBCTCG found that, compared to methotrexate containing chemotherapy
(non-ACT) or no chemotherapy (no-CT), anthracycline-containing chemotherapy (ACT) decreases overall mortality after breast
cancer (BC). However, increased cardiac mortality was observed but did not outweigh improvement in BC survival. The incidence
of cardiotoxicity after adjuvant BC treatment is reported with varying frequencies and still remains unknown.
Methods
Women younger than 60 years and diagnosed with lymph node positive BC between 1998 and 2002 were identified through the
Cancer Registry of the Regional Cancer Centre in the southeast health region of Sweden. Data on patient, tumor and treatment
characteristics were registered from the Cancer Registry and medical records. Information on cardiovascular toxicity (CVT) was
collected from the National Patient Registry, the Cause of Death Registry and the Prescribed Drugs Registry. ICD diagnoses for
cardiac diseases and hypertension were used. A predefined Kaplan Meier and Cox regression analysis and a post hoc
multivariate analysis to investigate confounders were performed. CVT was registered as event only if it occurred prior to BC
relapse.
Results
Of 524 eligible women, 329 were analyzed (mean age, 50 years; range, 28 to 60 years) and 195 were excluded due to missing
medical records (n=153), primary metastatic disease (n=18), trastuzumab treatment (n=6) and known heart disease at BC
diagnosis (n=13). Of those analyzed, 176 received ACT, 64 non-ACT and 89 no-CT. The cumulative CVT was 16 %. CVT was
observed in 12 % of those that received ACT, 14 % of those that received non-ACT and 25 % of those that received no-CT. CVT
was significantly higher among women that received no-CT (p=0.024) meanwhile BC relapse and BC mortality was significantly
higher among women treated with ACT compared to no-CT (p=0.02). According to Kaplan Meier curve, the risk for CVT among
women treated with ACT was higher than the rest the first 5 years after diagnosis. 13 out of 85 patients that received epirubicin
<450mg/m2 and 5 out of 54 that received epirubicin >450 mg/m2 developed CVT (p=0.302). A multivariate analysis adjusted for
age, treated side, obesity, smoking and received chemotherapy showed that age between 51 and 60 year was a significant risk
factor for developing CVT (p=0.035).
Conclusion
Clinically overt CVT 10-14 years after adjuvant BC treatment was diagnosed in 16% of women, but the subclinical incidence
remains unclear. Women that did not receive chemotherapy showed increased incidence of CVT, probably due to lower risk for
BC relapse and better overall survival. Age seems to be an independent risk factor for CVT. The incidence of clinical ACT CVT
was unexpected low, however, CVT was observed already in epirubicin < 450 mg/m2. BC relapse still remains the most important
risk factor for mortality among women with high risk BC and nowadays high-dose chemotherapy and dose dense treatments are
given, increasing the risk of CVT. It is essential to co-operate with cardiologists and investigate methods to identify patients that
are at risk of developing chemotherapy-related CVT and find suitable measures to prevent CVT influencing overall survival.

Title: One-tenth of patients younger than 40 years develop a permanent chemotherapy-induced ovarian function failure after
receiving adjuvant anthracycline-based chemotherapy with or without taxanes
Vivianne C Tjan-Heijnen1, Ingeborg J Vriens1, Ashley J Beijers1, Maureen J Aarts1, Maaike de Boer1, Joyce H Royen1 and Ron J
van Golde1. 1Maastricht University Medical Centre, Maastricht, Netherlands.
Body: Background
To assess the incidence and predictors of (permanent) chemotherapy-induced ovarian function failure (COFF) in premenopausal
women with hormone receptor positive breast cancer treated with adjuvant chemotherapy.
In our university hospital, patients with COFF and hormone-receptor positive breast cancer are monitored for ovarian function
recovery by 3-monthly FSH and estradiol blood levels (serum estradiol is measured by direct immunoassay). In this present
study, we collected data from the medical records of all premenopausal women with hormone-receptor positive breast cancer
treated with anthracycline-based chemotherapy, with or without the addition of taxanes, who were diagnosed in the years
2005-2012. To meet the definition of COFF, the amenorrhea and ovarian function suppression had to last 24 weeks since the
last menstruation before or during chemotherapy. Patients with hormone-receptor negative breast cancer were excluded.
Results
We identified 135 eligible women. Initial oral hormonal therapy consisted of tamoxifen (n= 116) or aromatase inhibitors (AI, n=16,
of whom 1 patient younger than 40 years). Median follow-up of the included patients was 25 months (range 3-69 months). The
majority of women was older than 40 years (80%). Permanent or temporary COFF was present in 95.6% of patients; that is, in
97.2% of patients of 40 years versus in 88.8% of patients < 40 years of age, which was not different between age-groups.
However, permanent COFF was significantly more often present in women 40 years (75%) as compared with 11.1% of women
< 40 years ( P < 0.03). Patients who developed a permanent COFF had a mean age of 47.4 (SD 3.9) years, whereas patients
who developed a temporary COFF had a mean age of 38.0 (SD 6.5).
In 57% of the patients, premenopausal hormone levels were the first evidence of ovarian function recovery. The second-last FSH
and estradiol values of patients who had an ovarian function recovery were still clearly in postmenopausal range (Figures will be
shown at the meeting).
Conclusion
COFF is seen in 89% of patients < 40 years, but in the majority it was reversible. This is reassuring for those with a childwish. As
in a significant proportion of patients FSH and estradiol values are the first sign of ovarian function recovery, close monitoring of
ovarian function is required if ovarian function suppression is considered an additional effective hormonal treatment, and with
respect to indication of non-hormonal contraceptive devices. We would not recommend AI as single hormonal treatment in young
patients with COFF.

Title: Oncologist treatment choices in patients with early stage invasive lobular breast carcinoma - a survey
Carmel Jacobs1, Mark Clemons1, Mohamed FK Ibrahim2, Christina Addison1, Jean-Michel Caudrelier1, Ian D Graham1, Brain
Hutton1 and Angel Arnaout1. 1Ottawa Hospital Cancer Centre & University of Ottawa, Ottawa, ON, Canada and 2St Vincent's
University Hospital, Dublin, Ireland.
Body: Introduction: Invasive lobular carcinoma (ILC) is common and accounts for 5-15% of all breast cancers. ILC has distinct
clinical and histological features that separate it from invasive ductal carcinoma (IDC) with regards to its breast imaging
characteristics, patterns of recurrence and sensitivity to systemic therapy. ILC presents challenges to the physician in many
aspects of local-regional and systemic therapy choices. We surveyed breast cancer physicians on their beliefs and practice
patterns on issues around the management of ILC.
Methods: A questionnaire was developed and circulated electronically using a modified Dillman technique to surgical, radiation,
and medical oncologists across Canada and Ireland.
Results: The questionnaire was completed by 91 of 429 physicians (21% response rate). Response rate by specialty was 25/69
(36%), 21/54 (41%) and 45/306 (13%) for surgical, radiation and medical oncologists respectively. Most surgeon responders
(77%) would feel "uncomfortable making treatment decisions for ILC with a mammogram alone" and 100% would be "more
comfortable with an MRI". Although 55% reported treating ILC as they would IDC, 22% of surgeons will purposefully obtain larger
gross margins intra-operatively. Some radiation oncologists believe ILC is an independent risk factor for local-regional recurrence
after breast conserving surgery (49%), and after mastectomy (29%). 33% of radiation oncologists would offer radiation therapy
after mastectomy specifically because of the ILC subtype even in the absence of usual indications for radiotherapy. Most medical
oncologists treat ILC comparably to IDC with the factors having the largest influence on systemic treatment decisions being
tumour stage, hormone receptor status and HER-2 status. 51% of medical oncologists treat ILC with adjuvant chemotherapy as
they do for IDC at least most of the time, while 40% use neoadjuvant chemotherapy as frequently as they do for IDC at least
most of the time. 75% of medical oncologists manage the hormonal treatment of ILC as they do IDC most of the time or
always.
Conclusions: There remains significant clinical equipoise in the local-regional and systemic management of ILC. This survey has
demonstrated wide variations in both beliefs and practices of management for ILC. Clinical guidance, developed on clinical trials
specifically assessing the management of ILC, is required.

Title: RANK/RANKL expression by immunohistochemistry (IHC) in young breast cancer (BC) patients and during pregnancy:
Association with clinicopathologic features, gene expression profiles and patient outcome
Hatem A Azim, Jr1, Fedro A Peccatori2, Sylvain Brohee1, Daniel Branstetter3, Giancarlo Pruneri2, Sherene Loi4, Giuseppe Viale2,
Bill Dougall3 and Christos Sotiriou1. 1Jules Bordet Institute, Brussels, Belgium; 2European Institute of Oncology, Milan, Italy;
3
Amgen Inc, Thousand Oaks and 4Peter MacCallum Cancer Center, Melbourne, Australia.
Body: Background & Objectives: RANKL is a major paracrine effector of the mitogenic action of progesterone in mammary
epithelium via its receptor RANK. Increased mammary tumor formation following pregnancy was observed in transgenic mice with
gain of function in RANK, a process that was arrested using a RANKL inhibitor. Based on epidemiological studies, pregnancy
increases BC risk on the short term with pregnancy-associated BC associated with poor prognosis. Here, we report for the first
time the expression of RANK/RANKL in young BC patients using IHC, and its association with diagnosis during pregnancy and
prognosis. We also evaluate genes and pathways that are activated in RANK/RANKL expressing tumors.
Patients & Methods: 195 young BC patients were included; of whom 65 were diagnosed during pregnancy. All patients had
central pathologic review and 85% had available gene expression data using Affymetrix. RANK/RANKL expression by IHC on the
primary tumor and adjacent normal tissue was performed at Amgen laboratories, blinded for clinical data. IHC was performed with
antibodies against human RANK (N-1H8) and human RANKL (M366) using the H-score. The difference in expression of
RANK/RANKL between pregnant and non-pregnant patients and the association with clinicopathologic features were examined.
We evaluated genes and pathways that are associated with RANK/RANKL expression as a continuous variable in a linear
regression model. Finally, we tested the association between RANK/RANKL expression and disease-free survival (DFS).
Results: Median age was 36 years (range: 28-47). RANKL expression was more prevalent in the pregnant group independent of
other pathologic features; both on the tumor (mean H score: 32 vs. 8) and adjacent normal tissue (mean H score: 87.3 vs. 32.9,
both p<0.001). 18.7% of pregnant and 5.3% of non-pregnant patients had 10% of cells with RANKL expression 3+. RANKL
expression was significantly higher in PgR+, well differentiated, and luminal-A tumors, with negative correlation with Ki-67 (all
p<0.001). RANK expression was higher on normal compared to tumor (23.6 vs. 14.2, p=0.003), with no differences according to
pregnancy status. RANK expression was higher in triple negative and poorly differentiated tumors (all p<0.001). Using FDR<0.05,
151 and 1207 genes were significantly correlated with RANKL and RANK expression by IHC, respectively. A positive correlation
was observed between mRNA and IHC expression of RANKL (r=0.89, p<0.001) and RANK (r=0.19, p=0.01). High RANKL
expression was associated with pathways related to mammary gland development, bone resorption, T-cell proliferation and
regulation of chemotaxis, while RANK expression was associated with immune response and proliferation pathways. At a median
follow-up of 65 months, neither RANK nor RANKL expression was associated with DFS.
Conclusions: RANKL expression is higher during pregnancy both in normal breast tissue and primary tumor and is associated
with important biological processes. These results support the preclinical data suggesting RANKL as a key player in the crosstalk
between pregnancy and BC guiding further development of RANKL-targeted therapy.

Title: Exploring biomarkers of response to zoledronic acid in breast cancer from clinical trial result of neoadjuvant chemotherapy
with zoledronic acid: JONIE-1 study
Takafumi Sangai1, Takashi Ishikawa2, Norio Kohno3, Daishu Miura4, Eiichi Sato5, Hiroshi Kaise2, Masato Suzuki6, Yoshie
Hasegawa7, Hirokazu Tanino8, Jun Horiguchi9 and Kohei Akazawa10. 1Chiba University Graduate School of Medicine, Chiba,
Japan; 2Tokyo Medical University, Tokyo, Japan; 3Kobe Kaisei Hospital, Kobe, Hyogo, Japan; 4Toranomon Hospital, Tokyo,
Japan; 5Tokyo Medical University, Institute of Medical Science, Tokyo, Japan; 6Teikyo University Chiba Medical Center, Chiba,
Japan; 7Hirosaki Municipal Hospital, Aomori, Japan; 8Gunma University Hospital, Ota, Gunma, Japan and 9Niigata University
Medical and Dental Hospital, Niigata, Japan.
Body: Background: Preclinical and clinical data have shown that zoledronic acid (ZOL) increases antitumor effect of
chemotherapy (CT) in breast cancer (BC). We previously reported that the addition of ZOL to neoadjuvant CT is potentially
beneficial in postmenopausal patients with triple-negative BC in JONIE-1 Study (0% pCR rate in CT versus 50% in CT combing
ZOL, p=0.029). Our data suggest that there might be subpopulation that responds to ZOL addition; however, there is lack of
evidence to explain mechanisms, emphasizing the need for exploring biomarkers of response using clinical samples. We
hypothesized three mechanisms of antitumor effect of ZOL.
1. ZOL inhibits Src signaling pathway directory in BC cell.
2. ZOL suppresses releasing Insulin-like growth factor-1 (IGF-1) from the bone and decreases chemoresistance in BC.
3. ZOL alters immune response to BC especially through TAM which induces cytotoxic T cell infiltration.
Patients and Methods: We investigated the relationship between clinicopathological features and tumor shrinkage in 178 Stage
IIA-IIIB HER-2-negative BC patients from the JONIE-1 adjuvant phase III trial comparing CT (FEC100 q3w 4 cycles followed by
weekly paclitaxel for 12 cycles) versus CT combining ZOL (4mg q3-4w). To evaluate Src activation, IGF-1 receptor (IGF-1R)
activation, M1/ M2 macrophage infiltration, and cytotoxic T cells infiltration, we performed immunohistochemistry and two
pathologists were independently performed assessment. Formalin fixed, paraffin embedded core needle biopsy sample at
diagnosis and surgical specimen after neoadjuvant CT were serially sectioned at 4 micrometer and processed for HE staining and
immunohistochemistry using primary antibodies as follows; Rabbit anti-Src mAb (36D10, Cell signaling), Rabbit polyclonal
anti-p-IGF-IR Ab (Tyr 1161) (Santa-Cruz Biotechnology), Mouse anti-CD80 mAb, (UMAB65, OriGene Technologies), Mouse
anti-CD163 mAb (10D6, Lwica Biosystems), Mouse Anti-CD8 mAb (C8/144B, Dako).
Results: All immunohistochemistry were successfully performed. Src activation was recognized as peripheral membrane
localization. Stained slides are now under evaluation by pathologist. Associations between clinicopathological features and the
effect of CT with ZOL will be under investigation.
Discussion: Src activation was observed in more than 70% of triple negative BC. Multiple cellular functions of Src are mediated
by Ras which is the main target of ZOL in osteoclasts. IGF-1R was overexpressed in approximately 50% of BC and IGF-1
signaling protects BC cell from CT by proliferative and anti-apoptotic effects. IGF-1 is the most released growth factor from the
bone matrix during bone resorption and ZOL administration resulted in a significant decrease in IGF-1. Subclinical inflammation
induced by macrophages in adipose tissue play an important role in promoting BC growth. It has been reported that ZOL reverted
tumor-infiltrating macrophages (TAM) phenotype from M2 to M1. Therefore tumor infiltrating lymphocytes would be dominantly
reverted to cytotoxic T cells for inhibiting tumor growth as well. We will present the results and discuss antitumor effects of ZOL at
the meeting.

Title: Change of HER2 status following neoadjuvant systemic therapy in primary breast cancer patients
Atsushi Yoshida1, Naoki Hayashi1, Hiroshi Yagata1, Koyu Suzuki1, Seigo Nakamura2 and Hideko Yamauchi1. 1St Luke's
Internationall Hospital, Tokyo, Japan and 2Showa University School of Medicine, Tokyo, Japan.
Body: Background: The incidence of change in HER2 status in primary breast cancer after neoadjuvant chemotherapy (NAC)
and whether the prognosis is affected by the change in HER2 status is not well known.
Patients and Methods: Five hundred and eighty-eight patients who were treated with anthracycline- and/or taxane-based NAC
and had non-pathologic complete response between 2001 and 2008 in our hospital were enrolled. Human epidermal growth
factor receptor-2 (HER2) status was assessed in specimen by needle biopsy before NAC and on the residual tumor of surgical
specimen. We determine the impact of change in HER2 status on recurrence-free survival (RFS). Twenty-eight patients had
received trastuzumab with NAC, and 57 patients had received trastuzumab as adjuvant therapy. HER2-positive was defined as
3+ by immunohistochemistry and/or amplification by fluorescent in situ hybridization. Association between change in HER2 status
after NAC and clinicopathologic factors, including age, clinical T stage, estrogen receptor (ER), progesterone receptor (PR),
Nuclear grade (NG), lymphovascular invasion (LVI) and Trastuzumab usage (NAC and Adjuvant setting) were determined.
Result: A median follow-up period was 57 months (range, 3 to 131 months). Four hundred eighty-nine of the 588 patients (83.1%)
had HER2-negative tumors and 99 patients (16.8%) had HER2-positive tumor before NAC. In 11 of the 489 patients (2.2%)
HER2-negative changed to HER2-positive.In 33 of the 99 patients (30%) HER2-positive changed to HER2-negative. In
clinicopathologic factors, ER and PR positive before NAC were associated with incidence of change in HER2 status after NAC.
Receiving trastuzumab was not correlated with incidence of change in HER2 status. In terms of RFS, there was no difference
between patients with and without change in HER2 status in both of the 489 patients with HER2-negative tumors and 99 patients
with HER2-positive tumors before NAC (p=0.26, p=0.23, respectively).
Conclusion: We herein reported the incidence of change in HER2 status after NAC with the largest sample size. However,
change in HER2 status did not seems to affect prognosis. Further prospective study is needed to confirm the prognostic impact of
change in HER2 status.

Title: Changes in Ki67 expression in breast cancer during the menstrual cycle and menopause
Takashi Fujita1, Masataka Sawaki1, Masaya Hattori1, Naoto Kondo1, Akiyo Yoshimura1, Mari Ichikawa1, Yayoi Adachi1, Tomoka
Hisada1, Haruru Kotani1, Junko Ishigro1 and Hiroji Iwata1. 1AIchi Cancer Center Hospital, Nagoya, Japan.
Body: Background
Previous studies have shown that the menstrual cycle phase can influence PgR status of breast cancer. But data on whether the
menstrual cycle phase affects Ki67 expression is inconsistent. This study aims to compare the Ki67 expression on
ultrasonography guided vacuum-assisted breast biopsy (US-guided VABB) with matched breast cancer surgical specimens.
In 120 breast cancer patients without neoadjuvant chemotherapy who underwent US-guided VABB and surgical resection from
April 2008 and March 2012 at Aichi Cancer Center Hospital, we examined the concordance of Ki67 level between US-guided
VABB and surgical specimen. All the US-guided VABB were performed using 11-gauge Mammotome. In this study, the Ki67
cut-off level for positivity was defined at 20%.
Two phases of the menstrual cycle were pre-defined as indicated; phase 1 (low estrogen) days 2735 or 16; phase 2 (high and
intermediate estrogen) days 726 (Hayes BP, et al. Breast Cancer Res Treat 2013).
We defined the three groups as follows: the non-matching menstrual phase group (different menstrual cycle phase at the time of
biopsy and surgery: n=18), the matching menstrual phase group (same menstrual cycle phase at the time of biopsy and surgery:
n=25), and the post-menstrual group (n=77).
We evaluated the discordance of Ki67 expression between US-guided VABB and surgical specimens in the three groups.
Results
A differential expression of Ki67 was found in 13 patients and the concordance rate of Ki67 expression between US-guided VABB
and surgical specimens was 89.2% with a Kappa statistic value of 0.78. (The concordance rate of ER, PgR, and HER2 status
were 96.4%, 90.2%, and 97.0%, respectively.)
There were no major differences in tumor and patient characteristics (age, pathological tumor size, and number of biopsy
specimens) between the non-matching menstrual phase group and the matching menstrual phase group.
The discordance rate of Ki67 expression for the non-matching menstrual phase group, the matching menstrual phase group, and
the post-menstrual group were 22.2%, 4.0%, and 10.4%, respectively. In the patients with ER positive tumors, the discordance
rate of Ki67 expression for the non-matching menstrual phase group, the matching menstrual phase group, and the
post-menstrual group were 23.5%, 4.8%, and 10.9%, respectively. The discordance rate of Ki67 expression tended to be higher
in the non-matching menstrual phase group than in the matching menstrual phase group. (p=0.11)
Conclusions
Though limited by the low number of patients, our study suggested that the menstrual cycle could affect Ki67 expression as
patients with different menstrual cycle phase at the time of biopsy and surgery show discordant results.
Prospective evaluation of Ki67 expression in premenopausal patients with ER positive tumor is needed.

Title: Potential biomarkers of response to primary antiangiogenic and hormonal therapy in post-menopausal women with
hormone-positive, HER2-negative primary breast cancer
Helena Verdaguer1, Serafin Morales2, Valent Navarro1, Alba Martinez Lopez1, Anna Petit1, Fina Climent1, Oriol Casanovas1 and
Snia Pernas1. 1Institut Catal d'Oncologia, Hospital de Bellvitge, IDIBELL, Hospitalet del Llobregat, Barcelona, Spain and
2
Hospital Arnau de Vilanova, Lleida, Spain.
Body: Introduction The role of antiangiogenic therapy in primary hormonal therapy in HER2-negative ER-positive early breast
cancer is unknown. Potential biomarkers of response to antiangiogenic therapy are lacking. A phase I clinical trial was conducted
with sunitinib and exemestane given at conventional dose (25 mg/d) during 6 months, before surgery. 18 patients were enrolled,
15 in dose level 0 of sunitinib (25 mg/d) and 3 in dose level 1 (37.5 mg/d). Results were presented in SABCS 2011. Main toxicities
were: asthenia (50%), leucopenia (28%, all grade 2), diarrhea (28%), mucositis (22%), and hypertension (22%). 10 patients
achieved radiological partial response (56%) and 8 patients stable disease (44%). 7 patients (38.89%) obtained a pathological
downstaging. Potential biomarkers of response to antiangiogenic therapy are presented.
Materials and methods Tissue samples were obtained by fine-needle aspiration or core needle biopsy before starting treatment,
one month after and at surgery. All samples were formalin-fixed paraffin-embedded. Ki67, phospho-ERK and mean vessel density
(by CD34), were analyzed by immunostaining. At the same time points, plasma levels of angiopoietin 2 (ANG2), soluble VEGFR2
and VEGF were analyzed by ELISA. Basal levels and its changes over time were evaluated and correlated with clinical outcomes.
Results Changes in Ki67 were observed, with a median value of 16.44%pre surgery and 12.78% post surgery (p=0.062). A
significant decrease in mean vessel density was not observed.
Basal levels of plasmatic biomarkers are shown in the next table:
Basal levels of plasmatic biomarkers
SD
PR
Path Downst
No Path Downst
ANG 2
2396+/-650
3455+/-1394
0.08
3184+/-1610
2818+/-768
0.6
VEGF
110+/-125
77+/-66
0.57
94+/-63
88+/-121
0.9
VEGFR2
10570+/-225
12110+/-4394
0.35
12140+/-4321
10980+/-3196
0.55
Values of biomarkers are average in ng/ml Standard Deviation. Differences analyzed by students t-test.
Abreviations: SD= stable disease. PR= partial response. p = p value. Path. Downst.= Pathological downstaging.
Furthermore, there was a significant decrease in VEGFR2 mean levels after one month of treatment (p=0.0046): 122843449
ng/ml at baseline; 81483216 ng/ml at one month and 77323052 ng/ml at 6 months. Differences between basal and one month
determination were significant (p<0.01), but no differences were seen between 1 month and 6 months, showing a relevant early
decrease of VEGFR2 plasma levels. In contrast, levels of VEGF did not change significantly over time and had no association
with clinical outcomes.
Conclusions Baseline ANG2 levels have a promising predictive value of response in this phase I trial of neoadjuvant
combination of sunitinib plus exemestane. There is a significant early decrease in VEGFR2 with treatment with sunitinib. These
results should be validated in further studies to improve the selection of patients for antiangiogenic+hormonal therapy.

Title: Feasibility of the PROSIGNA multigene test in core biopsies and comparison to corresponding surgical breast cancer
sections
Aleix Prat1,2, Patricia Galvn1, Wesley Buckingham3, Maria Vidal1,2, Sherley Daz2, Paolo Nuciforo2, Sean Ferree3, Barbara
Adamo4, Santiago Ramon y Cajal2 and Vicente Peg2. 1Vall d'Hebron Institute of Oncology (VHIO), Barcelona, Spain; 2Vall
dHebron University Hospital; 3Nanostring Technologies and 4Hospital Clnic i Provincial.
Body: Background: The PROSIGNA (PAM50) gene assay has been validated on formalin-fixed paraffin embedded (FFPE)
surgical resection specimens (SRS) to identify the intrinsic subtypes of breast cancer and to estimate the 10-year risk of
recurrence (ROR) in post-menopausal patients treated with adjuvant endocrine therapy. However, demonstration of the ability to
perform PAM50 assay in diagnostic core biopsy specimens (CBS) before primary surgery and/or systemic therapy could be
clinically useful. The objectives of this study were to 1) evaluate the feasibility of performing the PAM50 assay in CBS and 2)
compare the PAM50 results from paired CBS and SRS.
Methods and materials: Baseline tissue surface area, cellularity and RNA yield (obtained after 10 FFPE 10m sections) were
determined in CBS from 30 newly diagnosed breast cancer patients. The tissue volume requirements determined from these
samples reflected the lower 95% confidence limits of a minimum RNA concentration of >20ng/L. The RNA yield and assay pass
rate of the established tissue volume requirements were then tested in 30 independent CBSs. Intrinsic subtype concordance, and
ROR score variability, were determined from 1) multiple extractions of the same CBS (10 independent cases for a total of 84
extractions) and 2) multiple CBS of the same tumor (30 independent cases for a total of 79 CBS). To test the PAM50 assay
concordance between paired CBS and SRS, the following PAM50 data were evaluated in an independent and retrospective set of
33 paired samples: 4-class subtype classification (Luminal A, Luminal B, HER2-enriched and Basal-like), 3-class subtype
classification (Luminal A/B, HER2-enriched and Basal-like), ROR score (0-100), proliferation score and the correlation to each
subtype centroid. Correlation and concordance between CBS and SRS were estimated using Pearson coefficients and multi-rater
kappa values, respectively.
Results: Baseline median surface area, cellularity and RNA yield concentration were 10.2 mm2, 45% and 155.3 ng/l,
respectively. Correlation of surface area and cellularity with RNA yield concentration was low (Person coefficient <0.25). Minimum
tissue requirements were determined based on surface area: >12 mm2 = 2 10-micron slides, 6-12 mm2 = 4 slides; <6 mm2 = 8
slides. Subtype calls on multiple extractions from the same CBS were 98% concordant (82/84) and the average ROR score
standard deviation (SD) was 2.2 units. Subtype calls on different CBS of the same tumor were 94% concordant (74/79) and the
average ROR score SD was 6.8 units. All 7 discordant cases were between Luminal A and B calls. Comparison between paired
CBS and SRS revealed correlations of 0.90 (range 0.90-0.98) for ROR scores, proliferation scores and subtype centroid
correlations. Intrinsic subtype concordance between paired CBS and SRS was 87% (kappa=0.81) and 97% (kappa= 0.91) for the
4-subtype and 3-subtype classifications, respectively. Finally, the overall PAM50 assay pass rate in CBS was >95%.
Conclusions: The PAM50 assay in CBS is feasible and measurements are comparable with surgical resections, which suggest
that PAM50 can be performed on diagnostic core biopsy tissues.

Title: Association of baseline pro-inflammatory (IL-6, CRP) and coagulation (D-dimer) markers with baseline functional status in
women with breast cancer (BC) undergoing chemotherapy
Yuan Yuan1, Nilesh Vora2, Tao Feng1, Joanne Mortimer1, Thehang Luu1, George Somlo1, Joseph Chao1, Vivi Tran1, Shu Mi1, Tim
Synold1, James Waisman1, Laura Zavala1, Vani Katheria1 and Arti Hurria1. 1City of Hope, Duarte, CA and 2Long Beach Memorial
Medical Center, Long Beach, CA.
Body: Background: Pro-inflammatory and coagulation factors such as IL-6, CRP and D-dimer serve as biomarkers for aging.
The utility of these markers as biologic correlates of physical function in patients with BC is not known. This study was performed
to determine if baseline serum markers of inflammation (IL-6, CRP) and coagulation (D-dimer) correlate with baseline functional
status in women with stage I-III BC requiring chemotherapy (chemo). Methods: This is a prospective longitudinal study that
enrolled 153 women across all age groups with BC who had pre-chemotherapy peripheral blood captured for IL-6, CRP, and
D-dimer and a baseline assessment of the following functional status measures: activities of daily living (Medical Outcomes Study
[MOS] Physical Health); instrumental activities of daily living (IADL); self-rated Karnofsky performance status (KPS);
physician-rated KPS; number of falls in last 6 months; and Timed Up and Go (TUG). Peripheral blood samples were collected for
measurement of IL-6, CRP and D-dimer. Quantitative IL-6 and CRP levels were obtained using NOVEX immunoassay
(Invitrogen) and D-dimer levels were measured with NanopiaD-dimer(Sekisui). Univariate analyses were performed to describe
correlations of these three biomarkers and 6 measures of physical function. Results: 153 patients (mean age of 57.5 y, range
30-81 y) with stage I- III BC (Stages I [n=35; 23%], II [n=82; 54%], III [n=36; 24%]) were enrolled. Chemo regimens include:
doxorubicin+cyclophosphamide/ paclitaxel(AC-T: 44%), docetaxel/cyclophosphamide (TC: 35%),
docetaxel/carboplatin/trastuzumab (TCH: 7%) and other regimen(14%). Scores for the physical function measures are as follow:
MOS (median 89, range 0-100); IADL (median 14, range 4-14); self-rated KPS (median 90, range 60-100); physician-rated KPS
(median 100, range 80-100); TUG (median 9 seconds, range 5-18). Serum biomarkers measurements and distributions are listed
in table 1. There were associations between decreased physical function by IADL and increased IL-6 (p<0.01); decreased MOS
and increased D-dimer (p<0.01); increased number of falls and increased CRP (p=0.02) and D-dimer (p=0.04); increased TUG
and increased IL-6 (p<0.01), CRP (p<0.01) and D-dimer (p=0.06) (table 2). Physician and patient-rated KPS did not correlate with
IL-6, CRP and D-dimer level. Conclusions: Baseline measures of inflammation and coagulation correlate with physical function
measures among patients with breast cancer. Future analyses evaluating the association between aging biomarkers and
measures of physical function with subsequent risk of chemotherapy toxicity is underway.
Table 1. Serum biomarkers measurement at baseline prior to initiation of chemotherapy
Biomarker
Mean
Standard Deviation
Median
Range
IL-6 (pg/ml)
4.3
5.1
3.0
0-48.0
CRP(g/ml)
5.7
7.9
2.8
0.1-48.4
D-dimer(g/ml)
0.7
0.6
0.6
0.1-3.3
Table 2. Univariate analysis of measures of physical functions versus biomarkers

Variables
Spearman Coefficient
p value
MOS vs D-dimer
-0.21
<0.01
IADL vs IL-6
-0.27
<0.01
No. of falls vs D-dimer
0.16
0.04
No. of falls vs CRP
0.19
0.02
TUG vs D-dimer
0.15
0.06
TUG vs IL-6
0.26
<0.01
TUG vs CRP
0.23
<0.01

Title: BRCAness and prognosis in triple-negative breast cancer patients treated with neoadjuvant chemotherapy
Hirokazu Tanino1, Yoshimasa Kosaka1, Norihiko Sengoku1, Mina Waraya1, Hiroshi Nishimiya1, Hirozhi Katoh1, Mariko Kikuchi1,
Naoko Minatani1, Saeko Teraoka1, Takumo Enomoto1, Takeo Sato2 and Masahiko Masahiko2. 1Kitasato University Hospital,
Sagamihara, Kanagawa, Japan and 2Kitasato University Hospital, Sagamihara, Kanagawa, Japan.
Body: Microarrays have shown that triple-negative breast cancers (TNBCs) are a heterogeneous group of disorders that have
differential responses to chemotherapy. On the other hand, BRCA genes play an important role in DNA damage repair. Gene
mutations and methylation of BRCAs causes functional abnormalities, leading to defects in DNA repair capacity. This state is
called "BRCAness." In this study, we addressed BRCAness, therapeutic effects, recurrence, and prognosis in patients with
TNBCs who were treated with neoadjuvant chemotherapy.
We enrolled 40 patients with TNBC who were treated with neoadjuvant chemotherapy (anthracyclines alone in 3 patients and
anthracyclines plus taxanes in 37 patients) at our hospital between April 2006 and October 2012. BRCAness was determined by
preoperative core needle biopsy (CNB) specimens and surgical specimens. Genes from those specimens were amplified by
multiplex ligation-dependent probe amplification (MLPA), and the amplicons were scored after separation by electrophoresis. With
a cutoff value of 0.5, values of 0.5 or more were determined as the BRCA1-like Type (BRCAness) and those of less than 0.5 as
the Sporadic Type to analyze clinical effects, pathological complete response (pCR) rate, recurrence, and prognosis.
With regard to therapeutic effects of neoadjuvant chemotherapy, pCR (ypT0/Tis/N0) was observed in 15 patients and non-pCR in
25 patients (pCR rate: 37.5%). Twelve patients had recurrence after surgery, and 8 of whom died of the original disease.
(1) The BRCA1-like Type accounted for 22 patients while the Sporadic Type accounted for 18 patients in CNB specimens. No
major differences were observed between the BRCA1-like Type and Sporadic Type with regards to the pCR rate (7/22 vs. 5/18).
Those two types had equivalent results in recurrence rate and prognosis.
(2) Among the 24 non-pCR patients whose BRCA status could be determined by surgical specimens, 9 were found to be of the
BRCA1-like Type. Patients with a BRCA1-like tumor had more recurrences (7/9 vs. 5/15), and their relapse-free survival was also
lower (p<0.05). No association with prognosis was found. Six patients whose BRCA status of CNB specimens was of the
BRCA1-like Type and that of surgical specimens turned to be of the Sporadic Type were better in recurrence (p<0.01) and
prognosis (p<0.05), compared with seven patients whose BRCA status of surgical specimens remained to be of the BRVA-1 like
Type.
Patients with TNBC who achieved a pCR by neoadjuvant chemotherapy had a better prognosis even if the type is BRCA1-like. In
contrast, the recurrence rate was higher when residual tumor remained after neoadjuvant chemotherapy and when the BRCA
status became BRCA1-like. New clinical trials assessing the true recurrence (TR) rate of BRCA are expected since neither
platinum-containing drugs nor poly (ADP-ribose) polymerase (PARP) inhibitors are effective against tumors with nonfunctional
BRCA genes.

Title: Assessment of the prognostic and predictive ability of a gene signature compared to histological grade in estrogen receptor
positive, HER2 negative breast cancer
Takayuki Iwamoto1, Catherine Kelly2, Giampalo Bianchini3, Takeo Mizoo1, Tomohiro Nogami1, Takayuki Motoki1, Tadahiko Shien1,
Naruto Taira1, Naoki Hayashi4, Naoki Niikura5, Toshiyoshi Fujiwara1, Hiroyoshi Doihara1 and Junji Matsuoka1. 1Okayama
University Hospital; 2Mater Misericordiae University Hospital; 3San Raffaele Hospital; 4St Luke's International Hospital and 5Tokai
University Hospital.
Body: Background: Genomic biomarkers have been widely adopted to assist in clinical decision making regarding chemotherapy
use in estrogen receptor (ER)-positive, HER2-negative breast cancer. First generation genomic signatures (FGGS) serve
predominately as prognostic biomarkers and secondarily have a role in prediction of chemotherapy response. The majority of the
FGGSs provide similar prognostic information which mainly capture in different ways tumor proliferation. While several studies
have compared the prognostic value of FGGSs to clinico-pathological variables, few studies have performed a similar comparison
for their predictive value. For this reason, we aimed to compare both the prognostic and predictive value of histological grade and
the genomic marker.
Methods: We retrieved publicly available cDNA microarray data from 1,373 primary ER+/HER2- breast cancers (n=721 treated
with various or unknown, n=350 untreated node negative, n=302 treated with neoadjuvant chemotherapy). We developed a
genomic signature simulated from recurrence online (http://www.recurrenceonline.com/) to calculate recurrence score and
recurrence risk using pre-defined sets of genes by cDNA microarray (B Gyorffy Breast Cancer Res Treat 2012). Breast cancers
were categorized as low, intermediate or high risk for distant recurrence using grade and genomic signature. We compared the
prognostic and predictive information provided by histological grade to the genomic signature. The outcome of interest in
untreated patient was distant event free survival. The outcome of interest in the anthracycline-taxane treated patients was
pathological complete response (pCR) in breast and axilla.
Results: Fifty five, 28 and 17% breast cancers were classified as low, intermediate and high risk by genomic signature and 22, 59
and 19% as grade I, II and III respectively. The genomic signature classified 11% of grade I/II cancers (126/1108) and only 42%
of grade III cancers (112/265) as high risk, and 29% of GradeIII (77/265) as low risk. Univariate analysis in the untreated cohort,
showed both histological grade (overall p=0.007) and the genomic signature (p<0.001) could predict prognosis. In multivariate
analyses for tumor size, age, grade and genomic signature, only the genomic signature remained statistically significant for
prognosis. As expected a significantly higher rate of pCR was observed in histological grade III cancers (15.9%) compared to
grade I (3.4%) and II (3.8%) after neoadjuvant chemotherapy (NAC). Results were similar using the genomic signature with pCR
rates of 4.6%, 5.7% and 16.5% for low, intermediate and high risk, respectively. Grade I and II cancers (n=189) classified as high
risk by the genomic signature had a pCR rate of only 2.8%. Instead, the grade III tumors which were also defined at high risk by
the genomic signature had a pCR rate of 26.5%. Multivariate predictive models showed neither biomarker retained statistical
significance in predictive response to NAC.
Interpretation: The genomic signature was better at identifying low risk cases compared to histological grade alone. There was no
difference in prediction of NAC response between either biomarker. Better predictive biomarkers for NAC response are needed.

Title: Prognostic significance of breast cancer index (BCI) in node-positive hormone receptor positive early breast cancer: NCIC
CTG MA.14
Dennis Sgroi1, Paul Goss1, Judy-Anne Chapman2, Elizabeth Richardson1, Shemeica Binns1, Yi Zhang3, Cathy Schnabel3, Mark
Erlander3, Kathy Pritchard4, Lei Han2, Lois Sheperd2 and Michael Pollack5. 1Massachusetts General Hospital, Boston, MA;
2
Queens University, Kingston, ON, Canada; 3bioTheranostics, Inc, San Diego, CA; 4Sunnybrook Odette Cancer Centre, University
of Toronto, Toronto, ON, Canada and 5McGill University, Montreal, QC, Canada.
Body: Background: The continuous linear Breast Cancer Index (BCI) risk index combines the ratio of genes HOXB13 to IL17BR
(H/I) and the molecular grade index (MGI) (Zhang et al, Clinical Cancer Research, 2013). The BCI signature was developed for
node-negative breast cancer patients treated with tamoxifen. We examine here whether linear BCI is prognostic for node-positive
hormone-receptor positive tamoxifen-treated patients.
Methods: MA.14 randomly assigned 667 hormone positive (HR+), postmenopausal women to 5 years of tamoxifen (TAM) +/- 2
years of octreotide LAR (TAM-OCT). A representative subgroup of 299 patients underwent gene expression profiling by RT-PCR
for linear BCI. We performed exploratory analyses restricted to node positive patients. The primary objective was to assess the
prognostic effect of BCI on relapse-free survival (RFS). RFS was defined as the time from randomization to the time of recurrence
of the primary disease alone, including local and ipsilateral nodal recurrence and metastatic disease, and censoring at longest
follow-up or death from another cause. With a median 9.8 years follow-up, the association of BCI with RFS was assessed by
multivariate Cox regression including treatment, stratification factors (other than nodal status), and baseline patient and tumor
characteristics. Patients were defined to be low risk based on BCI if the adjusted Cox survival was >95%, where adjustment was
by trial treatment, stratification factors, and baseline patient and tumor characteristics, including IGF-1, IGFBP-3, and C-peptide.
Results: 292 of 299 patient samples passed internal analytical quality control; 116 node positive ER+ve patients had 34 (29.3%)
relapses, with adjusted Cox survival at 9.6 years of 87.8%. Fifty-two of the 116 patients (45%) did not receive adjuvant
chemotherapy, and experienced 11 (21%) RFS events. In the 116 patients, higher continuous BCI value was associated with
shorter RFS (p=0.002): hazard ratio (HR) 1.49 (95% CI 1.16-1.91). Smaller pathologic T had significantly (p=0.03) better RFS
HR=0.39, (95%CI 0.17-0.90). With MA.14 patient mean BCI of 5.09532, Cox survival at 4.1 years was 95.2%; 17/34 (50%) who
recurred had failed by this time.
Discussion: In this subgroup analysis, we found that BCI and tumor size were significant prognostic factors for node-positive
hormone-receptor positive patients who were treated with tamoxifen.

Title: The neuronal protein sortilin is expressed in aggressive breast cancers and participates in tumor cell growth and invasion
Jay Pundavela1, Severine Roselli1, Yohann Demont2, Sam Faulkner1, John Attia3,4, Sheridan Keene1, Marjorie M Walker3 and
Hubert Hondermarck1. 1School of BioMedical Sciences & Pharmacy, Callaghan, New South Wales, Australia; 2INSERM U872,
Centre de Recherche des Cordeliers, 15 rue de l'Ecole de Medecine, Paris, France; 3School of Public Health & Medicine,
Callaghan, New South Wales, Australia and 4Hunter Medical Research Institute, Callaghan, New South Wales, Australia.
Body: The membrane protein sortilin is involved in intracellular trafficking and has emerged as a key player in the regulation of
neuronal viability and function. Few studies have suggested that sortilin may also be implicated in cancer, but its expression in
human tumors and potential value as a therapeutic target is unknown. In this study, the level of sortilin was analyzed in a series of
318 clinically annotated breast cancers and 53 normal adjacent tissues by immunohistochemistry. Sortilin was specifically
localized in epithelial cells and was detected in 65% of cancers compared to 46% of normal tissues (p=0.0088). Sortilin was
detected in 79% of invasive ductal carcinomas compared to only 54% of invasive lobular carcinomas (p<0.0001). Interestingly,
sortilin was associated with lymph node involvement (p=0.0093), suggesting a relationship with metastatic potential and poor
outcome. In Log-Linear modeling, two-way analyses confirmed the association of sortilin with histological type (ductal vs. lobular
invasive carcinomas) and lymph node invasion (p=0.002 and p=0.096 respectively). In vitro, sortilin was detected in a range of
cancer cell lines using RTPCR and Western-blotting. Functional investigation using RNA interference revealed that a decrease in
sortilin level in the highly metastatic breast cancer cell line MDA-MB-231 resulted in a reduced proliferation and invasiveness, but
had no effect on cell viability or apoptosis. Together, these data reveal that sortilin is expressed in aggressive breast cancers and
is a potential therapeutic target.

Title: No Show

Title: SOX10 and folate receptor alpha are frequently expressed in triple negative and progesterone receptor negative breast
cancers
Laura L Hoang1, Weimin Qi1, Charlie Yu1 and David Tacha1. 1Biocare Medical, Concord, CA.
Body: Background:
Specific biomarkers can be essential for developing effective treatments for aggressive breast cancers, especially triple negative
subtypes, for which treatment options are limited. Folate receptor alpha (FR), a critical membrane protein for DNA synthesis and
cell metabolism, has been suggested to participate in the transformation of breast cancer into aggressive subtypes. It has been
shown to be strongly associated with poor prognosis in triple negative breast cancers (TNBC) as well as estrogen receptor (ER)
positive and progesterone receptor (PR) negative subtypes.
SOX10 is a nuclear transcription factor that participates in neural crest development and in the differentiation of cells of
melanocytic lineage. Data suggests that SOX10 may contribute in stem cell or progenitor cell maintenance. Recently, SOX10
expression has also been documented in benign breast myoepithelial cells and in aggressive breast cancers. The correlation of
FR and SOX10 in breast cancer is not fully known. This is the first study to compare FR and SOX10 immunohistochemical
profiles in breast cancers with emphasis in TNBC.
Design:
166 cases of whole breast cancer tissues were classified according to their ER, PR, and HER2 immunohistochemical (IHC)
status. These same cases were then IHC stained for mouse monoclonal SOX10 and FR. Cut-off values of 1% and 5% for
SOX10 and FR, respectively, were used to determine positivity.
Results:
SOX10 achieved a sensitivity of 42.1% (8/19) in ER+/PR-/HER2- cases and was negative in all ER+/PR-/HER2+ cases (p<0.05).
FR was positive in 7.6% (7/92) of ER+/PR+/HER2- cases and was negative in all ER+/PR+/HER2+ cases. SOX10 identified
more ER+/PR-/HER2- cases (42.1%, 8/19) than ER+/PR+/HER2+ cases (7.7%, 1/13) (p<0.05). Similarly, FR stained 52.6%
(10/19) of ER+/PR-/HER2- cases and was negative in all ER+/PR+/HER2+ cases (p<0.005). SOX10 and FR were observed in
3.3% (1/30) and 20% (6/30) of HER2+ cases, respectively.
In ER-/PR-/HER2- (triple negative) cases, both markers were highly expressed with 40.0% (10/25) and 52.0% (13/25) positive
cases with SOX10 and FR, respectively, with 24.0% (6/25) of cases positive with both markers. Approximately one half of TNBC
cases expressed SOX10 and FR; however, most SOX10 positive TNBC cases did not overlap with FR positive TNBC cases.
Table 1: SOX10 and FR expression in breast cancer subtypes
ER/PR/HER2 Classification
SOX10+ (%)
FR+ (%)
Co-expression of SOX10 and FR (%)
ER+/PR+/HER2+ (n=13)
1 (7.7%)
0 (0.0%)
0 (0.0%)
ER+/PR+/HER2- (n=92)
6 (6.5%)
7 (7.6%)
2 (2.2%)
ER+/PR-/HER2+ (n=10)
0 (0%)
5 (50.0%)
0 (0.0%)
ER+/PR-/HER2- (n=19)
8 (42.1%)
10 (52.6%)
5 (26.3%)
HER2+ (n=30)
1 (3.3%)
6 (20.0%)
0 (0.0%)
ER-/PR-/HER2- (n=25)
10 (40.0%)
13 (52.0%)
6 (24.0%)
Conclusion:
SOX10 and FR were frequently expressed in triple negative breast cancers and in progesterone receptor negative breast
cancers. Our data suggests that there may be different mechanisms by which SOX10 and FR are implicated in aggressive
breast cancers. These findings may help achieve a better understanding of the two different pathways involving stem cells
(SOX10) and growth factors (FR), their potential prognosis and their therapeutic management in the future.

Title: Expression of SOX10 in invasive ductal carcinoma of the breast
Laura L Hoang1, Jianmin Wang2, David Tacha1, Huijiao Chan3, Bing Wei3, Zhang Zhang3, Hong Bu3, David G Hicks4 and Ping
Tang4. 1Biocare, Inc, Concord, CA; 2RTI Health Solutions, Research Triangle Park, NC; 3Sichuan University, West China Hospital,
Chengdu, China and 4University of Rochester Medical Center, Rochester, NY.
Body: Background: The transcription factor SOX10 mediates the differentiation of neural crest-derived cells, and expression of
SOX10 detected by IHC analysis is primarily used to support the diagnosis of melanoma. Expression of SOX10 has been recently
reported in benign breast myoepithelial cells and triple negative breast cancer. The aim of the current study is to investigate the
expression pattern of SOX10 in a cohort of invasive ductal carcinoma (IDC) tumors, and analyze its relationship with different
clinicopathological features and clinical outcome. Methods: Four hundred-twenty eight cases of IDC of the breast diagnosed in
our institution between 1997 and 2008 and having follow-up information were included in this study. Immunohistochemical studies
for SOX10 (1% of tumor cells having nuclear staining designated as positive expression), ER, PR, HER2 and Ki-67 were
performed on 25 pre-constructed tissue microassay blocks. The relationship between SOX10 expression and the
clinicopathologic features, expression of ER, PR, HER2 and Ki-67, and clinical outcome were evaluated. Results: Among this
cohort of patients, the majority of them were greater than 50 years old, with grade 1 or 2 tumors that were less than 2 cm and
node negative. 81% of the tumors were ER positive, 75% were PR positive and 8% HER2 positive. The overall expression rate of
SOX10 was 18%; however, its expression rates were significantly higher in high grade tumors (31%), and tumors that were ER
(53%) and PR (41%) negative, and tumors with high Ki-67 expression (designed as >15% nuclear labeling, 47%). 51% of
ER-/PR- tumors, 64% of triple negative tumors and 68% of Basal-like tumors expressed SOX10. SOX10 expression was not
associated with patient age, tumor size, nodal status and over-expression of HER2. When we looked at the 4 different levels of
SOX10 expression (1-25%, 26-50%. 51-75% and 76-100% of cells with positive nuclear staining) in different subgroups of breast
cancer, we noted that the majority of all cases had high levels of SOX10 expression (76-100% cells with nuclear staining)
regardless of their ER, PR, HER2 and Ki-67 status. No significant difference in overall survival (p value=0.8553) and disease-free
survival (0.1810) between SOX10 positive and negative tumors was noted. Conclusion: Our data demonstrates a significant
association between SOX10 expression and high grade, ER/PR negative, triple negative or basal-like IDC, suggesting their
myoepithelial differentiation. Given that SOX10 has been recently described as a principle driver for melanoma, better
understanding of SOX10 expression and signalling would provide important insight in development of novel diagnostic and
therapeutic tools for SOX10 expressing tumors.

Title: Expression of high affinity folate receptor in breast cancer brain metastasis
Jose P Leone1, Rohit Bhargava1, Adrian Lee1 and Adam Brufsky1. 1University of Pittsburgh, Pittsburgh, PA.
Body: Background: Folic acid is required by proliferating cells for the synthesis of nucleotide bases among other tasks. The high
affinity folate receptor (HFR) is a membrane protein that is upregulated in several cancers of epithelial origin and rarely present in
most normal cells. High expression of HFR has been documented in ovarian, breast, kidney, uterine and lung cancers. In breast
cancer patients, HFR is overexpressed in 33% of primary tumors (PT) and this is associated with poor prognosis. The HFR
expression in breast cancer brain metastases (BCBM) is unknown. The aim of this study was to analyze the incidence of HFR
expression in BCBM and its role in the prognosis of this high-risk patient subgroup.
Methods: We analyzed a database of 42 patients who underwent craniotomy with resection of breast cancer brain metastasis
(BCBM). Of those, we collected 19 brain metastasis (BM) and 13 PT. HFR status was assessed by immunohistochemistry.
Log-Rank test analyzed differences in overall survival (OS) between groups.
Results: Median age was 51 years (range 24-74). Median follow-up was 4.2 years (range 0.6-18.5). HFR was positive in 4/19 BM
(21.1%) and in 1/13 PT (7.7%). Positive samples had low H-scores (range 1-50). Staining with H&E revealed that 11/19 BM
(57.9%) and 7/13 PT (53.8%) had apocrine differentiation. Analysis of OS showed no difference between patients with positive
HFR (median OS 48 months) and negative HFR (median OS 69 months) (P=0.25). Similarly, there was no difference in OS
between patients with apocrine differentiation (median OS 63 months) and those without apocrine differentiation (median OS 69
months) (P=0.49).
Conclusions: To the best of our knowledge, this is the first analysis of HFR expression in BCBM. While previous studies
associated the presence of HFR with worse prognosis; in this cohort of high-risk patients, HFR was positive in only 21.1% of BM
with low levels of positivity. Around half of our patients had apocrine differentiation. Neither the status of HFR nor the presence of
apocrine features had impact in OS in our cohort.

Title: No Show

Title: AQP3 expression predicts survival in patients with HER2-positive early breast cancer
Yee Soo Chae1, Soo Jung Lee1, Jeeyeon Lee2, Jin Hyang Jung2 and Ho Yong Park2. 1Kyungpook National University Medical
Center, Kyungpook National University School of Medicine, Daegu, Korea and 2Kyungpook National University Medical Center,
Kyungpook National University School of Medicine, Daegu.
Body: Background: Recent studies have revealed aquaporins (AQPs) as targets for novel anti-tumor therapy, since they likely
play a role in carcinogenesis, tumor progression, and invasion. Accordingly, we analyzed the prognostic impact of AQP3
expression and polymorphisms in a number of patients with early breast cancer (EBC).
Patients and Methods: AQP3 expression was investigated on the basis of the immunohistochemistry of tissue microarray
specimens from 447 EBC patients who underwent surgery between 2003 and 2008. We scored the staining intensity (0 through3)
and percentage of positive tumor cells (0 through 4), and the staining score was defined as sum of these scores used to
categorize the AQP3 expression as negative (0 through 2), weak (3 through 5), or strong (6 or more). For AQP3 polymorphisms,
seven SNPs (rs10813981, rs34391490, rs591810, rs2227285, rs2228332, rs17553719, and rs3860987) were selected using in
silico analysis and genotyped using the Sequenom MassARRAY.
Results: A total of 180 (40.3%) of the patients were identified as AQP5-positive (staining score >2), including 86 (19.2%) cases of
strong expression (stating score >5). In a univariate analysis, AQP3 expression was significantly associated with survival for the
patients with HER2-overexpressing EBC. Moreover, a multivariate survival analysis revealed that AQP3 expression was an
independent prognostic marker of disease-free survival (DFS: HR=3.137, 95%CI=1.079-9.125, p=0.036; Distant DFS: HR=2.784,
95%CI=0.921-8.414, p=0.070) for the HER2-overexpressing EBC patients. Meanwhile, none of selected AQP3 polymorphisms
were related with AQP3 expression in tumor tissue nor survival in the current study.
Conclusion: AQP3 expression in tumor tissue may be considered as a potential prognostic marker in patients with
HER2-overexpressing EBC after curative surgery.

Title: 2-Hydroxyestrone is associated with breast density measured by mammography and fat:water ratio magnetic resonance
imaging in women taking tamoxifen
Cynthia A Thomson1, Patricia A Thompson1, Betsy C Wertheim1, Denise Roe1, Marilyn T Marron1, John-Phillipe Galons2, Matthew
A Kupinski1, Maria I Altbach1, Gertraud Maskarinec3 and Alison Stopeck1. 1University of Arizona Cancer Center, Tucson, AZ;
2
University of Arizona Bio5 Institute, Tucson, AZ and 3University of Hawaii Cancer Center, Manoa, Honolulu, HI.
Body: Research Objectives and Rationale. Tamoxifen (TAM) use has been shown to reduce breast cancer recurrence with the
benefit greater in patients who experience a TAM-associated decrease in percent mammographic density(PMD); findings that
support PMD as a biomarker of response to TAM. PMD is a radiographic phenomenon of breast fibroglandular tissue that is
associated with breast cancer risk. PMD is inversely associated with body mass index (BMI) and sparse data have shown a weak
positive association with the sex hormone levels. Limited data exist evaluating the relationship between TAM, 2OHE1:16 a-OHE1
ratio (concentrations previously hypothesized to be associated with a reduced risk of breast cancer) and PBD.
Methods. Using cross-sectional baseline breast density (BD) from an ongoing prevention trial of diindolylmethane (DIM) in 121
women receiving TAM, we evaluated BD in relation to circulating TAM metabolites [TAM, endoxifen, 4-OH TAM, ND TAM],
estradiol (E2), sex hormone-binding globulin (SHBG) and urinary 2-OHE1 and 16-OHE1. PMD was assessed by mammography
(n=65; 54%) and also a novel, non-radiative, non-contrast magnetic resonance imaging-derived fat-water ratio (FWR-MRI) as the
fat fraction (Fra)50 and 80 (n=53; 44%) developed for repeat BD assessment in short intervals. This is our first report that BD
using digitized mammograms is correlated with FWR-MRI-derived measures designated Fra50 and Fra80; Spearman = 0.90
and 0.86, respectively, p < 0.001.
Results. As previously demonstrated, BMI was inversely correlated with all measures of BD. No association was shown between
TAM and TAM metabolites and BD or urinary 2OHE1. Further, we found no relationship between circulating E2 or SHBG
concentrations and BD. In contrast, urinary 2OHE1 levels were positively correlated with BD across all measures of density;
2OHE1 levels were most strongly correlated with BD measured by FW-MRI using Fra80 (Spearman Fra80=0.483; p=0.001
compared to Fra50 =0.431; p = 0.004 and PD=0.400; p=0.003). A significant, but weaker, correlation was observed for the
2OHE1:16OHE1 ratio and BD ( values 0.34-0.38). The magnitude of the relationship between 2OHE1 and BD was similar in pre
and post-menopausal women despite lower PBD after menopause.
Conclusions. Our results replicate earlier work from Maskarinec et al. wherein excreted 2OHE1 was an independent determinant
of BD. These data challenge the hypothesis proposed by Yager and Liehr that higher urinary 2OHE1 to 16OHE1 ratio would be
indicative of reduced hormone tumorigenesis. These results suggest a possible comparable binding affinity for the estrogen
receptor that may modify endogenous steroid hormones and their effects on BD. Our findings strengthen the arguments favoring
a better mechanistic understanding of BD, the biological determinants and their relationship to breast cancer. This is particularly
timely given new mandates to provide BD measures to all women undergoing mammography and recent findings that while BD is
associated with breast cancer risk high BD is not associated with greater breast cancer mortality.

Title: No Show

Title: A role of MACC1 expression and its regulation of the HGF/c-Met pathway in breast cancer
Aiko Sueta1, Yutaka Yamamoto1, Mitsuhiro Hayashi2, Takashi Takeshita2, Mutsuko Ibusuki2 and Hirotaka Iwase2. 1Targeting
Therapy for Breast Cancer, Kumamoto University Hospital, Kumamoto, Japan and 2Kumamoto University Graduate School of
Medical Science, Kumamoto, Japan.
Body: Background
The newly identified gene, metastasis-associated in colon cancer 1 (MACC1), is suggested to be a transcriptional regulator of the
receptor tyrosine kinase gene c-Met, leading to cancer progression and metastasis in colorectal cancer. Also in breast cancer,
aberrant hepatocyte growth factor (HGF) / c-Met signaling has been shown to contribute to worse prognosis and confer
resistance to endocrine therapy or trastuzumab treatment, however, little is known of the role of MACC1. Here, we report its
impact on the survival for breast cancer patients and the biological function in the cell lines.
Methods
A total of 300 breast cancer patients who received both surgery and adjuvant treatment at Kumamoto University Hospital between
2001 and 2009 were selected. We analyzed expressions of MACC1 by reverse transcription polymerase chain reaction (RT-PCR)
and immunohistochemistry (IHC) to evaluate the associations of its expression with breast cancer survival. In an in vitro study, the
expressions of MACC1 were examined by Western blotting in breast and colorectal cancer cell lines. After transfection with a
MACC1-harboing plasmid, we evaluated the activities of cMet protein and cell motility and proliferation. Further, the binding ability
of MACC1 to the cMet promoter was evaluated using chromatin immunoprecipitation (ChIP) assay.
Results
In survival analyses, reduced MACC1 expressions were associated with patient mortality. Cox proportional hazards model
showed that MACC1 mRNA (HR = 0.25, P = 0.001), MACC1 protein (HR = 0.37, P = 0.016), as well as axillary nodal status and
estrogen receptor status, were independent predictors of mortality. No significant correlations between MACC1 expression and
other clinicopathological factors were found. We found no strong positive correlation between MACC1 protein and c-Met mRNA
expression with a Spearmans coefficient of 0.16 (P = 0.0067). In the cell lines tested, MACC1 expression was much higher in
colorectal cancer cells (DLD-1) than breast cancer cells (MCF7 and MDA-MB-231). To investigate the impact of MACC1 on the
biological function of the cells, we transfected with MACC1 in breast cancer and colorectal cancer cells (SW480). MACC1
overexpression did not induce cMet expression in MCF7, whereas the corresponding cMet expression was upregulated in SW480
cells. Further, SW480 cells transfected with MACC1 showed enhanced migratory ability, whereas in MDA-MB-231 cells,
transfection of MACC1 had no impact on this ability. In ChIP assay, the binding of MACC1 to the cMet promoter region was
suggested in SW480 cells, but not in MCF7 cells.
Conclusions
Our findings provide some novel insights into the role of MACC1 for breast cancer, indicating that it plays different roles in breast
cancer and in several other cancers. The biological mechanism of MACC1 which underlies improvement of breast cancer
prognosis remains unelucidated. There is possibility that MACC1 does not act as the exclusive master regulator of the HGF/c-Met
signaling involved in disease progression in breast cancer. Further studies to validate our results are needed.

Title: C-myc and bcl-2 overexpression in HER2-positive breast carcinomas
Alexandra L Larson1, Denise M Scholtens1, Adriana A Rosca1, Virginia Kaklamani1 and Kalliopi P Siziopikou1. 1Northwestern
University Feinberg School of Medicine, Breast Pathology and Medical Oncology Sections and The Robert H. Lurie
Comprehensive Cancer Center, Chicago, IL.
Body: Background: Gene amplification plays a significant role in the transcriptional regulation of the genome; in breast cancer,
amplification of the HER2 gene occurs in 20-30% of patients and confers a poor prognosis. Trastuzumab is now widely used in
HER2-positive breast carcinomas and has significantly changed outcomes in this aggressive subtype of breast cancer. C-myc,
another oncogene, is reported to be amplified in breast cancer; c-myc regulates cell proliferation. Bcl-2, the prototype
anti-apoptotic gene, is overexpressed in human breast carcinomas. The anti-apoptotic action of bcl-2 is located downstream from
HER2 and may serve as a potential drug target. This study evaluated the rate of c-myc and bcl-2 overexpression in a cohort of
HER2-amplified breast carcinomas and correlated the dysregulation of these genes with patient demographics and histologic
characteristics.
Design: The study population consisted of 96 patients with HER2-positive invasive breast carcinoma treated with surgical
excision or mastectomy at Northwestern Memorial Hospital (2009-12) (mean age 53, range 18-80). Electronic medical records
were reviewed for patient demographics. Pathologic tumor characteristics (histologic type, size, grade, lymph node status) and
tumor marker profile at the time of diagnosis (ER, PR, p53 and ki-67) were evaluated. Tissue microarrays were constructed (3
cores from each case to account for tumor heterogeneity) for immunohistochemical evaluation of c-myc (Epitomics, clone V69)
and bcl-2 (DAKO, clone 124).
Results: Overall, these HER2-amplified breast carcinomas were almost exclusively infiltrating ductal carcinomas (92/96, 96%) of
high histologic grade: 68/96 (71%) were grade 3, while the other 28 were grade 2 tumors. C-myc was overexpressed in 49/96
(51%) of the cases and bcl-2 was expressed in 60/96 (62%). There was no association of c-myc or bcl-2 expression with age,
patient race, tumor grade, tumor size, lymph node status, p53 expression or ki-67 proliferation index. Expression of both c-myc
and bcl-2 was seen in 30/96 (31%) of the cases. Of interest, c-myc-positive/bcl-2-positive tumors more often had positive lymph
nodes and higher ki-67 proliferation index compared to tumors with c-myc-negative/bcl-2-negative phenotype (46.7% v 29.4%
and 75.8% v 56.2% respectively).
Conclusions: (1) C-myc is expressed in over 50% and bcl-2 in over 60% of HER2-amplified carcinomas. (2) C-myc and bcl-2
expression do not appear to have differential expression between age groups or racial groups and do not correlate with histologic
tumor characteristics. (3) HER2-amplified carcinomas expressing both c-myc and bcl-2 are highly proliferative tumors with a high
rate of lymph node positivity. These findings suggest that coupled c-myc and bcl-2 overexpression may contribute to the biologic
aggressiveness of HER2-amplified breast carcinomas. Additional studies into the molecular mechanisms that drive
HER2-amplified tumors are currently underway in this aggressive subtype of breast cancer.

Title: PDE5 as a novel biomarker and a potential therapeutic target for breast cancer
Ines Barone1, Antonella Campana1, Cinzia Giordano2, Marilena Lanzino1, Daniela Bonofiglio1, Balazs Gyorffy3, Stefania Catalano1
and Sebastiano And1. 1Health and Nutritional Sciences, University of Calabria, Arcavacata di Rende, CS, Italy; 2Centro Sanitario,
Arcavacata di Rende, CS, Italy and 3MTA TTK Lendlet Cancer Biomarker Research Group, Budapest, Hungary.
Body: Background: Phosphodiesterases are enzymes responsible for regulating second messenger signaling by hydrolyzing 3-5
cyclic guanosine monophosphate (cGMP), that activates specific pathways resulting in protein phosphorylation, ion fluxes, or
cyclic nucleotide hydrolysis to affect gene expression or other aspects of cellular activity. Previous studies have reported
increased PDE5 expression in multiple human carcinomas, including bladder, colon, lung and breast cancers, suggesting a role
for PDE5 in tumorigenesis. In addition, several in vitro observations have shown antiproliferative and proapoptotic effects of
sildenafil and other PDE5 inhibitors in cancer cell lines. However, very little is known about PDE5 expression in human breast
tumours and its potential role in breast cancer initiation and progression. We therefore propose to determine whether PDE5
expression may be predictor of outcome in breast cancer patients, and examine PDE5 impact on breast cancer phenotype in
vitro.
Methods: We employed MCF-10A normal breast epithelial cells, estrogen receptor (ER) -positive (MCF-7/ZR-75/T-47D) and
ER-negative (BT-20/MDA-MB-468/SKBR-3/MDA-MB-435) breast cancer cells. We used RT-PCR, immunoblotting and
immunofluorescence analyses for evaluating PDE5 expression. To examine PDE5 impact on breast cancer phenotype, MCF-7
cells were engineered to stably express PDE5 and four clones were selected. Cell proliferation was assessed by MTT and
anchorage-independent assays, motility and invasion by wound-healing, transmigration and matrigel-based invasion assays.
Retrospective analysis using 1959 breast cancer patients of the Metabric Project was performed to evaluate relationship between
PDE5 expression and overall survival by Cox proportional hazard regression.
Results: PDE5 mRNA and protein were constitutively expressed at high levels in all the examined tumor cell lines compared to
normal breast cells, except for the less motile and non-invasive MCF-7 cells. Interestingly, higher PDE5 expression was found in
more aggressive ER-negative cells. Stable overexpression of PDE5 did not affect proliferation of MCF-7 cells, while it significantly
increased motility and invasion of all the stable PDE5-transfected clones tested. Patients having high PDE5 expression had a
statistically significant poorer survival compared to patients with low PDE5 expression (p=0.014, HR= 1.2). A more relevant
discrimination was achieved in lymph node-negative patients (p=0.0015, HR= 1.6), suggesting that assessing PDE5 levels may
be helpful to identify a subgroup of early-stage breast cancer patients who are most likely at the highest risk of progression.
Conclusions: PDE5 expression may enhance cancer cell invasive potential, thereby representing prospectively a potential
molecular candidate as prognostic marker and target for breast cancer therapy.

Title: Identification of subgroups of triple negative breast cancer cells with selective responses to mTOR, CDK, mitotic and
proteasome inhibitors
Prson Gautam1, Leena Karhinen1, Agnieszka Szwajda1, Sawan Kumar Jha1, Bhagwan Yadav1, Tero Aittokallio1 and Krister
Wennerberg1. 1Institute for Molecular Medicine Finland (FIMM), Helsinki, Finland.
Body: Triple negative breast cancer (TNBC) is characterized by the lack of of estrogen, progesterone and HER2/ErbB2
receptors. It is a highly heterogeneous class of breast cancer and transcriptomics has recently been used to define 6 major
subtypes of TNBC. We studied a panel of 15 TNBC cell lines using a chemical biology approach where we measure the
responses to 306 approved and investigational oncology drugs. Clustering of cell lines based on their overall drug responses
resulted in a strikingly different grouping compared to the gene expression derived one, highlighting that the current TNBC
subtyping is not easily converted to differential sensitivities to drugs.
To further evaluate the nature of the drug responses and to differentiate between their cell growth and cytotoxic effects, we
multiplexed the standard cell viability readout in the cell line screening with detection of cytotoxicity. This simple multiplexed
readout identified several drug classes that previously had been assumed to be cytotoxic based on strong effects on cell viability
(cell numbers) while they in fact showed no or a very heterogeneous effect on cytotoxicity. Drug classes exhibiting this type of
response included mTOR inhibitors, cyclin-dependent kinase inhibitors (eg. alvociclib), mitotic inhibitors (eg. paclitaxel) as well as
proteasome inhibitors (eg. bortezomib) and RNA synthesis inhibitors (eg. dactinomycin).
Further investigation of these drug classes showed that their static effects were reversible and in some cases the cells even
overcame the inhibitory effect in the presence of the drug in a matter of a few days.
Given the non-toxic responses to major classes of anticancer compounds such as mTOR inhibitors, we performed combination
screens with these compounds to identify other drugs with which they may synergize to promote cancer cell specific killing.
Surprisingly, we instead found that mTOR inhibitors had an antagonistic effect on the activity of many other cancer drugs such as
different cytotoxic and antimitotic drugs, tyrosine kinase inhibitors, HDAC inhibitors and PARP inhibitors, suggesting that
combining these classes of drugs may be counterproductive also in the clinic. We also found out that accessing a cytotoxic
readout allowed us to identify effective synergistic drug combination concentrations that were not seen in cell viability readouts.
For example, these synergistic toxic combination responses were seen in DU4475 cells when the MEK inhibitor trametinib was
combined either with the PARP inhibitor iniparib or with the broad spectrum tyrosine kinase inhibitor ponatinib.
In conclusion, multiplexed cell viability cell death readouts in drug sensitivity testing yields novel critical information on single drug
and drug combination activities and liabilities. With this we were able to conclude that antimitotic, mTOR, CDK, proteasome and
metabolic inhibitors have a heterogeneous cytotoxic effect across the panel of TNBC cell lines in contrast to their homogenous
effect on metabolic inactivation.

Title: Exploiting Isobutyl-deoxynyboquinone-induced DNA damage responses and metabolic changes for breast cancer therapy
Mariya Ilcheva1, Lifen Cao1, Sandeep Burma1 and David Boothman1. 1UT Southwestern Medical Center, Dallas, TX.
Body: During oncogenic or cellular stress new genes are frequently (over)-expressed that could be exploited for targeted cancer
therapy. The enzyme, NAD(P)H:quinone oxidoreductase-1 (NQO1) is over-expressed in most solid cancers, including 60% of
primary and metastatic breast cancers regardless of subtype. Normally, NQO1 detoxifies quinones resulting in the formation of
stable hydroquinones that are subsequently excreted from the cell. However, NQO1 bioreduction can turn certain rare quinones,
such as -lapachone and deoxynyboquinone (DNQ), into potent cell death-inducing compounds. These agents cause severe
DNA damage, poly(ADP-ribose)polymerase-1 (PARP1) hyperactivation, NAD+/ATP loss, and programmed necrosis of
NQO1-expressing cancer cells. Although -lapachone (ARQ761) is in current clinical trials at UTSW, more potent tumor-selective
NQO1 compounds are needed. Based on its structure and mode of action, isobutyl-DNQ (IB-DNQ) was recently added to the
spectrum of NQO1 substrates as a more selective and potent anti-cancer agent whose mechanism of action remains to be
elucidated.
Although NQO1 expression is a major determinant of IB-DNQ-induced lethality, previously published results from our group
showed that increased catalase expression could cause marked cytoprotection. We conducted a screen for NQO1:Catalase
ratios in 266 breast tumor samples, and 143 normal breast samples, for a total of 409 specimens. We found that NQO1
expression was significantly elevated in breast tumors compared to normal tissue. In contrast, catalase expression was
suppressed in breast tumors versus adjacent normal tissue. These results predict that normal tissue, which typically has higher
catalase levels than cancer cells, could be selectively spared from IB-DNQ-induced toxicity. Thus, NQO1:Catalase ratios favor
use of IB-DNQ in breast cancers to exploit this large therapeutic window.
Since NQO1 bioactivatable drugs synergize with agents that damage DNA, we hypothesized that certain cancer
vulnerabilities (e.g., BRCA1-deficient breast cancers) that have elevated endogenous DNA damage would synergize with
IB-DNQ. Exposure of breast cancer cells with IB-DNQ induced DNA damage, PARP1 hyperactivation, dramatic loss of essential
nucleotides (NAD+/ATP), and -calpain-mediated programmed necrosis with 10X greater potency than -lapachone.
IB-DNQ-induced DNA double-strand breaks (DSBs) that occurred in cells in S/G2 phases were mainly repaired by error-free
homologous recombination (HR), and therefore BRCA1-deficient cancers, being HR defective, would be particularly vulnerable to
IB-DNQ treatment. Indeed, HCC1937 breast cancer cells, deficient in BRCA1, were extremely sensitive to low dose IB-DNQ due
to the overwhelming levels of IB-DNQ-induced DNA damage and their inability to repair it due to their compromised HR. In fact,
IB-DNQ was far superior to PARP inhibitors in targeting BRCA1-deficient cells. Studies in vivo showed equivalent antitumor
efficacy of IB-DNQ to -lapachone and DNQ, but with much greater potency at lower doses. These findings offer preclinical
proof-of-concept for IB-DNQ as a potent chemotherapeutic agent for the treatment of breast cancers, especially those deficient
in BRCA1. This research was supported by grant CA102972 to DAB.

Title: SUMO Inhibitors affect tumorigenesis of novel breast cancer xenograft model
Maria V Bogachek1, Jung M Park2, James P De Andrade1, Mikhail V Kulak1, Jeffrey R White1, Tong Wu1, Philip M Spanheimer1,
George W Woodfield1, Thomas B Bair3, Alicia K Olivier4 and Ronald J Weigel1,2,5. 1University of Iowa, Iowa City, IA; 2University of
Iowa, Iowa City, IA; 3Iowa Institute for Human Genetics, University of Iowa, Iowa City, IA; 4University of Iowa, Iowa City, IA and
5
University of Iowa, Iowa City, IA.
Body: A novel basal breast cancer cell line IOWA-1T was derived from chemotherapy resistant locally advanced breast cancer
tumor. The cells rapidly form large, skin-eroding xenografts in nude mice. The SUMO inhibitor anacardic acid (AA) effectively
cleared CD44+/hi/CD24-/low cancer stem cell (CSC) population in IOWA-1T and BT-20 basal cancer cell lines and delayed tumor
outgrowth of basal cancer xenografts. The effect of SUMO inhibitors to clear the CSC population was dependent upon the SUMO
unconjugated form of TFAP2A (Bogachek MV et al, Cancer Cell, 2014). Herein we show that tumors that eventually form from
IOWA-1T xenografts in mice treated with AA are not capable of developing secondary xenografts, confirming eradication of the
CSC population by SUMO inhibitors. As further mechanistic evidence for the SUMO pathway, transient knockdown of UBC9 and
PIAS1 SUMOylation enzymes repressed CD44 expression and increased tumor free and overall survival in mice inoculated with
IOWA-1T xenografts. Furthermore, CD44 downregulation was demonstrated in IOWA-1T cells after treatment in vitro with UBC9
inhibitor PYR-41 and PIAS1 inhibitor NSC-207895. Overall survival of mice with IOWA-1T xenografts was increased to 430.5
and 392 days with PYR-41 and NSC-207895 i.p. injections, respectively, compared to a vehicle treated control group 331 days
(p<0.05). By contrast, doxorubicin treatment was not able to extend survival of mice with IOWA-1T xenografts. These findings
establish the class of SUMO inhibitors as potential therapeutic drugs that eliminate the breast CSC population and may be
effective in basal breast cancer cases resistant to conventional chemotherapy.

Title: Eribulin mesylate promotes a mesenchymal-epithelial transition (MET) and effectively radio-sensitizes triple negative breast
cancer cells
Eileen P Connolly1, Youping Sun1 and Tom K Hei1. 1Colmbia University Medical Center, New York, NY.
Body: Purpose: Eribulin mesylate is a novel non-taxane microtubule dynamics inhibitor that irreversibly induces mitotic arrest,
and has been shown to promote a shift from mesenchymal to epithelial phenotypes in breast cancer. It has demonstrated
therapeutic activity in heavily pretreated patients with metastatic breast cancer, with a trend toward greater efficacy in patients
with triple negative breast cancer (TNBC). TNBC is an aggressive tumor with a mesenchymal phenotype and higher locoregional
recurrence (LRR) compared to estrogen receptor (ER) positive breast cancer. One postulated cause for higher LRR in TNBC is
that these tumors have a great proportion of cancer stem cells (CSC). CSCs are radio- and chemo-resistant when compared to
the bulk of the tumor, and are thought to be the source of both LRR and distant metastasis after treatment, thus an important
target for therapy. Microtubule inhibitors are known radiation sensitizers given that they arrest cells at G2/M, the most radiation
sensitive phase of the cell cycle. We hypothesize that because eribulin induces an irreversible mitotic arrest and promotes a
mesenchymal-to-epithelial transition (MET) it will enhance the radiation sensitivity of TNBC through reduction in CSCs and
inhibition of DNA damage repair (DDR) pathways.
Methods: Experiments were conducted in 3 subtypes of TNBC cells, all with p53 mutations and PTEN loss: SUM149 (basal B),
BT-549 (mesenchymal-like), and MDA-MB-468 (basal A). Cells were treated with 0.5nM-3nM Eribulin, alone or in combination
with increasing doses of radiation from 0-8 Gy. In vitro studies preformed included; cell survival assays, immunoblot analysis for
markers of MET, FACS analysis for cell cycle, cancer stem cell population, and apoptosis, and finally P-H2AX staining to
measure DDR.
Results: We found that combining ionizing radiation (IR) and eribulin is synergistic. The IC50 of eribulin combined with 6Gy IR,
was 0.5-1 nM, 10 fold less than drug alone, and a 40-60% decrease in cell viability was observed with the combination compared
to IR alone. A significant decrease in clonogenic survival was seen after 24 hours pretreatment with 1nM eribulin, with the
greatest efficacy seen in BT-549 cells. No significant increase in apoptosis was seen with the combination treatment by FACS
analysis though a significant delay in DDR was observed. Cell cycle analysis demonstrated eribulin increases the proportion of
cells in G2/M arrest following IR, as well as suppressed cell proliferation. 24-hour eribulin treatment also caused morphological
changes in culture consistent with transition from a mesenchymal to an epithelial phenotype, and IHC showed a decrease in
mesenchymal markers and increase in epithelial markers. Finally eribulin treatment decreased the proportion of CSCs by FACS
analysis by 80% in SUM149 cells and prevented an increase in CSC population following IR. Studies are on going to further
characterize the effect of combined eribulin and IR on DDR and the CSC population in TNBC cells.
Conclusions: These studies demonstrate that eribulin effectively radiosensitizes TNBC cells. We believe this occurs through
inhibition of DDR as well as promotion of MET which reduces the radiation resistant CSC population.

Title: Systemic statin treatment and intra-tumoral lipid metabolism - report from a window-of-opportunity breast cancer trial
Maria Feldt1, Julien Menard1, Mattias Belting1 and Signe Borgquist1. 1Institution of Clinical Sciences, Lund University, Lund,
Sweden.
Body: Background:
In recent years, attention has been drawn to the plasma cholesterol lowering drugs statins (HMGCR-inhibitors) due to their
exertion of pleiotropic intratumoral effects such as induction of apoptosis and inhibition of proliferation, generating a possible utility
in cancer prevention and treatment. However, the molecular mechanisms are complex and remain largely undefined. The aim of
this study was to assess the statin induced changes of the intra-tumoral levels of cholesterol, to gain enhanced understanding of
the role of the mevalonate pathway in cancer lipid metabolism.
Methods:
The study was designed as a window-of-opportunity trial, where 50 patients with primary invasive breast cancer were included,
and treated with 80 mg of atorvastatin/day for two weeks, between the time of diagnosis and surgery. From frozen tumor tissue
samples pre- and post atorvastatin treatment, lipids have been extracted, and the cholesterol levels have been measured using a
cholesterol quantification assay. In vitro experiments on MCF7-cells treated with atorvastatin was used for optimization and
comparison on the cellular level.
Results:
42 patients completed all study parts, and assessment of the cholesterol levels in the paired frozen tumor tissue was achievable
in 14 pairs. Following atorvastatin treatment, the intra-tumoral levels of total cholesterol was significantly increased (P=0.035),
with an increase in 11 of the pairs, and decrease in 3 pairs. Similar findings were observed in the in vitro experiments,
demonstrating increased levels of total cholesterol and storage in lipid droplets in the cells treated with atorvastatin.
Conclusions:
We have earlier reported statin-induced anti-proliferative effects in breast cancer. This study shows a statin-induced increase in
the intra-tumoral cholesterol levels, which may however rely on the increased cholesterol storage in lipid droplets, a phenomenon
observed in the in vitro experiments with MCF7 cells treated with atorvastatin. Our results suggest that the anti-proliferative
effects exerted by statins reluctantly reduce cell metabolism and consequently increase the storage of e.g. lipids.

Title: Safety and feasibility of neoadjuvant combined chemotherapy of breast cancer with paclitaxel carried in a lipid
nanoemulsion (LDE) associated with adriamycin and cyclophosphamide
Andre Mattar1, Luciana B Ferreira2, Luiz H Gebrim1, Roberto Hegg1 and Raul C Maranho2. 1Perola Byington Hospital, So Paulo,
Brazil and 2Lipid Metabolism Laboratory of the Heart Institute (InCor-HCMUSP), So Paulo, Brazil.
Body: Background
The introduction of neoadjuvant chemotherapy considerably changed the natural history of breast cancer by permitting a big
tumors to be operable, permitting less aggressive surgical procedures and when achieving complete pathological response
prolonging the disease-free period. However, the chemotherapy schemes bear high toxicity rates and are even life-threatening.
Neoadjuvant chemotherapy aims at reducing mortality and improving surgical options and offers an in vivo chemosensitivity
testing at the same time. It is the ideal setting for clinical and translational research. Compared to the classical adjuvant
treatment, it offers several advantages.
Methods
We performed a pilot study to evaluate the safety and feasibility of neoadjuvant combined chemotherapy of breast cancer with
paclitaxel carried in a lipid nanoemulsion (LDE) associated with adriamycin and cyclophosphamide in comparison to the classical
neoadjuvant chemotherapy paclitaxel, adriamicyn and cyclofosfamide (TAC). All the patients were women with recently
diagnosed breast cancer who needed primary chemotherapy as the first treatment selected from Perola Byington Hospital in So
Paulo - Brazil.
Results
From April 2006 to June 2008, 39 patients from a center for breast disorders (Prola Byington Hospital) were included in the
study. These patients were randomized to the standard chemotherapy scheme (control arm n=16) or LDE-paclitaxel (LDE arm
N=17). Six patients were excluded from the study because they did not complete at least 5 cycles of chemotherapy. We also
excluded three patients because of lack of follow up. The majority of patients had stage III (37% IIIA and 27% IIIB). There was no
significant difference among the groups, considering the clinical variables, so the sample was considered homogeneous. The
patients were grouped in four different groups regarding the expression of ER, PR and CerbB2: Luminal subtype, Triple Negative,
Her2-positive and Hybrid Luminal. We observed 37,5% of complete response in the control arm and 23,5% in the LDE arm. The
partial response were 43,7% in the control group and 58,8% in the LDE group. The toxicity particularly in grade 3 and 4 events
were substantially lower in LDE group: Nausea 11% vs 2,2%; Vomiting 7,3% vs 0.7% and neutropenia 8,8% vs 2,2%.
Conclusion
LDE scheme was effective and had lower grad 3 and 4 events than the TAC regiment. More studies are necessary to evaluate
this approach.

Title: Targeting triple-negative breast cancer with fibroblast growth factor receptor 2 allosteric inhibitor RPT835
Sergey A Tjulandin1, Ilya V Tsimafeyeu1, Mikhail Y Byakhov1, Wei Yin2, John Ludes-Meyers2 and Frits Daeyaert3. 1RPT835
Research Consortium, Moscow, Russian Federation; 2Altogen Labs, Austin, TX and 3FD Computing Molmo Services, Beerse,
Belgium.
Body: Background: FGFR2 is amplified in 4% of triple-negative breast cancers (TNBC). In addition, the mRNA expression levels
of FGFR2 were significantly increased in amplified vs nonamplified tumor samples, suggesting a potential role for FGFR2 in
TNBC (Turner et al. 2010). We evaluated efficacy of allosteric inhibition of FGFR2 in high- and low-expressing TNBC xenografts.
Methods: Immunocompromised mice were used for xenotransplantation of FGFR2 high-expressing TNBC cells (SUM52PE) or
FGFR2 low-expressing TNBC cells (HS578T). Forty animals with measurable tumors were selected on day 10 and randomized
into treatment groups (low-molecular weight allosteric inhibitor RPT835, 30 mg/kg; gavage, daily) or vehicle (water; gavage,
daily). Measurements of tumor volume (mm3) were performed by digital calipers every 3 days during 40 days after tumor
inoculation.
Results: RPT835 significantly inhibited aggressive growth of SUM52PE tumor xenograft (P<0.0001). At study day 31, mean tumor
volumes ( SE) were 2712.237 mm3 in the vehicle group, and 1080.749 mm3 in the study group. In addition, animals of the
study group received RPT835 from day 31 to day 40 with disease stabilization (no differences in tumor volume between days 31
and 40, P=0.167). The tumor growth curve shows a nearly exponential increase in median tumor volume up to day 31 in the
vehicle group and a high rate of slow growing tumors up to day 40 in the study group. In the HS578T xenograft study differences
in tumor volume between study and control groups were weak at day 31 (mean tumor volumes SE, 70389.1 mm3 vs.
1053179.8 mm3, respectively; P<0.001) and at day 40 (1104162.2 mm3 vs. 1592335 mm3; P=0.01).
Conclusions: The allosteric FGFR2 inhibitor RPT835 significantly impacts on growth of FGFR2-expressing TNBC.

Title: Novel flavonoid Anto-028 shows promising antitumor activity in preclinical models of breast cancer
Carlos Martinez-Perez1, Carol Ward1, Peter Mullen2, Graeme Cook3, Donald McPhail3, Arran K Turnbull1, David J Harrison2 and
Simon P Langdon1. 1University of Edinburgh, Edinburgh, United Kingdom; 2Pathology, Medical and Biological Sciences Building,
University of St Andrews, St Andrews, Fife, United Kingdom and 3Antoxis Limited, IMS Building, Aberdeen, United Kingdom.
Body: New drugs are required to treat both endocrine resistant and triple negative breast cancers (TNBC). Among new structures
being investigated are flavonoids, which are natural polyphenolic compounds with antitumor properties. Clinical trials are
underway to assess their application as chemopreventatives and as sensitisers to chemotherapy. Among flavonoids, myricetin
has shown particular promise, inducing cell cycle arrest and mitochondrial-dependent apoptosis in cancer cell line models. In this
study we evaluated a novel series of myricetin-derived flavonoids with improved antioxidant and mitochondrial targeting
properties. We hypothesised that these compounds might have potential in breast cancer management and in particular to treat
estrogen receptor positive (ER+) tumors, in which estrogen controls tumor growth at least partially through oxidative
stress-related mitochondrial signalling. We first assessed the effect of 8 flavonoids on 4 breast cancer cell lines (MDA-MB-231,
BT-549, MCF-7 and HBL-100) and 3 MCF-7-derived cell lines with reduced sensitivity to endocrine therapy
(LCC-1/LCC-2/LCC-9). The novel flavonoids were designed to assess the involvement of redox potential and mitochondrial
targeting through selective structure-activity changes. Anto-028 was the most potent compound identified with 4 to 140-fold lower
IC50 values than myricetin, as shown by sulforhodamine B (SRB) assays applied to assess antiproliferative effects. Although
endocrine-resistant cell lines were less sensitive, Anto-028 exerted a strong, ER-independent antiproliferative effect on both ER+
and TNBC cell lines, with IC50 values in the low micromolar range. Treatment for 8 hours exerted dose-dependent reduction in
cell viability and induction of cytotoxicity and apoptosis, with a 2 to 5-fold increase in caspase activation, as detected by
luminescence-based plate assays. The involvement of reactive oxygen species (ROS) regulation in these effects was
demonstrated by plate-based assays and microscopic detection of fluorescent probes for mitochondrial hydrogen peroxide and
superoxide. Results indicated that different species of ROS are sequentially generated by treatment with a range of
concentrations of Anto-028, suggesting a possible biphasic effect of the drug at different concentrations. Experiments in mice
implanted with subcutaneous MDA-MB-231 xenografts showed that Anto-028 can significantly inhibit tumour growth in vivo with
associated changes in percentage of viable areas within the xenografts (from 56% in controls to 37% in mice treated with
25mg/kg/day of Anto-028) and reduced Ki67 proliferation index (from 60% in controls to 36% in treated xenografts). Current
studies are assessing combinatorial effects with TRAIL and chemotherapy which have produced synergistic effects in ovarian
cancer models, with nanomolar concentrations of Anto-028 sensitising resistant cells to induce strong, significant antiproliferative
effects. In conclusion, Anto-028 is a novel flavonoid with promising antitumor activity in preclinical models of breast cancer.

Title: Structure-activity relationship of ruthenium (Ru) complexes to inhibit breast cancer growth and metastasis
Joji Iida1, Marc L Prazo2, Yifeng Lu2, Elisabeth T Bell-Loncella2 and Craig D Shriver3. 1Windber Research Institute, Windber, PA;
2
University of Pittsburgh, Johnstown, PA and 3Walter Reed Army National Military Medical Center, Bethesda, MD.
Body: Previous studies demonstrated that transition metal complexes inhibit cancer cell growth. We have examined a series of
ruthenium (Ru (II)) complexes for their ability to inhibit growth of breast cancer cells. Among three complexes with the general
formula [Ru(6-p-cym)(N-N)Cl]+ tested for their abilities to inhibit MDA-MB-231 cell growth, the complex with
o-phenalyenediamine (o-pda) as the bidentate nitrogen donor (N-N) ligand, [Ru(6-p-cym)(o-pda)Cl]+, showed significant
inhibitory activity in both a concentration- and a time-dependent manners. Although [Ru(6-p-cym)(o-pda)Cl]+ inhibited growth of
various human cancer types such as MCF-7 (breast), SK-Br3 (breast), B-cell lymphoma (Raji), osteosarcoma (HT1080), and
melanoma (Bowes) cells, two TN breast cancer cells, HCC1806 and HCC38, were resistant to this treatment, suggesting cell-type
specific functions of [Ru(6-p-cym)(o-pda)Cl]+ for inhibiting cell growth. When MDA-MB-231 cells were treated with
[Ru(6-p-cym)(o-pda)Cl]+, both cleavage of caspase-3 and release of HMGB-1 into conditioned medium increased in a
concentration dependent manner, suggesting [Ru(6-p-cym)(o-pda)Cl]+ induced both apoptosis and necrosis processes.
Importantly, [Ru(6-p-cym)(o-pda)Cl]+ synergistically inhibited MDA-MB-231 cell growth with cyclophosphamide but not
doxorubicin and paclitaxel. These results suggest that [Ru(6-p-cym)(o-pda)Cl]+ is a potent tumor growth inhibitor per se and
enhances tumoricidal activity of chemotherapeutic agents such as cyclophosphamide. Thus, Ru (II) complexes are promising
anti-cancer drugs which could be used alone and/or in combination with chemotherapeutic agents for breast cancer patients. The
opinion and assertions contained herein are the private views of the authors and are not to be construed as official or as
representing the views of the Department of the Army or the Department of Defense.

Title: Targeting protein-protein interactions in the proteasome assemblies as a novel strategy to treat triple negative breast
cancers
Maria Gaczynska1 and Pawel A Osmulski1. 1University of Texas Health Science Center, San Antonio, TX.
Body: The ubiquitin-proteasome pathway is responsible for the most of regulated intracellular protein degradation in human cells.
The proteasome, a multifunctional, multisubunit, modular proteolytic enzyme is an essential part of the pathway. The proteasome,
which comprises several proteolytic assemblies sharing a catalytic core, is a recognized anti-cancer drug target. Competitive
inhibitors such as bortezomib and carfilzomib bind to the active centers of the core and are successfully used to treat aggressive
blood cancers, most notably refractory/relapsed multiple myeloma and lymphomas. Unfortunately, so far breast cancers
performed disappointingly in clinical trials with the proteasome inhibitors, even if a recent genome-wide screen identified
proteasome addiction as vulnerability of triple negative breast cancers. Triple negative breast cancers are among the most deadly
and difficult to treat and such vulnerability is more than worth to explore. Here we aimed at searching for new concepts in
proteasome targeting, better suited for breast cancers than active-centers blocking drugs. The sophisticated structure of the
proteasome offers ample opportunities to design noncompetitive inhibitors. The most physiologically relevant 26S proteasome is
an assembly of a 20S catalytic core and a 19S regulatory particle "cap". The active sites bearing core is the canonical target of
proteasome targeting anti-cancer drugs. On the other hand, the 19S protein complex is responsible for recognizing and
processing the majority of intracellular protein substrates tagged for degradation by polyubiquitin. We designed small molecules
that instead of binding to the active sites of the core, target protein-protein interactions between the core and the cap. The lead
compound B1 in vitro prevents assembly of 20S and 19S components, destabilizes 26S proteasome and noncompetitively, most
likely allosterically, inhibits the 20S core at low-nanomolar concentrations. We tested the performance of B1 in the culture of triple
negative breast cancer MDA-MB-231 cells. Apparently, B1 at nanomolar concentrations suppresses the cell culture growth,
lowers the content of 26S proteasome and compromises activity of the ubiquitin-proteasome system, as manifested by the
accumulation of polyubiquitinated protein substrates in the cytosol of B1 treated cells. Importantly, B1 strongly synergizes with
bortezomib, even if these cells are relatively resistant to the treatment with bortezomib alone. The cytotoxic effect of the combined
treatment on the MDA-MB-231 cell culture is apparent with single-digit nanomolar concentrations of both drugs. Summarizing,
targeting protein-protein interactions in the proteasome assemblies with small molecules, alone or in combination with competitive
inhibitors, seems to provide a promising strategy to treat proteasome addicted triple negative breast cancers.

Title: A phase Ib study of the CXCR1/2 inhibitor reparixin in combination with weekly paclitaxel in metastatic HER2 negative
breast cancer First analysis
Anne F Schott1, Max S Wicha1, Raymond P Perez2, Giraldo Kato3, Tiffany Avery4, Massimo Cristofanilli4, James M Reuben5, R
Katherine Alpaugh6, Susan McCanna7, Pier Adelchi Ruffini7 and Lori J Goldstein6. 1Comprehensive Cancer Center, University of
Michigan, Ann Arbor, MI; 2University of Kansas Medical Research Center, Fairway, KS; 3Oncology Research Associates,
Pinnacle Oncology Hematology, Scottsdale, AZ; 4Thomas Jefferson University, Philadelphia, PA; 5MD Anderson Cancer Center,
Houston, TX; 6The Hospital of Fox Chase Cancer Center, Philadelphia, PA and 7Development, Domp s.p.a., Milano, Italy.
Body: Background. Experimental models and retrospective clinical observations point to cancer stem cells (CSCs) as the
culprits for tumor recurrence and metastasis. CSCs account for a small proportion of tumor cells, suggesting that their elimination
through pharmacological targeting would not necessarily translate in any sizeable tumor regression. Thus, an ideal CSCs
targeting agent should be a nontoxic molecule that can be safely administered in combination with chemotherapy to reduce tumor
burden and improve disease control. CXCR1, one of the receptors for CXCL8, has been identified on breast cancer (BC) CSCs
(Ginestier C et al., JCI 2010). Reparixin, an allosteric inhibitor of CXCR1 with a large safety database in non-cancer patients,
effectively targeted CSC in BC xenografts (Ginestier C et al., JCI 2010).
Methods. Patients were female aged > 18 years with HER-2 neg metastatic breast cancer (MBC), eligible for treatment with
paclitaxel (not taxane-refractory), had received up to 3 prior CT lines for advanced BC (not including neo/adjuvant chemotherapy),
had measurable disease according to RECIST 1.1, had ECOG PS of 0-1, had adequate organ function, and had no brain
metastases. Patients received a 3-day run-in with reparixin oral tablets 3 times daily (tid) followed by paclitaxel 80 mg/m2/week
(Days 1, 8, and 15 for 28-day cycle) + reparixin oral tablets tid for 21 days. Three dose levels of 3-6 subjects were explored: 400
mg, 800 mg and 1200 mg oral reparixin tid. A further 17 subjects were enrolled at the highest tolerated dose (total 20 patients).
Safety was assessed following one cycle. Treatment continued until disease progression, unacceptable toxicity or withdrawal of
consent. Primary endpoints were safety and tolerability, and pharmacokinetic (PK) profile of the combination treatment. Among
secondary endpoints, assessment of disease response every 2 cycles for indication of efficacy and correlative evaluations on
peripheral blood samples were conducted.
Results. From 02/12 to 04/14, 33 patients entered the study. PK of reparixin at 400 mg tid, tmax = 0.5-1.5 hr, t1/2= 1.7 hr; at 800
mg tid, tmax = 0.5-3 hr, t1/2= 4.6 hr; at 1200 mg tid, tmax = 0.5-2 hr, t1/2 = 1.6 h. Co-administration of reparixin on days 1 and 8 had
no effect on paclitaxel kinetics. Fifteen SAEs were recorded, none of which was related to Reparixin. Grade 3-4 adverse reactions
were recorded in 30% (10/33) patients including haematological toxicity (5/10). Only one patient discontinued treatment for a
reversible GI adverse reaction due to reparixin at the 1200 mg dose level. To date, 5 confirmed responses (2 CR, 3 PR) were
recorded among 18 patients who underwent at least 1 tumor assessment (at 8 weeks). Response duration was 20m+ and
3m+(for CR) and 9m+, 6m+, 2m+ (for PR). Final data will be presented at the meeting.
Conclusions. Combination treatment was safe and well tolerated at all dose levels without evidence of pharmacologic
interactions and the recommended dose for subsequent studies is 1200 mg tid. Efficacy was demonstrated both in hormone
receptor positive and triple receptor negative disease. A randomized phase II study of the combination versus single agent weekly
paclitaxel in patients with MBC is warranted.

Title: Functional genomic screening identifies USP11 as a novel therapeutic target in breast cancer
Darran P O'Connor1, Lisa Dwane1, Aisling E O'Connor1, Annette M Dirac2, Rene Bernards2 and William M Gallagher1. 1University
College, Dublin, Ireland and 2Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands.
Body: The estrogen receptor (ER) is the principal driver of growth and differentiation in breast cells and de-regulated receptor
function is a key feature of almost 75% of breast cancers. Here, we investigated the role of de-ubiquitinating enzymes (DUBs),
which act to remove ubiquitin moieties from proteins, in regulating the transcriptional activity of ER in breast cancer.
To identify DUBs involved in the regulation of ER transcriptional activity, we performed an RNAi loss-of-function screen using a
library of shRNA vectors targeting all known human DUB genes (108 genes/432 shRNAs in total). We found that suppression of a
number of DUBs markedly repressed or enhanced the activity of an estrogen-response-element (ERE) luciferase reporter to
estradiol (E2). In particular, suppression of the BRCA2-associated DUB, USP11, was found to down-regulate ER transcriptional
activity (both in the presence and absence of E2), as demonstrated by a pronounced decrease in estrogen-response element
(ERE) luciferase reporter activity. Subsequent validation of the screen using multiple individual hairpins and ZR-75-1 stable
USP11 knockdown cells confirmed the suppression of ERE-reporter activity and further revealed a notable reduction in
expression of the endogenous ER target genes TFF1 and PgR following USP11 knockdown. In vitro phenotypic analysis also
revealed a global decrease in cell survival, decreased ERK and AKT phosphorylation and increased sensitivity to DNA-damaging
agents in USP11 knockdown cell lines compared to non-targeting controls.
In silico analysis of publically available breast cancer gene expression datasets revealed a highly significant association between
high USP11 mRNA levels and poor prognosis. We observed a highly significant correlation between high expression of USP11
mRNA in ER positive patients and poor overall survival (OS)(HR 1.51, CI 1.07-2.14, p=0.018) and distant metastasis-free survival
(DMFS)(HR 1.35, CI 1.04-1.73, p=0.023). This correlation was also significant in ER positive patients who had received endocrine
therapy (OS, HR 3.4, CI 1.2-9.81, p=0.014), DMFS, HR 2.16, CI 1.23-3.8, p=0.0083). These results suggest a role for USP11 in
driving cellular growth and identify USP11 as a rationale and novel therapeutic target in breast cancer.

Title: Enrichment of janus kinase-2 (JAK2)-amplified tumor cell populations in triple-negative breast cancers (TNBC) during
chemotherapy treatment
Luis J Schwarz1, Justin M Balko1, Jennifer M Giltnane1, Monica V Estrada1, Melinda E Sanders1, Kai Wang2, Nancy U Lin3, Vicent
A Miller2, Philip Stephens2, Roman Yelensky2, Joseph A Pinto4, Henry L Gmez5, Melissa D Landis6, Jenny C Chang6 and Carlos
L Arteaga1. 1Vanderbilt University, Nashville, TN; 2Foundation Medicine, Cambridge, MA; 3Harvard University, Cambridge, MA;
4
Oncosalud-AUNA, Lima, Peru; 5Instituto Nacional de Enfermedades Neoplsicas, Lima, Peru and 6Houston Methodist Hospital,
Houston, TX.
Body: Neoadjuvant chemotherapy (NAC) is frequently used in the treatment of triple-negative breast cancer (TNBC).
Approximately 30% of TNBC patients achieve pathological complete response with excellent prognosis. However, the remaining
70% are at increased risk of recurrence with no molecularly targeted therapeutic options in the adjuvant setting. We performed
targeted next-generation sequencing in residual TNBC tumors after NAC to identify potentially actionable genomic alterations. As
previously reported, we found amplifications in JAK2 (JAK2AMP) in 7/68 (10.2%) post-NAC tumors compared with 1/30 (3%) in
the pre-NAC cohort; all cases were confirmed by JAK2fluorescence in situ hybridization (FISH). In contrast, JAK2AMP was
detected in 1% or less of primary untreated breast tumors in other independent cohorts.
Patients with JAK2AMP after NAC compared with JAK2NORMAL patients were younger (39.9yo vs 47.6yo), more frequently
pre-menopausal (71.4% vs 57.9%), and had little or no anti-tumor response to NAC (Miller & Payne I: 42.9% vs 15.8%). RNA
expression of the JAK2-activating ligand interleukin-6 (IL6) was also higher in these patients (p=0.008).
In matched untreated and post-NAC specimens, including 2 patient derived xenograft models generated from a single patients
pre- and post-NAC specimens, FISH analysis identified a subpopulation of tumor cells with JAK2AMP that was enriched after
NAC treatment. Preliminary data from RNA in situ hybridization (RNAScope) showed that tumor cells expressing high levels of
JAK2 are distinct from their counterparts, with higher IL6 expression, suggesting a paracrine signaling event.
We also explored the intra-tumor heterogeneity of JAK2AMP in cell lines and patient tumors, performing chemical and genetic
loss of function studies in 4 TNBC cell lines: HCC-1143 (JAK2DELETED/LOW), SUM-159PT (JAK2HIGH), HCC-38 (JAK2GAIN,
HIGH) and HCC-70 (JAK2AMP, HIGH).
In vitro, IL6 expression after adriamycin and/or docetaxel treatment was higher in the JAK2HIGH cell lines compared with the
changes registered in HCC-1143 (around 2, 4 and 100 fold increase from their respective IL-6 basal levels in HCC-38, HCC-70
and SUM159PT respectively, compared with 0.5 fold increase in HCC-1143).Treatment with chemotherapy also abrogated the
IL-6 downregulation produced by the JAK1/2 inhibitor ruxolitinib. Ruxolitinib decreased mammosphere formation by 50% in
SUM-159PT cells, with a 10% reduction of the CD24low/CD44high stem cell compartment. Interestingly, we observed no change
with ruxolitinib treatment in the JAK2GAIN/AMP cell lines. However, in HCC-38, siRNA knockdown of JAK2 reduced the
CD24low/CD44high compartment (around 15%) and mammosphere formation (around 80%).
These findings suggest that a JAK2AMP cell population may escape from chemotherapy-induced apoptosis resulting in lack of
response to treatment and eventual disease recurrence. Furthermore, chemotherapy may induce a wound-healing response in
the tumor, upregulating and/or selecting JAK2AMP cells via a paracrine or autocrine IL-6/JAK1/pSTAT3 signal. In vivo models
confirming these observations and further exploring the therapeutic potential of ruxolitinib to treat JAK2AMP TNBCs are
underway.

Title: TGF- receptor type III is a tumor promoter in mesenchymal-stem like triple negative breast cancer
Bojana Jovanovic1, J Scott Beeler1, Michael W Pickup2, Anna Chytil1, Agnieszka E Gorska1, William J Ashby1, Brian D Lehmann1,
Andries Zijlstra1, Jennifer A Pietenpol1 and Harold L Moses1. 1Vanderbilt University, Nashville, TN and 2University of San
Francisco, San Francisco, CA.
Body: Introduction: There is a major need to better understand the molecular basis of triple negative breast cancer (TNBC) in
order to develop effective therapeutic strategies. Using gene expression data from 587 TNBC patients we previously identified six
subtypes of the disease, among which a Mesenchymal-Stem Like (MSL) subtype. The MSL subtype has significantly higher
expression of the transforming growth factor beta (TGF-) pathway-associated genes relative to other subtypes, including the
TGF- receptor type III (TRIII). We hypothesize that TRIII is tumor promoter in mesenchymal-stem like TNBC cells.
Methods: Representative MSL cell lines SUM159, MDA-MB-231 and MDA-MB-157 were used to study the roles of TRIII in the
MSL subtype. We stably expressed short hairpin RNAs specific to TRIII (TRIII-KD). These cells were then used for xenograft
tumor studies in vivo; and migration, invasion, proliferation and three dimensional culture studies in vitro. Furthermore, we utilized
human gene expression datasets to examine TRIII expression patterns across all TNBC subtypes.
Results: TRIII was the most differentially expressed TGF- signaling gene in the MSL subtype. Silencing TRIII expression in
MSL cell lines significantly decreased cell motility and invasion. In addition, when TRIII-KD cells were grown in a three
dimensional (3D) culture system or nude mice, there was a loss of invasive protrusions and a significant decrease in xenograft
tumor growth, respectively. In pursuit of the mechanistic underpinnings for the observed TRIII-dependent phenotypes, we
discovered that integrin-2 was expressed at higher level in MSL cells after TRIII-KD. Stable knockdown of integrin-2 in
TRIII-KD MSL cells rescued the ability of the MSL cells to migrate and invade at the same level as MSL control cells.
Conclusions: We have found that TRIII is required for migration and invasion in vitro and xenograft growth in vivo. We also show
that TRIII-KD elevates expression of integrin-2, which is required for the reduced migration and invasion, as determined by
siRNA knockdown studies of both TRIII and integrin-2. Overall, our results indicate a potential mechanism in which TRIII
modulates integrin-2 expression to effect MSL cell migration, invasion, and tumorigenicity.

Title: Dual HER2 blockade of an activating driver HER3 mutation: A proof of principle study in a metastatic breast cancer patient
Francois-Clement Bidard1, Ck Ng2, Ezgi Tulukcuoglu1, S Piscuoglio2, Stephanie Descroix1, Laurent Mignot1, Jean-Louis Viovy1,
Paul Cottu1, Brigitte Sigal1, Anne Vincent-Salomon1, Britta Weigelt2, Jean-Yves Pierga1 and Jorge Reis-Filho2. 1Institut Curie,
Paris, France and 2Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background: HER3 is the only member of the EGFR family with an impaired kinase domain. Pre-clinical work has
suggested that activating hotspot point mutations in HER3/ERBB3 may constitute drivers of tumor progression. Mutant HER3
oncogenic activity was shown to be dependent on HER2 signaling. In vitro and in vivo experiments have suggested that
HER2-targeting therapies could be effective to block mutant HER3 oncogenic signaling. Through high-depth whole exome
sequencing analysis of the primary breast cancer and liver metastasis of a patient with stage IV disease, we identified a clonal,
activating HER3 mutation in both the primary tumor and metastasis, and sought to define whether this mutation would constitute a
driver of the disease and be a therapeutic target.
Patient and method: A 46 yo women was diagnosed in 2012 with de novo synchronous metastatic breast cancer. Biopsies of
the primary breast cancer and of the liver metastasis were performed before any treatment within the ESOPE prospective study
(NCT01956552). Pathologic examination revealed a high-grade invasive ductal carcinoma with low expression of estrogen and
progesterone receptors (both 20%) and intermediate HER2 (++) expression by immunohistochemical analysis without HER2
gene amplification by fluorescence in situ hybridization. While high-depth (250x) whole exome massively parallel sequencing
(HiSeq) of the primary tumor and liver metastasis was ongoing, the patient received two lines of chemotherapy (anthracyclineand taxane-based). Data analysis revealed that both the primary tumor and the liver metastasis displayed a clonal G284R HER3
mutation at a high allelic frequency compatible with the distribution of a driver genetic alteration. Single nucleotide polymorphism
(SNP6) arrays and sequencing confirmed the lack of HER2 gene amplification. At time of progression, after two lines of
chemotherapy (05/2014), the patient had a baseline blood draw and PET-CT and was administered trastuzumab (6 mg/kg q3w) in
combination with lapatinib (1250mg qd), with no chemotherapy.
Results: Treatment was well tolerated with no dose reduction. Protein-ligation assay (PLA) by rolling-circle amplification
performed at baseline showed HER2-HER3 heterodimers in circulating tumor cells. After only 15 days of treatment, PET-CT
displayed a complete metabolic response and a slight decrease of the diameter of liver metastasis (-20%). At day 21, no CTCs
were detected in 7.5ml of peripheral blood and CA15.3 dropped from 45 (baseline) to 32 (upper limit = 30). ctDNA was also
collected and is currently being analyzed. Dual blockade has been pursued; updated follow-up will be presented at the meeting.
Conclusion: HER3 mutations are particularly rare in breast cancer (<1% of breast cancers) and no trial is currently assessing the
relevance of HER2-blockade in this subgroup. On the basis of an extreme response to dual HER2 blockade with no
chemotherapy, we confirm the published preclinical data and suggest that these patients may benefit from anti-HER2 therapies. A
prospective global registry for HER3 mutated breast cancers cases and their response to specific systemic therapies may
facilitate the accrual of data to support the use of anti-HER2 agents in this patient population.

Title: Therapeutic targeting of ER in triple negative breast cancer
Jordan M Reese1,2, Malayannan Subramaniam1, Vera J Suman3, Xianglin Wu1, Vivian Negron4, Anne Gingery1, Kevin S Pitel1,
Sejal S Shah4, Heather E Cunliffe5, Ann E McCullough6, Barbara A Pockaj7, Thomas C Spelsberg1, Matthew P Goetz2,8, James N
Ingle8 and John R Hawse1. 1Mayo Clinic, Rochester, MN; 2Mayo Clinic, Rochester, MN; 3Mayo Clinic, Rochester, MN; 4Mayo
Clinic, Rochester, MN; 5Dunedin School of Medicine, University of Otago, Dunedin, New Zealand; 6Mayo Clinic, Scottsdale, AZ;
7
Mayo Clinic, Scottsdale, AZ and 8Mayo Clinic, Rochester, MN.
Body: Background: While the biological functions and clinical importance of ER are well understood in breast cancer, much
less is known about its most closely related family member, ER, particularly in the setting of triple negative disease. Additionally,
the ability to therapeutically target ER in triple negative breast cancer (TNBC) has not been fully explored.
Methods: Expression of ER protein was determined using a well characterized and validated ER specific monoclonal antibody
that only recognizes the full-length form of this receptor (PPG5/10) in a cohort of 71 TNBCs. To further define the biological
functions of ER in TNBC, novel ER expressing triple negative cell lines (MDA-MB-231 and Hs578T) were developed and
comprehensively characterized at the level of global gene expression profiling, modulation of important biological pathways,
cellular proliferation and response to targeted therapies.
Results: In TNBCs from 71 patients, nuclear and cytoplasmic ER was detected at moderate to high levels in 24% and 32% of
cases respectively. This moderate to high expression of both nuclear and cytoplasmic ER was associated with higher levels of
Ki67. Of the 17 tumors expressing ER, 13 (76%) were negative for androgen receptor expression. In the triple negative
MDA-MB-231-ER and Hs578T-ER cell lines, expression of ER led to inhibition of proliferation in response to both estrogen
and multiple ER specific agonists. Microarray analysis and RT-PCR profiling of these cells revealed that estrogen and ER
agonists highly induced the expression of multiple cystatins, a family of small secreted cysteine protease inhibitors, while
suppressing the expression of many interleukins. Conditioned media isolated from estrogen or ER agonist treated
MDA-MB-231-ER cells inhibited the proliferation rates and blocked TGF signaling in non-ER expressing TNBC cells, effects
that were completely reversed following depletion of cystatins from the conditioned media.
Conclusions: ER is expressed in a substantial proportion of TNBCs, and most do not express the androgen receptor. In TNBC,
where targeted therapies are lacking, our data suggest that estrogen or ER specific agonists would be expected to elicit
anti-tumor effects when ER is expressed. These tumor suppressive effects of ER appear to be mediated in part through the
actions of cystatins, and their inhibition of TGF signaling, suggesting that this family of secreted proteins may represent novel
biomarkers for monitoring ER specific drug responsiveness and/or patient outcomes.

Title: Targeting multiple pathways in breast cancer: Androgen receptor, HER2, and mTOR
Michael A Gordon1, Nicholas D'Amato1, Haihua Gu1,3, Darren Wong4, Anthony Elias2 and Jennifer K Richer1. 1University of
Colorado Anshutz Medical Campus, Aurora, CO; 2Wenzhou Medical University, University-town, Wenzhou, China and
3
Medivation Inc, San Francisco, CA.
Body: Background: Androgen receptor (AR) is expressed in 60% of breast cancers (BC) and current evidence indicates that
some BC become reliant on AR signaling for growth. AR activation affects numerous cellular functions and signaling cascades,
including the HER2/PI3K/mTOR pathway. Conversely, the HER2/PI3K/mTOR pathway regulates steroid hormone receptors
including AR. Therapeutics targeting this pathway such as the anti-HER2 trastuzumab (TRAS) and the anti-mTOR everolimus
(EVE) have shown significant clinical benefit; however the impact that AR inhibition may have on enhancing the efficacy of these
agents and a potential mechanism of synergy have not been explored in BC. The anti-androgen enzalutamide (ENZ) inhibits
nuclear entry of AR, is clinically effective in castrate-resistant prostate cancer, and is currently being tested in clinical trials for BC.
Hypothesis: Given the overlap in these pathways, simultaneous inhibition of AR pathway with agents targeting HER2/PI3K/mTOR
may be more effective in BC therapy.
Methods: HER2+ and triple-negative BC (TNBC) cell lines were treated with ENZ, TRAS, and EVE, either alone or in
combination, at three clinically relevant doses per drug. Cell proliferation was measured either by crystal violet (MDAMB453,
SUM225) or on an IncuCyte live cell imager (Essen Bioscience) with nuclear-red labeled cells (BT474, SKBR3, BT549). Drug
synergy was calculated using Calcusyn (Biosoft, Inc.) from IncuCyte proliferation data for selected cell lines (BT474, SKBR3,
BT549) by comparing combinations of multiple drug concentrations where a combination index (CI)<1 indicated synergy. Protein
expression of pathway components was measured by western blot, and gene expression was measured by RT-qPCR.
Results: ENZ inhibited proliferation of HER2+ BC cells, and a combination of ENZ plus TRAS inhibited proliferation more
effectively than either agent alone. The effect was synergistic in SKBR3 cells (CI<0.1), and additive in BT474 cells. Treatment of
a HER2+/ER+ cell line (BT474) as well as an ER-, HER2-non-amplified but overexpressing cell line (MDAMB453) with
dihydrotestosterone (DHT) caused increased expression of pHER3 and total HER3 protein, which was attenuated by addition of
ENZ. However, in HER2+/ER- cell lines (SKBR3 and SUM225), pHER3 and total HER3 levels did not change, but rather pHER2
levels increased in response to DHT and were attenuated with ENZ treatment. Additionally, ENZ inhibited proliferation of
TRAS-resistant SKBR3 cells. However, unlike the parental SKBR3 cells, ENZ did not affect pHER2 levels in the TRAS-resistant
cells. ENZ plus EVE treatment resulted in a significant decrease in proliferation compared to single agent; this effect was
synergistic in two HER2+ cell lines (SKBR3 CI<0.1, BT474 CI<0.1) as well as one TNBC cell line (BT549 CI<0.5). Treatment with
EVE alone caused a compensatory increase in AR protein and downstream AR gene expression in all cell lines. This increase
was attenuated with ENZ plus EVE combination treatment.
Conclusion: ENZ synergizes in vitro with FDA-approved BC therapies TRAS and EVE through distinct mechanisms. Combination
therapies containing ENZ may provide benefit for multiple BC subtypes, including HER2+ and triple negative BC.
Funded by DOD BCRP Clinical Translational Award BC120183 to JKR.

Title: Androgen receptor (AR): A novel target for radiosensitization and treatment in triple-negative breast cancers (TNBC)
Corey Speers1, Shuang G Zhao1, Meilan Liu1, Joseph Evans1, Prasanna Alluri1, Daniel F Hayes1, Felix Y Feng1 and Lori J Pierce1.
1
University of Michigan Hospital and Health System, Ann Arbor, MI.
Body: Background: Increased rates of locoregional recurrence have been observed in TNBC despite the use of chemotherapy
and radiation (RT). Thus, approaches that result in radiosensitizaton of TNBC are critically needed. We have previously
characterized the radiation response of 21 breast cancer cell (BCC) lines using clonogenic survival assays. We now pair this data
with high-throughput drug screen data available through cancer cell line encyclopedia studies to identify AR as a top target for
radiosensitization and assess AR inhibition as a radiosensitization strategy for TNBC.
Methods: Clonogenic survival assays were performed to determine the intrinsic RT sensitivity of 21 BCC lines (0-8 Gy RT). IC50
values were determined for 130 clinically available compounds and correlation coefficients were calculated using IC50 values (for
drug sensitivity) and SF-2Gy (for radiation sensitivity). Gene expression was measured using Affymetrix microarrays and protein
expression was measured using reverse-phase protein lysate arrays (RPPA) of human tumor samples (n=2,061) and BCC lines
(n=51). AR function was assessed using siRNA knockdown or inhibition with MDV3100 (enzalutamide). Kaplan-Meier analysis
was performed to determine the clinical impact of AR expression on local control and survival. A Cox proportional hazards model
was constructed to identify potential factors of survival, and multivariate analysis was used to determine variables most
significantly associated with LRF survival.
Results: Our radiosensitizer screen nominated bicalutamide as one of the most effective drugs in treating radioresistant BCC lines
(R2= 0.46, p-value <0.001). Recognizing that a subgroup of TNBC includes AR expressing tumors, we interrogated the
expression of AR in >2000 human breast tumor samples and found significant heterogeneity in AR expression with an increase
in TNBC (35% of tumors) compared to non-TNBC (28% of tumors). This same heterogeneity was also identified in human BCC
lines. There was a strong correlation between AR RNA expression and protein expression (R2= 0.72, p <0.0001). Inhibition of AR
using both siRNA and MDV3100 induced radiation sensitivity in vitro with an enhancement ratio (ER) of 1.35-1.42 in AR-positive
TNBC lines. No such radiosensitization was seen in AR-negative TNBC or ER-positive, AR-negative BCC lines.
Radiosensitization was at least partially dependent on impaired dsDNA break repair mediated by DNAPKcs. In vivo assessment
of tumor growth inhibition with RT and anti-AR strategies are currently underway. Clinically, analyses of patients with TNBC
showed that patients whose tumors had high expression of AR had markedly higher rates of LR after RT than patients with low
expression of AR (HR for LR 2.9-3.2, p-value <0.01, 2 independent datasets). There was no difference in LR in TNBC patients
not treated with RT when stratified by AR expression status. In multivariate analysis, AR expression was the variable most
significantly associated with worse LRF survival after RT with a HR of 3.58 (p-value < 0.01).
Conclusion: Our results implicate AR as a mediator of radioresistance in breast cancer and support the rationale for developing
clinical strategies to inhibit AR as a novel radiosensitizing target in TNBC.

Title: Targeting the c-myc/E2F1 pathway in TNBC promotes a DNA damage dependent synthetic lethality
Jason P Carey1 and Khandan Keyomarsi1. 1MD Anderson Cancer Center, Houston, TX.
Body: The c-myc/E2F1 pathway is highly upregulated in TNBC and dictates increased genomic instability, a higher frequency of
relapse and poor overall survival. C-myc is a potent oncogene with overexpression activity that influences several pathways
including metabolism, DNA replication, DNA damage, cell cycle and apoptosis. The essential role of c-myc in normal cells has
made therapeutic intervention elusive. However the role of c-myc influencing cell cycle progression has demonstrated that CDK
inhibition downregulates c-myc expression in TNBC breast cancer cells. We hypothesize that CDK inhibition downregulates
c-myc expression resulting in an accumulation of DNA damage thus increasing the susceptibility of TNBC cells to PARP
inhibition. A panel of TNBC cell lines was assessed for c-myc/E2F1 pathway expression and downregulation in the presence of
CDK inhibitor Dinaciclib. An MTT-based High Throughput Survival Assay (HTSA) was utilized to assess growth inhibition
response to drug treatment in TNBC cells. FACS analysis was performed on Dinaciclib treated TNBC cell lines to asses cell cycle
response to CDK inhibition. DNA damage and Homologous Recombination DNA repair was assessed by H2AX, BRCA1, RAD51
foci formation by immunofluorescence. Lentiviral knockdown of c-myc, E2F1, BRCA1 and RAD51 were performed to assess
susceptibility to PARP inhibition. Combination treatment of Dinaciclib + PARP inhibitor MK4827 was assessed via combination
index values calculated by Calcusyn. CDK inhibition resulted in downregulation of both c-myc and E2F1 in all TNBC cells. The
pan-CDK inhibitor Dinaciclib universally inhibited proliferation in all TNBC cell lines in the presence of the drug however drug
removal allowed TNBC cells to recover a normal proliferative profile. Dinaciclib induction increased H2AX foci formation while
decreasing BRCA1 and RAD51 foci formation resulting in an increase in DNA damage induced cell death that correlated with
c-myc downregulation. CDK inhibition correlated with downregulation of c-myc/E2F1 target genes. Lentiviral knockdown of c-myc
and not E2F1 induced susceptibility to PARP inhibitor MK4827. Combination CDK + PARP inhibitor resulted in a dose dependent
synthetic lethal increase in PARP inhibitor efficacy in both BRCA1 mutant and wild-type cells. Together these results suggest that
c-myc downregulation via CDK inhibition in combination with PARP inhibition may pose a novel combination therapy for TNBC
patients.

Title: Targeting voltage-gated sodium channels with the antiepileptic drug phenytoin inhibits breast tumor growth and metastasis
William J Brackenbury1, Michaela Nelson1, Ming Yang1, Adam A Dowle1 and Jerry R Thomas1. 1University of York, York, United
Kingdom.
Body: Voltage-gated Na+ channels (VGSCs) are heteromeric protein complexes containing pore-forming subunits and smaller,
non-pore-forming subunits. The subunits are multifunctional channel modulators and are members of the immunoglobulin
superfamily of cell adhesion molecules (CAMs). VGSCs are classically expressed in electrically excitable cells, e.g. neurons,
where they mediate action potential firing, neurite outgrowth and migration during development. An increasing body of evidence
indicates that VGSCs are also expressed in metastatic cells from a number of cancers, including breast cancer. The Nav1.5
subunit (encoded by SCN5A) is expressed in cancer cell lines, where it enhances migration and invasion. SCN5A is up-regulated
in tumor samples in several published datasets, associating with recurrence, metastasis, and reduced overall survival. We have
previously shown that the VGSC-blocking antiepileptic drug phenytoin inhibits the migration and invasion of metastatic
MDA-MB-231 cells in vitro. In addition, we have recently shown that the VGSC 1 subunit enhances breast tumor growth and
metastasis in a xenograft model of breast cancer. The purpose of the present study was to establish whether or not VGSCs might
be a viable therapeutic target by testing the effect of phenytoin on tumor growth and metastasis in vivo. We found that Nav1.5
expression was retained on MDA-MB-231 cells in orthotopic xenografts in immune-deficient mice. Mice were treated with
phenytoin (60 mg/kg) or vehicle by daily intraperitoneal injection for 3 weeks, starting 1 week following implantation of tumor cells.
Plasma phenytoin concentration was measured at the end of the experiment by liquid chromatography-mass spectrometry with
single reaction monitoring (LC-SRM-MS) using metaxalone as an internal standard. Phenytoin significantly reduced tumor growth,
detected by in vivo bioluminescent imaging and caliper measurements, without affecting animal weight. Phenytoin reduced the
density of Ki67-positive cycling cells in the primary tumors, but did not affect the density of apoptotic cells expressing activated
caspase-3, or the density of CD31-positive vessel structures. Phenytoin also reduced local invasion and the density of
MMP9-positive cells within primary tumors. Finally, phenytoin significantly reduced metastasis to the liver, lungs and spleen,
detected by bioluminescent imaging and GFP immunohistochemistry. This is the first study showing that phenytoin reduces tumor
growth and metastasis in vivo. Together, our data support the hypothesis that VGSCs are up-regulated in breast cancer, favoring
an invasive phenotype, and may thus be promising targets for therapeutic intervention. We propose that pharmacologically
targeting VGSCs should be further studied as a potentially novel, cost-effective, anti-cancer therapy. Repurposing FDA-approved
oral VGSC-blocking antiepileptic or antiarrhythmic drugs, e.g. phenytoin, may improve patient outcomes in the adjuvant setting.

Title: Erk5 as a therapeutic target in triple negative breast cancer (TNBC)
Mariska Sauzine Miranda, Fares Al Ejeh, Wei Shi, Peter Simpson, Sunil Lakhani and Kum Kum Khanna. 1University of
Queensland, Brisbane, Queensland, Australia; 2QIMR Berghofer Medical Research Institute, Brisbane, Queensland, Australia;
3
UQ Centre for Clinical Research (UQCCR), Brisbane, Queensland, Australia and 4Pathology Queensland, The Royal Brisbane &
Women's Hospital, Brisbane, Queensland, Australia.
Body: TNBC is a subtype of breast cancer known for its aggressive behavior and poor prognosis. TNBC accounts for 17-24% of
all invasive breast cancers and is a subtype that lacks the expression of hormone receptors; oestrogen [ER], progesterone [PR])
and epidermal growth factor receptor 2 (HER2) receptors. Currently there has been a shift from using empirically derived agents
that inhibit tumor cell growth and/or survival, to molecular therapies that target specific molecules that regulate these and other
important biological processes. Despite the success of some new-targeted therapies to treat breast cancer, the outlook for the
majority of patients with TNBC remains poor. Therefore we have utilized the KinexTM antibody array to resolve potential new
targetable "nodes". We have screened 43 primary breast cancer biopsies (16 TNBC, 16 ER/PR positive and 11 HER2-positive)
and 16 breast cancer cell lines for protein/phosphoprotein levels. Extracellular-signal-regulated kinase 5 (Erk5) is a member of the
MAPK family. Erk5 has a large, unique C-terminal-half not found in other MAPK family members and because of this unique
400-amino acid extension, ERK5 is also called big MAP kinase 1 (BMK1). By extensive in silico analyses, we identified that the
Erk5 metagene is prognostic and identifies TNBC/BLBC with poor prognosis. Our kinome study of primary breast cancer has
further confirmed that the MEK5-Erk5 pathway is upregulated in the TNBC subgroup compared to the other TNBC subgroups and
that pharmacological inhibition of Erk5 is therapeutic against primary MDA-MB-231 TNBC xenografts (1).
Our preliminary data has shown that Erk5 is overexpressed in the TNBC subtype at the protein level. We have also shown that
silencing Erk5 by siRNA or shRNA inhibits migration of TNBC cell lines in vitro. Expression of Erk5 (total and phosphoproteins)
will be investigated in primary breast tumours and metastases samples from patient by immunohistochemistry. We will also
overexpress Erk5 in non-metastatic breast cancer cell lines to study the effect of Erk5 overexpression on metastases. In an
elegant study, site directed mutagenesis was used to modify Erk5 and show that phosphorylations of the C-terminal domain of
kinase activated Erk5 is required for the transcriptional activity of Erk5. Using in vitro and in vivo models we will determine if the
transcriptional activity of activated Erk5 is a major player to drive metastatic phenotype. Moreover, whether kinase-dependent or
transcription-dependent activity of Erk5 on several pathways is sufficient to drive metastasis will also be determined.
Our results strongly suggest the MEK5-Erk5 pathway as a prognostic and therapeutic pathway in a poor-prognosis subset of
TNBC/BCLC. Characterization of this pathway in clinical cohorts particularly in TNBC tumors, mechanistic understanding of its
contribution to metastasis and therapeutic targeting in vivo would help develop a personalised approach to identify and treat
TNBC tumors and metastases with activated Erk5 pathway.
1. Al-Ejeh F, Miranda M et al. Kinome profiling reveals breast cancer heterogeneity and identifies targeted therapeutic
opportunities for triple negative breast cancer. Oncotarget. March 19, 2014.

Title: High MELK expression levels correlate with shorter overall survival in breast cancer
Takahiro Kogawa1,2, Diane D Liu4, Gaurav B Chauhan1,2, Juliana M Taliaferro3, Hiroko Masuda1,2, Steven J Van Laere5,6, Franois
Bertucci7, Yu Shen4, Naoto T Ueno1,2, Kevin N Dalvy3 and Chandra Bartholomeusz1,2. 1Morgan Welch Inflammatory Breast Cancer
Research Program and Clinic, University of Texas MD Anderson Cancer Center, Houston, TX; 2University of Texas MD Anderson
Cancer Center, Houston, TX; 3College of Pharmacy, University of Texas, Austin, TX; 4University of Texas MD Anderson Cancer
Center, Houston, TX; 5Translational Cancer Research Unit-Antwerp, Oncology Center, General Hospital Sint-Augustinus, Wilrijk,
Belgium; 6Laboratory of Gynecological Oncology, Catholic University, Leuven, Belgium and 7Institut Paoli-Calmettes, Marseille,
France.
Body: Background
Triple-negative breast cancer (TNBC) is thought to relapse and metastasize because cancer stem cells (CSCs) resist
conventional chemotherapy and radiotherapy and later give rise to secondary tumors. MELK (maternal embryonic leucine zipper
kinase), a protein kinase of the Snf1/AMPK kinase family, is known to play a critical role in promoting cell proliferation, CSC
maintenance, apoptosis, and transformation. MELK is frequently upregulated in basal-like breast cancers but is not expressed in
normal vital organs (Lin et al, Breast Cancer Res 2007;9:R17; Wang et al, eLife 2014;10.7554/eLife.01763). We hypothesized
that MELK is upregulated in TNBC and that high expression correlates with poor progression-free survival (PFS), distant
metastasis-free survival (DMFS), and overall survival (OS) in breast cancer.
Methods
This retrospective study involved the World Inflammatory Breast Cancer Consortium dataset comprising 389 patients from 3
institutions (The University of Texas MD Anderson Cancer Center, General Hospital Sint-Augustinus, and Institut
Paoli-Calmettes) with clinical and tumor data and gene expression (GE) profiles. We studied the 314 patients with stage I-III
breast cancer (84 TNBC and 230 non-TNBC) within this dataset. Hormone receptor (HR) and HER2 status was defined using the
GE data [ESR1 (probe set 205225_at), PGR (208305_at), ERBB2 (216836_s_at)]. We defined high MELK mRNA GE (probe set
204825_at) as 7.0 using a Martingale residual plot. Time-to-event endpoints were summarized using the Kaplan-Meier method
and compared between or among groups using the log-rank test. Multicovariate Cox proportional hazard models were applied to
assess the effect of covariates on survival endpoints.
Results
Median age of patients was 54 yrs (range, 24-89). Disease stage was III in 48.4% of patients, II in 30.9%, and I in 20.7%. Most
tumors had ductal histology (82.5%) and almost half had nuclear grade III (49.7%). MELK expression was significantly higher in
TNBC than in the HR+HER2-, HR+HER2+, and HR-HER2+ subtypes (p<0.0001, Kruskal-Wallis test). Median follow-up was 5.7
years. In univariate analysis, patients with MELK-overexpressing tumors had significantly shorter PFS, DMFS, and OS times than
patients with low MELK expression (table).
In multicovariate Cox regression model, high MELK expression did not have independent prognostic value for PFS (p=0.37) or
DMFS (p=0.29); however, compared with low expression, was associated with a higher risk of death (hazard ratio=1.79; 95%
CI=1.11-2.89; p=0.02), adjusted for tumor stage, IBC status and TNBC status and stratified by the study center.
Conclusion
High MELK expression was an independent prognostic factor of OS in breast cancer. Our preliminary results and those of others
suggest that MELK may be an important therapeutic target for breast cancer, particularly TNBC, where the MELK expression
level was significantly higher than in other subtypes; this warrants further investigation with a large dataset. This study justifies
developing MELK inhibitors in TNBC.
5-year DFS, DMFS and OS rates
Endpoint
MELK GE
Survival Probability
95%CI
Log-rank test p value
RFS
<7
73.3
64.7-80.1
0.0112
51.8
43.2-59.8
DMFS
OS
<7
75.3
66.8-82.0
53.5
44.8-61.5
<7
84.3
76.3-89.7
62.7
53.9-70.4
0.0081
0.0002

Title: Inhibition of Src increases radioiodide uptake in breast cancer cells by inhibiting phosphorylation of pituitary tumor
transforming gene binding factor (PBF)
Vikki L Poole1, Martin L Read1, Rachel J Watkins1, Bhavika Modasia1, Waraporn Imruetaicharoenchoke1, Kristien Boelaert1, Vicki
E Smith1 and Christopher J McCabe1. 1University of Birmingham, Birmingham, West Midlands, United Kingdom.
Body: Although not detectable in normal breast tissue, the sodium iodide symporter (NIS) has been found to be expressed in
70-80% of breast cancers. However, the majority of NIS is intracellular, leaving only 20-30% functional at the plasma membrane.
Whilst radioiodine therapy has been proposed as a potential treatment for breast cancer, effective therapy would require
increased levels of membranous NIS localisation in tumours. Previous work revealed that overexpression of pituitary tumor
transforming gene binding factor (PBF) in thyroid cells leads to the redistribution of NIS from the plasma membrane into
intracellular vesicles, thereby reducing radioiodide uptake, a process modulated by Src phosphorylation of PBF. Here we show
that PBF and NIS have a consistent relationship in breast cancer, with phosphorylation of PBF at residue Y174 being critical for
the association. Immunofluorescent microscopy revealed co-localisation between NIS and PBF in co-transfected MDA-MB-231,
MCF-7 and T47D cells, with increased intracellular staining for NIS compared to cells transfected with NIS alone. Phosphorylated
PBF was also observed to co-localise with NIS in T47D cells. Treatment with PP1, a Src inhibitor which modulates the
phosphorylation of PBF, led to increased NIS plasma membrane staining and less intracellular co-localisation with PBF.
Functional studies in MCF-7 and MDA-MB-231 cells demonstrated that PBF significantly repressed radioiodide uptake in cells
expressing exogenous NIS (25% and 30% reduction respectively; n=3, p<0.05). Treatment with PP1 restored the ability of MCF-7
and MDA-MB-231 cells to uptake I-125 (1.24 and 1.69 fold increase respectively; n=3, p<0.05). Combined all-trans retinoic acid
(ATRA) and dexamethasone treatment has previously been shown to enhance NIS expression and radioiodide uptake in MCF-7
cells. Importantly, in the face of NIS-induction via ATRA and dexamethasone, PBF retained its ability to repress iodide uptake
(31% reduction; n=3, p<0.05). This repression was overcome by PP1 treatment, which restored radioiodide uptake to vector only
transfected levels (1.22-fold increase; p=ns, n=3). In keeping with PP1 data, the Src inhibitors saracatinib and dasatinib inhibited
the phosphorylation of PBF in MCF-7 and MDA-MB-231 cells. Radioiodide uptake studies revealed that 1nM dasatinib was
capable of restoring the ability of MDA-MB-231 cells to uptake I-125 (2 fold increase; n=3, p<0.05) when expressing exogenous
NIS and PBF compared to vehicle only treated cells. 10nM saracatinib was also capable of restoring I-125 uptake but with less
potency than dasatinib. Inhibition of focal adhesion kinase (FAK) with PF573228 failed to ameliorate PBFs repression on I-125
uptake, suggesting that FAK is not responsible for PBF phosphorylation. Taken together, these data suggest that PBF alters the
subcellular location of NIS and potently represses its activity in breast cancer cells. Further, tyrosine phosphorylation of PBF by
Src modulates the ability of breast carcinoma cell-lines to take up radioiodide, with important implications for adapting NIS as a
potential therapy in breast cancer.



Title: Melatonin on angiogenesis in breast cancer
Debora C Zuccari1, Bruna V Jardim-Perassi1, Mateus R Loureno1, Gabriel M Doho1, Gabriela B Gelaleti2, Lvia C Ferreira2, Thaiz
F Borin2 and Marina G Moschetta1. 1Faculdade de Medicina de So Jos do Rio Preto, FAMERP, So Jos do Rio Preto, So
Paulo, Brazil and 2Universidade Estadual Paulista, So Jos do Rio Preto, So Paulo, Brazil.
Body: The rapid tumor growth results in hypoxia on tumor microenvironment, leading to a cascade of events that induce
angiogenesis and subsequent cancer progression. Thus, the identification of therapeutic agents that can inhibit angiogenesis is
essential for the control of tumor progression. Exogenous administration of melatonin, a hormone secreted by the pineal gland,
has been shown several oncostatics effects on different types of cancers. The aim of this study was to evaluate the effectiveness
of melatonin treatment on angiogenesis in breast cancer, in the in vitro and in vivo studies. In the in vitro study, breast cancer cell
lines (MCF-7 and MDA-MB-231) were treated with melatonin under cobalt chloride (CoCl2)-induced hypoxic conditions. Cell
viability was measured by MTT assay, the expression of hypoxia-inducible factor 1-alpha (HIF-1) and vascular endothelial
growth factor (VEGF-A) was assessed by real-time PCR and immunocytochemistry. Additionally, other proteins involved in
angiogenesis were evaluated by the Protein Array. In the in vivo study, the MDA-MB-231 cells were implanted in athymic nude
mice, which were treated with melatonin (40 mg/kg) for 21 days. The tumor was measured weekly and evaluation of angiogenesis
was performed by single-photon emission computed tomography (SPECT) with Tc-99m-HYNIC-VEGF-c, which is specific for
VEGF receptors (VEGFR2/VEGF3). Moreover, VEGFR2, VEGFR3, von Willebrand factor (vWF) and cell proliferation marker
(Ki-67) were evaluated in tumor tissue by immunohistochemistry, and other angiogenic proteins by Protein Array. Results from
the in vitro study showed that 1 mM of melatonin under hypoxic conditions (200 M CoCl2) led to decreased cell viability, protein
levels of HIF-1 and gene and protein expression of VEGF-A in both cell lines (p < 0.05). Among other proteins evaluated,
melatonin treatment under hypoxia resulted in a decrease of VEGF-C, VEGFR2, VEGFR3, matrix metalloproteinase 9 (MMP-9)
and angiogenin in MCF-7 cell line (p < 0.05). For MDA-MB-231, a significant reduction was observed for VEGFR2 protein,
epidermal growth receptor (EGFR), and angiogenin (p <0.05). In vivo study, mice treated with melatonin showed reduced tumor
growth compared to control animals (144.90 38.38 mm3 vs 282.00 88.53 mm3, p < 0.05). Furthermore, one animal showed
tumor regression during melatonin treatment (Day 7 = 27.38 mm3, 8.79 mm3 Day = 14, mm3 Day 21 = 4.8). SPECT detect less
radioactivity of Tc-99m-HYNIC-VEGF-c and consequent reduced expression of VEGFR2/3 in tumors treated with melatonin
(150,46 17,06 % vs 183,55 20,92 %) but statistical significance was not achieved (p > 0.05). The reduction of VEGFR2 in
tumors treated with melatonin was confirmed by immunohistochemistry (p <0.05), as well as the reduction of micro-vessel density
(vWF) and cell proliferation (ki-67) (p < 0.05). There was no change in the expression of the other proteins evaluated, however a
significant increase in EGFR and Insulin-like growth factor (IGF-I) was observed in tumors treated with melatonin (p< 0.05).
Taken together, our results showed that melatonin has an important anti-angiogenic effect, suggesting its potential therapeutic
action in breast cancer.
Support: FAPESP.

Title: No Show

Title: Endoplasmic reticulum resident protein transmembrane protein 33 (TMEM33) induces apoptosis via UPR signaling and
autophagy in breast cancer cells
Rong Hu1, Xiyuan Zhang1, Leena Hilakivi-Clarke1, Usha Kasid1 and Robert Clarke1. 1Georgetown University, Washington, DC.
Body: Breast cancer is the most common cancer type in women, with expected incidence of around 232,670 new cases in the
US alone each year. Endoplasmic reticulum stress (EnR stress) and the related unfolded protein response (UPR) are activated in
breast cancer cells to promote cell survival and endocrine resistance. TMEM33 is a novel transmembrane protein that resides in
the endoplasmic reticulum (EnR) and has been shown to activate two of the UPR branches (PERK; IRE1). However, the cellular
functions and underlying mechanism of this EnR resident protein remains largely unknown. In this study, we show that
overexpression of TMEM33 induces robust cell death in breast cancer cells. TMEM33 overexpression strongly activates the
pro-death JNK-p53 pathway, possibly through the activation of IRE1. We also observe a significant inhibition of the downstream
survivin (also called BIRC5), which is known to inhibit cell death by binding to caspases and blocking their activation. We further
show that either the blockage of JNK activation with a small molecule inhibitor, or the overexpression of survivin, protects cells
against TMEM33-induced apoptosis. In addition, we demonstrate that TMEM33 overexpression induces autophagy in breast
cancer cells (changes in LC3 and p62 expression). Inhibition of autophagy with either the autophagy inhibitor chloroquine, or by
knockdown of the autophagy gene Atg5, further sensitizes breast cancer cells to TMEM33 overexpression. Conversely, cell death
induced by TMEM33 is decreased by overexpression of the autophagy gene Beclin 1. The findings in this study demonstrate that
the novel EnR resident protein TMEM33 induces cell death by activating the IRE1-JNK-p53-survivin pathway in breast cancer
cells. Paradoxically, autophagy is also activated by TMEM33 but the prosurvival action is generally not sufficient to block all of the
cell death signaling.

Title: Expression of ING1b-derived peptide inhibits viability of breast cancer cells and promotes apoptosis
Oleksandr Boyko1 and Karl Riabowol1. 1University of Calgary, Calgary, AB, Canada.
Body: The ING1b protein is a type II tumor suppressor and a stoichiometric member of HDAC-containing protein complexes.
Primarily by altering chromatin structure, ING1b contributes to regulation of gene expression, senescence and apoptosis.
Mislocalization and decreased levels of ING1b are commonly observed in human tumors and cancer cell lines. Multiple
independent studies in varied cell types report that ING1b overexpression promotes apoptosis. Since the inactivation of apoptosis
pathways is frequent in cancer cells, modulating ING1b expression in tumors may serve as a viable approach for cancer therapy.
We have defined the minimal ING1b region that confers apoptotic function as estimated by the levels of PARP cleavage, FACS
analysis and using an Annexin V binding assay. We have established that ING1b-derived peptides containing its third alpha helix
(A3H) and NLS/NTS domains are able to induce apoptosis at levels comparable to those of full length ING1b. The A3H-NLS/NTS
peptide exhibited strong nucleolar localization characteristic of full length ING1b. Cells overexpressing the full length ING1b and
A3H-NLS/NTS peptide showed similar changes in cell morphology characteristic of apoptosis and exhibited increased levels of
PARP cleavage. While the A3H region that includes the N-terminal part of the highly conserved Lamin Interacting Domain (LID)
was necessary but not sufficient, the NLS/NTS domain that mediates nuclear and nucleolar localization of ING1b was required
and partially sufficient for induction of apoptosis. Adenoviral-mediated expression of A3H-NLS/NTS peptide in cells of
osteosarcoma, glioblastoma and breast cancer origins resulted in strong reduction of cell viability. Depending on the cell line,
treating cells with 45 MOI of virus expressing A3H-NLS/NTS peptide resulted in a 60 - 85% decrease in cancer cell survival
compared to cells treated with the control construct (Ad-GFP), and the highly transformed triple negative tumorigenic
MDA-MB-468 breast cancer line was sensitive to the peptide when infected with 5 MOI of A3H-NLS/NTS adenovirus. These
pro-apoptotic effects were found to be dose and time dependent. Furthermore, using p53 wild-type, p53 mutant and p53-null
cancer cell lines we demonstrated that the effects of A3H-NLS/NTS expression on cell survival and apoptosis were independent
of p53-status. The evaluation of the synergy between the A3H-NLS/NTS peptide and common chemotherapeutic agents is
currently ongoing. Our long-term goal is to develop ING1b-based therapeutics that can be used as an adjuvant therapy in
combination with the existing breast cancer treatments.

Title: CD4 Th1 cytokines and HER-2/HER-3 blockade induces tumor senescence in breast cancer
Cinthia Rosemblit1 and Brian J Czerniecki1. 1University of Pennsylvania, Philadelphia, PA.
Body: Background: HER-2 a molecular oncodriver in breast tumorigenesis is over expressed in 25% of human breast cancers,
and its expression correlates with enhanced tumor aggressiveness. While targeted therapies have improved outcomes many
patients become resistant or recur. We have recently established a progressive loss in anti-HER-2 CD4 Th1 responses during
disease progression and is associated with outcomes. The pleiotropic Th1 cytokines IFN- and TNF- have diverse effects on
tumor epithelial cells. In this study we sought to determine whether these Th1 cytokines induce senescence of HER-2 expressing
breast cancer cells and assess the impact of IFN- and TNF- with simultaneous HER-2 and HER-3 blockade.
Results: All breast cancer cell lines tested express IFN- and TNF- receptors measured by western blot analysis. The high and
intermediate HER-2 expressing cells are sensitive to tumor senescence induction when treated with combinations of IFN- and
TNF- in a dose dependent manner. Low HER-2 expressing cells were less sensitive to senescence induction as measured by
positive -galactosidase activity and the expression of p15INK4b and p16INK4a by western blot. Addition of IFN- and TNF-
treatment to HER-2-depleted cells by RNAi resulted in an increase senescent phenotype and was increased further when the
cells were double HER-2- and HER-3-depleted. To determine whether CD4 Th1 mediated effects on high HER-2 human breast
cancer cell lines, we co-cultured increasing number of HER-2 antigen-primed CD4+ T cells with high HER-2 human breast cancer
cells in a transwell cell culture system. This resulted in a dose-dependent senescence of breast cancer cells compared with CD4+
T cells primed either with immature dendritic cells (DC) or mature DC plus irrelevant (Class II BRAF) peptide. In addition,
SK-BR-3 breast cancer cells incubated with the supernatant of the CD4+ T cells-immature DC or mature DC co-culture
demonstrated similar results. CD4+ Th1-elaborated cytokines IFN- and TNF- in the supernatants were confirmed using ELISA.
Blocking antibodies against IFN- and TNF- demonstrated reduced senescence induction.
Conclusions: Our results establish a role for IFN- and TNF- in inducing tumor senescence in breast cancer. An effective CD4
Th1 responses combined with HER-2 and HER-3 blockade can significantly drive tumor senescence in breast cancer that can be
explored as treatment to effectively eliminate residual breast cancer cells and prevent recurrence.

Title: No Show

Title: Studies on tumor heterogeneity using a preclinical model of breast cancer caused by genetic alteration of the ATX-LPA axis
Lorenzo Federico1, Kang Jin Jeong1, Dong Zhang1, Zhenlin Ju1, Zechen Chong1, Jennifer B Dennison1 and Gordon B Mills1.
1
University of Texas MD Anderson Cancer Center, Houston, TX.
Body: Targeted alterations of the ATX/LPA signaling axis in mice creates a tumorigenic environment from which mammary
neoplasias can develop. Using mammary tumors derived from this model, we have established a transplantation procedure in
which fragments of different tumors are orthotopically transplanted in syngeneic mice. Under these conditions, primary murine
tumors displayed a wide spectrum of growth rates suggesting that alteration of ATX/LPA signaling axis engenders a variety of
different breast cancer subtypes with distinctive propensity to progress, which is similarly observed in human neoplastic disease.
Comprehensive analysis of high-throughput RNAseq and Reverse Phase Protein Array (RPPA) data revealed that despite being
passaged multiple times, tumors retained their distinct growth profiles and also maintained surprisingly stable molecular
characteristics. The remarkable consistency in terms of growth, transcriptional landscape, and protein expression demonstrates
that in this model a small fragment randomly taken from any region of the tumor can accurately and reproducibly regenerate an
entire developmental program when grown in vivo. It follows that each consistent molecular change detected in the tumor likely
represents either a marker of the tumor or its mechanistic oncodriver. Using unbiased multidimensional scaling analysis of
RNAseq data we have identified tumors that were strictly related to normal mammary tissue in terms of gene expression as well
as individual gene signatures that differentiate slow- vs fast-growing tumors. Furthermore, we have identified several stable
alterations at a protein level that are commonly found in human breast cancers including cMYC, BRCA2, PARP1, Claudin-7,
Her3, and E-Cadherin. These results demonstrate that an orthotopic transplantation procedure in immunocompetent hosts
represents a relevant preclinical platform to probe human breast cancer heterogeneity and pinpoint molecular alterations that are
mechanistically responsible for malignant behavior.

Title: Molecular characterization of a patient-derived xenograft (PDX) resource for triple negative breast cancer
Joel H Graber1, James G Keck1, Susan D Airhart1, Carol J Bult1 and Edison T Liu1. 1Jackson Laboratory, Bar Harbor, ME.
Body: The Jackson Laboratory (JAX) has developed a resource of human tumors implanted into immune deficient mice (patient
derived xenografts; PDX) as a platform for testing standard of care and novel therapeutic options for Triple Negative Breast
Cancer. PDX models provide an advantage over cell culture based models for testing therapeutic interventions because they
retain properties such as tumor cell heterogeneity that are critical to the biological properties of a patients tumor and response to
treatment.
Tumor material acquired from biopsy or surgical resection was implanted subcutaneously into the flank of immune deficient
NOD-scid IL2r gamma-chain null (NSG) mice. The PDX resource currently contains 21 established breast cancer PDX models
(12 TNBC) with 24 additional models currently in development. Two of the established TNBC PDX models have BRCA1
mutations. The median age of the patients from whom tumor material was obtained for all breast models is 53 (45-89).
Tumors that successfully engrafted were characterized for somatic mutations using the new JAX Clinical Cancer Panel, Copy
Number Variants using the Affymetrix human 6.0 SNP array, and gene expression using both Affymetrix U133 plus v2 and
RNA-Seq. Normalized gene expression was analyzed for characteristic patterns in a pan-cancer approach across all PDX models
and further compared with the previously identified TNBC molecular subtypes (Lehmann et al. 2011. JCI 121:2750-2767). The
combination of principal components analysis and classification via expression pattern resulted in putative matches of models to
most of the known molecularly defined subtypes of TNBC tumors.
Tumor bearing mice for the TNBC PDX models have been treated with docetaxel, cisplatin, cyclophosphamide and doxorubicin.
Preliminary studies of tumor response to these treatment regimes revealed systematic differences that can be correlated with
features of the genomic analysis, including expression subtype characterization.
The JAX collection of TNBC cancer PDX models is a well-annotated, publically available resource of models with deep genomic
characterization and standard of care therapy response data for use in the development of advanced therapeutic options.
Genomically defined subgroups within the collection suggest strategies to refine patient selection and treatment algorithms.
Information about the models along with summarized genomic data is publicly available at the Mouse Tumor Biology database
PDX web portal (http://tumor.informatics.jax.org).

Title: No Show

Title: The role of sphingosine kinase 1 in breast carcinogenesis
Yoshiko Shimizu1, Hideki Furuya1, Paulette M Tamashiro1, Kayoko Iino1, Charles J Rosser1 and Toshihiko Kawamori1. 1University
of Hawaii, Honolulu, HI.
Body: Introduction: The bioactive sphingolipids, ceramide/sphingosine and sphingosine 1-phosphate (S1P) possess opposite
effects on cell fate. Sphingosine kinase 1 (SphK1) is an enzyme that principally regulates the balance or the "rheostat" of the
sphingolipids by phosphorylating sphingosine to form S1P. In this study, we investigated the effect of SphK1 deficiency on
HER2/neu-induced mammary carcinogenesis using a MMTV-neu transgenic (Tg) mice model. Methods: MMTV-neu Tg mice
were crossbred with SphK1 knockout (KO) mice to generate MMTV-neu Tg mice with KO or wild-type for SphK1 gene. The
number and size of the palpable tumors were recorded weekly until mice reached 35 weeks of age or the largest tumor diameter
reached 20 mm in size. At necropsy, tumors and blood were collected for analysis. The sphingolipid profiles in blood were
analyzed using tandem mass spectrometry. Results: The incidences of mammary tumor development in the homozygous SphK1
KO (1/13; 8%) and heterozygous SphK1 KO (11/44; 26%) were significantly reduced (P=0.0112 and 0.0208, respectively) when
compared to wild-type mice (8/13; 62%). The mammary tumor multiplicity was significantly reduced in homozygous SphK1 KO
(0.080.08; P=0.0036) when compared to wild-type mice (0.690.17). The S1P levels in blood were significantly decreased in
homozygous SphK1 KO mice (P<0.0001), while sphingosine levels were significantly increased in the blood of heterozygous
SphK1 KO mice (P<0.001). In both homozygous and heterozygous SphK1 KO mice, the blood level of C16:0-Ceramide was
significantly increased (P<0.001). Conclusion: The results provide novel evidence that SphK1 mediates HER2/neu-induced
mammary carcinogenesis by regulating the ceramide/sphingosine-S1P "rheostat". Thus, we propose that SphK1 inhibition may
be a novel therapeutic option in the treatment of HER2 breast tumors.

Title: Anti-tumor efficacy of PI3K-mTOR pathway inhibitors in combination with PARP inhibitor plus carboplatin in
BRCA1-competent TNBC, beyond PTEN: A proof of concept study
Nandini Dey1, Yuliang Sun1, Jennifer Carlson1, Lori Friedman2, Pradip De1 and Brian Leyland-Jones1. 1Avera Research Institute,
Sioux Falls, SD and 2Genentech, San Francisco, CA.
Body: Background: Recently we reported that PI3K-mTOR inhibition potentiated anti-tumor effects in BRCA-competent TNBC
cells when combined with ABT888 (A) and carboplatin (C) (De et al., Neoplasia, 2014). Pro-proliferative and anti-apoptotic actions
of this pathway constitute one of the main effector signals of RAS. Here, we tested the anti-tumor effect of either single or dual
node blockade of PI3K-mTOR pathway when combined with A and C on BRCA-competent TNBC cells with WT-PTEN
background and activated RAS-RAF pathway. Materials & Methods: Athymic mice bearing TNBC xenograft tumors were treated
with pan PI3K inhibitor, GDC-0941 or PI3K-mTOR dual inhibitor, GDC-0980 alone or in combination with A and C.
Mechanism-based in vitro studies were performed using a panel of BRCA-wt/mutants TNBC cells with varying genetic
backgrounds. Results: Blocking a single nodal point of PI3K by GDC-0941 failed to significantly inhibit growth of pre-established
tumors (> 20%) even in combination with A and C in MDA-MB231 xenografts, while GDC-0980 potentiated an anti-tumor effect by
inhibiting tumor growth by 90%. In vitro treatment (1) decreased proliferation signals (pAKT, pP70S6K, p4EBP1, pS6RP), cell
cycle progression, vascular mimicry & 2D clonogenic growth, and (2) increased apoptosis markers (cl-caspase3, 9, BIM,
cl-PARP, & annexinV positivity). GDC-0980 in combination with A plus C concomitantly decreased Ki67, cl-caspse3, pVEGFR,
CD31, p4EBP1, and pS6RP in vivo (IHC). In contrast, combination of GDC-0941 with A plus C failed to affect the cell cycle,
apoptosis or 2D clonogenic growth in MDA-MB231 cells. Alternatively, Treatment with RAD001 increased pAKT while it perturbed
the activation of S6RP/4EBP1. Conclusion: Our data indicate that following dual inhibition of PI3K-mTOR, S6RP/4EBP1
de-phosphorylation tracks more consistently with the drugs tumor-growth inhibitory response rather than the upstream state of
AKT-activation. Unlike mTORC1 inhibitor RAD001, GDC-0980 potently eliminates (in vitro & in vivo) feed-back re-activation of the
pathway as, (1) it targets PI3K, reactivation of AKTT308 is blocked and (2) inhibition of the mTORC2 complex blocks the
reactivation of AKTS473. TNBC tumors with PTEN independent RAS/RAF mutation-mediated activation of PI3K-mTOR pathway
can be controlled by dual node blockade of the pathway when combined with a PARP inhibitor and carboplatin.

Title: Amplified DERE-mediated epigenetic repression in ER-mediated breast tumorigenesis
Pei-Yin Hsu1, Hang-Kai Hsu1, Yi-dong Chen1, Victor X Jin1, Zelton D Sharp1 and Tim H-M Huang1. 1UTHSCSA, San Antonio, TX.
Body: Epidemiological studies reveal that maternal exposure to estrogenic chemicals during pregnancy results in higher risk of
breast cancer in offspring. One of possible causal mechanisms is epigenetic reprogramming, but how estrogen/ER signaling
contributes to tumorigenesis through epigenetic machinery remains unclear. In this study, we demonstrate that the epigenetic
modulation mediated by estrogen-driven amplification of distant-acting regulatory elements, or enhancers, may be an explanation
addressing to this question. Enhancers contain transcription factor binding sites known to remotely regulate transcription through
chromatin looping or transvection. Using our in vitro model to mimic maternal estrogen exposure and a "Seq-to-Seek" strategy
integrating three next-generation sequencing approaches, we comprehensively mapped distant estrogen response elements
(DEREs) that remotely control transcription of target genes through chromatin proximity. Surprisingly, a densely mapped DERE
region located on chromosome 20q13 frequently amplifies in ER-positive breast cancer with poor prognosis. Progressive
accumulation of DERE copies was observed in normal breast progenitor and cancer cells chronically exposed to estrogen. Upon
estrogen stimulation, these aberrantly amplified DEREs are clustered as a potential transcriptional depot for suppressing target
gene expression through altering chromatin interactions, leading to accumulation of repressive histone marks (e.g. trimethylated
H3K27 and H3K9), polycomb repressive complex 2, the presence of heterochromatin protein 1 (HP1) and DNA methylation,
following prevention of RNA polymerase II and active histone mark- H3K4me3 binding. Furthermore, neutralization of HP1
function can significantly attenuate estrogen-driven DERE clustering and DERE-mediated repressive chromatin interactions,
resulting in inhibition of cell proliferation. Deletion of the interested DERE regions using CRISPR/Cas9 genomic editing system
further demonstrated that DERE-driven remote transcriptional control play a crucial role in tumor cell growth. Our data support a
model in which amplified DEREs preferentially induce repressive epigenetic modulation of target genes. These findings indicate
that 20q13 DERE region is a potential transcriptionally repressive domain whose aberrant amplification can result in suppressing
expression of tumor suppressor genes, leading to tumorigenesis. In summary, our findings suggest that amplification of DNA
regulatory elements can profoundly alter target transcriptome during tumorigenesis and amplified DEREs can be used as
potential prognostic markers for endocrine resistance and predicative for progeny at risk of breast cancer.

Title: An exhaustive algorithm for detecting copy number aberrations and large structural variants in whole-genome mate-pair
sequencing data
Yan W Asmann1, Chen Wang2, Brian M Necela1, Xianfeng Chen2, Jean-Pierre A Kocher2, Matthew J Maurer2, Thomas M
Habermann2, Susan L Slager2, Andrew L Feldman2, Anne J Novak2, James R Cerhan2, Edith A Perez1 and E Aubrey Thompson1.
1
Mayo Clinic, Jacksonville, FL and 2Mayo Clinic, Rochester, MN.
Body: Objectives and Rationale: Structural variants (SV) including large copy number aberrations (CNV), translocations,
inversions, and large insertions and deletions (INDEL) play a critical role in tumorigenesis and progression. In fact, we now know
that tumors can be categorized according to the size of mutations harbored. In several cancers, including ovarian and breast
cancer, the large structural mutations, rather than single site mutations, play a dominate role in tumor etiology. Therefore, it's
critical to implement reliable algorithm for the detection of structural variants in DNA sequencing data. Mate-pair sequencing is a
protocol specifically implemented for detection of the whole-genome level structural variants. It requires less sequencing depth
therefor is cost effective, and enables the detection of CNVs, translocations, and inversions simultaneously. However, so far there
has been no reliable bioinformatics pipeline for the analyses of the mate-pair sequencing data.
Methods: Our novel algorithm, the SnowShoes-SV, is an exhaustive search algorithm designed specifically for mate-pair DNA
sequencing data analyses. It calls the SVs based on disconcordant read pairs. The false SVs are filtered according to the
following criteria: (i) the number of the supporting read pairs; (ii) the lack of reads from control data that implicate SV at the same
region; (iii) the mappability and uniqueness of the region based on data from the ENCODE project; (iv) consistencies of the
mapping orientations of the supporting read pairs; (v) the similar sizes between sequencing library and the two end read clusters.
Results: Using a set of samples previously genotyped by aCGH, the SnowShoes-SV successfully detected all known CNVs and
other SVs. It also identified copy number neutral translocations and inversions previously not identified by aCGH. In addition, the
algorithm nominated novel SVs which are to be validated by PCR.
Conclusions: SnowShoes-SV is a highly sensitive and specific algorithm for SV detection from the mate-pair DNA sequencing
data.

Title: Targeting ErbB2 with human PEPD
Lu Yang1, Yun Li1, Arup Bhattacharya1 and Yuesheng Zhang1. 1Roswell Park Cancer Institute, Buffalo, NY.
Body: ErbB2, also known as Her2 or Neu, belongs to the ErbB family of plasma membrane-bound receptor tyrosine kinases,
which also include ErbB1, ErbB3 and ErbB4. ErbB2 is best known for its involvement in human breast cancer. ErbB2 gene
amplification occurs in 20% of breast cancer, and ErbB2 amplification or overexpression is a strong predictor of poor disease
prognosis. ErbB2-targeted therapies, particularly humanized monoclonal antibody trastuzumab (Ttzm) in combination with
chemotherapy, have shown considerable clinical efficacy. However, primary and secondary resistance remains a clinical
challenge, and Ttzm, produced in mammalian cells, is very expensive.
We have found that human prolidase, also known as peptidase D (PEPD) among several other names, binds to ErbB2 with high
affinity (Kd = 7 nM) and binds as a homodimer (493 amino acids per subunit) to subdomain 3 in the extracellular domain of
ErbB2. Each monomer of PEPD binds to one copy of ErbB2. However, PEPD is a weak ErbB1 binder (Kd = 5 M) and does not
bind to ErbB3 or ErbB4. PEPD is the first-ever natural ligand of ErbB2, and unlike the other ligands of ErbB receptors, it is devoid
of an EGF motif. PEPD has been long known to hydrolyze dipeptides with proline or hydroxylproline at the carboxy terminus, but
the dipeptidase activity of PEPD is not involved in ErbB2/ErbB1 modulation. In cells overexpressing ErbB2, where both activated
dimers and inactive monomers of ErbB2 exist, as ErbB2 overexpression causes spontaneous dimerization, auto-tyrosine
phosphorylation and recruitment and activation of downstream signals, PEPD rapidly binds to ErbB2 homodimers (<10 min) and
silences the existing ErbB2-SRC signaling, a key oncogenic pathway of ErbB2, by disrupting SRC association with ErbB2. In
contrast, PEPD binds to ErbB2 monomers relatively slowly (>30 min), but this binding causes ErbB2 dimerization, ErbB2
phosphorylation and downstream signaling. PEPD binding to ErbB2 subsequently causes pronounced ErbB2 depletion, resulting
from its internalization and degradation. PEPD also strongly inhibits the DNA synthesis, anchorage-independent growth and
invasion of cells that overexpress ErbB2, but has no effect on cells without overexpression of ErbB2. In fact, cells become
sensitized to inhibition by PEPD upon achieving stable ErbB2 overexpression. Thus, the overall impact of PEPD on ErbB2 is
inhibitory, and PEPD targets cells addicted to ErbB2. In ErbB2-overexpressing cells, at equimolar concentrations, PEPD was
more effective than Ttzm in driving ErbB2 depletion, but is weaker than Ttzm in stimulating ErbB2 phosphorylation. In mouse
tumor models, PEPD administered by intraperitoneal injection (Monday, Wednesday, Friday) at 0.2-2 mg/kg body weight strongly
inhibited the growth of ErbB2-overexpressing tumors, but had no impact on tumors without ErbB2 overexpression, and the
PEPD-treated mice showed no adverse effects.
Given that the findings described above were made using human PEPD generated in bacteria, there is a distinct possibility that
recombinant human PEPD may be a low cost alternative to Ttzm. Further investigation of the antitumor activity of PEPD and its
modulation of ErbB2 signaling is warranted.

Title: Single-cell TOF-SIMS reveals that human breast cancer stem cells have significantly lower content of palmitoleic acid
compared to their counterpart non-stem cancer cells
Yoshimi Ide1,2, Michihiko Waki2, Itsuko Ishizaki3, Yasuyuki Nagata2,4, Yumiko Taki1, Yuko Hosokawa1, Ryoichi Matsunuma1,
Hiroyuki Ogura1, Norihiko Shiiya1, Noriaki Sanada3 and Mitsutoshi Setou2. 1Hamamatsu University School of Medicine,
Hamamatsu, Shizuoka, Japan; 2Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan and 3ULVAC-PHI,
Chigasaki, Kanagawa, Japan.
Body: Background
Distinguishing individual cancers according to their biochemical heterogeneity have provided much useful information to clinical
site. Recently, the cancer stem cell (CSC) theory has been accepted as a concept that explains the mechanism of cancer
recurrence and resistance to treatment. To characterize such particular cell populations in heterogeneous tissues, we conducted
combination of fluorescence activated cell sorting (FACS) and time-of-flight secondary-ion mass spectrometry (TOF-SIMS) and
applied the method to analyses of breast CSCs. TOF-SIMS, which enables to visualize the composition of molecules with mass
over 100 Da that were obtained in specimen, has been employed to analyze surface of industrial materials and biomaterials. This
method is thus suitable for performing single cell analysis of membranous lipids.
Methods
Breast cancer specimens surgically resected from two patients were enzymatically dispersed into cells. They were labeled with
fluorescence-conjugated antibodies of CD45, CD44, and CD24. The cells of CD45-CD44+ CD24- were sorted as CSCs with
FACS as well as CD45-CD44- CD24+ cells as non-stem cancer cells (NSCCs). TOF-SIMS analysis and fatty acid analysis was
performed according to our previous study published in Surface and Interface Analysis (1). The surface of the sorted cells was
analyzed by a PHI TRIFT V (ULVAC-PHI Inc., Kanagawa, Japan) TOF-SIMS instrument. Primary ion beam is irradiated to the
surface of the samples and, secondary ions derived from samples are calculated by time-of-flight with the information of the place
where the molecular ions were ejected. Negative secondary ions were obtained with a mass range of m/z 01850. Mass spectra
were analyzed by WinCadenceN software (ULVAC-PHI Inc.) to obtain ion counts and ion images. Integrated ion intensities of FA
were normalized using phosphoric acid intensity. The Welchs t-test was used to compare the normalized ion counts with P-value
< 0.05 taken as statistically significant.
Results
FACS analyses successfully collected CD45-/CD44+/CD24- CSCs and CD45-/CD44-/CD24+ non-stem cancer cells (NSCCs) in
both two cases, which were corresponding to o.33% and 0.74% of all cells in case 1, and 0.14% and 1.14% in case 2. TOF-SIMS
analyses visualized phosphoric acids and four fatty acid (FA) species in the sorted CSCs and NSCCs. These ions probably came
from membranous phosphopolipids and they were uniformly detected from the locus where the cell attached. Integrated ion
intensity of palmitoleic acids [FA(16:1)] of CSCs normalized by phosphoric acids signals were significantly decreased than that of
CD45-/CD44-/CD24+ NSCCs as a counterpart. Therefore, our novel method successfully provided lipid composition analysis of
individual cells classified with complicated combination of marker expressions in clinical specimens composed of heterogeneous
cellular populations, and characterized lipid composition of CSCs.
Reference
1. Nagata Y, Ishizaki I, Waki M, Ide Y, Hossen A, Ohnishi K, et al. Glutaraldehyde Fixation Method for Single-Cell Lipid Analysis
by Time-of-Flight Secondary Ion-Mass Spectrometry. Surface and interface analysis : SIA. 2014:DOI: 10.1002/sia/5522.

Title: Post-transcriptional coordination of gene expression during breast cancer tumorigenesis
Laura Simone Bisogno1 and Jack D Keene1. 1Duke University, Durham, NC.
Body: We investigate mechanisms of RNA regulation central to human breast cancer progression. Aberrant gene expression is
known to be an important factor in cancer, yet little is known about the role RNA-binding proteins (RBPs) play in disease onset
and progression. Sequence specific RBPs coordinately regulate subsets of functionally related mRNAs within ribonucleoprotein
complexes (RNPs), which are remodeled in response to cellular perturbations, allowing for the rapid and coordinated translation
of proteins that have common functions. These post-transcriptional events robustly influence expression patterns of
proto-oncogenes, growth factors, cytokines, and cell cycle regulators by influencing both mRNA stability and translation.
Therefore, understanding the post-transcriptional layer of gene regulation is critical to understanding cancer development and
progression. We have identified significant transcriptomic changes during the stepwise transition from primary mammary epithelial
cells to a fully malignant state, as well as coordinated RNA dynamics of transient cellular RNP complexes. For this analysis, we
generated a model of human breast cancer formation in which normal mammary epithelial cells were first immortalized through
the expression of hTERT, p53DD, Cyclin D1, CDK4 R24C and C-MYCT58A, and then subsequently transformed by the addition of
H-RASG12V. RNA-sequencing analysis demonstrated that the genes most significantly changed in this model are those involved in
cell adhesion. In fact, while primary cells have a gene expression pattern typical of normal epithelial cells, both immortalized and
transformed cells exhibit an mRNA expression pattern typical of mesenchymal cells. Interestingly, we found that N-cadherin
protein, as well as other prototypical mesenchymal proteins, are only robustly expressed in the fully malignant cells, but not in
immortalized cells. This suggests coordinated translational regulation of mRNAs that are essential for activating the
epithelial-to-mesenchymal transition (EMT). Many of these EMT-related mRNAs are targets of both the HuR protein and
microRNAs. HuR is a translational activator that competes with microRNAs and is mislocalized in many cancers. To integrate the
quantitative RNA dynamics of HuR and microRNAs on a global scale, we used RNP-Immunoprecipitation followed by
high-throughput sequencing to identify and quantify the remodeling of mRNA subsets associated with HuR and the Ago2/RISC
complex in our breast cancer progression model. This integrative work provides global information about the underlying biology of
carcinogenesis at the level of post-transcriptional coordination of gene expression, resulting in a more comprehensive
understanding of the many layers of complex gene regulation. Therefore, the results from this study may ultimately direct our
ability to counter the RNA regulatory changes that underlie malignancy.

Title: Low dose of CDB-2914 and CDB-4124 efficiently inhibit the growth of T47D spheroid induced by pre-menopausal and
post-menopausal concentrations of estrogen and progesterone
Oukseub Lee1, Mi Ran Choi1, Susan E Clare1 and Seema A Khan1. 1Northwestern University Feinberg School of Medicine,
Chicago, IL.
Body: Background: The anti-progestins (RU-486, CDB-2914 and CDB-4124) may have potential to prevent estrogen receptor
(ER) and progesterone receptor (PR) positive breast cancer. Physiological estradiol (E2), and progesterone (P4) levels are
different in pre- and post-menopausal women. The purpose of this study was to determine: 1) Whether the physiological female
hormones at pre- and postmenopausal concentrations affect the growth of an in vitro model of breast cancer, i.e., T47D
spheroids, and 2) Whether anti-progestins at pharmacological concentrations work as growth inhibitors in high and low hormonal
environments.
Methods: T47D cells were grown as spheroids in the presence of serum concentrations of E2 and P4 consistent with pre- and
postmenopause. Premenopausal luteal phase concentrations were 262 pM (71.3 pg/mL) of E2 + 18 nM (5.66 ng/mL) of P4;
postmenopausal concentrations were 122 pM (33.2 pg/mL) of E2 + 3 nM (1.08 ng/mL) of P4. T47D cells (5000 per well) were
seeded in 1.5% agarose coated 96 well plates, and grown in phenol red-free mammary epithelial cell growth basal medium
(Lonza) supplemented with 10% double charcoal -stripped FBS. Hormones and anti-progestin treatments started 24 hrs after cell
seeding. Three concentrations (50, 250, 1000 nM) of the anti-progestins were tested. Images of each spheroid were taken daily
for 14 days, and the sizes of spheroids were analyzed by area (Pixel) using ImageJ software.
Results: Results are presented in Table 1.
Table 1. Relative spheroid growth compared to the hormone treatment alone
Hormone treatments
Drug treatments (nM)
RU-486
CDB-2914
CDB-4124
50
0.8
0.56
0.7
250
0.72
0.63
0.68
1000
0.84
0.64
0.6
50
0.95
0.76
250
0.9
0.83
0.74
1000
0.96
0.85
0.71
Premenopausal condition
Postmenopausal condition
Premenopausal and postmenopausal hormone concentrations stimulated spheroid growth by twofold and 1.7 fold higher,
respectively, when compared to the vehicle control (0.1% DMSO) at14 days.
In the premenopausal condition, RU-486 showed moderate inhibition of spheroid growth at all three concentrations; CDB-2914
was more efficient at inhibition than CDB-4124 at 50 nM. At 250 nM and 1000 nM, CDB-2914 and CDB-4124 showed similar
efficacy.
In the postmenopausal condition, RU-486 showed very minor inhibition on spheroid growth at all three concentrations; CDB-2914
showed significantly higher inhibition than CDB-4124 at 50 nM. At 250 nM and 1000 nM, CDB-4124 was more efficient than
CDB-2914.
Conclusions: Our results indicate that T47D spheroids will grow in postmenopausal hormone concentrations, but growth is
enhanced in premenopausal hormone conditions. RU-486 did not produce effective inhibition at postmenopausal hormone levels;
however pharmacological concentrations (50, 250 and 1000 nM) of both CDB-2914 and CDB-4124 efficiently decrease the
spheroid growth induced by both premenopausal and postmenopausal hormone levels. These data suggest that low doses of
CDB-2914 and CDB-4124 should be further investigated for ER/PR positive breast cancer prevention and therapy.

Title: Blocking a key region in the HER2 subdomain III inhibits the HER2-network in patients with resistance to HER2-targeted
therapy
Barbara Schroeder1, Menendez Javier2, Espinoza Ingrid3 and Lupu Ruth1. 1Mayo Clinic Cancer Center, Rochester, MN; 2Catalan
Institute of Oncology and Girona BioMedical Research Institute, Girona, Catalonia, Spain and 3University of Mississippi Medical
Center, Jackson, MS.
Body: The HER2 receptor is over-expressed in about 25% of breast carcinomas and correlates with more aggressive breast
cancer disease, poor prognosis as well as resistance to chemo- and endocrine therapies. The strong correlation between HER2
levels/activity and the malignancy of the disease made it a preferred target for anti-cancer therapy. For example, Trastuzumab,
which targets HER2 homo-dimerization and Pertuzumab, which interferes with HER2 hetero-dimerization, are in clinical use.
Unfortunately, Trastuzumab monotherapy is only partially effective and the majority of breast cancer patients who initially respond
to Trastuzumab rapidly develop resistance to the drug. Thus, novel strategies/agents are in need, especially drugs that prevent
hetero-dimerization with other HER family members.
Structural analyses revealed that sub-domain III in the extracellular domain (ECD) of the HER family members is responsible for
dimerization. Therefore, we hypothesized that disabling the dimerization loop in the HER2-ECD sub-domain III would ultimately
convert HER2 to a non-functional receptor. Therefore, we introduced a series of deletions in the HER2-ECD sub-domain III and
determined their oncogenic properties compared to HER2wt expressing cells. Importantly, a small deletion (16 amino acids) of
the HER2 extracellular domain (named HER26) abolished its homo- and hetero-dimerization and profoundly affected
HER2-catalyzed activation of the HER network, both in the context of HER2 over-expression and ligand-induced trans-activation
of HER2. Expression of the HER26 variant failed to promote anchorage-independent growth and interfered with the
activation/Tyr phosphorylation of HER1, HER2 and HER3. Moreover, this mutant failed to promote resistance to Paclitaxel
treatment in HER2-overexpressing breast cancer cells.
To determine the molecular mechanisms underlying this behavior, we assessed the mutant and wt expressing cells lines
morphologically and biochemically and demonstrated that the HER26 is absent from the plasma membrane (PM), most likely
due to a intracellular trafficking defect that mistargets the receptor to or traps it in an endomembrane compartment. Compared to
HER2wt-expressing cells, the HER26 mutant proves to be even more effective in inhibiting the oncogenic properties of the
receptor than the current drugs of choice such as Trastuzumab, Pertuzumab and Cetuximab alone or in combination as
measured by anchorage-independent growth.
These findings reveal that the HER2-ECD bears an essential "activating" region that is indispensable for HER2 homo-and
heterodimerization. Elimination of this "activating" element in HER2 seems to recapitulate and greatly improve the combined
actions of Trastuzumab and Pertuzumab. These findings offer a strong rationale for developing this peptide sequence into a
valuable anti-HER2 therapeutic drug.

Title: Contemporary risk of local, regional and contralateral breast cancer recurrence
Kim C Aalders1, Annelotte CM van Bommel2, Thijs van Dalen1, Gabe S Sonke3, Paul J van Diest4, Liesbeth J Boersma5 and
Margriet van der Heiden-van der Loo6. 1Diakonessenhuis, Utrecht, Netherlands; 2Leiden University Medical Center, Leiden,
Zuid-Holland, Netherlands; 3Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands; 4University Medical Center,
Utrecht, Netherlands; 5University Hospital Maastricht, (GROW Maastro Clinic), Maastricht, Limburg, Netherlands and
6
Comprehensive Cancer Centre the Netherlands (IKNL), Utrecht, Netherlands.
Body: Background
Long-term follow-up of breast cancer patients aims to detect curable recurrence, and focuses on ipsilateral in-breast recurrence
(LR), regional lymph node recurrence (RR) and contralateral breast cancer (CBC). In recent years there is mounting evidence of
a decrease in locoregional recurrence rates. Non-surgical-treatment modalities have evolved extensively, while surgery has
become less invasive over the last fifteen years. The present study aimed to address contemporary loco-regional recurrence
rates evaluating time trends and the role of contributing factors.
The Netherlands Cancer Registry was searched for all female patients diagnosed and operated for a unilateral primary breast
cancer (pT1-2,anyN,M0) between 1-1-2003 and 31-12-2006. Exclusion criteria were previous cancer, neo-adjuvant chemotherapy
or incurable disease. Data on 5-year follow-up were available from hospital records and included the first site of recurrence and
contralateral breast cancer (CBC). The 5-year risk of developing LR, RR and CBC were estimated using Kaplan Meier curves.
Patients were censored at time of death, lost to follow-up or the development of distant metastases. Prognostic influence of
various patient- and disease characteristics was assessed.
Results
A total of 35.006 eligible patients were identified. The 5-year rates of LR, RR, and CBC are presented in Table 1. The risk of CBC
was higher than LR and RR. Over time, the rates decreased significantly for all three endpoints.
Table 1. Overall 5-year risk of local, regional and contralateral recurrence and distant metastases over time (period 2003-2006)
Local recurrence(a)
Regional recurrence
Contralateral breast cancer
no. of events
rate (%)
no. of events
rate (%)
no. of events
rate (%)
2003 (n=8933)
185
2,40%
86
1,11%
227
3,08%
2004 (n=9048)
181
2,35%
83
1,07%
189
2,51%
2005 (n=9055)
144
1,84%
75
0,95%
190
2,49%
2006 (n=7970)
131
1,87%
50
0,70%
146
2,05%
Overall (n=35.006)
641
2,12%
294
0,96%
752
2,55%
(a)Local recurrence (ipsilateral in-breast recurrence + new primary). Rates represent Kaplan Meier estimates
The LR-rate was lower with breast conserving surgery (BCS) vs. amputation (1.8% vs. 2.5%), T1a-b vs. T1c-T2 tumors (2.0% vs.
2.5%), ER+ vs. ER- tumors (1.8% vs. 3.5%) and inversely related with age (highest in pts. <35 yrs: 2.9%). LR rate seemed
independent of HER2 status.
The 5-year RR-rate was 0.9% for N0 patients, and decreased from 1.0% to 0.7% over time. The risk of RR after amputation
decreased from 1.8% to 0.9% over time, but was higher than after BCS (1.6% vs. 0.6%). Overall, the RR-rate was highest in the
N>1 group (1.4%) and the triple negative group (2.0%).
The CBC-rate was lower for patients who received chemotherapy (CT) than for patients who did not (1.6% vs. 3.1%). The
CBC-rate only decreased over the years in the CT-group (3.7% to 2.5%).
Conclusions
Loco-regional recurrence rates have decreased substantially in recent years and have become very low. For the vast majority of
patients the risk of LR is substantially lower than the risk of CBC and the risk of RR is rarely larger than 1.0%. These low rates
might reflect improvements in systemic treatment.

Title: The body mass index interacts with the prognostic effect of the progesterone receptor in patients with a luminal HER2
Kathleen Van Asten1, Anneleen Lintermans1, Annouschka Laenen2, Chantal Remmerie3, An Poppe4, Giuseppe Floris5, Hans
Wildiers6, Marie-Rose Christiaens2 and Patrick Neven4. 1KU Leuven, Oncology, Leuven, Belgium; 2KU Leuven, InterUniversity
Center for Biostatistics and Statistical Bioinformatics, Leuven, Belgium; 3University Hospitals Leuven, Leuven, Belgium;
4
University Hospitals Leuven, Leuven, Belgium; 5University Hospitals Leuven, Leuven, Belgium and 6University Hospitals Leuven,
Leuven, Belgium.
Body: Background
Obesity (body mass index (BMI) >30) increases breast cancer (BC) risk and promotes BC metastases probably through high
estrogen levels. A high BMI as a risk and prognostic factor is consistently reported in postmenopausal women. Obese women
develop particularly hormone receptor positive, HER2 negative BC. In this study, we investigate the influence of BMI on the
prognostic effect of the progesterone receptor (PR) in postmenopausal patients with estrogen receptor (ER) positive, HER2
negative BC.
Women over 50 years of age diagnosed with primary operable BC between 2000 and 2012 at University Hospitals Leuven were
retrieved from the database. Patients were subdivided into normal weight (<25 kg/m), overweight (>25 and 30) and obese
(>30). We investigated for each BMI category the distant metastasis free interval (DMFI) and BC specific survival (BCSS) by PR
status. Apart from the total cohort, subgroup analysis was performed for luminal A (grade 1-2) and luminal B HER2 negative BC
(grade 3). We used Fine and Grays competing risk regression for the analyses. Covariates were age at diagnosis, tumor size,
lymph node status, and therapy.
Results
In total, 3227 patients were analyzed for DMFI. For BCSS, 3192 patients were analyzed as patients with unknown cause of death
were excluded from the analysis. 2395 of all patients had luminal A and 832 had luminal B HER2 negative BC (table 1). Median
time of follow-up was 6.5 years.
Table 1: Percentage of luminal A and luminal B patients per BMI category
BMI<25
BMI >25 and 30
BMI >30
Luminal A
1105 (76%)
820 (75%)
470 (70%)
Luminal B
355 (24%)
278 (25%)
199 (30%)
BMI: body mass index

BMI interacts with the prognostic effect of PR for DMFI and BCSS in luminal HER2 negative BCs. Only patients with BMI <25 had
a reduction in the risk of distant metastasis and BC-related death if the tumor was PR positive as compared to patients with PR
negative BC. This was observed in the total cohort and only seen in the luminal B subgroup (table 2). A similar effect was
observed in obese patients but this did not reach statistical significance, and was mainly present during the first 5 years following
diagnosis. For the overweight patients, no difference in DMFI and BCSS was observed by PR status.
Table 2: Multivariate analysis for DMFI and BCSS for PR positive versus PR negative in the 3 BMI categories in the total and the
luminal B cohort.
DMFI (HR, 95% CI)
BCSS (HR, 95% CI)
0.25, 0.16 to 0.40
0.22, 0.12 to 0.40
Total cohort
BMI <25
BMI >25 and 30
1.24, 0.59 to 2.60
0.95, 0.40 to 2.25
BMI >30
0.70, 0.33 to 1.49
0.53, 0.21 to 1.33
BMI <25
0.19, 0.11 to 0.33
0.14, 0.06 to 0.31
BMI >25 and 30
1.23, 0.44 to 3.48
1.02, 0.36 to 2.85
BMI >30
0.82, 0.31 to 2.20
0.61, 0.18 to 2.00
luminal B
BCSS: breast cancer specific survival, BMI: body mass index, CI: confidence interval, DMFI: distant metastasis free interval, HR:
hazard ratio
Conclusion
Normal weight patients have a reduced risk of developing distant metastases and of BC-related death if the tumor is PR positive
compared to PR negative BC. No difference between PR positive and PR negative cases was observed in overweight BC
patients. This BMI-dependent prognostic effect of PR was limited to luminal B BC patients.

Title: Transcriptional profiling of breast cancer metastases identifies liver metastasis-selective genes associated with adverse
outcome in luminal A primary breast cancer
Siker Kimbung1, Ida Johansson1, Anna Danielsson3, Srinivas Veerla1, Suzanne Egyhazi4, Jonas Bergh4, Zakaria Einbeigi3, Barbro
Linderholm3,4, Elisabet Lidbrink4, Niklas Loman1,2, Per Malmstrm1,2, Martin Sderberg2, Thomas M Walz5, Mrten Fern1, Thomas
Hatschek4, Ingrid Hedenfalk1 and the TEX Study Group4. 1Lund University, Lund, Sweden; 2Skne University Hospital, Lund,
Sweden; 3Sahlgrenska University Hospital, Gothenburg, Sweden; 4Karolinska Institutet and Karolinska University Hospital, Solna,
Sweden and 5Linkping University Hospital, Linkping, Sweden.
Body: Background: The site of relapse is associated with the prognosis of metastatic breast cancer, but our understanding of the
molecular determinants of organ-specificity of metastasis is incomplete. This study aimed to provide additional insight into the
biology of breast cancer liver metastases and to identify liver metastasis-selective genes associated with outcome in primary
breast cancer.
Methods: A cohort of 304 women with locally advanced and metastatic breast cancer was studied. Estrogen receptor (ER),
progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) and Ki67 expression were quantified in primary
tumors (N=217) by immunohistochemistry and in situ hybridization on tissue microarrays, and molecular subtypes were assigned
according to the 2013 St Gallen guidelines. In addition, fine-needle aspirates of metastases (N=91) were subjected to whole
genome transcriptional profiling.
Results: Liver relapse was significantly associated with ER positivity (P<0.002) and the luminal B-like subtype (P<0.01) and was
prognostic of an inferior post-relapse survival (P<0.001). Transcriptional profiling revealed that the major variation in the
transcriptional landscape of breast cancer metastases was associated with the expression of hormone receptors and the tumor
molecular subtype. However, liver metastases displayed unique transcriptional fingerprints, characterized by down-regulation of
extracellular matrix (i.e. stromal) genes involved in adhesion and skeletal development. Importantly, we identified a subset of 17
liver metastasis-selective genes that displayed significantly decreased expression in primary tumors of high histological grade
(grade 3) and of the luminal B and basal subtypes (P<0.001). This 17-gene signature was significantly and independently
prognostic of shorter relapse-free (P =0.001) and overall (P=0.03) survival among patients with ER positive primary tumors.
Remarkably, the 17-gene signature remained an independent predictor of shorter relapse-free survival (P=0.004) among patients
with luminal A primary tumors.
Conclusion: These results highlight a possible role of stroma-related genes in breast cancer liver metastasis biology and validate
the prognostic relevance of extracellular matrix/stromal genes in hormone receptor positive primary breast cancer, specifically of
the luminal A subtype.

Title: Creation of a robust algorithm utilizing minimal gene sets normalized against a reference gene set to identify triple-negative
breast cancer (TNBC) subtypes
Rob S Seitz1, David R Hout1, Stephan W Morris1, Rebecca B Smith1, Brian D Lehmann2, Xi Chen2, Jennifer A Pietenpol2 and
Brian Z Ring1. 1Insight Genetics, Inc, Nashville, TN and 2Vanderbilt-Ingram Cancer Center, Vanderbilt University School of
Medicine, Nashville, TN.
Body: Introduction: Treatment of TNBC has been challenging due to the absence of well-characterized molecular targets and the
heterogeneity of the disease. Using TNBC gene expression (GE) microarray profiles, the Pietenpol group molecularly binned the
malignancy into six distinct subtypes: two basal-like [BL1, BL2], two mesenchymal-like [M, MSL], an immunomodulatory [IM], and
a luminal subtype expressing androgen receptor [LAR]). Importantly, initial evidence suggests specific TNBC subtypes exhibit
different sensitivities to various targeted and conventional chemotherapies (1). For example, BL1 and BL2 showed sensitivity and
resistance, respectively, to taxanes (2).
Background: The original TNBCtype algorithm was generated from a meta-analysis of existing GE data from tumors and
clustering the data into the six subtypes listed above and a seventh "unclassified" subtype (1). To transition the test into the clinic,
we have modified the method of classification by reducing the number of signature genes. We then downloaded a publicly
available GE dataset to validate the optimized assay on an independent cohort.
Methods: Gene set enrichment followed by shrunken centroid analysis were used for feature reduction, resulting in 258 genes
used for model building. Linear regression, targeted minimum loss based estimation, random forest, and elastic-net regularized
linear models were employed, with the latter giving the best fit with the least number of required genes. Models were created to
identify each class individually or together using a centroid model. Coefficient and cutoffs were established on a normalized
TNBC training data set consisting of 14 cohorts (N=386) and then applied to a seven cohort validation data set (N=201). Finally,
both the original 2188 gene and the optimized 101 gene classifier models were applied to an independent cohort (278 TNBC
patients treated neoadjuvantly with mitotic inhibitors: GSE41998) to identify the TNBC subtypes.
Results: In the validation cohorts used by Lehmann et al.(1), there was strong agreement between the 2188 and 101 gene
models (Fisher exact test, P<0.0001). Specificity for the individual class models ranged from 90% (M) to 100% (LAR), while the
101 gene centroid resulted in a 4% rate of cases discordantly classified compared to the original 2188 gene classifier. As was
seen in the initial clustering in the training data set, there was notable overlap between the subtyping of BL1 and M. On the
discovery cohorts normalized with the reference gene set, the specificity for the individual class models ranged from 79% (M) to
97% (BL1). For the 101 gene centroid model the misclassification error was 5%.When the additional independent TNBCcohort
was examined, the agreement between the 2188 and 101 gene models was 89%. Furthermore, patients defined as BL1 or BL2
by the minimal gene set matched previous observations of sensitivity and resistance, respectively, to mitotic inhibitors.
Conclusions: These initial results show that the minimal gene set shows agreement with the larger original TNBC algorithm and
recapitulates the initial clinical observations. Future work will determine the clinical utility of this assay for patient management.

Title: Prognostic value of the progesterone receptor by proliferation rate in patients with luminal HER2 negative breast cancer
Kathleen Van Asten1, Ben Van Calster2, Anneleen Lintermans1, Olivier Brouckaert3, Giuseppe Floris4, Hans Wildiers5 and Patrick
Neven3. 1KU Leuven, Oncology, Leuven, Belgium; 2KU Leuven, Development and Regeneration, Leuven, Belgium; 3University
Hospitals Leuven, Leuven, Belgium; 4University Hospitals Leuven, Leuven, Belgium and 5University Hospitals Leuven, Leuven,
Belgium.
Body: Background
Estrogen receptor (ER) positive, HER2 negative breast cancer (BC) can be classified into luminal A and luminal B-like tumors
according to tumor grade. Several evidences point to the fact that IHC expression of the progesterone receptor (PR) has
prognostic value. In this study, we assess to what extent a negative PR in luminal A BCs increases the risk of distant metastasis
and whether or not luminal B BC might have a good prognosis if PR is positive.
Women with primary operable ER positive, HER2 negative BC treated at University Hospitals Leuven between 2000 and 2009
were retrieved from our database. So called luminal A tumors were defined as grade 1-2 BC, whereas so called luminal B BC
were defined as grade 3 BC. Distant metastasis free interval (DMFI) and breast cancer specific survival (BCSS) were investigated
by their PR status. PR was considered negative if the semi-quantitative Allred score was 0-2. Before 2003 the semi-quantitative
H-score was used and a score <50 was considered negative. Statistical analysis for DMFI was performed using the cox
proportional hazards regression. For BCSS we used Fine & Grays competing risk regression. Covariates were age at diagnosis,
tumor size, lymph node status, and therapy.
Results
In total, 3294 patients from Leuven were analyzed. From this cohort, 285 patients experienced metastases (8.7%) and 172
patients died of BC (5.2%). Details are shown in table 1. The median age at diagnosis was 58 years with ages ranging from 22 to
95 years, 2358 patients (71.6%) were aged above 50 at diagnosis. The median follow-up period was 8.1 years.
Table 1: Number of patients that metastasized and died of BC by luminal subgroup and PR status.
luminal A
luminal B
Metastases
BC-related death
PR positive
110/2103 (5%)
61/2103 (3%)
PR negative
16/267 (6%)
9/267 (3%)
PR positive
120/786 (15%)
75/786 (10%)
PR negative
39/138 (28%)
27/138 (20%)
BC: breast cancer, PR: progesterone receptor

In Leuven, the reduction in risk of metastasis in patients with PR positive luminal A and luminal B BC was respectively 14%
(Hazard ratio (HR): 0.86, 95% confidence interval (CI) 0.52-1.51) and 47% (HR: 0.53, 95% CI 0.37-0.78) compared with PR
negative tumors. PR positive luminal A and luminal B BC patients had a 16% (HR: 0.84, 95 % CI 0.42-1.69) and 53% (HR: 0.47,
95% CI 0.30-0.75) reduction in the risk of BC-related death compared with PR negative tumors respectively. The same analysis
was also carried out for postmenopausal patients (older than 50 years) only. In this subcohort, PR positivity was associated with a
9% (HR 0.91, 95% CI 0.50 to 1.83) and 59% (HR 0.41, 95% CI 0.28 to 0.63) reduction in the risk of metastatic events in luminal A
and luminal B lesions, respectively. For BCSS, a 31% (HR 0.69, 95% CI 0.30 to 1.57) and 66% (HR 0.34, 95% CI 0.20 to 0.57)
reduction in the risk of BC-related death for respectively PR positive luminal A and luminal B BC patients was found.
Conclusion
These results suggest that the prognostic effect of PR in primary operable BC depends on the tumor grade. Compared with
luminal PR negative BC, PR positivity improves outcome more in luminal B than in luminal A lesions.

Title: The multidisciplinary application of genomics in clinical practice (MAGIC) survey: Identification of early stage hormone
receptor-positive (HR+), HER2 breast cancer (BC) patients in whom multigene assays may have their highest utility
Matti Aapro1, Juan Enrique Bargallo Rocha2, Michele De Laurentiis3, Roberto Elizalde4, Lszl Landherr5, Barbro Linderholm6,
Terry Mamounas7, Christos Markopoulos8, Miguel Martin9, Patrick Neven10, Alexander Petrovsky11, Dan Rea12, Roman Rouzier13,
Vincent Smit14, Christer Svedman15 and Christoph Thomssen16. 1Multidisciplinary Oncology Institute, Clinic of Genolier, Genolier,
Switzerland; 2Instituto Nacional de Cancerologia, Mexico City, Mexico; 3National Cancer Institute G. Pascale Foundation, Naples,
Italy; 4Hospital Dr. I. Pirovano, Buenos Aires, Argentina; 5Uzsoki Teaching Hospital, Budapest, Hungary; 6Sahlgrenska Academy
and University Hospital, Gothenburg, Sweden; 7University of Florida Health Cancer Center at Orlando Health, Orlando, FL;
8
Athens University Medical School, Athens, Greece; 9Medical Oncology Service, Hospital General Universitario Gregorio
Maran, Madrid, Spain; 10Multidisciplinary Breast Centre and Gynaecological Oncology, UZ Leuven, Leuven, Belgium; 11Russian
Cancer Research Center, Moscow, Russian Federation; 12School of Cancer Sciences, University of Birmingham, Birmingham,
United Kingdom; 13Institut Curie-Universit Versailles-Saint-Quentin, Paris-Saint-Cloud, France; 14Leiden University Medical
Center, Leiden, Netherlands; 15Medical Affairs, Genomic Health, Stockholm, Sweden and 16Martin Luther University of
Halle-Wittenberg, Halle (Saale), Germany.
Body: Background
Treatment recommendations for early stage HR+, HER2 BC patients depend on many factors. The MAGIC survey evaluated
which criteria clinicians use regarding the need for adjuvant chemotherapy (AdjCT) and showed that there was substantial
heterogeneity across clinicians and countries in treatment decisions (Aapro et al, EBCC 2014, abstract 24). Multigene assays
(MGA) help to make more-informed decisions by providing prognostic and predictive information beyond traditional parameters,
but are not always needed. The data presented here show for which BC patient profiles there is a high heterogeneity in treatment
recommendations. We suggest that MGAs may be useful to guide treatment recommendations in these cases.
Methods
From August 2013 until January 2014, physicians with 5 years experience in BC treatment and participating in multidisciplinary
teams were invited for the online MAGIC survey. The survey evaluated respondent characteristics and registered treatment
recommendations for randomly generated early BC patient profiles (n=672). A conjoint analysis was used to assess which patient
attributes were considered for treatment decisions.
Results
The survey was completed by 911 physicians from 52 countries, of whom 72% had >10 years experience. Their treatment
recommendations showed that for BC patient profiles with only high-risk or only low-risk characteristics, there was a high
consensus to recommend AdjCT or no AdjCT (endocrine treatment alone); 42% of the profiles had >75% probability of being
recommended AdjCT and 6% had >75% chance of being recommended no AdjCT.
If interactions between patient characteristics were not considered, age was ranked as the most important patient characteristic
for AdjCT decisions, followed by tumor grade, tumor size, nodal status, and expression of Ki67, estrogen receptor (ER), and
progesterone receptor. The combination of patient attributes and their interactions were, as expected, of importance; some
node-positive patients or patients with a Grade 3 tumor had >75% probability to be recommended no AdjCT (eg, older patients or
patients with a small [<2 cm] tumor). Conversely, some patients with small, Grade 1 tumors had >75% probability to be
recommended AdjCT (eg, young or node-positive patients).
In total, 104 patient profiles (15%) were identified for which treatment recommendations were highly heterogeneous, with a
probability of <50% for both an AdjCT treatment recommendation and no AdjCT as a treatment recommendation. These patient
profiles tended to have the following characteristics: >50 years old, tumor size <3 cm, Grade 1 or 2 tumor, high ER expression,
and Ki67 expression <20%.
Conclusions
There was substantial heterogeneity in treatment recommendations and an overall tendency to give chemo-endocrine rather than
endocrine treatment alone. The highest uncertainty in treatment decisions was seen in patients with intermediate risk by clinical
and pathological parameters. This opens questions concerning treatment decisions and in such cases MGAs may be most useful.

Title: Optimal method of detection and threshold for early intervention to prevent lymphoedema: A multi-centre prospective study
Nigel J Bundred1, Charlotte Stockton1, Katie Riches2, Linda Ashcroft3, Abigail Evans6, Anthony Skene7, Maria Bramley4, Tracey
Hodgkiss4, Arnie Purushotham5, Vaughan Keeley2 and BEA Investigators1. 1University Hospital of South Manchester NHS
Foundation Trust, Manchester, United Kingdom; 2Derby Hospitals NHS Foundation Trust, Derby, United Kingdom; 3Christie NHS
Foundation Trust, Manchester, United Kingdom; 4Pennine Acute Hospitals NHS Trust, Manchester, United Kingdom; 5Guy's and
St Thomas' NHS Foundation Trust, London, United Kingdom; 6Poole Hospital NHS Foundation Trust, Poole, United Kingdom and
7
Royal Bournemouth & Christchurch Hospitals NHS Foundation Trust, Bournemouth, United Kingdom.
Body: Introduction
Women who undergo axillary surgery are at risk of developing lymphoedema. Early detection is recommended by measuring arm
volume from a baseline before surgery to enable early intervention. The optimal measurement method to enable early detection
and time to intervention are unclear. This prospective multi-centre study compares multi-frequency bioimpedance spectroscopy
(BIS, ImpediMed) with the validated perometer method to determine which test is more sensitive for detecting the optimal
threshold to prevent lymphoedema.
Methods
Participants (N = 960) undergoing axillary clearance at 9 UK centres have pre-operative and regular arm volume measurements
post-surgery (1, 3, 6, 9 & 12 months, then 6 monthly), by the validated arm perometry compared with BIS (L-Dex) measurements
as well as self-reported symptoms questionnaire. Change in arm volume was calculated using relative arm volume change
(RAVC). The predictors of lymphoedema development and optimal method were assessed.
Results
Currently 612 patients, median age 55 (range 24 to 90) years, have 6 month follow-up data and 327 have 18 month follow-up
data. Seventy six percent were ER positive and received endocrine therapy, 84% percent received radiotherapy and 67%
received chemotherapy in addition to surgery. Lymphoedema by 18 months was detected in 19% (n=79) of women by perometry
(10% RAVC) and a change in L-Dex of 10 was observed in 31% of women. A moderate correlation between perometer and BIS
at 3 months (r=0.40) and 6 months (r=0.60), with a sensitivity of 73% and specificity of 84% was found.
Univariate analysis revealed a threshold for early intervention to prevent lymphoedema was RAVC 5%-<10% (p=0.03).
Multivariate analysis indicated that Oestrogen Receptor (ER) negative breast cancer (p=0.01, hazard ratio (HR)=0.43, 95%
confidence interval (CI)=0.24 to 0.84), number of positive nodes (p=0.01, HR=1.05, 95% CI=1.01 to 1.09) and a measurement of
5%-<10% (p=0.04, HR=1.67, 95% CI=1.03 to 3.54) at 6 months after surgery predicted development of lymphoedema. Further
investigation of why ER negative patients are at increased risk of developing lymphoedema is planned.
Conclusions
The optimal threshold for early intervention to prevent progression to lymphoedema is 5%-<10% relative arm volume change by
perometry. Further data on the sensitivity of BIS will be obtained in this study. Arm volume measurements remain necessary
before and after ANC to allow early intervention.
(Funded by NIHR Programme Grant).

Title: Postpartum breast cancer demonstrates increased liver and brain metastasis with a proposed role for postpartum involution
Virginia F Borges1, Erica Goddard1, Ann H Partridge2 and Pepper Schedin1. 1University of Colorado, Denver, CO and
2
Dana-Farber Cancer Institute, Boston, MA.
Body: Background: Postpartum breast cancer, which we have defined as a breast cancer diagnosis within 5 years of giving birth
(PPBC), has a 3-fold increased risk for metastasis and death when compared to nulliparous women that is independently
significant when age at diagnosis, tumor stage, receptor status and year at diagnosis are accounted for. Models of postpartum
breast cancer developed by our group have shown that the process of breast involution after lactation ceases or after pregnancy
if lactation does not occurs is a key event that drives tumors toward increase metastasis. To further investigate the increased
metastatic potential of postpartum involution, we performed a nested cohort study of metastatic young womens breast cancer.
We have also developed an immune-competent mouse model of postpartum breast cancer to fully characterize the metastatic
capacity of the postpartum involution milieu. Hypothesis: The reduced survival of postpartum breast cancer is due, in part, to an
altered pattern of metastatic spread to the liver driven by postpartum involution. Methods: Two independent cohorts from the
University of Colorado (UC) and Dana Farber Cancer Institute (DFCI) were utilized to identify a nested cohort of women with first
presentation of systemic metastasis. Location of metastasis was verified and frequency of end organ involvement recorded for the
pre-determined targets of liver, bone, brain and lungs. The cohort was compared by parity status of nulliparous or PPBC. For
determining the contribution of postpartum involution to metastatic spread and investigate mechanism, an intracardiac
immune-competent mouse model of postpartum breast cancer metastasis was developed. Results: Cases of young women
diagnosed 45 that had documented metastatic disease and for which site of metastasis could be identified were included (n=79).
Cases with missing parity data or for whom site specific information was not available were excluded. Overall, the dominant organ
for metastatic involvement at first presentation of metastatic disease was the bones. In PPBC, liver was the second most common
site of metastasis (35%) and higher than in nulliparous women (24%) where lung metastases were the second most frequent.
Individual organs were evaluated by parity status and a notable skewing of frequency of liver and brain metastasis to PPBC was
seen. 63% versus 37% of the cases with liver mets and 73% versus 27% of cases with brain involvement were PPBC versus
nullipara at first metastatic presentation. No differences were seen in lung or bone metastasis between the parity groups. To test
whether the postpartum involution supports increased liver metastasis, mice were injected with mammary tumor cells into the left
ventricle. Similar to the human data, significantly increased liver metastasis, but not bone or lung was observed in the postpartum
group compared to age-matched nulliparous controls.
Conclusion: PPBC demonstrates enrichment for liver and brain metastasis that is in part confirmed through an animal model and
supports the role of postpartum involution in increased metastasis and death. Further studies into the mechanism by which
increased liver metastasis occur in postpartum breast cancer patients are underway.

Title: Genetic polymorphisms (SNPs) as predictive markers for paclitaxel-induced peripheral neuropathy (PNP) and
capecitabine-induced hand-foot syndrome (HFS) in HER-2 negative metastatic breast cancer patients
Siu W Lam1, Charlotte N Frederiks1, Tahar van der Straaten2, Steffen M de Groot3, Agnes Jager4, Monique MEM Bos5, Sabine C
Linn6, Joan van den Bosch7, Hans J Braun8, Ankie MT van der Velden9, Maartje Los10, Jojanneke EA Portielje11, Judith R Kroep2,
Aafke H Honkoop12, Carolien H Smorenburg13, Bea Tanis14, Johanna MGH van Riel15, Jetske M Meerum Terwogt16, Marien O den
Boer17, Joep Douma18, Frank Jeurissen19, Johan Berends20, Henk-Jan Guchelaar2 and Epie Boven1. 1Medical Oncology, VU
University Medical Center, Amsterdam, Noord-Holland, Netherlands; 2Leiden University Medical Center, Leiden, Netherlands;
3
Comprehensive Cancer Centre the Netherlands, Amsterdam, Noord-Holland, Netherlands; 4Erasmus MC Daniel den Hoed
Cancer Center, Rotterdam, Netherlands; 5Reinier de Graaf Hospital, Delft, Netherlands; 6Netherlands Cancer Institute, Antoni van
Leeuwenhoek Hospital, Amsterdam, Noord-Holland, Netherlands; 7Albert Schweitzer Hospital, Dordrecht, Netherlands; 8Vlietland
Hospital, Schiedam, Netherlands; 9Tergooi Hospitals, Hilversum, Netherlands; 10St Antonius Hospital, Nieuwegein, Netherlands;
11
Haga Hospital, Hague, Netherlands; 12Isala Clinics, Zwolle, Netherlands; 13Medical Center Alkmaar, Alkmaar, Netherlands;
14
Groene Hart Hospital, Gouda, Netherlands; 15St Elisabeth Hospital, Tilburg, Netherlands; 16Onze Lieve Vrouwe Gasthuis,
Amsterdam, Noord-Holland, Netherlands; 17Laurentius Hospital, Roermond, Netherlands; 18Rijnstate Hospital, Arnhem,
Netherlands; 19Medical Center Haaglanden, Hague, Netherlands and 20Gemini Hospital, Den Helder, Netherlands.
Body: Background
Newly identified SNPs in genes encoding metabolizing enzymes (CYP2C8, CYP3A4) and drug target (TUBB2A) of paclitaxel
have been associated with PNP1-3. A recent GWAS has found novel SNPs in EPHA5 and FGD4 possibly predictive for PNP4.
Likewise, SNPs in genes (CDA, CES2) involved in capecitabine activation may play a role in HFS5. Here, we attempted to confirm
these SNPs as predictive markers for PNP and HFS in patients (pts) receiving first-line paclitaxel (T) and bevacizumab (A)
without or with capecitabine (X).
In the phase II ATX trial (NTR1348;BOOG2006-06), 312 pts were randomized 1:1 to AT (T 90 mg/m2 d1, 8, 15 & A 10 mg/kg d1,
15 q4w x 6 cycles A 15 mg/kg d1 q3w for next cycles) or ATX (T 90 mg/m2 d1, 8, A 15 mg/kg d1 & X 825 mg/m2 bid d114
q3w x 8 cycles the same dose of A & X q3w for next cycles). Toxicity was graded by using NCI CTCAE v3 at each cycle.
Germline DNA was isolated for genotyping CYP2C8*3 (1196A>G & 416G>A), CYP3A4*22 (522-191C>T), EPHA5 2895G>A,
FGD4 2044-236G>A, CES2 823C>G, TUBB2A 101C>T, TUBB2A 112G>A, CDA 943insC and CDA 451C>T. Results of TUBB2A
and CDA SNPs will be presented at the meeting.
The association between SNPs and toxicity was analyzed for 1) maximum grade of treatment-related toxicity per patient by
Pearsons 2 or Fisher's exact test; 2) cumulative dose level of drugs until first grade 1 toxicity or first dose reduction by
Kaplan-Meier curve and Gehan-Breslow test.
Results
188 pts had SNPs genotyped; characteristics reflected the trial cohort. Table 1 shows the maximum grade of PNP. When grouped
into grades 01 and 23, no differences in rates of PNP were noted among SNPs. Median cumulative dose of T until grade 1
PNP was 1,800 mg/m2 (95% CI 1,5342,065). Carriers of the CYP2C8 416 A-allele had a significantly lower cumulative dose of T
until grade 1 PNP compared to those with a homozygous G/G genotype (Gehan-Breslow p=0.011). Carriers of the FGD4
2044-236 A-allele had a significantly higher cumulative dose until first dose reduction of T compared to those with a homozygous
G/G genotype (Gehan-Breslow p=0.037).
Table 1. toxicity by maximum grade per patient
Grade 0
Paclitaxel-treated patients (N=188)
PNP, No. (%)
62 (33)
33 (36)
41 (22)
17 (9)
Capecitabine-treated patients (N=93)
HFS, No. (%)
17 (18)
21 (23)
25 (27)
30 (32)
For ATX-treated pts, rates of HFS are shown in Table 1. With regard to maximum grade of HFS, there was a trend of a lower rate
of grade 23 HFS for CES 823C/G compared to C/C genotype (33% vs 64%, Fishers exact p=0.059). The median cumulative
dose of X until grade 1 HFS was 98.7 g/m2 (95%CI 82.1115). CES 823C>G was not associated with the cumulative dose of X
until grade 1 HFS or dose reduction.
Conclusions
Our results indicate that CYP2C8 416G>A and FGD4 2044-236G>A might be predictive markers for paclitaxel-induced
neurotoxicity, whereas reported associations of other SNPs with toxicity could not be confirmed.
1Hertz et al. Ann Oncol 2013, 2De Graan et al. Clin Cancer Res 2013, 3Leandro-Garcia et al. Clin Cancer Res 2012, 4Baldwin et
al. Clin Cancer Res 2012, 5Caronia et al. Clin Cancer Res 2011
Financial support from Roche Netherlands.

Title: Prospective study of the impact of the Prosigna assay on adjuvant clinical decision-making in an unselected population of
women with estrogen receptor-positive, HER2-negative, node-negative breast cancer: A GEICAM study
Miguel Martn1,2, Milagros Gonzlez-Rivera2, Serafn Morales3, Juan de la Haba4, Luca Gonzlez-Cortijo5, Luis Manso6, Joan
Albanell7, Antonio Gonzlez-Martn8, Snia Gonzlez9, Angels Arcusa10, Luis de la Cruz-Merino11, Federico Rojo12, Maria Vidal13,14,
Uxue Goicoechea15, Patricia Galvn14, Rosala Caballero15, Eva Carrasco15, Steven Michalopoulos16, John Hornberger16,17 and
Aleix Prat13,14. 1Hospital Gregorio Maran, Madrid, Spain; 2Laboratory of Traslational Oncology. Hospital Gregorio Maran,
Madrid, Spain; 3Hospital Arnau de Vilanova, Lleida, Spain; 4Hospital Reina Sofa, Crdoba, Spain; 5Hospital Quirn, Madrid,
Spain; 6Hospital Doce Octubre, Madrid, Spain; 7Hospital del Mar, Barcelona, Spain; 8MD Anderson Cancer Center, Madrid, Spain;
9
Hospital Mutua de Terrassa, Barcelona, Spain; 10Consorci Sanitari de Terrassa, Barcelona, Spain; 11Hospital Vrgen de la
Macarena, Sevilla, Spain; 12Fundacin Jimnez Daz, Madrid, Spain; 13Hospital de la Vall d'Hebron, Barcelona, Spain;
14
Translational Genomics Group, Vall dHebron Institute of Oncology (VHIO), Barcelona, Spain; 15Spanish Breast Cancer
Research Group (GEICAM), San Sebastin de los Reyes, Madrid, Spain; 16CEDAR Associates, Menlo Park, CA and 17Stanford
University School of Medicine, Palo Alto, CA.
Body: Background: Prosigna (PAM50) is a standardized test that measures the expression levels of 50 classifier genes in
formalin-fixed paraffin-embedded (FFPE) breast tissue tumor samples. It provides subtype classification based on the
fundamental biology of an individual patients tumor (referred to as "intrinsic subtyping"), as well as a prognostic score (referred to
as "risk of recurrence [ROR] score") that predicts the probability of distant recurrence over 10 years. This decision impact study
examines whether the Prosigna test influences both physician and patient adjuvant treatment selection, beyond standard
immunohistochemistry (IHC) testing.
Methods: The analytic sample was comprised of postmenopausal women with node-negative, estrogen-receptor positive (ER+),
HER2 negative (HER2-), early-stage breast cancer with tumors <5 cm (T1-T2). FFPE surgical specimens were analyzed using
Prosigna in a central laboratory. Patients were classified according to the intrinsic tumor subtype (i.e., Luminal A, Luminal B,
HER2-enriched, basal-like) and ROR score (low, intermediate or high risk groups). The primary endpoint was the effect of the
Prosigna test on oncologists treatment recommendations, and the actual treatment received by patients (hormonal therapy
[HT], chemotherapy [CT], chemotherapy and hormonal therapy [CHT]). All samples were analyzed in two independent
laboratories to measure site-to-site concordance. Prosigna subtypes were compared with IHC intrinsic subtypes based on the
St. Gallen 2013 criteria (cut-points: PgR20% and Ki6714%).
Results: A total of 200 patients met eligibility criteria and were enrolled in the study between June 2013 and January 2014.
According to Prosigna results, intrinsic tumor subtypes of these patients were distributed as follows: 129 Luminal A (64.5%), 66
Luminal B (33.0%), 3 HER2-enriched (1.5%) and 2 basal-like (1.0%). Modifications to the adjuvant treatment recommendations
by ROR score can be seen in the following table:
Change in physician pre- to post- Prosigna recommendation
Low ROR (N=101), N (%) Intermediate ROR (N=66), N (%) High ROR (N=33), N (%) TOTAL (N=200), N (%)
HT to CHT 0 (0.0)
8 (12.1)
10 (30.3)
18 (9.0)
CHT to HT 11 (10.9)
9 (13.6)
0 (0.0)
20 (10.0)
17 (25.7)
10 (30.3)
38 (19.0)
TOTAL
11 (10.9)
Treatment decisions changed for 19.0% of all patients: 10.0% and 9.0% of patients went from CHT to HT and HT to CHT,
respectively. The percentage of patients who received chemotherapy in the low, intermediate and high risk groups was 5%, 36%
and 88%, respectively. Both the central and each local laboratory analyzed the samples using IHC. We found 60% concordance
between central IHC and Prosigna intrinsic subtypes (Kappa=0.2365). Prosigna results were consistent across labs (Kappa =
0.89).
Conclusions: The Prosigna test can be reliably performed in hospital laboratories to provide useful information beyond standard
clinical-pathological variables that oncologists can use to optimize adjuvant treatment decisions in clinical practice. Subtype
determined using IHC is not an interchangeable proxy for subtype determined by Prosigna.
*Two last authors have contributed equally to the study.

Title: UK OPTIMA-prelim study demonstrates economic value in more clinical evaluation of multi-parameter prognostic tests in
early breast cancer
Peter S Hall1, Alison F Smith1, Armando Vargas-Palacios1, Robert C Stein3, John Bartlett2, Jane Bayani2, Andrea Marshall4, Janet
A Dunn4, Amy F Campbell4, Carrie Cunningham5, Leila Rooshenas6, Monika Sobol5, Adrienne Morgan7, Christopher Poole8,
Sarah E Pinder9, David A Cameron5, Nigel Stallard8, Jenny Donovan6, Luke Hugh-Davies11, Helena Earl12, Andreas Makris12,
Claire Hulme1 and Christopher McCabe10. 1Leeds Institute of Health Sciences, University of Leeds, Leeds, United Kingdom;
2
Ontario Institute for Cancer Research; 3National Institute for Health Research University College London Hospitals BioMedical
Research Centre, London, United Kingdom; 4Warwick Clinical Trials Unit, University of Warwick; 5Cancer Research Centre,
University of Edinburgh, Edinburgh, United Kingdom; 6University of Bristol, Bristol, United Kingdom; 7Independant Cancer
Patients' Voice; 8University of Warwick, Coventry, United Kingdom; 9King's College London; 10University of Alberta, Edmonton,
AB, Canada; 11Addenbrooke's Hospital, Cambridge, United Kingdom and 12Mount Vernon Cancer Centre, Northwood, Middlesex,
United Kingdom.
Body: Background
There is uncertainty about the benefit of chemotherapy for some patients with ER-positive HER2-negative early breast cancer.
Multi-parameter assays of gene expression may enhance the value of chemotherapy through personalised treatment decisions.
An economic evaluation was undertaken in the context of the feasibility phase of an RCT (OPTIMA prelim) designed to validate
prospectively the use of such an assay as a treatment decision tool in the UK National Health Service (NHS). The aim of the
economic evaluation was to confirm value in an ongoing RCT and optimise its design for economic endpoints. Comparators
included (i) All patients treated with chemotherapy, (ii) Oncotype DX, (iii) MammaPrint/BluePrint and (iv) Prosigna.
Methods
A model-based cost-effectiveness analysis was conducted to the standards of the UK National Institute for Care Excellence
(NICE) reference case. A Markov model was constructed to simulate the care pathway of a cohort of patients with characteristics
identified in the OPTIMA prelim study or, where unavailable, from the published literature. The costs (GBP) and benefits (QALYs)
were estimated over a time horizon of the patient life-time. Alternative scenarios of recurrence rates and chemotherapy effect
were explored in patients identified high or low risk by the tests and treated with and without chemotherapy. Scenarios included
estimates based on the SWOG-8814 trial, the EBCTCG and outcomes forecasted using Adjuvant! Online. Uncertainty introduced
by discrepancy in patient selection between tests was modelled using a Bayesian decision analytic framework. Probabilistic
sensitivity analysis and value of information analysis was conducted using Monte Carlo simulation.
Results
There were 285 randomised patients. Multi-parameter analyses were performed on tumour samples and baseline factors were
included in the model. The cost-effectiveness of all tests was uncertain. Uncertainty was predominantly driven by assumptions
about long term recurrence rates in test-selected groups and the ability of tests to predict benefit from chemotherapy. The
relationship between recurrence-free survival and life expectancy in test-selected groups and in patients who did or did not
receive adjuvant chemotherapy was also important. The incremental cost-effectiveness ratio (ICER) for Oncotype DX compared
with chemotherapy for all was cost-effective in many scenarios, ranging from GBP26,000 per QALY to resulting in increased
QALYs with cost savings (dominate), depending on assumptions. The value of information analysis placed high societal value in
further research into recurrence-free survival for test-directed chemotherapy, irrespective of the test evaluated.
Conclusion
There is substantial value in prospective comparative research into all tests evaluated, including long term outcomes, to resolve
uncertainties in the clinical and economic optimal choice of test.
Acknowledgements
This project was funded by the National Institute for Health Research Health Technology Assessment (NIHR HTA) Programme
(project number 10/34/01). The views and opinions expressed therein are those of the authors and do not necessarily reflect
those of the HTA programme, NIHR, NHS or the Department of Health.


Title: Is there prognostic significance of tumor cellularity in primary non-treated breast carcinoma?
Emily S Reisenbichler1, William Dupont1, Plummer Dale1 and Omar Hameed1. 1Vanderbilt University, Nashville, TN.
Body: Background: Many factors such as tumor size, grade, lymph node and receptor status, either independently or in
combination, as with the Nottingham Prognosic Index (NPI), are known to predict outcomes in non-treated breast cancer. With the
growing use of neoadjuvant therapies, additional prognostic indicators have been identified for evaluating treated carcinomas.
Many post-treatment methods of analysis rely on tumor cellularity (TC) either alone, as in the Miller-Payne system, or in
combination with other tumor features, as in the Residual Cancer Burden (RCB) to predict distant relapse-free survival (RFS). It is
not clear however, whether TC can predict outcomes in non-treated breast carcinoma. The goal of this study was to evaluate the
prognostic value of TC in this particular setting.
Design: TC (%), excluding foci of necrosis and in-situ carcinoma, was determined from histologic review of a representative tumor
section in the primary excision of 366 invasive breast carcinomas and categorized into quartiles. Prior detailed histology review
included tumor size (TS), histological type and grade, receptor and lymph node status, RFS and overall survival (OS). Nottingham
Prognostic Index (NPI) was calculated for each case (0.2 x tumor size (cm) + lymph node stage (1, node negative; 2, 1-3 positive
nodes; 3, 4 positive nodes) + histologic grade).
Results: Mean patient age was 58 yr (range, 21-91) and median follow-up was 87 mo (range, 0.7-165). Invasive ductal carcinoma
of no special type constituted 80% of cases, invasive lobular carcinoma 10%, and other special types of carcinoma, 10%.
Nottingham grades I, II and III, represented 25%, 41% and 32% of the cases, respectively (unknown in 4). Mean NPI was 3.93
(range, 2.066.8). Estrogen receptor was positive in 66% and negative in 25% of cases (unknown in 9%). TC ranged from 2-99%
(mean 47.6%). As expected, NPI was predictive of OS (p=0.000; hazard ratio 1.726; 95% confidence interval 1.45-2.05) and RFS
(p=0.000; hazard ratio 2.011; 95% confidence interval 1.62-2.50). TC, unadjusted for other covariates was not predictive of OS or
RFS (Table 1). The same analysis of ER positive and negative subgroups continued to show no relation of TC to OS or RFS
(Table 2). When adjusted for NPI, TC still showed no significant relation to OS or RFS (data not shown).
Table 1
TC Quartile
OS, unadjusted
RFS, unadjusted
Hazard Ratio*
P Value
0.779
0.315
0.48-1.27
1.073
0.769
0.67-1.73
0.978
0.934
0.58-1.65
0.968
0.913
0.54-1.75
0.921
0.798
0.49-1.73
1.038
0.913
0.53-2.01
Table 2
Estrogen Positive Carcinomas
TC Quartile
OS, unadjusted
Hazard Ratio*
P Value
0.824
0.581
0.42-1.64
1.224
0.535
0.65-2.21
RFS, unadjusted
1.564
0.164
0.83-2.94
0.654
0.299
0.29-1.46
0.813
0.602
0.37-1.77
0.891
0.778
0.40-1.98
Estrogen Negative Carcinomas

TC Quartile
OS, unadjusted
RFS, unadjusted
Hazard Ratio*
P Value
1.090
0.839
0.78-2.49
0.708
0.414
0.31-1.62
0.488
0.138
0.19-1.26
3.048
0.031
1.10-8.41
1.212
0.739
0.39-3.76
0.820
0.759
0.23-2.91
*Relative to 1st quartile

Conclusion: Despite its utility in the neoadjuvant setting, TC does not offer the same prognostic value in the setting of untreated
tumors and is not predictive of OS or RFS in primary non-treated carcinomas.

Title: Bone mineral density change on aromatase inhibitors as a predictor of breast cancer recurrence
Hilary L Martin1,2, John A Davidson1,2, Francis Yap2, Kim Chung2, Muhammad A Khattak1 and Andrew D Redfern1,2. 1Royal Perth
Hospital, Perth, WA, Australia and 2University of Western Australia, Perth, WA, Australia.
Body: Background
Aromatase inhibitors (AIs) reduce the risk of breast cancer recurrence in hormone receptor positive breast cancers by blocking
the production of estrogen. Low estrogen states are also associated with reduction of bone mineral density. AIs have been shown
to lead to reduction in bone mineral density although the degree of change in bone mineral density varies between individuals. In
this study we investigate the association between change in bone mineral density and recurrence in patients treated with AIs.
Methods
This was a retrospective cohort study utilizing a single centre breast unit database. 327 patients were identified who had an initial
bone density T score result at time of commencement of AI and a subsequent result after commencement of AI treatment. There
were 145 patients who had the bone density raw score available in g/cm. Logistic regression (Stata 9) was used to predict
recurrence. Two pathological prognostic factors, tumor size and number of lymph nodes, were shown to predict recurrence in this
data set and were therefore used in multivariate testing. Baseline and sequential data on lumbar spine and hip T score and bone
mineral density (BMD) in the same patient were analyzed for predicting recurrence and hazard of failure using a Cox model. Bone
density in the models was adjusted to show either an alteration of T score of 1.0 or in BMD of 0.01g/cm.
Results
The mean lumbar T score difference was a reduction of 0.26 in sequential measurements (95% CI 0.2 to 0.31). No baseline T
score or BMD showed any significance in predicting recurrence or the hazard of recurrence. Logistic regression modeling showed
a reduction in lumbar spine T score following commencement of AI of 1.0 would decrease the odds of recurrence by 0.28 (95%CI
0.1 to 0.75). A 0.01g/cm decrease in lumbar BMD would decrease the risk of recurrence by 0.79 (95% CI 0.68 to 0.94). Hip T
score if decreased by a 1.0 would decrease the odds of recurrence by 0.27 (95% CI 0.08 to 0.86) and a 0.01g/cm decrease in
hip BMD would decrease the odds of recurrence by 0.81 (95% CI 0.66 to 0.99). The hazard for recurrence from commencement
of AI was also significant in the lumbar spine data for a decrease of 1.0 in the second T scores (HR 0.2 p 0.002) and for the
lumbar BMD reduction of 0.01g/cm (HR 0.78 p<0.0001). The hip T scores and BMD did not reach statistical significance for
predicting the hazard for recurrence.
Conclusion
Our data supports the hypothesis that reduction in bone density post commencement of aromatase inhibitor therapy is associated
with reduced recurrence of breast cancer. It is likely that this association is the result of the effect of low circulating estrogen,
resulting both in reduced risk of recurrence and lower bone density. This effect on bone density may prove a useful surrogate to
measure efficacy of aromatase blockade and consequent anticancer effect on residual breast cancer.

Title: Deconstructing the TNM staging system for breast cancer
Jigar A Patel1, Matthew T Hueman1, Dechang Chen2 and Donald E Henson2. 1Walter Reed National Military Medical Center,
Bethesda, MD and 2Uniformed Services University of the Health Sciences, Bethesda, MD.
Body: Background
The TNM staging system is a standard classification for recording extent of disease in breast cancer. However, with progress in
understanding tumor biology, it is unknown how new prognostic factors that will eventually be integrated with the TNM will affect
its predictive ability. Our objective was to show the impact on 10-year survival rates for breast cancer as different combinations of
prognostic factors are integrated into the TNM.
Methods:
Data were obtained from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute for
the years 1991 through 2000. After exclusions, 132,339 cases of female breast cancer were available. An ensemble clustering
algorithm was used to calculate survival after including additional prognostic factors listed in SEER in the TNM. Combinations of
the following 6 factors were sequentially added to the TNM: tumor grade, ER/PR status, age at diagnosis, racial/ethnic group, and
histological tumor type.
Results:
Survival rates amongst some tumors with the same TNM stage varied as new factors were integrated into the TNM. Factors
associated with favorable outcome usually were associated with better survival than factors associated with less favorable
outcome for each stage group with varying degrees. There were 4 different tumor combinations that represented 4 different TNM
stages that all corresponded to a 90% 10-year survival when additional factors were added to the TNM stage. Integration of
additional prognostic factors led to a crossover in survival of some stage groups. In one combination (T1, N2, grade 1, ER+, PR+,
age <50: 131111) patients who were assigned stage IIIA had a 10-year rate of 90%, which qualifies for a stage I category.
Survival Crossover in TNM Staging
10-year Survival (%)
Prognostic Factor Combination
TNM Stage
Number of Patients
90
113
IA
18993
90
112222
IA
2246
90
21211
IIA
5101
90
122112
IIA
4019
90
31111
IIB
682
90
131111
IIIA
82
58
2232221
IIB
1103
58
22322213
IIB
92
58
3331111
IIIA
128
58
14211111
IIIC
64
**Abbreviated table. Prognostic Factor Combination in order from left to right: T, N, grade, ER status, PR status, age, race,
histological type. * T1 = 1; T2 = 2; T3 = 3; N0 = 1; N1= 2; N2 = 3; N3 = 4; Grade 1 = 1; Grade 2 = 2; Grade 3 = 3; ER+ = 1; ER- =
2; PR+ = 1; PR- = 2; Age <= 50 = 1; Age >50 = 2;
Conclusions:
Integrating new prognostic factors into the TNM always changed the outcome. Survival rates, therefore, are relative and depend
on the selection of prognostic factors. Adding new factors selected different cohorts from the population which had a
heterogeneous population of cancer survivors. These cohorts usually had different survival rates compared with the overall
population from which they were drawn. Integrating combinations of prognostic factors revealed frequent crossover of stage
groups at 10 years, which is a violation of a staging system and could impact the interpretation of clinical trials.

Title: NSAS BC02 substudy of chemo-induced amenorrhea (CIA) in premenopausal women who received either taxane alone or
AC followed by taxane as a postoperative chemotherapy
Fumikata Hara1, Hirofumi Mukai2, Toru Watanabe3, Yukari Uemura4 and Yasuo Ohashi5. 1NHO Shikoku Cancer Center,
Matsuyama, Ehime, Japan; 2National Cancer Center Hospital East, Kashiwa, Chiba, Japan; 3Hamamatsu Oncology Center,
Hamamatsu, Shizuoka, Japan; 4University of Tokyo, Bunkyo, Tokyo, Japan and 5Chuo University, Bunkyo, Tokyo, Japan.
Body: Background: Chemotherapy has a direct cytotoxic effect to breast cancer cells as well as ovarian suppression. In
NSABP-B30 study, Swain, et al demonstrated that CIA contributed to reduce recurrence and prolong overall survival in
premenopausal women with ER positive breast cancer. Thus far, incidence of CIA by anthracycline and cyclophosphamide (AC),
or CMF has been reported. However, there has been no report on CIA by taxane alone therapy. Therefore, it is critically important
to evaluate the incidence of the CIA in NSAS-BC02 (Watanabe T, ASCO2009) which compared taxane alone (q3w Docetaxel
75mg/m2 x8: DTX:, q3w Paclitaxel 175mg/m2 x8: PTX to AC -> taxane (q3wAC 60/600mg/m2 x4 -> DTX x4: ACD, q3wAC x4
-> PTX x4: ACP) in postoperative patients with node-positive breast cancer. In addition, we examined the relationship between
CIA and prognosis in this substudy.
Methods: Menstrual status of all women participating in NSAS BC02 was assessed at study entry, every cycle during
chemotherapy, at 2 months after protocol treatment, and then at every 6 months until 5 years. After 5 years, menstrual status was
assessed annually. Women who were having regular menstrual cycles (premenopause) or irregular menstrual cycle
(perimenopause) at study entry were included in this CIA substudy. We defined CIA as having no menstrual cycle for at least 6
months after chemotherapy.
Results: Of the 1049 women enrolled in NSAS BC02, 395 were analyzed, including 315 with premenopause and 80 with
perimenopause. Median age was 44.2 years old (42-62). Mean body mass index was 22.7 (15.4-38.4). Tumor characteristics
were pathological stage I/IIA/IIB/IIIA 12.7%/39.0%/37.0%/11.4%, ER positivity 56.0% and PgR positivity 54.4%. Of 395 women,
287 (72.7%) was CIA due to protocol treatment. Regarding the type of protocol regimen, proportion of the CIA was 76.9% in ACP,
75.2% in ACD, 62.8% in PTX and 75.2% in DTX. There was no significant difference of CIA frequency between AC followed by
taxane and taxane alone 76.0vs 69.4%, respectively; p=0.14 . Predictive factors of CIA were ACD against PTX (odds
ratio (OR): 2.15), age increase by 5 years (OR: 1.50), and ER negativity (OR: 2.08) according to logistic regression analysis. In
terms of effect to prognosis, CIA was an independent prognostic factor for disease-free survival (DFS) and overall survival in
overall population of substudy according to multivariate Cox analysis. To eliminate guarantee time bias (GTB) (Giobbie-Hurder et
al. JCO 2013), we used time-dependent Cox model. As a result, CIA was not statistically significant prognostic factor of DFS,
even in the subgroup analysis of both ER positive and ER negative patients (Table 1).
Conclusion: Although there has been no data on CIA by taxation alone regimen, eight cycles of taxation treatment caused a high
frequency of CIA in premenopausal women with breast cancer (PTX62.8% and DTX75.2%). It would be cautious to conclude that
CIA was statistically significant association with prognosis, because it might be due to GTB.
Table 1: The effect of CIA on DFS by time-dependent Cox model
DFS in time-dependent Cox model
Subjects
Hazard ratio
95% Confidence interval
p value
Overall
0.77
0.52-1.16
0.21
ER positive
0.60
0.31-1.16
0.13
ER negative
0.82
0.47-1.43
0.49

Title: The 70-gene signature affects adjuvant systemic treatment decisions in breast cancer patients: A population-based,
observational study
Anne Kuijer1, Annelotte CM van Bommel2, Margriet van der Heiden- van der Loo3, Carolien A Drukker4 and Thijs van Dalen1.
1
Diakonessenhuis, Utrecht, Netherlands; 2Leiden University Medical Center, Leiden, Netherlands; 3Comprehensive Cancer
Centre the Netherlands, Utrecht, Netherlands and 4Netherlands Cancer Institute, Amsterdam, Noord-Holland, Netherlands.
Body: Background
Gene signatures (GS), such as the 70-gene signature (MammaPrint), are used as an adjunct to clinicopathological factors to
predict outcome in breast cancer patients. According to the current Dutch national guidelines GS can be used in case of
ambivalence regarding the benefit of adjuvant chemotherapy (ACT). While the impact of GS on the individual patient is well
established, less is known about the impact on predefined patient cohorts in terms of an increase or decrease of the proportion of
patients who receive ACT.
Methods
Patients surgically treated for primary breast cancer between November 2011 and April 2013 were selected from the Netherlands
Cancer Registry. The administration of ACT in these patients was evaluated in relation to the use of the 70-gene signature (In the
Netherlands the 70-gene signature is the most frequently used gene expression profile). Based on the Dutch guidelines clinicians
might be ambivalent regarding the administration of ACT in the following subgroups: patients < 70 yrs with pN0, BRI tumours
>2cm (group A), patients <70 yrs. with pN0, BRII tumours >1cm (group B) and patients < 70 yrs classified as pN1micro, BR I or II
(group C).
Results
A total of 13.122 patients were identified in the Cancer Registry. The 70 gene signature was used in 794 patients; 19 in group A,
227 in group B, 62 in group C, and 456 (57.4%) did not fulfil the ambivalence criteria. In the latter patients, clinicopathological
characteristics were contributed as follows: BR III tumours > 1cm (n=204), age > 70 (n=71), tumours<1cm (n=88), HER2+ (n=92),
ER- (n=83) and N+ (n=166). In the predefined groups A, B and C, ACT was administered less when the 70-gene signature had
been used (P= 0.019, 0.016 and 0.117 respectively; see table 1). The administration of ACT was in line with the test result of the
70-gene signature in 89%, 86% and 84% of the patients in group A,B and C respectively. In categories other than the predefined
A,B and C an inverse relation was seen: more ACT was given when a GS was used (38% without and 54% with a GS) and
adherence to the gene expression test result was lower.
Table 1: administration of adjuvant chemotherapy in relation to the use of MammaPrint
N
Group A
Broup B
Group C
Chemotherapy (%)
No Mammaprint
122
61 (47%)
Mammaprint
19
4 (21%)
Low riks/High risk
15/3
ACT in line with GS
89%
No Mammaprint
1229
507 (41%)
MammaPrint
226
75 (33%)
Low/ risk/High risk
125/74
ACT in line with GS
86%
No MammaPrint
397
209 (53%)
MammaPrint
62
26 (42%)
Low risk/High risk
34/23
ACT in line with GS
84%
p-value
0.019
0.016
0.117
Conclusion
The proportion of patients who receive ACT decreased when a GS was used in predefined cohorts of patients for whom
ambivalence exists regarding the use of ACT. The majority of patients for whom a GS was used did not fit these predefined
categories and in subsets an inverse relation was seen: the use of a gene signature was associated with a higher chance of
receiving ACT.

Title: No Show

Title: Activated form of the estrogen receptor (ER) in breast cancer (BC) and its correlation with prognosis
Erard Gilles1, Jacques Bosq2, Charline Alleaume3, Alexander Zukiwski4, Emile Hutt5 and Jacques Bonneterre5. 1Invivis
Pharmaceuticlas, Bridgewater, NJ; 2Gustave Roussy Institute, Villejuif, France; 3Biodoxis, Romainville, France; 4Arno
Therapeutics, Flemington, NJ and 5Centre Oscar Lambret, Lille, France.
Body: Background: About 50% of ER positive (ERpos) BCs are resistant to hormone treatment. In absence of ligand, ERs are
evenly distributed in nuclei in normal tissue. Upon ligand binding, ERs dimerizes and form a discrete focal subnuclear distribution
pattern (FDP), which is associated with transcriptional activation of ER and can be visualized with high powered microscopy. We
have developed an IHC method to characterize the FDP in archival BC specimens (ASCO 2013 abst#592). We hypothesized
that, in BC, the presence/absence of FDP of ER could predict anti-estrogen (anti-E) activity. We could determine two tumor
phenotypes for ERpos tumors: a diffuse nuclear ER staining or "D-ER" corresponding to the expression of non-functional ER,
which is the pattern observed in vitro or in vivo when no ligand are bound to steroid receptors (SR); D-ER thus is thought to
predict lack of treatment effect of anti-E. And an aggregated nuclear pattern which corresponds to a similar pattern observed in
vitro or in vivo when ligand is bound to ER; A-ER would suggest that ER is activated and a potential target to anti-Es.
Methods: A previously reported study (ASCO 2013 abst #592) was expanded from 254 evaluable cases to 755 with paraffin
embedded formalin fixed (PEFF) BC specimens with clinical and pathology data. Specimens were analyzed for standard HES,
ER, progesterone receptor (PR) and Ki67. The A-ER and D-ER nuclear patterns were analyzed at 1000x magnification.
Results: Mean age; 57 (17 -89). Histology: ductal 85% lobular 13%, other 2%; 82% ERpos and 78% either PRApos or PRBpos,
10% ERpos and 6% PRpos only. 92% of ERpos cases had received anti-Es; Adj. Chemotherapy 36%, Stage: I 47%, II 45%, III 8
%. Grade: I 25%, II 52%, III 23%. Median follow up 42 months. ER status was D-ER in 71% and A-ER in 29% of the specimens.
With DFS defined as time to PD or death (5 year cut off), 125/755 events were observed. ERpos was better that ERneg (HR=
0.36, p = 0.00001). Within ERpos tumors group, in univariate analysis, a time-dependent Cox model showed that A-ER pattern
was associated with better DFS vs D-ER pattern (HR = 0.03, p=0.02, time interaction = 0.01). A-ER was not correlated with SBR
Grade, and was associated with its anisonucleosis (Aniso) index (0.02), with histology (ductal, 0.005) but not age, stage or HER2
status. Ki67 testing is ongoing. In monovariate analysis stage (0.00004), grade (p < 10-6), PR (p < 10-6) were prognostic on DFS,
but not histology and HER2. In a time-dependent multivariate Cox model, A-ER remained an independent predictor (p = 0.013),
with grade (p =0.001 with Mitotic Index 0.002, Differentiation 0.023, Aniso NS), stage 0.001, time interaction 0.009), PR and
HER2 NS.
Conclusions: This study supports the hypothesis that anti-Es are mainly active in BC with A-ER pattern, which is targetable by
anti-Es. Independent statistical significance was reached after adjusting for well-established prognostic factors. Given the 10-year
hormonal treatment adjuvant recommendation guidelines, a better assay than the simple ER status determination would have
important implications in BC management.

Title: Prognostic significance of the interferon metagene in node-negative breast cancer depends on the molecular subtype
Marcus Schmidt1, Leonie van de Sandt2, Karolina Edlund3, Isabel Sicking1, Marco Battista1, Anne-Sophie Heimes1, Antje
Lebrecht1, Gerald Hoffmann1, Mathias Gehrmann4, Jrg Rahnenfhrer2 and Jan G Hengstler3. 1University Hospital Mainz, Mainz,
Germany; 2Technical University Dortmund, Dortmund, Germany; 3Leibniz Research Centre for Working Environment and Human
Factors (IfADo) at Dortmund TU, Dortmund, Germany and 4Bayer GmbH, Leverkusen, Germany.
Body: Background: Interferons are crucial for adaptive immunity and play an important role as central coordinators of
tumor-immune system interactions. We examined the subtype specific prognostic significance of an interferon (IFN) metagene in
node-negative breast cancer.
Methods: Using microarray based gene-expression data, we identified co-regulated genes related to biological processes. After
hierarchical clustering, we defined an interferon (IFN) metagene which was composed of 36 interferon-stimulated genes. The
subtype specific prognostic role of the IFN metagene was analysed in four previously published cohorts (Mainz, Rotterdam,
Transbig, Yu) of node-negative breast cancer patients not treated with adjuvant therapy (n=824). A meta-analysis of previously
published cohorts was performed using a random effects model. Prognostic significance of the IFN metagene for metastasis-free
survival (MFS) was examined in different molecular subtypes: luminal A (ER+/HER2-/aurora kinase A [AURKA]low, luminal B
(ER+/HER2-/AURKAhigh), basal-like (ER-/HER2-), and HER2+.
Results: Prognostic significance of the IFN metagene was restricted to the HER2+ positive molecular subtype (HR 0.50, 95% CI
0.28-0.88, P=0.0056). Prognostic effects were not seen in luminal A (HR 1.00, 95% CI 0.65-1.53, P=0.8819), luminal B (HR 1.00,
95% CI 0.76-1.32, P=0.9770) or basal-like (HR 0.87, 95% CI 0.66-1.15, P=0.3282) carcinomas of the breast.
Conclusions: The prognostic significance of the interferon metagene in node-negative breast cancer is subtype specific and
confined to the HER2+ molecular subtype. A higher expression of the IFN metagene is associated with improved outcome in
HER2+ breast cancer.

Title: Low body mass index (BMI) is associated with poor survival in Japanese patients with early breast cancer; an exploratory
analysis of prospective randomized phase III trials N-SAS BC02 and 03
Yoichi Naito1, Yasuo Ohashi2, Isao Yokota3, Toru Watanabe4, Hiroji Iwata5, Shozo Ohsumi6, Shinji Ohno7, Yasuo Hozumi8,
Seiichiro Yamamoto9, Masato Takahashi10, Tomohiko Aihara11 and Hirofumi Mukai1. 1National Cancer Center Hospital East,
Kashiwa, Japan; 2Chuo University, Tokyo, Japan; 3University of Tokyo, Tokyo, Japan; 4Hamamatsu Oncology Center,
Hamamatsu, Japan; 5Aichi Cancer Center Hospital, Nagoya, Japan; 6National Hospital Organization Shikoku Cancer Center,
Matsuyama, Japan; 7National Hospital Organization Kyushu Cancer Center, Fukuoka, Japan; 8Jichi Medical University,
Shimotsuke, Japan; 9National Cancer Center Research Center for Cancer Prevention and Screening, Tokyo, Japan; 10Hokkaido
Cancer Center, Sapporo, Japan and 11Aihara Hospital, Minoh, Japan.
Body: Background: Obesity is reported to be associated with worse prognosis in early breast cancer. However, there is little data
regarding the impact of low BMI on survival in patients with breast cancer. As obesity is rare and low BMI is relatively common in
Japanese population compared to Caucasians, Japanese cohort is suitable to assess the impact of low BMI on survival in
patients with early breast cancer. Recently an exploratory analysis of a small Japanese randomized phase II trial (JFMC 34-0601)
suggested that low BMI was associated with a decreased overall response rate to neoadjuvant endocrine therapy with
exemestane. We further explored the impact of low BMI on survival in patients with early breast cancer using a dataset of
randomized phase III trials in Japan.
Methods: Patients included in prospective randomized phase III trial N-SAS BC02 and BC03 were retrospectively analyzed.
N-SAS BC02 investigated four arms of adjuvant chemotherapy consisted of taxane alone or in combination with
anthracycline-containing regimen (median follow up of 6.1 years). NSAS BC03 compared anastorozole with tamoxifen as
adjuvant endocrine therapy (median follow up of 6.4 years). The correlation of BMI and overall survival was exploratory analyzed.
This study was supported by the Public Health Research Center Foundation CSPOR.
Results: A total of 1726 patients were included in our study. Median age was 56 (24 82) years, 71.2% of tumors were ER
positive, and 9.7% were HER2 overexpressed. Lymph node metastases were observed in 76% of patients. Mean value of BMI
was 23.3 and only 4.6% of patients had BMI over 30. 33.1% of patients had BMI under 22 and 4.8% had BMI under 18.5. In the
univariate Cox proportional hazard model, lower BMI was significantly associated with worse prognosis (BMI<27 vs >27, HR 0.55,
95% CI 0.32 0.93, p = 0.025). The same trend was observed in multivariate analysis (HR 0.61, p = 0.064).
Conclusion: We confirmed that obese patients were relatively rare in Japanese patients with early breast cancer. In this
non-obese population, lower BMI was correlated with worse prognosis. However these results should be cautiously interpreted.
Our findings suggest that there may be an optimal BMI in patients with early breast cancer and it should be confirmed by another
cohort.

Title: Functional subtyping with BluePrint 80-gene profile identifies two distinct triple positive subtypes with and without
trastuzumab/chemo-sensitivity: Implications for treatment from the NBRST registry
Pat Whitworth1, Jennifer Beatty2, Paul Baron3, Paul Richards4, James Pellicane5, Angela Mislowsky6, Charles Nash7, Laura Lee8,
Mary Murray9, Femke de Snoo10, Lisette Stork-Sloots10, Sarah Untch10, Mark Gittleman11, Stephanie Akbari12 and Peter Beitsch13.
1
Nashville Breast Center, Nashville, TN; 2Breast Place, Charleston, SC; 3Breast & Melanoma Specialists of Charleston,
Charleston, SC; 4Blue Ridge Cancer Care, Roanoke, VA; 5Virginia Breast Center, Midlothian, VA; 6Coastal Carolina Breast
Center, Murrells Inlet, SC; 7Northeast Georgia Medical Center, Gainesville, GA; 8Comprehensive Cancer Center, Palm Springs,
CA; 9Akron General Hospital, Akron, OH; 10Agendia Inc, Irvine, CA; 11Breast Care Specialists, Allentown, PA; 12Virginia Hospital
Center, Arlington, VA and 13Dallas Surgical Group, Dallas, TX.
Body: Background
Classification by molecular subtype can aid in the selection of therapy for patients with breast cancer. However at present, the
methodology for molecular subtyping is not standardized and the methodology and interpretation of results vary between different
laboratories. Subtype is being assigned using conventional immunohistochemistry (IHC) and fluorescence in situ hybridization
(FISH) ("conventional subtype") or molecularly using gene expression profiling. The aim of the current prospective NBRST study
is to compare chemosensitivity as defined by pathological Complete Response (pCR), or endocrine sensitivity as defined by
partial response (PR) using the 80-gene BluePrint functional subtype profile vs. conventional IHC/FISH subtyping.
Methods
MINDACT is an international, prospective, randomized, phase III trial investigating the clinical utility of MammaPrint in selecting
patients with early BC for adjuvant chemotherapy (CT), which enrolled 6,694 patients. Molecular subtyping data were obtained by
MammaPrint and BluePrint (Agendia, Amsterdam, the Netherlands) on frozen samples (n=6,694) classifying patients in the
following subtypes: Luminal A (Luminal-type/MammaPrint Low Risk); Luminal B (Luminal-type/MammaPrint High Risk);
HER2-type; and Basal-type. ER, PgR, HER2 and Ki67 protein status were centrally assessed on FFPE blocks by IHC and/or
FISH in the European Institute of Oncology, Milan, Italy (n=5,740; 86%). Patients were also classified according to the St. 2013
Gallen recommendations [Goldhirsch et al. 2013], which recognizes an additional category (Luminal B-like HER2+).
Results
Ki67 is often used as biomarker to distinguish Luminal A from Luminal B subgroups. The concordance between MammaPrint and
centrally assessed Ki67 in Luminal-type patients is 60%, with a score of 0.26 (95% CI 0.24 0.28) indicating that Ki67 and
MammaPrint cannot reliably substitute for each other. When using a cut-point of 20% instead of 14% the concordance increased
to 78%, with a score of 0.44 (95% CI 0.410.47).
There is a relatively large group of clinical HER2+ cases that are BluePrint Luminal-type (208 out of 541; 38%) indicating that
tumor expression of the Luminal profile is dominant compared with expression of the HER2 profile. These patients have high IHC
ER results and all except for 1 fall into the group that St Gallen separately defines as Luminal B-like HER2-positive. These
patients may have lower response to trastuzumab [von Minckwitz et al. JCO 2012].
98 out of 622 BluePrint Basal-type patients are clinical Luminal HER2-. 2/3 of these patients have low centrally assessed IHC PR
expression and 1/3 have low centrally assessed ER expression (1% and <10%).
Conclusions
Molecular subtyping using BluePrint and MammaPrint leads to a reclassification of 22% (113/515) of tumors. The re-classification
of patients leading to re-assignment to more responsive vs. less responsive groups is most prominent in classically assessed
triple positive patients where 46% of patients are re-assigned to the less responsive BP Luminal-type group (pCR rate of 7%) vs.
46% of patients assigned to the responsive BP HER2-type group (pCR rate of 49%). These findings confirm the more accurate
identification of molecular subgroups for treatment decision by BluePrint functional subtype classifier which may therefore serve
as a better guide for neo-adjuvant treatment than standard, local IHC/FISH assay.

Title: Serumbased test to identify patients with early relapse treated with adjuvant hormonal therapy
Heidi Fiegl1, Christian Marth1, Krista Meyer2, Julia Grigorieva2 and Heinrich Roder2. 1Innsbruck Medical University, Innsbruck,
Tyrol, Austria and 2Biodesix, Boulder, CO.
Body: Background
Prediction of relapse in hormone receptor-positive patients treated with adjuvant hormonal therapy is an area of active research.
Several tissue based genomic tests have been developed and are used in clinical practice [e.g. OncotypeDX (Genomic Health),
MammaPrint (Agendia)] to evaluate the risk of recurrence in early stage breast cancer. However, a costeffective serum based
test, that would allow identification of patients at high risk of early relapse, is of clinical interest.
Methods
We used MALDI ToF Mass spectrometry to obtain mass spectra from pre-surgery serum samples from 499 patients treated with
adjuvant hormonal therapy. Spectra were subjected to pre-processing and 84 peaks (features) were selected for the analysis. We
used a novel proprietary approach, utilizing recent advances in learning theory, to create a diagnostic test to classify patients as
Early Relapse or No Early Relapse. The method creates many multivariate classifiers that are filtered for performance and
combined using logistic regression with dropout regularization into a single master classifier. To avoid bias introduced through a
particular split of training and test, many realizations of the development set are created. The performance of each master
classifier is examined and the classifiers are combined using a majority vote procedure to serve as the final test. The method
allows the use of smaller training sets and minimizes overfitting.
Results
22 out of 499 patients had an early relapse in < 5 years. Samples from these patients were matched, mindful of treatment and
HER2 status, with 22 samples from patients without relapse with the longest duration of relapse free survival (RFS), to serve as
the development set. The remaining samples, including those from patients with late relapses (> 5 years), were set aside for
additional testing. The performance of the 200 master classifiers created from the 200 test/training splits of the development set
was evaluated showing a median overall accuracy of 70%, specificity of 73%, sensitivity of 67%, and hazard ratio (HR) of 3.5.
The final classifier was created using the 200 master classifiers from the realizations of Training and Test splits which were
combined using the majority vote procedure. Classification of patients in the development set from the majority vote of master
classifiers resulted in a significant separation in survival curves (log-rank p<0.0001, HR 6.2 Median RFS Early Relapse 2.6 years,
No Early Relapse not reached) with overall accuracy of 79%, specificity of 82%, and sensitivity of 77%. In the combined
population of all patients, the separation was also significant (log-rank p=0.028, HR 2.0, median RFS not reached in both
classifications). In the multivariate analysis of the overall population, classification remained significant (p=0.007, adjusted HR
3.5) along with menopausal status, nodal status, and tumor size.
Conclusions
We created a classifier from pre-surgery serum samples that can identify patients at risk of early relapse (<5 years) when treated
with adjuvant hormonal therapies. Such a test would have clinical utility in identifying patients who may need a revised adjuvant
treatment strategy.

Title: Age independently predicts opposite disease-specific survivals in luminal A breast cancer patients diagnosed at younger
and older than 50 years of age
Yuanbin Ru1, Richard J Mural1, Patricia S Steeg2, Hallgeir Rui3, Craig D Shriver4 and Hai Hu1. 1Windber Research Institute,
Windber, PA; 2National Cancer Institute, Bethesda, MD; 3Thomas Jefferson University, Philadelphia, PA and 4Walter Reed
National Military Medical Center, Bethesda, MD.
Body: BACKGROUND
It has been reported that women diagnosed with invasive breast cancer (IBC) in their 50s or 60s have better survival than either
younger or older women. Young women with IBCs often have a more aggressive phenotype which contributes to worse survival.
But there is supporting evidence in the literature for both sides whether age alone is an intrinsic driver of the poor outcome in
young women. The availability of large-scale IBC data provided a good opportunity to address this issue.
METHODS
Clinicopathologic and gene expression data from 2 public datasets, including METABRIC from the European Genome-Phenome
Archive (n=1992) and The Cancer Genome Atlas-Breast Cancer project (TCGA-BC, n=980) from National Cancer Institute, were
used in this study. PAM50 was used to derive intrinsic subtypes based on microarray and RNA-Seq data. For a given phenotype,
Fishers exact test was used for its association with age, and 2-sample test for equality of proportions with continuity correction
between young and older patients. Kolmogorov-Smirnovs test was used for equality of age distributions between phenotypes.
Disease-specific survival (DSS) was examined for relationship with age, adjusted for race, AJCC stage, nodal metastasis, tumor
size, grade, and subtype where applicable. Kaplan-Meier estimate and log-rank test were used to generate and compare survival
curves, respectively. Cox proportional hazards model was used for univariate and multivariate analyses and to calculate hazard
ratios (HRs).
RESULTS
Firstly, we confirmed that more aggressive IBCs were enriched in younger patients. Younger patients (<50 years) were diagnosed
with more basal-like subtype (P<0.05) and more node+ diseases (P<0.05) compared with their older counterparts (50 years) in
both datasets. Secondly, we found no significant difference in DSS between these 2 age groups after adjusting for subtype and
other clinicopathologic variables. Finally, we examined the effect of age on DSS within the younger and older patient groups
separately. After adjusting for the effects of other clinicopathologic variables in the METABRIC dataset, the continuously
increasing age was associated with better (HR=0.897, P=0.039) and worse (HR=1.033, P=0.0079) DSS in Luminal A subtype in
the younger and older patient groups, respectively. Age was also associated with worse DSS in Luminal B (HR=1.018, P=0.086)
and all (HR=1.014, P=0.030) subtypes in older patients. The results were validated in the TCGA-BC data (younger: HR=0.877
and P=0.022 in Luminal A; older: HR=1.070 and P=0.066 in Luminal A, HR=1.081 and P=0.034 in Luminal B, HR=1.055 and
P=0.0052 in all patients).
CONCLUSION
Our results suggested that age was an independent predictor of better DSS in younger patients, but worse DSS in older patients,
especially those with Luminal A subtype. The opposite effects of age on DSS in younger and older patients warrant further
molecular studies.
The views expressed in this abstract are those of the authors and do not reflect the official policy of the Department of Defense,
or U.S. Government.


Title: No Show

Title: Glucose-regulated protein 78 and C/-EBP homologous protein predict disease-free survival and responsiveness to
chemotherapy in breast cancer
Yi-Zi Zheng1,2, Zhi-Gang Cao1,2, Xin Hu1 and Zhi-Ming Shao1,2,3. 1Key Laboratory of Breast Cancer in Shanghai, Fudan University
Shanghai Cancer Center, Fudan University, Shanghai, China; 2Shanghai Medical College, Fudan University, Shanghai, China
and 3Institutes of BioMedical Sciences, Fudan University, Shanghai, China.
Body: Background
Cancer cells are generally under endoplasmic reticulum(ER) stress. Notably, the ability of cells to respond to ER stress is critical
for cell survival, and chronic or unresolved ER stress can lead to apoptosis. Glucose-regulated protein (GRP) 78 and C/-EBP
homologous protein (CHOP) are commonly used as markers of endoplasmic reticulum (ER) stress. As an ER chaperone, GRP78
functions as a potent anti-apoptotic factor and confers drug resistance, whereas CHOP is a key initiating factor of ER
stress-related cell death. The clinical implications of GRP78 and CHOP, however, have not been fully studied in breast cancer. In
this study, we aimed to investigate the predictive value of GRP78 and CHOP in breast cancer patients who underwent adjuvant
chemotherapy.
Methods: An immunohistochemistry screen for GRP78 and CHOP was performed using a tissue microarray (TMA) containing
250 tumors from female patients diagnosed with invasive ductal breast carcinoma at the Fudan University Shanghai Cancer
Center. The staining results were scored semi-quantitatively, and a prediction model was constructed to verify the hypothesis.
Results: In this retrospective study cohort, positive GRP78 staining was detected in 52.6% (n = 112; 52.6% positive, 47.4%
negative) of tumors, and CHOP staining was present in 56.3% (n = 120; 56.3% positive, 43.7% negative) of cases. In
Kaplan-Meier analysis, CHOP correlated with prolonged disease-free survival (DFS; P = 0.001), whereas GRP78 showed an
opposite association (P < 0.001). Moreover, in a GRP78-positive subset, CHOP overexpression correlated with a lower risk of
recurrence. A further multivariate COX analysis revealed that positive GRP78 staining cases exhibited a higher likelihood for
disease events (HR = 4.573; 95% CI: 2.291-9.128; P < 0.001), while CHOP positivity was indicative of lower risk for recurrence
(HR = 0.385, 95% CI: 0.215-0.688; P = 0.001). In the receiver operating characteristic (ROC) analysis, the prediction capability of
the predictive model combining the above two markers surpassed that of a traditional model (P = 0.0085 for the area under the
curve comparison). Within the anthracycline-treatment subgroup, the combined GRP78 and CHOP exhibited similar predictive
significance.
Conclusions: Collectively, our findings suggest a tight association between ER stress markers and clinical outcomes for female
patients diagnosed with invasive ductal breast carcinoma. Adding these two ER stress markers to traditional prognostic factors
provides a more sensitive and accurate predictive model for female patients with invasive ductal breast cancer. Moreover, the
combination of GRP78 and CHOP has predictive value for responsiveness to anthracycline-based adjuvant breast cancer
chemotherapy. Thus, our findings provide insights into further applications of GRP78 and CHOP, potentially providing additional
predictive information for oncologists with regard to choosing treatment regimens.

Title: Prognostic and predictive value of an integrated mRNA-lncRNA signature in triple-negative breast cancer: A comprehensive
transcriptome analysis
Yizhou Jiang1, Yi-Rong Liu1, Ke-Da Yu1, Xin Hu1, Xiao-En Xu1, Ling Yao1 and Zhi-Ming Shao1. 1Fudan University Shanghai
Cancer Center, Shanghai, China.
Body: Purpose
Triple-negative breast cancer (TNBC) is a highly diverse group of disease, and clinical outcome of patients with TNBC is highly
variable. Due to the heterogeneity of TNBCs, there is limited molecular signature routinely used for predicting the risk of disease
recurrence and benefit of adjuvant chemotherapy. Our study aims to develop and validate a RNA signature, integrating
messenger RNAs (mRNAs) and long non-coding RNAs (lncRNAs) together, for TNBC patients to improve risk stratification and
avoid unnecessary adjuvant therapy.
Methods
Using transcriptome microarrays, we analyzed 33 paired TNBC and adjacent normal breast tissues, and identified 1,644 mRNAs
and 1,047 lncRNAs which were differentially expressed between tumors and normal tissues. We further determined the
expression of these mRNAs and lncRNAs in an additional 134 TNBC samples using transcriptome microarrays, and confirmed
their associations with patients' recurrence-free survival (RFS). Using the LASSO Cox regression model, we built an integrated
mRNA-lncRNA signature incorporating seven mRNAs and three lncRNAs. Prognostic and predictive accuracy of the signature
was tested in the training set of 167 TNBC patients and further validated in an independent validation set of 143 TNBC patients.
Results[/br]In the training set, we identified 36 mRNAs and 32 lncRNAs which were tumor-specific and significantly associated
with patients' RFS. Using the LASSO Cox regression model, an integrated mRNA-lncRNA signature based on seven mRNAs
(CCR4, CTSB, ERO1L, HIF1A, IGFR1R, SRD5A1 and TFF1) and three lncRNAs (n381928, n333541 and
TCONS_I2_00013109-XLOC_I2_007048) was developed in the training set and subsequently validated in the validation set.
Patients were classified to the high-risk (high risk of recurrence) and low-risk (low risk of recurrence) groups according to their
scores in the signature. In the training set, multivariate analysis showed that the predicted high-risk group had higher risk of
developing recurrent disease within five years of surgery than the low-risk patients (hazard ratio [HR] = 6.12, 95% confidence
interval [CI] 2.76-13.56, P<0.001). In the validation set, the predicted high-risk group also had poorer RFS in multivariate analysis
(HR = 5.09, 95% CI 1.82-10.96, P<0.001). Furthermore, analyzing the areas under the time-dependent receiver operating curve
for five-year RFS, we proved the integrated mRNA-lncRNA signature had better prognostic value than the seven-mRNA-only
signature and clinicopathological risk factors in both the training and validation sets. Finally, Cox proportional hazards models
were utilized to test the interaction between adjuvant chemotherapy and different risk groups. Patients in the low-risk group had a
more favorable response to adjuvant chemotherapy both in the training and validation sets.
Conclusion
The integrated mRNA-lncRNA signature is a reliable tool for predicting disease recurrence and benefit of adjuvant chemotherapy
in TNBC patients. If further validated in larger population, it could facilitate patient counseling and individualize treatment of
TNBC.


Title: The thrombin clotting pathway is upregulated in the stroma of pre-invasive breast cancer and further upregulated in
aggressive invasive breast cancer phenotypes
Hudhaifah Shaker1,2, Nigel J Bundred1, Harith Albadry3, Sarah L Nicholson3, Susan Pritchard3, Karin Jirstrm4, Goran Landberg2
and Cliona C Kirwan1. 1University of Manchester, Manchester Academic Health Science Centre, University Hospital of South
Manchester, Manchester, United Kingdom; 2Breakthrough Breast Cancer Unit, Cancer Research UK Institute Manchester,
Manchester, United Kingdom; 3University Hospital of South Manchester, Manchester, United Kingdom and 4Oncology and
Pathology, Lund University, Skane University Hospital, Lund, Sweden.
Body: BACKGROUND
Components of the thrombin (extrinsic) clotting pathway are upregulated in cancer. The clotting pathway factors tissue factor (TF)
and Thrombin promote tumour progression through protease activated receptors PAR2 and PAR1 respectively.
AIMS
To determine if tumour expression (epithelial and stromal) of a procoagulant phenotype is associated with aggressive breast
cancer phenotypes and reduced survival.
METHODS
Tumour expression of TF, thrombin, PAR1 and PAR2 was determined by immunohistochemistry in two cohorts.
PROSPECTIVE STUDY
Early invasive breast cancer (n=199), ductal carcinoma in situ (DCIS, n=42) and normal breast tissue samples (n=121).
RETROSPECTIVE STUDY
Early invasive breast cancer patients (n=144) with median follow-up of 69 (range 4 to 91) months.
Procoagulant phenotype expression was correlated with tumour grade, proliferation (Ki67), ER and HER2 status (both cohorts),
survival and recurrence (retrospective cohort).
RESULTS
PROSPECTIVE STUDY
Epithelium
Thrombin (p<0.01) but not TF, PAR1 or PAR2 was increased in invasive cancer compared to DCIS and normal breast tissue.
Stroma
TF, Thrombin, PAR1 and PAR2 were increased in the stroma of DCIS compared to normal breast stroma (p<0.05, all).
In invasive breast cancer, TF was increased in invasive cancer compared to DCIS and compared to normal breast tissue (p<0.01,
both). Thrombin, PAR1 and PAR2 were increased in invasive cancer compared to normal breast tissue (p<0.01, all).
TF, thrombin, PAR1 and PAR2 were increased in high proliferating (p<0.01, all) and high grade cancer (p<0.01, all). TF (p=0.02)
and PAR1 (p<0.01) were increased in ER negative cancer. TF, thrombin and PAR2 was increased in HER2 positive cancer
(p<0.01, all).
RETROSPECTIVE STUDY
Stroma
As with the prospective study, thrombin and PAR2 expression was increased in high proliferating cancer (p<0.05, both) and
thrombin was increased in high grade cancer (p<0.05). PAR1, TF and thrombin expression was increased in ER negative cancer
(p<0.05, all) and PAR2 was increased in HER2 positive cancer (p=0.05).
Overall (OS) and disease-free survival (DFS)
PAR1 stromal expression was an independent predictor of reduced OS (HR 3.3, 95% CI 1.3-8.3, p=0.01) but did not correlate
with DFS.
There was no association between epithelial PAR1 expression or epithelial or stromal TF, thrombin or PAR2 expression and DFS
or OS.
CONCLUSION
Stromal upregulation of the thrombin pathway occurs in in-situ cancer, implying cancer-stromal communication at the pre-invasive
stage. Stromal thrombin pathway components may have a role in the transition of pre-invasive to invasive cancer.
Stromal (but not epithelial) thrombin pathway upregulation is associated with aggressive invasive breast cancer phenotypes and
reduced survival. The thrombin pathway may provide a novel therapeutic target, particularly in ER negative, HER2 positive breast
cancer.

Title: Linking genotype to clinical outcome in breast cancer by combining NGS and gene chip data
Balazs Gyorffy1,2,3, Lorinc Pongor1 and Mate Kormos1. 1MTA TTK Lendlet Cancer Biomarker Research Group, Budapest,
Hungary; 2MTA-SE Research Group for Pediatrics and Nephrology and 3Semmelweis University.
Body: Introduction: Next generation sequencing (NGS) provides the possibility to measure mutational status for any part of any
gene. However, because of scarce data available to date, linking these mutations to relevant clinical outcome in a large number
of patients is not possible.
Aim: Our goal was to combine available genotype data generated by using NGS with gene expression data generated by gene
chips to establish a framework to assess the effect of genotype on clinical outcome.
Methods: NGS data generated by the TCGA consortia and publicly available gene chip data obtained from the GEO and EGA
repositories were utilized. NGS data was processed using MuTect, SNPeff, GRCh37 and R. RNA-seq data was normalized using
DEseq. Gene chip data was MAS5 normalized. Generation of the transcriptomic fingerprint for mutation status was computed by
ROC utilizing the RNA-seq data. In the gene chip data, the average expression of significant genes identified was designated as
a metagene for the given genotype. Correlation to survival for this metagene was assessed by computing Cox regression and
plotting Kaplan-Meier survival plots. Finally, we have set up an online interface to enable running the analysis for any selected
gene.
Results: The database contains 332 NGS samples containing mutational status for 22,938 genes and RNA-seq data for 10,987
genes. The gene chip database contains 5,934 patients with 10,987 genes plus detailed clinical characteristics and survival data.
We evaluated correlation to outcome for previously identified genes harboring the ten most common somatic mutations in breast
cancer. Of these, TP53 (n of mutations out of 332=93, hazard rate=0.51, p<1E-16), AKT1 (n=18, HR=1.6, p=1.6E-15), PIK3CA
(n=119, HR=1.5, p=8.5E-12), MAP3K1 (n=20, HR=1.4, p=1.3E-08), CDH1 (n=34, HR=1.3, p=4.4E-07), and RB1 (n=21, HR=1.3,
p=7E-06) reached statistical significance while PI3K, PTEN, CDKN1B and GATA3 were not significant or had insufficient number
of mutated samples.
Discussion: By connecting genotype to gene expression signature and employing this signature for survival analysis we have set
up a pipeline enabling the functional validation of a discovered mutation for any gene in a large breast cancer cohort.

Title: Elevated neutrophil lymphocyte ratio predicts survival in breast cancer
Cher Hui Koh1, Nirmala Bhoo-Pathy2,3, Khoon Leong Ng1, Mee Hoong See1, Gie Hooi Tan1, Suniza Jamaris1 and Nur Aishah
Taib1. 1Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia; 2Julius Centre University of Malaya, Faculty of
Medicine, University of Malaya, Kuala Lumpur, Malaysia and 3Julius Center for Health Sciences and Primary Care, University
Medical Center Utrecht, Utrecht, Netherlands.
Body: Background: Host inflammatory response affects disease progression and survival in cancer. While the elevation of
neutrophil/lymphocyte ratio (NLR) and platelet/lymphocyte ratio (PLR) have been associated with poor prognosis in colorectal
cancer, evidence on prognostic significance of these indicators in breast cancer is sparse. We determined whether the initial,
pretreatment NLR and PLR impact survival of patients presenting with primary breast cancer.
Methods: Of 2059 consecutive patients, newly diagnosed with breast cancer in University Malaya Medical Centre, Malaysia
between January 2000 and December 2008, we only included 1447 patients with available differential blood count at time of
diagnosis (70%). Data on tumor characteristics and treatment were obtained from the hospital-based breast cancer registry
whereas vital status was verified with the national mortality registry. Patients were stratified into quintiles of NLR and PLR.
Differences in demography, tumor characteristics and treatment patterns between the quintiles were assessed. Relative survival
rates (RSR) were estimated using the Malaysian population mortality data. Multivariable Cox regression was used to determine
the independent prognostic significance of NLR and PLR.
Results: Median age at presentation was 52 years, whereas median tumor size at diagnosis was 3.5 cm with 48% of patients
presenting with axillary lymph node involvement. Both NLR and PLR were positively correlated (p<0.001). Compared to patients
in the lowest NLR quintile (NLR1.38), those in the highest quintile (NLR 3.95) were younger at diagnosis (median: 49 years
versus 56 years; p<0.001), and presented with bigger tumors (median: 5.0 cm versus 3.0 cm: p<0.001), axillary lymph node
metastasis (55.2% versus 45.9%; p=0.035), distant metastases at diagnosis (32.0% versus 7.2%: p<0.001), and higher tumor
grades (49.6% versus 33.9%: p=0.016). A total of 598 deaths were observed during 9248 person-years of follow-up,
corresponding with a 5-year RSR of 70.3% (95%CI: 67.7%-72.8%). Higher NLR quintiles were significantly associated with poorer
survival; 5-year RSRs were 75.6% (95%CI: 68.9%-81.4%) in quintile 1, 79.4% (95%CI: 74.5%-83.7%) in quintile 2, 72.3%
(95%CI: 66.5%-77.5%) in quintile 3, 65.5% (95%CI: 59.7%-70.9%) in quintile 4, and 51.8% (95%CI: 44.4%-58.7%) in quintile 5.
Following adjustment for age, ethnicity, AJCC6 stage, PLR, tumor grade, lymphovascular invasion, hormonal receptor status,
locoregional management, chemotherapy, and hormone therapy, the hazard ratios (HR) for the second to fifth quintiles of NLR
compared to the first quintile were 1.06 (95%CI:0.801.40), 1.15 (95%CI:0.861.55), 1.59 (95%CI:1.212.10), and 2.42
(95%CI:1.813.21) respectively; p for linear trend test <0.001. Patients in the highest PLR quintile were also significantly
associated with decreased survival compared to those in the lowest quintile; 5-RSR: 53.2% (95%CI: 46.9%-59.1%) versus 76.9%
(95%CI: 70.9%-82.1%), respectively. Nevertheless, this association was not significant following multivariable adjustment (HR:
0.99, 95%CI:0.76-1.29).
Conclusion: High NLR seems to be an independent prognostic factor for breast cancer but not PLR. These findings warrant
further validation.

Title: Prognostic value of axillary nodal ratio after neoadjuvant chemotherapy of AC followed by docetaxel: A multicenter
retrospective cohort study
Se Hyun Kim1, Jee Hyun Kim1, Tae-Yong Kim2, In Sil Choi3, Yee Soo Chae4, Sun Kyung Baek5, Seok Yun Kang6, In Hae Park7,
Yoon Ji Choi8, Soohyeon Lee9, Joo Hyuk Sohn9, Yeon-Hee Park10, Young-Hyuck Im10, Jin-Hee Ahn11, Sung-Bae Kim11 and Kyung
Hae Jung11. 1Seoul National University Bundang Hospital, Seong-nam, Korea; 2Seoul National University Hospital, Seoul, Korea;
3
SMG-SNU Boramae Medical Center, Seoul, Korea; 4Kyungpook National University Hospital, Daegu, Korea; 5Kyung Hee
University Medical Center, Seoul, Korea; 6Ajou University Hospital, Suwon, Korea; 7National Cancer Center, Goyang, Korea;
8
Korea University Ansan Hospital, Seoul, Korea; 9Severance Hospital, Seoul, Korea; 10Samsung Medical Center, Seoul, Korea
and 11Asan Medical Center, Seoul, Korea.
Body: Background: The ratio of involved to retrieved lymph nodes (LNR) is suggested as a prognostic factor in operable breast
cancer. However, there are conflicting results regarding its clinical significance after neoadjuvant chemotherapy. We investigated
the prognostic value of LNR with a thorough evaluation of potential prognostic factors in a large cohort constructed from Health
Insurance Review and Assessment Service database of Korea.
Patients and method: This retrospective analysis is based on the data of 814 patients with clinical stage II/III breast cancer
treated with four cycles of adriamycin/cyclophosphamide (AC) followed by four cycles of docetaxel (DOC) before surgery. We
evaluated the clinical significance of the LNR (3 categories: Low, 0-0.20 vs. Intermediate, 0.21-0.65 vs. High, 0.66 -1.00) using
Kaplan-Meier method, log-rank test, and Cox proportional hazard regression model.
Result: A total of 799 patients underwent breast surgery (Median age 45, range 16-74; Mastectomy 369, Lumpectomy 380, and
Others 50). Axillary lymph node dissection was performed in 704 (88.1%) patients. Pathologic complete response (pCR,
pT0/isN0) was achieved in 129 (16.1%) of 799 patients (HR+/HER2-, 34/373 [9.1%]; HER2+, 45/210 [21.4%]; TNBC 50/216
[23.1%]). The mean numbers of involved LN and retrieved LN were 2.70 (range 0-42) and 13.98 (range 1-64), respectively. The
mean LNR was 0.17 (Low, 574 [71.8%]; Intermediate, 170 [21.3%]; High, 55 [6.9%]). In univariate analysis, LNR was significantly
associated with worse relapse-free survival (3-yr RFS rate 84.8% in low vs. 66.2% in intermediate vs. 54.3% in high; P <0.0001,
log-rank test). In multivariate analysis, LNR was not significantly associated with recurrence after adjustment of other clinical
factors (Age, histologic grade, intrinsic subtype, ypT-stage, ypN-stage, lymphatic or vascular invasion, and pCR).
Multivariate analysis for relapse-free survival
P-value
HR
95%CI
AGE (<50, 50)
0.157
ypT-stage
<0.0001
ypN-stage
0.035
pCR (T0/isN0)
0.027
Lymphovascular invasion
0.040
Subtype
<0.0001
Histologic grade
0.001
LNR Low (0-0.20)
0.954
1.00
LNR Intermediate (0.21-0.65)
0.973
1.01
0.55-1.86
LNR High (0.66-1.00)
0.797
1.12
0.48-2.59
Conclusion: LNR is not superior to ypN-stage in predicting clinical outcome of breast cancer after neoadjuvant chemotherapy.

Title: p16INK4a expression and chemotherapy toxicity in women with early stage breast cancer
Hyman Muss1, Allison Deal1, Arti Hurria2, Natalia Mitin1, Chad Torrice1, Krishnamurthy Janakiraman1, Trevor Jolly1, Grant
Williams1, Shani Alston1, Jerard West1, Laura Zavala2, Vani Katheria2 and Norman Sharpless1. 1UNC Lineberger Comprehensive
Cancer Center, Chapel Hill, NC and 2City of Hope.
Body: Background: Increased expression of p16INK4a, a molecular marker of aging, is a hallmark of increased cellular
senescence in most mammalian tissues. In human peripheral blood T-lymphocytes, expression of p16INK4a increases 10-fold
between 20 and 80 years of age (Liu et al, Aging Cell, 2009). We hypothesized that higher "molecular age," as evidenced by
increased T-cell expression of p16INK4a at the time of initiation of breast cancer (BC) chemotherapy (CRx), predicts increased
treatment-related toxicity.
Methods: Patients (pts) with early breast cancer scheduled to receive neoadjuvant (NA) or adjuvant (Adj) chemotherapy (CRx)
had p16INK4a evaluation performed prior to treatment. Expression of p16INK4a mRNA in CD3+ T-lymphocytes was determined
using TaqMan real time quantitative reverse transcription polymerase chain reaction. Grade 3 and 4 (G3/4) hematologic (H) and
non-hematologic (NH) toxicities (NCI CTCAE version 4) were assessed during and within 4 weeks of completion of CRx.
Wilcoxon Rank Sum tests compared p16INK4a between groups.
Results: 93 pts with Stage I-III BC and complete toxicity data during and within 4 weeks after completion of CRx have been
accrued. Median age (range) was 52 (25-76). 48 (52%) of pts were hormone receptor (HR) positive, 21 (23%) HER-2 positive,
and 24 (26%) triple-negative. 39 (42%) received NA and 54 (58%) received Adj CRx. 57 pts (61%) received an anthracycline
containing combination and 21 (23%) cyclophosphamide/docetaxel (TC). 85 pts (91%) received pegfilgrastim. Overall, 69 pts
(74%) had a G3/4 toxicity during or 4 wks of CRx, 54% NH and 40% H. 22 pts (24%) were hospitalized for CRx related toxicity
and 11 (12%) of these admissions were for neutropenic fever (NF). 23 (25%) reported G3/4 fatigue during CRx and had
significantly higher p16INK4a values at baseline than those without fatigue (p=0.03). There was no significant association of
baseline p16INK4a and other NH or H toxicity or hospitalization in this cohort.
Conclusion: In this small sample of pts treated with anthracycline and non-anthracycline containing NA and Adj CRx regimens,
there was a significant association of baseline p16INK4a and G3/4 fatigue. The heterogeneity of treatment and use of pegfilgrastim
in almost all pts limits the power of this study to find significant relationships between p16INK4a and other toxicities. The cohort for
this study is being expanded to further explore p16INK4a as a predictor for G3/4 toxicity.
Support: Breast Cancer Treatment Foundation, New York, NY; University Cancer Research Fund, University of North Carolina,
Chapel Hill.

Title: Pre-surgical neutrophil-to-lymphocyte ratio (NLR) is a prognostic indicator of recurrence free and overall survival in breast
cancer patients undergoing primary surgery
Derbrenn O Connor1,2, Mark L Griffin1, Jenna S O'Sullivan2, Sean Millar2, Jo O'Keeffe1, Brian R Bird1,2, Sandra Deady3 and
Conleth G Murphy1,2. 1Bon Secours Hospital, Cork, Ireland; 2University College, Cork, Ireland and 3National Cancer Registry of
Ireland, Cork, Ireland.
Body: Background
There is growing evidence that elevated neutrophil-to-lymphocyte ratio (NLR) is an independent prognostic indicator associated
with poor survival in various cancers including colon cancer, ovarian cancer, esophageal cancer and gastric cancer. Several
studies in early breast cancer suggest that NLR at diagnosis may be an independent negative prognostic marker in this
population also. The aim of the current study is to evaluate the association between blood NLR immediately prior to surgery and
recurrence free and overall survival in breast cancer patients.
Methods
We performed a single institution, retrospective cohort study including all patients treated for invasive breast cancer amenable to
primary surgery at our institution between 1st January 2006 and 31st December 2010. Clinical and pathologic details were
collected from the patient medical records. Exclusion criteria included prior malignancy, chemotherapy receipt prior to surgery,
recent corticosteroid use, systemic autoimmune conditions, recent significant cardiovascular illness, infection or inflammatory
condition. NLR was calculated on the most recent complete blood count performed on the day of surgery or at the pre-surgical
assessment. Eligible patients were divided into high (4) and low (<4) NLR groups.
Results
We identified 357 patients, of whom 223 met eligibility criteria for analysis. At a median follow-up of 55.8 months, 18 patients
(8.1%) died and 32 (14.3%) experienced disease recurrence. Kaplan Meier survival curves revealed significantly inferior overall
survival (log-rank p=0.003) and recurrence free survival (log-rank p=0.01) in the high NLR group. Univariate Cox proportional
hazard regression demonstrated an increased risk of mortality and breast cancer recurrence with pre-treatment NLR4, with
hazard ratios of 5.49 (p=0.008, 95% CI 1.56 to 19.37) and 3.68 (p=0.016, 95% CI 1.28 to 10.58) respectively.
Conclusion
This study confirms pre-treatment NLR as a prognostic factor for breast recurrence and death among patients receiving curative
surgery for early breast cancer. Strict inclusion criteria reduced the likelihood of confounding due to comorbidities which might
affect NLR and be independently associated with poor outcomes. Our study supports the usefulness of NLR as a component of
the prognostic assessment of early breast cancer patients.

Title: SA02 trial: Results of a genomics-based prospective cohort in node-positive early breast cancer with good-prognosis
signature treated with adjuvant chemotherapy
Bertucci Franois1, Extra Jean-Marc1, Ferrero Jean-Marc2, Bachelot Thomas3, Autret Aurlie1, Boyer-Chammard Agns1 and
Viens Patrice1. 1Institut Paoli-Calmettes, Marseille, France; 2Centre Antoine Lacassagne, Nice, France and 3Centre Lon Brard,
Lyon, France.
Body: Background. Adjuvant chemotherapy (CT) for node-positive (N+) early breast cancer (EBC) is based upon an
anthracycline-taxane combination. However, the benefit of taxane is limited to a small population, but associated to morbid and
financial costs, making crucial the identification of patients likely to benefit or not from anthracycline-based regimen without
taxane. Using DNA microarrays to profile a retrospective series of 498 patients (pts) treated with adjuvant anthracycline-based CT
without taxane, we had identified and validated a gene expression signature (GES) associated with metastatic relapse. The
corresponding Relapse Score (RS) sorted the patients (pts) in two groups: the "good-prognosis" group (75% of pts) with a 5-year
metastasis-free survival (MFS) of 82%, and the "poor-prognosis" group (25% of pts) with a 56% 5-year MFS. We present here the
results of a prospective multicentric national cohort of 175 pts, SA02, initiated to analyze anthracycline-based adjuvant CT without
taxane in N+ EBC pts with a "good-prognosis" RS, with the aim of confirming their good prognosis in term of 5-year MFS.
Methods. Women with surgical EBC were screened for inclusion in 4 French hospitals. After diagnosis of lymph node
involvement, frozen tumor samples were used for hybridization on Affymetrix U133 Plus 2.0 microarrays, and the RS was defined.
RS-based "good-prognosis" pts were treated in the SA02 cohort and received 6 FEC100 cycles (Fluoro-uracile 500 mg/m2,
Epirubicin 100 mg/m2, Cyclophosphamide 500 mg/m2, every 21 days), followed by adjuvant radiotherapy, hormone therapy
and/or trastuzumab according to standard guidelines. Pts with a "poor-prognosis" or non-evaluable RS were not included in the
cohort. We present here an analysis with a median follow-up of 50 months.
Results: Between May 2007 and May 2010, samples from 175 eligible N+ EBC pts were collected for gene profiling. The profiling
failed for 54 samples (31%), due to insufficient RNA amount, poor RNA quality, poor labeling performance, or forgotten. 102 pts
(84%) were defined as "good-prognosis", while only 5 pts (4%) were defined as "poor-prognosis", and 14 pts (12%) could not be
assigned a prognostic group. The percentage of "good-prognosis" RS was higher than observed in the initial retrospective series.
On the 102 "good-prognosis", 88 were included in the SA02 cohort to receive 6 FEC100 cycles, 14 pts were not included in the
cohort due to investigator or patients decision. The mean time from the date of surgery to the onset of chemotherapy was 5.4
+/-1.5 weeks. With a median follow-up of 50 months, the 2-year and 4-year MFS are 98% (95% CI 91-99) and 96% (95% CI
89-99). The 2-year and 4-year overall survival are 100% and 99% (95% CI 92-100).
Conclusion. This analysis confirms that genomic analyses are feasible in clinical practice. The MFS results with a median
follow-up of 50 months are within the expected hypothesis; follow-up is needed to confirm or not these results at 5 years. Final
follow-up data will be available in 2015.

Title: Cardiorespiratory fitness (VO2max) before, during and after adjuvant treatment in breast cancer patients
Hanne Frydenberg1, Tora J Bettum1, Trygve Lofterd1, Elisabeth Edvardsen2,3, Vidar G Flote1, Sissi E Finstad4, Gro F
Bertheussen5, Ellen Schlichting6, Anne McTiernan7 and Inger Thune1,8. 1Cancer Center, Oslo University Hospital, Oslo, Norway;
2
Norwegian School of Sport Sciences, Oslo, Norway; 3Oslo University Hospital, Oslo, Norway; 4Norwegian Directorate of Health,
Oslo, Norway; 5St Olav University Hospital of Trondheim, Trondheim, Norway; 6Cancer Center, Section for breast Surgery, Oslo
University Hospital, Oslo, Norway; 7Fred Hutchinson Cancer Research Center, Seattle, WA and 8Institute of Community Medicine,
Faculty of Health Sciences, University of Troms, Troms, Norway.
Body: Background: Breast cancer treatment may result in reduced exercise capacity that may in turn lead to reduced maximum
oxygen consumption (VO2max). However, whether physical exercise can counteract any observed decline in VO2max in breast
cancer patients undergoing adjuvant breast cancer treatment, is less known.
Material & methods: The women participating in the Norwegian Energy Balance and Breast Cancer Aspect (EBBA)-II pilot study,
were aged 35-75 years and diagnosed with stage I-II breast cancer. Performing a maximum exercise test on a treadmill (modified
Balke protocol), VO2max was assessed at four times; preoperative, 6, 12 and 24 months postoperative. The patients were
randomized postoperative to a control group (n=31) or an intervention group (n=29) stratified by menopausal status. The 12
months exercise intervention program consisted of group-based exercise, 60 minutes twice a week and a minimum of 60 minutes
of individual exercise. Regression models were used to study the associations between treatment regime and VO2max.
Results: Breast cancer patients (n=60) with a mean age at diagnosis of 55.3 years (38.0-69.0 years), had a mean body mass
index of 25.1 kg/m2, and a mean preoperative VO2max of 32.4 ml/min/kg. Comparing the intervention group to the control group,
the intervention group maintained VO2max throughout the treatment period, and improved their VO2max with 7.8 % from 12 to 24
months postoperative (p=0.117), while the control group had a 15% reduction in VO2max 6 months after surgery (p<0.001), which
improved 14 % at 12 months and additionally 6 % at 24 months postoperative (p=0.025). Among those patients receiving
chemotherapy (60%), and being in the control group, a decline in VO2max of 22.9 % (p<0.001) at 6 months postoperative was
observed. In comparison, patients in the intervention group who received chemotherapy had a 4.5 % reduction in VO2max at 6
months postoperative (p = 0.159). Thereafter, in the control group, VO2max improved with 21.6 % at 12 months postoperative
(p=0.006), while in the intervention group VO2max improved with 13.4 % 24 months postoperative (p=0.038). Patients in the
intervention group who did not receive any chemotherapy increased their VO2max by 6% 6 months postoperative (p=0.174), while
patients in the control group who did not receive any chemotherapy had a reduction in VO2max of 2.1 % at 6 month postoperative
(p=0.630).
Conclusion: Our findings suggest that systematic physical training may counteract a decline in VO2max in breast cancer patients
receiving adjuvant treatment, including chemotherapy, and is of clinical interest, but needs to be replicated in larger studies.

Title: No Show

Title: Influence of lifestyle factors and tumor cell dissemination in 632 early breast cancer patients
Bahriye Aktas1, Anna Frackenpohl1, Siegfried Hauch2, Johann Kraus3, Hans Armin Kestler3, Rainer Klaus Kimmig1 and Sabine
Kasimir-Bauer1. 1University Hospital, Essen, Germany; 2AdnaGen AG, Langenhagen, Germany and 3University Ulm, Ulm,
Germany.
Body: Introduction: Influence of lifestyle behaviour in risk of developing breast cancer is supported by several lines. Data from
632 early breast cancer (EBC) patients were collected to evaluate the influence of lifestyle factors in progression free survival
(PFS) and overall survival (OS). Results of disseminated tumor cell (DTC) in bone marrow and circulating tumor cells (CTC) in
blood were available as well. A complete pathological data set and medical history were obtained. It was the purpose of the
present study to correlate these data to compare the findings.
Methods: We evaluated 629 bone marrow samples and 606 blood samples from EBC patients treatet between 2004 to 2010 at
the time of first diagnosis. All samples underwent immunomagnetic enrichment using the AdnaTest BreastCancerSelect
(AdnaGen AG, Germany) within 4 hours after blood withdrawal followed by RNA isolation and subsequent gene expression
analysis by reverse transcription and Multiplex-PCR in separated tumor cells using the AdnaTest BreastCancerDetect. CTCs
were analyzed for the three breast cancer associated markers: GA733-2, Muc-1, Her-2 and actin as an internal PCR control. BM
aspirates were analyzed for DTCs by immunocytochemistry using the pan-cytokeratin antibody A45-B/B3. Lifestyle data icluding
menopausal status, BMI, usage of Metformin, hormone replacement drugs, beta blockers and Bisphosphonates were collected by
accessing the patient files. Histological data of the primary tumor were available for each patient.
Results: The overall detection rate for CTCs was 15.41% (88/571 patients) and for DTC was 38.5% (242/628 patients),
respectively. The mean BMI of 428 patients was 26.4 in 81/574 premenopausal, 426/574 postmenopausal and 67/574
perimenopausal patients. Medical history of smoking (128/394 patients), using hormone replacement therapy (92/354 patients),
alcohol consumption (68/378 patients), having allergies (188/419), using Metformin (22/389 patients), taking beta blockers
(83/392 patients) and Bisphosphonates (201/526 patients) were compared to PFS and OS as well as the histological data of the
primary tumor.
Conclusion: Lifestyle factors seems to influence the outcome in our cohort of EBC patients as shown in previous studies. Final
data and results regarding to tumor cell dissemination compared to lifestyle behaviour will be available for the SABCS 2014.

Title: Survival of metastatic hormone receptor (HR) positive/HER2 negative; HER2+; and triple negative (TN) breast cancer
based on initial presentation
Wendie-Lou D den Brok1, Caroline Speers1, Gondara Lovedeep1, Emily Baxter1, Scott Tyldesley1 and Lohrisch Caroline1. 1British
Columbia Cancer Agency (BCCA), Vancouver, BC, Canada.
Body: Introduction: Median overall survival (OS) for patients (pts) with metastatic breast cancer (MBC) is described as 2-3 years,
but few reports compare OS for de novo versus relapsed disease. The recognized clinical subtypes of breast cancer (HR
positive[+] /HER2 negative; HER2+; TN) are known to have variable OS clinically, but this has not been systematically
documented. We hypothesized that pts relapsing from a non MBC stage (relapsed) might have different OS than those presenting
initially with metastases (de novo), and that OS would differ for the three clinical biomarker subgroups.
Methods: Using the Breast Cancer Outcomes Unit database, we identified all women diagnosed with MBC, de novo or relapsed,
in British Columbia between 01/2001 and 12/2009 and referred to the BC Cancer Agency. Review of medical records confirmed
ER, PR and HER2 status. Survival from MBC diagnosis was calculated for relapsed vs de novo in the three biomarker subgroups.
Results: After excluding pts with a synchronous or prior contralateral disease, we identified 3645 women with known ER. Median
follow up was 91 months. HER2 known (n=3010) and unknown (n=635, 17%) cases had the same median OS (17 months [m])
and HR status (72% ER+). Trastuzumab (T) was standard in MBC during this era, but fewer than 10% had adjuvant T (introduced
mid 2005) which explains the high number of HER2+ MBC cases. We previously reported longer OS for HER2+ disease
relapsing after adjuvant therapy without T (older cohort) than after adjuvant T (SABCS 2013 Lohrisch). HER2 unknown pts were
excluded from further survival analyses. Three percent of cases with known PR were ER negative/PR +. Therefore ER was used
as the main determinant of HR status.
For the entire cohort and all biomarker groups, OS was longer for denovo than for relapsed MBC.
Overall Survival - relapsed vs de novo MBC for each biomarker subtype
Relapsed MBC
Subgroups
All cases with known ER
N dead/total
2124/2311
Median OS, m
15
De novo MBC
N dead/total
Median OS, m
593/699
26
P-value
<.0001
HR+/ HER2 neg

All
821/912
24
290/364
34
<.0001
PR known
414/477
20
251/319
34
<.0001
PR unknown
407/435
28
39/45
33
0.1315
110/126
15
0.0001
TNBC
All
368/391
9
HER2+
All
935/1008
11
193/209
17
0.0061
ER neg
268/278
58/60
9.5
0.2191
ER+
667/730
15
135/149
25
0.0088
OS was longest for denovo HR+/HER2 negative (34m) and shortest for HER2+/ER negative (7m) and TN (9m) relapsers. The
difference in OS for relapsed vs denovo HER2+ disease was significant for HR+ but not HR negative cases, likely due to small
numbers in the latter subset.
Conclusion: Relapsers experience shorter OS than their denovo biomarker counterparts, possibly due to the selective pressure of
adjuvant therapy on disease biology. When restricted to pts who received systemic therapy for MBC, OS figures may be higher.
Novel therapies may decrease the total number of relapsers, and improve OS in MBC for all, but are unlikely to narrow the OS
gap between relapsed and denovo groups. Trials exploring therapies for MBC of all biomarker types should therefore stratify by
stage at initial diagnosis.

Title: No Show

Title: Association of metabolic syndrome, its components and multigene assays for recurrence risk
Hanh P Mai1, Stephanie Kliethermes1, Shikha Jain1, Shelly S Lo1, Ellen R Gaynor1, Kathy S Albain1 and Patricia Robinson1.
1
Loyola University Medical Center, Maywood, IL.
Body: Background:
There is an association of metabolic syndrome (MS) and its constituents (obesity; diabetes mellitus, DM; hypertension, HTN;
hyperlipidemia, HL) with breast cancer (BC) causation and outcomes. Previously we showed an impact of obesity on tumor
biology as defined by a multigene assay. Our objective was to study the association between MS, its components and tumor
biology, as determined by the 70 gene signature (70-GS) and the 21-gene Recurrence Score (RS).
Methods:
Consecutive patients with newly diagnosed ER+, lymph node-negative BC from 2005-2012 were studied. A 70-GS was done for
those pts with tumors that had either low or intermediate RS. Pearsons Chi-square tests for univariate analyses and logistic
regression for multivariate analysis were used.
Results:
Low or intermediate RS were found on tumors from 151 pts of which 133 had a 70-GS. The MS was present in 23/104 (22%) pts
with intermediate RS and 21/46 (46%) pts with low RS (p=0.004). DM, HTN, and obesity were each inversely associated with
21-gene RS in univariate analyses (p=0.002, p=0.003, p=0.004 respectively, see Table). However, MS and its individual
components were not significantly associated with the 70-GS. Upon adjustment for age and race, the association between MS
and RS was not significant (OR=0.64; p=0.28); however, DM (OR=0.35; p=0.01), HTN (OR=0.44; p=0.05) and obesity (OR=0.35;
p=0.01) remained significantly inversely associated with RS. Independent of age and race, patients with DM, HTN, or obesity
were more likely to be in the low-risk 21-gene RS group.
[Table]
High Risk 70-GS
Low Risk 70-GS
Metabolic Syndrome
p-value
Intermediate 21-gene RS
Low 21-gene RS
0.30
0.004
Yes
14(24%)
24(32%)
23(22%)
21(46%)
No
44(76%)
50(68%)
81(78%)
25(54%)
Diabetes
0.24
0.002
Yes
12(21%)
22(30%)
18(17%)
19(41%)
No
46(79%)
52(70%)
86(83%)
27(59%)
Hypertension
0.28
0.003
Yes
36(62%)
39(53%)
52(50%)
35(76%)
No
22(38%)
35(47%)
52(50%)
11(24%)
Obesity
p-value
0.27
0.004
Yes
35(60%)
38(51%)
51(49%)
34(74%)
No
23(40%)
37(49%)
54(51%)
12(26%)
Conclusions and Implications:

The association of MS and its components differs for the 70-GS and RS assays. There is significant association between MS
(and DM, HTN, obesity) and RS, but no association with 70-GS. Patients with MS (and components) are more likely to have a
biology (low RS) known to have less chemotherapy benefit. Given the reported impact of MS on BC incidence and outcomes
along with the global increased MS incidence, the MS may have profound impact on BC biology and treatment outcomes in the
coming years. It is also possible that multigene assays currently in use for prognosis and prediction may need refinement in the
presence of MS and/or its components.

Title: Histological grade provides significant prognostic information in the discrimination between luminal A-like and luminal B-like
HER-2 normal subtypes of breast cancer according to St Gallen 2013
Anna Ehinger1,2, Per Malmstrm1,3, Pr-Ola Bendahl1, Christopher W Elston4, Anna-Karin Falck1,5, Carina Forsare1, Dorthe
Grabau1,6, Lisa Rydn1,7, Olle Stl8 and Mrten Fern1. 1Division of Oncology and Pathology, Lund Cancer Center at Medicon
Village, Lund University, Lund, Sweden; 2Blekinge County Hospital, Karlskrona, Blekinge, Sweden; 3Skne University Hospital,
Lund, Sweden; 4Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom; 5Helsingborg Hospital, Helsingborg,
Sweden; 6Skne University Hospital, Lund, Sweden; 7Skne University Hospital, Lund, Sweden and 8Linkping University, Faculty
of Health Sciences, Linkping, Sweden.
Body: Objective: According to St Gallen recommendations from 2013, estrogen receptor (ER), progesterone receptor (PR),
HER-2, and Ki-67 defines two subtypes of ER-positive and HER-2 normal breast cancer (BC): Luminal A-like and Luminal B-like.
Patients with Luminal B-like BC are often recommended chemotherapy in addition to endocrine therapy, whereas endocrine
therapy may be sufficient for patients with Luminal A-like BC. Histological grade (G) 1, 2 and 3 are not included in the St Gallen
recommendations. Our unpublished data from a series of 161 premenopausal N0 BC patients with long-term follow-up show that
the classification of Luminal A-like vs Luminal B-like HER-2 normal BC is strongly associated to G. Luminal A-like BC is often G1
or G2 and Luminal B-like BC is usually G2 or G3. We also found that the few G3 (n=6) Luminal A-like cases had a prognosis
more similar to Luminal B-like and that the few G1 (n=2) Luminal B-like HER-2 normal cases had a prognosis more similar to
Luminal A-like. The aim of this study is to evaluate in other cohorts if these findings can be confirmed.
Methods: G and St Gallen subtypes were evaluated in three BC cohorts from altogether 547 pre- and postmenopausal
chemotherapy nave T1-2N0-N1M0 patients. The endpoint was distant disease-free survival with 10 years of follow-up. We
compared the Luminal A-like and the Luminal B-like HER-2 normal subtype definition according to the original St Gallen
recommendation from 2013 based on ER, PR, HER-2 and Ki-67 (St Gallen 2013) with our proposal, where ER-positive, HER-2
normal, G1 BC is defined as Luminal A-like, independent of Ki-67 and PR, and ER-positive, G3 BC is defined as Luminal B-like,
independent of Ki-67 and PR (St Gallen 2013+G). The importance of Ki-67 and PR for subtyping was thus restricted to G2 BC.
Results: The hazard ratio (HR) between Luminal B-like HER-2 normal (n=185) and Luminal A-like (n=362) defined according to St
Gallen 2013+G, was 2.3 (95% confidence interval (CI) 1.6-3.4; p<0.0001) compared to 1.6 (95% CI: 1.1-2.4; p=0.025) according
to St Gallen 2013. Twenty-five patients, classified as Luminal B-like HER-2 normal with St Gallen 2013 were G1 and
consequently reclassified as Luminal A-like with St Gallen 2013+G. None of these twenty-five patients developed metastases
during the follow-up period. Thirty-eight patients showed the opposite pattern (G3 Luminal A-like with St Gallen but Luminal B-like
according to St Gallen 2013+G). Seventeen of these patients developed distant metastases already during the first five years.
Conclusion: Our findings strongly suggest that ER-positive, HER-2 normal, and G1 BC is a good prognosis group, independent of
Ki-67 and PR, and should be treated as Luminal A-like BC, whereas ER-positive, HER-2 normal, and G3 BC should be
considered as a worse prognosis group, independent of Ki-67 and PR, and should be treated as Luminal B-like BC. Based on our
findings the importance of Ki-67 and PR is restricted to G2 BC for the discrimination between Luminal A-like and Luminal B-like
HER-2 normal subtypes of BC.

Title: Impact of the immunohistochemical subtypes of breast cancer on prediction of axillary metastasis: Experience of one breast
center in Argentina
Maria F Calvo1, Carola Allemand1, Francisco H Corrao1, Roberto Orti1, Liliana B Zamora1, Maria C Riggi1, Maria F Ilzarbe1, Jorge
Piccolini1, Alejandra Wernicke1, Sebastian Gogorza1, Gustavo Izbizky1 and Claudio Lorusso1. 1Hospital Italiano, Buenos Aires,
Argentina.
Body: INTRODUCTION
Axillary node metastasis is one of the most important prognostic factors to be considered in the treatment of Breast Cancer.
Although the association between axillary metastasis and pathologic tumor size has been extensively studied, the correlation
between the immunohistochemical (IHC) subtype and axillary compromise has not. The aim of this study was to evaluate the
correlation between the immunohistochemical subtype of Breast Cancer (BC) and axillary extension. As secondary outcomes, we
assessed disease-free (DFS) and overall survival (OS).
MATRERIALS AND METHODS
1413 consecutive patients who underwent surgery for invasive primary breast cancer at the Hospital Italiano de Buenos Aires
between the years of 2007 and 2012 were included. Patients presenting with stage IV disease were excluded. We analyzed the
clinical and pathologic data of patients who were submitted to either sentinel node biopsy (SNB) or axillary lymph node dissection
(ALND). Based on IHC, tumors were classified into four groups: Luminal A (RE+ RP+ HER2-, KI 67 <14%), Luminal B (RE+ RP+,
HER+/ KI 67 >15%), HER 2 (RE- RP- HER+) and Triple Negative (TNBC) breast cancer (RE- RP- HER-).
RESULTS
We evaluated 1413 patients, among which 1248 patients were eligible for inclusion and analysis. In this population, 386 patients
(31%) had axillary metastasis. By considering the Luminal A subtype population as our control group, we found that axillary
metastasis was significantly increased in the Luminal B and HER2 positive subtypes (p<0.0001), but not in the TNBC subtype
(p=0.4468). When adjusted by tumor size and IHC, in tumors smaller than 2 cm (pT1), the Luminal B and TN subtypes
significantly increase the risk of node metastasis with an OR 2.73 (CI95% 1.73 - 4.31, P > 0.000), and OR 2.05 (CI95% 1.13
-3.70, P=0.017) respectively. In the case of HER2 positive tumors, the odds ratio for axillary extension was 6.62 (CI95% 3.02 14.50, P > 0.000). The median follow-up was 29 months (17- 44 months), and the overall survival estimated by Kaplan-Meier was
91% (CI95% 87-94), with a disease-free survival of 62% (CI95% 28-83).
DISCUSSION
In this cohort, immunohistochemical subtype was an important independent predictor of axillary metastasis. Tumors smaller than
2 cm, that overexpress HER2 in absence of estrogen and progesterone receptor, have up to six times greater incidence of axillary
extension than those belonging to the Luminal A subtype. Luminal B and Triple Negative cancers on the other hand appear to
present twice the risk when compared to the Luminal A subtype.

Title: Predictive characteristics for extensive nodal involvement in patients with axillary lymph node metastases
Nicole C Verheuvel1, Ingrid Van den Hoven1, Hendrik WA Ooms1, Vivianne CG Tjan-Heijnen2, Rudi MH Roumen1,2 and Adri C
Voogd2. 1Mxima Medical Center, Veldhoven, Netherlands and 2Maastricht University, School GROW, Maastricht, Netherlands.
Body: Background
Staging of axillary lymph nodes in invasive breast cancer is an important prognostic indicator. Various prediction models have
been developed to predict the risk of not having additional axillary metastases in patients with a positive sentinel node, thereby
disregarding patients with a positive ultrasound. However, it is important to identify all patients with extensive nodal involvement,
defined as 3 or more positive axillary lymph nodes, in whom an axillary lymph node dissection cannot be omitted.
Aim
This study aims to identify factors predicting extensive nodal involvement in the axilla, with the emphasis on the method of axillary
staging; ultrasound guided lymph node biopsy versus sentinel node procedure.
Methods
All patients diagnosed with invasive breast cancer in the period between January 2006 and December 2011 at the breast center
of the Mxima Medical Center were studied. Univariate and multivariate regression analyses were performed. Variables with a
p-value of 0.10 in univariate analyses were entered in the multivariate model where a p-value of 0.05 was considered
statistically significant.
Results
We included 307 cases, representing 306 node positive patients, of whom 178 cases had 1 or 2 positive lymph nodes and 129
cases had 3 or more positive lymph nodes. Multivariate analyses showed that factors as a positive axillary ultrasound (OR=4.513;
95%CI=2.30-8.86), palpability of axillary lymph nodes (OR=2.143; 95%CI=1.04-4.42) and lymphovascular invasion (OR=3.622;
95%CI=1.63-7.81) are significantly associated with extensive nodal involvement in patients with invasive breast cancer.
Conclusion
This study has identified clinically important factors predicting extensive nodal involvement in patients with a positive lymph node
biopsy by either a sentinel lymph node procedure or an ultrasound guided lymph node biopsy. Hence, the role of axillary staging
by ultrasound should be redefined since it might play an important role in selecting patients with extensive nodal involvement
who, in our opinion, may still benefit from axillary treatment.

Title: Clinical outcomes of breast cancer patients with intermediate oncotype DX recurrence scores (RS): A review of the
Cleveland Clinic Experience
Lindsey M Goodman1, Alberto J Montero1, Lisa Rybicki1, Karen Mrazeck1, Benjamin Calhoun1, Raymond Tubbs1 and Halle
Moore1. 1Cleveland Clinic Foundation, Cleveland, OH.
Body: Background:
The 21 gene RT-PCR assay (Oncotype Dx, Genomic Health, Inc, Redwood City, CA) performed on formalin-fixed
paraffin-embedded tissue quantifies the likelihood of distant recurrence in patients with estrogen receptor (ER)-positive, axillary
lymph node-negative breast cancer. The 21 gene recurrence score (RS) also quantifies the magnitude of clinical benefit of
adjuvant cytotoxic chemotherapy to endocrine therapy. Patients with a high risk RS clearly benefit from the addition of adjuvant
chemotherapy, while patients with a low RS do not. However, approximately one-third of patients have an intermediate RS
(18-31) and the incremental benefit from adjuvant chemotherapy is unclear. The primary objective of this study was to evaluate
the clinical outcomes of early stage breast cancer patients with intermediate RS.
Methods: We identified 262 patients with intermediate RS (18-31) in a prospectively maintained registry of patients with Oncotype
Dx testing at the Cleveland Clinic from 2004-2013. ER status, progesterone receptor (PR) status, HER2 gene amplification,
treatment with chemotherapy, follow-up for distant recurrences and other clinical and pathological variables were collected for all
patients in the registry.
Results:
Patient characteristics are listed in Table 1. One hundred patients (38.2 %) were treated with chemotherapy plus endocrine
therapy, while 156 (59.5%) were treated with endocrine therapy alone. Treatment data was unavailable for 6 patients (2%).
Distant recurrence status is unknown for 2 patients. The overall rate of distant plus local recurrence for all 262 patients was 3.8%
(3 local and 7 distant recurrences). There were 6 and 4 recurrences respectively, for patients received chemotherapy plus
endocrine therapy (6%), and in patients treated with endocrine therapy alone (2.5%), which was not significantly different (p=.21).
Table 1. Patient Characteristics
Median age at diagnosis (Range)
58 (27-88)
Median RS (Range)
22 (18-31)
ER+ (%)
262 (99.6)
PR+ (%)
234 (89)
HER-2 amplified (%)
10 (3.8)
Grade (%)
1
84 (32.8)
136 (53.1)
36 (14.1)
Histology (%)
Ductal
187 (71.1%)
Lobular
45 (17.1%)
Mixed
31 (11.8%)
Median Tumor size (range)
1.6 cm (0.2-6.5)
Lymph Node Status (%)

pN0
238 (90.8)
1-3 positive nodes
24 (9.2)
Conclusions:
With a median follow-up of 45 months, our data indicate that the realized distant recurrence rate for patients with an intermediate
range RS was less than 5%. These data do not exclude the possibility that some patients with an intermediate RS may derive a
small incremental benefit from the addition of adjuvant chemotherapy. Larger, randomized prospective trials like TAILORx should
provide additional guidance for the management of this patient population.

Title: Androgen receptor expression in pre-menopausal early breast cancer patients treated with endocrine therapy within the
ABCSG-12 trial - a single center pilot analysis
Gabriel Rinnerthaler1, Anna M Knopp2, Cornelia Hauser-Kronberger2, Simon P Gampenrieder1, Patrick Morre1, Brigitte
Mlineritsch1, Christian Fesl3, Michael Gnant4 and Richard Greil1. 1Salzburg Cancer Research Institute with Laboratory of
Immunological and Molecular Cancer Research and Center for Clinical Cancer and Immunology Trials, Paracelsus Medical
University, Salzburg, Austria; 2Paracelsus Medical University, Salzburg, Austria; 3Austrian Breast & Colorectal Cancer Study
Group, Vienna, Austria and 4Breast Health Center, Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.
Body: Background: Estrogen-receptor (ER) positive breast cancer co-expresses androgen receptor (AR) in 75% to 95% of
cases. The Austrian Breast & Colorectal Cancer Study Group (ABCSG) trial 12 compared anastrozole plus goserelin with
tamoxifen plus goserelin in premenopausal patients with hormone receptor positive early breast cancer. In addition, patients were
randomized to receive zoledronic acid or not. The aim of this analysis was to investigate the effect of AR expression on outcome
in dependency of the treatment arm.
Patients and methods: AR expression was analyzed by immunohistochemistry using a mouse monoclonal antibody
(Novocastra) in formalin-fixed paraffin-embedded specimens from 194 early breast cancer patients treated within the ABCSG 12
trial at our institution. As there is no generally accepted cut-off level defining AR positivity, we used two definitions for this
analysis: A) positive staining of 10% of tumor cells B) an immune reactive score (IRS; intensity of staining X percentage of
stained cells) 3 according to the Remmele score used for ER/PR evaluation.
Results: A total of 194 patients were included in this analysis. In tissue samples collected before 2001 AR staining was very
weak, interpreted as a loss of antigenicity due to archival time and conditions. Therefore, 40 patients with tissue samples
achieved earlier than 2001 were excluded from further analysis. Seventy-nine percent (122 of 154) and 63% (97 of 154) of tumors
were AR positive by definition A and B, respectively. All except one of the AR positive tumors according to definition B, were also
positive according to definition A.
Disease free survival (DFS) and overall survival (OS) data
N (%)
DFS
OS
Definition A ( 10%)
AR positive
122 (79)
AR negative
32 (21)
HR 0.61 (0.21-1.72) P=0.3428
HR 0.17 (0.03-0.99) P=0.025
HR 0.88 (0.33-2.35) P=0.7994
HR 0.39 (0.07-2.34) P=0.2846
Definition B (IRS 3)
AR positive
97 (63)
AR negative
57 (37)
DFS and OS did not differ between both endocrine treatment arms in a Cox regression model tested for interaction between AR
expression and endocrine treatment.
Conclusion: In this pilot study patients with AR positive disease had a numerically better DFS and OS compared to AR negative
patients, but only prolongation of OS in patients with 10% AR positive tumor cells was statistically significant. AR expression did
not influence outcome between tamoxifen and anastrozole treated patients, but based on the small number of events, this results
have to be interpreted with caution. This data will be confirmed in a larger proportion of patients treated within the ABCSG-12 trial.

Title: Neutrophil/lymphocyte ratio (NLR) can be one of the useful predictive prognostic markers for the metastatic breast cancer
(MBC)
Takeshi Miyamoto1, Tomomi Fujisawa1, Akiko Morishita1 and Yasuhiro Yanagita1. 1Gunma Prefectural Cancer Center, Ota,
Gunma, Japan.
Body: Introduction: The possibility of tumor infiltlating lymphocytes as a predictive marker of the neoadjuvant chemotherapy for
breast cancer was discussed at SABCS 2013. Cancer microenvironment formed by the immune and inflammatory cells is noticed
to be one of the factors for tumor growth, invasion or metastasis. To figure out the macro inflammatory environment as an
extension of the microenvironment, the neutrophil / lymphocyte ratio (NLR) is a useful method and a simple indicator of systemic
inflammatory state. We have some reports that NLR can predict the prognosis in gastric cancer and colorectal cancer. But in
breast cancer, few reports can be seen. We examined the relationship of NLR and the risk of recurrence in curable breast cancer
in our hospital, and could not detect the specific relations. We hypothesized one of the reasons that resectable breast cancers are
not in systemic phase in the view of inflammatory or immune reaction. Therefore, the next, in metastatic breast cancer (MBC),
absolutely systemic disease, we tried to reveal the relationship of the NLR at the recurrence and the prognosis.
Purpose: To evaluate the NLR affects the Overall Survival (OS) of the patients of MBC or not.
Patients: From 2003 to 2013, we have 300 MBC patients in our hospital included 53 Stage4 patients at the first visit. Median
Disease free survival (DFS) is 911-day, the median OS after the recurrence is 1196-day. Average value of the NLR is 2.85. The
reasons of MBC are bone metastases, pleural and pulmonary metastases, liver metastasis, lymph node metastasis, central
nervous system (CNS) metastasis, unresectable metastatic chest wall recurrence, or other.
Result: By univariate analysis, NLR 3.7(p<0.01), DFS >1000 days, (p<0.01), liver metastases, (p<0.05), CNS metastasis
(p<0.01), and 2 or more organs metastasis (p<0.05) made a contribution to poor OS. Between these 5 factors, special relations
were not seen. Other factors, stage4, organ metastasis except for liver and CNS, tumor subtype and the age at recurrence had
no significant effects for OS. Out of these 5, not only DFS >1000(p<0.01) or CNS metastasis (p<0.01) but also NLR 3.7 (p<0.05)
were the independent prognostic factors by using multivariate analysis.
Discussion: The NLR was one of the prognostic factors which we can easily and simply examine by blood sample. The reason of
this fact, we suggest, is that the whole body micro-environment caused by the immune or inflammatory cells at MBC occurred
contributes to tumor growth. For hormone receptor positive MBC without life threading organ metastasis, we select hormonal
therapy first, usually. But, high NLR MBC patient have a possibility of selection for up-front chemotherapy even if without life
threading metastasis. High Ki67-index as well as high NLR may be useful for the prognostic biomarker, however, make no
decision for the chemotherapy agents. First, we need to accumulate further retrospective cases and plan the prospective study to
make sure of the adequate treatment divided by NLR.
Conclusion: The NLR 3.7 is one of the independent predictive prognostic factors for MBC as well as DFS >1000 days and CNS
metastasis.

Title: Do traditional prognostic factors in early breast cancer still have a role in the molecular era?
Valentina Rossi1, Paola Berchialla2, Ivana Sarotto1, Furio Maggiorotto1, Nicoletta Tomasi Cont1, Riccardo Ponzone1, Massimo
Aglietta1 and Filippo Montemurro1. 1Candiolo Cancer Institute FPO, IRCCS, Italy and 2University of Turin, Italy.
Body: Background
The combined immunohistochemical (IHC) analysis of the oestrogen receptor (ER), progesterone receptor (PgR), human
epidermal growth factor receptor 2 (HER2) and proliferation index (Ki67) allows the approximation of the molecularly-defined
breast cancer subtypes. As this classification is now increasingly being used to determine the adjuvant therapy choice in early
breast cancer patients, the additional role of traditional prognostic factors in each distinct tumour subtype deserves further
investigation.
Methods
A total of 1,202 women undergoing surgery for Early Breast Cancer (EBC) were selected for this retrospective analysis. Breast
cancer subtypes were defined as luminal A, luminal B, HER2 luminal, HER2 enriched and Triple Negative [TN] by combining IHC.
Kaplan-Meier survival analysis and log rank tests were used to determine Event Free Survival (EFS). Cox proportional hazard
models were used to estimate hazard ratios, adjusted for potential confounders.
Results
A total of 345 tumours (29%) were luminal A, 558 (46%) luminal B, 107 (9%) HER2 luminal, 76 (6%) HER2 enriched and 116
(10%) TN. At a median follow-up of 58 months (4-137 months) relapse events occurred in 38 (11%) patients with luminal A, 102
(18%) with luminal B, 33 (31%) with HER2 luminal, 32 (42%) with HER2 enriched and 40 (35%) with TN tumours. Cox
proportional hazard multivariable analysis identified the following independent associations with EFS in each subtype: 4 positive
axillary nodes (HR 4.96, 95% CI 1.97-12.48, p<0.001), tumour grading=3 (HR 2.97, 95% CI 1.00-8.79, p<0.049), adjuvant
radiotherapy (HR 0.46, 95% CI 0.22-0.95, p<0.035) in luminal A tumours; 4 positive axillary nodes (HR 3.97, 95% CI 2.34-6.74,
p<0.001), PgR<20% (HR 1.55, 95% CI 1.04-2.29, p<0.03), age at first diagnosis >35 years (HR 0.22, 95% CI 0.10-0.48,
p<0.001), adjuvant hormonotherapy (HR 0.43, 95% CI 0.22-0.85, p<0.01), adjuvant radiotherapy (HR 0.57, 95% CI 0.36-0.91,
p<0.02) in luminal B tumours; 4 positive axillary nodes (HR 5.22, 95% CI 1.91-14.28, p<0.001) and adjuvant trastuzumab (HR
0.41, 95% CI 0.17-0.95, p<0.039) in HER2 enriched, 4 positive axillary nodes in HER2 luminal (HR 4.78, 95% CI 1.84-12.39,
p<0.001) and TN (HR 4.12, 95% CI 1.96-8.65, p<0.001) tumours.
Conclusion
Conventional risk factors retain their independent value in a subtype-specific fashion in early breast cancer. Although growing
importance is being given to stratification based on biological characteristics, the integration of traditional factors facilitates a
better definition of risk categories in early breast cancer patients.

Title: Clinical predictors of long-term survival in Her2-positive (HER2+) metastatic breast cancer (MBC)
Pooja Murthy1, Kelley M Kidwell1, Dafydd G Thomas1, Jennifer J Griggs1, Sofia D Merajver1, Anne F Schott1, Jeffrey B Smerage1,
Catherine H Van Poznak1, Max Wicha1, Daniel F Hayes1 and N Lynn Henry1. 1University of Michigan Medical School, Ann Arbor,
MI.
Body: Introduction: Clinical observation suggests that a subset of patients with HER2+ MBC survive for prolonged periods when
treated with Her2-targeting regimens. We hypothesized that we could identify clinical and pathological factors associated with
prolonged survival.
Methods: An IRB approved, retrospective, single institution review of patients diagnosed with HER2+ MBC was performed.
Patients treated with pertuzumab or ado-trastuzumab emtansine were not included because recent FDA approval did not permit
long-term follow-up. Clinical and pathologic characteristics were abstracted from the medical record. Kaplan Meier curves were
constructed to evaluate time to progression after first metastasis, and overall survival from time of first metastasis. Cox
proportional hazards analysis was used to assess for factors associated with long-term survival. A p value of <0.05 is statistically
significant.
Results: Review identified 181 patients with HER2+ MBC. Median age was 47 (range 35-80). More than half (N=107) had
hormone receptor positive disease; 21% had received adjuvant trastuzumab; and 25% had stage IV disease at diagnosis. Median
overall survival from the time of MBC diagnosis was 4.2 yrs (range 0.1-15.5). Since the diagnosis of MBC, 70 (38%) survived for 5
or more yrs and 15 (8%) survived more than 10 yrs. One third (N=59) of patients had brain metastases. These patients had a
median survival of 1.5 yrs (range 0-12.5 yrs) with 14% (N=8) living for more than 5 yrs following diagnosis of brain metastasis.
Factors associated with decreased survival are listed in Table 1.
Factors Associated with Decreased Survival in Her2+ MBC
Factor
Hazard Ratio (95% C.I.)
P Value
Age (continuous)
1.03 (1.01-1.04)
0.0004
Time to recurrence (continuous)
0.99 (0.94-1.04)
0.74
Hormone receptor positive
0.70 (0.49-0.99)
0.042
Brain metastasis
1.09 (0.77-1.56)
0.63
Adjuvant anti-Her2 therapy
0.98 (0.61-1.58)
0.94
Multiple (vs single) sites of disease at initial diagnosis of MBC
1.66 (1.17-2.35)
0.005
Conclusions: In the treatment era of trastuzumab and lapatinib, 8% of patients within this cohort with HER2+ MBC lived more
than 10 yrs. Analysis of current standard clinical and pathologic characteristics are not predictive of survival duration. Identifying
factors associated with prolonged survival may provide insights for individualizing treatment selection.


Title: Predicting the likelihood of additional non sentinel lymph node metastasis in early breast cancer: Novel sentinel nodal
station status versus Singapore General Hospital nomogram
Sue Zann Lim1, Puay Hoon Tan2, Gay Hui Ho3, Preetha Madhukumar3, Yirong Sim1, Shaun Shi Yan Tan1, Cindy Lim4, Veronique
Kiak Mien Tan3 and Kong Wee Ong3. 1Singapore General Hospital, Singapore; 2Singapore General Hospital, Singapore; 3National
Cancer Centre Singapore, Singapore and 4National Cancer Centre, Singapore.
Body: Background: Sentinel lymph node biopsy (SLNB) has been widely used in early breast cancer patients for the detection of
axillary nodal metastasis. We were the first to describe 2 novel sentinel nodal stations (SNS) in relation to the intercostobrachial
nerve (ICB) and the medial pectoral neurovascular bundle (MP) at which sentinel lymph nodes (SLN) were consistently identified,
even only with the use of blue dye. In a pilot study involving 176 cases, we have shown that the ICB and MP SNS represent
sequential echelons of SLN draining the breast. It was observed that the status of the MP SNS can be used in predicting the
likelihood of additional non sentinel lymph node metastasis in early breast cancer. Thus, we aim to compare this against the
Singapore General Hospital (SGH) nomogram, the existing standard predictive model in the local population. The SGH
nomogram was developed from predictors in the Memorial Sloan-Kettering Cancer Centre (MSKCC) nomogram. It uses only 3
pathological parameters: lymphovascular invasion, number of positive and negative SLN. This has been shown to be at least
equal if not better than the MSKCC nomogram as a predictive model in the Singapore population.
Methods : All patients who underwent oncologic breast surgery and SLNB (using the SNS identification technique) at the
Department of Surgical Oncology, National Cancer Centre Singapore from February 2012 to December 2013 inclusive were
reviewed. Patients who fulfilled the following selection criteria were included in the study: [1] invasive ductal or lobular carcinoma,
[2] SLN identified in both ICB and MP SNS,[3] axillary clearance done with total lymph nodes 10, based on a positive SLNB.
The performance of the MP SNS status and SGH nomogram in predicting the likelihood of additional non sentinel lymph node
metastasis was compared with the calculation of the area under the receiver-operating characteristic curve (AUC).
Results: A total of 49 patients were identified. Majority of the patients had early breast cancers: 94% had tumour size 5cm and
71% had N1 disease. The median number of total SLN, ICB and MP nodes identified were 3 (range 2-14), 2 (range 1-7) and 1
(range 1-12) respectively. The median number of positive and negative SLN were both 1 (range 1-5 and 0-9 respectively). The
positive predictive value of MP SNS status for additional non sentinel lymph node metastasis was 76.5% (95% CI: 50.1-93.2).
The strong association was proven by an odds ratio of 7.15 (95% CI 1.86-27.50, p-value: 0.002). The negative predictive value of
MP SNS for eventual N stage was 93.8% (95% CI: 79.2-99.2). In most of the cases, the nodal stage remained at N1 in the
presence of negative MP node. The model with MP SNS status yielded an AUC of 0.706 (95% CI: 0.579-0.832) which was higher
than that of the SGH nomogram, 0.658 (95% CI: 0.503-0.813).
Conclusions: The novel MP SNS proved to be a single parameter which predicts the likelihood of additional non sentinel lymph
nodes metastasis better than the SGH nomogram. More importantly, from the clinical point of view, the MP SNS status can be
made available intra-operatively and hence guide the decision for further axillary dissection.

Title: Her2 FISH amplification in ER/PR/Her2 IHC negative breast cancer
Sunati Sahoo1 and Helena Hwang1. 1University of Texas Southwestern, Dallas, TX.
Body: Background:
Triple negative (ER negative, PR negative, Her2 negative) breast cancer is an aggressive cancer that is not likely to respond to
endocrine or anti-Her2 therapy. While Her2 positivity is a poor prognostic factor, patients with Her2 positive tumors are eligible for
anti-Her2 therapy. Her2 positivity is based on either immunohistochemistry (IHC) or fluorescence in situ hybridization (FISH)
result. Her2 IHC scores of 0+ and 1+ are negative, 2+ is equivocal, and 3+ is positive. The Her2 gene is amplified when the
Her2/CEP17 ratio is 2.0. While most tumors will show IHC/FISH concordance, some tumors may show discordant findings
between IHC staining and FISH. Previous studies have shown benefit from anti-Her2 therapy in tumors with discordant IHC/FISH
results. The purpose of this study was to determine the frequency of Her2 amplification in triple IHC negative breast cancer to
determine whether routine Her2 FISH testing is necessary in this subset of patients.
Methods:
After Institutional Review Board approval, the pathology databases were searched for cases of triple IHC negative breast cancer
from 2003-August 2013. All cases of triple IHC negative with corresponding Her2 FISH testing were included in this study. The
results of Her2 FISH were correlated with Her2 IHC. The clinical history and IHC slides for all discordant Her2 cases were
reviewed when available.
Results:
A total of 659 triple IHC negative breast cancer cases from 516 patients were found. The patients age at diagnosis ranged from
26-92. The cases were classified using the ratio of Her2/CEP17 per the 2013ASCO/CAP guidelines: non-amplified < 2.0,
amplified 2.0. As the copy number for Her2 was not always available, the 2013 ASCO/CAP criterion for amplification based on
the copy number was not used in reclassifying the cases. Based on the ratio, 20 tumors were amplified, 631 tumors were not
amplified, and 8 tumors had insufficient tissue for Her2 FISH testing. The Her2/CEP17 ratio for amplified cases ranged from 2.0 to
11.1 and for non-amplified cases from 0.5 to 1.9. Of these 20 patients, 10 received some trastuzumab treatment, 6 did not receive
any trastuzumab, and in 4 patients, there was no additional information available regarding treatment with trastuzumab. Of the 10
who received trastuzumab treatment, 2 died of disease (DOD), 26 months and 45 months after their diagnosis. The overall
survival of the other patients in this group was 23-106 months. Of the 6 who did not receive trastuzumab, one DOD after 76
months. The other 5 patients have overall survival ranging from 15-93 months. Of the patients with unknown Her2 targeted
therapy status, one is deceased, the status of 2 is unknown, and one patient is alive at 110 months.
Discussion:
Our study showed 3% of triple IHC negative breast cancers to be Her2 FISH amplified. Of the 20 tumors, 19 showed Her2/CEP17
between 2.0 to 4.2 with one case showing Her2/CEP17 of 11.1. Firm conclusions on the efficacy of trastuzumab treatment in this
small subset of patients with Her2 amplification but no protein overexpression cannot be drawn. However, the overall survival of
those treated and not treated is similar. Based on the small number of cases with Her2 IHC/FISH discordance, a compelling case
for routine testing of all Her2 IHC negative tumors cannot be made.

Title: TIMP-4 is a prognostic and predictive marker in triple-negative breast cancers
U Margaretha Wallon1, Jennifer L Sabol2, Vlasta Zemba-Palko2, James S DuHadaway1, Erica Sutanto-Ward1, Zonera A Ali2, Paul
B Gilman2, Robin M Ciocca2, Ned Z Carp2 and George C Prendergast1. 1Lankenau Institute for Medical Research, Wynnewood,
PA and 2Lankenau Medical Center, Wynnewood, PA.
Body: BACKGROUND Tissue inhibitor of metalloproteinase-4 (TIMP-4) is a secreted multi-functional protein associated with
poor survival prognosis among early-stage triple-negative breast cancers (TNBC). TNBC represent a highly aggressive form of
this disease with an unmet need for effective predictive markers and targeted therapy.
Extracellular TIMP-4 binds to the membrane bound tetraspanin CD63 and induces the activation of the tumor promoting
PI3K/AKT/mTOR pathway.
Here we report that TIMP-4 induced aggressive tumor growth and metastasis can be adverted by directly targeting TIMP-4 using
a newly developed monoclonal antibody (mAb) to sequester TIMP-4 and the varied responses to common chemotherapy (CTX)
regimen.
METHODS The role of elevated TIMP-4 in TNBC cell behavior was tested in cell culture and animal experiments using the
human breast cancer line MDA-MB-468. Cells with or without TIMP-4 added to the medium were used to determine the effects on
growth, clonogenic survival and response to chemotherapeutic agents such as adriamycin, Taxol, and the new TIMP-4 mAb. The
same cell-line was used to induce tumor growth in nude mice with or without TIMP-4 containing slow-release pellets implanted
into the mammary fatpad (mfp). Tumor growth and response to therapy was followed over a six-week period. Lungs, liver, spleen
and mfp were collected and analyzed for presence of human cells using a specific anti-human MHC I mAb.
Prospectively collected patient samples, in accordance with the IRB approved protocol, were tested for circulating levels of
TIMP-4 using a commercially available ELISA assay in samples collected prior to chemotherapy and at each treatment cycle. The
medical oncology staff recommended therapy without knowledge of TIMP-4 status.
RESULTS Augmentation of TIMP-4 levels in cell culture medium or the mfp of mice resulted in similar tumor phenotype as in
the clinic; fast growing tumors with accelerated disease progression.
Elevated TIMP-4 levels in the tumor environment resulted in a 1.5-fold increased growth rate with liver and/or lung metastasis in
25% of animals (N=16). No metastases were found in animals with normal TIMP-4 levels. Treating cell cultures or tumor-bearing
mice (i.p. injections) with our TIMP-4 mAb resulted in decelerated growth rate and no detectable metastatic disease in the
animals.
Results from patient samples demonstrated that circulating TIMP-4 levels in breast cancer patients remain elevated after
definitive surgery, indicating that TIMP-4 might continuously stimulate any remaining disseminated tumor cells. Adriamycin
containing regiments was the only CTX to suppress the TIMP-4 levels independent of primary tumor size and nodal status.
CONCLUSIONS Based on these clinical and experimental data we suggest that TIMP-4 may represent a prognostic and
predictive marker, and a therapeutic target for TNBC patients at highest risk. The presence of TIMP-4 identifies a patient
population likely to recur quickly due to continuous activation of the PI3K/AKT/mTOR pathway. Though adriamycin therapy can
reduce the TIMP-4 levels, the toxicity of this agent suggests that targeted therapy of the PI3K/AKT pathway and/or a biological
therapeutic approach directed against TIMP-4 may be of benefit in this subset of pts and should be further explored.

Title: High triple-negative breast cancer prevalence and poor outcome of hormone receptor positive breast cancer among young
Mexican women
Cynthia Villarreal-Garza1,2, Nancy Reynoso-Noveron1, Claudia Arce-Salinas1, Fernando Lara-Medina1, Enrique Bargallo1, Leticia
Mendoza-Galindo1, Federico Lasa1, Maria Teresa Ramirez1 and Alejandro Mohar1. 1Instituto Nacional de Cancerologia, Mexico,
DF, Mexico and 2Tecnologico de Monterrey, Monterrey, NL, Mexico.
Body: Background: Breast cancer (BC) among young women is an emerging public health issue in Mexico, as the proportion of
incident cases and deaths has been reported to be greater in women aged <40 years compared to high-income regions.
However, there is scarce information in Hispanic populations regarding the presentation and outcome of this subgroup of patients.
The purpose of our study was to compare the distribution of BC subtypes between age subgroups and to determine the
prognostic significance of young age stratified by molecular subtype based on ER, PR, and HER2 status.
Methods: This study included all consecutive patients with BC diagnosed at the National Cancer Institute in Mexico in the year
2007. A panel of ER, PR, and HER2 was used as a means of classifying BC into three molecular subtypes: (a) Hormone-receptor
(HR) positive and HER2-negative, (b) Triple-negative, and (c) HER2-positive. The Chi-square test was used to compare the
distribution of baseline characteristics among groups according to age (40 years vs. >40 years). Cox proportional hazard
analysis was applied to assess the association of clinical and pathological factors with recurrence free survival with follow-up
through 2013.
Results: Of the 696 analyzed patients, 134 (19.3%) were 40 years. For the whole group, locally advanced BC accounted for the
majority of the cases (61%), and there was no difference regarding clinical stage of presentation between age subgroups. The BC
subtype distribution differed among age subgroups (HR-pos/HER2-neg: 42% vs. 52%, triple-negative: 30% vs. 21%, HER2-pos:
24% vs. 23%, for young vs. non-young respectively). After a median follow-up of 74.6 months, 48% of young patients
experienced a recurrence compared to 39% of women >40 years. After subtype and clinical stage stratification, young age
remained a significant independent predictor of recurrence in patients with HR-pos/HER2-neg tumors (hazard ratio 1.71; 95%
CI:1.09-2.69; P = 0.019), but was not found to be a predictor of recurrence for patients with HER2-pos and triple negative
subtypes.
Conclusions: The results of our present study suggest that the poor survival associated with young age is not only driven by a
high proportion of triple-negative BC, but also by poor prognosis in the most prevalent HR-pos/HER2-neg subgroups. Possible
explanations for this finding are a greater proportion of Luminal B BC subtype and/or tamoxifen resistance in the young age
group. Further characterization of HR-pos BC should be pursued and aggressive therapeutic strategies must be considered for
treatment of young women with high-risk features.
It is critical to note the extremely high prevalence of BC among young women in our Institution, which comprised 19% of the total
population. To address this unmet, growing burden, we adopt the model of a specialized program for the care of young women
with BC, the first in Latin America, at the National Cancer Institute in Mexico. The goals of our program include optimizing
complex clinical care and supporting needs for young women and their families; promoting medical, biomedical, and social
research; and educating women and health professionals to promote early cancer detection and improved multidisciplinary
management.

Title: Recent weight gain and increased breast cancer risk varies by receptor classification among pre and postmenopausal
women
Graham A Colditz1, Heather Eliassen2, Adetunji T Toriola1, Susan E Hankinson3, Walter C Willett4, Loki Natarajan5 and Bernard
Rosner2. 1Washington University School of Medicine, St Louis, MO; 2Harvard Medical School, Boston, MA; 3University of
Massachusetts, Amherst, MA; 4Harvard School of Public Health, Boston, MA and 5Mores UCSD, Cancer Center, La Jolla, CA.
Body: Obesity is well established as a cause of postmenopausal breast cancer incidence and mortality. Furthermore, adiposity in
early life through premenopausal years reduces breast cancer incidence. However, studies using measures of adiposity at age
18 also report inverse relations with premenopausal breast cancer and for some but not all subtypes of breast cancer defined by
molecular status. To assess the relations of adiposity in childhood, in late adolescence, and in adult years as well as change in
weight in relation to total invasive breast cancer and subtype defined by receptor status we fit models based on the
Rosner-Colditz log-incidence model of breast cancer.
The Nurses Health Study cohort was established in 1976 when 121,701 female US registered nurses ages 30-55 responded to a
mail questionnaire inquiring about risk factors for breast cancer including reproductive factors, hormone use, anthropometric
variables, benign breast disease (BBD), and family history of breast cancer. The risk factors have been updated by repeat
questionnaires every 2 years. We followed a cohort of 77,232 women from 1980 to 2006 (1,445,578 person-years) documenting
4,196 incident cases of invasive breast cancer. ER and PR status were obtained from pathology reports and medical records. A
total of 2,033 ER+/PR+ tumors, 595 ER-/PR- tumors, 512 ER+/PR- tumors were identified among women with complete
information on breast cancer risk factors.
Overall, weight at age 18 was inversely related to incidence. The relative risk per 25 lb weight difference was 0.89 (0.85, 0.93).
After controlling for weight at age 18, long-term weight change was positively related to total incident breast cancer risk among
both premenopausal (RR per 25 lb weight gain since age 18=1.08; 95% CI=1.05-1.12) and postmenopausal women (RR per 25
lb weight gain since menopause=1.16; 95% CI=1.09-1.23).
In addition, we focus on the effect of short-term weight gain (over past 4 years) on breast cancer risk while controlling for weight
at age 18 and long-term change in weight during premenopause and postmenopause. We found a significant effect of short-term
weight change and breast cancer risk (RR=1.20; 95% CI=1.09-1.33) for a 4-year weight gain of 15 lbs vs no change ( 5lbs)
(RR per 25 lb weight gain=1.13; 95% CI=1.06-1.21, p<0.001). However, the effect was stronger for premenopausal women ( 15
lb weight gain vs no change, RR=1.38; 95% CI=1.13-1.69) (RR per 25 lb weight gain=1.26; 95% CI=1.08-1.48, p=0.004) than for
postmenopausal women ( 15 lb weight gain vs no change, RR=1.10; 95% CI=0.97-1.25) (RR per 25 lb weight gain=1.08; 95%
CI=1.00-1.16, p=0.063).
The effect of short-term weight gain during premenopause was stronger for ER+/PR- (RR per 25 lb weight gain=2.19; 95%
CI=1.33-3.61, p=0.002) and ER-/PR- breast cancer (RR per 25 lb weight gain=1.61; 95% CI=1.09-2.38, p=0.016) than for
ER+/PR+ breast cancer (RR per 25 lb weight gain=1.13; 95% CI=0.89-1.43, p=0.32).
In conclusion, adiposity in childhood has a protective lifelong relation to breast cancer risk, though mechanisms remain largely
unexplained. There are long-term deleterious effects of weight change both pre- and post-menopause and deleterious effects of
short-term weight gain during premenopausal years.

Title: Vitamin D levels, triple-negative breast cancer, and geography: A retrospective analysis of a large database of oncology
practices in the United-States (US)
Joyce G Habib1, Janet Espirito2, Robyn Harrell3, Brian Turnwald4, Joyce O'Shaughnessy5, Ervin Epstein6 and Debra Patt7.
1
Sammons Cancer Center, Texas Oncology, Dallas, TX; 2McKesson Specialty Health, Woodlands; 3McKesson Specialty Health,
Woodlands; 4McKesson Specialty Health; 5Sammons Cancer Center, Texas Oncology, The US Oncology Network, Dallas, TX;
6
Children's Hospital Oakland Research Institute, Oakland and 7Texas Oncology, The US Oncology Network, McKesson Specialty
Health, Austin, TX.
Body: BACKGROUND: The UVB-vitamin D-breast cancer (BC) hypothesis is supported by ecological studies demonstrating an
inverse correlation between sunlight exposure and BC incidence and mortality. Collective data from observational studies also
favor an inverse association between vitamin D status and BC risk, recurrence and mortality. Yet, inconsistencies exist regarding
the role of vitamin D in BC biology and its recently described association with triple-negative breast cancer (TNBC). Similarly, the
validity of the vitamin D-latitude paradigm has also been disputed. In this study, we aimed to determine the existing associations
between serum vitamin D levels, geographic distribution, clinical and pathological characteristics of BC patients (pts) across the
US.
METHODS: This was a retrospective analysis from the electronic health record database of pts diagnosed with BC between
1/2007-5/2013 across US Oncology Network practices. Practices were stratified based on their geographic location into one of
three categories: northern (>40 N), central (35-40N) and southern (<35N) latitude. For each pt, we collected the following:
age at diagnosis, BMI, smoking history, clinical stage, estrogen receptor (ER), progesterone receptor (PR), HER2 status, and the
first documented serum 25-hydroxyvitamin D (25-(OH)D) level, categorized as <20 deficient, 20-30 suboptimal and >30 optimal.
Statistical comparison was performed using Chi-squared tests for categorical variables and Kruskal-Wallis tests for continuous
variables. Logistic regression was used to predict the likelihood of vitamin D deficiency.
RESULTS: 20,338 BC pts with a documented vitamin D level were identified. Mean age at diagnosis was 58. Using age 60 as a
surrogate for menopausal status, pre- and post-menopausal pts were equally distributed. Stage and hormone receptor status
distribution were: 8%, 41%, 32%, 11% and 4% for stage 0, I, II,III and IV, respectively; 63%, 13% and 10% for ER+/HER2 -,
HER-2+ and TNBC, respectively. 17.6%, 27.8% and 54.5% of pts had deficient, insufficient and optimal vitamin D levels,
respectively. The covariates of age < 60 years (OR 1.24), advanced stage (OR 1.32 for stage II, OR 1.51 for stage III, and OR
1.81 for stage IV), TNBC (OR 1.45), BMI 25 (OR 2.02), current smoker (OR 2.11) and lower latitudes (OR 1.36 and 1.19 for
central and southern latitude, respectively) were independent predictors of first documented vitamin D deficiency in a multivariate
model.
CONCLUSION: Vitamin D deficiency may be associated with TNBC and central and southern latitudes. The influences of
differential vitamin D supplementation and potential variations in timing of testing warrant further investigation.

Title: Development of a cardiac toxicity prediction tool for HER2 (+) breast cancer patients receiving trastuzumab
Jeffrey A Sulpher1, George Dranitsaris2, Freya Crawley1, Franco Dattilo1, Maya Kovacs1, Christopher Johnson3 and Susan Dent1.
1
Ottawa Hospital Cancer Centre, Ottawa, ON, Canada; 2University of Ottawa, Ottawa, ON, Canada and 3Ottawa Hospital, Ottawa,
ON, Canada.
Body: Background: In HER2 positive breast cancer, the addition of trastuzumab prolongs overall survival in both early stage and
metastatic disease. However, moderate to severe cardiac toxicity is a potentially serious complication which can lead to dose
reductions, delays, hospitalizations, and premature discontinuation of treatment. Patient care could be substantially improved if
such cardiac events are accurately predicted through the use of validated and easy-to-use mathematical models. In this study,
the development of a model to predict the risk of cardiac toxicity prior to initiation of trastuzumab therapy is described.
Methods: Medical records of 498 HER2 positive breast cancer patients who received trastuzumab at the Ottawa Hospital Cancer
Centre were identified. Charts were reviewed for potential cardiac toxicity risk factors and cardiac events. Potential cardiac toxicity
variables included: cardiac risk factors, medications, previous chemotherapy/radiation, baseline left ventricular ejection fraction
(LVEF), and exposure to anthracyclines. Cardiac toxicity was defined as: decreased LVEF greater or equal to 10% and to less
than 50%, referral to the cardiac oncology service, or clinical symptoms of heart failure. General linear modeling for a discrete
bivariate outcome was used to identify risk factors for cardiac toxicity using a backwards elimination process. Internal validation of
the final regression coefficients was done using nonparametric bootstrapping. A risk scoring algorithm (range 0-100) was then
derived from the final model coefficients. A receiver operating characteristic (ROC) curve analysis was then undertaken to
measure the predictive accuracy of the final scoring algorithm.
Results: Baseline LVEF, concomitant use of any cardiac medication or lipid lower drugs and doxorubicin based chemotherapy
were identified as being important predictors for cardiac toxicity. The ROC curve analysis indicated good predictive accuracy with
an area under the curve of 0.69 (95%CI: 0.64 to 0.74). Prior to the initiation of trastuzumab, patients with risk scores greater than
45 units would be considered at high risk for developing cardiac toxicity (likelihood ratio = 2.9).
Conclusions: Risk of cardiac toxicity with trastuzumab is increased in patients with a low baseline LVEF, history of coronary artery
disease and hyperlipidemia, as well as those receiving doxorubicin. The planned external validation and eventual clinical
application of this prediction tool will be an important source of risk information for the practicing oncologist, and may enhance
patient care by optimizing preventative therapies and selecting high risk patients for cardiac imaging and cardiology follow-up.

Title: Volumetric breast density improves breast cancer risk prediction
Jennifer A Harvey1, George J Stukenborg1, Wendy F Cohn1, Kathy L Repich1, Olivier Alonzo2, Wendy M Novicoff1, Martin D Yaffe2
and William A Knaus3. 1University of Virginia, Charlottesville, VA; 2Sunnybrook Health Sciences Centre, Toronto, ON, Canada and
3
Northshore University Research Institute, Evanston, IL.
Body: Introduction: There is increasing interest in implementing personalized breast cancer screening strategies rather than
relying on population based guidelines. Most risk models do not include breast density and two models that do rely on subjective
BI-RADS categories; all have limited discriminatory ability (C-statistics ranging from 0.60-0.74). Our aim was to develop a model
that includes an automated objective and numeric volumetric measurement of breast density with other known risk factors to
improve risk prediction.
Methods: A case-control study design was used to evaluate the association between risk factors and breast cancer diagnosis. All
women diagnosed with breast cancer during 2003-13 with a digital contralateral mammogram at the University of Virginia at the
time of diagnosis were eligible as cases. All women without a breast cancer diagnosis but with a digital mammogram at UVA
during 2003-2008 were eligible as controls. Risk factor information was collected using a self-reported electronic questionnaire.
Mean automated volumetric breast density (Volpara, NZ) was calculated for each patient as a percentage. Controls were
matched to cases in a 2:1 ratio based on age group, race, and education, using the GREEDY algorithm. Case-control selections
were made using the weighted sum of the absolute differences between the case and control matching factors. Conditional
logistic regression using the partial likelihood function from Cox proportional hazards regression was used to fit risk prediction
equations to the matched case-control study dataset, with stratification for each case matched set. A full model was estimated
including all available covariates for use as a model performance reference standard. A reduced model was then estimated
including covariates in the full model that had a Wald Chi-Square/degrees of freedom ratio > 1.0. The performance of the full and
reduced models was measured using the C index and the maximum R-Square statistic.
Results: The study enrolled 3,445 women; 839 cases and 2,606 controls. Multivariable analysis was conducted using 825 cases
and 1,628 controls with 1 or more breast studies reported for the surveyed population. The matching process yielded balanced
matching factor values between cases and controls, with no significant differences in age group (p = 0.95), race (p = 0.13), or
education (p = 0.86). The full prediction model (with 97 df) yielded a C index of 0.88, and an R-Square of 0.53. The reduced
model (with 15 df) had a C index of 0.83 and an R-Square of 0.54. Variables in the reduced model included: mean breast density;
biopsy showing ADH, ALH/LCIS; BMI; use of HRT, contraceptives, NSAIDS; smoking; exercise; parity; diabetes; family history of
breast cancer, HBOC, Li-Fraumeni, or Cowden Syndromes and/or BRCA mutation. Mean volumetric breast density was a leading
independent predictor of case status in both the full (p<0.0001) and reduced models (p=0.0212).
Conclusion: Volumetric breast density with other known risk factors may provide more accurate individual risk assessment,
enabling clinicians to develop patient centered risk based screening protocols that better inform decision making while including
patient preferences. The next steps require independent validation of the risk model in other populations.

Title: Value of adding single-nucleotide polymorphism panel markers to phenotypic algorithms of breast cancer risk
Gillian Dite1, Richard Allman2 and John L Hopper1. 1Centre for Epidemiology and Biostatistics Melbourne School of Population
and Global Health, University of Melbourne, Melbourne, VIC, Australia; 2Genetic Technologies Ltd, Fitzroy, VIC, Australia and
3
Centre for Epidemiology and Biostatistics Melbourne School of Population and Global Health, University of Melbourne,
Melbourne, VIC, Australia.
Body: Currently, breast cancer risk prediction algorithms incorporate phenotypic risk factors, including age, family history,
reproductive history and benign breast disease.
Since 2007, genome-wide association studies have identified a growing number of single-nucleotide polymorphisms (SNPs) that
are associated with an increased risk of breast cancer. While no single SNP is very informative on its own, a polygenic approach
to genetic screening could improve estimates of individual risk, creating the possibility of screening strategies individually tailored
to each woman.
We studied cases and controls recruited to the Australian Breast Cancer Family Registry, to investigate whether a SNP risk score
derived from a panel of 77 SNPs can improve risk estimates obtained from the commonly used breast cancer risk prediction
models (BCRAT, IBIS, BRCAPRO and BOADICEA).
SNP risk scores were calculated using previously published Odds Ratios (ORs) and risk-allele frequencies based on the
assumption of independence of additive risks on the log OR scale. SNP risk scores were calculated by multiplying adjusted risk
scores for each of the SNPs. Combined risk scores were calculated by multiplying the SNP risk score and the algorithm derived
risk scores, under the assumption of independence.
Overall, the inclusion of data on 77 SNPs improved the area under the receiver operating characteristic curve (AUC) for all of the
breast cancer risk prediction models: Preliminary data analysis provided AUC values as follows: BCRAT+SNP = 0.656,
IBIS+SNP= 0.628, BRCAPRO+SNP= 0.689, BOADICEA+SNP= 0.698.
The breast cancer risk prediction models examined place different weighting on the phenotypic components. For example, the
IBIS model is weighted towards uncommon high-risk phenotypes, while the BCRAT is weighted towards lower risks in the general
population. A combination of the 77 SNP score and the most appropriate model might substantially improve the ability to identify
high-risk women in different screening populations.

Title: Balancing the harms and benefits of radiation therapy for DCIS: A decision analysis examining the risk of
radiation-associated sarcoma
Marquita R Decker1, Joseph F Levy2, Lee G Wilke1, David J Vanness2 and Heather B Neuman1. 1University of Wisconsin,
Madison, WI and 2University of Wisconsin, Madison, WI.
Body: INTRODUCTION:
More than 60,000 women are diagnosed with ductal carcinoma in situ (DCIS) annually and offered the option of breast conserving
surgery (BCS), often including radiation (RT) to reduce local recurrence. Although the incidence of radiation-associated sarcoma
(RAS) is low (0.05-0.25% at 10 years), the low mortality associated with DCIS and large number of DCIS diagnoses means that
an increasingly large number of women are at risk of RAS. This study sought to weigh the risk of RAS with the benefits of
BCS+RT for DCIS.
METHODS: A second-order Monte Carlo micro-simulation model of women ages 35 and older with DCIS was constructed. The
decision analysis compared harm-benefit ratios of sarcoma-related deaths per breast cancer deaths averted within 20 years of
treatment with BCS+RT versus BCS alone. Stratified analyses were performed by age group to account for differential life
expectancy. To generate parameter estimates for model inputs, Bayesian network meta-analysis was used to synthesize rates of
DCIS and invasive recurrence from clinical trials of BCS+RT and BCS alone using a Weibull specification. Sarcoma incidence
was estimated non-parametrically using SEER. Constant hazard rates for breast cancer mortality after invasive recurrence and
RAS mortality were estimated from clinical trials. To account for uncertainty, probabilistic sensitivity analysis was conducted using
10,000 Monte Carlo samples and 95% credible intervals (CrI) were constructed for event rates and harm-benefit ratios.
RESULTS: The micro-simulation model of an age-distributed cohort demonstrated that 1 in 840 women with DCIS (95%CrI 1:648
to 1:3522) would develop RAS within 20 years after treatment with BCS+RT. Overall, there would be 1 RAS-related death for
every 12 breast cancer deaths averted (95%CrI 1:7 to 1:19) by the addition of RT to BCS. Stratified analysis demonstrated that
the harm-benefit ratio was higher in women <75 years of age, with more RAS-related deaths caused per breast cancer deaths
averted. The model was most impacted by parameter estimates for rates of invasive recurrence, breast cancer mortality after
invasive recurrence, and RAS incidence rates.
CONCLUSIONS:
The risk of developing a RAS following BCS+RT for DCIS should not be overlooked. This may be especially true for women at
low risk of recurrence and younger women (<75 years in our model). These findings contribute to the ongoing conversation about
consequences of overtreatment of DCIS, and should be incorporated into shared-decision making discussions regarding the
optimal management of DCIS for a given patient.
Age-Stratified Incremental Harm-Benefit Ratios for BCS+RT versus BCS Alone
Age Group
RAS Deaths: Breast Cancer Deaths Averted*

Posterior Median Ratio (95% CrI)
Overall
1:12 (1:7 to 1:19)
35 to 54
1:10 (1:6 to 1:14 )
55 to 74
1:11 (1:7 to 1:15)
75+
1:17 (1:9 to 1:24)
*Probabilistic sensitivity analysis using 10,000 second-order parameter samples with a 20 year time horizon

Title: The development of a prediction tool for moderate to severe diarrhea in HER-2/hormone positive metastatic breast cancer
(MBC) patients receiving lapatinib in combination with letrozole (L-L)
George Dranitsaris1 and Mario E Locouture2. 1Augmentium Pharma Consulting and 2Memorial Sloan Kettering Cancer Center.
Body: Background: For patients with HER-2/hormone positive MBC, the addition of lapatinib to letrozole is associated
improvements in tumour response rates and a prolongation of progression free survival. However, moderate to severe diarrhea (
grade 2) is a potentially serious toxicity which can lead to dose reductions, delays, hospitalizations and even the premature
discontinuation of treatment. Patient care could substantially be improved if these diarrhea events could be accurately predicted
through the use of validated and easy-to-use mathematical models. In this study, the development of a repeated measures model
to predict the risk of grade 2 diarrhea prior to each month of L-L therapy is described.
Methods Data from 111 patients who received the L-L combination as part of a clinical trial were reviewed [Johnston, 2009].
Generalized estimating equations (GEE) were used to develop the final risk model using a backwards elimination process.
Internal validation of the final regression coefficients was done using nonparametric bootstrapping. A risk scoring algorithm (range
0-250) was then derived from the final model coefficients. A receiver operating characteristic curve (ROC) analysis was then
undertaken to measure the predictive accuracy of the final scoring algorithm.
Results: Presence of skin and lung metastases at baseline, cumulative lapatinib dose and Hg level (nadir) were identified as
being important predictors for grade 2 diarrhea. There was also a negative association between time on therapy and risk of
diarrhea where a higher frequency was observed in the first few months. The ROC analysis indicated good predictive accuracy
with an area under the curve of 0.80 (95%CI: 0.72 0.88). Prior to each new month of therapy, patients with risk scores > 125
units would be considered at high risk for developing grade 2 diarrhea.
Conclusions: Risk of grade 2 diarrhea is associated with cumulative lapatinib exposure, disease related factors as well as Hg
level. The planned external validation and eventual clinical application of this prediction tool will be an important source of risk
information for the practicing oncologist and can enhance patient care by optimizing preventative therapies earlier in a proactive
manner.

Title: Can proton therapy for localized mediastinal Hodgkins lymphoma reduce second breast cancer incidence?
Samy R Horn1, Victor Pernin1, Nathalie Fournier-Bidoz1 and Youlia Kirova1. 1Institut Curie, Paris, France.
Body: Introduction
Secondary breast cancer (SBC) is a recognized late complication of radiation therapy, after treatment for Hodgkin lymphoma
(HL), with reported relative risks as high as 50 with older techniques (Mantle Field). Within the past, it has been estimated that
reduction in dose and volume will reduce the incidence of late complications such as secondary neoplasms and cardiovascular
disease. We evaluate the impact of mediastinal proton therapy on potential limitation of SBC risk.
Material & methods
For 14 young female patients with early-stage, mediastinal HL, treated with chemotherapy and involved-field raditation therapy
(IFRT), we simulated similar treatment plans with conformal radiotherapy (CRT), helical tomotherapy (HT) and proton therapy at
the dose of 30 Gy. We report the respective doses to the breasts. Treatment plans were not specifically designed for breast
sparing.
Results
Proton therapy significantly lowered the dose to the breasts, with mean doses of 2,76/1,53 Gy to the left and right breast
respectively with proton therapy (vs. 4,95/3,88 Gy with HT and vs. 5,56/3,58 Gy with CRT). Proton therapy best limited lower
doses (V4Gy and V10Gy) compared to CRT, while HT could limit the higher doses (V20Gy) at the expense of larger volume
irradiated at low doses (V4Gy).
Conclusion
Relative reduction in mean doses to the left and right breasts was of 50 and 57% respectively with proton therapy compared to
CRT. Relation between radiation dose and SBC seems to be linear, and reduction in SBC should be in such proportion with
proton therapy. Proton therapy seems highly interesting for breast sparing after curative treatment for HL, but these results need
to be confirmed by individualized risk estimations and prospective trials.

Title: Body mass index and prognosis after breast cancer diagnosis in Japanese women
Toshinari Yamashita1, Tomoyuki Aruga1, Hiromi Miyamoto1, Kazumi Horiguchi1, Yayoi Honda1, Nami Idera1, Risa Goto1 and
Katsumasa Kuroi1. 1Tokyo Metropolitan Komagome Hospital, Tokyo, Japan.
Body: Background: Studies conducted mainly in Western countries have reported a relationship between body mass index
(BMI) and prognosis among women with breast cancer. Only a few studies have been conducted in Japan so far because the
percentage of high BMI is low. In the present retrospective study, we investigated the associations between BMI and the clinical
characteristics and prognosis among breast cancer patients.
Methods: We analyzed 1,744 breast cancer patients who started treatment between 2004 and 2012 at a single hospital in Japan.
All patients with ductal carcinoma in situ, male breast cancer as well as metachronous and synchronous bilateral breast cancer
were excluded. Median age was 57 years (range 2391). The number of patients less than 50 years old was 496. World Health
Organization BMI classifications were used: Underweight, less than18.5 kg/m2, n=157; Normal, 18.524.9 kg/m2, n=1181;
Overweight, 2529.9 kg/m2, n=316; and Obese, more than or equal to 30kg/m2, n=90. The Cox proportional hazards model was
used to estimate hazard ratios for recurrence free survival (RFS) in relation to BMI classifications.
Results: Median follow up interval was 4.2 years. During the follow-up period, 126 breast cancer recurrences were observed.
BMI classification correlated with clinical tumor size (cT) significantly and BMI classification tended to correlate with lymph node
metastases and estrogen receptor (ER) status.
Among patients less than 50 years old, the RFS of those with BMI 25.0 kg/m2 was compared to that of patients with BMI <25.0
kg/m2. In multivariate analyses, BMI classification was one of the significant factors (p=0.02) along with lymph node metastases
(p=0.0001) and ER status (p=0.04).
However in patients aged 50 years or over BMI category was not a significant factor (p=0.12).
Conclusions: It has been reported that higher BMI is a risk factor for breast cancer recurrence among postmenopausal patients.
Our results suggest that higher BMI is also associated with an increased risk of breast cancer recurrence among premenopausal
patients. It raises the possibility that maintaining an appropriate body weight improves the prognosis in premenopausal patients
after they have been diagnosed.

Title: The comparison of the distribution of breast cancer risk factor between Chinese women, non-Chinese Asian and Caucasian
women in the screening cohort of Athena Breast Health Network
Bo Pan1,2, Jeffrey Tice1, Qiang Sun2, Celia Kaplan1, Zhou Yidong2, Yali Xu2, Songjie Shen2, Changjun Wang2, Alexandra
Solomon1, Lauren Ryan1, Paige Kendall1, Timothy Henderson1, Laura Esserman1, Beth Crawford1, Athena Breast Health Network1
and Laura van 't Veer1. 1Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, CA and
2
Peking Union Medical College Hospital, Beijing, China.
Body: Background: A major area of innovation in breast cancer (BC) is improving risk models for screening efforts. The
University of California Athena Breast Health Network uses several risk models, including the Gail model [PMID: 10491430]. The
Breast Cancer Surveillance Consortium (BCSC) model [PMID: 18316752] which integrated information on breast density was
developed and validated by UCSF investigators and reports Asian-specific risk. The Peking Union Medical College (PUMC)
model [PMID: 22662004] is developed from a case-control study in China and reports BC risk in Chinese women. Risk factors
and their respective weights included in these models are different.
Methods: From December 2012 to April 2014, 2,305 women without previous history of BC and consented to research were
enrolled in the UCSF Athena screening cohort. Questions asked included risk factors used in Gail, BCSC and PUMC model, such
as age, age of menarche (AOM), age at first live birth, body mass index (BMI), breast biopsy, breast density, hormone
replacement therapy (HRT), oral contraceptives (OCP). Women were considered high risk when: Gail > 1.67% 5-year risk, BCSC
> 1.67% 5-year risk, PUMC score > 30 (equals >0.20% 1-year risk) respectively. The distribution of the risk factors and the high
risk population percentage were compared between Chinese women versus non-Chinese Asian and Caucasian women.
Results: 402 Asian women comprise 17.4% of all 2,305 Athena screening cases with 234 Chinese, and 168 non-Chinese Asian
(NCA). Differences in risk factor distribution were observed for the following: positive family history was observed 23% for
Caucasian, 15% for Chinese and 13% for NCA (p=0.001), and previous breast biopsy was 27%, 17% and 22% respectively
(p=0.002). Ever use of HRT was 36% in Caucasian, 17% in Chinese and 11% in NCA (p<0.001). Ever give birth was 68% in
Chinese, 65% in NCA, and 57% in Caucasian (p=0.001), while the age at first live birth <30 was 35%, 33%, and 26%
correspondingly (p=0.001). Breast density appeared to be higher in Asian women (p=0.095). The high risk proportions by each
model are given in Table 1.
Conclusion: Chinese and NCA women have a lower proportion of high risk by the BCSC and PUMC model compared to the
Caucasian women, whereas by Gail model these proportions appear to be similar.
Table 1 Percentage of high risk population by ethnicity for different risk models
Risk models
Chinese (n=234)
non-Chinese Asian (n=168)
Caucasian (n=1,492)
P Value
Risk
low
high
low
high
low
high
Gail
137 (58.6%)
97 (41.4%)
107 (63.7%)
61 (36.3%)
884 (59.2%)
608 (40.8%)
0.509
BCSC
211 (90.2%)
23 (9.8%)
152 (90.5%)
16 (9.5%)
864 (58.0%)
626 (42.0%)
0.000
PUMC
167 (71.7%)
66 (28.3%)
108 (64.7%)
59 (35.3%)
781 (52.6%)
704 (47.4%)
0.000

Title: Risk of chemotherapy-induced febrile neutropenia (FN) in patients (pts) with non-metastatic breast cancer (BC) and
documented risk factors for FN
Derek Weycker1, Xiaoyan Li2, Richard Barron2, Hongsheng Wu1, Phuong Khanh Morrow2, Hairong Xu2, Maureen Reiner2, Jacob
Garcia2, Shivani Mhatre3 and Gary Lyman4. 1Policy Analysis Inc (PAI), Brookline, MA; 2Amgen Inc, Thousand Oaks, CA;
3
University of Houston, Houston, TX and 4Hutchinson Institute for Cancer Outcomes Research, Seattle, WA.
Body: Background: Clinical practice guidelines recommend primary prophylactic colony-stimulating factor (CSF) for pts with
cancer receiving myelosuppressive chemotherapy when their risk of FN is high (20%). Evaluating FN risk in pts who receive
regimens that are not documented as high risk in guidelines can be challenging; these pts may be at a high risk of developing FN
based on a combination of regimen and pt risk factors. This retrospective study estimated FN risk among subgroups of pts with
non-metastatic BC receiving 1 of 3 commonly used, non-high risk chemotherapy regimens.
Methods: Pt-level data from 2 US healthcare claims databases comprising medical and outpatient pharmacy claims from
commercial and Medicare supplemental plans were pooled. Eligible pts were 18 years old and had initiated a course of TC,
TCH, or non-dose-dense AC/AC-T for non-metastatic BC between July 1, 2003 and June 30, 2012. Occurrence of FN during any
cycle of the first chemotherapy course was identified using diagnosis codes for neutropenia, fever, or infection. Risk factors,
documented in guidelines and the published literature, were evaluated during the 12 months before chemotherapy initiation. The
percentage of pts, FN risk, and relative risk (RR, compared to pts with no risk factors) for subgroups of pts with specific risk
factors are presented.
Results: 50,893 pts were included in the analysis. FN risks by chemotherapy regimen and subgroup are shown in the table.
TC N=26,266
TCH N=9,105
AC/AC-T N=15,522
% Pts FN Risk (RR, 95% CI)
Any risk factor
78.1
14.2 (1.4, 1.3-1.5)
74.5
16.9 (1.3, 1.2-1.5)
73.6
16.6 (1.6, 1.4-1.7)
Age 65
15.1
16.3 (1.6, 1.4-1.7)
11.5
16.3 (1.3, 1.1-1.5)
11.4
17.9 (1.7, 1.5-1.9)
Cardiovascular disease
16.8
16.9 (1.6, 1.5-1.8)
14.6
20.2 (1.6, 1.4-1.8)
16.6
19.5 (1.8, 1.6-2.1)
Diabetes
11.8
17.4 (1.7, 1.5-1.9)
10.2
21.3 (1.7, 1.4-2.0)
11.1
20.4 (1.9, 1.7-2.2)
Liver disease
1.9
18.7 (1.8, 1.5-2.2)
2.2
20.8 (1.6, 1.2-2.2)
2.0
17.8 (1.7, 1.3-2.2)
Lung disease
2.0
23.4 (2.3, 1.9-2.7)
1.5
30.7 (2.4, 1.8-3.2)
1.9
27.1 (2.6, 2.1-3.2)
Renal disease
1.3
22.7 (2.2, 1.8-2.7)
0.9
25.0 (2.0, 1.3-2.9)
1.0
26.8 (2.5, 1.9-3.3)
Osteoarthritis
7.0
19.6 (1.9, 1.7-2.1)
6.1
21.1 (1.7, 1.4-2.0)
6.0
20.8 (2.0, 1.7-2.3)
Thyroid disorder
13.0
15.4 (1.5, 1.3-1.7)
11.9
17.4 (1.4, 1.2-1.6)
10.6
17.2 (1.6, 1.4-1.9)
33.2
11.7 (1.1, 1.0-1.2)
33.9
14.1 (1.1, 1.0-1.3)
33.7
14.2 (1.3, 1.2-1.5)
24.6
13.8 (1.3, 1.2-1.5)
23.2
17.4 (1.4, 1.2-1.6)
22.9
16.0 (1.5, 1.3-1.7)
12.5
16.3 (1.6, 1.4-1.8)
11.2
19.8 (1.6, 1.3-1.8)
11.1
21.3 (2.0, 1.8-2.3)
4+
7.8
22.8 (2.2, 2.0-2.5)
6.1
25.0 (2.0, 1.6-2.3)
5.9
24.8 (2.3, 2.0-2.7)
Reference group (0 risk factor)
21.9
10.4 (NA)
25.5
12.7 (NA)
26.4
11.6 (NA)
Chronic comorbidities
Number of risk factors
Selected risk factors are presented.

A high proportion of pts (74%-78%) had 1 risk factor for FN, and these pts had a higher FN risk than pts with no risk factors.
55.4%, 56.7%, and 42.2% of all pts who received TC, TCH, or AC/AC-T, respectively, received CSF prophylaxis in cycle 1.
Conclusions: FN risk assessments are needed for pts who are receiving non-high risk regimens.

Title: The KRAS-variant, multiple breast cancer risk, and estrogen withdrawal
Joanne B Weidhaas1, Terri Mcveigh2, Robert Pilarski3, Nicola Miller2, Joann Sweasy4, Michael Kerin2, Sunitha Nallur4, Jackie
Sadofsky4, Antonio Nemec4, Karl Sweeney2 and Zelterman Daniel4. 1UCLA, Los Angeles, CA; 2National University of Ireland
Galway, Ireland; 3Ohio State University, OH and 4Yale University, New Haven, CT.
Body: Background: The KRAS-variant is a germ-line, functional microRNA binding site mutation in KRAS that leads to increased
risk of premenopausal triple negative breast cancer (TNBC) (1), as well as double primary breast/ovarian cancer in women (2).
Evidence of an association with post-menopausal (PM) cancer onset (3), as well as altered BC biology in hormone replacement
therapy (HRT) users (4), led us to investigate the hypothesis that estrogen exposure alters BC risk and biology in KRAS-variant
individuals.
Methods: Women with a confirmed diagnosis of BC were invited to join a study of the KRAS-variant and hormones, and asked to
complete a questionnaire, send in a DNA sample and submit pathologic information. For the study 1712 BC survivors were
eligible for inclusion. An additional cohort of unaffected women, positive for the KRAS-variant (n=80), were included from Ohio
State University as the control population. Differences were evaluated through 2 tailed t-test and Chi squared analysis.
Additionally, isogenic, perfectly matched cell lines +/- or -/- for the KRAS-variant were created to study transformation in vitro.
Results: Of the 1712 study participants, 17.4% (298) had the KRAS-variant. Compared to controls, KRAS-variant BC patients had
significantly lower Body Mass Index (BMI) (p<0.0001), were significantly likely to have fewer live births (p=0.0028), and to be
older at their first birth (p=0.0070). There was no difference in age of testing between the BC and control KRAS-variant cohorts.
Excluding 68 (3.9%) women with other known BC associated mutations (BRCA1, BRCA2, p53, and PTEN), KRAS-variant BC
patients were significantly more likely to be diagnosed with a second, independent BC compared to non-variant BC patients
(p=0.0046). No differences were found between the variant and non-variant groups in synchronous versus metachronous
presentation, time between first and second breast cancer diagnoses, time between diagnosis and enrollment, or lobular
histology.
For women diagnosed when PM in our cohort (n=765), tumor biology in KRAS-variant women was significantly impacted by HRT
history. In KRAS-variant PM BC patients that stopped HRT before their BC diagnosis (ex-HRT users), the proportion developing
TNBC was significantly higher than for non-variant ex-HRT users (35.5% versus 7.3%, p<0.0001)). The proportion of
KRAS-variant ex-HRT users developing TNBC was also significantly higher than for KRAS-variant women who were never on
HRT (11.4%, p=0.020) or for those who were on HRT when diagnosed with BC (3.6%, p<0.0001). In addition, KRAS-variant
ex-HRT users had significantly higher-grade tumors than non-variant ex-HRT users (p=0.0286).
Finally, cell line studies with isogenic KRAS-variant cell lines demonstrated that estrogen withdrawal led to significantly enhanced
cell line growth, and apparent transformation for the KRAS-variant+/- line compared to the non-variant -/- line.
Conclusions: Based on our studies, it appears that estrogen withdrawal may be a risk for breast cancer development and
aggressive breast tumor biology for women with the KRAS-variant.
1. Paranjape, et al. (2011). Lancet Oncology. 12(4):377-86. Dec 5.
2. Pilarski, et al. (2012). PLos ONE 7(5): e37891.
3. Ratner, et al. (2011). Oncogene 31(42): 4559-66.
4. Cerne, et al. (2012). BMC Cancer 12(105).

Title: The contribution of common genetic variation to breast cancer risk among women receiving tamoxifen or raloxifene within
the National Surgical Adjuvant Breast and Bowel Project (NSABP) P-1 and P-2 trials
Celine M Vachon1, Daniel J Schaid1, James N Ingle1, Matthew P Goetz1, Donald L Wickerham2,3, Michiaki Kubo4, Erin E Carlson1,
Soonmyung Paik3, Norman Wolmark3,5, Yusuke Nakamura6, Liewei Wang1, Richard M Weinshilboum1 and Fergus J Couch1.
1
Mayo Clinic, Rochester, MN; 2Section of Cancer Genetics and Prevention, Allegheny General Hospital, Pittsburgh, PA; 3National
Surgical Adjuvant Breast and Bowel Project, Pittsburgh, PA; 4RIKEN Center for Genomic Medicine, Tokyo, Japan; 5Allegheny
General Hospital, Pittsburgh, PA and 6University of Chicago Medical Center, Chicago, IL.
Body: Purpose: Tamoxifen and raloxifene, are primary prevention strategies for women at high risk of breast cancer
(Visvanathan, 2013). Recent advances in genetic studies of breast cancer risk have identified common susceptibility loci that
explain 14% of familial risk for breast cancer in the general population (Michailidou, 2013). However, it is not known if these loci
are risk factors for breast cancer among high-risk women treated with SERMs for breast cancer prevention. We hypothesized that
the large risk reduction associated with SERMs, coupled with the fact that several breast cancer loci correlate with family history,
may limit the contribution of these common genetic loci to breast cancer in this high risk population. We present the first report to
evaluate 75 established breast cancer susceptibility loci, in the context of a polygenic risk score (PRS), as a risk factor for breast
cancer among high risk women taking raloxifene and tamoxifen for breast cancer prevention.
Methods: We conducted a matched case-control study of 594 cases (i.e., participants who developed breast cancer while on
SERM therapy) and 1,171 matched controls selected from the 33,000 participants enrolled in the NSABP P-1 and P-2 breast
cancer prevention trials. Genotypes of 75 single nucleotide polymorphisms (SNPs) were available from a genome-wide
association study conducted at the RIKEN Center for Genomic Medicine. We formed a quantitative PRS from reported per-SNP
odds ratios (OR) for the 75 susceptibility loci. Conditional logistic regression was used to examine the PRS as a risk factor for
breast cancer and to assess whether the PRS and breast cancer association differed by treatment type, family history, or other
clinical characteristics. Analyses also examined associations of PRS with invasive vs. in situ cancer and ER-positive vs.
ER-negative cancer.
Results: The PRS ranged from 3.98 to 7.74, and a one unit change in PRS was associated with a 42% increase in breast cancer
(OR=1.42; 95% CI: 1.18-1.70; P = 0.0002). There was evidence of a stronger association of PRS with breast cancer among
women with no first-degree family history (OR=1.62, 95% CI: 1.18-2.21) compared to those with a positive family history
(OR=1.32, 95% CI: 1.06-1.66) (Pintx<0.05). The PRS also appeared a stronger risk factor for ER-positive (OR=1.59, 95% CI:
1.25-2.02, P < 0.0002) vs. ER-negative (OR=1.05, 95% CI: 0.68-1.62, P=0.84) breast cancer, although differences did not reach
statistical significance (Pintx=0.10). PRS and breast cancer associations were similar across tamoxifen and raloxifene treatments,
age at trial entry, 5-year predicted Gail model risk, hysterectomy status, BMI, presence of atypical hyperplasia and invasive vs. in
situ cancer.
Conclusion: A polygenic risk score composed of 75 loci was a risk factor for ER-positive breast cancer, especially in the absence
of a first-degree family history of breast cancer. Further, the PRS associations with breast cancer were similar for women taking
tamoxifen or raloxifene for prevention. These data suggest that common genetic variation adds information on risk of ER-positive
breast cancer in a high-risk population receiving SERMs.

Title: Statins and breast cancer stage and mortality in the Womens Health Initiative cohort
Monica P Arun1, Pinkal Desai2, Amy Lehman3, Marilyn L Kwan4, JoAnn E Manson5, Sayeh Lavasani6, Sylvia
Wasswertheil-Smoller7, Gloria E Sarto8, Meryl S LeBoff9, Jane Cauley10, Michele L Cote6, Jennifer Beebe-Dimmer6, Allison Jay11,
Rowan T Chlebowski12 and Michael S Simon6. 1Boston Medical Center, Boston, MA; 2Weill Cornell Medical College, New York,
NY; 3Ohio State University, Columbus, OH; 4Kaiser Permanente of Northern California, Oakland, CA; 5Brigham and Women's
Hospital, Boston, MA; 6Karmanos Cancer Institute, Detroit, MI; 7Albert Einstein College of Medicine, Bronx, NY; 8University of
Wisconsin, Madison, WI; 9Brigham and Women's Hospital, Boston, MA; 10University of Pittsburgh, Pittsburgh, PA; 11Mayo Clinic,
Rochester, MN and 12David Geffen School of Medicine at UCLA, Torrance, CA.
Body: Background: Statins are the most widely prescribed cholesterol-lowering drugs in the United States, with approximately
25% of US adults using statins by 2008. The anti-carcinogenic effect of statins may reduce the metastatic potential of cancer cells
leading to stage migration with statin users more likely diagnosed with early rather than late stage cancer. We evaluated the
effects of statins on breast cancer stage migration and breast cancer-specific mortality in the Womens Health Initiative (WHI)
Clinical Trial and Observational Study. Methods: The study population included 128,675 postmenopausal women aged 50 to 79
years, with 7,883 newly-diagnosed pathologically-confirmed cases of in situ (19%), local stage (61%), regional stage (19%) and
distant stage (1%) breast cancer and 401 deaths due to breast cancer after an average of 11.5 (SD=3.7) years of follow-up. To
reduce the possibility of detection bias, we excluded women who did not report a mammogram within 5 years of study entry, and
who had no health insurance or medical care provider (n=28,237). Stage was coded using criteria implemented in the
Surveillance, Epidemiology and End Results Program and stratified into early stage (in situ and local) vs. late stage (regional and
distant). Information on statin use prior to breast cancer diagnosis was collected at baseline and years one, three, six, and nine
years post-baseline. Self- and interviewer-administered questionnaires were used to collect risk factor information. Cause of
death was based on medical record review by physician adjudicators. Cox proportional hazards regression models were used to
calculate hazard ratios (HR) with 95% confidence intervals (CI) to evaluate the relationship between statin use as a
time-dependent exposure and diagnosis of late stage breast cancer and breast cancer-specific mortality adjusting for important
confounders. For these models, participants who had early stage breast cancer or who died during follow-up were censored at
time of diagnosis or death, respectively. We also evaluated the effect of statins stratified by estrogen receptor (ER) status.
Statistical tests were two-sided. Results: Statins were used by 10,474 women (8%) at baseline. In the multivariable-adjusted
time-dependent model, statin use was associated with a marginal reduction in risk of late stage breast cancer (HR 0.84, 95% CI:
0.70-1.02, p=0.082) and a reduction in risk of late stage estrogen receptor positive breast cancer (HR 0.79, 95% CI: 0.63-0.99,
p=0.044). Statin use was also associated with a marginal reduction in breast cancer-specific mortality although it did not reach
statistical significance (HR 0.59, 95% CI: 0.32-1.06, p=0.075). Conclusions: In the WHI, statin use was associated with a modest
reduction in risk of late stage estrogen positive breast cancer and a non-significant decrease in breast cancer mortality. We are
currently performing sensitivity analyses on these models. Studies using other large data sets with longer follow-up and
information on cancer-directed therapy are needed to further evaluate this possible association.


Title: Histologic features of benign breast biopsy tissue and association with ER positive and ER negative breast cancer in the
Mayo BBD cohort study
Amy C Degnim1, Derek C Radisky1, Robert A Vierkant1, Ryan D Frank1, Marlene H Frost1, Vernon S Pankratz1, Celine M Vachon1,
Tanya L Hoskin1, Julie M Cunningham1, Chen Wang1, Jean-Pierre Kocher1, Teresa M Allers1, Joanne L Johnson1, Tina J Hieken1,
Karthik Ghosh1, Lynn C Hartmann1 and Daniel W Visscher1. 1Mayo Clinic, Rochester, MN.
Body: Introduction: Current models to predict breast cancer risk do not differentiate risk for estrogen receptor (ER) positive and
negative breast cancer (BC), despite growing evidence that these tumors are biologically very different. We hypothesized that
women with ER+ BC cancers have different clinical risk factors and histologic findings on prior benign breast biopsies than those
with ER- BC.
Methods: After IRB approval, we examined associations of age at benign biopsy and histologic features of the benign biopsy with
ER status of incident BCs within the Mayo Benign Breast Disease Cohort. Benign biopsy slides were reviewed for extent of
lobular involution and degree of epithelial proliferation by a single breast pathologist blinded to BC events. Invasive BCs occurring
within 15 years after benign biopsy were classified as ER+ if ER staining was >1%. BC case-only associations with ER status
were evaluated using multivariate logistic regression. Full-cohort hazard ratios (HR) and 95% confidence intervals (CI) for risk of
ER-specific subtypes were estimated using Cox proportion hazards regression.
Results: Among 13,410 women undergoing a benign breast biopsy from 1967-2001, 656 invasive BCs (459 ER+, 106 ER, 106
unknown) occurred within 15 years. Women who developed ER+ and ER- BCs were similar in age at the time of their prior benign
breast biopsy (p=0.34). Although benign biopsies in women who later developed ER+ BC were more likely to show complete
involution (23% vs 15% for ER- BC), this was not statistically significant (p=0.06). However, the degree of epithelial proliferation
was significantly associated with ER status of later BCs (p=0.001), with ER+ BCs more likely than ER- BCs to have had a prior
biopsy with atypical hyperplasia (16% vs 8%), and ER+ BCs less likely than ER- BCs to have had a prior biopsy with proliferative
disease without atypia (33% vs 52%); this association remained after multivariate adjustment (p=0.003). We further pursued the
association of epithelial proliferation with differential risk of ER+ and ER- BC in our overall cohort of 13,410 women (Table 1).
Compared to women with non-proliferative disease, women with proliferative disease +/- atypia had 2-fold hazard ratios for ERBC, whereas hazard ratios for ER+ BC were higher in women with atypical hyperplasia (4-fold) compared to proliferative
disease.
Hazard Ratio of ER+ and ER- BC within 15 years of BBD based upon degree of epithelial proliferation at benign biopsy
Cancer Type
Group
N Events
Hazard Ratio (95% CI)
p-value
Invasive ER- 15 years
NP
8478
42
1.00 (ref)
<0.0001
PD
4229
54
2.63 (1.76, 3.93)
AH
703
2.60 (1.22, 5.54)
NP
8478
232
1.00 (ref)
PD
4229
151
1.33 (1.09, 1.64)
AH
703
74
4.41 (3.39, 5.73)
Invasive ER+ 15 years
<0.0001
NP=nonproliferative disease, PD=proliferative disease without atypia; AH=atypical hyperplasia

Conclusion: ER+ and ER- breast cancers appear to have different features on prior benign breast biopsy, with atypical
hyperplasia showing increased risk for both types of breast cancer, but a greater risk for ER+ tumors.

Title: The birth outcomes of pre-menopausal breast cancer survivors: Do they have a greater prevalence of delivering a preterm
infant?
Kristin Z Black1 and Diane L Rowley1. 1University of North Carolina, Gillings School of Global Public Health, Chapel Hill, NC.
Body: BACKGROUND: Due to the advancement of screening and treatment options for cancer, more people are able to live
fruitful lives after a cancer diagnosis, yet for pre-menopausal breast cancer survivors the effects of disease and treatment on birth
outcomes is not well documented. POPULATION: Linked North Carolina birth record-cancer registry data were used to examine
the birth outcomes of pre-menopausal breast cancer survivors. Out of the 2,213,464 eligible live births that occurred between
1990 and 2009 in North Carolina, 539 of the mothers are breast cancer survivors and 10.6% (n=235,262) of the mothers
experienced a preterm birth (which is below the national average of 12%). A vast majority of the women have a high school
diploma or are college educated (81.4%; n=1,796,594), 14.0% (n=309,208) of the women reported that they smoked during
pregnancy, and about two-thirds of the women were not married at the time of the birth of their child (67.7%; n=1,499,053). A
majority of the study population is non-Hispanic White (62.6%; n=1,385,393) followed by non-Hispanic Blacks (24.0%;
n=531,584), Hispanics/Latinos (9.7%; 215,224), and non-Hispanics of other races (3.7%; n=81,250). METHODS: The aim of this
study was to determine if breast cancer survivors of reproductive age (ages 18-49) who had a live birth after their diagnosis have
a greater prevalence of preterm birth than women who were not diagnosed with breast cancer. Binomial regression was used to
estimate the exposure-outcome association in this case-cohort study. FINDINGS: The crude prevalence of preterm birth for
pre-menopausal breast cancer survivors is 2.01 (95% CI: 1.71-2.36) times the crude prevalence of preterm birth for women who
were not diagnosed with breast cancer. When the data were stratified by race/ethnicity, the prevalence of preterm birth for
pre-menopausal breast cancer survivors compared to women not diagnosed with breast cancer within each racial/ethnic group is
2.27 (1.85-2.79) for Whites, 1.45 (1.10-1.91) for Blacks, 2.23 (0.64-7.81) for Hispanics/Latinos, and 1.83 (0.52-6.50) for other
races. Controlling for the mothers education level, marital status, and smoking status during pregnancy, the prevalence of
preterm birth for pre-menopausal breast cancer survivors compared to women not diagnosed with breast cancer within each
racial/ethnic group is 2.37 (1.93-2.91) for Whites, 1.50 (1.14-1.98) for Blacks, 2.28 (0.65-7.97) for Hispanics/Latinos, and 1.79
(0.51-6.31) for other races. CONCLUSION: Women diagnosed with breast cancer during their reproductive years are potentially
at greater risk of experiencing a preterm birth and may benefit from targeted preconception health interventions.

Title: No Show

Title: History of chemoprevention in patients with newly diagnosed breast cancers
Jennifer Chun1, Freya Schnabel1, Shira Schwartz1, Karen Hiotis1, Amber Guth1 and Deborah Axelrod1. 1NYU Langone Medical
Body: Background:
Chemoprevention (including tamoxifen, raloxifene and exemestane) is a strategy to reduce breast cancer incidence in high risk
women. Studies have shown at least a 50% decrease in the incidence of breast cancer in users of these drugs. The benefit is
limited to a reduction in the incidence of ER/PR positive breast cancer. However, there is a growing population of women who
have used these agents for primary breast cancer prevention, and a larger population of breast cancer survivors who have used
these drugs as part of their systemic treatment. The purpose of this study was to identify a contemporary cohort of women with
newly diagnosed breast cancers who had utilized chemoprevention strategies and describe their patterns of disease.
Methods:
The Breast Cancer Database at our institution was queried for patients who reported use of chemopreventive drugs and
developed breast cancer between 1/2010-4/2014. Patients were divided into primary and secondary chemoprevention groups (no
previous history of breast cancer and previous history of breast cancer, respectively). Descriptive statistics were utilized.
Results:
Out of a total of 1782 patients with newly diagnosed breast cancer, there were 106 (6%) patients who had used a
chemopreventive agent. Out of these 106 patients, 91 (86%), had used the drug as part of their systemic therapy for prior breast
cancer, with a median of 11 years from the initial diagnosis to the diagnosis of a second breast cancer. The primary
chemoprevention group included women with risk based on family history and atypical hyperplasias. The majority of patients
(81%) were diagnosed with early stage disease (stage 0, 1). In both groups, the majority of cancers were ductal in origin (84%).
Eight of the 15 patients (53%) in the primary chemoprevention group were on treatment at the time of their cancer diagnosis;
while 29% of patients in the secondary group were on treatment at the time of diagnosis. In the secondary group, 49% were
contralateral second primary breast cancers and 44% were ipsilateral recurrences. Interestingly, the majority of cancers in both
groups were ER/PR positive.
Conclusions:
Our cohort of women who used chemoprevention drugs were overwhelmingly diagnosed with early stage breast cancer, likely
reflecting their commitment to screening. The majority of cancers were ER/PR positive. In this group, the choice of cancer
treatment will need to be modified in light of prior hormonal treatment. Many of the patients in the secondary group were past
users of prevention agents and further work is needed to clarify the duration of benefit of these drugs. In a similar vein, we look
forward to research efforts to define the optimal age to initiate primary chemoprevention in high risk women.

Title: Longitudinal comparison of weight change in breast cancer survivors to cancer-free women: a prospective study in women
with a familial risk of breast cancer
Amy L Gross1, Sarah A Davidovics1, Deborah K Armstrong2, Jennifer E Axilbund2, Betty J May1 and Kala Visvanathan1,2. 1Johns
Hopkins Bloomberg School of Public Health, Baltimore, MD and 2Johns Hopkins Sidney Kimmel Comprehensive Cancer Center,
Baltimore, MD.
Body: Background: Although postdiagnosis weight gain has been reported in breast cancer (BC) survivors from the general
population, weight gain has not been studied in survivors with a familial BC risk who tend to be younger, treated with
chemotherapy and more frequently diagnosed with an estrogen receptor (ER) negative BC. In addition, most studies of weight
change in survivors have not included a comparison group of cancer-free women and therefore are unable to distinguish the
impact of cancer or related treatment on weight gain versus increasing age. In this study we examine the change in weight among
survivors compared to cancer-free women from the same high-risk cohort.
Methods: In an ongoing prospective cohort of women at Johns Hopkins with a family history of BC, ovarian cancer, and/or a
BRCA1/2 mutation, we identified 303 survivors of incident stage 0-III BC and 552 cancer-free women who were 20 years at
enrollment, completed a baseline questionnaire and at least one follow-up questionnaire within 4 years. Linear and logistic
regression was used to estimate change in self-reported weight from baseline to follow-up between survivors and cancer-free
women. Survivors were categorized based on time between BC diagnosis and baseline, as well as BC treatment categories.
Results: Greater than 60 percent of BC survivors were diagnosed under 50 years, and 22.3% had ER negative tumors. Neither
weight nor body mass index (BMI) at baseline differed between survivors and cancer-free women; this was irrespective of tumor
subtype. Survivors were more likely than cancer-free women to have ever used statins (19.5% in survivors vs. 14.9% in
cancer-free women). Average change in weight during follow-up was 2.92 (95% CI 0.92, 4.91) pounds greater in survivors
compared to cancer-free women after adjustment for age, baseline BMI, menopausal status, physical activity, statin use, and
enrollment year. The greatest weight gain was observed in survivors diagnosed within 5 years prior to baseline, and treated with
chemotherapy and not hormone therapy compared to cancer-free women. This group of survivors was 2.12 (95% CI 1.16, 3.90)
times more likely than cancer-free women to have gained at least 11 lbs during follow-up. Further those survivors who had used
statins gained significantly more weight compared to cancer-free statin users and non-users with and without cancer (p for
interaction = 0.012); this increase in weight gain was largely driven by survivors who had received chemotherapy and statins (p
for interaction = 0.008).
Conclusion: In this study, BC survivors recently diagnosed and treated with chemotherapy were twice as likely to gain at least 11
lbs, compared to cancer-free women. This amount of weight gain has been linked to an increased risk of heart disease and
diabetes in women. Therefore survivors in these categories may benefit from early interventions aimed to reduce weight gain.
Intriguingly, we observed a statistically significant interaction for statin use on weight gain in chemotherapy-treated survivors,
which may reflect an underlying biological interaction between these agents in BC survivors. Given the prevalence of statin use in
this population, this needs to be explored further.

Title: Metformin increases survival in hormone receptor-positive, Her2-positive breast cancer patients with diabetes
Hee Jeong Kim1, Hyun Wook Kwon1, Jong Won Lee1, Sae Byul Lee1, Hee Seung Park1, Sei Hyun Ahn1 and Hae Na Shin1. 1asan
Medical Center, Seoul, Song Pa Gu, Korea.
Body: Purpose
Metformin use has recently been observed to decrease both the rate and mortality of breast cancer. Our study was aim to
determine whether metformin use is associated with survival in diabetic breast cancer patients by breast cancer subtype and
systemic treatment
Data from the Asan Medical Center Breast Cancer Database from 1997 to 2007 were analyzed. The study cohort comprised
6,967 nondiabetic patients, 202 diabetic patients treated with metformin, and 184 diabetic patients that did not receive metformin.
Patients who were divided into three groups by diabetes status and metformin use were also divided into four subgroups by
hormone receptor and HER2-neu status.
Results
In Kaplan-Meier analysis, the metformin group had a significantly better overall and cancer specific survival outcome compared
with non metformin diabetic group (P <0.005 for both). There was no difference in survival between the nondiabetic and
metformin groups. In multivariate analysis, Compared with metformin group, patients who did not receive metformin tended to
have a higher risk of metastasis with HR 5.37 (95%CI, 1.88 to 15.28) and breast cancer death with HR 6.51 (95% CI, 1.88 to
15.28) on the hormone receptor-positive and Her2-negative breast cancer. The significant survival benefit of metformin observed
in diabetic patients who received chemotherapy and endocrine therapy (HR for DFS 2.14; 95% CI 1.14 to 4.04) was not seen in
diabetic patients who did not receive these treatments.
Conclusion
Patients receiving metformin treatment when breast cancer diagnosis show a better prognosis only if they have hormone
receptor-positive, HER2-positive tumors. Metformin treatment might provide a survival benefit when added to systemic therapy in
diabetic patients.

Title: No Show

Title: Association between mammographic breast density and histologic features of benign breast disease
Karthik Ghosh1, Robert A Vierkant1, Ryan D Frank1, Daniel W Visscher1, Vernon S Pankratz1, Christopher G Scott1, Derek C
Radisky1, Marlene H Frost1, Lynn C Hartmann1, Amy C Degnim1 and Celine M Vachon1. 1Mayo Clinic, Rochester, MN.
Body: Background: Mammographic breast density (MBD) reflects the proportion of fibroglandular tissue in the breast, and is an
established risk factor for breast cancer. Benign breast disease (BBD) is also a known risk factor for breast cancer. We previously
showed an association between lobular involution (physiologic atrophy) and MBD in a large cohort study of women with BBD. We
performed a comprehensive evaluation of the association of histologic features of BBD with MBD. Methods: The Mayo BBD
cohort includes women, 18 to 85 years, who were diagnosed with BBD at Mayo Clinic in Rochester, MN between 01/01/1967 and
12/31/1991. For this study, we included women who had a mammogram within 6 months of BBD diagnosis, and diagnosed with
BBD between 1985 and 1991 when parenchymal pattern of breast density was clinically recorded. Risk factor information such as
age, and body mass index (BMI) was collected from medical records. Parenchymal pattern, assessed clinically and previously
used in multiple studies of MBD and risk, classified breast density based on extent and type of density, into four categories: N1
(non-dense, no ducts visible); P1 (ductal prominence occupying <25% breast); P2 (prominent ductal pattern occupying >25% of
the breast; DY (homogenous plaque-like areas of density). All assessments of benign breast tissue were performed by expert
breast pathologists blinded to both MBD and BBD reports. Histologic characteristics included overall impression (non-proliferative
disease (NPD), proliferative disease without atypia (PDWA), and atypical hyperplasia (AH)), proportion of normal lobules that
were involuted (no (0%), partial (1 to 74%), or complete (75%)), type of AH (ADH or ALH), ductal and lobular hyperplasia,
calcifications, sclerosing adenosis, columnar alteration, cyst, fibroadenoma, marked fibrosis, intra-ductal papilloma, radial scar
(number and size), duct ectasia, and mucocele like tumors. Associations of parenchymal pattern with BBD characteristics were
examined using multicategorical nominal logistic regression models. We first examined associations after adjustment for age and
BMI. All variables statistically significant (p<0.05) in these models were then simultaneously included in a fully-adjusted model.
Results: Of 2,257 women in the study, 14% were <40 years old, 52.7% between 40- 59 years, and 32.9% were 60 years. In this
sample, 55.2% of women had NPD, 36% had PDWA, and AH in 8.8%. MBD was classified as N1 in 21.8%, P1 in 12.6%, P2 in
23.7% and DY in 41.9%. Age- and BMI-adjusted analyses showed that there was an association of parenchymal pattern with
overall impression, lobular involution, ductal hyperplasia, sclerosing adenosis, columnar alteration, cyst, duct ectasia and fibrosis.
Multivariate analyses found that women of younger age (p<0.001) and with lower BMI (p<0.001), presence of fibrosis (p<0.001)
and no involution (p=0.016) were more likely to have P2 or DY parenchymal pattern than women without those characteristics.
Conclusion: Among women with benign breast disease, higher MBD is associated with younger age, lower BMI, fibrosis, and lack
of lobular involution. These findings, concordant with other studies of MBD, provide insight into underlying mechanisms by which
MBD and BBD contribute to breast cancer risk.

Title: Association between vitamin D supplementation and mammographic density change over time in women at high risk for
breast cancer
Katherine D Crew1,2,3, Tong Xiao2 and Mary Beth Terry2,3. 1Columbia University College of Physicians and Surgeons, New York,
NY; 2Columbia University Mailman School of Public Health, New York, NY and 3Herbert Irving Comprehensive Cancer Center,
New York, NY.
Body: Background: Vitamin D deficiency has been linked to breast cancer risk, but less is known about vitamin D and changes
over time in mammographic density (MD), a strong predictor of breast cancer risk. Studies that have evaluated the association
between MD and vitamin D have primarily been cross-sectional designs and focused on average-risk postmenopausal women.
Methods: Using data from a prospective cohort study (1991-2013), we examined whether vitamin D supplementation was
associated with MD at baseline and changes in MD over time. High-risk women had a first-degree family history of breast cancer,
atypical hyperplasia, lobular or ductal carcinoma in situ. They completed baseline questionnaires with self-reported vitamin D
supplement use (Y/N) and had serial mammograms with qualitative assessment of MD (BIRADS categories: 1=0-24%,
2=25-50%, 3=51-75%, 4=76-100%). GEE logistic regression and unordered polytomous regression models were used to assess
the association between change in MD in the short-term (<3 years) and long-term (3 years) with vitamin D use (stayed dense:
BIRADS 3/4 for both exams; stayed nondense: BIRADS 1/2; increased: BIRADS 1/2 to 3/4; decreased: BIRADS 3/4 to 1/2).
Primary confounders were included in every model (age, race, body mass index [BMI], menopausal status) and other additional
confounders were selected based on 10% change-of-coefficient rule.
Results: Of 1171 women who had vitamin D supplement information and a baseline mammogram, 615 had two mammograms
within 3 years from baseline and 461 had a long-term follow-up mammogram. Median age was 49 (range, 17-88), median BMI
23.6 kg/m2 (range, 14.9-53.4), and mean follow-up time 6 years (range, 9 months-18 years). Among women with a BMI<25, no
vitamin D supplementation was associated with dense baseline MD (BIRADS 3/4) after adjusting for age, race, menopausal
status, and annual household income (OR=1.61, 95% CI=1.12-2.33). Those who reported vitamin D use were about 50% less
likely to demonstrate long-term increases in MD (see table below).
Vitamin D and short-term and long-term mammographic density changes
Stay dense
Increase
Decrease
Stay nondense
n; OR (95% CI)
n; OR (95% CI)
n; OR (95% CI)
Vitamin D*
291
36; 0.80 (0.38-1.66)
60; 1.00 (0.53-1.89)
209; 1.41 (0.91-2.17)
Vitamin D**
291
36; 0.82 (0.37-1.80)
60; 0.74 (0.37-1.47)
209; 1.37 (0.87-2.16)
Vitamin D***
211
32; 0.46 (0.20-1.04)
69; 1.08 (0.58-1.99)
134; 1.32 (0.78-2.23)
Vitamin D****
211
32; 0.49 (0.21-1.14)
69; 1.12 (0.60-2.11)
134; 1.34 (0.78-2.30)
*Short-term: adjusted for age, BMI, race and menopausal status; **Short-term: additional adjustment for highest education level,
annual household income; ***Long-term: adjusted for age, BMI, race and menopausal status; ****Long-term: additional
adjustment for highest education level, age of first childbirth, time intervals from the first to the last mammogram
Discussion: Although vitamin D supplementation was not associated with short-term changes in MD, we did observe a trend
toward an association with long-term change among high-risk women. If replicated in larger studies, our study gives added
evidence that MD changes may need longer observation time.

Title: No Show

Title: Psychosocial adjustment, cancer worry and perceived risk in 6-13 year old girls from breast cancer families
Angela R Bradbury1, Linda Patrick-Miller4, Brian Egleston3, Lisa Schwartz2, Colleen B Sands1, Wendy Chung5, Gord Glendon10,
Lisa Tuchman6, Cindy Moore7, Paula Rauch7, Irene Andrulis10, Saundra Buys8, Caren J Frost8, Esther M John9, Theresa Keegan9,
Julia Knight10, Mary Beth Terry5 and Mary Daly3. 1University of Pennsylvania, Philadelphia, PA; 2Children's Hospital of
Philadelphia, Philadelphia, PA; 3Fox Chase Cancer Center, Philadelphia, PA; 4University of Chicago, Chicago, IL; 5Columbia
University, New York, NY; 6Children's National Medical Center, Washington, DC; 7Partners Healthcare; 8University of Utah;
9
Cancer Prevention Institute of California, Fremont, CA and 10Lunenfeld-Tanenbaum Research Institute.
Body: Background: Many parents discuss familial and genetic risk of breast cancer (BC) with offspring. What girls know and
perceive of BC risk and its psychosocial impact is unknown.
Methods: In the multisite LEGACY Girls Study, 6-13 YO daughters and their parents/guardians from BC families (FH+) and
families without BC (FH-) were recruited to examine early determinants of, and responses to BC risk. Parents/guardians
completed surveys reporting on their daughters psychosocial adjustment (PSA). Mothers and daughters 10-13 YO completed
surveys reporting their PSA and perceptions of breast cancer risk. We used linear and logistic regressions with variable selection.
Results: 731 parents/guardians reported on their own and the PSA of 898 daughters (450 FH+, 448 FH-), and 447 girls (10-13
YO) completed surveys (227 FH+, 220 FH-). FH+ girls have lower somatization and internalizing behaviors by parent/guardian
report than FH- girls (Table 1). Intrusive BC (IBC) worry, and avoidant BC (ABC) worry were significantly higher in FH+ girls. IBC
worry (coef=0.8, p=0.04) and ABC worry (coef 1.8, p=0.007) were higher in daughters whose mother had BC. Daughter perceived
stress (PS), anxiety, depression, somatization, and internalizing did not differ by mother BC history. In multivariable models,
daughter-anxiety was associated with mother-anxiety (Coef 1.4, p<0.0001) and BRCA+ family (coef -13.5, p=0.040).
Daughter-depression was associated with mother-anxiety (coef 1.0, p=0.009). Daughter-IBC worry was associated with
mother-IBC worry (coef 0.2, p<0.001). Daughter-ABC worry was associated with mother-IBC worry (coef 0.2, p=0.002) and being
FH+ (coef 1.3, p=0.02). Daughter-PS was associated with mother-depression (coef 0.2, p=0.003). Somatization was not
associated with any variables in multivariable models. FH+ girls were more likely to report themselves at higher risk for BC,
although many reported uncertainty about their own BC risk.
Table 1
Daughter Psychosocial Adjustment
FH+ Mean(SD)
FH-Mean(SD)
By parent/guardian report
n=437
n=431
Anxiety
46.7 (30.3)
49.1 (29.1)
Depression
51.2 (27.2)
52.8 (27.3)
Somatization
43.9 (30.7)**
50.3 (30.7)**
Internalizing
46.1 (29.6)*
50.8 (29.1)*
Daughter reported
n=228
n=222
Anxiety
40.2 (28.5)
42.4 (29.5)
Depression
32.4 (26.2)
32.4 (27.7)
Somatization
36.5 (25.5)
44.0 (27.7)
Internalizing
33.1 (25.5)
34.0 (26.7)
IBC Worry
2.0 (3.4)**
1.2 (2.4)**
ABC Worry
3.8 (5.8)**
2.0 (4.3)**
Perceived Stress
4.8 (2.8)
4.6 (2.8)
Daughter Perceived BC Risk
n=219
n=209
Higher risk for BC than peers
85 (39%)**
27 (13%)**
Same risk as peers
57 (26%)
54 (26%)
Lower risk than peers
19 (9%)
35 (17%)
Responded 'not sure'
58 (26)
93 (45)
* p<0.05, ** p<0.01
Conclusions: Pre-adolescent girls from BC families have lower somatization and internalizing behaviors by parent report, but
higher self-reported BC worry. Daughter general anxiety, depression and BC worry are associated with corresponding mother
affect. Some girls from BC families are aware of their increased risk and related research suggests this may increase with age.
Understanding how PSA and BC worry changes over time and the impact on health and risk behaviors can inform interventions to
optimize responses to growing up in families at familial and genetic risk for breast cancer.

Title: Bone mineral density(BMD) monitoring in postmenopausal women with early stage estrogen and/or progesterone receptor
positive breast cancer on aromastase inhibitor(AI) therapy at University of Toledo medical center. A quality improvement project
Asma Taj1, Sadik Khuder1 and Iman Mohamed1. 1University of Toledo, Toledo, OH.
Body: Background
More than 70% of breast cancer patients develop endocrine-responsive disease with estrogen receptor (ER)-positive or
progesterone receptor-positive tumors or both [1] and require endocrine treatment with either estrogen blockage or ablation. The
aromatase inhibitors are highly effective well-tolerated treatment for postmenopausal endocrine-responsive breast cancer.
However, their use is associated with accelerated bone loss and an increase in fracture risk(2, 3)
Method
Post menopausal patients with early stage breast cancer( stage I and stage II) which was ER and/or PR positive and who were
on AI therapy were included. Total patients 243.Stage I (101) and stage II (142).Agents used were Anastrazole, letrozole,
exemestane .BMD (bone mineral density monitoring)was done by DEXA(Dual-energy X-ray absorptiometry) scan, and compared
with national guidelines.
Results
243 patients were included. At 0 years(at start of AI therapy) 150(62%) did not get a DEXA scan,40(16.4%) had normal DEXA
scan, 48(19.8% )had osteopenia and 5( 2%) had osteoporosis. At 2 years from AI therapy 126(51.8%) did not get a DEXA scan,
35(14.4%) had normal DEXA scan, 73(30%) had osteopenia and 9(3.7%) had osteoporosis. At 5 years from AI therapy 180(74%)
did not get a DEXA scan,21(8.6%) had normal DEXA scan, 35(14.4%) had osteopenia and 7(2.8%) had osteoporosis.
Anastrazole and letrozole were equally associated with osteopenia/osteoporosis. Only 50% of patients with osteoporosis(10/21)
received treatment for osteoporosis.
Conclusions
Although there is evidence of negative impact of the aromatase inhibitors on bone, our data still show a poor application of the
recommendations in order to prevent osteoporosis related to the administration of these drugs. Part of the problem is the mixed
literature on diagnosis and management of osteoporosis. Also medicare covers for a DEXA scan only once very two years.
The national compliance for managing AI induced bone loss is very low which points out tthat this is a global problem. Our
suggestion is a more active implementation of the guidelines, also by means of a greater collaboration between the oncologist
and the specialist in osteoporosis, and the offer of a diagnostic and therapeutic pathway.

Title: A broad spectrum therapeutic strategy for TNBC revealed by a new pathway that coordinates oncogenic RTKs
Thomas F Westbrook1, Amritha Nair1, Tingting Sun1, Kristen L Karlin1, Jessica Kessler1, Ilenia Migliaccio1, Don X Nguyen5, Ronald
J Bernardi1, Alex Renwick1, Chad J Creighton1, Noah Dephoure3, Steven P Gygi3, Chad A Shaw1, Richard Gibbs1, David
Wheeler1, Rachel Schiff1, James G Christensen4, David J Shields4, C Kent Osborne1, Stephen J Elledge2, Susan G Hilsenbeck1
and Michael T Lewis1. 1Baylor College of Medicine, Houston, TX; 2Brigham and Women's Hospital, Harvard Medical School,
Boston, MA; 3Harvard Medical School, Boston, MA; 4Pfizer Global Research & Development, San Diego, CA and 5Yale University
School of Medicine, New Haven, CT.
Body: Triple-negative breast cancer (TNBC) is a collection of diseases with distinct clinical behaviors and heterogeneous
molecular features. Such clinical and genetic heterogeneity has called into question whether there are common pathogenic
mechanisms (and potential therapeutic targets) driving the TNBC subtype of breast cancer. Herein, we present evidence of a
novel tumor suppressor network that is frequently compromised in TNBC, and a broadly-effective strategy to target this pathway
for TNBC therapeutic intervention. Using an unbiased genetic screen, we identified a tumor suppressor network governing tumor
survival of TNBCs in vitro and in vivo. We define the tyrosine phosphatase PTPN12 as a core component in this network.
PTPN12 is a potent suppressor of mammary epithelial cell survival and transformation, and PTPN12 function is compromised in
more than 70% of human TNBCs. Notably, the tumorigenic and metastatic potential of PTPN12-deficient TNBCs is severely
impaired by restoring PTPN12, suggesting that strategies to mimic PTPN12 function have substantive therapeutic potential. Using
integrative proteomic, genetic, and pharmacologic approaches, we demonstrate that PTPN12 suppresses TNBC survival by
inhibiting multiple oncogenic receptor tyrosine kinases (TKs) including MET, PDGFR, and others. Frequent inactivation of
PTPN12 in human TNBC unleashes these oncogenic TKs in a concerted manner. Importantly, combination inhibitors targeting
these PTPN12-regulated TKs significantly impair TNBC cell survival and confer robust tumor regression across a panel of 18
patient-derived xenograft ("PDX") models of human TNBC. This suggests that TNBCs are broadly dependent on a distinct
combination of proto-oncogenic tyrosine kinases constrained by PTPN12. Collectively, these data identify PTPN12 as a
commonly inactivated tumor suppressor in TNBC and provide a rationale for combinatorially targeting select receptor tyrosine
kinases in TNBC and other cancers based on their defects in tyrosine phosphatase activity.

Title: Spatial distribution of veliparib measured by matrix assisted laser desorption ionization mass spectrometry in triple negative
breast cancer tumors
Imke H Bartelink1, Brendan Prideaux3, Byron Hann2, Gregor Krings2, Jean-Phillippe Coppe2, Lamorna Swigart-Brown2, Pei Rong
Evelyn Lee2, Laura Esserman2, Douglas Yee4, Denise Wolf2, Rada M Savic1 and Laura van t Veer2. 1University of California, San
Francisco, CA; 2UCSF Helen Diller Family Comprehensive of University of California San Francisco, San Francisco, CA; 3New
Jersey Medical School Public Health Research Institute Rutgers, The State University of New Jersey, Newark, NJ and 4Masonic
Cancer Center, Minneapolis, MN.
Body: Background: Veliparib, an inhibitor of Poly(ADP-ribose) polymerase (PARPi), in combination with DNA-damaging agents
showed significant efficacy, especially in triple negative breast cancer (TNBC) patients. However, in I-SPY 2, approximately 42%
of TNBC did not optimally respond to the veliparib based treatment. To exert therapeutic effect, drugs such as veliparib or the
DNA damaging agent carboplatin must reach cancer cells at an adequate concentration. Differences within the microenvironment
of the tumor (e.g. reduced vascular density and leaky endothelium) are responsible for variability in the uptake and distribution of
the drug in the tumor. We hypothesize that heterogeneous or insufficient drug concentrations in the solid tumor lead to
inadequate response to PARPi in some TNBC patients. First we performed a mouse xenograft study to determine the penetration
and distribution of veliparib and carboplatin in TNBC tumors in mice.
Methods: 1*106 MDA.MB.231, HCC70 or MDA.MB.436 cells were injected bilaterally into mamamary fat pad of a total of 36 Beige
SCID mice. When the tumors exceeded 200 mm3, veliparib (20mg/kg or 60mg/kg) or placebo was orally administered 3 times
daily for 3 days + Carboplatin 60mg/kg on day 1+2. These doses and dosing schedules were estimated to result equivalent
plasma and tumor concentrations in mice compared to the doses currently used in clinical trials. Mice were euthanized 3 hours
after the last dose of veliparib and tumors, and muscle tissues were obtained. In addition, control tissues were spiked with a
broad range of concentrations of veliparib. The spatial distribution of the drugs in the tumor tissue was measured using
Matrix-assisted laser desorption/ ionization mass spectrometric imaging (MALDI). The protonated molecular ion at m/z 245.1 was
used to construct 2D ion maps of the tissue showing relative quantitation of drug levels. H&E staining was performed to
characterize the tumors.
Results: Veliparib in control tissues was detected at very low concentrations with a range of detection between 100 fmol-1nmol.
After dosing, veliparib penetrates into the tumors and was detectable at both dose levels. However, drug distribution within the
tumors was observed to be inhomogeneous. In spots where the drug accumulated, necrosis was observed. Veliparib
accumulated in spots in the tumor and near the rim of the tumor. Differences between cell lines were observed, with the largest
accumulation at the rim in BRCA mutated cells.
Conclusions: Veliparib can be measured using MALDI with good specificity and sensitivity and using concentrations equivalent to
patients. TNBC tumors show largely heterogeneous distribution of PARPi, with accumulation in the pushing margings of the
tumor. Future analyses will determine whether this heterogeneity in drug distribution explains variability in response to PARPi
therapies.

Title: HACE1 loss results in hyperactive Rac signaling conferring resistance to HER2 targeted therapies
Erik T Goka1, Dana S Senderoff1, Gustavo Munguba1 and Marc E Lippman1. 1University of Miami Miller School of Medicine,
Miami, FL.
Body: HER2-targeted therapies have been instrumental in improving the treatment of HER2+ breast cancer. However, drug
resistance usually emerges and remains a big challenge in using anti-HER2 therapies. Previous studies have implicated
activation of the Rho GTPase Rac1 by PTEN loss or IGF-1R overexpression as a mechanism of resistance to HER2-targeted
therapies. We identified HACE1, an E3 ubiquitin ligase for Rac1, as a tumor suppressor capable of cooperating with HER2 to
transform normal mammary epithelial cells. While HACE1 loss alone resulted in enhanced Rac activation, HER2 activation of
Rac1 combined with HACE1 loss resulted in even higher levels of activated Rac resulting in the ability to form tumors in
immunocompromised mice. In this study, we show that loss of HACE1 confers resistance to the HER2 tyrosine kinase inhibitor
(TKI) Lapatinib due to sustained Rac signaling irrespective of EGFR/HER2 signaling. While Lapatinib inhibition alone is capable
of attenuating proliferation of HER2 overexpressing mammary epithelial cells, HACE1 loss continues to drive proliferation of in
vitro tumor formation (clonogenicity) as well in vivo tumor growth. We demonstrate that Lapatinib sensitivity can be restored using
a Rac inhibitor in HACE1 knockdown cells. Moreover, while the Rac inhibitor alone was capable of attenuating the effects of
HER2 overexpression and HACE1 loss, simultaneous inhibition of both HER2 and Rac signaling was superior to either
monotherapy alone. These results support Rac activation as a mechanism of resistance to HER2-targeted therapies and highlight
the use of a Rac inhibitor to treat HER2+ refractory breast cancers.

Title: OOTR-N007: A phase II neoadjuvant study of letrozole plus palbociclib in postmenopausal patients with ER positive, HER2
Louis WC Chow1,2,3, Chi-Kei Lam2 and Wings TY Loo2. 1Organisation for Oncology and Translational Research, Hong Kong;
2
UNIMED Medical Institute, Hong Kong and 3Institute for Applied Research in Medicine and Health, Macau University of Science
and Technology, Macau.
Body: Background : CDK4-associated kinase activity is required to maintain breast tumorigenesis. Virtually all ER-positive cell
lines harbour loss of p16ink4a. Low expression of CDK inhibitors p21 and p27 and high expression level of cyclin E and D1 have
all been associated with resistance to anti-estrogen therapy. In preclinical models, Rb deficiency is associated with resistance to
antiestrogen therapy. Palbociclib (PD 0332991, Pfizer Inc.) is an oral, potent, highly selective reversible inhibitor of CDK4/6 that
prevents cellular DNA synthesis by prohibiting progression from G1 into the S phase. Palbociclib in combination with letrozole has
shown promising activity in metastatic setting.
Methods : In an open-label, multi-center, single arm pilot study efficacy and safety of neo-adjuvant palbociclib 100 mg QD for 3
weeks plus letrozole 2.5 mg QD in 4 week cycles for 4 months was studied. Postmenopausal patients of mostly Chinese
ancestry, with histologically confirmed ER+, HER2- invasive breast cancer and tumor greater than 2 cm were registered. Patients
with T3N1, T4 or any N2, N3 were excluded.
Aim : The primary endpoint is Objective Response Rate (ORR). The secondary end-point include Pathologic Response Rate
(PRR), Disease-Free Survival (DFS), safety. Exploratory analysis of gene and protein expression in tumors and serial whole
blood is planned.
Results : As of June 2014, 11 patients were recruited. 9 patients completed treatment and 2 are still on study. Of the 9 patients
that completed study at time of this abstract, 1 patient had a complete pathological response (pCR) and 7 had a partial response
(ORR of 89%). Baseline Ki67 levels were low (median 18%), consistent with luminal type A disease. On average, patients
underwent surgery 24 days after completion of the study. Ki67 was measured at baseline, cycle 1 day 15, and at the time of
surgery. Median Ki67 at time of surgery was 10%. There were no significant changes in Ki67 that could be related to treatment
outcome based on this preliminary data analysis in surgical samples only. In the first 8 patients, whom all started at a dose of
125mg palbociclib, 4 patients developed grade 3/4 neutropenia in the absence of fever. Neutropenia was well manageable with
G-CSF and/or dose modification. Protocol amendment allowed starting dose of 100 mg with dose titration to 125mg after the first
cycle. 3 patients started at 100mg and grade 4 neutropenia without fever occurred in 1 subject. Another common side effect was
low grade mucositis (n=5).
Conclusion : Based on these initial results, the addition of palbociclib to neo-adjuvant letrozole appears to be safe and effective.
Historically, ORR in this patient population with letrozole alone is below 55%. The addition of palbociclib increased the preliminary
ORR to 89%. One patient had a pCR, uncommon for patients on neo-adjuvant endocrine therapy. Complete safety and efficacy
data will be included for at least 11 patients, including initial biomarker results. The study continues enrolment up to 45 patients.

Title: Tumor priming with cyclophosphamide for enhanced tumor delivery, penetration and anti-tumor activity of MM-302,
HER2-targeted liposomal doxorubicin
Elena Geretti1, Shannon C Leonard1, Nancy Dumont1, Christopher W Espelin1, Daniel F Gaddy1, Thomas J Wickham1 and Bart S
Hendriks1. 1Merrimack Pharmaceuticals, Inc, Cambridge, MA.
Body: MM-302, HER2-targeted PEGylated liposomal doxorubicin, is a liposomal antibody drug conjugate designed to target
doxorubicin to HER2-overexpressing cancer cells, while limiting uptake into non-target cells. Effective tumor delivery and
penetration are critical barriers to the clinical activity of nanomedicines, including liposomes. Cyclophosphamide has been
successfully combined with both doxorubicin and liposomal doxorubicin in breast cancer therapy, with the two drugs being
administered on the same day to patients. We have evaluated a novel sequential combination regimen of cyclophosphamide with
MM-302 with the goal of improving tumor delivery and penetration of MM-302.
Methods: Biodistribution studies were carried out in multiple tumor xenograft models to assess the delivery of MM-302 and free
doxorubicin, either as single agents or in combination with cyclophosphamide at different dosing schedules. The total doxorubicin
within tumors was quantified by HPLC and microscopically by determining the number of doxorubicin-positive nuclei within frozen
tumor sections. Induction of DNA damage/repair, tumor cell apoptosis, and changes in the tumor architecture in response to drug
treatment (tumor cell density and vascular parameters) were quantified by automated image analysis. Interstitial fluid pressure
measurements were carried out to evaluate changes in the tumor physiology upon cyclophosphamide treatment. Anti-tumor
activity studies in BT474-M3 tumor-bearing mice were performed to evaluate the ability of the different dosing regimens to inhibit
tumor growth.
Results: Pre-dosing of tumors with cyclophosphamide enhanced subsequent MM-302 delivery to tumor xenografts (2-3-fold)
without affecting delivery to non-target tissues, such as the heart and skin. We demonstrate that this effect is critically dependent
on the timing of cyclophosphamide administration. Analysis of cyclophosphamide-treated tumors suggests that the mechanism for
improved MM-302 delivery involves the induction of tumor cell apoptosis, reduction of overall tumor cell density, substantial
lowering of interstitial fluid pressure and increase in vascular perfusion. Finally, treatment of tumors xenografts with
cyclophosphamide followed by MM-302 resulted in a significantly greater tumor growth inhibition compared to either single agent
alone.
Conclusions: Rational combination of MM-302 with cyclophosphamide results in an active and tolerable regimen in preclinical
models that enhances the tumor delivery and activity of MM-302, without affecting doxorubicin exposure to non-target organs.
This novel sequential dosing strategy represents an advance in addressing the critical challenge for tumor delivery of
nanomedicines. This work provided data supporting the initiation of a clinical evaluation of the effect of cyclophosphamide on
MM-302 delivery as part of an on-going Phase 1 clinical trial of MM-302 in HER2-positive metastatic breast cancer
(http://clinicaltrials.gov/show/NCT01304797).

Title: VS-6063 (defactinib) and VS-4718 reduce cancer stem cells in models of breast cancer: Implications for clinical trials in the
neoadjuvant setting
Vihren N Kolev1, Sean McDermott2, Max Wicha2, Jonathan A Pachter1, Qunli Xu1 and David T Weaver1. 1Verastem Inc,
Cambridge, MA and 2University of Michigan, Comprehensive Cancer Center, Ann Arbor, MI.
Body: Cancer stem cells (CSCs) are an underlying cause of tumor progression and metastasis. In breast cancer, CSCs can be
identified by Aldehyde Dehydrogenase 1 (ALDH) or CD44-high/CD24-low expression. Neoadjuvant chemotherapy has been
shown to lead to an increase in CSCs in locally advanced breast cancer (Alamgeer et al., 2014, Br. Can. Res. R14). In addition,
the presence of CSCs in residual axillary disease is associated with a significantly worse prognosis following neoadjuvant
chemotherapy and surgery (Sakakibara et al. 2011, Cancer 3899, 2011). Currently, there are no approved therapies that
effectively target and kill CSCs. VS-6063 and VS-4718 are orally bioavailable small molecules that kill cancer stem cells through
the inhibition of Focal Adhesion Kinase (FAK). Both VS-6063 and VS-4718 have demonstrated preferential targeting of CSCs in
preclinical models and are currently in clinical development.
Here we report that VS-6063 and VS-4718 effectively kill CSCs in multiple models of breast cancer. In an ex vivo model, biopsies
from human breast tumors were obtained and cultured as primary explants within 24 hours of surgery. The primary explants were
incubated with VS-6063, VS-4718 or paclitaxel for 4 days. Treatment with either VS-6063 or VS-4718 decreased the proportion of
CSCs in contrast to paclitaxel. VS-6063 and VS-4718 diminished the self-renewal capacity of primary cultures from established
TNBC patient-derived xenografts as measured by tumorsphere assays. In a MDA-MB-231 mouse xenograft model, treatment with
VS-6063 decreased CSC more than 6-fold in an in vivo limiting dilutions assay. Similarly, using an imaging-based 4T1-luciferase
TNBC orthotopic model, both VS-6063 and VS-4718 diminished the size of metastatic nodules within two weeks.
CSCs are readily detectable in primary breast cancers at surgery, yet methods to detect these populations are still developing. A
multiplex assay for the CSC markers, ALDH1, CD44 and CD24, was explored with biopsies of primary tumor and matched lymph
node, and primary tumors taken pre- and post-neoadjuvant chemotherapy. Consistent with previously reported data, elevated
levels of ALDH were observed at higher levels post-treatment, and in lymph nodes. In addition, zones of ALDH+ and
CD44-high/CD24-low tumor cells can be mapped with the multiplex assay for the potential detection of CSCs in breast cancer
biopsies.
In summary, VS-6063 and VS-4718 diminish the CSC subpopulation in vitro, ex vivo and in xenograft models using a number of
functional and biomarker assays. This critical subpopulation of CSCs is detectable in residual tumor following neoadjuvant
therapy. Potentially, multiplex assays of CSC markers will be an improved means to monitor CSCs in clinical specimens.
CSC-targeted agents such as VS-6063 or VS-4718 should be clinically tested in the neoadjuvant setting to potentially delay time
to relapse and improve patient outcome.

Title: Perioperative administration of desmopressin (dDAVP) in breast cancer patients: A phase II dose-escalation study
Ruth S Weinberg1, Marcelo O Grecco1, Gimena S Ferro1, Debora Seigelshifer1, Nancy V Perroni1, Francisco J Terrier2, Analia
Sanchez Luceros3, Enzo Domenichini4, Marcelo D Guthmann5, Daniela Di Leo5, Eduardo Spitzer5, Graciela N Ciccia6, Ana V
Torbidoni7, Juan Garona7, Marina Pifano7, Giselle V Ripoll7, Roberto E Gomez5 and Daniel F Alonso7. 1'Eva Peron' Hospital, San
Martin; 2Italian Hospital, La Plata; 3National Academy of Medicine, Buenos Aires, Argentina; 4'Alexander Fleming' Institute,
Buenos Aires, Argentina; 5Elea Laboratories, Buenos Aires, Argentina; 6Chemo-Romikin, Buenos Aires, Argentina and 7National
University of Quilmes, Buenos Aires, Argentina.
Body: Background: dDAVP is a well known peptide analog of the antidiuretic hormone vasopressin, that has been used to
prevent bleeding during surgical procedures in patients with hemostatic disorders. It induces a rapid increase of hemostatic
mediators by stimulating their release from microvascular endothelial cells. In preclinical studies, dDAVP inhibited lymph node
and early blood-borne metastasis from aggressive mouse mammary tumors. Besides, perioperative administration of dDAVP
significantly prolonged disease-free and overall survival in a veterinary clinical trial enrolling dogs with locally advanced mammary
cancer. The compound is a selective agonist of V2 vasopressin receptors present on both endothelial and breast cancer cells.
Recent evidence indicated that dDAVP promotes tumor-mediated production of angiostatin and also activates endothelial release
of von Willebrand factor (vWF), which may cause apoptosis of micrometastatic cells. Considering its hemostatic and
antimetastatic properties, a phase II dose-escalation trial was performed in patients with breast cancer, administering a
lyophilized formulation of dDAVP by IV infusion in saline, before and after surgical resection of primary tumor.
Methods: Eligibility included otherwise healthy female patients between 18 and 65 years of age, histological/cytological diagnosis
of breast carcinoma (Stage 0, I, II), and mastectomy or lumpectomy with sentinel node biopsy, either requiring or not further
axillary dissection. dDAVP was administered in two IV infusions, the first 30-60 minutes before surgery and the second 24 hours
later. Five groups of at least 4 patients each received increasing total dDAVP doses of 0.5, 1.0, 1.25, 1.5 and 2.0 g/kg. Primary
endopoints were safety and tolerability in breast cancer patients undergoing surgery as first treatment, as well as selection of the
best dose for clinical use. Secondary endpoints included surgical bleeding, plasma levels of vWF, and circulating tumor cells
(CTC) as measured by quantitative PCR detection of cytokeratin 19 (CK-19) mRNA in whole blood.
Results: The trial accrued 21 patients from April 2012 to February 2014. Adverse events were reversible and observed from the
third dose level (1.25 g/kg), including nausea, hot flushing, skin rash, dyspnea and palpitations. Reactions were adequately
managed by slowing the infusion rate of dDAVP (over 30 minutes). A reduced intraoperative bleeding of up to 50% was noted
with increasing doses of dDAVP. Both vWF antigen and activity showed a rise after each dDAVP infusion, and maximum plasma
levels were obtained at the higher dose level of 2 g/kg. Interestingly, a preliminary analysis indicated a drop in CTC counts
24-48 hours after dDAVP treatment in patients with detectable CK-19 mRNA preoperative levels.
Conclusions: At the highest dose level evaluated (2 g/kg) perioperative dDAVP appeared safe when administered in two slow IV
infusions of 1 g/kg, before and after the surgical procedure. The available data suggest that treatment is associated with
reduction of intraoperative bleeding, higher circulating vWF levels and postoperative drop in CTC counts. Final results will be
available at the time of the meeting.
Clinical trial number: NCT01606072.

Title: Exploring the best therapeutic partner of triple negative breast cancers: Using different characteristics/dependent pathways
of triple negative breast cancer cell lines based on subgroups
Bora Lim1, David T Dicker1, Leah C Kline1 and Wafik S El-Deiry1. 1Penn State Hershey Cancer Institute, Hershey, PA.
Body: Triple negative breast cancer (TNBC) still remains as a difficult disease given the abscence of effective treatment modality,
despite some success in PARP inhibition in BRCA gene mutation patients. Recently, TNBC was elucidated to be a large group of
heterogenic subgroups based on their genetic/proteomic characters. Two main subgroups that consists of greater than 90% of
whole TNBC include mesenchymal like (TNMBC) and basal (TNBBC) subgroups. Mesenchymal like subgroup harbor
myoepithelial, epithelial-to-mesenchymal transition related phenotype, and genetic dependency on growth hormone pathway,
higher sensisitivy to death receptor inducing apoptosis pathway, as well as more dependency on PI3K-Akt-mTOR pathway. In
contrast, basal like subgroup harbor the characteristics that is related to a defective DNA repair system, cell cycles regulation,
and interestingly - dependency on RAS-MEK pathway for their survival and growth. Given explosive enrichment of understanding
such pathways and enumeration of pathway targeting inhibitors, as well as the ability of timely recognition of biomarkers in
patients sample which can differentiate subgroups of TNBC- offer an attractive strategy of novel therapeutics development in this
difficult to treat disease. We tested targeting both dependent pathway and core characteristics of TNBC using representative
panels of each subgroups: MDA-MB-468, BRCA1KD MCF10A, HCC1937 as in TNBBC panel showed effective killing when
various combination of pAkt and PI3K, mTOR inhibitors, CDK1/2 inhibitors, death receptor mediated apoptosis inducing agents
were introduced in combination. In contrast TNMBC MDA-MB-231, SUM159,BT549 showed effective killing when pERK and MEK
inhibitors, in combination with BMI-1 inhibitors were given. Further elucidation of mechanisms of action in each combinations will
further offer the ground of proper combination in subgroups of TNBC that could be further studied in animal model, and finally in
clinical development.

Title: A preclinical study to demonstrate the utility of fulvestrant and MLN0128 combination against HR+ and HER2+ breast
cancer
Shang Victoria Wu1, Hannah Lu1, Masaya Kai1, Noriko Kanaya1, Thenhang Luu1, Courtney Vito1, Laura Kruper1, Joanne Mortimer1
and Shiuan Chen1. 1Beckman Research Institute of the City of Hope, Duarte, CA.
Body: Background: Up to 10% of total breast cancers are positive for both hormone receptor [HR: estrogen receptor (ER)
and/or progesterone receptor (PR)] and HER2. These patients belong mainly to the luminal B subtype, which exhibits resistance
to endocrine therapy. Currently, systemic treatment for HR+/HER2+ breast cancer patients involves a combination of
chemotherapy and HER2-directed therapy. While these therapies improve outcomes, they are associated with pernicious side
effects.
Results and Discussion: As demonstrated by studies from a number of laboratories, in HR+/HER2+ cancer, ER is constitutively
activated. In other words, the estrogen ligand is not needed for ER activation. The constitutively activated ER, through its
non-genomic pathways, can stimulate both HER2 (a feed forward loop) and associated kinases. In patients with HR+/HER2+
tumors, genes in the PI3K and ER pathways have been altered. Activation of PI3K/Akt/mTOR by HER2 overexpression predicts
tumor progression in breast cancer [Zhou et al., Clin Cancer Res. 10: 6779 (2004)]. Therefore, PI3k/AKT/mTOR is considered as
an attractive therapeutic target for HR+/HER2+ breast cancer. Allosteric inhibitors of mTOR, such as RAD001 (Everlimus), only
target mTORC1 but not mTORC2, relieving the negative feedback loop in this pathway, and leading to the activation of AKT. Our
results suggest that to effectively treat HR+/HER2+ cancers, both mTORC1 and mTORC2 signaling must be suppressed. In
these contexts; MLN0128 (i.e., INK128 or "MLN") is a new ATP-competitive inhibitor of mTOR. It targets both mTORC1 and
mTORC2; and it does not interact with FKBP12 (an immunoregulatory protein). Consequently, MLN produces weaker
immune-suppressing effects than everolimus. Results from our cell culture experiments reveal that MLN is twenty times more
potent than everolimus against ER+/HER2+ cells. Furthermore, while MLN alone has been demonstrated to inhibit the
proliferation of ER+/HER2+ cells, we have observed more benefits when it is used as part of a combination. Fulvestrant (ICI) by
itself only suppresses the proliferation of ER+/HER2+ cells partially. However, when MLN and ICI are used together, our studies
revealed a synergistic effect.
In addition, HR+/HER2+ PDX models have been generated to identify novel molecular networks and to examine the in vivo action
of new targeted therapies against HR+/HER2+ cancer. We not only have verified the synergistic effect of MLN and ICI
combination in vivo, we also identified a set of important genes playing roles in the growth of HR+/HER2+ tumors through
RNA-Seq analysis.
Conclusion. Due to their synergistic and targeted action, this MLN and ICI combination could provide better clinical outcome and
less side-effects to HR+/HER2+ breast cancer patients, compared to the currently available options in the clinics.

Title: A phase I study evaluating AZD2014 in combination with fulvestrant in patients with ER+ advanced metastatic breast
cancer
Howard Burris III1, Patricia LoRusso2, W Larry Gluck3, Suzanne Jones4, Muaiad Kittaneh2, Erika Hamilton5, Stephen Green6,
Wendy Burke6, Donald Strickland4, Elisabeth Oelmann6 and Manish Patel7. 1Tennessee Oncology, PLLC/Sarah Cannon Research
Institute, Nashville, TN; 2Karmanos Cancer Institute, Detroit, MI; 3Greenville Health System Institute for Translational Oncology
Research, Greenville, SC; 4Sarah Cannon Research Institute, Nashville, TN; 5Tennessee Oncology, PLLC, Nashville, TN;
6
AstraZeneca, Macclesfield, United Kingdom and 7Florida Cancer Specialists/Sarah Cannon Research Institute, Ft. Myers, FL.
Body: Background: Preclinical data suggest that patients with ER+ Breast Cancer become less sensitive to hormonal therapy by
upregulation of the mTOR pathway. BOLERO-2 demonstrated that the combination of an allosteric mTOR inhibitor and an
aromatase inhibitor improves progression-free survival in postmenopausal women with hormone resistant advanced breast
cancer (NEJM 2012; 366:520-529). AZD2014 is a selective dual mTORC1 and mTORC2 inhibitor whilst fulvestrant is an estrogen
receptor antagonist that is approved for the treatment of postmenopausal women with disease progression following antiestrogen
therapy. This phase I trial assessed the safety, tolerability, pharmacokinetics, and preliminary efficacy of AZD2014 administered
in combination with fulvestrant.
Methods: Continuous BID or QD dosing of AZD2014 was administered in combination with fulvestrant 500 mg intramuscularly on
day 1 of each 28 day cycle. An additional 500 mg dose of fulvestrant was administered on day 15 of cycle 1 as per the approved
dosing schedule. Single and multiple dose AZD2014 and fulvestrant pharmacokinetic samples were obtained. Optional tumor
biopsies were also obtained.
Results: 43 patients (median age 61, range 32-82 years; prior chemo in the metastatic setting = 25; prior hormonal therapy = 43)
have been treated in 4 dosing cohorts: 50 mg BID = 13; 35 mg BID = 6; 100 mg QD = 10; and 75 mg QD = 14. Patients have
received 302+ treatment cycles (median 5 cycles/patient, range 1-21) and 4 patients continue on treatment. Review of preliminary
unvalidated safety and Pk data revealed that dose-limiting toxicities included rash/mucositis (1 pt) and hyperglycemia (1 pt) at 50
mg BID; fatigue (1 pt) and mucositis (1 pt) at 100 mg QD; and rash (1 pt) at 75 mg QD. Treatment-related toxicities (any grade)
include: fatigue (51%), mucositis (60%), rash (60%), nausea (47%), hyperglycemia (12%), and neutropenia (2%).
Pharmacokinetic data show that AZD2014 is rapidly absorbed with a median time to maximum concentration of 1-1.75 hours and
an estimated mean half-life of approximately 3.3-5.6 hours across the dose range. There is no evidence that co-administration of
fulvestrant has a clinically relevant impact on exposure to AZD2014. In the 26 patients with measureable disease, 5 confirmed
PRs have been observed (duration 4-18+ months) with an additional 14 patients experiencing stable disease for at least 6 months
of treatment.
Conclusion: Continuous QD (75mg) and BID (35 and 50 mg) administration of AZD2014 in combination with fulvestrant is
tolerable with clinical benefit observed in nearly half of the patients. Toxicities observed are broadly consistent with AEs observed
in other trials with antihormonal agents and other mTOR inhibitors, but seem to be well manageable. There were no new or other
additive toxicities, when the two drugs were combined and AZD2014 pharmacokinetic data are broadly consistent with what has
been previously observed for AZD2014 single agent. A randomized phase II trial of the combination is ongoing and an intermittent
weekly dosing schedule is being explored in additional patient cohorts in the current study.

Title: Phase Ib dose-escalation study of an Akt inhibitor ipatasertib (Ipat) in combination with docetaxel (Doc) or paclitaxel (Pac)
in patients (pts) with metastatic breast cancer (MBC)
Steven J Isakoff1, Johanna C Bendell2, Andrs Cervantes3, Jean-Charles Soria4, L R Molife5, Sandra M Sanabria-Bohorquez6,
Elizabeth A Punnoose6, Shidong Jia6, Premal Patel6 and Cristina Saura7. 1Massachusetts General Hospital Cancer Center,
Boston, MA; 2Sarah Cannon Research Institute, Nashville, TN; 3Hospital Clinico University of Valencia, Valencia, Spain; 4Institut
Gustave Roussy, Villejuif, France; 5Royal Marsden/Institute of Cancer Research, Sutton, United Kingdom; 6Genentech, San
Francisco, CA and 7Vall D'Hebron University Hospital, Barcelona, Spain.
Body: Background: The Akt pathway is frequently aberrantly activated in MBC (e.g. via PTEN loss, and/or alterations of PIK3CA,
AKT1, or AKT3); additionally, Akt activation may occur in response to chemotherapy, leading to cell survival and
chemoresistance. Ipat (GDC-0068) is a potent oral, ATP-competitive inhibitor of all Akt isoforms. In preclinical models, Ipat
synergistically combined with taxanes. In the Phase I dose-escalation single agent study, Ipat was given to pts including MBC,
and downregulated Akt signaling at doses 100 mg.
Methods: Eligible pts with MBC, treated with up to 3 prior systemic chemotherapy regimens, received Doc 75 mg/m2
intravenously (IV) on Day 1 with escalating doses of Ipat PO QD on Days 2-15 every 21 days or received Pac 90 mg/m2 IV on
Days 1, 8, and 15 with escalating doses of Ipat PO QD on Days 1-21 every 28 days. A standard 3+3 does-escalation design was
used, with an expansion cohort of HER2-negative/hormone receptor positive [HER2-/HR+] MBC patents (including triple-negative
MBC [TNBC]) in the Pac cohort at the recommended phase II dose. Pharmacokinetic (PK) and circulating tumor cell (CTC)
samples were collected. Archival tumors were assessed for PTEN by immunohistochemistry and for pathway-relevant mutations.
Results: As of 1 Apr 2014, 54 patients with multiple solid tumors, including 19 pts with MBC (TNBC: n=13, HER2-/HR+: n=4, and
HER2-positive: n=2) were enrolled, and pts with MBC received Ipat with Doc (n=5) or Pac (n=14). The common Grade 2
adverse events (AEs) related to Ipat in combination with Doc ( 10% and > 1 patient) were diarrhea (80%), nausea (60%) and
vomiting (40%), versus in combination with Pac ( 10% of pts) were diarrhea (43%), fatigue (29%) and hyperglycemia (14%). The
PKs of Ipat, Doc, or Pac were comparable to the single agents. Partial responses by RECIST v1.1 were seen in 5 pts, including
pts with HER2-/HR+ (n=2) or TNBC (n=3) who had previously progressed on Pac (n=4) or PI3K inhibitors (n=2) or had tumors
with PI3K/Akt alterations [PTEN loss (n=1), PIK3CA mut (n=2), or AKT1 mut (n=1)]. Time on study > 9 months occurred in 4 pts
(HER2-/HR+ and TNBC) who had progressed on prior Pac (n=3) or PI3K inhibitors (n=3), and/or had tumors with PIK3CA mut
(n=3).
Conclusions: Ipat in combination with Doc or Pac is well-tolerated and has a safety profile generally consistent with the single
agent. Anti-tumor activity with Ipat in combination with taxanes was seen in MBC, including HER2-/HR+ or TNBC with baseline
PI3K/Akt alterations. Updated safety, efficacy, and biomarker data will be presented.

Title: Manumycin A derived nanoparticle induced cytoplasmic vacuolation mediated cell death in triple negative breast cancer
Prajjal K Singha1, Srilakshmi Pandeswara1, Manjeri A Venkatachalam1 and Pothana Saikumar1. 1University of Texas Health
Science Center, San Antonio, TX; 2University of Texas Health Science Center, San Antonio, TX; 3University of Texas Health
Science Center, San Antonio, TX and 4University of Texas Health Science Center, San Antonio, TX.
Body: Background: Major treatment modalities of breast cancer therapy are endocrine, radiation and chemotherapies, and can
inflict apoptosis resistance effects in most clinically aggressive breast tumors. Among these tumors, triple negative breast cancers
(TNBC) are the major aggressive and treatment resistant subtypes. Therefore, it is important to find alternative modes of
treatment to eradicate these TNBC and reduce tumor burden. Earlier we have reported the novel role of Manumycin A (ManA), a
natural antibiotic produced by Streptomyces parvulus, induced cytoplasmic vacuolation mediated cell death in TNBC. In the
present study, nanoparticle based approach was used for delivering ManA to the tumor site to increase the efficacy and reduction
of the drug dose.
Methods: All cell lines were cultured according to the recommended procedures. Lecithin based nanoparticles were made for
both vehicle and ManA. Cell proliferation, transmission electron microscope (TEM), immunoblotting, tumor xenografts and lung
metastasis study were performed by standard methods.
Results: TEM imaging revealed spherical shaped homogenous size distribution of nanoparticles entrapped with vehicle (Veh-NP)
or ManA (ManA-NP). To determine the effect of ManA-NP induced cytoplasmic vacuolation mediated cell death in TNBC, results
of ManA-NP were compared to Veh-NP and ManA drug alone. Two fold reduction in the dose of ManA in ManA-NP induced
cytoplasmic vacuolation death in several TNBC were observed compared to Veh-NP and ManA drug groups. Indeed ManA-NP
caused significantly higher cell death than other groups. ManA-NP induced cytoplasmic vacuolation death was also associated
with increased endoplasmic reticulum (ER) stress markers, LC3, p62 proteins and accumulation of ubiquitinated proteins similar
to that of ManA alone. Importantly, ManA-NP reduced pAkt and increased PTEN and p21 proteins in TNBC. Notably, apoptotic as
well as autophagic inhibitors were unable to protect TNBC against ManA-NP induced cell death. Importantly, thiol antioxidant
N-alpha-acetyl-L-cysteine inhibited the formation of cytoplasmic vacuolation mediated cell death along with the induction of ER
stress, LC3, p62 and protein ubiquitination in TNBC. Interestingly, ManA-NP failed to induce cytoplasmic vacuolation mediated
cell death in slowly dividing normal human mammary epithelial cells even after treating for 72 h at same concentration that is
required for ManA drug alone to induce cytoplasmic vacuolation mediated cell death in rapidly proliferating TNBC. Most
importantly, ManA-NP reduced breast tumor growth derived from MDA-MB-231 cells in mice at 1 mg/ kg of body weight
compared to 5 mg/ kg of body weight of ManA and Veh-NP groups. Finally, ManA-NP completely impeded the lung metastasis of
MDA-MB-231 cells compared to Veh-NP mice. Moreover, lung sections of ManA-NP treated mice showed normal thin-walled
alveoli structure compared to Veh-NP treated lung sections where cancer cells invaded the lung tissue.
Conclusions: These results clearly indicate that ManA-NP enhances the therapeutic efficacy and induces cytoplasmic
vacuolation which serves as a "Trojan-Horse" like mechanism of cell death in TNBC.

Title: Targeting STAT3 with novel small molecule inhibitors to sensitize breast cancer cells to radiation therapy
Lili Wang1, Zhengduo Yang1, Qing Xia1, Haijun Chen2, Guoshuai Cai1, Christopher Wild2, Jia Zhou2 and Qiang Shen1. 1University
of Texas MD Anderson Cancer Center, Houston, TX and 2University of Texas Medical Branch, Galveston, TX.
Body: Radiation therapy plays an important role in controlling the growth and progression of breast cancer. However, its efficacy
is limited by the radiation-associated toxicity to normal tissue, and the intrinsic or acquired radioresistance developed in cancer
cells. It was well documented that radiation onto cancer tissues cause complex changes in gene expression patterns. Thus, it
may be possible to manipulate the expression of specific genes in cancer cells to increase radiosensitivity and reduce
radioresistance. Accumulating studies strongly demonstrated that signal transducers and activators of transcription 3 (STAT3) is
involved in cell survival, proliferation, inflammation, invasion, metastasis, angiogenesis and immune responses. Particularly,
STAT3 is activated by ionizing radiation. We hypothesized that blocking STAT3 will increase sensitivity to irradiation in cancer
cells. To this end, we synthesized a series of novel STAT3 inhibitors to address the challenge of radioresistance in breast cancer
cells including metastatic and triple-negative lines. Among them, the compound HJC0152 and HJC0123 displayed significant
inhibition of proliferation of MDA-MB-231, MCF-7 and MCF-7/Adr in dose and time dependent manners. HJC0152 and HJC0123
also induced apoptosis and necrosis in comparison to the control cells. We also found that these STAT3 inhibitors induce
apoptosis of MDA-MB-231, MCF-7 and MCF-7/Adr cells by inhibiting anti-apoptotic protein Bcl-2 expression, increasing the
expression of apoptotic effector protein caspase-3 and Bax. In addition, we found that HJC0152 and HJC0123 in combination
with X-ray irradiation induce G2/M cell cycle arrest in MDA-MB-231, MCF-7 andMCF-7/Adr cells. The new STAT3 inhibitors
increased the radiosensitivity of MDA-MB-231, MCF-7 and MCF-7/Adr cell lines, inhibited radiation-induced DNA damage repair,
and promoted cells to enter mitosis, a phase more sensitive to irradiation. These changes were accompanied with decreased
activation of STAT3 and decreased expression of the STAT3 downstream gene, Bcl-2. Our findings suggest that STAT3 blockade
may represent an effective strategy to overcome radiation resistance, using STAT3 inhibitors as radiation sensitizers to restore
the sensitivity of cancer cells to radiation therapy.

Title: Inducible suppression of insulin receptor substrate I inhibits IGF-I/insulin/estradiol dependent cell growth in MCF-7L breast
cancer cells
Xihong Zhang1, Sidhant Varma1 and Douglas Yee1. 1Masonic Cancer Center, University of Minnesota, Minneapolis, MN.
Body: Insulin receptor substrate (IRS) proteins are adaptor proteins downstream of insulin-like growth factor (IGF) and insulin
receptors. IRS proteins play a role in cancer biology, with IRS1 function linked to mitogenesis and survival and IRS2 associated
with metastasis. Moreover, this family of adaptor proteins is also involved in the signal transduction of many other transmembrane
receptors. Therefore IRS proteins could be potential therapeutic targets for cancer therapy. To test the function of IRS-1, we
created a doxycycline inducible IRS1 shRNA and stably transfected MCF-7L breast cancer cells. The level of IRS1 protein
knockdown varied within different clones from 50 to 95% reduction compared to non-induced cells. IRS2 mRNA and protein levels
were not significantly affected. We chose a high (3G5) and intermediate (3A7) IRS1 knockdown clones for further study.
Doxycycline treated 3G5 and 3A7, showed partial inhibition of IGF-I, IGF-II and insulin stimulated phosphorylation of
IRS1(Tyr1222) and AKT. pErk1,2 were slightly increased and pIGF-IR was unchanged compared to doxycycline non-treated
cells. IRS-2 tyrosine phosphorylation was stimulated when IRS-1 was suppressed and may account for the persistent
downstream cell signaling. Anti IGFIR antibody dalotuzumab (20g/ml) effectively inhibited cell signaling stimulated by IGF-I,
IGF-II and insulin in both doxycycline treated and non-treated cells. There was no obvious difference in cell signaling pattern
between plus and minus doxycycline treated cells within 24 hour time course. In monolayer cell growth assays, IGF-I, IGF-II,
insulin and estradiol growth stimulation were significantly inhibited when IRS-1 was suppressed in 3G5 cells. In 3A7 cells, which
have a lower level of IRS-1 suppression, growth stimulation was moderately inhibited. In anchorage independent growth assays
stimulated by IGF-I, colony number and size were inhibited by doxycycline-induced IRS-1 suppression with a reduction of 80% in
3G5 cells and 50% reduction in 3A7 cells. These results show that growth of MCF-7L cells are dependent on IRS-1 expression.
Reduced IRS1 can hinder cancer cell growth and tumorigenicity even when signaling was not impaired potentially due to
compensation by other adaptor proteins. We conclude that suppression of IRS-1 may inhibit several growth regulatory pathways
in breast cancer and could serve as a potential drug target.

Title: No Show

Title: Discordance in cyclin D1 changes after metformin exposure by different protein expression methods: Results from a
"Window of Opportunity" trial
Kevin Kalinsky1, Hanina Hibshoosh1, Tian Zheng1, Katherine D Crew1, Susan Refice1, Sheldon Feldman1, Bret Taback1, Eileen
Connolly1, Matthew Maurer1 and Dawn L Hershman1. 1Columbia University Medical Center, New York, NY.
Body: Background: Laboratory and population studies demonstrate that metformin offers a beneficial breast cancer (BC) effect.
In vitro, metformin has been shown to induce cycle cycle arrest. In a pre-surgical metformin trial of overweight/obese, multi-ethnic
BC patients, we reported no difference in tumor proliferation, as measured by ki-67. Reverse Phase Protein Array (RPPA) is a
high-throughput antibody-based technique to assess cellular protein activity in signaling networks. The goal of this study was to
assess changes in cyclin D1 by RPPA as compared to immunohistochemistry (IHC) in patients treated in a pre-surgical metformin
trial.
Methods: Metformin 1500mg PO daily (500mg am/1000 mg pm) was administered for 2-4 weeks prior to resection in 35 patients
with stage 0-III operable BC, BMI > 25 kg/m2, and no history of diabetes. All tumor analysis was performed on paraffin-embedded
tumor tissue. For RPPA, protein was extracted from pre- and post-metformin tissue, denatured by sodium dodecyl sulfate, and
printed on nitrocellulose-coated slides. Samples were probed with 160 antibodies associated with various cellular activities,
including cyclin D1. For RPPA, the cyclin D1 antibody used was by Santa Cruz (SC-718 rabbit polyclonal). For IHC, the cyclin D1
antibody used was by Ventana (SP4-R rabbit monoclonal). We analyzed changes in protein expression in tumor tissue of study
patients with those of untreated historical controls, matched by age, BMI, and tumor characteristics. For RPPA and IHC, paired
t-test was used to calculate within-group changes, and two-sample t-tests were used to compare between-group changes in
cases and controls (significance: p 0.05). For RPPA, multiple comparisons were adjusted for by fixing the false discovery rate
(FDR) at 25%. Pearsons correlation coefficient was used for correlation between RPPA and IHC
Results: Of the 35 metformin-treated patients, 32 were evaluable. The majority were Hispanic (80%). Metformin was
administered for a median of 23 days (range: 8-64). Of the invasive BCs (n=21/35), 80% of patients had HR+/HER2- BC. The 33
historical controls were well-matched. Adjusting for multiple comparisons, there was a statistically significant increase in cyclin D1
by RPPA after metformin vs. control [mean change from baseline after metformin: +0.065 (0.118) vs. mean change from baseline
in control: -0.044 (0.152), p=0.002]. There was no change in cyclin D1 by IHC after metformin vs. control [mean change from
baseline after metformin: -5.64 (21.5) vs. mean change from baseline in control: -6.37 (14.28), p =0.88]. There was no correlation
in cyclin D1 between IHC and RPPA (estimate = 0.09, p=0.31). In vitro assessment of cyclin D1 after metformin use in various
cell lines is ongoing.
Conclusions: We report no correlation between cyclin D1 between IHC and RPPA. Pre-clinical assessment of changes in the
cell cycle after metformin is ongoing. Interpreting results of proteomic findings based off of paraffin-embedded tissue should be
done so with caution.

Title: The effects of the BKM120 in combination with herceptin on HER2+ BC cells and BCSCs
Jin Zhang1,2,3, Feng Yu1,2,3, Jingjing Liu1,2,3, Xiaobei Zhang1,2,3 and Yunhui Hu1,2,3. 1Tianjin Medical University Cancer Institute and
Hospital, National Clinical Research Center of Cancer, Tianjin, China; 2Key Laboratory of Breast Cancer Prevention and Therapy
of Ministry of Education, Tianjin, China and 3Key Laboratory of Cancer Prevention and Therapy, Tianjin, China.
Body: Introduction: Breast cancer stem cells (BCSCs) are suspected to be responsible for tumour recurrence, metastasis as well
as chemo-resistance. Dysregulated PI3K/Akt signaling is implicated in the pathogenesis of a number of breast cancers, including
the HER2+ breast cancer. This study evaluated the efficacy of BKM120 monotherapy and BKM120 plus Herceptin in treatment of
normal cells and the BCSCs of the HER2+ breast cancer .
Methods:The BCSCs of SK-BR-3 cells were isolated by Serum free cells suspension culture. The impacts of BKM120
monotherapy and BKM120 combined with Herceptin on normal breast cancer cells and BCSCs were assayed by MTT assay,
wound-healing assay and plate clone formation assay. The mammosphere-forming efficiency (MFE) of breast cancer cells treated
with BKM120 and BKM120 plus Herceptin were also calculated. The nature of the drug interaction of BKM120 and Herceptin was
evaluated by using the combination index (CI) according to the method of Chou and Talalay. The impacts of BKM120
monotherapy and BKM120 combined with Herceptin in PI3K/AKT pathway on the normal cancer cells and BCSCs of SK-BR-3
were assayed by Western bloting.
Results: BKM120 showed significant antiproliferative activity in the BCSCs, as well as in normal SK-BR-3 cells. After treated with
BKM120, invasion abilities of SK-BR-3 cells was significantly weakened(P<0.01).The colony forming efficiency and the MFE was
also decreased compared with the control (P<0.01). Additionally, the pAKT and pS6 expression levels were inhibited. The effects
of BKM120 in combination with Herceptin were investigated, which was called Combination Index, showed synergism between
the two agents in BCSCs as well as in the normal cells. Furthermore, BKM120 plus Herceptin showed more significantly
suppression on following aspects in BCSCs and normal cancer cells : proliferation, invasion ability, colony forming as well as
MFE. In addition, the levels of pAKT,pS6 in the BKM120 plus Herceptin group were decreased more obviously than those in
control and monotherapy group respectively(p<0.05).
Conclusions:BKM120 can inhibit the growth of normal cancer cells and BCSCs of SK-BR-3, by inhibiting cell proliferation,
impairing the invasion abilities and the MFE, and reducing the expression of pAKT(Thr308,S473)and pS6(Thr389).Moreover,
these results suggest that the combination of PI3K Inhibitor BKM120 with Herception may provide a novel therapy strategy for
HER2+ BC.

Title: Local excision without radiation for ductal carcinoma in situ: 12-year results from the ECOG E5194 study
Lawrence J Solin1, Robert Gray2, Lorie L Hughes3, William C Wood4, Mary Ann Lowen5, Sunil Badve6, Frederick L Baehner7,
James N Ingle8, Edith A Perez9, Abram Recht10, Joseph Sparano11, Kathy Miller6 and Nancy E Davidson12. 1Albert Einstein
Healthcare Network, Philadelphia, PA; 2ECOG Coordinating Center, Boston, MA; 3Harris Radiation Therapy Center at Gordon
Hospital, Calhoun, GA; 4Emory University, Atlanta, GA; 5Mercy Medical Center, Springfield, MA; 6Indiana University, Indianapolis,
IN; 7University of California, San Francisco, CA; 8Mayo Clinic, Rochester, MN; 9Mayo Clinic, Jacksonville, FL; 10Beth Israel
Deaconess Medical Center, Boston, MA; 11Montefiore Medical Center, Bronx, NY and 12University of Pittsburgh, Pittsburgh, PA.
Body: Background: The ECOG E5194 study was a prospective trial designed to evaluate surgical excision (lumpectomy)
without radiation for selected women with ductal carcinoma in situ (DCIS) of the breast with low risk clinical and pathologic
features.
Methods: Eligible patients were enrolled on two study cohorts (not randomized): (1) low or intermediate grade DCIS, tumor size <
2.5 cm; or (2) high grade DCIS, tumor size < 1.0 cm. Cohort assignment was based on pathology assessment from the treating
institution. Protocol specifications included surgical excision of the DCIS tumor with a minimum negative margin width of at least 3
mm or no tumor on re-excision. Radiation treatment was not allowed. From April 1997 to October 2002, 665 evaluable patients
were enrolled through ECOG or NCCTG (561 in Cohort 1; 104 in Cohort 2). Tamoxifen was optional (not randomized) beginning
in May 2000, and was given to 30% of the patients. The primary study endpoint was the rate of developing an ipsilateral breast
event (IBE), defined as local recurrence of DCIS or invasive carcinoma in the treated breast. The median follow-up was 12.3
years. We have previously reported 7-year results (L. Hughes, J Clin Oncol 27:5319, 2009; median follow-up 6.3 years; 66 IBEs),
and we herein provide 12-year results.
Results: Median patient age was 60 years and 58 years for Cohort 1 and Cohort 2, respectively. Tumor size was < 10 mm for
79% and 80% of patients, respectively. The minimum negative margin width was > 5 mm for 64% and 69% of patients,
respectively. There were 99 IBEs, of which 51 (52%) were an invasive IBE. The IBE and invasive IBE rates increased over time
in both cohorts (see Table). The 12-year rates of an IBE were 14.4% for Cohort 1 and 24.6% for Cohort 2 (p = 0.003), and for an
invasive IBE, 7.5% and 13.4%, respectively (p = 0.08). No difference was seen for the 12-year rates of overall survival (84.0% vs
82.8%; p = .96) or contralateral breast events (6.7% vs 12.0%; p = 0.16). On multivariate analysis, study cohort (hazard ratio =
1.81; p = 0.01) and tumor size (p = 0.01) were statistically significant for an IBE, and study cohort was borderline statistically
significant for an invasive IBE (p = 0.08). On central pathology review (75% of cases), neither grade nor comedo necrosis was
associated with the risk of an IBE or invasive IBE (all p > 0.15). Salvage treatment at the time of an IBE included mastectomy for
42% (31/74) and 64% (16/25) of the patients, respectively.
Conclusions: For these selected patients with favorable DCIS based on clinical and pathologic characteristics treated with
surgical excision without radiation, the rates of an IBE and an invasive IBE continued to increase through at least 12 years of
follow-up.
IBE Rates According To Study Cohort.
Cohort 1 (Low or Intermediate Grade)
Cohort 2 (High Grade)
Time
IBE
Invasive IBE
IBE
Invasive IBE
At 5 years
6.0% (4.0%, 8.1%)
2.7% (1.3%, 4.1%)
15.0% (7.7%, 21.7%)
5.3% (0.8%, 9.7%)
At 7 years
9.5% (7.0%, 12.0%)
4.8% (2.9%, 6.6%)
18.2% (10.6%, 25.8%)
7.6% (2.2%, 13.0%)
At 10 years
12.5% (9.5%, 15.4%)
6.4% (4.2%, 8.6%)
24.6% (15.7%, 33.4%)
13.4% (5.9%, 20.9%)
At 12 years
14.4% (11.2%, 17.6%)
7.5% (5.1%, 10.0%)
24.6% (15.7%, 33.4%)
13.4% (5.9%, 20.9%)

Title: The extent of lobular carcinoma in situ (LCIS) at surgical excision predicts for the development of subsequent breast cancer
Jessica C Gooch1, Elena Guerini-Rocco1, Sujata Patil1, Starr B Koslow1, Marina De Brot1, Anna Y Park1, Camilla A Boafo1, Jorge
S Reis-Filho1 and Tari A King1. 1Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Introduction
LCIS is considered both a risk indicator and a non-obligate precursor of invasive breast cancer. A diagnosis of LCIS confers a risk
that is 8-10x higher than that of the general population; yet we remain unable to predict risk of progression and hence lack the
tools to make personalized treatment recommendations. Here we examine the relationship between pathologic features and
extent of LCIS at surgical excision with risk of subsequent breast cancer.
Methods
From a prospective database of 1032 patients with LCIS diagnosed from 1980-2009, we identified patients with and without a
subsequent cancer diagnosis and created a nested case-control cohort study. A case was defined as a breast cancer diagnosis
6 months after LCIS diagnosis. All cases were matched to 1 control (cancer free) based on age at diagnosis of LCIS +/- 5 yrs
and length of follow up. All H&E slides from the first diagnosis of LCIS were requested for central pathology review. Comparisons
between cases and controls were done using conditional logistic regression analysis.
Results
At median followup of 79 mos (2-368mos) 156 pts (15%) in the parent database have developed breast cancer. Case-control
matching resulted in a study cohort of 72 cases and 274 controls for which the original LCIS diagnostic slides were available.
There were no significant differences in clinical characteristics between the study cohort and the parent population. Median time
to cancer among cases was 3.6 yrs (0.5-12.7yrs). Median cancer-free follow up for controls was 9.9 yrs (1.1-19.8yrs). No
significant differences in any of the features examined were observed between cases and controls (Table). Although the median
number of slides with LCIS did not differ, the ratio of the total number of slides with LCIS over the total number of slides reviewed
was significantly associated with case-control status. Cases had a significantly higher ratio (median 0.5, (0.3-1.0)) than controls
(median 0.3, (0-1); p=0.003). On conditional logistic regression analysis, a ratio of >0.5 was associated with a 2.7 (95%CI,
1.4-4.9) greater odds for cancer development when compared to a ratio of <0.25 (p=0.008).
Conclusion
In this nested case-control study, the quantity of LCIS at excision, as measured by the ratio of the number of slides with LCIS
over the total number of slides reviewed, was found to be a significant predictor of subsequent breast cancer. Patients with LCIS
in >50% of slides reviewed had a 2.7 increased odds for cancer; suggesting that the extent of LCIS in a biopsy specimen should
be considered when counseling patients about future risk and risk reducing options.
Table 1
Cases n=72
Controls n=274
p-value
Bilateral LCIS at diagnosis
2 (3%)
6 (2%)
0.81
Multicentric LCIS at diagnosis
7 (10%)
25 (9%)
0.68
Recurrent LCIS over time
9 (13%)
24 (9%)
0.30
Median # slides with LCIS
4 (1-19)
3 (1-41)
0.42
Median # TDLU with LCIS
11 (1-129)
6 (1-342)
0.40
Cell Type
A
A/B
B
24 (33%)
45 (63%)
3 (4%)
99 (36%)
145 (53%)
30 (11%)
0.70
Nuclear Grade
1
14 (19%)
54 (75%)
58 (21%)
189 (69%)
0.54
2
3
4 (6%)
27 (10%)
Necrosis
6 (2%)
NA
Calcifications
27 (38%)
114 (42%)
0.57
Ductal Extension
60 (83%)
249 (91%)
0.09
ALH
12 (17%)
31 (11%)
0.19

Title: The prognostic role of HER2 expression in ductal breast carcinoma in situ
Signe Borgquist1, Wenjing Zhou2, Karin Jirstrm3, Rose-Marie Amini4, Thomas Sollie5, Therese Srlie6, Salma Butt1, Carl
Blomqvist7 and Fredrik Wrnberg2. 1Lund University, Lund, Sweden; 2Uppsala University, Uppsala, Sweden; 3Lund University,
Lund, Sweden; 4Uppsala University, Uppsala, Sweden; 5rebro University, rebro, Sweden; 6Institute for Cancer Research, Oslo
University Hospital, Radium Hospitalet, Oslo, Norway and 7Helsinki University Central Hospital, Helsinki, Finland.
Body: Background: HER2 is a well established prognostic and predictive factor in invasive breast cancer. The role of HER2 in
ductal breast carcinoma in situ (DCIS) is much debated and recent data have suggested that HER2 is mainly related to in situ
recurrences. This contrasts the proposed role of HER2 in the progression from in situ to invasive cancer. Our aim was to study
HER2 as a prognostic factor in a large population based cohort of DCIS.
Methods: All 458 women diagnosed with a primary DCIS 1986-2004 in two Swedish regions were included and tissue microarrays
constructed. Silver-enhanced in situ hybridisation (SISH) and immunohistochemistry (IHC) were used for detection of HER2
amplification and IHC expression. HER2 status and its relation to invasive breast cancer recurrence (IBCR) (ipsilateral or
contralateral invasive events and regional or distant metastasis) and ipsilateral events (IBE) were studied. Kaplan-Meier survival
analyses and Cox proportional hazards regression models were used. Adjustments were made for age, size, radiotherapy and
ER status.
Results: Mean follow up was 184 months. DCIS was screening detected in 75.5% of cases. Breast conserving surgery (BCS) was
performed in 78.6% of whom 44.0% received postoperative radiotherapy. No women had hormonal or chemotherapy. A total of
106 IBCRs and 105 IBEs were identified. 54 IBEs were in situ and 51 invasive cancer. Eighteen women died from breast cancer
and another 114 had died from other causes. 420 tumours could be classified using available SISH or IHC data; 132 were HER2
positive (31.4%) and 288 HER2 negative. SISH and IHC data were concordant in 296 of 332 (89.2%) available cases. HER2
positivity was related to size, grade and ER and PR negativity.
The risk of IBCR was statistically significantly lower subsequent to a HER2 positive DCIS, hazard ratio (HR) 0.53 (95% CI
0.31-0.90), log rank p=0.01. However, the curves did not separate until after almost ten years. HRs after adjustments were similar
to the crude analyses. The risk of any IBE was not statistically differently changed by HER2 status. In women undergoing BCS,
HR was 1.18 (0.77-1.82), log rank p=0.44. But interestingly, divided by type of IBE, HER2-positivity showed an increased risk of
in situ IBEs, HR 1.64 (0.92-2.90), log rank p=0.09 and, a decreased risk of invasive IBEs, HR 0.76 (0.39-1.50), log rank p=0.43.
These correlations were however not statistically significant.
Risk of invasive and local recurrence by HER2 status in a population based cohort of women with a primary DCIS
HER2 positive
HER2 negative
Invasive Breast Cancer Recurrence
HR (95% CI)
Reference 1.0
All patients (n=420)(events=101)
0.53 (0.31-0.90)
Ref
BCS (n=324) (events=94)
1.18 (0.77-1.82)
Ref
BCS, in situ IBEs (events=48)
1.64 (0.92-2.90)
Ref
BCS, invasive IBEs (events=46)
0.76 (0.39-1.50)
Ref
Ipsilateral new Breast Events (IBE)
Conclusions: In this long term follow-up DCIS cohort positive HER2 status in the primary DCIS predicted a statistically lower risk
of IBCR. This effect was seen from ten years after primary surgery and onwards. The risk of an in situ IBE was increased and the
risk of an invasive IBE was decreased if the primary DCIS was HER2 positive. All together, this challenges the role of HER2 as a
driving force of the progression from in situ to invasive cancer.

Title: Metformin for the treatment of ductal carcinoma in situ (DCIS)
Atreyi Dasgupta1, David G Edwards1, Frances S Kittrell1, Susan G Hilsenbeck1, Daniel Medina1 and Sao Jiralerspong1. 1Baylor
College of Medicine, Houston, TX.
Body: BACKGROUND: Each year about 60,000 new cases of breast ductal carcinoma in situ (DCIS) are diagnosed in the United
States. DCIS significantly increases the risk of invasive breast cancer. To reduce this risk, tamoxifen is offered to patients with
ER-positive DCIS in the adjuvant setting. In contrast, no adjuvant therapy exists for ER-negative DCIS. Metformin is a frontline
therapy for type 2 diabetes and its safety and side effect profiles are well documented. Preclinical, epidemiologic, and
retrospective data all support an antitumor effect of metformin in breast cancer. Metformin may act via activation of AMP kinase
(AMPK), a sensor of cellular energy, which in turn inhibits mTOR activity and shuts down global protein synthesis. Previous data
from our group showed that metformin treatment is associated with a higher rate of pathologic complete response in diabetic
breast cancer patients receiving preoperative chemotherapy. In the current study, using preclinical models, we explored the
effects of metformin in DCIS and its progression to invasive breast cancer.
EXPERIMENTAL DESIGN AND METHODS: We tested the anti-tumor activity of metformin in terms of proliferation and invasion
using both DCIS cell lines and a mouse intraductal (MIND) transplantation model previously developed by us. The latter involves
intraductal injection of human DCIS cells into mouse mammary glands, and importantly recapitulates the progression of DCIS to
invasive cancer seen in human breast cancer. Western blot and immunofluorescent staining for AMP kinase, mTOR, and other
relevant targets, as well as flow cytometry, were used to investigate mechanism.
RESULTS: Metformin significantly inhibited cell proliferation, migration, and invasion in all DCIS cell lines tested. In the mouse
model, using metformin at a dose that is readily achievable in human patients, metformin significantly reduced the tumor burden
in terms of median number of ducts filled, from 11 to 2 using DCIS.com cells (p < 0.001) and from 10 to 3 using SUM.225 cells (p
= 0.004). In addition, importantly, metformin inhibited the progression of DCIS lesions to invasive breast cancer significantly, with
a reduction in invasive lesions from 74 to 17% (p = 0.003). In mechanistic studies, metformin showed activation of AMPK and
subsequent inactivation of mTOR and other downstream targets. Treatment with metformin did not lead to significant cell cycle
arrest but rather to increased cell death. We found that cell death by apoptosis increased only modestly with metformin treatment,
while cell death by necrosis increased significantly.
CONCLUSION: Metformin inhibits the growth of DCIS and its progression to invasive breast cancer. To our knowledge, this is the
first reported evidence of metformins anti-tumor activity in DCIS. Our data support the exciting prospect of developing metformin
clinically for the treatment of DCIS.

Title: Clinical-pathological features and treatment modalities associated with disease recurrence in DCIS and micro-invasive
carcinoma
Angela Toss1, Adam Berger2, Fran Guiles2, Jocelyn Andrel Sendecki2, Nicole L Simone2, Rani P Anne2, Tiffany P Avery2, Rebecca
J Jaslow2, Juan P Palazzo2, Melissa A Lazar2, Theodore N Tsangaris2 and Massimo Cristofanilli2. 1University of Modena and
Reggio Emilia, Modena, Italy and 2Thomas Jefferson University Hospital, Philadelphia, PA.
Body: INTRODUCTION
The primary aim in the management of DCIS is the prevention of recurrence and contralateral tumor (CT). Previous studies
reported that younger patients and African Americans (AAs) experience a higher risk of recurrence, second tumors and show
worse overall survival. Nevertheless, risk factors for DCIS recurrence, treatments and outcome are still widely discussed. The aim
of our analysis was to identify clinical-pathological features and treatment modalities associated with recurrence in DCIS and
micro-invasive carcinoma (MIC).
METHODS
In the Thomas Jefferson University Tumor Registry, we identified 820 patients with DCIS and 61 with MIC treated between 2003
and 2013. Associations between recurrence and demographic factors, histopathological features and treatment were assessed.
RESULTS
The median age was 59 years with a racial distribution of 69.3% white, 19.5% AAs and 5.7% Asian. There was no significant
difference in age at diagnosis by ethnicity and, 64.8% was ER/PR positive and 10.1% was HER2 positive. The associations of
age and ethnicity with hormone status and HER2 status were not statistically significant. To date, 73 (8%) patients developed
locoregional recurrence or CT, 50 (68.5%) of which were in situ lesions. Only one patient had MIC in primary lesion. There was
no significant difference in age, while white women had higher recurrence rate than Asians and AAs (9% vs. 4%). ER/PR
negative women were more likely to develop recurrence than ER/PR positive (16% vs. 8%). The hormonal status of primary DCIS
and recurrences was concordant in 89% of cases. Moreover, 43.5% of recurrences were HER2 3+ and 23% were HER2 2+
(FISH unknown). Mastectomy was performed in 26.3% of patients and 73.1% had conservative surgery. Women who received
conservative surgery were significantly older, while there were no significant ethnic differences. Among women who underwent
conservative surgery, the 41.4% that received radiation therapy (RT) were significantly less likely to develop recurrence than the
53.2% that did not. However, there was no significant difference in recurrence between mastectomy and conservative surgery
with RT. Among ER/PR positive patients, the 30% that received preventive hormone therapy was significantly less likely to
develop recurrence than the 61.4% that did not.
CONCLUSION
Our study confirms that ER/PR negative DCIS is associated with significantly higher loco-regional recurrence and standard
preventive modalities reduce the risk. White women appear to have a higher recurrence probably related to risk factors (e.g.
obesity) while the distribution of histological subtypes of DCIS, age at diagnosis, MIC, and treatment modalities did not
significantly differ among diverse ethnic groups. Interestingly, a high rate of HER2 positive recurrences has been reported in our
sample, suggesting that HER2 may represent a potential biomarker for DCIS at high risk of recurrence and therefore HER-2
targeted therapeutic interventions (e.g. vaccines, trastuzumab) can contribute to prevent it. Further analyses are needed to
confirm the correlation between age/ethnicity and DCIS outcome. Better define the subgroup at worse prognosis could help to
identify biomarkers predictive of recurrence or second tumors.

Title: Tamoxifen acceptance by DCIS patients and effect on subsequent ipsilateral and contralateral breast events
Anna M Higham1, Irene B Helenowski1, Shruti R Zaveri1, Daniel H Schneider1, Nora M Hansen1, Kevin P Bethke1 and Seema A
Khan1. 1Lynn Sage Breast Program, Northwestern University Feinberg School of Medicine, Chicago, IL.
Body: Introduction: Existing data demonstrate that the use of endocrine therapy decreases the risk of in-breast recurrence and
contralateral new primary cancers. However, the acceptance of tamoxifen is relatively low in women with DCIS, with reported
acceptance rates ranging from 32% to 67%. We reported 67% tamoxifen acceptance in 2005 (Nakhlis F, JACS, p688); we now
re-examine this question to assess the impact of patient and tumor characteristics, and the use of tamoxifen, on new breast
events in women with DCIS undergoing breast conserving therapy (BCT) or mastectomy.
Methods: We conducted a retrospective review using a prospective database; 695 patients with DCIS were identified between
1998 and 2009. Patient and tumor characteristics included age, tumor grade, tumor size, margin status, re-excision status, use of
BCT, adjuvant radiation therapy (RT) and tamoxifen therapy. Women undergoing bilateral mastectomy were excluded for analysis
of in-breast events. The demographic data were examined for differences in continuous variables using the Wilcoxon rank sum
test, and differences in categorical variables between groups were assessed using Fishers exact test. Multivariate odds ratios
were obtained using logistic regression.
Results: The mean age of the patient population was 61.5 11.4 (median and range: 61 (30-99). 44.9% of patients were over
the age of 50. The median follow up time was 60 months. Among the BCT group, 58.5% of 348 women with ER+ DCIS accepted
tamoxifen therapy, compared to 50.7% of 71 mastectomy patients with ER+ DCIS. 188 women (27.1%) underwent mastectomy
(122 unilateral and 66 bilateral); 507 (72.9%) received BCT. Among all women with complete radiation and tamoxifen data, the
number of new events was 88 (16.9%). Of these, 8 (1.5%) were in the mastectomy group, and 80 (17.9%) in the BCT group
(p=0.0007). Ipsilateral events occurred in 1/99 (1%) woman undergoing mastectomy. In the BCT group complete radiation and
tamoxifen data were available on 448 women, of whom 54 (12%) experienced ipsilateral events (p<0.0001). The multivariate
odds ratio (OR) for ipsilateral breast events in the BCT group receiving RT+ surgery, was 0.66 (95%CI 0.25-1.71); among those
receiving tamoxifen + surgery it was 0.65 (95% CI 0.20-2.1) and for the combination of RT, tamoxifen, and surgery it was 0.23
(95% CI 0.08-0.65, p=0.006). In a model that included these treatment parameters, patient age, DCIS size and grade were
non-significant whereas the presence of free margins was highly significant (OR 0.27, 95%CI 0.10-0.71, p=0.008). Overall, 29%
of new breast events were invasive. Contralateral events occurred in 14/467 women (3%) of BCT and mastectomy patients with
hormone receptor positive DCIS, but eight of these were non-compliant.
Conclusions: The use of optimal DCIS therapy (complete excision, RT, tamoxifen) decreased the odds of ipsilateral breast
events by 73%, whereas RT alone or tamoxifen alone in addition to surgery were less effective. Despite relatively high
acceptance of tamoxifen, lack of compliance remains an issue and will be further examined.


Title: The DCIS Score - Potential for healthcare savings?
Rebekah Young1, Kimberly Gergelis1, Shalom Kalnicki1 and Jana L Fox1. 1Einstein/Montefiore Center for Cancer Care, Bronx, NY.
Body: Introduction
The Oncotype Dx Recurrence Score for DCIS (DCIS Score) is a 12-gene assay derived from the original Oncotype DX test that
evaluates recurrence risk among women with invasive carcinoma of the breast. The DCIS Score provides a local recurrence risk
estimate at 10 years following lumpectomy for ductal carcinoma in situ (DCIS). Results can help guide decisions regarding
adjuvant radiation (RT). Foregoing RT can be a source of significant healthcare savings. We investigated the actual healthcare
dollar savings to-date in our patient population.
Methods
We evaluated patients in whom the DCIS Score was ordered (x) and calculated total cost of testing. Potential cost of RT was that
of IMRT as reimbursed by Medicare for a hypofractionated (16 fraction) course, multiplied by x. Many of our large-breasted
patients require IMRT to increase dose homogeneity and to limit dose to normal tissues. We also calculated potential cost with 3D
conformal (3D-CRT). Total potential cost was the sum of testing and treatment costs, determined for each modality. The number
of patients ultimately treated (y) was also multiplied by these costs. Total actual cost was the sum of test expenses and actual
treatment costs. Savings was the difference between the total actual and total potential cost.
Results
From February, 2012 to May, 2014 the DCIS Score was performed in 38 patients (x= 38). Median age was 66 (40-85). Grade was
low in 39%, intermediate in 45%, and high in 16%. 50% had necrosis present, and the median size of DCIS was 0.5cm (0.1 3.1cm). The total cost of testing was $4125 * 38 = $156,750. IMRT reimburses at approximately $23, 000 and 3D-CRT at
approximately $11,000 per treatment. (Medicare reimbursement rates can vary among states.) The potential total cost of RT
ranged from $418,000 - $874, 000; testing brought the total potential costs to $574, 750 - $1,030, 750. Upon receipt of the test
results, 12 (y) patients ultimately underwent therapy. IMRT was given in 11 patients and 3D-CRT in 1, for a total treatment cost of
$264, 000. Therefore, total actual expenditures were $420, 750. Savings amounted to (574, 750 420, 750) = $154, 000 to
(1,030, 750 420, 750) = $ 610, 000.
Conclusions
In the era of rising healthcare costs, it is imperative to examine instances of possible over-treatment. The DCIS Score has the
potential to save not only healthcare dollars, but to spare patients radiation side effects, time lost from work, and transportation
expenses. While there are costs associated with the assay, if ordered judiciously, these can be offset by the subsequent savings
from eliminating treatment.

Title: A class I histone deacetylase inhibitor, entinostat, enhances lapatinib efficacy in both HER2-overexpressing inflammatory
and non-inflammatory breast cancer cells through FOXO3-mediated Bim1 expression
Jangsoon Lee1, Chandra Bartholomeusz1, Gabriel N Hortobagyi1, Peter Ordentlich2 and Naoto T Ueno1. 1UT MD Anderson
Cancer Center, Houston, TX and 2Syndax Pharmaceuticals, Inc, Waltham, MA.
Body: Background: Human epidermal growth factor receptor 2 (HER2) overexpression has been reported in 15%-20% of breast
cancers and is associated with shorter survival and worse clinical outcome. Inflammatory breast cancer (IBC), a very aggressive
subtype of advanced breast cancer, accounts for approximately 2% of all breast cancers and 8%-10% of all breast cancer-related
deaths in the United States. Recent study represented that about 60% of IBCs overexpress HER2. Although there are effective
HER2-targeted agents, patients with HER2+ breast cancer often have intrinsic and acquired resistance to the anti-HER2 agents
that are currently approved by the U.S. Food and Drug Administration. Therefore, novel combination strategies are needed to
treat HER2+ breast cancers that develop drug resistance. To this end, we investigated the combinational effect of entinostat, an
oral isoform-selective histone deacetylase type I inhibitor, and lapatinib, a HER2/epidermal growth factor receptor dual tyrosine
kinase inhibitor, in HER2+ inflammatory breast cancer (IBC) and non-IBC breast cancer cells. Methods: We assessed the
combinational synergistic effect and its mechanism via CellTiter Blue assay, flow cytometry, anchorage-independent growth,
quantitative real-time polymerase chain reaction, small interfering RNA, Western blotting, and mammary fat pad xenograft mouse
models. Results: We found that compared with entinostat or lapatinib alone, the two drugs in combination synergistically inhibited
tumor cell proliferation (P < 0.001), reduced in vitro colony formation (P < 0.05), and resulted in significant in vivo tumor shrinkage
or growth inhibition in both IBC and non-IBC xenograft mouse models (SUM190 and BT474, P < 0.001). The synergistic antitumor
activity of the entinostat-lapatinib combination was due to downregulation of phosphorylated Akt, which induced the transcriptional
activity of FOXO3, resulting in the induction of Bim1 (a BH3 domain-containing pro-apoptotic protein). Furthermore, entinostat
sensitized trastuzumab/lapatinib-resistant HER2+ cells to the trastuzumab-lapatinib combination and enhanced the
anti-proliferation effect compared with single-agent with lapatinib or combination treatment with lapatinib and trastuzumab.
Conclusion: Taken together, our data provide evidence that entinostat has enhanced antitumor effect in combination with the
HER2-targeted reagent lapatinib and results in the induction of apoptosis by FOXO3-mediated Bim1 expression. Our findings
justify conducting a clinical trial (clinicaltrial.gov, NCT01434303) of combinational treatment with entinostat, lapatinib, and
trastuzumab in patients with HER2+ IBC or non-IBC that is resistant to trastuzumab-based treatment.

Title: Genomic profiling by FoundationOne analysis of inflammatory breast cancer cases reveals a high frequency of clinically
relevant genomic alterations (GA)
Norma A Palma1, Siraj M Ali1, Kai Wang1, Juliann Chmeleiki1, Gary Palmer1, Deborah Morosini1, Jeffrey S Ross1,2, Vincent A
Miller1, Phil J Stephens1 and Massimo Cristofannilli3,4. 1Foundation Medicine, Cambridge, MA; 2Albany Medical College, Albany,
NY; 3Kimmel Cancer Center, Thomas Jefferson University Cancer Center, Philadelphia, PA and 4Inflammatory Breast Cancer
International Consortium.
Body: Background: Inflammatory breast cancer (IBC) is a distinct clinicopathologic entity that carries a worse prognosis relative
to non-IBC breast cancer even when matched for biomarkers (ER/PR/HER2). Genomic profiling of IBC cases may identify
alterations that suggest response to targeted therapies, but is best implemented by an integrated NGS assay capable of detecting
all classes of genomic alterations (GA).
rearranged in cancer were fully sequenced to high, uniform coverage from a commercial CLIA-certified laboratory (Foundation
Medicine).
Results: Of 2,208 clinical breast cancer cases assayed, 55 IBC cases were identified, and of the 50 cases for which hormone
receptor and HER2/neu status were know, 34% were ER-/PR-/HER2- (TNBC). IBC cases harbored 274 GA with an average of
5.0 GA/tumor (range 1-15). At least one alteration associated with an FDA approved therapy or clinical trial was identified in 53/55
(96%) of cases, yielding an average of 2.6 clinically relevant GA/case. Genes most frequently altered were TP53 (60%), MYC
(31%), PIK3CA (25%), ERBB2 (20%), FGFR1 (18%) and PTEN (16%). MYC amplifications were present in 24% of the 2,208
clinical breast carcinoma cases. In the TNBC subset of IBC, 8/19 (42%) pt samples showed MYC amplification (median copy
number 8X, range 7-20) as compared to 9/36 (25%) in non-TNBC pt samples (median copy number 7X, range 6-21).
Within this prospective series, treatment decisions were made based on FoundationOne results. A58-year old pt with likely
secondary IBC harbored two ERBB2 activating base substitutions (V777L and S310F), but without amplification of ERBB2), and
had durable response to lapatinib (Ali et al. JCO 2014). In another case, a 53 year old pt presented with ERBB2-amplified IBC
now refractory to HER2-targeted therapy. FoundationOne testing revealed an activating EGFR mutation (L858R) as well as the
previously described ERBB2 amplification, suggesting that the EGFR alteration may underlie the acquired resistance. The pt
responded to erlotinib monotherapy for 8 months. (Ali et al. Clin Br Ca, 2013). Clinical follow-up for additional patients is ongoing.
Conclusions: 96% of IBC cases harbored at least one alteration that suggests responsiveness to agents that are FDA approved
or being studied in clinical trials. IBC cases also frequently had MYC amplifications (31%), but this may reflect a high percentage
of TNBC-IBC cases with MYC amplifications (42%) in this series. Given the limited treatment options and poor prognosis of
patients with metastatic IBC, the FoundationOne assay with comprehensive NGS-based genomic profiling has the potential to
identify new treatment paradigms and address an unmet clinical need for this disease.



Title: A novel link between anti-apoptotic signaling, NFB, and SMAD7 in IBC pathobiology
Myron K Evans1,2, Scott J Sauer1, Amy J Aldrich1, Joseph Geradts2, Peter Vermeulen4, Luc Dirix4, Steven Van Laere4 and
Gayathri R Devi1,2,3. 1Duke University, Durham, NC; 2Duke University, Durham, NC; 3Duke Cancer Institute, Duke University,
Durham, NC and 4Translational Cancer Research Unit, General Hospital Sint-Augustinus, Antwerp, Wilrijk, Belgium.
Body: Background: Inflammatory breast cancer (IBC) has the highest lethality amongst all subtypes of breast cancer and
develops rapid therapeutic resistance. High NFB activation has been identified as a distinct molecular determinant in IBC
pathobiology; however, the precise sequence of its activation and functional consequence in IBC remains unknown. Our previous
work identified increased expression of the X-linked inhibitor of apoptosis protein (XIAP) due to altered translation in IBC, while
other studies have noted a crosstalk between XIAP and NFB. We hypothesized that XIAP drives NFB activation in IBC
promoting therapeutic resistance and tumorigenesis.
Methods: NFB phosphorylation, nuclear translocation, and target gene expression were evaluated in triple-negative SUM149
IBC cells with targeted overexpression or knockdown of XIAP. Using specific point mutants, we assessed the domain and
mechanism of XIAP-mediated NFB activation in IBC. We evaluated proliferation and viability in 2D and 3D culture of SUM149
cells treated with JSH-23, a small molecule inhibitor of NFB nuclear translocation. We monitored the effects of XIAP
overexpression or knockdown on in vivo tumorigenicity in IBC xenograft models by measuring tumor growth and NFB signaling.
IHC analysis of XIAP and NFB was performed on tumor microarrays containing both non-IBC and IBC.
Results: Knockdown of XIAP significantly decreased NFB activation in IBC cells. Domain analysis revealed the necessity of the
BIR1 domain of XIAP and TAB1:IKK complex formation in activating NFB. NFB antagonism inhibited proliferation of cells and
sensitized therapy-resistant, XIAP overexpressing cells to targeted therapy. Loss of XIAP inhibited tumor growth of SUM149
tumor cells, correlating with decreased ALDH activity and varied epithelial-mesenchymal characteristics in these cells, while
overexpression of XIAP significantly enhanced tumor growth of SUM149 cells. Further analysis revealed altered SMAD7
expression in XIAP knockdown cells, revealing crosstalk between XIAP, NFB, and TGF signaling in IBC. IHC analysis of XIAP
expression in invasive non-IBC tumors correlated with triple-negative status as well as increased grade and stage of tumors. In
IBC tumors, XIAP expression associated with increased NFB.
Conclusions: In summary, our studies reveal that XIAP expression is necessary for NFB activation in IBC and is critical for IBC
development and progression. This study provides a novel insight into how an anti-apoptotic protein may regulate survival
signaling and disease progression and may guide further research into innovative inhibitors of this interaction.

Title: CCR5 antagonists suppresses the migration and invasion of human inflammatory breast cancer cells
Zhaomei Mu1, Xuanmao Jiao1, Richard G Pestell1 and Massimo Cristofanilli1. 1Kimmel Cancer Center, Thomas Jefferson
University, Philadelphia, PA and 2Inflammatory Breast Cancer International Consortium.
Body: Background: Inflammatory breast cancer (IBC) is a rare and highly metastatic variant of breast cancer with a poor
prognosis and lower survival rate. The disease has a peculiar pattern of early recurrence to soft tissue, bone and central nervous
system irrespective of molecular subtype. Currently, there are no specific therapeutic options for IBC patients, particularly there
are no therapeutic agents aimed at controlling metastatic spread. Chemokine CCL5 and its receptor CCR5 play a significant role
in breast cancer progression and metastasis. We have recently shown that CCR5 promotes breast cancer invasiveness and
metastatic potential, and CCR5 inhibition by the FDA approved CCR5 antagonists Maraviroc and Vicriviroc reduced in vivo
metastasis in a basal-like breast cancer model (Cancer Res 2012;72:3839-3850). Here, we examined the expression of CCR5, as
well as the effects of CCR5 inhibition on the migration and invasion in vitro in human IBC cells.
Methods: IBC cell lines SUM149, SUM190, and FC-IBC-02 derived from pleural effusion fluid of an IBC patient were examined
for the CCR5 expression by immunofluorescence staining. The effects of CCR5 antagonists (Maraviroc and Vicriviroc) on cell
migration and invasion in vitro were examined using matrigel-coated Boyden chamber assay in FC-IBC-02 cells, which are
triple-negative, basal-like, cancer stem cell phenotype, and rapidly developed primary tumors and metastasis in vivo (Breast
Cancer Res Treat 2013;140:2333). Cells were treated by Maraviroc and Vicriviroc at 100 M.
Results: CCR5 was expressed at low positive component comprising only about 5-7 % of the total cell population in SUM 149
cells. CCR5 expression was not detected in SUM190 cells. CCR5 was expressed at significantly higher levels with a higher
percentage of positive population (50-60%) in FC-IBC-02 cells compared with SUM149 cells. FC-IBC-02 cells have a relatively
high percentage of CCR5 positive cells in suspension culture. CCR5 inhibition by both CCR5 antagonists Maraviroc and Vicriviroc
significantly inhibited the migration and invasion of IBC cells in vitro. Maraviroc and Vicriviroc demonstrated effective agents in
controlling migrations (55% and 60% relative inhibition).
Conclusions: CCR5 is highly expressed in human IBC cells. CCR5 antagonists demonstrated the inhibition of migration and
invasion of IBC cells. Further studies are warranted to determine the effects of CCR5 antagonists in combination with standard
treatment in the ability to control IBC progression and metastasis. These results may suggest a potential antimetastatic role for
CCR5-inhibitors in IBC.

Title: High miR-19a serum levels correlate with favorable prognosis in patients with metastatic HER2+ inflammatory breast cancer
and may result from an effective antibody-dependent cell-mediated cytotoxicity induced by trastuzumab
Simone Anfossi1,2, Antonio Giordano1,2, Lei Huo1,2, Ricardo H Alvarez1,2, Vicente Valero1,2, Gabriel N Hortobagyi1, Wendy A
Woodward1,2, Naoto T Ueno1,2, George A Calin1 and James M Reuben1,2. 1University of Texas MD Anderson Cancer Center,
Houston, TX and 2Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, Houston, TX.
Body: Background: IBC is a rare but highly aggressive form of locally advanced breast cancer (5-year OS rate: 40.5% IBC vs
85% non-IBC patients) accounting for 10% of all breast cancer deaths. To date, no unique molecular diagnostic or prognostic
biomarker has been identified for IBC. Increasing evidence supports the potential value of miRNA as prognostic and predictive
serum biomarker in cancer. We found that IBC cells expressed high levels of miR-19a and patients with metastatic IBC HER2+
(MIBC HER2+) and high miR-19a serum levels had better prognosis than patients with MIBC HER2+ and low miR-19a serum
levels. As one of the mechanisms of action of trastuzumab is the induction of antibody-dependent cell-mediated cytotoxicity
(ADCC), we hypothesized that the increased miR-19a serum levels in MIBC HER2+ patients with favorable clinical outcome could
result from an effective ADCC and be used as biomarker to monitor the response to trastuzumab.
Methods: Total RNA was isolated using the Total RNA Purification Kit (Invitrogen, Norgen Biotek). MiR-19a levels in tumor
tissue, serum, cell lines and supernatants were evaluated by qRT-PCR (Applied Biosystems). ADCC was evaluated by Annexin
V-FITC Apoptosis Detection Kit I (BD Pharmingen).
Results: Microarray was performed in IBC (n=23) and non-IBC (n=24) tumors and normal tissue (n=12). Microarray showed
higher miR-19a expression in IBC compared with non-IBC (p=0.028) and normal tissue (p=0.0002). The two IBC cell lines
SUM149 (triple receptor-negative) and KPL-4 (HER2-amplified) expressed higher levels of miR-19a compared with the non-IBC
cell lines MDA-231 (triple receptor-negative), SKBR3 (HER2-amplified), and MCF-7 (HER2-non-amplified) (p<0.001, p<0.001,
p<0.001 and p<0.05, p<0.05, p<0.01 respectively). To assess whether miR-19a could be released from IBC cells upon NK
cell-mediated ADCC, we performed a NK cytotoxicity test using the NK-resistant KPL-4 cells and the NK-sensitive SKBR3 and
the NK-sensitive MCF-7 cells as control. Co-incubation with trastuzumab induced increased MCF-7, SKBR3 and KPL-4 cell death
(2.1-fold, 2.4-fold and 3.5-fold respectively) and accordingly increased miR-19a levels in their supernatants (MCF-7: 3-fold,
p=0.017; SKBR3: 6-fold, p=0.005 and KPL-4: 8-fold, p=0.0001). The pattern of miR-19a levels in the supernatants correlated with
that expressed at cellular levels (MCF-7<SKBR3<KPL-4). We measured miR-19a serum levels in patients with MIBC HER2+
(n=27) and metastatic nonIBC HER2+ (n=24). Patients with MIBC HER2+ and high miR-19a level had longer PFS (10.3 vs 3.2
months, p=0.022) and OS (median not reached vs 11.2 months, p=0.003) than patients with low miR-19a levels. Patients with
metastatic nonIBC HER2+ and high miR-19a levels had longer OS (32.9 vs 13.3 months; p=0.015) than patients with low
miR-19a levels (32.9 vs 13.3 months; p=0.015).
Conclusion: High miR-19a serum levels are associated with favorable prognosis in patients with MIBC HER2+ and could result
from an effective NK cell-mediated ADCC. MiR-19a may represent a novel serum biomarker to monitor the response to
trastuzumab therapy and predict clinical outcome in patients with MIBC HER2+.

Title: Risk factors for developing inflammatory breast cancer: Unique trends among a single patient population
Randie E White1, Laura E Warren2, Jennifer R Bellon2, Faina Nakhlis2, Heather A Jacene1, Eren D Yeh1, Judith Hirshfield-Bartek1,
Beth Overmoyer1 and Inflammatory Breast Cancer International Consortium3. 1Dana-Farber Cancer Institute, Boston, MA;
2
Brigham and Women's Hospital, Boston, MA and 3Inflammatory Breast Cancer International Consortium.
Body: Introduction: Inflammatory breast cancer (IBC) is a virulent form of breast cancer, characterized by skin erythema and
edema associated with enlargement of the breast rapidly occurring within 3-6 months (mo). Because of its rarity (<5% incidence),
there is a paucity of data identifying epidemiologic risk factors, which may shed light on the potential causes of IBC, and guide
prevention strategies.
Methods: This study utilized retrospective data obtained from an IRB approved database of 275 patients (pts) with IBC evaluated
at Dana Farber Cancer Institute (DFCI) from 1997-2012. Pts with confirmed invasive breast cancer had documented clinical
characteristics of IBC clinically staged as T4d. The statistical software JMP 10 was used to perform statistical tests. Chi Square
Tests, Fishers Exact Tests, and descriptive statistics were compiled.
Results: The mean age of diagnosis among our study population was 50.2 years (yrs). IBC pts were more frequently diagnosed
when premenopausal (55%) versus (v) postmenopausal (45%); 25% of pts had metastases upon presentation. The majority of
patients (77%) were overweight (BMI 25-29.9) or obese (BMI30). More premenopausal pts (80.3%) had a BMI >25 v
postmenopausal pts (73.5%). We observed no association with BRCA status among those undergoing genetic testing (13%
BRCA positive (pos); 54 tested); however 52% of pts had a family history of breast cancer (5% BRCA pos). The majority of pts
(81%) did not undergo genetic testing. The most common IBC subtype was HER2 pos (40%); 19% were triple negative (neg), and
16% were hormone receptor (HR) pos/HER2 neg. We also observed a trend of a longer duration of symptoms prior to diagnosis
among younger pts. The mean age of pts who experienced >6 months (mo) of symptoms prior to diagnosis was 42 yrs v the
mean age of 51 yrs among those experiencing < 2 mo of symptoms prior to diagnosis. Pts also displayed a temporal trend in
diagnosis dependent upon time of year. 57% of pts were diagnosed with IBC during the warmer temperatures (Mar 21 Sept 20)
compared with 43% diagnosed during the cooler temperatures (Sept 21 Mar 20).
Conclusion: This retrospective epidemiologic analysis demonstrated various trends in a single population of IBC pts. The
association of high BMI and risk of developing IBC among premenopausal women contrasts with that seen in the non-IBC group,
i.e. high BMI is a risk factor for non-IBC only among postmenopausal women. Targeting the obesity crisis may be a means of
reducing the risk of developing IBC among younger women. A differentiating feature of IBC is the rapid onset of signs and
symptoms of IBC, and yet, younger pts had a significant delay in diagnosis of >6 mo compared with older pts. This emphasizes
the urgency to educate both pts and providers about IBC and facilitate rapid diagnostic procedures. The seasonal relationship
with diagnosis observed in this cohort of IBC pts is intriguing. Investigators have hypothesized an infectious etiology contributing
to the development of IBC, namely infection by viruses or bacterial pathogens may play a role in the pathophysiology of IBC.
These unique trends seen in the DFCI IBC population deserve further investigation.

Title: Clinical outcomes of triple negative inflammatory breast cancer treated with contemporary anthracycline and taxane
preoperative therapy support further investigation of therapeutic targets
Beth Overmoyer1, Hao Guo1, Laura E Warren2, Jennifer R Bellon2, Kornelia Polyak1, Faina Nakhlis2, Judith Hirshfield-Bartek1,
Heather Jacene1, Eren D Yeh1, Meredith Regan1 and Inflammatory Breast Cancer International Consortium3. 1Dana-Farber
Cancer Institute, Boston, MA; 2Brigham and Women's Hospital, Boston, MA and 3Inflammatory Breast Cancer International
Consortium.
Body: Background: Approximately 40-50% of inflammatory breast cancer (IBC) is triple negative (TN), defined as estrogen and
progesterone receptor and HER2 negative. The poor prognosis associated with TN- IBC lends itself to active investigation of
novel therapies to improve outcome. Dr. Kornelia Polyaks laboratory has demonstrated a significant association of JAK2/STAT3
pathway activity in TN-IBC (Overmoyer, Cancer Res 2012). In preparation of designing a clinical trial investigating the effect of
JAK2 inhibition by ruxolitinib on biologic parameters and subsequent use in neoadjuvant chemotherapy (NAC) for TN-IBC, we
determined historical outcomes resulting from a single institutions contemporary standard treatment of TN-IBC.
Methods: Among the 273 pts enrolled in the IRB approved IBC database at the Dana Farber Cancer Institute, 28 pts were
identified with Stage III (T4d,NX,M0)TN-IBC diagnosed from 1/1/1999 to 12/31/2011 who were treated with standard NAC
including anthracycline, cyclophosphamide and taxane. Time to treatment failure (TTF) was defined from diagnosis (dx) to first
progression or recurrence; time to distant metastasis (TDM) was defined from dx to first metastasis at a distant site; overall
survival was defined from dx to death from any cause. Subsequent to NAC, 25 pts underwent modified radical mastectomy
(MRM) and radiation. For those 25 pts, disease-free survival (DFS) was defined from MRM to first recurrence or death; time to
local/regional recurrence (LRR) was defined from MRM with death as competing risk. All time-to event endpoints were censored
at the date last known alive if an event was not observed.
Results: Among 28 patients, the median TTF was 19 months (mo) with 67% free from progression/recurrence at 1 year (yr) after
dx (95% CI, 51-87%). Median TDM was 20 mo with 78% free from distant metastases at 1 yr (CI 64-95%). Most patients (13/21)
had multiple sites of first distant metastasis including: lung (7), contralateral axilla (7), bone (6), liver (3) and CNS (3). Median OS
was 34 mo since dx (Table). Among 25 patients who underwent MRM, median DFS was 15 mo with 1-yr DFS of 58% (42-82%);
the cumulative probability of LRR was 13% and 33% at 1 and 2 yr.
table1
OS probability
95% CI
1-year
96%
89%-100%
2-year
73%
57%-92%
3-year
49%
32%-75%
Conclusions: This retrospective analysis of clinical outcomes of Stage III TN-IBC treated with contemporary anthracycline/taxane
regimes is consistent with previously reported outcomes using historical NAC regimens. (Li, et al, the Oncologist 2012). These
dismal rates of 49% 3-year OS from diagnosis and 58% 1-yr DFS following MRM demand more active investigation into novel
targeting agents which can be combined with standard NAC specifically for the treatment of TN-IBC; a disease that has no known
therapeutic target. For this reason DFCI is actively investigating the role of inhibiting the JAK2/STAT3 pathway using ruxolitinib in
conjunction with standard weekly paclitaxel followed by AC as NAC for TN-IBC. Clinical trial information: NCT01796197.

Title: No Show

Title: A comparative analysis of primary tumor resection in men and women with stage IV breast cancer
Nasreen A Vohra1, Swapnil D Kachare1, Timothy L Fitzgerald1, Jan H Wong1 and Mahvish Muzaffar1. 1East Carolina University,
Brody School of Medicine, Greenville, NC.
Body: Background: Primary tumor resection (PTR) for metastatic female breast cancer continues to be debated. Given the rarity
of male breast cancer, treatment paradigms for female breast cancer are extended to the management of male breast cancer.
Whether the role of PTR in men with metastatic breast cancer is similar to that in women remains unclear. We sought to compare
these two populations using a large, national database.
Methods:
All patients with Stage IV breast cancer between the years 1988-2011 in the SEER database were identified. Uni and multivariate
descriptive and survival analyses were performed.
Results: A total of 41,601 patients with stage IV breast cancer were identified; 98.9% (n=41,162) females, 1.1% (n=439) males.
On average, female patients were younger (63 vs. 66y) and more often White (78 vs. 74%), p0.02. Tumors in male patients
were more likely to be hormonally positive, with varying breakdown of T and N-stages and histologic subtypes as compared to
tumors in females, p0.05. Males were more likely to undergo PTR (51 vs.40%, p<0.05), however both males and females had
similar rates of radiotherapy (35 vs. 32%, p=0.35). Among male patients, those who received PTR were of similar age to those
who did not receive PTR (p=0.64), but had a greater representation of White patients (p=0.04). There were differences in T-stage,
N-stage and hormonal status between men who did and did not receive PTR, p0.05. Men receiving surgery were also more
likely to receive radiation therapy (38 vs. 32%, p=0.003). In women, all demographic and tumor-related factors were significantly
different between those who did and did not undergo PTR. On univariate analysis, surgery was associated with improved
disease-specific median survival in both men (36 vs. 21 mths) and women (34 vs. 18 mths), p<0.05. Younger age, White race,
lower T and N-stage, lower grade, hormonal positivity, mucinous histology, and radiation therapy were associated with improved
disease specific survival (DSS) in females, while only lower T-stage, hormone positivity and mucinous histology were associated
with improved DSS in men. On multivariate analysis, a lack of resection of the primary tumor remained independently associated
with increased mortality in men (HR 1.91, p<0.05) and women (HR 1.6, p<0.05). Over the study period there was a decrease in
the rate of surgery in both men and women, p0.0006, but only women were found to have a statistically significant improvement
in DSS with surgery over time.
Conclusion: Regardless of gender, patients with metastatic breast cancer who underwent primary tumor resection had a
significant improvement in DSS. Factors associated with DSS varied between male and female patients, but the reasons for this
difference are unclear. A well designed randomized trial including both genders will help determine the utility of PTR in stage IV
Characteristics of Stage IV Breast Cancer in Males: Surgery vs. No surgery (S = p.05)
Surgery (%)
No surgery (%)
White race
78.4
68.7
T1 stage
15.8
7.4
N1 stage
27.5
38.6
ER +
72.5
57.6
PR +
57.2
45.2
Her2 +
1.4
3.7
0.25
Grade II
34.7
26.3
Infiltrating ductal carcinoma
76.6
56.2
Radiation therapy
37.8
32.3
Year of diagnosis
P-value
1988-1992
8.1
7.4
2008-2011
25.2
36.9

Title: The microRNA miR-141 is a key regulator of brain metastasis from breast cancer
Bisrat G Debeb1, Lara Lacerda1, Simone Anfossi1, Parmeswaran Diagaradjane1, Khoi Chu1, Lei Huo1, Caimiao Wei1, Richard A
Larson1, Adam R Wolfe1, Wei Xu1, Daniel L Smith1, Li Li1, Cristina Ivan1, Pamela K Allen1, Xiang H Zhang2, George A Calin1,
Savitri Krishnamurthy1, Naoto T Ueno1, Thomas A Buchholz1, James M Reuben1 and Wendy A Woodward1. 1University of Texas
MD Anderson Cancer Center, Houston, TX and 2Baylor College of Medicine, Houston, TX.
Body: Purpose: Brain metastasis poses a major treatment challenge and remains an unmet clinical need. Finding novel
therapies to prevent and treat brain metastases requires an understanding of the biology and molecular basis of the process,
which currently is constrained by a dearth of experimental models and specific therapeutic targets. The purpose of this study was
to develop preclinical models and identify molecular mediators of brain metastasis from breast cancer.
Methods: We used MDA-MB-IBC3 (ER-/HER2+), SUM149 (ER-/HER2-), MCF7 (ER+/HER2-), SUM159 (ER-/HER2-) and
MDA-MB-231 (ER-/HER2-) cell lines for this study. GFP-labeled cells were injected via tail vein into SCID/beige mice and
metastatic colonization to the brain and lung evaluated by fluorescent stereomicroscope and histology 8-weeks after injection.
miRNA microarray was performed with miRNA 3.0 Array. Stable knockdown of miR-141 was achieved with the lentiviral miRZip
system. MiR-141 serum levels in 105 breast cancer patients were measured using quantitative PCR.
Results: We developed novel brain metastasis models in which tail-vein injection of both triple-negative and a
HER2-overexpressing inflammatory breast cancer lines led to a high rate of brain metastases (67%) in SCID/Beige mice
(SUM149, 6 of 9 mice; MDA-MB-IBC3, 10 of 15 mice). Sub-lines derived from lung or brain metastases in these models were
morphologically and molecularly distinct. The brain metastasis-derived sublines showed epithelial morphology and overexpressed
epithelial markers and miR-141 while sublines from lung metastases showed mesenchymal morphology and overexpressed
mesenchymal markers. Knockdown of miR-141 significantly inhibited metastatic colonization to the brain compared to controls
(miR-141 knockdown vs. control: SUM149, 0 of 8 mice vs. 6 of 9 mice, p=0.009; MDA-MB-IBC3, 2 of 14 mice vs. 10 of 15 mice,
p=0.007) but it did not affect colonization to the lung. Importantly, ectopic expression of miR-141 in non-expressing MDA-MB-231
significantly enhanced brain metastatic colonization (5 of 9 mice vs. 0 of 10 mice, P=0.02). On multivariate analyses high serum
level of miR-141 was an independent predictor of progression free survival [HR 4.8 (95%CI, 2.6-8.7), P<0.001] and overall
survival [HR 7.0 (95%CI 3.5-15.1), P<0.001] in patients with metastatic breast cancer.
Conclusion: We demonstrated high rates of brain metastases from a heterogeneous group of cell lines that have not previously
been associated with brain metastases, demonstrated miR-141 as a key regulator of brain metastasis and provided clinical
evidence supporting the prognostic relevance of miR-141. We propose that miR-141 should be examined as a biomarker and
potential target in the prevention and treatment of brain metastases from breast cancer.

Title: Oncolytic viral therapy enhances the survival of mice in a novel model of breast cancer brain metastases
W Hans Meisen1, Samuel Dubin1, Steven Sizemore1, Haritha Mathsyaraja1, Katie Thies1, Norm Lehman1, Peter Boyer1, Alena C
Jaime-Ramirez1, J Bradley Elder1, Kimerly Powell1, Michael Ostrowski1 and Balveen Kaur1. 1James Comprehensive Cancer
Center The Ohio State University Medical Center, Columbus, OH.
Body: Breast cancer [BC] is one of the leading causes of brain metastases. The 2 year survival rate of patients with breast
cancer brain metastases [BM] is less than 2%. Oncolytic viruses exploit the aberrant molecular and genetic pathways found in
cancer cells to selectively replicate in and destroy tumors while sparing normal tissues. Here, we demonstrate the oncolytic
Herpes Simplex Virus [HSV-1], 34.5ENVE, can specifically target and destroy BC brain metastases. The 34.5ENVE virus
expresses anti-angiogenic Vstat120 and its replication is transcriptionally driven by the cancer specific promoter Nestin. Vstat120
expression is reduced in brain, renal, and gastric cancers, however its expression status in BC is not known. Analysis of The
Cancer Genome Atlas revealed a 52% reduction in Vstat120 expression in invasive ductal breast carcinomas [n=69] compared to
normal breast tissue [n=389; P<0.0001]. Reduced Vstat120 expression was also associated with decreased disease free survival
in BC patients [n=324; P<0.03]. An examination of Vstat120 expression in 50 breast cancer cell lines from the Neve et al dataset
showed Vstat120 mRNA levels were reduced in 38% of BC cell lines compared to the MCF-10A epithelial cell line [19 of 50 cell
lines]. These analyses suggested BC patients may benefit from the re-expression of anti-angiogenic Vstat120. Nestin is
up-regulated in several metastatic cancers, and its expression correlates with reduced survival in BC patients. In a cohort of 166
patients stratified by median Nestin expression, we observed Nestin to be significantly associated with an increased incidence of
brain and lung metastases [n=164; P<0.02]. Additional analysis of the Neve et al. microarray dataset showed Nestin was
upregulated in 100% of the BC cell lines examined [50 of 50]. These results suggest that Nestin expression may be a strong
therapeutic target for BC. 34.5ENVE was cytotoxic to human BC cells of varying subtypes in vitro including the HER2+ and triple
negative BCs known to frequently metastasize to the brain. Since 34.5ENVE replication is driven by a Nestin promoter, we
compared the cytotoxicity of 34.5ENVE with a similar virus lacking Nestin driven ICP34.5 expression. We observed a 54.14% and
24.44% increase in killing in the MDA-MB-468 and MDA-MB-231 cell lines in the Nestin-driven 34.5ENVE virus as compared to a
virus lacking ICP34.5, respectively [P<0.001]. This is the first study to specifically use Nestin expression to target BC. To test the
therapeutic efficacy of 34.5ENVE against BM in vivo, we created a novel, immune competent BC BM model using Met-1 and
DB-7 murine BC cell lines. Intracranial implantation of these cells resulted in tumors which recapitulated the human BM tumor
biology. Treatment of mice with established Met-1 BM tumors with a single, intratumoral dose of 34.5ENVE resulted in significant
tumor regression [via MRI] and increased survival. Similarly, mice bearing intracranial DB-7 tumors treated with a single dose of
34.5ENVE showed a doubling of median survival compared to control treated mice [median survival 17 days vs 34 days,
respectively; P<0.0004]. The results of these studies warrant further investigation of oncolytic 34.5ENVE viral therapy to treat
established BC brain metastases.

Title: Intrinsic subtypes and MRI patterns in brain metastasis associated with breast cancer
Nicole Williams1, Vinay Varadan2, Aditi Vadodkar3, Kristy Miskimen3, Stephanie Kim1, Shaveta Vinayak1, Paula Silverman1, Jill
Barnholtz-Sloan2, Andrew Sloan1, Cheryl Thompson2, Lisa Rogers1, Hannah Gilmore1 and Lyndsay Harris1. 1University
Hospitals/Case Western Reserve, Cleveland, OH; 2Case Comprehensive Cancer Center, Cleveland, OH and 3Case Western
Reserve University, Cleveland, OH.
Body: Background: Breast cancer is the 2nd most common cancer to metastasize to the brain. The development of brain
metastasis (BM) is associated with a lower median survival compared with other locations of metastasis and carries with it high
morbidity and reduced quality of life. There are limited treatment options and virtually no approved targeted therapies for this
disease. We have previously reported specific pathways enriched in BM compared to primary tumors and now further this study
to examine pathways by intrinsic subtype and location of and number of BM.
Methods: Archival FFPE material was obtained from BM using pathology records; clinical data, including MRI images, was
retrieved from medical records and institutional tumor registry under an IRB protocol. Tumor DNA/RNA was extracted from 2 mm
cores macrodissected from the FFPE tissue blocks using the Qiagen AllPrep DNA/RNA FFPE kit. Gene expression profiling was
performed using Affymetrix Human Transcriptome Array 2.0 microarray on BM samples for which sufficient RNA was available.
Gene-level expression quantification was derived after RMA normalization using the Affymetrix Transcriptome Analysis Console.
PAM50 subtypes were assigned by clustering samples using median-subtracted PAM50 gene expression levels. Differential gene
expression was estimated using the non-parametric Mann-Whitney test, followed by assessment of false discovery rate using the
Banjamini-Hochberg FDR methodology. Pathway enrichment analysis was performed using the NCI Pathway Interaction
Database.
Results: Gene expression profiling showed the following intrinsic subtype distribution among all BM: luminal A 32% (19/59),
luminal B 31% (18/59), HER2 enriched 7% (4/59), and basal subtype 31% (18/59). At time of development of BM 64% (38/59) of
patients presented with a single lesion compared to 36% (21/59) of patients who presented with multiple lesions (p=0.12).
Thirty-nine percent (7/18) of patients with basal subtypes were observed to present with multiple BM compared to 61% (11/18)
non-basal subtypes (p=0.25). In addition, 12% (13/59) of BM were found to be exclusively dural-based lesions. They appeared
more frequently in the luminal subtypes [11/13 vs 2/13; p=0.06]. A total of 314 genes were differentially expressed (Wilcox pval <
0.05; FDR < 0.01) between the basal and luminal subtypes. As expected, we found that the FOXA transcription factor network
was up-regulated in luminal when compared to the basal subtype, whereas the FOXM1 transcription factor network was
up-regulated in the basals. We also found a total of 28 genes to be significantly differentially expressed (Wilcox pval < 0.05; FDR
< 0.01) between the dural and non-dural BM. The beta1 integrin and syndecan-1 pathways were significantly enriched, along with
angiogenesis and lymphatic endothelium pathways. Key genes in these pathways (COL1A1, COL1A2, COL3A1, CDH11, were
found to be at least 2-fold up-regulated in the dural BM compared to non-dural BM.
Conclusion: Identifying pathways that are differentially expressed between intrinsic subtypes may help us develop new targeted
therapies to provide more treatment options for breast cancer patients with brain metastasis.

Title: Phase 1/2a study of glutathione PEGylated liposomal doxorubicin (2B3-101) in breast cancer patients with brain
metastases
Philippe G Aftimos1, Bojana Milojkovic-Kerklaan2, Vronique Diras3, Sevilay Altintas4, Carey Anders5, Monica Arnedos6, Hans
Gelderblom7, Patricia Soetekouw8, Werner Gladdines9, Pieter Gaillard9, Carlos de Sousa9, Agnes Jager10, Myra van Linde11,
Ahmad Awada1, Jan Schellens2 and Dieta Brandsma12. 1Institut Jules Bordet, Universit Libre de Bruxelles, Medical Oncology
Clinic, Brussels, Belgium; 2Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland,
Netherlands; 3Institute Curie,, Paris, France; 4Antwerp University Hospital, Antwerp, Belgium; 5Lineberger Comprehensive Cancer
Center, University of North Carolina, Chapel Hill, NC; 6Gustave Roussy Cancer Institute, Villejuif, France; 7Leiden University
Medical Center, Leiden, Netherlands; 8Maastricht University Medical Center,, Maastricht, Netherlands; 9to-BBB technologies B.V.,
Leiden, Netherlands; 10Erasmus MC Daniel den Hoed Cancer Center, Rotterdam, Netherlands; 11VU Medical Center, Amsterdam,
Noord-Holland, Netherlands and 12Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, Noord-Holland,
Netherlands.
Body: Background:
The incidence of brain metastases (BM) in breast cancer (BC) patients (pts) has increased over the past decade. The brain is
regarded as a sanctuary site for metastatic cells which are partially protected from drugs by the blood-brain barrier (BBB).
2B3-101 is a doxorubicin (DOX) liposomal formulation that uses glutathione transporters on the BBB to penetrate the brain.
Non-clinical studies have shown a 5-fold enhanced delivery of DOX to the brain after IV administration of 2B3-101 compared to
liposomal DOX, without signs of cardio- or neurotoxicity.
Methods:
This phase 1/2a open label study assessed the safety, tolerability, MTD, PK, and anti-tumor activity of single agent 2B3-101 in pts
with BM of solid tumors, or high-grade gliomas.
For this analyses BCBM pts (n=25) were included. These pts received 2B3-101 at a starting dose of either 40 (n=3) or 50 (n=22)
mg/m2 IV every 3 weeks, until disease progression or unacceptable toxicity. Anti-tumor activity was assessed by RECIST 1.1.
Patients with HER2-positive BCBM were treated with concurrent trastuzumab.
Results:
As of May 30, 2014, 88 cycles (median 2, range 1-10) of 2B3-101 alone or with trastuzumab were administered to 25 heavily
pretreated BC pts, 3 pts are still on treatment. Median age was 47 (3361) years and 18 (72%) pts had HER2+ disease. Pts had
received a median of 7 (415) prior regimens; 18 (72%) had received prior radiation therapy to the brain.
In phase 1, 2B3-101 alone or with trastuzumab was well tolerated up to a dose intensity of 15 mg/m2/wk. Cycle 1 MTD was not
reached. Phase 2a dose of 50 mg/m2 was selected based upon tolerability after repeated dosing. The most frequent reported
treatment emergent AEs are qualitatively consistent with conventional PEGylated liposomal DOX and were ( grade 2):
neutropenia (59%), palmar plantar erythrodysesthesia (PPE) (50%), fatigue (45%), stomatitis (22%), and infusion reaction (18%).
Notable Grade 34 AEs were neutropenia (35%) and PPE (13%). All AEs were transient and manageable with dose delays,
reductions and standard medication. 57% of pts had 2B3-101 dose delays and 39% of pts required dose reductions. 2B3-101
showed no neuro- or cardiotoxicity. PK data showed non-linear exposure of 2B3-101 without signs of accumulation upon repeat
dosing, a mean half-life of 69h (range 43-120h) and independent of trastuzumab co-treatment.
Best overall and intracranial tumor responses of 2B3-101 in 23 evaluable BCBM pts (92% of total) are summarized in Table 1.
Table 1: Anti-tumor activity results of 2B3-101 in BCBM patients
Population
Number of evaluable Overall partial

pts
response
Overall stable
disease
Intracranial tumor response of 12-weeks overall PFS

20%.
rate
Overall
23
2 (9%)
11 (48%)
4 (17%)
48%
"Luminal"
0 (0%)
2 (50%)
1 (25%)
50%
16
2 (13%)
9 (56%)
3 (19%)
56%
HER2
positive
TNBC
0 (0%)
0 (0%)
0 (0%)
0%
Conclusions:
2B3-101 alone or with trastuzumab is safe and well tolerated and shows intra- and extracranial anti-tumor activity in heavily
pretreated BC pts. A 12-week PFS rate of 56% in HER2+ BCBM was observed, which warrants further clinical studies. An
international, multicenter, randomized, controlled phase IIb study in HER2+ BCBM is being planned.
NCT01386580, sponsored by to-BBB technologies BV.

Title: New graded prognostic assessment (GPA) in breast cancer brain metastases in a contemporary cohort
Manmeet S Ahluwalia1, Ming Chi1, Vyshak Alva Venur1, Thomas Budd1, Lilyana Angelov1, Samuel Chao1, Paul Elson1, Gene H
Barnett1 and Jame Abraham1. 1Cleveland Clinic, Cleveland, OH.
Body: Background:
Breast cancer is the second most common cause of brain metastases (BCBM). We evaluated prognostic factors for overall
survival (OS) in contemporary cohort of BCBM patients treated at a tertiary care institution.
Methods:
With IRB approval, Cleveland Clinics database was used to identify BCBM patients treated between 2000 and 2013. OS from the
diagnosis of BCBM was the primary end point. Cox proportional hazards models with stepwise variable selection were used for
data analysis.
Results:
562 female patients were included for this analysis. Karnofsky performance scale (KPS) was 90-100 in 204 patients (41%), 70-80
in 223 (44%) and <70 in 76 (15%) patients. Two hundred, eighty nine (52%) patients were treated with whole brain radiation
therapy (WBRT), 89 (16%) with stereotactic radiosurgery (SRS), 66 (12%) with WBRT and SRS, 44 (8%) underwent surgery and
WBRT, 16 (3%) underwent surgery and SRS. Median OS for the entire cohort was 11.2 months (95% C.I., 10.0-13.1). Median OS
in Luminal B (triple positive), Her 2 positive, Luminal A (ER/PR positive, Her 2 negative) and basal (triple negative) patients was
23.3, 14.5, 10.0 and 8.3 months respectively. Disease specific graded prognostic assessment (GPA) for BCBM is based on KPS,
age at diagnosis, and subtype. Overall it was prognostic for OS (p <.0001), however separation between groups was variable.
Starting de-novo, KPS, age at diagnosis of BCBM, and subtype were again identified as prognostic, although the scoring was
different. Additional independent predictors were the number of sites of extracranial metastases, leptomeningeal disease,
control of primary cancer and location of BCBM were found to be independently prognostic for survival. A new GPA was
formulated by assigning "points" (weights) to the levels within each of these seven factors that are proportional to the regression
coefficient estimates in the final model and then adding the total number of points present.
Prognostic factors in BCBM on multivariate analysis
Factor
Points
KPS
80-100
13
<70
Hazard Ratio
0.41 (0.32-0.52)
<0.001
Breast Cancer Subtype ( basal as reference)

Luminal B
10
0.49 (0.32-0.74)
<0.001
HER2
0.58 (0.43-0.77)
0.007
Luminal A
0.63 (0.48-0.77)
0.002
Basal
0
0.65 (0.56-0.77)
<0.001
0.51 (0.38-0.67)
<0.001
0.68 (0.53-0.88)
0.003
Number of sites of Extra-Cranial Metastases

0 or 1
12
2-3
>3
Leptomeningeal Disease
No
10
Yes
Primary Controlled
Yes
No
Age at Diagnosis of BCBM

<50
>/=50
Location of Brain Metastasis

Infratentorial or Supratentorial
Both
0.75 (0.59-0.96)
0.02
0.76 (0.61-0.96)
0.02
The new GPA consists of four groups: unfavorable, intermediate 1, intermediate 2 and favorable with OS of 2.7, 9.1, 18.5 and
29.5 months respectively.
New GPA and Conventional GPA
New GPA
Group
Number of points
Number
Median OS (months)
Unfavorable
</=21
84 (19%)
2.7
Intermediate-1
22-36
153 (34%)
9.1
Intermediate-2
37-45
133 (30%)
18.5
Favorable
>45
78 (17%)
29.5
Conventional GPA
Unfavorable
0-1
55 (12%)
4.8
Intermediate-1
1.5-2
130 (29%)
9.1
Intermediate-2
2.5-3
148 (33%)
12.9
Favorable
3.5-4
116 (26%)
20.4
Conclusions:
A new GPA for BCBM is proposed.

Title: Impact of early detection of brain metastasis in metastatic breast cancer patients: A single institutional experience
Satomi Matsuo1, Junichiro Watanabe1, Koichi Mitsuya1 and Yoko Nakasu1. 1Shizuoka Cancer Center, Shizuoka, Japan.
Body: Background/Introduction
Since brain metastasis (BM) is commonly seen in metastatic breast cancer (MBC) patients (pts), it may be an important
prognostic factor. It has been believed that screening MRI would not improve a pts outcome, however, an appropriate diagnosis
and early initiation of therapy, especially screening MRI followed by stereotactic irradiation (STI), might contribute to a pts
survival.
We reviewed our medical records for 589 MBC pts who were treated between September 2002 and March 2014. COX regression
analyses were applied to identify the survival risk factors, and the Kaplan-Meier method with a log-rank test was utilized to
evaluate the survival rates.
Results
Upon checking our medical records, 187 pts, or 37.1% of MBC pts, developed BM with the median time to BM of 585.0 days
(95% confidence interval [CI] 501.0-647.0). The tumor subtypes of primary lesion were luminal, 44.9%; luminal-HER2, 14.9%;
HER2+, 21.3%; triple-negative (TN), 16.0% and unavailable, 2.6%. The median overall survival (OS) from the diagnosis of BM
was as follws: all pts, 292.0 days (95%CI 220.0-345.0); luminal, 212.0 days (150.0-293.0); luminal-HER2, 400.0 days
(258.0-613.0); HER2+, 531.0 days (423.0-670.0) and TN, 127.0 days (43.0-185.0). Accompanying visceral and/or bone lesion(s)
at the diagnosis of BM were as follows: lung, 49.7%; liver, 45.9% and bone, 64.7%. The development of BM in the HER2+ pt was
significantly less frequently associated with progression of known visceral/bone lesions than other subtypes (HER2+ 41.0%
versus luminal 64.2% or TN 70.0%; P<0.01, Chi-square test), and the pts who developed BM without progression of other
lesion(s) showed a significantly superior OS (hazard ratio [HR] 0.48, 95%CI 0.32-0.70, P<0.001) regardless of the subtype. The
91 pts with BM (48.6%) whose lesions were detected by screening MRI showed a significantly superior OS from BM than did the
pts with BM with any symptoms (median 359.0 days, 95%CI 249.0-439.0 versus 222.0 days, 95%CI 159.0-292.0; P<0.05). STI
was performed as initial therapy in 20.3% of pts, and improved their OS from BM in all subtypes (median 439.0 days with STI,
293.0 days without STI; P<0.05), STI or the use of a tyrosine kinase inhibitor (TKI) significantly improved the OS from BM in
HER2+ pts (median 661.0 days with STI versus 423.0 days without STI; P<0.05, median 622.0 days with TKI versus 287.0 days
without TKI; P<0.01) The univariate COX analysis indicated that HER2+, 2>ECOG performance status at the diagnosis of BM, no
other progressive lesions at the diagnosis of BM, <5 brain lesions and STI were favorable factors for the OS. When limited to
HER2+ pts, the multivariate COX analysis showed that the lack of other progressive lesions at the diagnosis of BM (HR 0.20,
95%CI 0.09-0.47, P<0.001) and STI (HR 0.18, 95%CI 0.06-0.56, P<0.01) markedly decreased the risk of death for HER2+ pts.
Conclusions
From our institutional review, there seemed to be no special strategy for improving the OS of luminal or TN pts, however, HER2+
pts showed an improved OS after BM following the early detection and appropriate therapies for the BM. We conclude that
HER2+MBC pts are good candidates for a BM screening program.

Title: Predict subsequent brain metastasis in patients with metastatic breast cancer: External validation of a published nomogram
and competing risk regression analysis
Ludivine Genre1, Henri Roch2, Monia Ouali3, Thomas Filleron3 and Florence Dalenc2. 1Claudius Regaud Institute, Toulouse,
France; 2Claudius Regaud Institute, Toulouse, France and 3Claudius Regaud Institut, Toulouse, France.
Body: Backgroud: Brain metastasis (BM) is a fatal event that nontheless alter seriously survival of patients suffering from breast
cancer but also reduce their quality of life. Selection of an enriched patients population at high risk for BM is warranted to
develope preventive strategies and/or to evaluate the impact early treatment of BM in prospective trials.
Methods: Electronic medical records of patients, treated in the Institut Claudius Regaud, with metastatic breast cancer and
without BM at stade IV diagnosis or in the first month, were retrospectively reviewed for the period between January 2005 and
December 2012. We first study the Graesslins nomogram (J Clin Oncol., 2010; 28(12): 2032-37) characteristics in our patients.
The discrimination prediction of subsequent BM was evaluated by the area under the receiver operating characteristic curve
(AUC) and we performed the calibration. Moreover, we have evaluated average and maximal errors between predictions and
observations obtained from the calibration curve. Then, competing risk analysis was used to identify prognostic factors with time
to BM appearance and death before BM.
Results: We identified 446 patients without BM at stage IV diagnosis or in the first month, 70 of them developed subsequent BM.
Young age ( 50 years) at the diagnosis of breast cancer (p=0.01), ductal carcinoma (p=0.02), negative status of hormone
receptor (p<0.0001), HER2 overexpression/ amplification (p<0.0001), grade III of primary tumour (p=0.04) and number of
metastatic sites (p=0.05) were significatively associated with subsequent BM. The external validation of Graesslins nomogram
shows a good discrimination with an AUC of 0.695 [95%CI, 0.61-0.77]. The calibration is correct with an E max = 0.076 and an E
avg = 0.37. Interestingly, in our study the cumulative incidence of BM is 5.48%, 12.95% and 18.15% at 1, 3 and 5 years after the
diagnosis of stade IV. In contrast, the cumulative incidence of death before the diagnosis of BM is 16.08%, 43.97% and 56.40%
respectively. In multivariate analysis, HER2 overexpression/amplification (p<0.0001), triple negative status (p=0.027) and number
of metastasis sites (p=0.037) are associated with BM, while age > 50 years (p=0.015) and triple negative status (p<0.001) are
associated with death before the BM diagnosis.
Conclusion : We have validate the robustness of Graesslins nomogram to predict subsequent BM in patients with metastatic
breast cancer and suggest the best utility in patients with HER2+ breast cancer, looking forward new systemic therapies to control
non BM in triple negative breast carcinoma.

Title: Prognostic factor of HER2-positive breast cancer patients developed brain metastasis: A multicenter retrospective analysis
Naoki Hayashi1, Naoki Niikura2, Norikazu Masuda3, Seiki Takashima4, Rikiya Nakamura5, Ken-Ichi Watanabe6, Chizuko
Kanbayashi7, Mayumi Ishida8, Yasuo Hozumi9, Michiko Tsuneizumi10, Naoto Kondo11, Yoichi Naito12,13, Yayoi Honda14, Akira
Matsui15, Tomomi Fujisawa16, Risa Oshitanai2, Hiroyuki Yasojima3, Hideko Yamauchi1, Shigehira Saji17 and Hiroji Iwata11. 1St
Luke's International Hospital, Tokyo, Japan; 2Tokai University School of Medicine, Kanagawa, Japan; 3Breast Oncology, NHO
Osaka National Hospital, Osaka, Japan; 4National Hospital Organization Shikoku Cancer Center, Japan; 5Chiba Cancer Center
Hospital, Chiba, Japan; 6Hokkaido Cancer Center, Japan; 7Niigata Cancer Center Hospital, Niigata, Japan; 8National Kyushu
Cancer Center, Japan; 9Jichi Medical University, Japan; 10Shizuoka General Hospital, Shizuoka, Japan; 11Aichi Cancer Center,
Nagoya, Japan; 12National Cancer Center Exploratory Oncology Research and Clinical Trial Center, Japan; 13National Cancer
Center Hospital East, Japan; 14Cancer and Infectious Diseases Center, Tokyo Metropolitan Komagome Hospital, Tokyo, Japan;
15
National Hospital Organization, Tokyo Medical Center,, Japan; 16Gunma Prefectural Cancer Center, Ota, Gunma, Japan and
17
Kyoto University Graduate School of Medicine, Japan.
Body: Background:
HER2-positive breast cancer has a high risk of developing brain metastasis compared to other subtypes of breast cancer.
However, the clinical course and prognostic factors of HER2-positive breast cancer patients with brain metastases are not well
known because of the relatively small population. The aim of this study was to determine clinicopathological factors associated
with prognosis of HER2-positive patients developed brain metastasis.
Methods:
A retrospective large dataset of 432 HER2-positive patients who were diagnosed with brain metastases between 2001 and 2012
were collected from 24 institutions of the Japan Clinical Oncology Group: Breast Cancer Study Group. We assessed the
clinicopathological factors associated with prognosis of these populations with brain metastases.
Results:
The median age of the 432 patients was 54 years (range, 2086 years). Of the patients, 162 patients (37.5%) had
ER-positive/HER2-positive (ER+HER2+) breast cancer and 270 patients (62.5%) had ER-negative/HER2-positive (ER-HER2+)
breast cancer. Nineteen of the 162 patients with ER+HER2+ (12%) and 53 of the 270 patients with ER-HER2+ (20%) underwent
surgery for brain metastases. After the diagnosis of brain metastasis, 108 patients with ER+HER2+ (63%) and 175 patients with
ER-HER2+ (64%) received HER2-targeting agents, including trastsuzumab and/or lapatinib.
The median brain metastasis-free survival period from the diagnosis of primary breast cancer was 33.5 month in both subtypes.
In 63.4% of patients with ER+HER2+subtype and 75.6% of patients with ER-HER2+, brain metastases were detected within 2
years after development of first distant metastasis. Eighty-four patients with ER+HER2+ subtype (52%) and 133 patients with
ER-HER2+ (49%) had more than 3 brain metastases at the diagnosis.
The median survival period after developing brain metastasis was 16.5 months (95% confidence interval [CI], 11.921.1 months)
in patients with ER+HER2+ and 11.5 months (95% CI, 9.113.8 months) in patients with ER-HER2+ (p = 0.117).
Patients with more than 3 brain metastases had significantly shorter OS period than patients with equal or less than 3 brain
metastases in both of ER+HER2+ (p < 0.001) and ER-HER2+ (p = 0.018). According to receiving HER2-targeting agents,
patients receiving both of trastsuzumab and lapatinib had significantly longer survival period than patients who had received
trastsuzumab alone, lapatinib alone, or no HER2-targeting agent (p < 0.001).
Conclusions:
Our results showed that HER2-positive patients with more than 3 brain metastases at the diagnosis had poor prognosis
regardless of ER-positivity, and receiving both of trastsuzumab and lapatinib might improve their survival. Further studies are
needed to determine the best treatment strategy including these HER2-targeting agents for these populations.

Title: SPARC expression in brain metastases of breast cancer patients
Isabell Witzel1, Jakob Matschke2, Markus Glatzel1, Karin Milde-Langosch1, Elena Laakmann1, Sibylle Loibl4, Berit M Pfitzner3,
Carsten Denkert3 and Volkmar Mller1. 1University Medical Center, Hamburg, Germany; 2Institute of Neuropathology, University
Medical Center, Hamburg, Germany; 3Institute of Pathology, University Medicine of Berlin, Berlin, Germany and 4German Breast
Group, Neu-Isenburg, Hessen, Germany.
Body: Background: The incidence of brain metastases in breast cancer patients is rising and has become a major clinical
challenge in the last years with so far limited therapeutic options. Therefore, further insights into the biology of brain tropism are
important. Secreted protein acidic and rich in cysteine (SPARC) is an albumin-binding protein. A differential expression of SPARC
in breast tumor tissue and its surrounding stroma compared to normal tissues has been described and might be associated with
poor prognosis. In this study, tissues of brain metastases were evaluated for SPARC expression by immunohistochemistry using
a standardized immunoreactive score (IRS).
Methods: 138 tissue samples of brain metastases were available for construction of a tissue-microarray (TMA) and evaluation of
SPARC expression. Immunohistochemical staining for SPARC was carried out as previously described (Sinn et al. Ann Onc,
2014) using the antibody Novocastra NCL-O-NECTIN; Clone: 15G12; 1:100). For stromal SPARC expression, the intensity of
staining was evaluated by a four-tier scoring system (negative, weak, moderate and strong). Cytoplasmic SPARC expression was
evaluated by the percentage of tumor cells with cytoplasmatic staining (0 % = "0", 1-10 % = "1", 11-50 % = "2", 51-80 % = "3", 81100 % = "4"). The staining intensity was evaluated as negative (0), weak (1), moderate (2) or strong (3). The numeric values were
multiplied, resulting in an immunoreactivity score (IRS) ranging from 0 to 12. The definition for high cytoplasmic SPARC was IRS
3 (at least weak intensity in > 50 % of tumor cells or at least strong intensity in 1-10 % of tumor cells). Staining of normal brain
tissue and comparison to primary breast tumors is still ongoing.
Results: Cytoplasmic SPARC expression was detectable in 104 cases (78%), 30 cases were negative (22%). Stromal SPARC
intensity was strong in 65 (63%) cases, moderate in 35 (34%) cases and low in 4 cases (4%). There was a weak association
between stromal and cytoplasmic SPARC expression (r=0.22, p=0.010). No significant association between SPARC expression
and clinicopathological parameters could be observed. 43.5% (n=54) of brain metastases were HER2-positive, 39.5% (n=49)
were triple-negative and 16.9 % (n=21) were HR+/HER2 negative. Cytoplasmic SPARC expression was slightly higher in HER2
positive compared to triple-negative brain metastases. Regarding survival analysis, strong stromal SPARC expression was
associated with shorter overall survival from first diagnosis of brain metastases but did not reach statistical significance (HR 4.2,
95%-CI 0.5 34, p=0.18).
Conclusion:
SPARC is frequently expressed in brain metastases of breast cancer patients. It might provide additional prognostic information in
patients with brain metastases.

Title: Activity of T-DM1 in HER2-positive breast cancer brain metastases
Rupert Bartsch1, Anna S Berghoff1, Ursula Vogl2, Margaretha Rudas1, Elisabeth Bergen1, Michael Gnant1, Karin Dieckmann1,
Katja Pinker1, Zsuzsanna Bago-Horvath1, Arik Galid3, Leopold Oehler2, Christoph C Zielinksi1, Guenther G Steger1 and Matthias
Preusser1. 1Medical University of Vienna; 2St Joseph Hospital Vienna and 3Hanusch Hospital Vienna.
Body: Background
Different local therapy options such as radiotherapy, radiosurgery and neurosurgery remain the mainstay of brain metastases
(BM) management, especially in case of oligometastatic or symptomatic disease. Recently, the LANDSCAPE study established
lapatinib plus capecitabine (LapCap) as potential standard option for primary systemic treatment in oligosymptomatic patients
(pts) with multiple Her2-positive BM. Limited evidence exists with regards to the potential activity of antibodies in BM. Due to a
disruption of the blood-brain-barrier at the metastatic site, it is anticipated that even large molecules such as antibodies may
penetrate into the central-nervous system (CNS). On the other hand, the CNS is an immune-privileged organ which may hamper
activity of trastuzumab. This problem may be overcome by the use of T-DM1.
T-DM1 is an antibody-drug conjugate linking trastuzumab (T) to an anti-microtubule agent providing higher activity and lower
toxicity as compared to LapCap. Here, we investigated the activity of T-DM1 in newly diagnosed or progressive BM.
Nine pts (median age 55 years) with Her2-positive BM treated at two Austrian centres were included. All pts had received prior
treatment with T, five pts (55.6%) had already received lapatinib, and two pts (22.2%) pertuzumab as well.
In two asymptomatic pts, T-DM1 was administered as primary therapy, while seven pts had documented CNS progression upon
prior local treatment. T-DM1 was administered every three weeks at a dose of 3.6 mg/kg. Restaging was conducted every twelve
weeks with MRI or whenever symptoms of disease progression occurred.
Results
Median follow-up was 6 months and median brain metastases-free survival 11 months. Seven pts (two with primary treatment and
five receiving T-DM1 upon CNS progression) are currently assessable for CNS response. 3/7 pts (42.9%) had partial remission,
one patient progressing upon prior local therapy had stable disease lasting for 10 months, and one patient had stable disease for
5 month. One patient progressing of prior WBRT had minor response of BM on MRI but no reduction of brain oedema and
increasing cortisol doses and was therefore deemed PD. The other patient with primary PD was also progressing after WBRT.
Conclusion
This prospectively sampled case series again indicates that systemic therapy offers activity in Her2-positive BM. LapCap remains
the standard of care but T-DM1 offers relevant clinical activity and should be investigated within the context of larger clinical
studies.

Title: No Show

Title: Graded prognostic assessment for triple negative breast cancer brain metastases
Ming Chi1, Vyshak Alva Venur1, Jame Abraham1, Thomas G Budd1, Paul Elson1 and Manmeet S Ahluwalia1. 1Taussig Cancer
Institute, Cleveland Clinic, Cleveland, OH.
Body: Background:
Brain metastases is a serious complication of triple negative breast cancer (TBCBM). We evaluated prognostic factors for overall
survival (OS) in contemporary cohort of TBCBM patients treated at a tertiary care institution.
Methods:
With IRB approval, Cleveland Clinics database was used to identify BCBM patients treated between 2000 and 2013. OS from the
diagnosis of TBCBM was the primary end point. Cox proportional hazards models with stepwise variable selection were used for
data analysis.
Results:
One hundred forty three female patients were included for this analysis. Karnofsky performance scale (KPS) was 90-100 in 52
patients (39%), 70-80 in 50 (38%) and <70 in 30 (23%) patients. Sixty eight (49%) patients were treated with whole brain radiation
therapy (WBRT), 26 (19%) with stereotactic radiosurgery (SRS), 18 (13%) with WBRT and SRS, 13 (9%) underwent surgery and
WBRT, 5 (4%) underwent surgery and SRS. Seventy nine percent (113/143) of patients are reported to have died. Overall
survival was 8.3 months (95% C.I. 5.5-10.6). Disease specific graded prognostic assessment (GPA) for breast cancer brain
metastases is based on KPS, age at diagnosis, and subtype. By definition triple negative patients fall into the lowest two GPA
score groups. GPA was prognostic for survival (p=.05), however separation between groups was variable. Starting de-novo, KPS
(but different coding than in the GPA), leptomeningeal disease, and the number of extracranial sites of disease at the time of
diagnosis of TBCBM were found to be independently prognostic for survival (proportional hazards model, stratified by when brain
metastases were diagnosed, and a stepwise selection algorithm). A new GPA was formulated by assigning "points" (weights) to
the levels within each of these three factors that are proportional to the regression coefficient estimates in the final model and
then adding the total number of points present.
Prognostic Factors Impacting Survival of TBCBM
Factor
Number of points
KPS
80-100
</= 70
No. Extra-Cranial Mets

0 or 1
>/= 2
Leptomeningeal Disease
No
Yes
Hazard Ratio
0.31 (0.21-0.48)
<.0001
0.42 (0.28-0.64)
<.0001
0.43 (0.26-0.71)
0.001
The new GPA consists of three groups: unfavorable, intermediate and favorable with OS of 2.7, 9.1 and 13.6 months respectively.
New GPA
GPA
Number of points
Number
Median OS ( months)
Favorable
36 (27%)
13.6
Intermediate
4-5
59 (45%)
9.1
Unfavorable
</= 3
36 (27%)
2.7
<0.0001
Conclusions:
A novel GPA for TBCBM is proposed.

Title: MicroRNA-34a suppresses breast cancer bone metastasis by inhibiting osteoclastogenesis and targeting tgif2
Jing Y Krzeszinski1, Wei Wei1, HoangDinh Huynh1, Zixue Jin1, Xunde Wang1, Tsung-Cheng Chang2, Xian-jin Xie3,4, Lin He5,
Lingegowda S Mangala6,7, Gabriel Lopez-Berestein7,8, Anil K Sood6,7,9, Joshua T Mendell2,3 and Yihong Wan1,3. 1University of
Texas Southwestern Medical Center, Dallas, TX; 2University of Texas Southwestern Medical Center, Dallas, TX; 3Simmons
Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX; 4University of Texas Southwestern Medical Center,
Dallas, TX; 5University of California at Berkeley, Berkeley, CA; 6University of Texas MD Anderson Cancer Center, Houston, TX;
7
Center for RNA Interference and Non-Coding RNA, University of Texas MD Anderson Cancer Center, Houston, TX; 8University
of Texas MD Anderson Cancer Center, Houston, TX and 9University of Texas MD Anderson Cancer Center, Houston, TX.
Body: Introduction:
Approximately 70% of people living with metastatic breast cancer have metastases to their bones. Osteolytic bone metastasis is a
common, debilitating and essentially incurable skeletal complication of breast cancer, in which tumor cells migrate to and destroy
bones, causing extreme pain, fractures, life-threatening hypercalcemia, limited mobility and eventually motality. The bone
resorbing osteoclasts significantly contribute to this process. MicroRNAs (miRNAs) play important roles in physiology and
disease, and present tremendous therapeutic potential. Nonetheless, how miRNAs regulate skeletal biology is underexplored. We
hypothesize that miRNAs that can suppress osteoclast function may ameliorate breast cancer bone metastasis.
Methods:
A strategy of ex vivo osteoclast differentiation from mouse bone marrow cells was used to examine the levels of several
cancer-related miRNAs during a time course of osteoclastogenesis, and test the effects of miR-34a mimic or inhibitor.
Micro-Computed Tomography was conducted to quantify bone mass. ELISA analyses were performed to measure serum bone
resorption and bone formation markers. Ovariectomy was employed as a model for postmenopausal osteoporosis. As a model for
bone metastases, luciferase labelled bone-metastasis prone MDA-MB-231 human breast cancer cell subline were injected into
the left cardiac ventrical of nu/nu mice. Bone metastases were detected and quantified weekly post injection by bioluminescence
imaging.
Results:
MiR-34a is down-regulated during osteoclast differentiation. Osteoclastic miR-34a over-expressing transgenic mice exhibit lower
bone resorption and higher bone mass. Conversely, miR-34a knockout and heterozygous mice exhibit elevated bone resorption
and reduced bone mass. Consequently, ovariectomy-induced osteoporosis, as well as bone metastasis of breast cancer are
diminished in osteoclastic miR-34a transgenic mice.
Pharmacologically, systemic delivery of miR-34a mimics via a chitosan nanoparticle vehicle (miR-34a-CH) can target multiple
tissues. Our bio-distribution analysis shows that miR-34a-CH delivery to the bone marrow is among the highest compared to
other tissues. Administration of miR-34a nanoparticles not only diminished ovariectomy induced bone loss but also attenuated
breast cancer bone metastasis.
Systemic miR-34a-CH delivery affected neither tumor growth when cancer cells were injected subcutaneously nor tumor
metastasis to other organs such as lung when cancer cells were injected intravenously. These results further support the notion
that the suppression of bone metastases by miR-34a-CH was mediated by its inhibition of the bone metastatic niche rather than
the cancer cells.
Mechanistically, we identify the homeodomain transcription factor Tgif2 as an essential direct miR-34a target that is
pro-osteoclastogenic. Tgif2 deletion reduces bone resorption and abolishes miR-34a regulation.
Conclusion:
Using mouse genetic, pharmacological and disease models, we have identified mir-34a as a novel and critical suppressor of
osteoclastogenesis, bone resorption and the bone metastatic niche, revealing miR-34a as a potential new therapeutic strategy to
confer skeletal protection and ameliorate bone metastasis of breast cancer.

Title: Inhibition of osteolytic tumor growth by 5-FdU-alendronate, a bisphosphonate conjugate that maintains bone formation:
Implications for treatment of osteolytic bone lesions
Christian Schem1, Rob J Tower2, Philipp Kneissl1, Anna C Rambow1, Graeme M Campbell2, Thorsten Heilmann1, Anna Trauzold4,
Walter Jonat1, Claus C Gler2, Sarah Schott3 and Sanjay Tiwari2. 1University Medical Centre Schleswig-Holstein, Campus Kiel,
Kiel, Schleswig-Holstein, Germany; 2Section BioMedical Imaging, University Hospital Schleswig-Holstein, Campus Kiel, Kiel,
Schleswig-Holstein, Germany; 3University Hospital Heidelberg, Heidelberg, Germany, Heidelberg, Baden-Wrttemberg, Germany
and 4Institute for Experimental Cancer Research, University of Kiel, Kiel, Schleswig-Holstein, Germany.
Body: Patients with breast cancer bone metastases suffer significant morbity from skeletal-related events but an effective agent
that inhibits bone resorption while maintaining bone formation has not been identified. In this study we tested the therapeutic
efficacy of a novel drug, 5-FdU-ale, a conjugate between the anti-metabolite 5-FdU and the bisphosphonate alendronate, in a
mouse model of breast cancer bone metastases. Administration of 5-FdU-ale in vitro induces cell cycle arrest similar to treatment
with unconjugated 5-FdU. In vivo, mice harboring bone lesions treated with 5-FdU-ale showed a reduction in tumor size not
observed with administration of either Alendronate or 5-FdU. Since osteolysis mediated release of growth factors from bone
contribute to tumor growth, we show 5-FdU-ale treatment significantly reduces bone resorption, although uniquely, the inhibition
of osteoclast activity is not mediated through inhibition of prenylation of small GTPases. Furthermore, and in contrast to
Alendronate, there is no concomitant decrease in bone formation activity, as determined by serum osteocaclin levels. This finding
is supported by micro-CT analyses which reveal significantly higher bone volume and histologically, pockets of tumor cells are
observed largely confined to regions of marrow space between endochondral and trabecular bone. Taken together, we conclude
that 5-FdU-ale has potent anti-tumor efficacy in osteolytic bone lesions mediated uniquely through bone formation activity of
osteoblasts and inhibition of bone resorption. The study identifies the important role of osteoblast in not only preventing the
osteolytic metastatic phenotype but as a means of harnessing the therapeutic potential of bone formation to treat osteolytic
lesions.

Publication Number: PD1-1
Title: Tumor infiltrating lymphocytes and correlation with outcome in the Cher-LOB study
Maria Vittoria Dieci1,2, Giancarlo Bisagni3, Katia Cagossi4, Alberto Bottini5, Samanta Sarti6, Federico Piacentini7, Pierfranco
Conte1,2 and Valentina Guarneri1,2. 1Oncology and gastroenterology, University of Padova, Padova, Italy; 2Medical Oncology 2,
Istituto Oncologico Veneto IRCCS, Padova, Italy; 3Azienda Ospedaliera ASMN, IRCCS, Reggio Emilia, Italy; 4Ramazzini Hospital,
Carpi, Italy; 5AO Istituti Ospitalieri, Cremona, Italy; 6Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori, Meldola,
Italy and 7University Hospital, Modena, Italy.
Body: Background: Tumor infiltrating lymphocytes (TIL) are emerging as a strong prognostic and predictive factor for breast
cancer, especially for the HER2-positive and triple negative subtypes (Loi S, Ann Oncol 2014; Dieci MV, Ann Oncol 2014). Here
we report the results of the TIL biomarker analysis performed in the CherLOB study.
Methods: The phase II neoadjuvant CherLOB study (Guarneri, J Clin Oncol 2012) randomized 121 HER2-positive, stage II-IIIA
breast cancer patients to anthracyclines/taxane-based chemotherapy plus trastuzumab (arm A), lapatinib (arm B), or both (arm
C). Primary endpoint was pathological complete response (pCR). Hematoxylin and eosin-stained slides from both pre-treatment
biopsies and post-treatment surgical samples were centralized and evaluated for the % of intratumoral (It) and stromal (Str) TIL
as previously described (Denkert C, J Clin Oncol 2010). Samples were classified as lymphocyte-predominant (LP) if ItTIL and/or
StrTIL >=60% and as non-LP if ItTIL and StrTIL <60%.
Results: Pre-treatment TIL evaluation was available for 105 of the 118 CherLOB patients who were assessable for pathological
response. Both ItTIL and StrTIL as continuous variables (per 10% increase) were associated with a higher probability of achieving
a pCR (adjusted OR: 2.64, 95%CI 1.46-4.79, p=0.001 and 1.32 95%CI 1.08-1.6, p=0.006 for ItTIL and StrTIL, respectively). pCR
rates were significantly higher in LP compared to non-LP cases (59% vs 27%, p=0.011). According to treatment, TIL effect was
more evident in patients treated with HER2 double-blockade (arm C). According to estrogen receptor (ER) status, no difference in
pCR rates between LP and non-LP cases was observed in the ER-positive population, whereas pCR rate was more than doubled
for ER-negative LP compared to ER-negative non-LP patients (table 1).
Table 1
overall
Arm A
Arm B
Arm C
ER+
ER-
n tot
pCR rate %
LP
17
59
non-LP
88
27
LP
40
non-LP
27
26
LP
50
non-LP
28
21
LP
83
non-LP
33
33
LP
33
non-LP
59
25
LP
11
73
non-LP
29
31
p
0.011
0.52
0.15
0.022
0.67
0.02
Overall, 71 of the 121 CherLOB patients had residual disease at surgery: for 54 of them, paired pre-treatment and post-treatment
TIL were available. No significant changes in ItTIL and StrTIL levels were observed before and after treatment. However, six
cases presented a LP phenotype on the residual disease; all but one of them started from a non-LP pre-treatment phenotype and
received lapatinib as part of the neoadjuvant treatment (4 arm B, 1 arm C).

Conclusions: In this analysis, TIL predicted the achievement of pCR for early HER2-positive patients undergoing neoadjuvant
chemotherapy plus anti-HER2 agents. TIL predictive effect seems limited to ER-negative patients. Combinations of chemotherapy
plus anti-HER2 agents containing lapatinib may be able to convert a non-LP into a LP tumor. Updating of follow-up is ongoing,
correlations between TIL and survival will be presented at the meeting.

Title: Impact of tumor-infiltrating B-cell clonal diversity on response to neoadjuvant therapy in triple negative and HER2+ breast
cancer treated on CALGB (Alliance) 40601 and 40603
Michael D Iglesia1, Benjamin G Vincent1, Jonathan S Serody1, Lisa A Carey1, William T Barry2, William M Sikov3, Clifford A Hudis4,
Eric M Winer5 and Charles M Perou1. 1University of North Carolina, Chapel Hill, NC; 2Alliance Statistics and Data Center,
Dana-Farber Cancer Institute, Boston, MA; 3Alpert Medical School of Brown University, Providence, RI; 4Memorial Sloan Kettering
Cancer Center, New York, NY and 5Dana-Farber Cancer Institute, Boston, MA.
Body: Background: Tumor infiltrating lymphocytes (TILs) are associated with improved outcomes in breast (BrCa) and ovarian
cancer (OvCa). This benefit is largely restricted to the basal-like and HER2-enriched subtypes of BrCa and the immunoreactive
subtype of OvCa. It is not known whether TILs respond to a small subset of antigens, similar to an antiviral or antibacterial
response, or if the response is nonspecific. We developed a novel method to assess B-cell population diversity by analyzing
B-cell receptor (BCR) sequence complexity in mRNA-seq datasets derived from tumor biopsies. B-cells in a subset of basal-like
and HER2-enriched BrCa showed high expression of immunoglobulins coinciding with reduced BCR diversity consistent with a
restricted epitope-driven immune response. Analysis of DNA patterns from B-cells in basal-like and HER2-enriched BrCa showed
a greater prevalence of BCR somatic hypermutation (SHM) suggestive of an antigen-restricted response (Iglesia et al, CCR
2014). Here, we studied the impact of this adaptive immune response on treatment response.
Methods: Using two neoadjuvant cooperative group trials in triple-negative (TNBC) and HER2-positive (HER2+) BrCa, we
evaluated BCR diversity (as assessed by SHM diversity) as a continuous variable and as a binary variable (diverse/restricted)
relative to BCR expression (the Restriction Index) in pre-treatment tumor samples from 265 patients with HER2+ BrCa treated on
CALGB 40601, a randomized phase III trial of paclitaxel plus trastuzumab +/- lapatinib, and 443 patients with TNBC treated on
CALGB 40603, a randomized phase II trial of standard chemotherapy +/- carboplatin and/or bevacizumab. We examined the
relationship between a restricted immune response and pCR rate, the primary endpoint of both studies, overall and within
molecular subtypes.
Results: In HER2+ BrCa, the combination of high immunoglobulin expression and lower sequence diversity (high Restriction
Index) was observed in 28% of the pre-treatment biopsies, and varied by intrinsic subtype, with the greatest prevalence in the
HER2-enriched subset (n=80, 46% vs 20% in all others). BCR restriction predicted improved pCR rates in all patients (67%
versus 37%, p <0.0001). It remained significant in the HER2-enriched subset (n = 82, p=0.0086). The impact of Restriction Index
on response to chemotherapy in TNBC is being analyzed and will be presented along with multivariate models to adjust for other
patient and disease characteristics and explore potential interactions.
Conclusions: The presence of a restricted diversity B-cell response in HER2+ breast cancer correlates with improved response
to neoadjuvant chemotherapy plus HER2-targeted therapy, which may in part explain its impact on prognosis. We will determine if
a similar correlation exists with chemotherapy response in TNBC. This suggests that immunomodulatory therapies supporting a
B-cell response may be a promising therapeutic approach to targeting these tumors.

Title: The significance of tumor infiltrating lymphocyte density, subset composition and organization in breast cancer
Karen Willard-Gallo1, Laurence Buisseret1, Soizic Garaud1, Chunyan Gu-Trantien1, Alexandre de Wind1, Sbastien Duquenne1,
Denis Larsimont1, Christos Sotiriou1 and Martine Piccart1. 1Institut Jules Bordet, Universit Libre de Bruxelles, Brussels, Belgium.
Body: The clinical relevance of tumor infiltrating lymphocytes (TIL) in breast cancer (BC) has been clearly established by recent
large clinical trials (presented at SABCS 2013). The relationship between protective immunity, observed in some patients, and
critical features of lymphoid subset composition and organization remain unknown. Our recent work revealed that tertiary
lymphoid structures (TLS), principally composed of T cells, B cells and dendritic cells, are present in the peri-tumoral regions of
breast tumors and associated with a CD4+ follicular helper T (Tfh) cell presence. Through retrospective analyses, we determined
that TLS were associated with good clinical outcomes in both the neo-adjuvant and adjuvant settings. To gain insight into the
immune components linked with a significant TIL and TLS presence, we initiated a prospective flow cytometric study. We
systematically immunophenotyped T and B cell TIL in breast tissues from tumors (n=125), non-adjacent non-tumor tissues
(NANT, n=115) and normal tissue from mammary reductions (n=26) on the day of surgery. TIL organization and spatial
distribution was subsequently analyzed by immunohistochemistry (IHC) and immunofluorescence (IF) on paraffin sections for a
subset of patients (n=78). The fresh tissue analyses revealed that TIL density was a continuum across the 125 patients analyzed.
A cutoff for TIL-positive and -negative tumors was set at 58 CD45+ TIL per mm3 of tissue based on the number of CD45+ cells
present in the remarkably similar normal and NANT tissues. Applying this threshold to BC, 65% were TIL-positive with
approximately one-third considered extensively infiltrated. TIL-positive tumors are characterized by an increase in CD4+ T cells
and CD19+/CD20+ B cells. The median CD4/CD8 ratio was >1 in TIL-positive compared to <1 in TIL-negative tumors and NANT.
CD4+ and CD8+ T cells were predominantly CD45RO+ memory cells, with a significant proportion expressing PD-1. Infiltrating B
cells were approximately 50% memory cells in contrast to <15% in normal and NANT tissues. Extensively infiltrated tumors were
more frequently associated with Tfh and follicular B cells resident in TLS. IF analysis by confocal microscopy found that TLS
resident cells included specific lymphocyte subsets: marginal B cells (CD20+CD27+IgD), follicular mantle B cells (CD20+IgD+),
germinal center B cells (CD20+Ki67+CD35+CD21+PD-1+) and germinal center Tfh cells (CD4+CD200+TIGIT+CXCL13+) in a B
cell zone surrounded by a T cell zone containing CD4+ and CD8+ T cells. To determine whether the flow cytometry data was
correlated with routine pathology, sections of the same tumors were stained by H&E and CD45 (total leukocytes) or CD3 plus
CD19 (T cells + B cells; >95% of CD45+ cells) IHC and scored by trained pathologists. The best correlation was observed
between CD3/CD20 flow cytometry and CD3/CD19 IHC, the latter also associated with lower inter-observer variability. TIL density
was positively correlated with proliferation (Ki67 & histological grade) and hormone receptor negativity while TLS were more
frequent in younger women. These data suggest that organized immune responses in TLS adjacent to the tumor bed provide an
effective location for generating anti-tumor memory T and B cell responses.

Title: Somatic leukemogenic mutations associated with infiltrating white blood cells in breast cancer patients
Elizabeth A Comen1, Maria Kleppe1, Hannah Wen1, Britta Weigelt1, Lennart Bastian1, Brian Blum1, Franck T Rapaport1, Matt
Keller1, Nicolas Socci1, Agnes Viale1, Daoqi You1, Robert Benezra1, Edi Brogi1, Jorge Reis-Filho1, Michael Berger1, Ross Levine1
and Larry Norton1. 1Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background: In the last few decades, theoretical models of cancer growth and progression have long focused on the
aberrations of cancer cells alone, such as the abnormal mitotic and invasive characteristics of cancer cells. More recent research
across multiple solid tumors suggests a critical interplay between solid tumors and immune regulating cells. Mounting evidence
suggests that the immune system can tip the scales of cancer progression, eliciting either an anti-tumor or pro-tumor immune
response depending upon varying stimulating and inhibitory factors. Here, we are the first to demonstrate novel mutations
including leukemogenic mutations among tumor infiltrating lymphocytes in breast cancer patients.
Methods: We obtained 17 primary breast cancer samples from patients who presented for either a lumpectomy or mastectomy as
part of an IRB approved biospecimen protocol. Of the 17 patient samples, 13 had triple negative breast cancer, 2 had ER+,
HER2+ disease, and 2 had ER+, HER2- disease. In the 17 samples, we used fluorescent activated cell sorting to separate
CD45-positive hematopoietic cells from CD45-negative epithelial cells. We then performed exome sequencing of tumor-infiltrating
hematopoietic cells to investigate for the presence of pathogenic mutations in tumor-associated leukocytes. In this first step, we
identified candidate mutations in known cancer genes, including BCOR, NOTCH2, TET2, NF1, EZH2, and JAK1. As a validation
step, we then performed capture-based sequencing of tumor-infiltrating leukocytes in 20 breast cancer samples matched to each
patients germline DNA sample (buccal swab). In 10 of the 20 patients, we identified and validated somatic mutations. Of note, 6
of these patients harbored mutations known to be associated with leukemia, including DNTM3A, TET2, and BCOR. Most of these
mutations were present in at least 5-20% of reads. This suggests that these mutations were present in enriched subclones and
were not rare alleles occurring in a minority of hematopoietic stem cells. Lastly, we performed 454 deep sequencing analysis of
microdissected tumor DNA samples and confirmed the absence of these mutations in breast cancer cells.
Conclusion: Our data demonstrate somatic mutations in tumor infiltrating leukocytes in breast tumors which were not identified in
matched germline or tumor DNA samples. Notably, some of these mutations have been implicated in the pathogenesis of
lymphoid and myeloid malignancies. This observation suggests a unique relationship between cancer cells and mutant infiltrating
leukocytes. We are now investigating the functional interaction between cancer cells and hematopoietic cells. Our findings
reframe our understanding of carcinogenesis and offer novel opportunities for cancer detection and treatment.

Title: Characterizing the Tumor Immune MicroEnvironment (TIME) in high-risk ductal carcinoma in situ
Michael J Campbell1, Rita Mukhtar1, Ekene Obi-Okoye1, Booyeon Han1, Vickram Tandon1, Sarah Zheng1, Zelos Zhu1, Max
Endicott1, Max Wicha2, Linda Lindstrom1, Alfred Au1, Frederick Baehner1, Joe Gray3 and Laura Esserman1. 1University of
California, San Francisco, CA; 2University of Michigan, Ann Arbor, MI and 3Oregon Health & Science University, Portland, OR.
Body: Background: Ductal carcinoma in situ (DCIS) of the breast is a premalignant condition. Although DCIS is treated as an
obligate precursor of invasive ductal carcinoma, the rate and latency of progression from DCIS to invasive breast cancer (IBC) in
the absence of treatment are unknown. DCIS is not one condition, but rather a spectrum of disease and although DCIS itself is
not a lethal condition, women with DCIS are at higher risk of developing subsequent IBC over a time period of 1-20 years
depending on DCIS subtype. Features of DCIS that are associated with high risk of recurrence include large size (> 5cm), high
grade, comedo necrosis, palpable mass, hormone receptor (HR) negativity, and HER2 positivity. The objective of this study was
to characterize the tumor immune microenvironment (TIME) of these high-risk DCIS lesions.
Methods: Forty-eight cases of high grade DCIS, enriched for large, confluent lesions and history of recurrence were age matched
with 64 cases of non-high grade DCIS. IHC analyses were performed as single or two-color stains for the following antigens:
CD68, CD8, CD4, CD20, HLA-DR, CD115, FoxP3, PCNA, Mac387, MRC1, ALDH, CD24, CD44, Ki-67, and HER2. HR status
was determined from ER and PR staining results in pathology reports. A Nuance multispectral imaging system was used to image
and spectrally unmix each stain. Protocols for automated image analysis were developed using CellProfiler software.
Associations between immune cell populations and clinical parameters (tumor palpability, recurrence, HR status, HER2 status,
and Van Nuys score [12-point scale: margins, age, size, grade]) were identified with non-parametric Spearman correlation tests.
Results: We found a high macrophage infiltrate associated with a high Van Nuys score, palpability, and high Ki-67. High CD115
(CSF-1 receptor) was associated with HER2+, high Ki-67, and recurrence. Mac387+ cells and FoxP3+ regulatory T cells (Treg)
were associated with high Van Nuys score, comedo necrosis, high Ki-67, HR- and HER2+. Interestingly, both Mac387 and
CD115 were expressed on tumor cells as well as macrophages and high CD115 staining on tumor cells was associated with
recurrence. The presence of CD8+HLA-DR-negative T cells throughout a section was associated with high Van Nuys score, HR-,
HER2+, and recurrence. In contrast, CD8+ T cells within the nests of tumor cells were negatively associated with Van Nuys
score, palpability, and comedo necrosis. A tumor immune microenvironment score (TIME score) was developed based on the
proportions of various immune cell populations. A high TIME score was significantly associated with high Van Nuys scores as well
as with recurrence.
Summary: These results demonstrate that high risk DCIS features (palpability, high Van Nuys score, high proliferation, HR-,
HER2+, and increased recurrence) are associated with a suppressed tumor immune micro-environment (high FoxP3+ cells,
CD68+Mac387+ cells, CD8+HLA-DR-neg T cells, and upregulated CD115). These high risk lesions truly represent an opportunity
to prevent cancer. Identifying these high risk lesions with the help of tumor immune microenvironment markers and manipulating
the DCIS TIME via local or systemic immunotherapeutic strategies may represent an ideal preventative intervention.

Title: Inhibition of PD-L1 by MPDL3280A leads to clinical activity in patients with metastatic triple-negative breast cancer
Leisha A Emens1, Fadi S Braiteh2, Philippe Cassier3, Jean-Pierre DeLord4, Joseph Paul Eder5, Xiaodong Shen6, Yuanyuan Xiao6,
Yan Wang6, Priti S Hegde6, Daniel S Chen6 and Ian Krop7. 1Johns Hopkins University, Baltimore, MD; 2Comprehensive Cancer
Centers of Nevada, Las Vegas, NV; 3Centre Lon Brard, Lyon, France; 4Institut Claudius Regaud, Toulouse, France; 5Yale
School of Medicine, New Haven, CT; 6Genentech, San Francisco, CA and 7Dana-Farber Cancer Institute, Boston, MA.
Body: Background: Metastatic triple-negative breast cancer (TNBC) is associated with a poor prognosis and has no targeted
therapy options. Programmed death-ligand 1 (PD-L1) expression is more prevalent in TNBC than in other breast cancer
subtypes. PD-L1 may protect cancer cells from immune-mediated destruction by binding to its receptors PD-1 and B7.1.
MPDL3280A, a human anti-PD-L1 monoclonal antibody with an engineered Fc-domain designed for optimized efficacy and
safety, blocks PD-L1 activity and restores tumor-specific T-cell immunity.
Methods: This multicenter Phase I study selectively enrolled a cohort of patients with PD-L1-positive metastatic TNBC. PD-L1
positivity was centrally evaluated based on the PD-L1 immunohistochemistry (IHC) status of tumor-infiltrating immune cells (ICs).
Eligible patients received MPDL3280A 15 or 20 mg/kg IV every 3 weeks for up to 1 year. The objective response rate (ORR) was
assessed by RECIST v1.1. MPDL3280A immune correlates were evaluated for tumor and circulating biomarkers. The clinical
data cutoff was January 1, 2014.
Results: Twelve patients were treated with MPDL3280A and evaluable for safety. Patients had ECOG PS 0-1 (33% ECOG PS
1), and the median age was 55 years (range, 29-72 years). All 4 patients (33%) with visceral metastases had liver metastases,
and 1 patient (8%) had bone metastases. 92% of patients received 2 prior therapies. Prior chemotherapies included
anthracycline (92%), taxane (75%) and platinum (42%). Grade 3-4 treatment-related adverse events (AEs) occurred in 8% of
patients (1 event of Grade 3 adrenal insufficiency). One patient had an immune-related AE (Grade 2 pyrexia). No
treatment-related deaths were observed. Nine patients were dosed by November 16, 2013 and evaluable for efficacy. The ORR
was 33%, including 1 CR and 2 PRs. Responders included patients with visceral metastases at baseline. At the time of the
clinical data cut-off, all responses occurred within 6 weeks of the first dosing of MPDL3280A, and all of the responses were
ongoing. The median duration of response had not been reached. One patient achieved stable disease as best response. Two
additional patients had tumor shrinkage (-43% and -44% change in target lesions, respectively) but were not counted as RECIST
responders due to the appearance of new lesions, which is likely attributable to pseudoprogression. Preliminary
pharmacodynamic biomarkers related to MPDL3280A activity, including circulating plasma markers and tumor immunomonitoring
with CD8 IHC, will be presented along with updated clinical data.
Conclusions: MPDL3280A treatment was well tolerated and was associated with objective clinical activity in patients with
pretreated metastatic TNBC. Further evaluation of the safety and clinical activity of MPDL3280A in both PD-L1-positive and
PD-L1-negative metastatic TNBC is ongoing. Clinical trial information: NCT01375842.

Title: Differential expression of innate and adaptive immune responses in TNBC outcome
Yesim Gokmen-Polar1, Xianyin Lai1 and Sunil Badve1. 1Indiana University School of Medicine, Indianapolis, IN.
Body: Introduction:
Clinical management of patients with triple-negative breast cancers (TNBCs) presents a significant challenge. TNBCs tend to
relapse early and exhibit poor prognosis compared to other breast cancer subtypes. Recent studies emphasize the influence of
the immune system in prognosis of TNBCs. In particular, the percentage of tumor infiltrating lymphocytes (TILs) is associated with
breast cancer prognosis and outcomes after conventional therapy in triple negative tumors. However, the biology underlying this
immune response is unclear.
Methods:
To identify differential expression of key immune response markers in TNBC with clinical outcome, we have performed a
three-step approach as follows: 1) quantitative LC-MS/MS proteomic analysis using 16 formalin-fixed paraffin-embedded (FFPE)
triple negative tumors with prognostic outcome and Ingenuity Pathway Analysis (IPA) performed to determine the biological
processes and networks, 2) in silico analysis of prognostic value (10 year- DFMS) with the expression levels of 96 immune
response genes using large cohorts of publicly available gene expression datasets (11 Affymetrix datasets), and 3) integrating the
quantitative LC-MS/MS proteomic analysis with the in silico analysis to further understand the relation between the type of
immune response including innate and adaptive immune responses.
Results:
LC-MS/MS proteomic analysis identified and quantified a total of 1,560 proteins of which 254 were differentially expressed in
TNBC tumors with poor prognosis compared to good prognosis (P0.05). IPA analysis revealed inflammatory response and
immunological processes as top differentially regulated biological functions in these tumors [P ranging between 7.41E-10 8.05E-03 (53 proteins) and 2.59E-09 - 5.80E-03 (69 proteins), respectively)]. The top altered immunologic networks included
"Immune cell trafficking" and "cell-mediated immune response". Among the differentially regulated markers from both analyses,
osteopontin (OPN or SPP1), STAT1, and MX1 remained highly significant in both proteomic and in silico analyses (P<0.008).
Interestingly, many components of the innate immunity (i.e TLRs) and adaptive immunity (Th2 response, Th17, T-reg, T cell
activation, and related cytokines) did not reach statistical significance in combinatorial analysis.
Conclusion:
Integrated in silico and proteomic analysis provides clues regarding the relative importance of immune pathways associated with
response in breast cancer. This will enable targeting of immune therapies to improve the functionality of the pathways and
ultimately in choosing the right patient TNBC subgroup for potential therapeutics, thereby improving the clinical outcome.

Title: The effect on overall and disease-free survival (OS & DFS) by adding bevacizumab and/or antimetabolites to standard
neoadjuvant chemotherapy: NSABP Protocol B-40
Harry D Bear1,2, Gong Tang1,3, Priya Rastogi1,4, Charles E Geyer1,2, Qing Liu1,3, Andr Robidoux1,5, James N Atkins1,6, Luis
Baez-Diaz1,7, Adam M Brufsky1,8, Rita S Mehta1,9, Louis Fehrenbacher1,10, James A Young1,11, Francis M Senecal1,12, Rakesh
Gaur1,13, Richard G Margolese1,14, Paul T Adams1,15, Howard M Gross1,16, Joseph P Costantino1,3, Soonmyung Paik1,17, Sandra M
Swain1,18, Eleftherios P Mamounas1,19 and Norman Wolmark1,20. 1National Surgical Adjuvant Breast and Bowel Project (NSABP),
Pittsburgh, PA; 2Medical College of Virginia School of Medicine and the Massey Cancer Center, Virginia Commonwealth
University; 3NSABP Biostatistical Center, University of Pittsburgh Graduate School of Public Health; 4University of Pittsburgh
Cancer Institute, University of Pittsburgh, PA; 5Centre Hospitalier de l'Universit de Montral (CHUM); 6South East Cancer
Control Consortium Community Clinical Oncology Program (CCOP); 7Minority-Based CCOP, San Juan; 8University of Pittsburgh,
Magee Womens Hospital, PA; 9University of California, Irvine, CA; 10Kaiser Permanente Oncology Clinical Trials, Northern
California; 11Colorado Cancer Research Program; 12Northwest Medical Specialties; 13CCOP, Kansas City; 14Jewish General
Hospital, McGill University; 15Genesys Regional Medical Center; 16CCOP, Dayton; 17Severance BioMedical Science Institute and
Yonsei University College of Medicine; 18Washington Cancer Institute, Medstar Washington Hospital Center, Washington, DC;
19
UF Cancer Center at Orlando Health and 20Allegheny Cancer Center at Allegheny General Hospital.
Body: Purpose
The NSABP B-40 study was designed to determine whether the addition of capecitabine (X) or gemcitabine (G) to neoadjuvant
docetaxel (T) followed by AC increased pathologic complete response (pCR) rates and improved outcomes of women with
operable, HER2-negative breast cancer. In addition, B-40 was to determine whether the addition of neoadjuvant plus adjuvant
bevacizumab (Bev) to T-based regimens followed by AC would increase pCR and improve outcomes. The pCR results and
toxicities have been reported previously.
The DFS and OS results are reported here for the first time.
Methods
Patients (Pts) received one of three T-based regimens for 4 cycles: T 100 mg/m2 day 1; T 75 mg/m2 day 1 and X 825 mg/m2 BID
days 1-14; or T 75 mg/m2 day 1 and G 1000 mg/m2 days 1 and 8. Pts then received preoperative AC x 4 cycles. Pts randomly
assigned to the Bev groups received Bev at 15mg/kg, q3wks x 6 with the first 6 courses of neoadjuvant chemotherapy and were
to resume Bev post-operatively for 10 doses. 1,206 pts were assigned and 1,163 of them were clinically eligible and had
follow-up. The cutoff date for data presented here was March 31, 2014. The median follow-up was 4.7 years.
Results
Neither X nor G added to neoadjuvant chemotherapy increased DFS or OS. 50% of Bev pts for whom treatment data were
available (577) completed 10 doses of post-op Bev; 26% did not start Bev post-op. Addition of Bev significantly increased OS
(HR=0.70, p=0.01) and marginally increased DFS (HR=0.81, p=0.06). The effect of Bev was most notable in women with
hormone receptor positive (HR+) breast cancers (DFS HR=0.71, p=0.04; OS HR=0.63, p=0.03).
Conclusions
Addition of G or X to neoadjuvant T + AC had no significant impact on DFS or OS. Bev added to neoadjuvant chemotherapy and
continued post-operatively marginally increased DFS in the overall cohort, with a significant increase in DFS in the HR+ subset.
The addition of bevacizumab significantly improved OS for women with operable HER-2-negative breast cancer. In agreement
with previously reported results for pCR, the improvement in DFS and OS was seen preferentially in women with HR+ breast
cancers.
Support
NCI PHS grants U10-CA-37377, -69974, -12027, -69651, -44066, and -44066-26, Genentech Inc, Roche Laboratories Inc, Eli
Lilly, and Precision Therapeutics, Inc.

Title: Final efficacy and updated safety results of the randomized phase III BEATRICE trial evaluating adjuvant bevacizumab
(BEV)-containing therapy for early triple-negative breast cancer (TNBC)
Richard Bell1, Julia Brown2, Mahesh Parmar3, Mark Toi4, Thomas Suter5, Guenther Steger6, Xavier Pivot7, John Mackey8,
Christian Jackisch9, Rebecca Dent10,11, Peter Hall12, Almut Mecke12, Leilani Morales13, Louise Provencher14, Elzbieta
Staroslawska15, Roberto Hegg16, Laurence Vanlemmens17, Andreas Kirsch18, Andreas Schneeweiss19, Norikazu Masuda20,
Friedrich Overkamp21 and David Cameron22. 1Deakin University, Geelong, Australia; 2Clinical Trials Research Unit, University of
Leeds, Leeds, United Kingdom; 3Medical Research Council Clinical Trials Unit, London, United Kingdom; 4Kyoto University,
Faculty of Medicine, Kyoto, Japan; 5Swiss Cardiovascular Center, Bern University Hospital, Bern, Switzerland; 6Division of
Oncology, Medical University of Vienna, Vienna, Austria; 7Service Oncologie Medicale, University Hospital Jean Minjoz,
Besanon, France; 8Cross Cancer Institute, Edmonton, Canada; 9Klinikum Offenbach, Offenbach, Germany; 10National Cancer
Center, Singapore, Singapore; 11Sunnybrook Health Sciences Center and University of Toronto, Toronto, Canada; 12Leeds
Institute of Health Sciences, University of Leeds, Leeds, United Kingdom; 13F. Hoffmann-La Roche Ltd, Basel, Switzerland;
14
Centre des Maladies du Sein Deschnes-Fabia, CHU de Qubec, Hpital du Saint-Sacrement, Qubec, Canada; 15St John's
Cancer Center, Lublin, Poland; 16Perola Byington Hospital/FMUSP, So Paolo, Brazil; 17Centre Oscar Lambret, Lille, France;
18
Onkologischer Schwerpunktam Oskar-Helene-Heim, Berlin, Germany; 19National Center for Tumor Diseases, University
Hospital, Heidelberg, Germany; 20Breast Oncology, NHO Osaka National Hospital, Osaka, Japan; 21Praxis und Tagesklinik fr
Onkologie, Oncologianova, Recklinghausen, Germany and 22University of Edinburgh and Cancer Services, NHS Lothian,
Edinburgh, United Kingdom.
Body: Background: In the metastatic and neoadjuvant settings, BEV improves PFS and pCR rates, respectively, when
combined with chemotherapy (CT) for breast cancer. However, in the adjuvant setting invasive disease-free survival (IDFS;
primary outcome measure) was not improved by adding 1 year of BEV to CT in the open-label randomized phase III BEATRICE
trial in early TNBC (stratified hazard ratio [HR] 0.87 [95% CI 0.721.07]). We report final efficacy and updated safety results.
Methods: Eligible patients (pts) had centrally confirmed triple-negative and/or basal-like operable primary invasive breast cancer
(pT1apT3). Investigators selected anthracycline (anth)- and/or taxane (tax)-based CT for each pt. After definitive surgery, pts
were randomized 1:1 to receive 4 cycles of either CT alone or the same CT + 1 year of BEV (5 mg/kg/wk equivalent).
Stratification factors were nodal status (0 vs 13 vs 4 involved nodes), selected CT (anth vs tax vs both), hormone receptor
status (negative vs low), and surgery (breast conserving vs mastectomy). Secondary outcome measures included overall survival
(OS) and safety (CTCAE v3.0). Final OS analysis was prespecified after median follow-up of 5 years, 76 mo after the first pt
was randomized, or after 340 deaths, whichever occurred first.
Results: At the data cut-off (June 30, 2014), median follow-up was 56 mo; 293 of 2591 randomized pts had died (86% of
estimated events for the final analysis). There was no statistically significant difference in OS between treatment arms either
overall or in prespecified subgroups (Table). 5-year OS rates were 88% (95% CI 85.789.6%) with CT alone and 88% (95% CI
86.089.8%) with CTBEV. Updated IDFS results (exploratory analysis) were consistent with the primary IDFS analysis. 5-year
IDFS rates were 77% (95% CI 74.479.4%) with CT alone vs 80% (95% CI 77.281.9%) with CTBEV. New grade 3 AEs
occurred in 4.6% and 4.5% of pts, respectively, in the period from 18 mo after first study dose to study end.
Category
All
Age, y
ECOG PS
Menopausal status
Tumor size, cm
Subgroup
All (N=2591)
<40 (N=484)
40<65 (N=1868)
65 (N=239)
0 (N=2388)
1 (N=192)
Post (N=1250)
Pre (N=1341)
0<2 (N=939)
OS HR (95% CI)
0.94 (0.751.18)
0.88 (0.511.54)
0.89 (0.681.17)
1.50 (0.733.05)
0.98 (0.771.25)
0.68 (0.311.49)
1.16 (0.831.62)
0.79 (0.581.08)
1.17 (0.691.98)
No. of positive lymph nodes
25 (N=1514)
0.93 (0.701.23)
5 (N=132)
0 (N=1640)
0.84 (0.451.58)
0.79 (0.561.13)
1.01 (0.651.56)
1.14 (0.751.73)
13 (N=638)
Adjuvant CT
4 (N=313)
Anth (N=947)
Anth + tax (N=1508)
Hormone receptor status
Tax (N=136)
ER and PgR negative (N=2453)
ER and/or PgR low (N=138)
Surgery
Breast conserving (N=1644)

Mastectomy (N=947)
0.83 (0.541.28)
1.03 (0.781.37)
0.63 (0.241.66)
0.93 (0.731.17)
1.42 (0.464.34)
0.93 (0.671.30)
0.95 (0.691.31)
Conclusions: The final OS analysis after 293 deaths showed no significant benefit from the addition of BEV to standard adjuvant
CT for early TNBC. Updated IDFS results were similar to the primary IDFS results, showing no difference between treatments.
Late-onset AEs were rare in both groups. 5-year IDFS and OS rates suggest that the prognosis for pts with TNBC is better than
previously thought.

Title: ARTemis: A randomised trial of bevacizumab with neo-adjuvant chemotherapy for patients with HER2-negative early breast
cancer
Helena M Earl1,2, Louise Hiller3, Janet A Dunn3, Clare Blenkinsop3, Louise Grybowicz4, Anne-Laure Vallier4, Jean Abraham1,2,5,
Jeremy Thomas6, Elena Provenzano2,5, Luke Hughes-Davies5, Karen McAdam7, Stephen Chan8, Rizvana Ahmad9, Tamas
Hickish10, Stephen Houston11, Daniel Rea12, John Bartlett13,14, Carlos Caldas1,2,15, David Cameron14 and Larry Hayward6.
1
University of Cambridge, Cambridge, United Kingdom; 2NIHR Cambridge BioMedical Research Centre, Cambridge, United
Kingdom; 3Warwick Clinical Trials Unit, University of Warwick, Coventry, United Kingdom; 4Cambridge Clinical Trials UnitCancer
Theme, Cambridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom; 5Cambridge University Hospitals
NHS Foundation Trust, Cambridge, United Kingdom; 6Western General Hospital, Edinburgh, United Kingdom; 7Peterborough City
Hospital, Peterborough, United Kingdom; 8Nottingham University Hospital (City Campus), Nottingham, United Kingdom; 9West
Middlesex University Hospital, Isleworth, United Kingdom; 10Royal Bournemouth Hospital, Bournemouth, United Kingdom; 11Royal
Surrey County Hospital, Guildford, United Kingdom; 12Sandwell and West Birmingham NHS Trust and University Hospital
Birmingham NHS Foundation Trust, Birmingham, United Kingdom; 13Ontario Institute for Cancer Research, Ontario, Canada;
14
Edinburgh Cancer Research Centre, University of Edinburgh, Edinburgh, United Kingdom and 15Cancer Research UK
Cambridge Research Institute, Cambridge, United Kingdom.
Body: Background: Bevacizumab (bev) has been used with neo-adjuvant chemotherapy (NACT) in breast cancer trials.
Geparquinto reported benefit for bev in triple negative (neg) patients (pts) (pathological complete response (pCR) 36.4% vs
27.8% p=0.02), as did CALGB 40603 (pCR 52% vs 44%, p=0.057), although NSABP-B40 showed benefit in ER positive (pos) pts
(pCR 23.3% vs 15.2%, p=0.008).
Methods: ARTemis is a randomised phase 3 trial adding bev to NACT (docetaxel (D)-FEC). Pts with HER2-neg invasive breast
cancer were eligible. Stratification was by age, ER status (neg:weak pos:strong pos), tumour size (T2:T3/4), clinical involvement
of axillary nodes and inflammatory/locally advanced disease. Pts were randomised (1:1) to bev+D-FEC or D-FEC. The primary
endpoint was pCR, defined as no residual invasive cancer in the breast or axillary lymph nodes after NACT. 800 pts were
required to detect 10% differences in pCR rates; 85% power, 5% alpha level.
Results: 800 pts were randomised from 66 UK centres (May 2009 to Jan 2013). 68% were <50 years old, 19% had inflammatory
and/or locally advanced disease, 79% of tumours <50mm, 52% clinical node pos and 33% ER-neg. A 2-reader independent
review of pathology reports determined whether pCR had been achieved or, at least, minimal residual disease (MRD) status.
Significantly more pts on bev+D-FEC had a pCR (22% vs 17%; adjusted p=0.03) (see table). pCR rates differed significantly
across ER groups (neg 38%, weak pos 39%, strong pos 7%; p<0.0001). Treatment effect of bev remained significant after
adjustment for ER (p=0.03). Similarly significantly more pts on bev+D-FEC had a pCR or MRD (36% vs 29%; adjusted p=0.035).
Rates differed significantly across ER groups (neg 51%, weak pos 58%, strong pos 18%; p<0.0001). Treatment effect of bev
remained significant after adjustment for ER (p=0.03).
$ pCR in all breast tumours AND absence of disease in ax LNs in all breast tumours
DFEC
Bev+DFEC
% (95%CI)
% (95%CI)
(n=66/393)
(n=87/388)
p*
17% (13-21%) 22% (18-27%) 0.03

ER neg (Allred 0-2) (n=253) 32% (24-41)
44% (36-54)
ER weak pos (Allred 3-5) (n=67) 26% (13-44)
52% (34-69)
ER strong pos (Allred 6-8) (n=461)

pCR or MRD in all breast tumours
7% (4-11)
6% (3-10)
(n=114/394)
(n=138/388)
29% (25-34%) 36% (31-41%) 0.035

ER neg (Allred 0-2) (n=254) 45% (36-54)
56% (47-65)
ER weak pos (Allred 3-5) (n=67) 44% (27-62)
73% (54-87)
ER strong pos (Allred 6-8) (n=461) 18% (13-23)
19% (14-24)
* Adjusted for stratification variables. $ Primary endpoint for the ARTemis trial
Conclusions: ARTemis showed a significant improvement in both pCR and MRD rates with the addition of bev to D-FEC. ER-neg
and ER-weak pos / HER2-neg breast cancer pts appeared to benefit most from bev, whilst pCR and MRD rates in ER-strong pos
pts were lower and did not appear to benefit from bev. Our results are similar to those reported in Geparquinto and CALGB
40603.

Title: Subgroup efficacy analyses of the randomized phase III TANIA trial evaluating continued or reintroduced bevacizumab
(BEV) after 1st-line BEV for HER2-negative locally recurrent/metastatic breast cancer (LR/mBC)
Fabio Puglisi1, Javier Cortes2, Eduard Vrdoljak3, Joseph Gligorov4, Norbert Marschner5, Christoph Zielinski6, Michele de
Laurentiis7, Etienne Brain8, Christelle Lvy9, Anja Welt10, Zsuzsanna Kahan11, Moshe Inbar12, Sabine de Ducla13, Ulrich
Freudensprung13 and Gunter von Minckwitz14,15. 1University Hospital of Udine, Udine, Italy; 2Vall d'Hebron University Hospital,
Barcelona, Spain; 3University Hospital Split, Split, Croatia; 4APHP Tenon, IUC-UPMC, Paris, France; 5Outpatient Cancer Center,
Freiburg, Germany; 6Comprehensive Cancer Center, Medical University Vienna and CECOG, Vienna, Austria; 7INT Fondazione
G Pascale, Naples, Italy; 8Institut Curie Hpital Ren Huguenin, Saint-Cloud, France; 9Centre Franois Baclesse, Caen,
France; 10West German Cancer Center, University Hospital, University Duisburg, Essen, Germany; 11University of Szeged,
Szeged, Hungary; 12Tel Aviv Sourasky Medical Center, Tel Aviv, Israel; 13F. Hoffmann-La Roche Ltd, Basel, Switzerland;
14
German Breast Group, Neu-Isenburg, Hessen, Germany and 15University Women's Hospital, Frankfurt, Germany.
Body: BACKGROUND: The open-label randomized phase III TANIA trial demonstrated statistically significantly improved
progression-free survival (PFS; primary endpoint) with the addition of BEV to 2nd-line chemotherapy (CT) in patients (pts) with
HER2-negative LR/mBC progressing after 1st-line BEV-containing therapy. We describe efficacy in clinically relevant subgroups
to assess consistency of the BEV treatment effect.
METHODS: Pts whose HER2-negative LR/mBC had progressed during/after 1st-line BEVCT were randomized 1:1 to
investigators chosen 2nd-line single-agent CT given either alone or with BEV (15 mg/kg q3w or 10 mg/kg q2w). 2nd-line therapy
was continued until disease progression (PD), unacceptable toxicity or consent withdrawal. At PD, BEV was continued with
3rd-line CT (investigators choice) in pts initially randomized to BEVCT but was not permitted in pts randomized to CT alone. The
primary endpoint was PFS from randomization to 2nd-line PD/death. Sample size was calculated based on a log-rank test
assuming median PFS of 79.3 mo with a corresponding hazard ratio (HR) of 0.75. PFS events were required in 384 of 488 pts
for 80% power at 2-sided =0.05. Subgroup analyses of the primary endpoint were prespecified and included subgroups defined
by stratification factors.
RESULTS: From Jan 2011 to Apr 2013, 494 pts were enrolled. The data cut-off for the primary analysis was Dec 20, 2013
(median follow-up: CT 15.9 mo; BEVCT 16.1 mo). The PFS benefit seen in the overall population was observed consistently in
most subgroups.
Median
Pts with PFS events/Total
2nd-line PFS, Unstratified HR (95% CI)
No. of pts (%)
mo
Subgroup
CT
BEVCT
CT BEVCT
All
203/247 (82) 204/247 (83) 4.2 6.3
0.75 (0.61-0.93)a
Hormone receptor statusb Triple negative
56/60 (93)
2.1 4.9
0.59 (0.40-0.88)
Positive, HER2 negative
147/187 (79) 159/191 (83) 4.7 6.7
0.84 (0.67-1.05)
<6
61/69 (88)
3.9 5.1
0.62 (0.43-0.90)
142/178 (80) 150/179 (84) 4.6 6.4
0.82 (0.65-1.03)
1.5ULN
167/207 (81) 168/210 (80) 4.4 6.3
0.77 (0.62-0.96)
>1.5ULN
36/40 (90)
36/37 (97)
2.1 5.8
0.73 (0.46-1.16)
Taxane
25/32 (78)
26/32 (81)
3.2 6.9
0.55 (0.31-0.98)
Non-taxane non-vinorelbine 156/191 (82) 151/188 (80) 4.4 6.0
0.84 (0.68-1.06)
Vinorelbine
22/24 (92)
2.4 6.5
0.41 (0.22-0.75)
Capecitabine
117/146 (80) 125/151 (83) 4.9 6.2
0.92 (0.72-1.19)
1st-line PFS, mob

LDHb
CT
choiceb
CT choice (not stratified)
45/56 (80)
54/68 (79)
27/27 (100)
BEV-free interval, wk
aStratified. bStratification
Non-capecitabine
86/101 (85)
79/96 (82)
3.2 6.5
0.55 (0.40-0.75)
138/165 (84) 130/149 (87) 4.2 5.8
0.82 (0.64-1.04)
>6
65/81 (80)
0.72 (0.51-1.00)
74/98 (76)
4.4 7.6
factor.
CONCLUSIONS: PFS was statistically significantly improved with the addition of BEV to 2nd-line CT in BEV-pretreated pts,
meeting the primary objective of TANIA. This effect was seen consistently within subgroups based on stratification factors. Within
the limitations of exploratory subgroup analyses with small sample sizes, further subgroup analyses may suggest some potential
inconsistencies in treatment effect. These hypothesis-generating observations require further exploration of potential differences
in disease and pt characteristics in TANIA, which may have influenced physicians CT choice.

Title: Epidemiology of stage IV breast cancer patients: A review of the National Cancer Database 2000-2011
Jessica Gries1, Van Do TH2 and Peter Silberstein1,2. 1Creighton University School of Medicine, Omaha, NE and 2VA Omaha
Medical Center, Omaha, NE.
Body: Background: Breast cancer patients tend to have distinctive characteristics that change as the cancer progresses to stage
IV, especially in race and socioeconomic status. A previous SEER-based study reported blacks were diagnosed with a more
advanced stage and larger tumor size compared to white patients'. This is the largest study to determine multiple factors
associated with patients presenting with stage IV breast cancer.
Methods: A population-based study was conducted using the National Cancer Database (2000-2011), which contains 70% of all
cancer diagnoses in the United States from 1658 American College of Surgeons Accredited-Hospitals. The initial diagnosis of
stage IV disease in women represented 3.56% (81,476) of the total patient population (2,294,058). The demographics of the
stage IV patients were compared to the entire database (all stages) of breast cancer patients using the chi square test.
Results: There was an increased incidence of stage IV Breast Cancer in patients with the following characteristics: Black, no
insurance, Medicaid, Medicare, less educated, lower household income, and higher comorbidity (Table-1). Stage IV cancer
incidence was higher in black patients (17%) compared to all stages (10%). There were 5,137 more black patients with stage IV
than expected. Stage IV patients are three 3x more likely to have no insurance and 2x as likely to have Medicaid when compared
to all stages. Stage IV patients tend to have less education and were twice as likely to have 2+ comorbidities. There was no
increased incidence of stage IV disease in either Hispanics or in different age groups, and only a slight difference in distance
traveled (not shown in table).
Conclusion: The following characteristics are more common in patients with stage IV disease: low income, Black, no insurance,
Medicaid, less education, and higher comorbidity. Patients who were White, had private insurance, higher education and income
status, and lower comorbidity had less representation in stage IV disease. These factors influence the occurrence of a more
advanced stage of the disease. Identifying or more promptly treating patients in high-risk populations may reduce their incidence
of stage IV disease.
'Silber, et, Al. "Characteristics Associated With Differences in Survival Among Black and White Women With Breast Cancer."
JAMA 310.4 (2013): 389. JAMA. Web. 26 Jan. 2014.
Table-1: Epidemiology of Stage IV Breast Cancer Patients (2000-2011)
Stage (%)
All Stages (%)
Observed/ Expected # of Patients**
White
74
80
60,519/ 65,425
Black
17
10
13,611/ 8,474
Hispanic
4,219/ 3,748
None
4,792/ 1,792
Medicaid
10
8,326/ 3,911
Medicare
39
35
32,614/ 28,843
Private
40
54
32,614/ 43,916
>31%
16
11
12,609/ 8,962
23-30.9%
17
15
14,169/ 11,895
18-22%
16
15
12,991/ 12,058
12-17.9%
23
24
18,854/ 19,717
Race/ Ethnicity*
Insurance*
Education* % Without HS Degree*

(Lower value=more education)

<12%
24
30
19,168/ 24,606
<$28,000
12
9,330/ 6,274
$28,000-32,999
13
11
10,422/ 8,718
$33,000-38,999
17
16
14,188/ 13,199
$39,000-48,999
23
23
18,585/ 18,658
>$49,000
31
37
25,269/ 30,391
64
66
51,832/ 54,019
11
8,855 /7,007
2+
3,178 /1,548
Household Income*
(per year)
Comorbidity*
* p-value < .0001 **Observed # of patients / % of all stages in category x total # of stage IV patients (Expected)

Title: Evaluation and clinical impact of intra-tumor heterogeneity (ITH) in primary HER2-overexpressing breast cancers
(HER2+BC) treated with adjuvant trastuzumab and chemotherapy (CT)
Sherene Loi1, Peter Savas1, Ingrid Lonnstedt2, Debora Fumagalli3, Franco Caramia1, Jason Li1, Roberto Salgado3, Andrew
Rowan4, Fabrice Andre5, Carsten Denkert6, Patrick Neven7, Sibylle Loibl8, Christos Sotiriou3, Charles Swanton4 and Terence P
Speed2. 1Peter MacCallum Cancer Centre, East Melbourne, Victoria, Australia; 2Walter and Eliza Hall Medical Research Institute,
Parkville, Victoria, Australia; 3Institute Jules Bordet, Brussels, Belgium; 4Translational Cancer Therapeutics Laboratory, Cancer
Research, London, United Kingdom; 5Gustave Roussy, Villejuif, France; 6Charit Hospital, Berlin, Germany; 7UZ Leuven,
Gasthuisberg, Leuven, Belgium and 8German Breast Group Forschungs GmbH, Neu-Isenburg, Hessen, Germany.
Body: Background: Despite the success of trastuzumab, when added to CT, for the treatment of women with primary HER2+
BC, diversity in clinical responses and survival are still observed. ITH or the presence of subclonal populations is theorized to be
a cause of drug resistance. In this study, we investigated if ITH that could be estimated from a single sample taken at primary
surgery from patients with newly-diagnosed HER2+ disease and if ITH was associated with poorer clinical outcomes despite
adjuvant trastuzumab-based treatment.
Methods: Fifty-two frozen tumor and matched germline samples were taken from patients diagnosed with primary HER2+BC and
treated with adjuvant trastuzumab and CT with a median follow-up of 6.3yrs (range 1.5-13). There were 10 (19%) invasive
relapses, 7 (13%) distant relapses and 5 (9.6%) deaths. We performed tumor and germline whole exome sequencing (WES),
tumor and germline copy number (CN) data profiling (Affymetrix SNP6.0), tumor gene expression (Affymetrix U133 2.0Plus).
Variants were called with MuTect, CNs were called using GISTIC. We developed a novel bioinformatics method that integrated
WES variant allele fractions, purity estimates and SNP data to delineate the presence of single or multiple genetically distinct
subclones. We confirmed our estimated tumor ploidy using flow cytometry on 9 samples. We investigated the association
between the presence of ITH and survival using multivariate Cox regression analyses.
Results: We observed in HER2+ BCs the presence of substantial chromosomal instability, numerous CN alterations and
aneuploidy. With regards to their genomic architecture, 76.9% (40/52) displayed the presence of at least one aneuploid subclone,
where the median ploidy was 3.3 (range:1.7-6.3) copies. Multiple distinct subclones or the presence of ITH could be clearly
detected in 35/52 (67.3%) samples, 8/52 (15.4%) had evidence of only a single clone, whilst in 9/52 (17.3%) samples their
genomic structure could not be definitively determined. The existence of ITH was significantly associated with invasive relapse
after adjustment (HR:4.38;95%CI:1.61-7.14;p=0.002) with estrogen receptor and nodal status also remaining significant in the
Cox modelin contrast, the absence of ITH was associated with excellent outcomes (no distant relapses). Other surrogates of
genomic instability (gene expression signature [CIN70] and SNP signature [Genomic Instability Index]) were not significantly
associated with survival in this dataset.
As exploratory analyses, potential genetic drivers of ITH were investigated. PIK3CA (n=13,p=0.008), MED1 (n=3,p=0.03) and
IKZF1 (n=3,p=0.03) mutations, amplifications in 8q22.2 (n=7,p=0.008) as well as deletion in 1q24.2 (n=5, p=0.006) were
associated with the presence of a single clone, whereas we did not find any genetic drivers with significant associations with ITH.
Conclusions: We show for the first time that the presence of ITH in primary HER2+ tumors is associated with worse outcomes
despite adjuvant trastuzumab and CT. These findings should be evaluated in larger cohorts.

Title: Broad exonic DNA diversity is associated with resistance to taxane-FAC chemotherapy in triple negative breast cancer
Tingting Jiang1, Weiwei Shi1, William F Symmans2, Charles Li1, James Platt1, Rosanna Lau2, Vikram B Wali1, Richard Lifton1,
Lajos Pusztai1 and Christos Hatzis1. 1Yale University, New Haven, CT and 2MD Anderson Cancer Center, Houston, TX.
Body: Purpose: Previous efforts to develop transcriptional markers of chemotherapy sensitivity in TNBC had limited success due
to the heterogeneity of this disease. The purpose of this study was to identify genomic differences between extremely
chemotherapy sensitive and highly chemotherapy resistant TNBC through whole exome sequencing and to assess measures of
genomic heterogeneity as predictive markers.
Methods: Twenty nine cases were selected from a prospectively collected cohort of fine needle aspiration specimens obtained
before preoperative chemotherapy (MDACC) to represent two extreme response cohorts including pathologic complete response
(pCR, N=17) or extensive residual cancer (eRD, N=12). DNA was extracted from specimens stored in RNAlater, exomes were
captured using NimbleGen SeqCap EZ Exome Library preparation and paired-end sequencing of 75 base pair fragments was
performed on Illumina HiSeq 2000. Alignment and variant calling were performed with BWA and GATK haplotype caller. Variants
were filtered against the 1000 Genomes and TCGA normal breast samples to identify candidate somatic variants. Fisher-exact
test was used to identify variant genes associated with sensitivity to chemotherapy. We calculated overall mutational load and
normal-adjusted clonal entropy of driver mutations as measures of genome heterogeneity. The chi-squared test was used to
compare differences in mutational spectra and genome heterogeneity between the two response groups.
Results: The mean coverage was over 150X and 93% of target regions had > 20X coverage. The number of non-silent COSMIC
mutations was similar in tumors from the two response groups (pCR: 63, range 49-82; eRD: 59, range 43-78) as well as the
number of novel non-silent mutations (eRD: 223, range 113-388; pCR: 192, range 125-293). Gene level aggregation of variants
identified 4 genes (MUC21, SLCO5A1, LRBA, STNE1) with response-associated mutational patterns. However, mutations were
non recurrent and p-values were modest, ranging from 0.04 to 0.005. We observed greater overall mutational load and subclonal
heterogeneity (clonal entropy of cancer related mutations) associated with eRD compared to pCR. Both measures suggest that
higher genomic DNA diversity is associated with chemotherapy resistance. However, some genes (BRCA1 and MKI67) had
higher mutational load (sum of minor allele frequencies per gene) associated with pCR compared to eRD. In general, a higher
proportion of C>T transition (P=0.011) and lower A>G transition (P=0.028) was associated with eRD. The same mutational
spectrum shift was previously described in the comparison of trunk and branch driver events suggesting that eRD tumors may
undergo heterogeneous branched evolution.
Conclusion: We observed greater genomic diversity and distinctive mutational spectra in original pre-treatment samples of TNBC
that were associated with extensive residual disease compared to pCR. Our analysis suggests that broad measures of genomic
diversity may serve as markers of resistance to chemotherapy.

Title: Impact of neoadjuvant chemotherapy on the clonal composition of breast cancer
Matthew P Goetz1, Michael T Barrett2, Krishna R Kalari1, Vera J Suman1, Sarah A McLaughlin3, Alvaro Moreno-Aspitia3, Ann M
Moyer1, Donald W Northfelt2, Richard J Gray2, Jason Sinnwell1, Douglas Mahoney1, Poulami Barman1, Peter Vedell1, Xiaojia
Tang1, Kevin Thompson1, Travis Dockter1, Katie Jones1, Sara J Felten1, Amy Conners1, Jeanette Eckel-Passow1, Hughes Sicotte1,
Steven N Hart1, Jia Yu1, Daniel W Visscher1, Eric D Wieben1, Cloann Schultz1, Minetta C Liu1, James N Ingle1, Liewei Wang1,
Richard W Weinshilboum1 and Judy C Boughey1. 1Mayo Clinic, Rochester, MN; 2Mayo Clinic, Scottsdale, AZ and 3Mayo Clinic,
Jacksonville, FL.
Body: Background
Cancer genomic investigations have identified recurrent genomic aberrations critical for cancer initiation, progression, and
metastases. However, these investigations are typically performed in isolation, and the effects of treatment on the clonal selection
of tumor cells are mostly unknown. We hypothesized that molecular profiling of residual tumors after neoadjuvant chemotherapy
(NAC) would identify new drug targets/pathways in patients at high risk for disease recurrence. To better identify clonal
populations of resistant breast cancer cells, we utilized DNA content-based flow sorting of nuclei to identify and isolate clonal
populations for aCGH and next generation sequencing (NGS) both before and after NAC.
Methods
The Breast Cancer Genome Guided Therapy Study (BEAUTY) (NCT 02022202) is a prospective study of patients with high-risk
breast cancer treated with neoadjuvant 12 weekly paclitaxel (T) +/- trastuzumab followed by 4 cycles of anthracycline based
chemotherapy. Tumor tissue from baseline, residual disease from surgery, distant metastases, and patient derived xenografts
(PDX) are obtained for cell sorting by DNA ploidy, aCGH, RNA and exome sequencing.
Results: 140 patients have been enrolled, 104 have completed surgery and 30 unique PDX have been established corresponding
to 26 patients prior to chemotherapy and 4 from residual disease at surgery. Baseline exome and RNA sequencing is complete in
140. Currently, genomic analyses of flow sorted matched baseline, surgical, PDX, and distant disease samples are available in 6
patients. Substantial genomic variation was observed in the surgical sample compared to the primary tumor including gain of
oncogenic drivers (EGFR) and loss of negative regulators (ATG5) (Table). The PDX recapitulated these events with excellent
fidelity compared to the corresponding human tumor. In patients with TNBC, RNA seq obained from matched samples
demonstrated changes in immune related pathways. Evaluation of drug targets/pathways identified in the resistant tumors are
ongoing using the PDX and sequencing of the remaining matched baseline/surgical disease will be reported.
Clonal changes occurring over time in patients with residual disease or disease recurrence after NAC
Tumor
Subtype
Residual
Disease Status
Cancer Burden
TNBC (AR
subtype)
Contralateral lymph node

recurrence at 150 days
2p25.2 - p25.1 amplicon lost at recurrence; 6p21.32 -p21.31

amplicon lost at recurrence
TNBC (BL 1)
Bone/liver/lymph node
recurrence at 135 days
5q11.2 amplicon gained at surgery; 12p13.33 - p13.2 amplicon

gained at surgery
TNBC (BL 1)
Disease-free at day 150
6q21 amplicon lost at sugery
TNBC (BL 2)
Brain recurrence at 390 days
Chr 2 chromothripsis at surgery
Luminal B
Progression during
chemotherapy
7p12.1 - p11.2 amplicon gained at surgery
Disease-free at day 360
9q33.2 amplicon lost at surgery; 15q13.1 - q13.3 amplicon lost

at surgery; 15q22.2 - q24.1 amplicon lost at surgery
Luminal HER2 3
Clonal Aberration Changes (baseline and post NAC)
Conclusions: We observed substantial evolutionary changes in residual breast tumors remaining after NAC. Our findings suggest
that a comprehensive assessment of the mutational landscape that has evolved during NAC can inform drug development in high
risk breast cancer patients.

Title: Reliability of whole exome sequencing for assessing intratumor heterogeneity from breast tumor biopsies
Weiwei Shi1, Anees Chagpar1, Tingting Jiang1, Donald R Lannin1, Brigid Killelea1, Nina Horowitz1, Raymond Lim2, James Platt1,
Charlotte KY Ng2, Vikram B Wali1, Britta Weigelt2, Jorge S Reis-Filho2, Christos Hatzis1 and Lajos Pusztai1. 1Yale University, New
Haven, CT and 2Memorial Sloan Kettering Cancer Center, New York, NY.
Body: BACKGROUND: False positive findings introduced by analytical noise in sequencing and bioinformatics pipelines
constitute a challenge for accurate massively parallel sequencing (MPS). Reports of intratumor genomic heterogeneity based on
MPS rarely estimate the impact of false positive mutation calls. The purpose of this study was to measure apparent genomic
heterogeneity in different regions of the same tumor and to assess the technical noise in variant calling in replicate sequencing of
the same DNA.
METHODS: Three anatomically distinct biopsies were obtained from 3 different regions of 11 breast cancers (33 samples)
including 6 low/intermediate grade, estrogen receptor (ER)-positive and 5 high-grade, triple-negative (TNBC) cancers. DNA from
8 different biopsies was split and independently processed on different days to obtain technical replicates. The NimbleGen
SeqCap EZ Exome Library preparation method was used for exome capture and paired-end sequencing of 75 base pair
fragments was performed on Illumina HiSeq 2000. Read alignment and variant calling were performed with BWA and GATK
haplotype caller. Concordance in variant calls and in minor allele frequencies (MAF) was assessed in the 3 biopsies of the same
tumor and 8 technical replicates. We adjusted for uneven sequence coverage and analyzed known germline variants from
dbSNP, known cancer related variants from COSMIC separately from novel variants (i.e. not previously reported in dbSNP or
COSMIC). We estimated intratumor genomic heterogeneity of genes after removing alterations identified in areas where mapping
is difficult and variant calls that had low analytical reliability in the technical replicates.
RESULTS: The mean coverage was over 150X and > 90% of target regions had 20X coverage. We validated the specificity
(98.2%) and sensitivity (86.7%) of the variant calling pipeline on the GIAB reference data. The concordance for germ line SNPs
and variants in COSMIC in technical replicates was 94.9% and 92.7%, respectively. Novel variants had very low concordance,
55.9%, in technical replicates. The concordance between MAF estimates from the technical replicates was high (0.974, 0.957 and
0.969 for single nucleotide variations, insertions and deletions, respectively). The concordance for germline SNPs and COSMIC
variants in pairwise comparisons of biopsies from the same tumor was 93.2% and 91.1%. For known variants, lower concordance
was observed in TNBC (88.3%-98.5%) compared to ER-positive tumors (93.6%-98.8%, P<0.05) indicating greater intratumor
heterogeneity. We identified variants in a small number of genes (DNAH9, PPM1E, and MAP3K1) that were called inconsistently
in most technical replicates, even after excluding low mappability regions. We assessed intratumor heterogeneity in the triplicate
biopsies, after excluding the technically unreliable variants. On average, two different biopsies from the same tumor shared 14272
+/-1379 common variants and differed in 816 +/- 416 variants.
CONCLUSION: We observed heterogeneity to be slightly greater in high-grade, ER-negative compared to low-grade, ER-positive
breast cancers. Differences in variants observed in multiple biopsies of the same tumor are only slightly greater than those
expected by technical noise alone.

Title: Whole exome sequencing (WES) of HER2+ metastatic breast cancer (MBC) from patients with or without prior trastuzumab
(T): A correlative analysis of TBCRC003
Nikhil Wagle1, Nancy U Lin1, Andrea L Richardson3, Ignaty Leshchiner2, Ingrid A Mayer4, Andres Forero-Torres5, Timothy J
Hobday6, Elizabeth C Dees7, Rita Nanda8, Mothaffar F Rimawi9, Hao Guo1, William T Barry1, Ron Bose11, Wei Shen11, Antonio C
Wolff10, Stacey B Gabriel2, Levi A Garraway1, Eric P Winer1 and Ian E Krop1. 1Dana-Farber Cancer Institute, Boston, MA; 2Broad
Institute, Cambridge, MA; 3Brigham and Women's Hospital, Boston, MA; 4Vanderbilt-Ingram Cancer Center, Nashville, TN;
5
University of Alabama, Birmingham, AL; 6Mayo Clinic, Rochester, MN; 7UNC Lineberger Comprehensive Cancer Center, Chapel
Hill, NC; 8University of Chicago, Chicago, IL; 9Baylor College of Medicine, Houston, TX; 10Johns Hopkins Hospital, Baltimore, MD
and 11Washington University School of Medicine, St Louis, MO.
Body: Background: Although the spectrum of genomic alterations in primary, treatment-nave breast tumors has been
described, the genomic landscape of HER2+ MBC remains underexplored. Furthermore, tumor genomic alterations that arise
after progression on anti-HER2 therapy are largely unknown.
Methods: We prospectively collected metastatic tumor biopsies from patients (pts) enrolled on TBCRC003 (NCT00470704), a
phase II study evaluating the combination of lapatinib (L) and T in pts with HER2+ MBC who had varying degrees of prior T
exposure. We performed WES on baseline metastatic biopsies and normal DNA from 57 pts. In 36 pts, we also performed WES
on pre-treatment primary tumors. Tumors were analyzed for point mutations, insertions/deletions, and copy number alterations.
Results: Total accrual was 116 pts. 87 pts were registered in one of two efficacy cohorts: Cohort 1 included pts w no prior T for
MBC. Pts with prior adjuvant T were included if the interval from last T to 1st recurrence > 12 months. Cohort 2 included pts with
1-2 prior lines of T for MBC or recurrence within 12 months of adjuvant T. An additional 29 pts were enrolled in a biomarker cohort
(Cohort 3). Per-protocol efficacy analyses for 85 pts deemed evaluable are shown below:
Objective Response Rate
Clinical Benefit Rate
Median Time to Progression
Cohort 1
50%
(90% CI 33.8-66.2%)
57.5%
(95% CI 40.9-73.0)
7.4 months
Cohort 2
22.2%
(90% CI 11.2-37.1%)
42.2%
(95% CI 27.7-57.8)
5.3 months
As we previously reported (Wagle et al, ASCO 2014), across 57 metastatic tumors, significant recurrently mutated genes were
TP53 (n=30; 53%) and PIK3CA (n=19; 33%). The frequency of mutant TP53 and PIK3CA was not significantly different from 119
primary, treatment-nave HER2+ tumors sequenced in the TCGA study (50%, p=0.8 and 27%, p=0.5, respectively). Recurrent
copy number alterations were also similar to TCGA data.
Comparing the 38 pts who received any prior T with the 19 pts who did not, there was no significant difference in the incidence of
mutant TP53 (53% vs 53%, p=1.0) and PIK3CA (37% vs 26%, p=0.6).
We identified mutations in the HER2 kinase domain in 4/38 pts who received prior T (11%), as compared to 0/19 T-nave pts.
HER2 kinase domain mutations have been identified in 2% of HER2-negative cancers but <1% of primary HER2+ cancers. 3 of
the mutations were L755S, which has been shown to be resistant to L and sensitive to irreversible HER2 inhibitors. The 4th
mutation was D742N, a novel kinase domain mutation. None of the 4 pts with HER2 kinase domain mutations had an objective
response, though 1 pt had stable disease for 29 weeks.
An analysis comparing paired archival primary tumors and baseline metastatic biopsies from 36 pts to identify genomic alterations
acquired or enriched in the metastatic tumors will be presented.
Conclusions: We present an analysis of the genomic landscape of HER2+ MBC, including comparisons between matched
primary tumors and metastatic biopsies. Somatic HER2 kinase mutations in pts with HER2+ MBC treated with prior T suggests
that these mutations may be involved in resistance to T, and may predict poor response to additional anti-HER2 therapy with
combined L and T. Novel therapeutic approaches may be required for these pts.

Title: ConvertHER: Evolution of genomic alterations from primary to metastatic breast cancer
Ana Maria Gonzalez-Angulo1, Ana Lluch2, Agda K Eterovic1, Angel Guerrero3, Xiaofeng Zheng1, Ramon Perez4, Shuying Liu1,
Jos I Chacn5, Ken Chen1, Silvia Antolin6, Gordon B Mills1, Jaime Ferrer2, Octavio Burgues2, Begona Bermejo2, Elia Munoz7,
Rosalia Caballero8, Eva Carrasco8, Eduardo Martinez9 and Funda Meric-Bernstam1. 1University of Texas MD Anderson Cancer
Center; 2INCLIVA BioMedical Research Institute, Hospital Clnico de Valencia; 3Instituto Valenciano de Oncologa; 4Clnica
Quirn; 5Hospital Virgen de la Salud; 6Complejo Hospitalario Universitario A Corua, A Corua, Spain; 7Hospital de La Plana;
8
GEICAM, San Sebastin de los Reyes and 9Hospital Provincial de Castelln.
Body: Background: Changes in breast cancer receptor status over disease progression and treatment have been described to a
point that could alter response to therapy. There is growing interest in delivering biomarker/genomically-based targeted therapies.
We aimed to determine the concordance of genomic alterations between primary (P) and metastatic (M) breast cancer in a
prospective collection study.
Methods: Targeted capture and next-generation sequencing was performed on formalin-fixed paraffin-embedded (FFPE)
samples, profiling 202 cancer relevant genes in 61 pairs (primary and corresponding recurrence/metastasis). Tumors were
classified at baseline as [hormone receptor (HR)+/HER2-, HR+/HER2+, HER-/HER2+, and triple negative breast cancer (TNBC)].
We aligned data to human reference assembly hg19 using Burrows-Wheeler Aligner's (BWA) and removed duplicated reads. We
identified somatic mutations variants and called copy number alterations (CNA) using an algorithm which reports gain or loss
status of each exon. Alterations potentially targetable with established or investigational therapeutics were considered
"actionable."
Results: Of the 61 cases, 15% changed breast cancer subtype. Of 747 mutations detected in 156 genes, 309 (41%) were
discordant. Median number of mutations were 10 (range 6-11) in P and 8 (range 6-10) in M. Most common mutations occurred at
NOTCH2, PCLO, MAP3K1, MLL3, NOTCH4, CRIPAK, TP53, PIK3CACSMD1, and FLG. Mutations were less common in
HR-/HER2+ tumors in both P and M. Mutation discordance was not different in cases of changed breast cancer subtype (P=.31).
Of 986 CNA detected in 173 genes, 758 (77%) were discordant. There was an increased frequency of EGFR1, ERBB2, FGFR3,
CRIPAK, MEN1 and WT1 amplifications in M. CNA were less common in HR-/HER2+ tumors in both P and M. CNA discordances
were more common in cases of changed breast cancer subtype (P<.0001) and driven by HER2- tumors. Fifty-eight (95%)
patients had actionable alterations that could inform targeted treatment options.
Conclusion: Deep targeted exome sequencing of cancer-related genes revealed potentially targeted alterations. We found 41%
and 77% mutation and CNA discordance between P and M. CNA were more common when breast cancer subtype changed.

Title: Plasma circulating tumor DNA as an alternative to metastatic biopsies for mutational analyses in breast cancer
Michail Ignatiadis1, Franoise Rothe1, Jean-Franois Laes2, Diether Lambrechts3,4, Dominiek Smeets3,4, Delphine Vincent1, Marion
Maetens1, Debora Fumagalli1, Stefan Michiels5, Stylianos Drisis1, Carine Moerman1, Jean-Pol Detiffe2, Denis Larsimont1, Ahmad
Awada1, Martine Piccart1 and Christos Sotiriou1. 1Institut Jules Bordet, Universit Libre de Bruxelles, Belgium; 2OncoDNA,
Gosselies, Belgium; 3Katholieke Universiteit Leuven, Belgium; 4Vesalius Research Center, VIB, Belgium and 5Gustave Roussy,
Univ. Paris-Sud, France.
Body: Background: Molecular screening programs are using next-generation sequencing (NGS) for cancer gene panels in
metastatic biopsies. We interrogated whether plasma can be used as an alternative to metastatic biopsies.
Patients and methods: The Ion AmpliSeqTM Cancer Hotspot Panel v2 (Ion Torrent), covering approximately 2,800 COSMIC
mutations from 50 cancer genes was used to analyze 70 primary and/or metastases and 29 plasma samples from 17 metastatic
breast cancer patients. The targeted coverage for tissue DNA was 1000x and for plasma circulating DNA 25000x. Whole blood
normal DNA was used to exclude germline variants. The Illumina technology was used for independent validation.
Results: Twelve patients had estrogen receptor (ER)+/ human epidermal growth factor receptor 2 (HER2)-, 1 ER+/HER2+, 2
ER-/HER2+ and 2 ER-/HER2- tumors. Evaluable NGS results were obtained for 61 primary/metastases and 29 plasma samples
from 17 patients. When primary/metastases were analyzed, 12 of 17 patients had at least 1 mutation (median 1 mutation per
patient, range 0-2) in either p53, PIK3CA, PTEN, AKT1 or IDH2 gene. When plasma was analyzed, 11 of 17 patients had at least
1 mutation (median 1 mutation per patient, range 0-2) in either p53, PIK3CA, PTEN, AKT1, IDH2 and SMAD4. All
primary/metastases/plasma mutations were independently validated using the illumina technology. When we focused on
metastases and plasma samples collected at the same time point, we observed that in 4 patients, no mutation was identified in
either metastases or plasma, in 9 patients the same mutations were identified in metastases and plasma, in 2 patients a mutation
was identified in metastases but not in plasma and in 2 patients a mutation was identified in plasma but not in metastases
(Table1). Thus, in 13 of 17 (76%) patients, metastases and plasma analysis provided concordant results whereas in 4 of 17
(24%) demonstated discordant results providing complementary information (Table1). Conclusion: Plasma can be tested as an
alternative tissue source in molecular screening programs.
Table Mutational status of synchronous metastatic biopsies and plasma samples analysed using the Ion AmpliSeqTM Cancer
Hotspot Panel v2
Patient ID
Gene
Mutation
Metastasis (MAF)
Plasma (MAF)
PTEN
p.Q171E
YES (27.5%)
YES (25.9%)
SMAD4
p.E394*
NO
YES (10.9%)
NO
NO
4
6
PIK3CA
p.H1047R
YES (20.5-47.7%)
YES (4.6%)
TP53
p.V274A
YES (35.6-56.1%)
YES (20.7%)
IDH2
p.R140R
YES (28.2)
YES (0.5%)
10
PIK3CA
p.H1047R
YES (19.7%)
NO
11
PIK3CA
p.E453K
YES (4.6-17.8%)
YES (2.8%)
11
PIK3CA
p.E453K
YES (13.1-23%)
YES (3.4%)
14
PIK3CA
p.H1047R
YES (24.8%)
NO
14
PIK3CA
p.H1047R
YES (0-13.8%)
YES (0.5%)
16
TP53
p.Y103*
YES (59.8-86.3%)
YES (49%)
17
NO
NO
19
NO
NO
20
PIK3CA
p.E545K
NO
YES (14.3%)
30
TP53
p.M237K
YES (27.6-50%)
YES (51.8%)
37
TP53
p.H193L
YES (61.6-82.9%)
YES (55.5%)
38
AKT1
p.E17K
YES (26-68.2%)
YES (10%)
38
TP53
p.R248W
YES (23.6-56.8%)
YES (5.9%)
39
TP53
p.R136H
YES (7.5%)
NO
NO
NO
40
MAF: Mutant allele frequency; For patients with multiple metastases samples at the same time point, the MAF range is provided.
Patients 11 & 14 had 2 timepoints with synchronous metastases/plasma samples.

Title: Circulating tumor DNA and circulating tumor cells in metastatic triple negative breast cancer patients
Francois-Clement Bidard1, Jordan Madic1, Anna Kiialainen2, Fabian Birzele2, Guillemette Ramey3, Quentin Leroy1, Thomas Rio
Frio1, Virginie Raynal1, Virginie Bernard1, Alban Lermine1, Inga Clausen2, Nicolas Giroud1, Roland Schmucki2, Carsten Horn2,
Olivia Spleiss2, Olivier Lantz1, Marc-Henri Stern1, Martin Weisser4, Ronald Lebofsky1 and Jean-Yves Pierga1. 1Institut Curie, Paris,
France; 2Roche Pharma Research and Early Development (pRED), Innovation Center, Basel, Switzerland; 3Institut Roche de
Recherche et Mdecine Translationnelle, Boulogne-Billancourt, France and 4Roche Pharma Research and Early Development
(pRED), Innovation Center, Penzberg, Germany.
Body: Background
Preliminary reports suggested that circulating tumor DNA (ctDNA) can be used as a prognostic marker in a way akin to circulating
tumor cells (CTC) in metastatic breast cancer patients. However ctDNA detection is often performed on multiple mutations,
combining heterogeneous techniques. Here we used the high prevalence of TP53 mutations in triple-negative metastatic breast
cancer (TNMBC) to compare CTC and ctDNA detection rates and prognostic value.
Methods
A cohort of 40 patients treated at the Institut Curie (Paris, France) was enrolled before starting a new line of treatment for
TNMBC. CTC were detected by the CellSearch system (in 7.5 mL of blood). Using massively parallel sequencing (NGS), TP53
mutations were first characterized in tumor tissue, then in plasma DNA extracted from fresh frozen plasma samples (from 15-20
mL of blood). ctDNA detection was performed using high depth targeted sequencing using two platforms in parallel (Illumina
HiSeq 2500 and Roche 454). Libraries for Illumina were prepared following the TAm-Seq procedure (Forshew et al, Sci Transl
Med 2012), with preamplification of all coding TP53 exons and flanking untranslated regions followed by both paired-end 150bp
Illumina and 454/Roche sequencing. CTC, ctDNA and usual patient characteristics were correlated with time to progression (TTP)
and overall survival (OS).
Results
Archived tumor (FFPE or frozen) tissue was available for 36 patients, and 31 were successfully sequenced: TP53 mutations were
found in 27 patients. As measured on the Illumina platform, ctDNA was detected in 21/27 patients (81%), ranging from 48 to
648,000 copies/mL of plasma (median 1620). Mutant allele fraction in circulating cell-free DNA ranged from 2 to 70% (median
5%). Comparison between ctDNA levels measured by Illumina and 454/Roche platforms in plasma displayed a good correlation
(R = 0.903), with a single discordance. 1 CTC were detected in 19 of these 27 patients (70%). Strikingly, high ctDNA levels had
prognostic impact neither on OS, nor on TTP, whatever the dataset used (Illumina or 454) whereas CTC5/7.5 mL were
correlated with OS (p=0.04) and marginally with TTP (p=0.06). Other known usual factors, such as poor performance status,
elevated LDH and number of previous treatment lines had also significant prognostic factors in this cohort. CTC and ctDNA early
changes during treatment were available for 12 patients and changes (increase/decrease) of the two biomarkers were globally
similar.
Conclusion
Demonstrating a good sensitivity (81%), ctDNA by the TAm-Seq is more frequently detected than CTCs in the 27 TNMBC with
TP53 mutations. The observed correlation between the 2 massively parallel sequencing approaches suggested that ctDNA levels
data were quantitative. In contrast to other usual prognostic factors, baseline ctDNA level did not demonstrate a prognostic
impact,in this proof-of-principle study, suggesting that mechanisms of ctDNA release in TNMBC rely on biological features that do
not dramatically impact patients outcome.

Title: Timing, severity and risk factors for arthralgia in the IBIS-II trial: A retrospective and exploratory analysis
Ivana Sestak1, Anthony Howell2, John F Forbes3, Patrick Neven4 and Jack Cuzick1. 1Centre for Cancer Prevention, Queen Mary
University, London, United Kingdom; 2Institute of Cancer Sciences, University of Manchester, Manchester, United Kingdom;
3
School of Medicine and Public Health, University of Newcastle, Newcastle, Australia and 4UZ Leuven, Leuven, Belgium.
Body: Background: Arthralgia is a well known side effect of aromatase inhibitors and low oestrogen levels and postmenopausal
status are associated with this event. Anastrozole reduced the incidence of oestrogen receptor positive, invasive breast cancer by
58% in the IBIS-II trial. However, timing, severity and risk factors for arthralgia have not been assessed in detail in this trial.
Methods: The IBIS-II trial randomised postmenopausal women at high risk to receive 1mg anastrozole or matching placebo for 5
years. Date of occurrence of arthralgia along with severity (mild, moderate, severe) were recorded at each yearly follow-up visit.
Age, body mass index (BMI), and previous hormone replacement therapy (HRT) were investigated as potential risk factors for
arthralgia. All analyses were done by the use of logistic regression.
Results: 3864 postmenopausal women (anastrozole: 1920, placebo: 1944) were enrolled in the IBIS-II trial. 58.5% of women
randomised to anastrozole reported arthralgia at any time during the trial compared with 52.8% on placebo (OR=1.26 (1.11-1.43),
P=0.0004). The majority of arthralgias were reported within the first 18 months of randomisation, with a decline thereafter (Table).
17.5% of women who reported arthralgia withdrew from the trial compared to 13.9% without any of these symptoms (OR=1.31
(1.10-1.57)), and the withdrawal was significantly greater for those with severe symptoms compared to mild (OR=5.97
(4.27-8.33)). Women who used HRT before trial entry had a significant higher risk of developing arthralgia than their counterparts
irrespective of allocated treatment (OR=1.45 (1.27-1.64), P<0.001). Increasing BMI (lowest vs. highest BMI group: OR=1.30
(1.11-1.53)) and age (lowest vs. highest age group: OR=1.23 (1.01-1.50)) were also significant risk factors for arthralgia. HRT and
BMI remained highly significant in a multivariate model.
Conclusion: Arthalgia was common in the IBIS-II trial irrespective of treatment. However it increased in severity in the active
treatment arm mainly in the 18 month period after randomisation. Severe arthralgia led to significantly more withdrawals from the
trial than mild symptoms irrespective of treatment. Major risk factors for arthralgia in both arms were previous HRT use and
obesity.
Number and percentages (%) of arthralgia of any severity according to treatment allocation and time point.
Overall
0-18 months
18-30 months
30-42 months
42-54 months
>54 months
1920
1150
842
673
543
401
Number with arthralgia 1123
770
170
86
53
44
Percentage (%)
58.5
67.0
20.2
12.8
9.8
11.0
1944
1294
979
782
618
481
Number with arthralgia 1026
650
153
109
69
45
Percentage (%)
52.8
50.3
15.6
13.9
11.2
9.4
OR (95% CI) (A vs. P)
1.26 (1.11-1.43) 1.33 (1.17-1.52) 1.29 (1.01-1.65) 0.92 (0.67-1.25) 0.87 (0.59-1.29) 1.17 (0.74-1.86)
Anastrozole
Women at risk
Placebo
Women at risk

Title: Cardiovascular toxicity following aromatase inhibitor use in 13,273 survivors cared for in a HMO
Reina Haque1, Joanne E Schottinger1, Jiaxiao Shi1, Joanie Chung1, Chantal Avila1, Britta Amundsen1 and Rowan T Chlebowski2.
1
Kaiser Permanente Southern California, Pasadena, CA and 2Los Angeles BioMedical Research Institute at the Harbor-UCLA
Medical Center, Torrance, CA.
Body: Background
Aromatase inhibitors (AIs) reduce breast cancer incidence in primary prevention trials (MAP3, IBIS2). However, controversy
regarding AIs influence on cardiovascular disease (MI, angina, and cardiac failure) (Amir et al JNCI 2011) could limit use in
prevention settings.
Methods
We assembled a cohort of 13,273 postmenopausal breast cancer patients initially CVD (cardiovascular disease)-free at diagnosis
in a large managed care organization. Women were diagnosed 1991-2010, and followed through 2012. The outcome, CVD risk
was compared across endocrine treatments (AI, tamoxifen [TAM], both, or neither). Information on demographics, comorbidity
(diabetes, hypertension, etc.), and covariate medications (antihyperlipidemics, antihypertensives, and other CVD drugs) were
available from electronic medical records. We conducted Cox models using time-dependent endocrine drug use variables
adjusted for age, demographics, comorbidity, and CVD drug use, cancer treatment, tumor characteristics and tumor laterality.
Results
Among the 13,273 cohort, postmenopausal women who used AIs exclusively had a similar risk of ischemic disease (HR=0.97,
95% CI: 0.78-1.22) and stroke (HR=0.97, 95% CI: 0.70-1.33) versus those who used TAM only (HR=1.00, reference). However,
women who used AIs only had a higher risk of other CVD disease combined (CHF, cardiomyopathy, dysrthymia, valvular
dysfunction, pericarditis) (HR=1.26, 95% CI: 1.11-1.43) than those exposed to TAM only. The risk of other CVD disease was
greater among women exposed to sequential TAM and AI treatment. The results are based on 3,711 CVD events occurring in
72,886 woman-years of follow-up.
Tamoxifen Only
AI Only
Both Tam and AI
No Hormones
HR (95% Cl)
HR (95% Cl)
HR (95% Cl)
HR (95% Cl)
Ischemic
1.00 (ref)
0.97 (0.78 - 1.22)
1.04 (0.83 - 1.29)
1.02 (0.84 - 1.23)
Stroke
1.00 (ref)
0.97 (0.70 - 1.33)
1.02 (0.75 - 1.37)
0.93 (0.70 - 1.24)
Other CVD
1.00 (ref)
1.26 (1.11 - 1.43)
1.28 (1.13 - 1.44)
1.23 (1.09 - 1.38)
COMPOSITE CVD
1.00 (ref)
1.15 (1.04 - 1.28)
1.19 (1.07 - 1.31)
1.14 (1.04 - 1.25)
MUTUALLY EXCLUSIVE CATEGORIES
Based on a subset of 7,982 patients who underwent breast irradiation, the risk of CVD overall was greater among women who
used AIs only and received left-sided irradiation (HR=1.21, 95% CI: 1.02-1.44).
Tamoxifen Only
AI Only
Both Tam and AI
No Hormones
HR (95% Cl)
HR (95% Cl)
HR (95% Cl)
HR (95% Cl)
1.00 (ref)
1.16 (1.00 - 1.34)
1.20 (1.04 - 1.38)
1.08 (0.95 - 1.23)
1.00 (ref)
1.21 (1.02 - 1.44)
1.16 (0.98 - 1.37)
1.14 (0.93 - 1.40)
Right-sided breast irradiation

COMPOSITE CVD
Left-sided breast irradiation
COMPOSITE CVD
Discussion
These results indicate that variation exists in the type of CVD events that occur in breast cancer patients receiving AIs in
comparison to tamoxifen users. For example, the risk of ischemic disease or stroke was not elevated in those who used AIs only
versus TAM users. However, overall CVD events were greater in women who used AIs only (or sequentially after TAM),
especially if they received left-sided breast irradiation. While these observational study results require cautious interpretation, they
provide a basis for comparing the benefits and risks of endocrine treatments.


Title: Prevalence, health care utilization and costs of concomitant depression among breast cancer survivors
Diana D Jeffery1. 1Defense Health Agency, Healthcare Operations Directorate, Clinical Support Division, Falls Church, VA.
Body: Purpose: Concomitant conditions in a cohort of non-elderly breast cancer survivors (BCS) were examined with a focus on
the prevalence, healthcare utilization, and costs of diagnosed depression. Little research has been conducted on how
concomitant depression among cancer survivors impacts utilization and costs of care.
Methods. Using administrative claims data from the Military Health System Data Repository (MHSDR), a cohort of 2,851 BCS
was identified with at least 2 years survival from the time of diagnosis. Concomitant conditions were based on ICD-9 codes;
codes 296.2, 296.3, 298.0, and 311 were used to identify depression. Fiscal year 2009 was used as the index year to calculate
healthcare utilization and costs. Bivariate analyses and logistic regression analysis were used to examine group differences and
predictors of having received a diagnosis of depression.
Findings. The most common concomitant chronic conditions in the BCS cohort were hypertension (50.0%), mood disorders or
adjustment disorders (37.5%), heart disease (23.0%), diabetes (19.9%), history of tobacco use (19.7%), asthma or chronic
obstructive pulmonary disease (16%), and obesity (16.8%). About 15.9% of the BCS were diagnosed with depression in the year
prior to, at the time of, or in the 2-year follow up period after the cancer diagnosis. With bivariate analysis, significant differences
were found between BCS with and without depression: those with depression had higher mean number of hospital stays (.33 vs
.11), mean number of bed days (1.94 vs .58), mean number of ambulatory visits (34.26 vs 20.42), and mean number of pharmacy
prescriptions (45.49 vs 27.60). For follow up care, BCS with a diagnosis of depression cost, on average, $7174 more annually
then those without a diagnosis of depression ($15,471 vs $8,297). No demographic characteristics significantly increased the
likelihood of having received a diagnosis of depression.
Discussion. The results show much higher annual health care utilization and costs for BCS diagnosed with depression compared
to BCS without a diagnosis of depression. These findings may reflect the health care plan provided to military-related
beneficiaries, a plan that has few restrictions for cancer follow up care if medically ordered. Claims data contains no information
about cancer stage, a correlate of health care utilization and costs. Overall, the findings provide empirical evidence that there is a
fiscal incentive to screen and manage mild symptoms prior to patients meeting diagnostic criteria for clinical depression.
Assumedly, timely screening and rapid intervention will lead to improved quality of life for the patient, decreased utilization of
health care resources, and cost savings for health care plans. In this respect, the findings support the 2014 ASCO
recommendations regarding screening and treatment for depression and anxiety. Conversely, the data intimate that adoption of
the ASCO recommendations will lead to higher costs as more individuals are positively screened and referred for treatment. Who,
what, or how such services will be afforded merits sustained inquiry.
The opinions expressed herein are those of the author and are not necessarily representative of the opinions or policies of the
Department of Defense.

Title: Blood draws, injections, blood pressure readings in the at-risk arm, and flying might not be associated with increases in arm
volume: A prospective study
Chantal M Ferguson1, Cynthia L Miller1, Nora Horick2, Melissa N Skolny1, Meyha N Swaroop1, Lauren S Jammallo1, Jean A
O'Toole3, Michelle C Specht4 and Alphonse G Taghian1. 1Massachusetts General Hospital, Boston, MA; 2Massachusetts General
Hospital, Boston, MA; 3Massachusetts General Hospital, Boston, MA and 4Massachusetts General Hospital, Boston, MA.
Body: Introduction:
Breast cancer related lymphedema (BCRL) is a swelling caused by compromise of the lymphatic system after breast cancer
treatment. Commonly-cited risk factors include treatment related variables such as axillary lymph node dissection (ALND) and
regional lymph node radiation (RLNR), and patient characteristics including BMI. Patients are often advised to avoid blood draws,
injections, and blood pressure cuffs on their at-risk arm, airplane travel, and extensive exercise to reduce the risk of developing
BCRL; however, data demonstrating the efficacy of such avoidance strategies do not exist. We sought to determine the impact of
blood draws, injections, and blood pressure readings in the at-risk arm, and flying on increases in arm volume in a large,
prospective cohort of patients.
Methods:
522 patients who underwent treatment for unilateral breast cancer between were included. Patients were prospectively screened
for BCRL with Perometer arm measurements pre-operatively, post-operatively, and at 3-8 month intervals thereafter. At each
measurement patients were asked to report number of blood draws, injections, and blood pressure readings in the at-risk arm,
and number of flights since the last measurement, and their responses were assessed for association with relative volume
change (RVC). RVC was analyzed as a continuous variable for association with risk factors.
Results:
522 patients with 2033 post operative measurements were included. Patients were followed for a median of 23 months and 4
post-operative measurements, with a minimum of 1 post-operative measurement and a maximum of 14. 5.56%. 76.8% (401/522)
underwent lumpectomy, 23.2% (121/522) underwent mastectomy. 70% (366/522) underwent sentinel lymph node biopsy, and
19% (98/522) underwent ALND. 62.4% (352/521) received radiation to the breast/ chest wall only, and 21.5% (112/521) also
received regional lymph node radiation. By univariate analysis, there was no significant association between RVC increase and
undergoing one or more blood draws (p=0.36), blood pressure (p=0.88), injections (p=0.79), or number of flights (p=0.89). ALND
was significantly associated with increases in arm volume (p=0.0017) by univariate analysis and older age at diagnosis was
associated with increased RVC with borderline significance (p=0.059).
Number of assessments with and without risk events since last measurement
# since last measurement
Blood Draw
Blood Pressure
Injection
Number of flights
None
91.2% (1796/1969)
83% (1633/1967)
97.6% (1923/1969)
72% (1461/2031)
1 or more
8.8% (173/1969)
17% (334/1967)
2.4% (46/1969)
28% (570/2031)
Conclusions:
In our patient population, non-treatment related risk factors including blood draws, blood pressures, and injections in the at-risk
arm, and flying were not significantly associated with increases in arm volume. This data can be used to help improve and refine
patient education regarding the importance of risk-reducing practices after breast cancer treatment.

Title: Characteristics associated with nonadherence to medications for hypertension, diabetes, and dyslipidemias among breast
cancer survivors
Gregory S Calip1,2 and Denise M Boudreau2,3,4. 1University of Illinois at Chicago, Center for Pharmacoepidemiology and
Pharmacoeconomic Research, Chicago, IL; 2University of Washington, Seattle, WA; 3University of Washington, School of
Pharmacy, Seattle, WA and 4Group Health Research Institute, Seattle, WA.
Body: Background: Comorbidity among breast cancer (BC) survivors increases risk of overall mortality and recent reports have
described some comorbid conditions to possibly influence risk of second BC events and BC-specific mortality. Therefore, clinical
management of chronic conditions such as hypertension (HTN), diabetes mellitus (DM) and dyslipidemias in BC survivors may be
important to both cancer- and non-cancer-related outcomes. Medication adherence for chronic conditions such as diabetes is
poor in general populations and quality of care may be further impacted by BC diagnosis/treatment. The objective of this study
was to describe characteristics associated with nonadherence to medications to treat HTN, DM and dyslipidemias among BC
survivors enrolled in a large, integrated health plan.
Methods: Retrospective cohort of 4,216 BC survivors, the Commonly Used Medications and Breast Cancer Outcomes (COMBO)
study at Group Health Cooperative. Women in our analysis were diagnosed with stages I-II BC between 1990-2008 and alive and
without recurrence or second primary BC in the year 365 days post-BC diagnosis (days 366-730). Medication users were
identified by 1 dispensings of antihypertensives, oral DM medications, and/or statins. Data on incident BC, patient
characteristics, and medications was obtained via linkage to the western Washington SEER registry, medical record and
automated health plan data including pharmacy dispensing records. Medication adherence was measured using medication
possession ratio (MPR) and classified as non-adherent if MPR<0.80. We estimated odds ratios (ORs) and 95% confidence
intervals (CIs) for nonadherence vs adherence to antihypertensives, oral DM medications and statins by various characteristics
using multivariate logistic regression.
Results: We identified 1,929 prevalent users of antihypertensives (n=1,779), DM medications (n=499) and/or statins (n=1,072).
37% were non-adherent to antihypertensives; 75% were non-adherent to DM medications; 39% were non-adherent to statins. In
adjusted models, younger women (ages <50, 50-64) were more likely to be non-adherent to all 3 therapeutic classes compared to
older women (ages 65). Women who received radiation therapy (OR=1.21 95% CI 1.00-1.47) or endocrine therapy (OR=1.25
95% CI 1.03-1.52) were more likely to be non-adherent to antihypertensives; women treated with chemotherapy (OR=1.67 95%
CI 1.03-2.69) were more likely to be non-adherent to DM medications. Greater BMI (P=0.001) and more frequent primary care
provider visits (2 vs 0-1 only; OR=0.30 95% CI 0.24-0.38) were associated with better adherence to antihypertensives. Likewise,
higher Charlson comorbidity scores (2 vs <2) were associated with greater adherence to DM medications (OR=0.49 95% CI
0.23-0.83) and statins (OR=0.54 95% CI 0.28-1.02).
Conclusion: In this population-based cohort of BC survivors, nonadherence to medications for HTN, DM, and dyslipidemias was
associated with younger age, and impact of specific BC treatments on medication adherence varied by therapeutic indication.
Additional research is warranted to target patients in need of medication management as well as to explore patient preferences
and provider factors that may influence medication adherence.

Title: A randomized placebo-controlled trial of acupuncture and gabapentin for hot flashes among breast cancer survivors
Jun J Mao1, Sharon X Xie1, Marjorie A Bowman2, Deborah Bruner3, Susan Q Li1, Angela DeMichele1 and John T Farrar1.
1
Perelman School of Medicine at University of Pennsylvania, Philadelphia, PA; 2Wright State University School of Medicine,
Dayton, OH and 3Emory University School of Nursing, Atlanta, GA.
Body: Purpose
Hot flashes are a common and debilitating symptom negatively affecting the quality of life of breast cancer survivors. We sought
to compare the short and long term effects of electro-acupuncture (EA) vs. gabapentin for hot flashes among breast cancer
survivors.
We conducted a randomized controlled trial of EA vs. gabapentin vs. placebos (sham acupuncture [SA] or placebo medication) in
women with breast cancer who had completed primary cancer treatments and experienced bothersome hot flashes twice daily or
greater. Acupuncturists performed ten EA/SA treatments over eight weeks using a manualized protocol with 2 Hz
electro-stimulation delivered by a TENS unit. Acupuncturists administered SA using Streitberger (non-penetrating) needles at
non-traditional acupuncture points without electro-stimulation. Gabapentin (900 mg daily) or placebo medication were continued
for eight weeks and then weaned off. The primary endpoint was the hot flash composite score measured by the daily diary at the
end of the intervention (Week 8). A secondary endpoint, durability of response, was evaluated at Week 24 from randomization.
Longitudinal mixed effects models were used to evaluate change in outcomes over time and group differences.
Results
Of 120 randomly assigned patients, the mean age was 52.3, 75% were White, 12.5% were peri-menopausal, and
20%/25%/37.5% had natural/surgically/chemically induced menopause, respectively. By Week 8, significant group differences
were observed. Mean reduction in hot flash composite scores was greatest in the EA group, followed by SA and gabapentin, with
placebo medication having the lowest reduction in hot flashes (-7.4 vs. 5.9 vs. 5.2 vs. 3.4, p=0.0003). By Week 24 and off
treatment, reduction in hot flashes in the EA group persisted whereas the hot flashes in the gabapentin group retuned to baseline.
The reduction in hot flashes was greatest in the EA group, followed by SA, placebo medication, and gabapentin (-8.5 vs. 6.1 vs.
4.6 vs. 2.8, p=0.0024). No serious adverse events were reported in any groups. The gabapentin group had the highest
percentage of participants reporting treatment- related adverse events followed by placebo medication, EA, and SA (48.4% vs.
29.0% vs. 19.3% vs. 3.2%, p=0.004).
Conclusion: Electro-acupuncture was more effective than gabapentin, sham acupuncture, or placebo medication in reducing hot
flashes for breast cancer survivors both short term and long term with few side effects. Gabapentin produced significant short
term reduction in hot flashes that did not persist off medication and was associated with more side effects.
Clinical Trial Registration: NCT01005108.

Title: Results from the phase 2 trial of ridaforolimus, dalotuzumab, and exemestane compared to ridaforolimus and exemestane
in advanced breast cancer
Hope S Rugo1, Olivier Tredan2, Jungsil Ro3, Serafin Morales4, Antonino Musolino5, Noemia Afonso6, Marta Ferreira7, Kyong Hwa
Park8, Javier Cortes9, Antoinette R Tan10, Joanne L Blum11, Lamar Eaton12, Christine K Gause13, Adelle (Zhen) Wang14, Ellie Im15,
David J Mauro12 and Jos Baselga16. 1University of California, San Francisco Medical Center, San Francisco, CA; 2Centre Lon
Brard, Lyon, France; 3National Cancer Center, Goyang-si, Gyeonggi-do, Korea; 4Hospital Arnau de Vilanova, Lleida, Spain;
5
University Hospital of Parma, Parma, Italy; 6Instituto Portugus de Oncologia Porto, Porto, Portugal; 7Instituto Portugus de
Oncologia Francisco Gentil, Porto, Portugal; 8Korea University Ansan Hospital, Seoul, Korea; 9Vall d'Hebron University Hospital,
Institute of Oncology, Barcelona, Spain; 10Rutgers Cancer Institute of New Jersey, New Brunswick, NJ; 11Baylor Sammons Cancer
Center, Dallas, TX; 12Merck & Co., Inc, Whitehouse Station, NJ; 13Merck Research Laboratories, North Wales, PA; 14Merck Sharp
& Dohme, Beijing, China; 15Merck & Co., Inc, Kenilworth, NJ and 16Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Introduction: The combination of a mammalian target of rapamycin (mTOR) inhibitor and an aromatase inhibitor has been
shown to significantly increase progression-free survival (PFS) in patients with estrogen receptor-positive (ER+) advanced or
metastatic breast cancer. Ridaforolimus is an alternative mTOR inhibitor with high potency and specificity. We hypothesized that
triplet therapy with ridaforolimus, dalotuzumab (a humanized monoclonal antibody targeting the IGF-1 receptor [IGFR]), and
exemestane (R/D/E) would be more effective than doublet therapy with ridaforolimus and exemestane (R/E).
Methods: This phase 2, randomized, open-label trial enrolled 80 postmenopausal patients who had high-proliferation (KI67
staining) ER+ breast cancer that had progressed following treatment with a nonsteroidal aromatase inhibitor. Patients received
either triplet therapy, at the previously determined maximum tolerated dose of oral ridaforolimus 10 mg QD5, dalotuzumab 10
mg/kg/week IV, and oral exemestane 25 mg/day (R/D/E, n=40), or doublet therapy with R 30 mg QD5 and E 25 mg/day (R/E,
n=40). Dose increases of R to 20 or 40 mg QD5 were permitted in the R/D/E or R/E arms, respectively, in the absence of grade
2 stomatitis after cycle 1. The R dose could be reduced in either arm for toxicity. The primary endpoint was PFS in the ITT
population by central review. Adverse events (AE) of clinical interest (Tier 1) included stomatitis, pneumonitis, hearing loss, and
hyperglycemia.
Results: Baseline characteristics were balanced between treatment groups. The median PFS was 23.3 (95% CI, 8.71, 38.43)
weeks for R/D/E versus 31.9 (95% CI, 16.00, 39.29) weeks in the R/E arm (hazard ratio, 1.18; 80% CI, 0.81-1.72; P=0.565). All
patients experienced at least one AE. 5 (12.8%) and 3 (7.5%) patients in the R/D/E and R/E arms, respectively, discontinued the
study because of AE. Serious drug-related AE occurred in 2.6% of the R/D/E arm and 15% of the R/E arm. Dose modifications
due to AE occurred in 10.3% and 50% in the R/D/E and R/E arms, respectively (difference -39.7%; 95% CI, -56.7, -20.4). Tier 1
AE were primarily grade 1-2 in severity. Stomatitis occurred in 76.9% (30/39 patients) in the R/D/E arm vs 95.0% (38/40 patients)
in the R/E arm (P=0.021), and grade 3-4 stomatitis was similar between arms (23.1% vs 25%). Pneumonitis occurred in 5.1% vs
22.5% (P=0.027) and hearing loss occurred in 1 patient in each treatment arm (2.6% vs 2.5%), all grade 1-2. Hyperglycemia
occurred at a similar rate in both treatment arms (28.2% vs 27.5%), with grade 3-4 events in 4 (10.3%) and 3 (7.5%) patients in
the R/D/E and R/E arms, respectively.
Conclusions: The combination of R 10 mg QD5, D, and E did not improve PFS when compared to R 30 mg QD5 plus E. The
incidence rates of AE were lower in the R/D/E arm than the R/E arm for most categories of adverse events, likely because of the
higher dose of R in the R/E arm. The efficacy reported for R/E in this study is similar to that reported in previous studies
evaluating mTOR inhibitors in combination with exemestane in ABC. Overlapping toxicities and lower doses likely contributed to
the lack of improved PFS with the addition of the IGFR inhibitor to this combination.

Title: Ph1b study of the PI3K inhibitor taselisib (GDC-0032) in combination with letrozole in patients with hormone
receptor-positive advanced breast cancer
Cristina Saura1, Jasgit Sachdev2, Manish R Patel3, Andres Cervantes4, Dejan Juric5, Jeffrey R Infante6, Donald Richards7, Sandra
Sanabria10, Xuyang Lu10, Joseph Ware10, Timothy R Wilson10, Hema Parmar10, Jerry Y Hsu10, Mafalda Oliveira1, Eric P Winer8,
Daniel D Von Hoff2, Jose Baselga9 and Ian E Krop8. 1Vall d'Hebron University Hospital, Vall d'Hebron Institute of Oncology
Barcelona, Barcelona, Spain; 2Virginia G. Piper Cancer Center/TGen, Scottsdale, AZ; 3Sarah Cannon Research Institute/Florida
Cancer Specialists, Sarasota, FL; 4Institute of Health Research INCLIVA, University of Valencia, Valencia, Spain; 5Massachusetts
General Hospital Cancer Center, Boston, MA; 6Sarah Cannon Research Institute and Tennessee Oncology, PLLC, Nashville, TN;
7
Texas Oncology, Tyler, TX; 8Dana-Farber Cancer Institute, Boston, MA; 9Memorial Sloan Kettering Cancer Center, New York,
NY and 10Genentech, San Francisco, CA.
Body: Background: Taselisib (GDC-0032) is a next-generation PI3K inhibitor with increased anti-tumor activity against PIK3CA
mutant (MT) cancers. Taselisib is an orally bioavailable, potent, and selective inhibitor of Class I PI3K alpha, delta, and gamma
isoforms, with 30-fold less inhibition of the PI3K beta isoform relative to the PI3K alpha isoform. Preclinical data show that
taselisib has enhanced activity against PI3K alpha isoform (PIK3CA) MT breast cancer cell lines and enhanced antitumor activity
when combined with letrozole. Clinical data with single-agent taselisib also showed increased tumor shrinkage in patients with
PIK3CA MT breast cancer as compared to patients with PIK3CA wildtype (WT) breast cancer.
Material and Methods: A Phase 1b dose escalation study was conducted with evaluation of taselisib doses ranging from 6-9 mg
QD in combination with letrozole 2.5mg QD in a modified 3+3 design. A dose expansion cohort was conducted with taselisib 6 mg
QD. Safety and tolerability of GDC-0032 was assessed, as well as pharmacokinetics (PK), pharmacodynamic (PD) assessment
by FDG-PET, and anti-tumor activity by RECIST.
Results: As of 31 January 2014, 28 patients were enrolled onto this study with the completion of dose escalation and the dose
expansion cohort. No dose limiting toxicities (DLTs) were observed at either the 6 mg (n = 20) or 9 mg (n = 8) dose levels.
Adverse events (AEs) assessed by the investigator as related to taselisib in 10% of patients (any grade) included diarrhea,
nausea, stomatitis, fatigue, rash, decreased appetite, hyperglycemia, dysgeusia, mucosal inflammation, vomiting, muscle
spasms, asthenia, dry mouth, dry skin, pruritus, and aspartate aminotransferase increased. Grade 3 and 4 adverse events
assessed by the investigator as drug-related and occurring in greater than one patient included diarrhea (14%), hyperglycemia
(7%), and mucosal inflammation (7%). No apparent PK interactions were observed between taselisib and letrozole. The median
number of prior systemic therapies was six, and promising efficacy data has been observed in these heavily pretreated patients.
Metabolic partial responses via FDG-PET ( 20% decrease in mean SUVmax) were observed in 11 out of 18 patients assessed
(61%). Confirmed partial responses by RECIST have been observed at both the 6mg and 9mg taselisib dose levels. For patients
with measurable disease at baseline, the overall response rate of 38% was observed in patients with PIK3CA MT breast cancer
and 9% in patients with PIK3CA WT breast cancer. Updated data on safety, PD, efficacy, and biomarker correlates will be
presented.
Conclusions: The combination of taselisib and letrozole is a well-tolerated regimen with promising preliminary efficacy in PIK3CA
MT breast cancer patients. This preliminary Ph1b clinical data is consistent with taselisib preclinical and single-agent clinical data
showing increased anti-tumor activity for taselisib in PIK3CA MT breast cancer as compared to PIK3CA WT breast cancer.
Taselisib is being further investigated in the neoadjuvant setting in combination with letrozole in the LORELEI study in patients
with untreated hormone receptor-positive breast cancer.

Title: Phase I trial: PI3K inhibitor BYL719 plus aromatase inhibitor (AI) for patients with hormone receptor-positive (HR+)
Payal D Shah1, Mary Ellen Moynahan1, Shanu Modi1, Betty Ann Caravella1, Farrah M Datko2, Stephen Zamora1, Elizabeth
Comen1, Theresa Gilewski1, Steven M Sugarman1, Gabriella D'Andrea1, Diana E Lake1, Shari B Goldfarb1, Sujata Patil1, Anne
Covey1, Michael F Berger1, Mario E Lacouture1, Larry Norton1, Clifford A Hudis1, Jose Baselga1, Sarat Chandarlapaty1 and Maura
N Dickler1. 1Memorial Sloan Kettering Cancer Center, New York, NY and 2Front Range Cancer Specialists, Fort Collins, CO.
Body: Background: Phosphatidylinositol 3-kinase (PI3K) hyperactivation plays a role in endocrine therapy resistance. Adding an
-selective PI3K inhibitor (BYL719) to hormonal therapy may therefore overcome resistance in HR+ MBC. We report results from
a phase I study to evaluate the safety and preliminary efficacy of BYL719 plus an AI in patients (pts) with HR+ MBC. Methods:
This 3+3 dose-escalation trial studied daily oral BYL719 added to standard dose letrozole (L, Arm A) or exemestane (E, Arm B),
and later examined intermittent dosing (Arm C, L + BYL719 every other week; Arm D, E + BYL719 on 5 of 7 days weekly). Pts
with HR+ MBC, any/no PIK3CA mutation, and on L/E were eligible. A cycle (C) was 28 days (d). Endpoints were dose-limiting
toxicity (DLT), tolerability (CTCAE 4.0), and efficacy (RECIST v1.1). Paired tumor biopsies were performed for genomic and
proteomic correlatives. Serial plasma was collected to quantify cell-free (cf) DNA and mutant allele fraction. Results: 14 pts
(median (M) age: 55 (30-69) yrs), 7 each on Arms A and B, received a M of 76d (6-312+) of BYL719 + L or E. All were evaluable
for toxicity, 10 for response. PIK3CA status was mutant(MT)/wild-type(WT)/unknown in 8/5/1 pts. M number of prior MBC
therapies was 2 (1-12) in Arm A, 6 (2-14) in Arm B. Arms had similar toxicities. On Arm A, BYL719 was given at 300mg daily
(DL0) to 3 pts who completed the 28d DLT period. 2 pts had 3 distinct DLTs: maculopapular rash (N=1), hyperglycemia (N=1),
abdominal pain (N=1). Dose was de-escalated (DL-1=250mg) with no DLT in 3 enrolled pts. On Arm B, DL0, 1 pt experienced
DLT (maculopapular rash) of 3 initially enrolled pts. Arm B expansion at DL0 had 1 additional pt with DLT (rash). Clinically
significant, treatment-related toxicities included grade (G)4: none; G3: maculopapular rash (N=8, including 1 pt treated at DL-1),
hyperglycemia (N=1) and G1/2: fatigue (N=7), nausea (N=7), and hyperglycemia (N=6). Toxicity required 6 dose reductions in
4pts and discontinuation in 2 pts. M duration on study for PIK3CA MT vs. WT was 169.5d vs. 69.5d, respectively. In pts with
PIK3CAMT MBC evaluable for response (n=7), 6 had clinical benefit: 1 PR (pt heavily pre-treated, including prior L, MBC to
liver, Arm A, now C10+ after DLT); SD (n=5, included -29.9%, -19%, -12%), and POD (n=1). In pts with PIK3CA-WT MBC
evaluable for response (n=3), 2 had SD (no changes +/-5%) and 1 had POD. Serial cfDNA analysis in 4 pts with SD or PR
demonstrated a decrease of >90% in the PIK3CA mutant allele fraction on treatment. Due to toxicity seen with continuous
BYL719, the study was amended to explore intermittent dosing schedules (Arm C, L; Arm D, E; DL0=250mg), with 5 pts enrolled,
3 of whom have completed the DLT period with no DLTs, and 1 pt with G3 rash. Correlative studies including serial cf DNA
collection from these pts is ongoing. Conclusions: Continuously dosed BYL719 with L or E shows promising antitumor activity.
Skin toxicity warranted evaluation of alternative schedules. Mutant allele fraction may be an early predictor of response and may
serve as a pharmacodynamic marker during intermittent treatment. Safety, efficacy, and correlative data from study Arm C and
Arm D will also be presented.


Title: Phase I study of the PI3K inhibitor BYL719 plus fulvestrant in patients with PIK3CA-altered and wild type ER+/HER2locally advanced or metastatic breast cancer
Filip Janku1, Dejan Juric2, Javier Cortes3, Hope Rugo4, Howard A Burris5, Martin Schuler6, Barbara Deschler-Baier7, Mark R
Middleton8, Marta Gil-Martin9, Jordan Berlin10, Eric Winer11, Douglas Bootle12, Lars Blumenstein12, David Demanse12, Christina
Coughlin13, Cornelia Quadt13 and Jos Baselga14. 1University of Texas MD Anderson Cancer Center, Houston, TX;
2
Massachusetts General Hospital, Boston, MA; 3Vall d'Hebron Institute of Oncology, Barcelona, Spain; 4University of California,
San Francisco, CA; 5Sarah Cannon Research Institute and Tennessee Oncology, PLLC, Nashville, TN; 6West German Cancer
Center, University Hospital, Essen, Germany; 7CCC MF, Uni-Klinikum Wrzburg, Wrzburg, Germany; 8Oxford National Institute
for Health Research BioMedical Research Centre, Churchill Hospital, Oxford, United Kingdom; 9Catalan Institute of
Oncology-IDIBELL, Barcelona, Spain; 10Vanderbilt Cancer Center, Nashville, TN; 11Dana-Farber Cancer Institute, Boston, MA;
12
Novartis Pharma AG, Basel, Switzerland; 13Novartis Pharmaceuticals Corporation, East Hanover, NJ and 14Memorial Sloan
Kettering Cancer Center, New York, NY.
Body: Background:
BYL719 selectively inhibits the -isoform of Class I PI3K. PI3K is encoded by PIK3CA, a frequently altered gene in human
cancers. Preclinical data indicate BYL719 may be more effective in patients (pts) with PIK3CA-altered tumors; however there are
data to suggest that PIK3CA-wild-type (wt) tumors may also be sensitive to BYL719. Here, we present updated data from the
Phase I study of BYL719 + fulvestrant in pts with PIK3CA-altered or -wt ER+/HER2 locally advanced/metastatic breast cancer
(BC) (NCT01219699).
Methods:
Adult women with PIK3CA-altered (mutation or amplification) ER+/HER2 BC received continuous oral BYL719 (300400
mg/day; 28-day cycles) + fixed-dose fulvestrant (500 mg every 4 weeks, plus an additional dose 2 weeks after first dose) during
dose escalation and expansion. Pts with PIK3CA-wt ER+/HER2 BC were enrolled into the dose expansion to receive BYL719
400 mg/day + fulvestrant. A Bayesian logistic regression model with overdose control guided dose escalation. Primary objective:
to determine the maximum tolerated dose (MTD) and/or recommended phase II dose (RP2D) of BYL719 in combination with
fulvestrant, which was declared previously as 400 mg/day. An expansion cohort at the MTD assessed safety (CTCAE v4.0),
tolerability, pharmacokinetics (PK), and preliminary efficacy (RECIST v1.0).
Results:
As of May 2, 2014, 64 pts (PIK3CA-altered n=41; PIK3CA-wt n=16; PIK3CA status unknown/pending n=7) received BYL719
300400 mg/day + fulvestrant. Median number of prior antineoplastic therapies: 5 (range: 112) for pts with PIK3CA-altered
tumors and 5 (range: 416) for pts with PIK3CA-wt tumors. Prior fulvestrant treatment: 19 (46%) and 7 (44%) pts with
PIK3CA-altered and -wt tumors, respectively. Overall, the most common (25%) adverse events (AEs; all grades/all doses)
suspected to be study drug-related were hyperglycemia (41%), diarrhea (34%), nausea (30%), and vomiting (25%). The most
common (>10%) study drug-related Grade 3/4 AEs (all doses) were maculopapular rash (14%) and hyperglycemia (13%).
Preliminary antitumor activity was observed in this trial. At data cut-off, partial responses (PRs) were observed in 2 patients with
PIK3CA-altered tumors evaluable for response (2/33, 6%), but no PRs were observed in the 15 evaluable patients with
PIK3CA-wt tumors. Duration of exposure was >16 weeks in 24 (59%) patients with PIK3CA-altered tumors and in 5 (31%)
patients with PIK3CA-wt tumors. PK and exposure of BYL719 + fulvestrant was similar to that observed with single-agent BYL719
at the same dose levels. At data cut-off, treatment was ongoing in 20 (49%) and 2 (13%) pts with PIK3CA-altered and -wt tumors,
respectively.
Conclusions:
BYL719 + fulvestrant demonstrated a favorable safety profile in pts with PIK3CA-altered and -wt ER+/HER2 BC, with mostly
on-target effects (i.e. hyperglycemia, rash). Preliminary clinical activity was seen in pts with PIK3CA-altered and -wt tumors, but
confirmed PRs were only observed in pts with PIK3CA-altered tumors. The low number of pts with PIK3CA-wt tumors limits
further conclusion.

Title: A phase I study of BKM120 and fulvestrant in postmenopausal women with estrogen receptor positive metastatic breast
cancer
Cynthia X Ma1, Jingqin Luo1, Michael Naughton1, Foluso Ademuyiwa1, Rama Suresh1, Timothy Pluard1, Gayathri Nagaraj1, Kaitlin
Arnold1, Craig Lockhart1 and Matthew J Ellis1. 1Washington University School of Medicine, St Louis, MO.
Body: Background
BKM120, an oral pan-Phosphatidylinositol-3-kinase (PI3K) inhibitor, plus fulvestrant (F) induced synergistic anti-tumor effect in
preclinical studies of estrogen receptor positive (ER+) breast cancer. We therefore conducted a phase I trial of BKM120 and F in
postmenopausal women with ER+ metastatic breast cancer (MBC) to determine the maximum tolerated dose (MTD), tolerability
and preliminary efficacy.
Methods
A 3+3 phase I design was chosen for phase IA with BKM120 administered orally to define MTD [Table 1]. Cycle (C) length was
28 days. F 500mg was administered intramuscularly on C1 day (D)1 and D15 then on D1 of each subsequent cycles. Two
expansion cohorts, phase IB (intermittent dosing: 5 days on and 2 days off) and Cohort C (continuous dosing: daily) of BKM120 at
MTD, was initiated to further assess the tolerability and efficacy. Patients (pts) with ER+ MBC with measurable disease were
eligible. No more than 3 lines of systemic therapy in the metastatic setting were allowed in phase 1B or Cohort C. Tumor
measurement occurred every 3 cycles. Adverse events (AEs) were assessed by CTCAE 4.0 and response by RECIST 1.1.
Results
Thirty one pts, with median age of 58 (range: 34-71) years, prior exposure to a median of 1 (range: 0-9) endocrine and 0 (range:
0-2) chemo regimens, were enrolled. Majority of pts (83%) had visceral metastasis. Thirty, 25, and 22 pts were evaluable for AE,
response and clinical benefit (CB), respectively. Most C1 AEs were grade (G)1 or 2, except 1 G3 diarrhea. No DLT occurred in
C1. However, G2/3 AEs required BKM120 interruption and/or reduction in C2 or beyond occurred in 16 (53%) pts, including 9
ALT/AST elevation (G2 10%, G3 17%, G4 3%), 7 rash (G3 23%), and 1 pt each with G2 confusion, G3 hyperglycemia, G2
pneumonitis, and G3 tremor. As of June 3, 2014, 1 pt withdrew consent, 19 pts discontinued therapy due to progressive disease
(PD) (n=15) or AE (n=4), 11 pts continue to receive BKM120/F. Fourteen (63.6%, 95% CI: 43.0 - 80.3%) of the 22 evaluable pts
derived CB, including 7 partial response (PR) and 7 prolonged stable disease (SD) > 6 months. Archival tumor and circulating
tumor DNA are being analyzed for presentation. Data will be updated.
Table 1 Study enrollment and results
N pt BKM120 interrupted/reduced
(AE)*
PD
PR (Cycle
completed)
SD (Cycle
completed)
NE
1A DL1 3 80 mg daily
2 (confusion, ALT)
1 (22)
1 (a)
1A DL2 6 100 mg daily
3 (ALT, rash)
2 (13,3 (a))
2 (9, 7)
Cohort
N BKM120
1A DL2b 2
100 mg
intermittent
2 (6, 6)
1B
10
100 mg
intermittent
6 (AST, ALT, rash Hyperglycemia)
2 (15+, 12)
5 (16+, 9+, 9, 6, 6)
3 (a, a, b)
10 100 mg daily
5 (rash, ALT, diarrhea, tremor)
3 (9+, 8+, 4+)
3 (3+, 3+, 3+)#
2 (1+, 2+)
16
13
Total (N) 31
*All occurred in C2+ except the diarrhea, CBR: 63%, NE: Not evaluable, a: off due to AE, b: withdrew consent, # NE for CBR
Conclusion
BKM120 100mg, administered daily or intermittently, plus F was tolerable without DLT during C1. Grade 2/3 AST/ALT and rash
were common in both dosing schedules in subsequent cycles resulting in BKM120 interruption/reduction. Promising activity
observed in this trial warrants further development of this combination in ER+ BC. Phase III studies are ongoing in pts with ER+
MBC progressed on aromoatase inhibitor.

Title: PTEN and PIK3CA but not p4EBP1 are associated with low rates of pathological complete response (pCR) to trastuzumab
based chemotherapy in primary HER2-overexpressing breast cancer
Sibylle Loibl1,9, Silvia Darb-Esfahani2, Alexander Klimowicz3, Gunter von Minckwitz1, Bianca Lederer1, Jens Huober4, Arndt
Hartmann5, Holger Eidtmann6, Berit Maria Pfitzner2, Peter A Fasching7, Katharina Tiemann8, Christian Jackisch9, Keyur Mehta1,
Michael Untch10 and Carsten Denkert2. 1German Breast Group, Neu-Isenburg, Hessen, Germany; 2Institute of Pathology, Charit
University of Berlin, Berlin, Germany; 3Functional Tissue Imaging Unit, Translational Labs Tom Baker Cancer Centre, Calgary;
4
University Hospital, Ulm, Germany; 5University Hospital Erlangen; 6University Hospital Schleswig-Holstein, Campus Kiel;
7
University Hospital, Erlangen, Germany; 8University Schleswig-Holstein, Campus Kiel; 9Sana Klinikum Offenbach and 10Helios
Klinikum Berlin-Buch.
Body: Background: Phosphatidylinositol 3-kinase (PI3K)/AKT pathway and phosphatase and tensin homolog deleted on
chromosome 10 (PTEN) aberrations are common in breast cancer (BC). PIK3CA mutation is associated with lower pCR rates
especially in patients treated with double anti-HER2 therapy (Loibl et al. JCO accepted). Another mechanism of resistance for
trastuzumab-based treatment could be loss of PTEN, resulting also in downstream activation of the pathway. We investigated the
correlation between PIK3CA, PTEN, p4EBP1 (phosphorylated E4 binding protein 1) and pCR in pts receiving neoadjuvant
therapy.
Methods: In addition to PIK3CA we retrospectively evaluated PTEN and p4EBP1 by immunohistochemistry in HER2+ patients
who received EC followed by docetaxel within the G4 study (Untch et al. 2011). The G4 study demonstrated a higher pCR rate by
adding trastuzumab to chemotherapy. HER2 and hormone receptors (HR) were centrally assessed. PTEN was assessed using
the automated quantitative immunofluorescence analysis (Aqua) using an antibody (Cell Signalling Technology). p4EBP1
assessed by immunohistochemistry with an immunreactive score ranging from 0-12. PTEN was categorized by an optimized
cut-off determined by the cut-off finder software (http://molpath.charite.de/cutoff/) and p4EBP1 was measured as a continuous
variable and correlated with PIK3CA genotype and pCR (ypT0, ypN0). Central HER2+ve cases with a tumor cell content of 20%
were selected (n=181).
Results: Median age was 48years (22-77); HR+ve 51%; Grade 3, 47.4%; pCR rate 32%. p4EBP1 analysis was available from
137 and PTEN from 108 patients. PIK3CA genotype was available in 83 of these patients. 58 pts had PIK3CA and PTEN
assessable, 14/58 had a PIK3CA mutation (mut) (25%).
Overall, pCR rate in PTEN low tumors was 27.6% vs 57.1% in PTEN high (p=0.010). Within the PIK3CA mut cohort 13/14
(92.9%) tumors were PTEN low. Within the PIK3CA wild-type (wt) cohort 30/44 (68.2%) were PTEN low (p=0.066). The tumours
were grouped into 4 subsets using PIK3CA (mut vs wt) and PTEN (low vs. high). pCR rate was 57.1% (8/14) in PTEN
high/PIK3CA wt cohort and decreased to 15.4% in the PTEN low/PIK3CA mut cohort. The group with either PTEN low and
PIK3CA wt or PTEN high and PIK3CA mut had a pCR rate of 32.2% (p=0.015). In multivariable analysis after adjustment for
baseline parameters PTEN was an independent predictor for pCR in the complete cohort (OR 12.3 [95% 1.82-82.9] p=0.010) and
in PIK3CA wt cohort (OR 10.3 [95% CI 1.31-81.62] p=0.027). Within the HR+ve group PTEN low tumors had a pCR rate of 22.2%
vs 61.5% in the PTEN high group (p=0.007). In multivariable analysis PTEN was independently predictive for pCR in the HR+ve
group (OR 49.5 [95% 3.4-707] p=0.004). p4EBP1 correlated weakly with PIK3CA (p=0.049) but not with PTEN. There was no
association of p4EBP1 with pCR.
Conclusion: Low PTEN was significantly associated with lower pCR and added independent predictive information not only in
the overall HER2+ve cohort but also in the PIK3CA wt and HR+ve cohort. It will be confirmed in a larger sample size if PTEN
assessment adds information to PIK3CA genotype to select patients with low pCR rates after trastuzumab therapy.

Title: HER2/PIK3CAH1047R transgenic mammary tumors develop acquired resistance to triple therapy with trastuzumab,
pertuzumab and PI3K inhibitors via multiple mechanisms
Ariella B Hanker1, Benjamin Bulen1, Monica Red Brewer1, Christian D Young1, Kirsten M Farrar1, Rebecca S Cook1, Thomas P
Stricker1 and Carlos L Arteaga1. 1Vanderbilt University, Nashville, TN.
Body: HER2 amplification and activating mutations in PIK3CA, the gene encoding the p110 subunit of PI3K, often co-occur in
breast cancer. We generated a transgenic mouse model of HER2-overexpressing (HER2+), PIK3CAH1047R-mutant breast cancer.
In these mice, PIK3CAH1047R accelerates HER2-mediated mammary epithelial transformation and metastatic progression, confers
stem cell-like properties to HER2-overexpressing cancers and generates resistance to the combination of trastuzumab and
pertuzumab (Hanker et al. PNAS 2013). HER2+/PIK3CA tumor growth was inhibited by treatment with the HER2 antibodies
trastuzumab and pertuzumab in combination with the pan-PI3K inhibitor BKM120 (TPB). We sought to discover mechanisms of
acquired resistance to the triple therapy by long-term treatment of established HER2+/PIK3CA tumors. We used tumor
transplants derived from two HER2+/PIK3CA transgenic mice, #564 and #635. Tumor transplants from model 564 were initially
growth inhibited by TPB, but did not regress. A subset of 564 transplants (3/11) resumed growth in the presence of continuous
TPB therapy. All transplants (n=9) from model 635 regressed to a volume of <100 mm3 within 6 weeks of treatment. All tumors
recurred and 2 tumors continued growth when re-treated with TPB. Resistance was maintained following passaging in mice and
tumors were cross-resistant to trastuzumab/pertuzumab/BYL719, a p110-specific inhibitor. TPB-resistant tumor 635-2
expressed p95 HER2, which was not detected in untreated tumors. In contrast, HER2 expression was significantly reduced in
TPB-resistant tumor 635-3. P-AKT remained suppressed in some resistant tumors, but was restored in others. Short-term TPB
treatment strongly suppressed P-S6 in sensitive tumors, whereas P-S6 was no longer inhibited in all TPB-resistant tumors from
both models. We are currently performing whole-exome sequencing and RNA-sequencing on TPB-resistant vs. untreated tumors
in order to identify additional mechanisms of resistance. In parallel, we established human HER2+, PIK3CA-mutant cell lines
(MDA-MB 453, UACC893, and HCC1954) resistant to TPB by long-term treatment (>5 months) in the presence of the three
drugs. Similar to the TPB-resistant tumors, P-S6 was no longer inhibited following TPB treatment in the resistant cell lines.
Treatment with the TORC1/2 inhibitor MLN0128 abolished levels of P-S6 in HER2+/PIK3CAH1047R tumors. Combined treatment
with MLN0128 and TPB inhibited growth of the drug-resistant tumors. Interestingly, Both TPB-resistant HER2+/PIK3CAH1047R
tumor lines displayed resistance to the antibody-drug conjugate trastuzumab-DM1 (T-DM1) in vitro and in vivo, despite
maintenance of HER2 overexpression. In addition, HCC1954 cells selected for resistance to TPB in culture were 66-fold less
sensitive to T-DM1 than parental cells, despite maintaining equal levels of HER2 by western blot. These data suggest that
multiple mechanisms may contribute to resistance to dual HER2 and PI3K blockade, including re-activation of mTOR signaling.
We speculate that a similar heterogeneity of resistance mechanisms may occur in HER2+/PIK3CA-mutant metastases in
patients.

Title: The long noncoding RNA M41 promotes aggressiveness and tamoxifen resistance in ER-positive breast cancers
Felix Y Feng1, Teng Ma1, Corey Speers1, Matthew K Iyer1, Shuang Zhao1, John R Prensner1, James M Rae1, Lori J Pierce1 and
Arul M Chinnaiyan1. 1University of Michigan, Ann Arbor, MI.
Body: Background: Long noncoding RNAs (lncRNAs) have recently been associated with the development and progression of a
variety of human cancers. To date, the interplay between known oncogenic drivers, such as estrogen receptor (ER), and lncRNAs
has not been well described. In this study, we identify M41 as the top outlier lncRNA in ER-positive vs ER-negative breast cancer
and investigate its role in preclinical cancer phenotypes and clinical outcomes.
Methods and Materials: RNA sequencing was performed on 89 breast cancer samples and cell lines, including 42 ER+ cases,
and a modified cancer outlier analysis was used to identify lncRNAs enriched in ER-positive disease. To assess ER regulation of
the top enriched lncRNA (M41), ChIP-Seq and ChIP-PCR was used to detect binding of ER to M41 promoter and qPCR was
used to determine changes in M41 expression following 10 nM estradiol treatment in MCF7 and T47D cells. Following knockdown
via siRNA, the impact of M41 expression was assessed on cell invasion, migration, proliferation, and anchorage-independent
growth. The impact of M41 knockdown on tamoxifen sensitivity was assessed by cell proliferation studies in MCF7 cells with
acquired tamoxifen resistance. Lastly, clinical associations between M41 expression and grade/node status, as well as event-free
survival (EFS), was determined using ANOVA and Kaplan-Meier analyses of TCGA samples.
Results: M41, an uncharacterized lncRNA located on chr21q22.2, was identified as the top outlier lncRNA in ER-positive vs
ER-negative breast cancer. M41 demonstrated outlier expression (RPKM values>50) in 15% of ER-positive cancers, and was not
significantly expressed in normal breast tissue. ChIP studies show that ER robustly binds to the M41 promoter. Estradiol
stimulation significantly increased M41 expression in a time-dependent manner. Knockdown of M41 significantly inhibited all
assessed oncogenic phenotypes in the ER-positive MCF7 and T47D cells, with a 60-80% decrease in both invasion and
anchorage-independent growth, but had no effect in the ER-negative MDA-MB-231 cell line (which has minimal M41 expression).
M41 expression was greater than 10-fold higher in tamoxifen-resistant MCF7 cells compared to parental controls (p<0.001), and
knockdown of M41 restored tamoxifen sensitivity on cell proliferation studies; studies on the mechanism of M41-mediated
tamoxifen resistance are ongoing. M41 overexpression was significantly correlated with node positivity, increasing grade, and
luminal B subtype in ER-positive breast cancer samples (p<0.001). In TCGA samples, M41 overexpression was significantly
associated with decreased EFS (p=0.003).
Conclusion: We have identified M41 as an ER-associated oncogenic lncRNA that contributes to preclinical cancer phenotype,
promotes tamoxifen resistance in cell line models, and associates with poor outcomes in clinical samples. We suggest that M41
represents a novel biomarker candidate for the prognosis of ER-positive breast cancers and provides new insight into the
biological complexity of breast tumor biology.

Title: FoxA1 gene amplification in ER+ breast cancer mediates endocrine resistance by increasing IL-8
Xiaoyong Fu1,2,3,4, Rinath Jeselsohn6, Emporia F Hollingsworth5, Dolores Lopez-Terrada5, Chad J Creighton2,3, Agostina
Nardone1,2,3, Martin Shea1,2,3, Laura M Heiser7, Pavana Anur8, Nicholas Wang7, Catie Grasso7, Paul Spellman8, Carolina
Gutierrez5, Mothaffar F Rimawi1,2,3, Susan G Hilsenbeck1,2,3, Joe W Gray7, Myles Brown6, C K Osborne1,2,3,4 and Rachel Schiff1,2,3,4.
1
Lester and Sue Smith Breast Center, Baylor College of Medicine, Houston, TX; 2Dan L. Duncan Cancer Center, Baylor College
of Medicine, Houston, TX; 3Baylor College of Medicine, Houston, TX; 4Baylor College of Medicine, Houston, TX; 5Baylor College
of Medicine, Houston, TX; 6Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA; 7Oregon Health & Science
University, Portland, OR and 8Oregon Health and Science University, Portland, OR.
Body: Background: ER transcriptional programming is associated with fundamental changes when endocrine resistance
develops. The Forkhead transcription factor, FoxA1, is a pioneer factor for ER-DNA binding. We hypothesize that FoxA1 plays a
critical role in ER transcriptional reprogramming in endocrine resistance by augmenting itself and the specific downstream
effectors. Methods: Next generation sequencing was applied to characterize a panel of endocrine-resistant (Endo-R) cell models.
Genomic PCR amplification and FISH assays were developed to measure FoxA1 copy number gain (CNG). Q-RT-PCR, Western
blots, IHC, ELISA, and cytokine arrays were used to determine the levels of FoxA1 and IL-8 in cell culture and in vivo xenograft
tumors. Effects of gene knockdown (ER, FoxA1, or IL-8) or inducible FoxA1 overexpression on ER and growth factor receptor
(GFR) downstream signaling were determined by cell growth and Western blots. ER and FoxA1 binding at the IL-8 gene locus
was measured by ChIP-qPCR. ChIP-seq analysis was integrated with RNA-seq data. Kaplan-Meier analysis evaluated the
predictive role of FoxA1in ER+ breast tumors. Results: Exome-seq revealed that FoxA1 is the most highly amplified gene in
TamR vs. P cells from two independent MCF7 models. Genomic PCR and FISH also indicate FoxA1 CNG in Endo-R models of
ZR75-1 and BT474. Increased FoxA1 expression was found in multiple Endo-R cells and in MCF7L Endo-R xenograft tumors.
Cytokines, especially IL-8, are more highly expressed in multiple Endo-R cell models, similar to our previous microarray data from
MCF7 Endo-R xenograft tumors. FoxA1 forced overexpression significantly induced IL-8 expression in MCF7L-P cells. It also
activated multiple GFR downstream signaling pathways, and conferred endocrine resistance. Conversely, knockdown of either
FoxA1 or ER significantly decreased IL-8 levels in TamR cells, and inhibited cell growth in both P and TamR cells. Knockdown of
IL-8 in TamR cells substantially inhibited GFR downstream signaling, and was more cytotoxic than in P cells. A novel
FoxA1-binding site (10 kb at 5UTR of IL-8) recruited more FoxA1 and p300 in MCF7L-TamR than -P cells. ChIP-seq shows a
general enhancement of FoxA1 binding around the genes (within 20 kb) that are differentially expressed in TamR vs. P cells. We
identified a FoxA1 CNG-associated gene signature from TCGA breast tumors that predicts worse relapse-free survival (RFS) in
Tam-treated ER+ tumors (from Loi et al). Meta-analysis showed that FoxA1 mRNA levels in the top 25th percentile predict worse
RFS in ER+ patients treated with Tam (N=615), but not in systemically untreated patients (N=500). FoxA1 CNG and
overexpression in clinical specimens by using our newly developed FISH and IHC assays are currently being investigated.
Conclusions: FoxA1 gene amplification was enriched in two independent MCF7 Tam-R cell models. Clonal selection of FoxA1
gene amplification may occur and lead to endocrine resistance. High levels of FoxA1 may mediate endocrine resistance by
directly inducing IL-8. The data suggest that IL-8 signaling is a component of a cytokine loop controlled by the FoxA1/ER
transcriptional reprogramming, which might be exploited in therapeutics to overcome endocrine resistance.

Title: Recurrent ESR1 fusion transcripts are associated with endocrine resistance in estrogen receptor positive, HER2 negative
breast cancer
Jennifer M Giltnane1, Justin M Balko1, Thomas L Stricker1, Christian Young1, M Valeria Estrada1, Nikhil Wagle2, Eliezer van Allen2,
X Jasmine Mu2, Violeta Sanchez1, Jaime Farley1, Kerry Fitzgerald3, Armin Graber3, Joseph A Pinto7, Franco Doimi4, Henry
Gmez4, Monica Rizzo5, Thomas B Julian6, Vandana Abramson1, Ingrid Mayer1, Mark Kelley1, Ashwini Yenamandra1, Ferrin C
Wheeler1, Melinda Sanders1, Levi Garraway2, Ingrid Meszoely1 and Carlos L Arteaga1. 1Vanderbilt-Ingram Cancer Center,
Vanderbilt University, Nashville, TN; 2Broad Institute of Harvard & MIT, Cambridge, MA; 3Genoptix, Carlsbad, CA; 4Instituto
Nacional de Enfermedades Neoplsicas, Lima, Peru; 5Emory University, Atlanta, GA; 6Allegheny General Hospital and NSABP,
Pittsburgh, PA and 7Oncosalud-AUNA, Lima, Peru.
Body: Breast cancer proliferation measured by Ki67 immunohistochemistry after short-term antiestrogen therapy has been
shown to correlate with disease-free survival. This suggests the use of biomarkers of the early effects of endocrine therapy on
ER+ tumors will identify resistant cancers. Thus, we hypothesized that profiling operable ER+ tumors after short term treatment
with an aromatase inhibitor would discover actionable molecular alterations causally associated with resistance to estrogen
deprivation.
We performed whole exome sequencing, RNA-Seq and quantitative immunofluorescence (QIF) of ER, PR, HER2, and Ki67 in
biopsies from 130 patients with an operable ER+/HER2 breast cancer that had received letrozole for 10-21 days prior to surgery.
Tumors were categorized by the natural log of 2-week post-letrozole Ki67 as sensitive, intermediate, or resistant.
We sequenced RNA from 50 frozen tumors and performed fusion transcript analysis using 4 programmatic algorithms (dRanger,
TopHat, DeFuse, Chimera Scan), resulting in 304 candidate gene fusions in 44 tumors. Primers with universal sequencing tags
were designed against 3 and 5 sites of breakpoints mapping to RefSeq exon coding regions (n=187); fusion sequences were
amplified by qRT-PCR from tumor and breast cancer cell line RNA. Single or multiple distinct product bands were visualized by
gel electrophoresis in 96 tumor samples and Sanger-sequenced. Results were mapped to the human RNA reference
transcriptome using BLAST.
Overall, 9% of putative fusion transcripts (n=27 from 16 unique tumors) were validated by mapping to the open reading frames of
predicted 3 and 5 genes. Fusion transcripts called by more than one program were more likely to validate (13 of 24 redundant
versus 14 of 269 unique; p<0.001). ESR1 fusions in 4 tumors mapped to chromosome 6q25.1, involving the 5 UTR of ESR1 and
3 exons of AKAP12, c6orf211, and CCDC170 (c6orf97). The ESR1:CCDC170 fusion was also detected in MDA361 and MCF7
cells as previously published, as well as in BT474 cells. FISH for multiple probes at 6q25.1 demonstrated structural
rearrangements but not amplification in primary tumors and breast cancer cell lines.
Using the 2-week Ki67 to stratify for response to treatment, the validated ESR1 fusions were present only in tumors that
maintained high (7.4%) to intermediate (>2.7%) Ki67 labeling indices upon estrogen deprivation with letrozole (p=0.01). PR
expression was lower (p=0.003) and ER expression higher (p=0.05) in ESR1 fusion+ tumors compared to fusion negative tumors.
RNA extracted from 14 additional tumors were screened for ESR1 fusions by qRT-PCR and the ESR1:CCDC170 fusion was
validated in 1 of 8 resistant/intermediate and 0 of 6 sensitive tumors.
In summary, biomarkers of early response to antiestrogens are needed in order to identify ER+ cancers that are treatment
resistant. In a prospective trial of operable ER+/HER2 breast tumors, we discovered recurrent intrachromosomal ESR1 fusion
transcripts associated with intrinsic resistance to estrogen deprivation with letrozole. Additional work investigating the genomic
basis and function of the fusion transcripts is underway.

Title: ESR1 gene fusions implicated in endocrine therapy resistance of ER+ breast cancer
Jieya Shao1, Jin Zhang1, Robert J Crowder1, Rodrigo Goncalves1, Chanpheng Phommaly1, Breast AWG and Network4, Charles M
Perou2, Christopher A Maher1, E Aubrey Thompson3 and Matthew J Ellis1. 1Washington University School of Medicine, St Louis,
MO; 2University of North Carolina, Chapel Hill, NC; 3Mayo Clinic, Jackosonville, FL and 4Cancer Genome Atlas.
Body: Background. We recently identified an in-frame ESR1 translocation in a tumor and PDX pair from a patient with endocrine
therapy resistant advanced disease. The fusion gene preserved the N-terminal 365aa of ESR1 containing the intact activation
function 1 (AF1), DNA-binding domain (DBD) and hinge, followed by the C-terminal transactivation domain of YAP1. The
ESR1-365>YAP1 fusion drove estrogen-independent and anti-estrogen-resistant tumor cell growth (Cell Reports 4:1116, 2013).
We therefore sought to further investigate the role of ESR1 fusion genes in breast cancer.
Methods. 711 cases from TCGA, 82 ER+ cases from neoadjuvant endocrine therapy trials and 25 ER+ systemic relapse samples
were screened for fusion genes using RNAseq. Fusion genes involving ESR1 were subsequently assayed for estrogen response
element (ERE)-mediated transcriptional activity and ability to drive estradiol independent breast cancer cell growth.
Results. Multiple ESR1 fusions were identified in mostly luminal B cancers. ESR1 fusions resulted most frequently from
rearrangements involving another gene on Chromosome (Chr) 6 but also fusions with genes on other chromosomes. The fusions
involving partners on Chr6 that retained an intact N terminal AF1 and DNA binding domain (DBD) of ESR1 (1-253aa) included
an Out-of-Frame (OF) fusion event with CCDC170 (a gene immediately centromeric to ESR1), POLH (in-frame or IF), AKAP12
(IF), PCMT1 (OF), SYNE1 (OF) and GPR126 (OF). Also identified were inter-chromosomal translocations involving
Chr12p-NOP2 (IF) and ChrX-PCDH11X (IF) and Chr7q-AKR1D1 (OF). The ESR1 fusions with transcriptional potential (because
of a retained DNA binding domain) included variable 5 ESR1 exons that preserved 253aa, 365aa, 412aa or 458aa of ESR1
sequence: in all cases disrupting ligand binding though loss of C terminal sequence. The effect of the IF and OF 3 sequences on
the transcriptional activity of the relevant ESR1 fragment was highly variable. The most transcriptionally active fusion (more active
than the relevant ESR1 fragment alone) was the ESR1-365>YAP1 as well as ESR1-365>PCDH11X and ESR1-253>CCDC170.
This is remarkable since PCDH11X is not considered a transcription factor and the CCDC170 sequence was OF in both cases
suggesting the activity of the fusion was due to "neomorphic" properties. Two fusions, NOP2 and POLH were inactive in the ERE
reporter assay, yet both stimulated estradiol independent growth. In these cases we suggest that the ESR1 locus can function as
a "promoter trap" which allows the identification of genes with previously unknown functions in endocrine therapy resistance.
Conclusions: In luminal-type breast cancer the ESR1 gene was fused to multiple 3 partners with remarkably heterogeneous
functions. The two most transcriptionally active, ESR1-365>YAP1 and ESR1-365>PCDH11X, were both identified in endocrine
therapy refractory advanced disease suggesting a role in fatal disease progression.

Title: Profiling of ESR1-mutated metastatic breast cancers by FoundationOne allows a broad genomic understanding for
potential clinical implications
Norma A Palma1, Siraj Ali1, Garrett Frampton1, Kai Wang1, Hannah Gilmore2, Julio Peguero6, Lyndsay N Harris2, Massimo
Cristofanilli5, Juliann Chmielecki1, Jeffrey S Ross1,4, Deborah Morosini1, Vincent A Miller1, Phil J Stephens1, Gary Palmer1 and
Joyce O'Shaughnessy3. 1Foundation Medicine, Cambridge, MA; 2Case Western Reserve University, Cleveland, OH; 3Baylor
Charles A. Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, TX; 4Albany Medical College, Albany, NY; 5Thomas
Jefferson University, Kimmel Cancer Center, Philadelphia, PA and 6Oncology Consultants Research, Houston, TX.
Body: Background: Estrogen receptor (ER) inhibition is an important treatment option for advanced breast cancer (BC) pts.
Recent studies describe recurring somatic mutations in codons 537 and 538 within the ligand-binding domain (LBD) of 20% of
ER+ metastatic disease, but not treatment-nave ER+ cancer or ER-negative disease, that render the ER constitutively active and
confer partial resistance to endocrine agents. Few studies have described the clinicopathologic and genomic covariates that
accompany the ESR1 genomic alterations (GAs). Correlating genomic events as captured by FoundationOne in pts with
ESR1-mutated BCs with clinical history may provide clinical insight into these alterations.
rearranged in cancer were fully sequenced to high, uniform coverage using FoundationOne (Foundation Medicine, Cambridge,
MA).
Results: 176 of the 2,208 (7.9%) BC pt cases harbored ESR1 alterations. 1127 short variants (SV) were detected in 176 patient
samples for an average of 6.4 GA/sample. 16.5% SV (186/1127) were ESR1 GAs with an average of 1.0 ESR1 GAs/sample.
ESR1 GAs consisted of base substitutions (77%) and amplifications (20%, median copy number 9X, range 6-28). Base
substitutions occurred at codons 538 (71/145) and 537 (70/145) and at two other novel sites, 341 (2/145) and 563 (1/145). A
patient-derived xenograft study suggested that tumors refractory to ER inhibitors and harboring ESR1 amplification could be
responsive to higher doses of estradiol. (Li et al, Cell Reports, 2013). The most frequently co-occurring GAs were PIK3CA
(37.5%), GATA3 (22.7%), TP53 (24.4%), MAP3K1 (10.2%) and CDH1 (7.9). Collection of data on tissue sites sequenced,
treatments and outcomes on these pts with ESR1 GAs are ongoing. Of note, one pt had recurrent disease following adjuvant
tamoxifen and letrozole. FoundationOne testing of her recurrent bony disease revealed an ESR1 Y537 alteration along with
mutations in PIK3CA, APC, and RAD51. She has been on fulvestrant monotherapy with stable metastatic disease 13 + months.
Conclusions: We found through FoundationOne testing that ESR1 GAs are reasonably common in advanced BCs, with base
substitutions accounting for 77% of ESR1 GAs and amplifications for 20%. Identification and characterization of ESR1-mutated
advanced BC pts by comprehensive genomic profiling capable of detecting both base substitutions and copy number changes
may identify clinically relevant GAs. Clinical correlation is pending.

Title: Estrogen receptor (ESR1) mutations confer resistance to hormone therapy using a common mechanism
Luca Gelsomino1, Guowei Gu1, Yassine Rechoum1, Sebastiano Ando'2 and Suzanne AW Fuqua1. 1Baylor College of Medicine,
Houston, TX and 2Universita' della Calabria, Arcavacata di Rende, Cosenza, Italy.
Body: Background: The idea that somatic estrogen receptor gene (ESR1) mutations could play an important role in the evolution
of hormone-responsive breast cancers was proposed by us with our original identification of two ESR1 mutations at residues
K303 and Y537. Technical issues with ESR1 mutation detection and the resulting paucity of reports in tumors led many to
assume that ESR1 mutations were not there. However, with recent Next Generation Sequencing of metastatic tumors, mutation
of ESR1 is now an accepted certainty. ESR1 mutant allele frequencies vary over a wide dynamic range, and are usually a
minority population within tumors. Therefore how does a minor subclonal tumor population drive resistance in metastatic tumors?
Methods: MCF-7 cells expressing endogeous wild-type ER was transduced with ESR1 mutants K303R, Y537N, Y537S, and
D538G lentivirus and stable clones selected. ER transcriptional assays and growth in soft agar were performed. Digital drop (dd)
PCR and primer extension snp detection were used to ascertain mutant:WT ESR1 allele frequency in cell lines, 200 primary
tumors from patients treated with tamoxifen monotherapy, and 20 metastatic breast tumors.
Results: Mutant ER constitutive transcriptional activity was fully antagonized by the antiestrogens tamoxifen or fulvestrant in
MCF-7 stable transfectants. In contrast, soft agar growth of all ESR1 mutant-expressing cells was unexpectedly and completely
resistant to the growth inhibitory effects of tamoxifen, although mutant-expressing cells were a minority subpopulation in the
stable clones. Therefore, in cells with WT ER co-expression, the mutant resistant phenotype dominates. We found that
phosphorylation of IGF1R; was constitutively increased in all ESR1-mutant expressing cells. Treatment with a specific IGF1R
inhibitor in combination with tamoxifen drastically restored hormone sensitivity in cells expressing the ESR1 mutations. These
results suggest a convergence in resistance mechanisms between the K303R and Y537 ESR1 mutation hot spots. We are
exploring whether the dominant mutant ESR1 resistant phenotype occurs via activation of paracrine mediators, and have
identified altered IGF-1 and interleukin 6 signaling in mutant-expressing cells. Mutation detection in a retrospective cohort of
primary and metastatic breast tumors is ongoing and will be presented.
Conclusions: We hypothesize that the selection of dominant-acting ESR1 mutations in tumors is a key event in breast cancer
progression, potentially due to the selective pressure of antiestrogens. The dominant-resistant phenotype of ESR1 mutants in a
majority WT background supports the subclonal evolution of ESR1 mutations in breast cancer recurrence. A common resistance
mechanism (like consitutive IGF1R activation) should enable biologic targeting of ESR1 mutation-positive metastatic patients a
feasible clinical goal.
Support: NIH/NCI R01 CA72038 and CPRIT RP1210732 to SAWF.

Title: In-silico discovery of novel estrogen receptor- inhibitors as potential therapeutics for tamoxifen resistant breast cancer
Kriti Singh1, Ravi SN Munuganti1, Miriam Butler1, Artem Cherkasov1 and Paul S Rennie1. 1Vancouver Prostate Centre, University
of British Columbia, Vancouver, BC, Canada.
Body: Estrogen receptor- (ER) positive breast cancer (BCa) represents 75% of all invasive BCas. Conventional ER-directed
drug Tamoxifen targets the estrogen binding pocket (EBP) of the receptor. However, over prolonged periods of treatment, the
therapeutic efficacy of Tamoxifen declines due to development of resistance. Numerous factors are causative for this
phenomenon, including recently reported mutations in the receptor (T537S). Therefore, there is an urgency to develop novel
anti-ER therapeutics that exhibit entirely different mode of ER inhibition. A promising alternative strategy is to prevent
receptor-coactivator interaction and block further crucial steps in ER activity. ER-coactivator interface should be less prone to
adaptive mutations as any mutations at this site would also likely block the coactivator recruitment, we therefore targeted the
Activation Function-2 (AF2) site, a coactivator binding pocket on ER, called to overcome the limitations of Tamoxifen.
Although AF2 is a shallow surface pocket, the pharmacophore-rich features of this site make it a druggable target. To identify
potential ER AF2 inhibitors, virtual screening was performed. Initial hits were subjected to lead optimization and more potent
analogues were rationally designed by exploiting critical features of this site. Potential compounds were tested for their ability to
inhibit ER transcriptional activity using the T47D-KBluc cell line stably transfected with an ER-specific luciferase reporter.
Consequently, the lead compound VPC-16339 (IC50=8.24M) was identified. The direct binding between VPC-16339 and the
receptor was confirmed by Biolayer Interferometry assay. More importantly, VPC-16339 prevents coactivator recruitment at the
AF2 pocket in a dose dependent manner as measured by TR-FRET coactivator recruitment assay. Increasing concentrations of
estradiol did not affect the IC50 of the lead compound, thereby ruling out the possibility of VPC-16339 binding to EBP.
VPC-16339 demonstrated a strong anti-proliferative effect on MCF7 and Tamoxifen resistant cells, with no effect on ER- HeLa
cells, suggesting its selective ER-mediated action, as further validated by ER luciferase assay in Tamoxifen resistant cells.
VPC-16339 effectively inhibits mRNA and protein expression levels of the estrogen dependent genes such as pS2,CathD and
CDC2. Due to AF2-guided mechanism, VPC-16339 successfully overcomes Tamoxifen resistance and inhibits the constitutively
active Tamoxifen resistant form of ER (T537S).
In summary, we report VPC-16339 as an ER AF2 specific inhibitor with promising anti-proliferative effect in BCa cell lines
including Tamoxifen resistant cell lines. VPC-16339 effectively inhibits the mutant form of the receptor (T537S) which is
responsible for acquired endocrine resistance. It can be anticipated that ER AF2 inhibitors will provide an alternative therapeutic
strategy that can be applied concurrently or simultaneously with current anti-ER treatments for BCa patients with advanced
disease.

Publication Number: OT1-1-01
Title: LORELEI: A Phase II randomized, double-blind study of neoadjuvant letrozole plus taselisib (GDC-0032) versus letrozole
plus placebo in postmenopausal women with ER-positive/HER2-negative, early stage breast cancer
Cristina Saura1, Evandro de Azambuja2, Peter Dubsky3, Mafalda Oliveira1, Kamal S Saini4, Christian Fes5, Ray S Lin6, Timothy R
Wilson6, Jill Fredickson6, Hema Parmar6, Jerry Y Hsu6, Martine Piccard2, Michael Gnant3 and Jose Baselga7. 1Vall d'Hebron
University Hospital, Barcelona, Spain; 2Jules Bordet Institute, Brussel, Belgium; 3Medical University of Vienna, Vienna, Austria;
4
Breast International Group, Brussel, Belgium; 5Austrian Breast and Colorectal Cancer Study Group, Vienna, Austria; 6Genentech,
San Francisco, CA and 7Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background: Taselisib (GDC-0032) is an orally bioavailable, potent, and selective inhibitor of Class I PI3-kinase (PI3K)
alpha, gamma, and delta isoforms, with 30-fold less inhibition of the PI3K beta isoform relative to the PI3K alpha isoform.
Preclinical data show that taselisib has enhanced activity against PIK3CA mutant cancer cell lines. Clinical data have also
demonstrated confirmed partial responses in patients with PIK3CA mutant breast cancer treated with single-agent taselisib.
Preclinical and clinical data also show enhanced antitumor activity when taselisib is combined with either letrozole or fulvestrant.
Study design: LORELEI is a phase II, two-arm, randomized, double-blind, multicenter, neoadjuvant study of letrozole and taselisib
versus letrozole and placebo in postmenopausal women with newly diagnosed ER+/HER2-, untreated, stage I-III operable breast
cancer. Other relevant eligibility criteria include tumor size 2 cm, unilateral disease, ECOG PS 1, and available and evaluable
tumor tissue for central review of PIK3CA mutation analysis. Patients will be randomized (1:1) to receive continuous daily
letrozole (2.5 mg) with either placebo or taselisib (4mg on a 5 days on/ 2 days off schedule) for 16 weeks. Study treatment is
followed by surgery. Adjuvant treatment will be given as per physicians discretion. Stratification at randomization is based on
tumor size and nodal status.
Endpoints: The co-primary endpoints are overall objective response rate (ORR) by centrally assessed breast magnetic resonance
imaging (MRI) via modified RECIST criteria and pathologic complete response (pCR) rate in breast and axilla at time of surgery in
all enrolled patients and PIK3CA mutant (MT) patients. Secondary endpoints include ORR by centrally assessed MRI and pCR
rate in PIK3CA wild-type (WT) patients. Other secondary endpoints performed in all enrolled patients and separately as per
PIK3CA mutation status include: assessment of ORR using breast ultrasound, clinical breast exam (i.e. palpation) and
mammography; changes in Ki67 levels from baseline to week 3, baseline to surgery and week 3 to surgery; centrally assessed
preoperative endocrine prognostic index (PEPI) score; changes in enhancing tumor volume from baseline to surgery as
measured by breast MRI via central assessment. Exploratory analyses include expression of biomarkers predictive of response to
letrozole plus taselisib from tumor tissue or blood.
Statistical methods:
The sample size was calculated to detect an absolute percentage increase of 24% in ORR via MRI (40% in the letrozole-placebo
arm vs. 64% in the letrozole-taselisib arm in the PIK3CA MT cohort) with 80% power at 16% two-sided significance level. The
sample size will also detect an absolute percentage increase of 18% in pCR rate (1% in the letrozole-placebo arm vs 19% in the
letrozole-taselisib arm in the PIK3CA MT cohort) with 80% power at 4% two-sided significance level.
Target accrual: Approximately 330 pts at 110 global sites across Europe, North and South America, and Asia-Pacific.
Reference Study ID Numbers: GO28888/BIG-3-13/SOLTI 1205/ABCSG 38.

Title: PI3K-Akt-mTOR pathway analysis to obtain further insight in the efficacy of everolimus in combination with exemestane in
metastatic, ER-positive breast cancer: A Dutch breast cancer research group (BOOG) study
Dinja T Kruger1, Karin Beelen2, Connie R Jimenez1, Maurice PHM Jansen3, Stefan Sleijfer3, Sabine C Linn2 and Epie Boven1. 1VU
University Medical Center, Amsterdam, Noord Holland, Netherlands; 2Netherlands Cancer Institute, Amsterdam, Noord-Holland,
Netherlands and 3Erasmus University Medical Center, Rotterdam, Zuid Holland, Netherlands.
Body: Background
In patients with hormone receptor-positive breast cancer, activation of the PI3K-Akt-mTOR pathway is associated with resistance
against endocrine therapy. Previous research has shown that genetic aberrations in this pathway occur frequently, including
mutation and/or amplification in PI3K subunits or PI3K effectors as well as loss of lipid phosphatases (Fu X, et al. The Breast;
2013). Central review of the BOLERO-2 randomized phase III trial in which patients refractory to a non-steroidal aromatase
inhibitor were randomized between exemestane combined with the mTOR inhibitor everolimus versus exemestane and placebo
has shown a progression-free survival (PFS) of, respectively, 11.0 and 4.1 months [hazard ratio = 0.38 (95% confidence interval
0.31-0.48; log-rank P <0.0001)] (Yardley DA, et al. Adv Ther; 2013). The combination, however, is known to cause more adverse
events and is associated with additional costs as compared to exemestane alone. In the present study we will explore whether
there are biomarkers that might indicate which patients most likely benefit from co-targeting PI3K and ER pathways.
Trial design/Aims
This is a Dutch prospective, open-label, single-arm, investigator-initiated, multicenter trial in which approximately 30 hospitals will
participate. A total of 175 patients will be included for baseline blood sampling and archival tumor tissue collection. From 50
patients, a fresh tumor biopsy is required at baseline and from 30 out of 50, another tumor biopsy will be collected upon
progressive disease. Exploratory biomarker assessment includes immunohistochemistry (total and phosphorylated PI3K, AKT,
mTOR, p70S6K and 4EBP1), tissue phosphoproteomics and circulating tumor DNA (mutations). The results of the biomarker
analysis will be compared with clinicopathological characteristics and PFS.
Eligibility
Postmenopausal patients with hormone receptor-positive, HER2-negative, locally advanced or metastatic breast cancer,
refractory to anastrozole or letrozole will be included. No previous treatment with exemestane or mTOR inhibitor for advanced
disease is allowed and Informed consent must be signed before enrollment.
Statistical methods
Since the majority of the tests involve the use of new techniques, the study will be mainly explorative in design. The association
between potential biomarkers and clinicopathological characteristics will be tested using Fisher exact test or the Mann-Whitney
test. PFS curves will be drawn using the Kaplan-Meier method. PFS in association with potential biomarkers will be tested using
Cox proportional hazard regression analysis.
Present and target accrual
Recently, the study has been opened for inclusion. A period of 2 years is planned for patient enrollment. Up to May 2014, two
patients were included.
ClinicalTrials.gov identifier
NCT02109913
Financial support is received from Novartis, the Netherlands.

Title: Phase Ib dose allocation study of oral administration of lucitanib given in combination with fulvestrant in patients with
estrogen receptor-positive and FGFR1-amplified or non-amplified metastatic breast cancer (INES)
Mario Campone1, Thomas Bachelot2, Fabrice Andr3, Chadi Saba4, Valrie Agrapart4, Marie-Jeanne Pierrat4, Frdric Dubois4,
Thibault Chesnel4 and Camille Poirot4. 1Institut de Cancrologie de l'Ouest Centre Ren Gauducheau, Saint-Herblain, France;
2
Centre Lon Brard Centre de Lutte Contre le Cancer (CLCC) de Lyon, Lyon, France; 3Institut Gustave Roussy-Breast Cancer
Unit 39 rue C Desmoulins, Villejuif, France and 4Institut de Recherche International Servier 53, rue Carnot, Suresnes, France.
Body: Background: Lucitanib is a potent, oral inhibitor of the tyrosine kinase activity of fibroblast growth factor receptors 1-3
(FGFR1-3), vascular endothelial growth factor receptors 1-3 (VEGFR1-3) and platelet-derived growth factor receptors /
(PDGFR/). FGF aberrancy, as defined as FGFR1-or 11q- amplification, is a hallmark genomic alteration in breast cancer,
observed at a frequency of up to 25% of patients and is strongly associated with luminal B type. Breast cancer patients with
measurable disease and FGF aberrancy treated in the ongoing Phase 1/2 clinical trial of lucitanib monotherapy experienced an
overall response rate of 50% (6 out of 12 patients). FGFR1-knock down was shown to decrease cell proliferation and reverse
resistance to endocrine therapy in a FGFR1-amplified breast cancer cell line, hence supporting the idea of combining lucitanib
with an endocrine agent such as fulvestrant, at the time of resistance. This has led to this Phase Ib study of lucitanib in
combination with fulvestrant in metastatic breast cancer.
Trial design: INES is a multicenter, open-label, 2 -part study to assess the tolerability of lucitanib in terms of Maximum Tolerated
Dose (MTD) and Dose-Limiting Toxicities (DLTs) when administered with fulvestrant. A Continual Reassessment Method (CRM)
will be used for the 1st part. A minimum of 3 patients will be enrolled at the initial dose level of 10 mg daily in combination with
fulvestrant. Additional doses of 12.5 mg and 15 mg of lucitanib will be tested with the option of deescalating to 7.5 mg in case of
DLTs. A minimum of 9 patients will be included at the MTD. In the 2nd part, 2 cohorts will be opened: fourteen FGF+ patients
(FGFR1-or 11q- amplification), and fourteen non-amplified patients. All patients will receive fulvestrant 500 mg monthly and
lucitanib at the recommended dose (RD) until unacceptable toxicity according to the investigator, disease progression or patient
withdrawal. The main objective is to identify the Phase II RD when lucitanib is combined with fulvestrant. Secondary objectives
include determination the Pharmacokinetic (PK) profile of lucitanib and metabolites; Measurement of tumour response;
Description of the pharmacodynamic (PD) profile of lucitanib and investigation of any potential exposure dose-response
relationships for safety, efficacy and PD.
Eligibility Criteria: Patients with estrogen receptor-positive, HER2 negative, breast cancer after progression or recurrence on
prior therapy including fulvestrant. Patients should have ECOG performance status 0 or 1. Patients with uncontrolled
hypertension are not eligible. For part 2, the presence of a metastatic site for biopsy to assess the presence of FGFR1- and/or
11q- amplification, which will be analysed centrally using FISH, is required.
Conclusion: INES is a phase Ib trial testing lucitanib in combination with fulvestrant in order to select the RD for phase II and
seek preliminary efficacy signal in FGFR1- or 11q- amplified or non- amplified patients. As of June 2014, 3 patients have been
enrolled in the 10 mg dose escalation cohort.

Title: OlympiA, Neo-Olympia and OlympiAD: Randomized phase III trials of olaparib in patients (pts) with breast cancer (BC) and
a germline BRCA1/2 mutation (gBRCAm)
Mark Robson1, Andrew Tutt2, Judith Balmaa3, 12, Bella Kaufman4, Judy Garber5, Charles Geyer6, James Ford7, Priyanka Sharma8,
Mary Stuart9, Helen Mann9 and Peter A Fasching10,11. 1Memorial Sloan Kettering Cancer Center, New York, NY; 2King's College
London School of Medicine and Institute of Cancer Reseach, London, United Kingdom; 3University Hospital Vall d'Hebron,
Barcelona, Spain; 4Sheba Medical Center, Tel Hashomer, Israel; 5Dana-Farber Cancer Institute, Boston, MA; 6Virginia
Commonwealth University, Massey Cancer Center, Richmond, VA; 7Stanford University School of Medicine, Stanford, CA;
8
University of Kansas Medical Center, Westwood, KS; 9AstraZeneca, Macclesfield, Cheshire, United Kingdom; 10University
Hospital, Erlangen, Germany; 11German Breast Group, Neu Isenburg, Hessen, Germany and 12SOLTI Breast Cancer Research
Group.
Body: Background A Phase II study showed that the PARP inhibitor olaparib (400 mg bid; capsules) exerts antitumor activity in
BC pts with a gBRCAm (Tutt et al Lancet 2010). Three Phase III trials of olaparib monotherapy have been initiated in BC pts with
a gBRCAm: OlympiA (NCT02032823), Neo-Olympia (D081EC00005), OlympiAD (NCT02000622).
Trial design
OlympiA
Neo-Olympia
OlympiAD
Setting
Adjuvant therapy for

high-risk, primary
TNBC
Neo-adjuvant therapy for primary TNBC
Metastatic BC (mBC)
Design
Randomized (1:1),
double-blind,
parallel-group
Randomized (1:1:1), three-arm,

parallel-group
Randomized (2:1), open-label
Olaparib
monotherapy
arm
300 mg bid (tablet)
300 mg bid (tablet) (Arm A)*
300 mg bid (tablet)
Comparator
arm(s)
Primary endpoint
Placebo
Placebo + weekly paclitaxel 80 mg/m2 for 12 Physician's choice of capecitabine 2500

wks (Arm B)* Olaparib 100 mg bid (tablet) +
mg/m2 (d1-14 q21d), vinorelbine 30
weekly paclitaxel 80 mg/m2 for 12 wks (Arm mg/m2 (d1, d8 q21d) or eribulin 1.4 mg/m2
C)*
(d1, d8 q21d)
IDFS
pCR rate
PFS (BICR)
OS, DDFS, incidence

of new cancers
OS, EFS, DDFS, ORR at 12 wks
OS, PFS2, ORR
HRQoL
HRQoL
HRQoL
Other objectives
Safety, tolerability
Target
recruitment (pts)
1320
300
310
Secondary
endpoints
*Curative-intent surgery to be performed after 12 wks; pts will then receive olaparib 300 mg bid (Arm A), placebo (Arm B), or
either weekly paclitaxel 80 mg/m2 for 12 wks (then olaparib 300 mg bid) or olaparib 300 mg bid (Arm C). BICR, blinded
independent central review; d, days; DDFS, distant disease-free survival; EFS, event-free survival; IDFS, invasive disease-free
survival; ORR, objective response rate; pCR, pathological complete response; q, every; PFS2, time to second disease
progression or death; TNBC, triple-negative BC
For each trial, eligible pts will have a BRCAm and will undergo gBRCAm testing (Myriad Integrated BRACAnalysis) as part of the
trial. For OlympiA, pts must be at high risk of recurrence and have completed local treatment and either neoadjuvant (without
pCR) or adjuvant chemotherapy. Neo-Olympia pts can have operable, locally advanced or inflammatory BC, must have a tumor
>2cm by clinical exam (or >1cm by radiological exam) and have completed four cycles of anthracycline plus carboplatin without
progression. OlympiAD pts can have TNBC or HER2 BC, and must have received prior anthracycline and taxane in the adjuvant
or metastatic setting, and 2 chemotherapy lines for mBC. OlympiA pts will be treated for up to 12 months (m); efficacy will be
assessed q3m up to 24m, then q6m up to 60m, then q12m. Neo-Olympia pts will be treated for 12 wks (w) pre-surgery, then for
40w post-surgery. In OlympiAD, PFS will be assessed by RECIST v1.1; radiologic exams will be performed at baseline, q6w up to
6m, then q12w until progression. In OlympiA and OlympiAD, IDFS and PFS will be analyzed using stratified log-rank tests; for
Neo-Olympia, pCR rate will be analyzed with an adjusted logistic regression model. Primary analyses will be undertaken after 330
IDFS events (OlympiA), surgery (Neo-Olympia) and 230 PFS events (OlympiAD). Enrollment began in Mar 2014 for OlympiAD,
Apr 2014 for OlympiA and is expected to begin in Q3 2014 for Neo-Olympia.

Title: Phase III study of palbociclib in combination with exemestane vs. capecitabine, in hormonal receptor (HR) positive/HER2
negative metastatic breast cancer (MBC) patients with resistance to non-steroidal aromatase inhibitors (NSAI): PEARL study
(GEICAM/2013-02_CECOG/BC.1.3.006)
Miguel Martn3, Semir Beslija4, Eva Carrasco1, Zsuzanna Kahan5, M Jos Escudero1, Istvan Lang6, Begoa Bermejo7, Moshe
Inbar8, Jos Ignacio Chacn9, Dan Jinga10, Jos ngel Garca-Saenz11, Juan de la Haba12, Serafn Morales13, Miguel Gil14, Laura
Murillo17, Antonio Antn15, Manuel Ruiz-Borrego16, Christoph Zielinski18, Gnther Steger18 and Bella Nisenbaum19.
1
GEICAM-Spanish Breast Cancer Group, San Sebastin de los Reyes, Madrid, Spain; 2CECOG-Central European Cooperative
Oncology Group, Vienna, Austria; 3Instituto de Investigacin Sanitaria Gregorio Maran-Universidad Complutense de Madrid,
Madrid, Spain; 4Institute of Oncology, Sarajevo, Bosnia and Herzegowina; 5Onkotherpis Klinika, Szeged, Hungary; 6National
Institute of Oncology, Budapest, Hungary; 7Hospital Clnico Universitario de Valencia, Valencia, Spain; 8Tel Aviv Sourasky
Medical Center, Tel Aviv, Israel; 9Complejo Hospitalario Virgen de la Salud, Toledo, Spain; 10Emergency University Hospital
Bucharest, Bucharest, Romania; 11Hospital Universitario Clnico San Carlos, Madrid, Spain; 12Hospital Universitario Reina Sofa,
Crdoba, Spain; 13Hospital Universitario Arnau de Vilanova, Lleida, Spain; 14Institut Catal d'Oncologia (ICO), L'Hospitalet de
Llobregat, Spain; 15Hospital Universitario Miguel Servet, Zaragoza, Spain; 16Hospital Universitario Virgen del Roco, Sevilla,
Spain; 17Hospital Clnico Universitario Lozano Blesa, Zaragoza, Aragn, Spain; 18Medical University of Vienna, Vienna, Austria
and 19Meir Medical Center, Kfar Saba, Israel.
Body: Background: Endocrine therapy (ET) is the cornerstone treatment for HRpositive, HER2-negative breast cancer (BC)
patients. AIs have become the treatment of choice in postmenopausal patients. The high response rates with ET in these patients
are partially undermined by the resistance developed by most of them over time. On early disease recurrence/progression to AIs,
the treatment options include other AI, estrogen-receptor antagonists or chemotherapy (being capecitabine one of the best
options). Preclinical data suggest that ER+/HER2- BC are dependent on cyclin-dependent kinases 4/6 (CDK4/6) function; the
inhibition of this target may be effective in delaying/reverting endocrine resistance. Palbociclib is an oral novel CDK4/6 inhibitor
that seems to be synergistic with ET in preclinical and clinical studies.
Trial Design: This is an international (6 countries) randomized phase III study. Patients are randomized 1:1 to exemestane (25
mg daily) plus palbociclib (125 mg daily x3 weeks every 4 weeks) vs. capecitabine (1,250 mg/m2 twice daily x2 weeks every 3
weeks). Postmenopausal patients with HR+/HER2- MBC are eligible if resistant to previous NSAI (letrozole or anastrozole)
defined as: recurrence while on or within 12 months after the end of adjuvant treatment or progression while on or within 1 month
after the end of treatment for MBC. Previous chemotherapy is permitted either in the (neo)adjuvant setting and/or as first line for
MBC. Patients must have measurable disease according to RECIST 1.1 or lytic bone lesions in the absence of measurable
disease. The primary objective is Progression-Free Survival (PFS); secondary objectives are overall survival, response rate,
clinical benefit rate, response duration, safety, quality of life and biomarkers defined changes. The study will recruit 348 patients
to detect a difference of 2.75 months in the median PFS (from 6 to 8.75 months; hazard ratio= 0.686), with a power of 80% and a
5% two sided significance level. The study started recruitment in March 2014 and 14 patients have been included so far
(ClinTrials.gov reference NCT02028507).

Title: A phase I study of LDE225 in combination with docetaxel in patients with triple negative (TN) advanced breast cancer
(ABC): GEICAM/2012-12 (EDALINE study)
Miguel Martin1, Manuel Ruiz-Borrego2, Jos M Trigo3, Silvia Antolin4, Jose A Garcia-Saenz5, Andres Hernando6, Alberto Ocaa7,
Federico Rojo8, Sara Lopez-Tarruella9, Jesus Corral2, Nuria Ribelles3, Lourdes Calvo4, Fernando Moreno5, Rosalia Caballero6 and
Eva Carrasco6. 1Instituto de Investigacion Sanitaria Gregorio Maraon, Universidad Complutense, Madrid, Spain; 2Hospital
Universitario Virgen del Rocio, Sevilla, Spain; 3Hospital Clinico Universitario Virgen de la Victoria, Malaga, Spain; 4Complejo
Hospitalario Universitario A Corua, A Corua, Spain; 5Hospital Clinico San Carlos, Madrid, Spain; 6GEICAM, San Sebastin de
los Reyes, Madrid, Spain; 7Complejo Hospitalario Universitario de Albacete, Albacete, Spain; 8Fundacion Jimenez Diaz, Madrid,
Spain and 9Instituto de Investigacion Sanitaria Gregorio Maraon, Madrid, Spain.
Body: Background: LDE225 is a potent and selective oral inhibitor of Smo, a key component of the hedgehog (Hh) signaling
pathway. Up-regulation of the Hh pathway is implicated in the genesis of a wide range of tumors including triple negative breast
cancer. Here we report an ongoing phase I study exploring the combination of LDE225 with docetaxel in TN ABC patients to
identify the Maximum Tolerated Dose (MTD) and the Recommended Phase II Dose (RP2D) (ClinicalTrials.gov Identifier:
NCT02027376).
Trial Design: Eligibility criteria include patients with TN ABC candidates to receive treatment with docetaxel that have received a
maximum of 3 prior chemotherapy regimens. Those patients with CNS involvement are also candidates if treated and clinically
stable. Treatment consists of 21-day cycles with docetaxel 75mg/m2 on day 1, every 21 day and LDE225 once daily. We use a
standard 3+3 design in sequential cohorts (3 dose levels (DL) of LDE225: 400mg once daily (DL1), 600mg once daily (DL2),
800mg once daily (DL3); and a DL-1: LDE225 400mg once daily and docetaxel 60mg/m2 every three weeks). The primary
endpoint is the MTD and RP2D of the combination; secondary endpoints include evaluation of safety and tolerability, in addition to
pharmacodynamic (PD) and pharmacokinetic (PK) studies. Patients are treated until radiologic or symptomatic progression or
unacceptable toxicity occurs. PK will be performed to evaluate whether LDE225 influences the pharmacology of docetaxel. PD
assessments include Hg gene expression signature associated to pathway activation in tumor samples and changes in Smo
related biomarkers in skin and blood correlative samples. Efficacy will be measured in terms of time to progression and objective
response rate. A minimum of 9 and a maximum of 18 patients will be included in this phase I. The study is approved by ERBs and
Competent Authority and already recruiting patients (two patients included in DL1).

Title: A phase III study of abemaciclib (LY2835219) combined with fulvestrant in women with hormone receptor positive (HR+),
human epidermal growth factor receptor 2 negative (HER2-) breast cancer (MONARCH 2)
Antonio Llombart1, Masakazu Toi2, Suzanne R Klise3, Martin Frenzel3, Edward M Chan3 and George W Sledge4. 1Hospital Arnau
de Vilanova, Valencia, Spain; 2Kyoto University, Kyoto, Japan; 3Eli Lilly and Company, Indianapolis, IN and 4Stanford University,
Stanford, CA.
Body: Background: Abemaciclib (LY2835219), an oral drug administered twice daily on a continuous schedule, is a cell cycle
inhibitor of both CDK4 and CDK6. In Study I3Y-MC-JPBA, abemaciclib demonstrated evidence of single agent activity in a
tumor-specific cohort of patients with metastatic breast cancer (MBC) and a median of 7 prior therapies; all responses observed
were in women with HR+ disease. Abemaciclib also demonstrated an acceptable safety profile both as a single agent and in
combination with fulvestrant. Based on these results, abemaciclib has been entered into a Phase III study (MONARCH 2) in
combination with fulvestrant for women with locally advanced or metastatic hormone receptor positive (HR+), human epidermal
growth factor receptor 2 negative (HER2-) breast cancer.
Trial design: MONARCH 2 (NCT02107703) is a randomized, double-blind, placebo-controlled Phase III study of fulvestrant with
or without abemaciclib for women with HR+, HER2- locally advanced (not amenable to curative treatment by surgery) or
metastatic breast cancer. Patients will be randomized 2:1 (Arm A [abemaciclib plus fulvestrant]: Arm B [placebo plus fulvestrant]
with stratification based on the nature of disease (visceral metastases versus bone only metastases versus other) and sensitivity
to endocrine therapy (no prior endocrine therapy versus primary resistance versus secondary resistance). Abemaciclib (200 mg
every 12 hours [Q12H] on Days 1 to 28 of a 28-day cycle) or placebo will be given orally and fulvestrant 500 mg will be
administered intramuscularly on Days 1 and 15 of Cycle 1, then on Day 1 of Cycle 2 and subsequent cycles.
Eligibility criteria: Postmenopausal women with HR+, HER2- inoperable locally advanced or metastatic breast cancer who have
either relapsed after prior endocrine therapy or have not received prior endocrine therapy are eligible. Patients are required to
have either measurable disease or non-measurable bone only disease, adequate organ function and an Eastern Cooperative
Oncology Group (ECOG) performance status (PS) of 1.
Specific aims: The primary objective of MONARCH 2 is to compare PFS between two treatment arms: abemaciclib plus
fulvestrant versus placebo plus fulvestrant for women with HR+, HER2- locally advanced or metastatic breast cancer. Secondary
objectives are to compare overall survival, objective response rate, clinical benefit rate, safety, pharmacokinetics and quality of
life.
Statistical methods: The study has 90% power to detect an increase in PFS of approximately 42% (hazard ratio = .703).
Assuming a median PFS of 6.5 mos. in the control arm, this corresponds to a 2.75 month increase in the median PFS to 9.25
mos. PFS and OS will be hierarchically tested to maintain an overall type I error rate of 2.5%.
Present accrual and target accrual: Target accrual is approximately 550 patients.
Contact information: For further information please contact 1-877-CTLILLY (1-877-285-4559).

Title: Master regulator (MR)-directed therapy in residual breast cancer patient derived xenografts (PDXs)
Kevin Kalinsky1, Prabhjot Mundi1, Dawn L Hershman1, Eileen Connolly1, Katherine D Crew1, Hanina Hibshoosh1, Andrea Calfiano1
and Matthew Maurer1. 1Columbia University Medical Center, New York, NY.
Body: Background: Neoadjuvant chemotherapy (NACT) is standard of care in operable breast cancer (BC). There are no
therapies that have demonstrated benefit in patients who failed to achieve a pathologic complete response (pCR). There remains
an unmet need to identify appropriate targets in which to direct therapy in this high-risk population to decrease BC mortality.
Computational approaches to reverse engineering of cell signaling networks from patient tumor-specific RNA expression data can
identify key molecular dependencies that result in the malignant phenotype, i.e. MRs. These approaches out-perform traditional
methods, such as identifying mutated or over-expressed genes, in determining true tumor dependencies. Successful targeting of
MRs may be a means for rational selection of targeted drugs in the operable BC setting.
Trial Design: A single center pilot study that will accrue 35 subjects with BC who will have received standard NACT and at least
1.5 cm of residual disease. At the time of definitive surgery, resected tissue will be assessed for residual disease and then
allocated. After reserving required tissue for pathologic diagnosis, 0.5 cm3 will be procured for xenografting and 0.5 cm3 will be
flash frozen for RNA extraction. Tissue for xenografting is immediately sent to a Champions Oncology site for implantation. We
will use fresh frozen tissue to perform RNA-seq. Previously validated computational algorithms at our center, including MARINa
and VIPER, will be used to interrogate the expression data and identify the top MRs in individual tumors.
Upon maturation of a Champions Oncology Tumor Graft model, the second generation expansion group will be divided into four
cohorts. We plan to test two drugs that target tumor-specific MRs, one vehicle control, and one negative control (paclitaxel, as
these tumors have demonstrated paclitaxel resistance) in each model.
Eligibility Criteria
1. Patients 18 years with newly diagnosed BC, deemed candidates for definitive surgery.
2. Subjects must have received standard NACT with taxane (paclitaxel or docetaxel; herceptin +/- pertuzumab if HER2+) and/or
adriamycin/cytoxan.
3. Clinical or radiographic evidence of 1.5 cm of residual BC.
4. ECOG PS 2
Specific Aims:
1) To computationally infer MRs in individual patient tumors resistant to NACT
2) To determine if targeting MRs results in superior tumor growth inhibition in PDX models
Statistical Methods: In the PDX model, pre- and post-treatment tumor volume (TV) is calculated: TV= width2 x length x 0.52, and
is standardized as a tumor growth inhibition (TGI) percentage compared to the vehicle control. Since we anticipate the
MR-directed arms will demonstrate a TGI (80%) twice that of the paclitaxel arm (40%), we have 80% power to detect a difference
in 18 patients (one-sided alpha, 0.05). We assume a xenograft take-rate of 50% in the post-NACT setting and thus aim for an N of
35. Final data analysis will be adjusted using ANOVA.
Present and Target Accrual: 35 patients, projected to accrue in 18-24 months. Estimate based on number of locally advanced
BC patients seen, competing studies, and likelihood of participation. The study will open in Fall 2014.

Title: Phase I/II trial of ruxolitinib in combination with trastuzumab in metastatic HER2 positive breast cancer
Kevin Kalinsky1, Dow-Chung Chi1, Shing Lee1, Amy Tiersten2, Della Makower3, Ellen Chuang4, Katherine D Crew1, Dawn L
Hershman, Jose Silva2, Andrea Califano1 and Matthew Maurer1. 1Columbia Universtiy Medical Center, New York, NY; 2Icahn
School of Medicine at Mount Sinai; 3Albert Einstein College of Medicine, Bronx, NY and 4Weill Cornell Medical Center, New York,
NY.
Body: Background
Integrated analysis of whole genome RNAi screening with computationally reverse engineered interactome models identified
IL6/JAK/STAT as a master regulator pathway essential for growth of ErbB2/HER2 positive breast cancer. Ruxolitinib (R),
FDA-approved treatment for myelofibrosis, inhibits JAK1 and JAK2. The combination of R plus Trastuzumab (T) is synergistic in
tumor growth inhibition in mouse xenografts of HER2 amplified breast cancer cell lines. These data provide a strong rationale for
studying the efficacy of combination R and T in a clinical trial.
Trial Design
A multi-center, open-label, phase I/II (P1/2) trial of R plus T in HER2+ metastatic breast cancer (MBC) who have progressed on
T-based therapy. P1 will be an adaptive design with 10 patients, using the time-to-event continual reassessment method. The
recommended P2 dose (RP2D) will be used in a non-randomized, open-label P2 trial with 30 evaluable patients (pts). Given the
anticipated limited overlapping toxicities, approximately 36 pts (range: 32-40) are expected for the P1/2. The duration of a
treatment cycle will be 21 days. R will be taken orally twice a day continuously. The P1 dosing range will be 10-25 mg BID (dose
level 0: 20 mg BID). T will be administered on Day 1 of each cycle at standard dosing. Objective Response Rate (ORR) will be
assessed by imaging every 9 weeks. Blood samples will be obtained for biomarker analysis, pre-treatment, on-treatment on
C2D1, and then at progression. Pre-treatment biopsies from archival tissue or new biopsy, on treatment biopsy on C2D1, and
upon progression of disease will be discussed with pts with accessible disease.
Main Eligibility Criteria:
1. HER2 positive MBC
2. Progression on >2 HER2-directed therapy in metastatic setting, including Pertuzumab and T-DM1
3. Measurable or non-measurable disease
4. LVEF >50%.
5. No history of prior JAK2 inhibitor
6. No HIV-positive or active infection
7. No concurrent medications that are potent CYP3A4 inhibitor or inducer
Specific Aims
1. Primary: P1: MTD of combined R + T. P2: Progression Free Survival (PFS)
2. Secondary: a) Clinical: ORR, clinical benefit rate (CBR), and tolerability. Pts will be stratified by hormone receptor (HR) status
to explore differences in efficacy between HR+ and HR-.
b) Explore potential predictive tumor and blood-based predictive biomarkers at baseline, on treatment, and progression: (tumor:
pSTAT3 expression); serum: IL-6, IL-8, C-reactive protein; circulating tumor cell pSTAT3 expression; and tumor gene expression.
Statistical Methods
Assuming a historical PFS of 8 weeks with single-agent agent HER2-targeted therapy in HER2+ MBC after progressing on
T-based therapy, we predict that pts receiving the combination of R plus T will have a PFS of at least 13 weeks. With a 2-sided
alpha of 0.05, we have 80% power to detect a difference with 30 pts.
Target Accrual
Sample Size: 32-40 pts; projected over 2 years at 4 sites: Columbia, Einstein, Mount Sinai, and Cornell. Trial will start accruing
July 2014.


Title: The B-YOND study: A phase II three-arm randomized trial of the combination of tamoxifen plus goserelin acetate with
BYL719 or buparlisib (BKM120) in premenopausal patients with hormone receptor-positive/HER2-negative locally advanced or
metastatic breast cancer
Yen-Shen Lu1, Yeon Hee Park2, Zhimin Shao3 and Roberta Valenti4. 1National Taiwan University Hospital, Taipei, Taiwan;
2
Samsung Medical Center, Seoul, Korea; 3Fudan University Cancer Hospital, Shanghai, China and 4Novartis Pharma AG, Basel,
Switzerland.
Body: Background:
The incidence of breast cancer (BC) has been rapidly increasing in many countries in the last decades. In contrast to Western
countries, approximately 50% of BC patients in Eastern countries are premenopausal, with hormone receptor-positive (HR+)
disease, or luminal subtype by molecular classification. Benefit from the most recent advances in endocrine treatments for
patients with HR+ BC is currently confined to postmenopausal women. For premenopausal metastatic BC (MBC) tamoxifen
and/or ovary ablation/suppression remain the recommended first-line therapies for metastatic disease through category 2
evidence, calling for further clinical investigations.
In the context of growing evidence of the role of PI3K/AKT/mTOR pathway inhibition in enhancing and extending the benefit of
endocrine therapies in HR+ MBC pre-clinical and clinical models, the B-YOND study is a phase II, three-arm randomized trial
aimed at exploring the PI3K inhibitors buparlisib and BYL719 in combination with tamoxifen in the context of ovarian suppression
in premenopausal patients with MBC.
Design & Objectives:
Premenopausal women with HR+/HER2-negative locally advanced or MBC who a) are newly diagnosed or b) recurred during or
after adjuvant treatment with tamoxifen monotherapy, and received no endocrine treatment in the metastatic setting, will be
randomized to receive tamoxifen plus goserelin (control arm, Arm 3) or the same combination with buparlisib (Arm 2) or BYL719
(Arm 1) until progression. Stratification based on previous treatment with tamoxifen and presence of liver and/or lung metastasis
will be applied. Primary objective is the efficacy comparison of Arm 1 vs Arm 3 and Arm 2 vs Arm 3 in terms of 9-month
progression free survival (PFS) rate. Secondary objectives include additional efficacy evaluations (median PFS, overall response
rate, clinical benefit), safety and tolerability, quality of life, and pharmacokinetics.
Statistical Methods:
Based on previous reports, the estimated proportion of patients who are alive without progression at 9 months is 50% for the
control arm (P0=0.50), and we hypothesize that this proportion is 75% for both experimental arms (P1=0.75). Using a two-sided
0.05 level of significance for both the comparisons [Arm 1 vs Arm 3; Arm 2 vs Arm 3], with an overall type-I error not greater than
0.10, with 80% power, the estimated patient numbers needed is 58 patients for each arm. With an assumption that approximately
10% patients will be lost to follow-up, a total of 192 patients will need to be randomized to the three treatment arms in 1:1:1 ratio.
Accrual
This study will be open for recruitment in Taiwan, South Korea, China, Hong Kong and Thailand and potentially extended to
additional Eastern countries. Enrollment started in May 2014 and will last for an estimated period of 18 months.

Title: A phase 3, open-label, randomized, parallel, 2-arm multi-center study of the oral PARP inhibitor BMN 673 versus
physicians choice in germline BRCA mutation subjects with locally advanced and/or metastatic breast cancer (EMBRACA study)
Jennifer K Litton1, Joanne L Blum2, Wolfgang Eiermann3, Young-Hyuck Im4, Miguel Martin5, Lida Mina6, Henri Roch7, Hope S
Rugo8, Frances Visco9, Charlie Zhang10, Nathalie A Lokker10 and Debra L Lounsbury10. 1MD Anderson Cancer Center, Houston,
TX; 2Baylor Sammons Cancer Center, Texas Oncology, US Oncology, Dallas, TX; 3Interdisziplinares Onkologisches Zentrum
Munchen, Munich, Germany; 4Samsung Medical Center, Seoul, Korea; 5Hospital General Universitario Gregorio Maran,
Madrid, Spain; 6Indiana University School of Medicine, Indianapolis, IN; 7Institut Universitaire du Cancer Toulouse, Toulouse,
France; 8UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA; 9National Breast Cancer Coalition,
Washington, DC and 10BioMarin Pharmaceutical Inc, Novato, CA.
Body: Background: Poly-ADP-ribose polymerase (PARP) represents a family of enzymes of which at least two (PARP1 and
PARP2) play important roles in DNA repair. PARP inhibition induces synthetic lethality in tumor cells bearing mutations in the
genes encoding breast cancer susceptibility gene 1 (BRCA 1) and breast cancer susceptibility gene 2 (BRCA 2), both of which
are key components in the pathway of homologous recombination DNA repair. BMN 673 is the most potent preclinical PARP
inhibitor described to date with the highest efficiency at trapping PARP-DNA complexes (Murai et al, 2014). BMN 673 is a novel
and highly potent PARP inhibitor and has shown promising single-agent anti-tumor efficacy in several tumor types in an ongoing
Phase 1/2 clinical study.
Methods: The purpose of this multi-center, international, open-label, 2:1 randomized Phase 3 trial (EMBRACA) is to compare the
safety and efficacy of BMN 673 versus protocol-specific physicians choice treatment (capecitabine, eribulin, gemcitabine or
vinorelbine) in subjects who have locally advanced and/or metastatic breast cancer with germline BRCA mutations. The primary
objective of the study is to compare progression free survival (PFS) of subjects treated with BMN 673 as a monotherapy relative
to those treated with protocol-specific physicians choice treatment. Secondary objectives include objective response rate (ORR),
overall survival (OS), safety and pharmacokinetics of BMN 673. Exploratory objectives include duration of response (DOR) and
health-related quality of life assessment. Patients may be eligible if they are 18 years or older, have histologically or cytologically
confirmed carcinoma of the breast, locally advanced and/or metastatic disease appropriate for systemic single cytotoxic
chemotherapy, documentation of a deleterious or pathogenic germline BRCA1 or BRCA2 mutation, 2 prior
chemotherapy-inclusive regimens for locally advanced and/or metastatic disease, prior treatment with a taxane and/or
anthracycline in the adjuvant or metastatic setting, ECOG performance status 1, and no prior platinum treatment for metastatic
disease. Patients (n=429) will be randomized 2:1 to receive either BMN 673 oral capsules once daily (1.0 mg/day) in 21-day
cycles or protocol-specific physicians choice treatment. All eligible subjects will receive study drug treatment until disease
progression or unacceptable toxicity. This trial is enrolling patients from the United States, Europe, Israel, Asia/Pacific, and South
America (NCT01945775).

Title: A Phase 2, randomized, open-label, multicenter, safety and efficacy study of oral lucitanib in patients with metastatic breast
cancer with alterations in the FGF pathway
Maysa Abu-Khalaf1, Ingrid Mayer2, Jason B Litten3, Mitch Raponi3, Andrew R Allen3, Lajos Pusztai1 and Carlos L Arteaga2. 1Yale
Cancer Center, New Haven, CT; 2Vanderbilt-Ingram Cancer Center, Nashville, TN and 3Clovis Oncology, Inc, San Francisco, CA.
Body: BACKGROUND: Lucitanib is a potent, oral inhibitor of the tyrosine kinase activity of Fibroblast Growth Factor Receptors
1-3 (FGFR1-3), Vascular Endothelial Growth Factor Receptors 1-3 (VEGFR1-3) and Platelet-Derived Growth Factor Receptors
A/B (PDGFRA/B). Aberrant FGF signaling, as defined as FGFR1 and/or 11q amplification, is a hallmark genomic alteration in
breast cancer, observed at a combined frequency of up to 25% of patients. Breast cancer patients with measurable disease and
aberrant FGF signaling treated in the ongoing Phase 1/2 clinical trial of lucitanib monotherapy experienced an ORR of 50% and a
DCR of 100%. This compelling clinical activity has led to the initiation of a global clinical development program for lucitanib in
breast cancer.
TRIAL DESIGN: The current study is comparing PFS for doses of lucitanib (10 or 15 mg daily) in patients with FGF-aberrant
metastatic breast cancer after failure of currently available standard therapies. Approximately 160 patients will be randomized 1:1
to the 10 mg and 15 mg daily dosing groups and stratified by FGF pathway alteration (FGFR1 or 11q amplification) and prior
anti-VEGF therapy (yes or no). FGFR1 and 11q (containing the FGF ligands 3, 4, and 19) amplification is determined locally for
patient enrolment and confirmed by central laboratory fluorescent in situ hybridization (FISH) testing. A total of 130 PFS events
provides 80% power at a 2-sided significance level of 0.05 to detect a hazard ratio of 0.60 for comparing the 10 mg and 15 mg
dose groups. Secondary objectives are ORR, DoR, DCR, OS, PROs, safety and population PK. Exploratory endpoints include
tissue and blood-based biomarkers that may be predictive of response or primary resistance to treatment with lucitanib.
ELIGIBILITY CRITERIA: Histologically or cytologically confirmed FGF-aberrant metastatic breast cancer relapsed or refractory to
approved standard available treatment. ECOG 0 or 1. Normal organ function. Patients with uncontrolled hypertension are
excluded.

Title: A phase 2, 2-stage, 2-cohort study of the oral PARP inhibitor BMN 673 in patients with germline BRCA mutation and locally
advanced and/or metastatic breast cancer (ABRAZO study)
Nicholas C Turner1, Judith Balmana2, Susan M Domchek3, Frances Visco4, Charlie Zhang5, Nathalie A Lokker5, Debra L
Lounsbury5 and Mark E Robson6. 1Royal Marsden Hospital and Institute of Cancer Research, London, United Kingdom; 2Vall
Hebron Institute of Oncology, Barcelona, Spain; 3Basser Research Center for BRCA, University of Pennsylvania, Philadelphia,
PA; 4National Breast Cancer Coalition, Washington, DC; 5BioMarin Pharmaceutical Inc, Novato, CA and 6Memorial Sloan
Kettering Cancer Center, Weill Cornell Medical College, New York, NY.
Body: Background: Poly-ADP-ribose polymerase (PARP) represents a family of enzymes of which at least two (PARP1 and
PARP2) play important roles in DNA repair. PARP inhibition induces synthetic lethality in tumor cells bearing mutations in the
genes BRCA1 and BRCA2, both of which are key components in the homologous recombination DNA double-strand breaks
repair pathway. BMN 673 is the most potent preclinical PARP inhibitor described to date with the highest efficiency at trapping
PARP-DNA complexes (Murai et al, 2014). BMN 673 has shown promising single-agent anti-tumor efficacy in several tumor types
in an ongoing Phase 1/2 clinical study.
Methods: The purpose of this Phase 2 trial (ABRAZO) is to evaluate the safety and efficacy of BMN 673 in patients with locally
advanced or metastatic breast cancer with a deleterious germline BRCA 1 or BRCA 2 mutation. This study is an open-label,
2-stage, 2-cohort Phase 2 study using a Southwest Oncology Group (SWOG) 2-stage design. Eligible subjects will be assigned to
either Cohort 1 (n=70) or 2 (n=70) based on prior chemotherapy exposure for metastatic disease: Cohort 1) Subjects who have
previously responded to a platinum-containing regimen for metastatic disease with disease progression > 8 weeks following the
last dose of platinum; or Cohort 2) Subjects who have received > 2 prior chemotherapy regimens and who have had no prior
platinum therapy for metastatic disease. The primary objective is to determine the objective response rate (ORR) of BMN 673 as
a single agent for each cohort. The secondary objectives of the study are to determine the following for each cohort: clinical
benefit rate (CBR), duration of response (DOR), progression free survival (PFS), and overall survival (OS). Health-related quality
of life assessment is an exploratory objective. Patients may be eligible if they are 18 years or older, have histologically or
cytologically confirmed carcinoma of the breast, locally advanced and/or metastatic disease, deleterious or pathogenic germline
BRCA1 or BRCA2 mutation, prior chemotherapy for metastatic disease based on the above Cohort 1 or Cohort 2 inclusion
criteria, ECOG performance status 1, and no central nervous system (CNS) metastasis except adequately treated brain
metastasis documented by baseline CT or MRI scan that has not progressed since previous scans and that does not require
corticosteroids for management of CNS symptoms. Eligible patients will receive BMN 673 oral capsules once daily (1.0 mg/day)
continuously in 21-day cycles until disease progression or unacceptable toxicity. This trial is enrolling patients from the United
States and European countries (NCT02034916).

Title: A phase II study of dovitinib as salvage therapy in patients with stage IV inflammatory breast cancer HER2-negative with
local or distant relapse
Ricardo H Alvarez1, Mariana Chavez-MacGregor1, Joe Ensor1, James L Murray1, Kimberly Koenig1, Savitri Krishnamurty1, Antony
Lucci1, Gildy V Babiera1, Wendy Woodward1, Gary J Whitman1, Summer A Jackson1, Michael Shi2, Kenneth Culver2, James L
Reuben1, Naoto T Ueno1 and Vicente Valero1. 1University of Texas MD Anderson Cancer Center, Houston, TX and 2Novartis,
East Hanover, NJ.
Body: Background:
Inflammatory breast cancer (IBC) is an aggressive form of breast cancer that accounts for 3 to 5% of all invasive breast tumors in
the United States. IBC possesses an increase of proangiogenic factors including vascular endothelial growth factor (VEGF), basic
fibroblast growth factor (bFGF), and platelet derived growth factor (PDGF) as compared with non-IBC. In particular, FGF family
receptors play a critical role in tumorigenesis, morphogenesis, and inducers of angiogenesis. Dovitinib (TKI258) is an oral tyrosine
kinase inhibitor with in vitro IC50 values of approximately 10 nmol/L against FGFR1-3, VEGFR1-3, and PDGFR. These structurally
related receptors are important for the growth and survival of endothelial cells during tumor angiogenesis.
Trial Design:
This is a single institution, single arm phase II study. Patients receive a single daily oral dose of dovitinib 500 mg for 5
consecutive days, followed by a 2-day rest period (5 days on/2 days off schedule).
Eligibility:
Patients have histological confirmation of breast carcinoma with a clinical diagnosis of IBC based on presence of inflammatory
changes in the involved breast, including diffuse erythema and edema (peau dorange). Pathological evidence of dermal
lymphatic invasion should be noted but is not required for diagnosis. HER2-negative. ECOG PS 0-2. Baseline MUGA or
echocardiogram scans with LVEF of > 50%. Normal hematology, liver and kidney function laboratory studies. Patients must have
received at least 2 chemotherapy lines for metastatic disease and have relapsed.
Research Hypothesis:
Dovitinib has antitumor activity in patients with HER2-negative advanced IBC.
Specific Aims:
Primary objective: to determine the disease control rate (CR, PR, and SD). Secondary objective: to evaluate safety profile.
Exploratory biomarkers: circulating tumor cells (CTC), CTC undergoing EMT, and cancer stem cells
The primary endpoint is the six-month disease control rate (ORR) as defined by RECIST 1.1. A response is anyone who
experiences SD, CR or PR in the first 6 months. We will conduct this study with Simons two-stage design using the mini-max
criterion and the response rate will be estimated accordingly. It is assumed that dovitinib will have a target ORR of 30%. An ORR
of 10% or lower is considered a failure based on the typical ORR with a second line regimen for IBC and the new regimen will be
rejected under this circumstance. When the probability of accepting a "bad" regimen (i.e. response rate < 10%) is 0.05 and the
probability of rejecting a "good" regimen (i.e. response rate > 30%) is also 0.10, Simons design to minimize the maximum sample
size requires 22 patients in the first stage. If two or less patients respond to the treatment, the trial will be stopped and the
regimen will be declared as ineffective. If at least three of the first 22 patients respond to the treatment, 11 additional patients will
be entered in the study to reach a total of 33 patients. By the end of the study, the new regimen will be rejected if response rate is
less than or equal to 6 out of 33 patients and will be accepted otherwise.
Present Accrual and Target Accrual:
A total of 22 patients were accrued. Target accrual is 33 patients.

Title: A randomized, multicenter, phase II study of ipatasertib (Ipat, GDC-0068), an inhibitor of Akt, in combination with paclitaxel
(Pac) as front-line treatment for patients (pts) with metastatic triple-negative breast cancer (TNBC)
Steven J Isakoff1, Sung-Bae Kim2, Seock-Ah Im3, Rebecca A Dent4, Tiffany A Traina5, Vicente Valero6, Cristina Saura7, Sreeni
Yalamanchili8, Premal Patel8 and Mafalda Oliveira7. 1Massachusetts General Hospital Cancer Center, Boston, MA; 2Asan Medical
Center, University of Ulsan College of Medicine, Seoul, Korea; 3Seoul National University College of Medicine, Seoul, Korea;
4
National Cancer Center Singapore, Duke NUS, Singapore, Singapore; 5Memorial Sloan Kettering Cancer Center, New York, NY;
6
University of Texas MD Anderson Cancer Center, Houston, TX; 7Vall d'Hebron University Hospital, Vall d'Hebron Institute of
Oncology (VHIO), Barcelona, Spain and 8Genentech, San Francisco, CA.
Body: Background: The PI3K/Akt pathway is often activated in TNBC through loss of PTEN expression, low INPP4B expression,
and/or increased AKT3 amplification. Activation of Akt may then lead to chemoresistance; thus, inhibition of Akt signaling may
result in improved efficacy of chemotherapy in TNBC. Ipat (GDC-0068) is a potent ATP-competitive small molecule inhibitor of all
Akt isoforms. In preclinical breast cancer models, the combination of Ipat with taxanes enhanced efficacy, and in a Phase Ib
study, the combination of Ipat with Pac was well-tolerated and resulted in clinical responses.
Trial Design: This is a randomized, double-blinded, placebo controlled, international, multicenter, Phase II study designed to
estimate the efficacy and safety profile of Ipat combined with Pac versus placebo combined with Pac in pts with metastatic TNBC.
Pts will receive Ipat or placebo 400 mg orally once daily on Days 1 to 21 of each 28-day cycle with Pac 80 mg/m2 on Days 1, 8,
and 15 of each 28-day cycle. Treatment will continue until disease progression, intolerable toxicity, withdrawal, or study
completion. Pts will then be followed every 3 months for survival. Archival tumors will be assessed for PTEN expression by
immunohistochemistry.
Key Eligibility: Pts 18 years, ECOG 0 or 1, with histologically documented TNBC that is inoperable locally advanced or
metastatic and not amenable to curative resection are eligible. Additional eligibility criteria include availability of a tumor
specimen, measurable disease per RECIST v1.1, and adequate hematologic and organ function within 14 days of study. Any
previous therapy for TNBC is excluded, except for prior neoadjuvant or adjuvant chemotherapy and/or radiation completed 6
months prior to study. Pts with known brain or spinal cord metastases are also excluded.
Objectives: The primary objective is progression-free survival (PFS) in all TNBC pts and in TNBC pts with PTEN-low tumors.
Secondary objectives include estimation of overall survival (OS), objective response rate (ORR), duration of ORR, safety,
pharmacokinetics (PK), patient-reported outcomes (PROs), and biomarkers.
Statistical Methods: Approximately 120 pts will be randomized 1:1 and stratified by prior adjuvant/neoadjuvant treatment including
chemotherapy and/or radiation (yes vs. no), disease free interval from last dose of chemotherapy ( 12 months vs. > 12 months),
and tumor PTEN status (low/null vs. moderate vs. high). Primary and secondary efficacy analyses will include all randomized pts,
grouped by treatment at randomization. KaplanMeier curves will be produced for analyses of PFS, OS, and duration of
response, and stratified log rank tests will be used to compare treatments.
Accrural: This study is open for accrual.

Title: Circulating tumor DNA in plasma as a surrogate for tumor biopsy to identify tumor genetic alterations in patients with
multi-focal metastatic breast cancer
Mary D Chamberlin1, Todd W Miller1, Jennifer R Bean1, Richard J Barth1, Kari M Rosenkranz1, Jonathan D Marotti1, John M
Gemery1 and Jiang Gui1. 1Dartmouth-Hitchcock Medical Center, Lebanon, NH.
Body: Background: While effective targeted therapies exist for patients with ER/PR + (anti-estrogens) and HER2+ (anti-HER2
agents) disease, and some triple-negative cancers respond to DNA-damaging chemotherapy, many patients eventually exhibit
disease refractory to all standard breast cancer therapies, particularly in the metastatic setting. Expanding catalogs of
tumor-targeted therapies are being developed, and tumor genetics are playing an increasing role in patient selection. However,
tumors can exhibit intra- and inter-tumor genetic heterogeneity. If a biopsied tumor is not genetically reflective of all tumors within
a patient, then the optimal therapy may be overlooked. Identifying targetable genetic changes for which there are drugs, using
non-invasive procedures, will be an increasing challenge for medical oncologists. One way to potentially overcome this issue is
through cell-free circulating tumor DNA, which is detectable in the bloodstream. This study represents the first step in this
important process: evaluating plasma DNA as a potential route to non-invasive identification of genetic mutations in patients with
metastatic breast cancer with several tumors.
Design: Patients with new or progressive metastatic breast cancer with >=3 sites of biopsy-able disease are enrolled. Large bore,
large volume blood draw for PT/ INR and plasma DNA will be obtained. Primary tumor will be biopsied if present. Biopsies of >=3
tumors in >=2 different organ sites will be required. Tumor histology and ER/PR/her2 status will be determined. Tumor tissue and
plasma will undergo DNA sequencing.
Eligibility: Measureable new or progressive metastatic breast cancer by CT and bone scan or PET scan.
>=3 sites of disease with >=2 organ sites appropriate for biopsy. Prior therapy allowed, but all specimens must be obtained prior
to change in therapy.
Primary objective: To determine whether the genetic mutations in plasma DNA are reflective of the genetic mutations present in
biopsies of all tumors.
Secondary objectives: To determine A) the amount of plasma DNA required to capture all somatic genetic mutations in tumor
samples, and B) the extent of genetic heterogeneity between primary and metastatic tumors.
Primary endpoint: Rate of genetic concordance between plasma DNA and >=3 tumor within each patient.
Present accrual: 2 Target accrual: 10.

Title: Trastuzumab in HER2-negative early breast cancer as adjuvant treatment for circulating tumor cells (CTCs) (Treat CTC)
Michail Ignatiadis1, Carlo Messina2, Saskia Litiere2, Dimitris Mavroudis3, Christian Dittrich4, Anthony Kong5, Wolfgang Janni6,
Christos Sotiriou1, Martine Piccart1, Jean-Yves Pierga7 and Brigitte Rack8. 1Institut Jules Bordet, Universit Libre de Bruxelles,
Belgium; 2European Organization for Research and Treatment of Cancer (EORTC), Belgium; 3University Hospital of Heraklion,
Greece; 4LBI-ACR VIEnna LBC-TO, Kaiser Franz Josef-Spital, Austria; 5Oxford University Hospitals NHS Trust Churchill
Hospital, United Kingdom; 6Universittsklinikum Ulm, Germany; 7Institut Curie, France and 8Ludwig-Maximilians-University,
Munich, Germany.
Body: Background:
The presence of Circulating Tumor Cells (CTCs) in metastatic breast cancer (BC) is associated with worse clinical outcome.
Recent data showed an association between CTC(s) detection and reduced disease-free and overall survival in early disease.
Patients with persisting CTC(s) after (neo)adjuvant chemotherapy might benefit from additional systemic treatment.
Trastuzumab is a part of the standard of care for patients with HER2-positive BC. Recent data have reinforced the hypothesis that
the therapeutic effect of trastuzumab depends on immune-related mechanisms. It has been demonstrated that trastuzumab
eliminated CTC(s), irrespective of the HER2 status of the primary tumor and of CTC(s) and this was associated with improved
relapse-free survival (Bozionellou et al, Clin Cancer Res 2004, Georgoulias et al Ann Oncol 2012).
The Treat CTC trial is designed to explore the effect of trastuzumab in patients with HER2-negative early BC and persisting
CTC(s) after (neo)adjuvant chemotherapy and surgery.
Trial Design:
Treat CTC trial is a multicentre (6 countries, 92 centers) European randomized phase II trial, sponsored by the EORTC and run
under the BIG umbrella. It will assess the efficacy of trastuzumab in eliminating persisting CTC(s) after the completion of
(neo)adjuvant chemotherapy and surgery in patients with HER2-negative early BC. Eligible patients will be randomized in a 1:1
ratio to either 6 cycles of trastuzumab or observation.
Eligibility criteria:
- Adequately excised HER2-negative early BC
- Evidence of CTC(s) detection using the CellSearch technology after completion of (neo)adjuvant chemotherapy
- Completion of adjuvant chemotherapy for node-positive disease or neoadjuvant chemotherapy with residual invasive disease in
breast or lymph nodes (no complete pathological response)
Specific aims:
The primary objective is to evaluate whether trastuzumab decreases the detection rate of CTCs in patients with HER2-negative
primary BC by comparing the trastuzumab treated arm to the observation arm. Furthermore, clinical outcomes as measured by
Recurrence Free Interval (RFI), Invasive Disease Free Survival (IDFS), Disease Free Survival (DFS) and Overall Survival (OS)
between the trastuzumab and observation arms will be compared.
Present accrual and target accrual:
It is estimated that 2175 women will be registered to include 174 patients eligible for randomization in a 1:1 ratio. Accrual is
expected to be completed in 2 years. Treat CTC started patient screening in May 2013 in Belgium, in March 2014 in Germany
and in June 2014 in France. An update of the proportion of patients screened versus patients randomized will be presented
during SABCS.
Methods:
The primary test will be a one-sided test to compare the trastuzumab arm to the observation arm for the CTC(s) detection rate at
week 18 (superiority test). The comparison for the primary endpoint will be performed on the intention-to-treat population using a
one-sided test with overall of 0.1. The odds ratio and its confidence interval will be estimated using a logistic regression model.
The comparison of RFI, IDFS, DFS and OS will be done using a two-sided test in a proportional hazards model for cause specific
hazard, adjusted for the stratification factors.

Title: The DETECT-study concept: Treatment based on the phenotype of circulating tumor cells in HER2-negative metastatic
breast cancer
Bernadette AS Jaeger1, Susanne Albrecht1, Fabienne Schochter1, Carola A Melcher2, Carsten Hagenbeck2, Thomas WP Friedl1,
Brigitte Rack3, Volkmar Mller4, Peter A Fasching5, Wolfgang Janni1 and Tanja Fehm2. 1University Hospital Ulm, Ulm, Germany;
2
University Hospital Duesseldorf, Duesseldorf, Germany; 3University Hospital LMU-Munich, Munich, Germany; 4University
Hospital Hamburg-Eppendorf, Hamburg, Germany and 5University Hospital Erlangen, Friedrich-Alexander University
Erlangen-Nuremberg, Comprehensive Cancer Center Erlangen-EMN, Erlangen, Germany.
Body: Background: The prognostic impact of circulating tumor cells (CTC) in metastatic breast cancer (MBC) is well
demonstrated. The role of CTCs in predicting specific treatment response and the importance of CTC phenotypes for therapeutic
decisions will be investigated within the DETECT-study concept.
Trial Design and eligibility criteria: The DETECT studies are prospective, multicenter, open-label clinical trials designed for
patients with HER2-negative MBC and evidence of CTCs in the peripheral blood. DETECT III is a two-arm study for patients with
HER2-positve CTCs, randomized to physicians choice therapy (chemotherapy or endocrine treatment) with or without additional
HER2-targeted treatment with lapatinib. DETECT IV combines tow single-arm studies aimed at patients with HER2-negative
CTCs. Postmenopausal patients with hormone-receptor-positive MBC will be treated with the mTOR-inhibitor everolimus in
combination with an endocrine therapy of physicians choice (everolimus cohort), whereas patients with triple-negative or
hormone-receptor-positive MBC and indication to chemotherapy will receive eribulin (eribulin cohort).
Specific aims: The primary objective of the trials is to estimate the clinical efficacy of treatments, assessed by the CTC clearance
rate for DETECT III and by progression-free survival (PFS) for DETECT IV.
Methods: Prevalence of CTCs at various time points as well as the HER2 status of CTCs are assessed using the FDA-approved
CellSearch System (Veridex, USA). After immunomagnetic enrichment with an anti-EpCam-antibody, cells were labelled with
anti-CK8/18/19 and anti-CD45 antibodies to distinguish epithelial cells from leucocytes. A fluorescein conjugate antibody with
anti-CK-Fluorescein Isothiocyanate (FITC) was used for HER2 phenotyping. The cut-off for CTC-positivity was 1 CTC and for
HER2 1 CTC with strong HER2-staining (+++). Survival endpoints will be estimated using the Kaplan-Meier method.
Present and target accrual: Overall, about 2000 patients with HER2-negative MBC will have to be screened for CTCs to be able
to recruit 228 patients with HER2-positive CTCs for DETECT III (which started in February 2012), 400 patients with
HER2-negative CTCs for DETECT IV- everolimus cohort (which started in December 2013) and 120 patients for DETECT IVeribulin cohort (which will start in the second half of 2014). 907 patients have been recruited for CTC screening until June 2014.
Perspectives: One screening for CTCs offers different treatment options for patients with HER2-negative MBC and evidence of
CTCs within the DETECT-study concept. DETEC III is the first study to investigate a personalized targeted treatment based on
the phenotype of CTCs. The addition of a HER2-targeted therapy in case of HER2-positive CTCs is innovative and in case of
success will lead to new treatment strategies in MBC. DETECT IV complements DETECT III with regard to additional therapy
indications.

Title: TP53 mutants in circulating tumor DNA and follow-up of BRCA1 mutation carriers: The CirCA01 study
Francois-Clement Bidard1, Dominique Stoppa-Lyonnet1, Catherine Nogues2, Olivier Caron3, Suzette Delaloge3, Catherine
Dugast4, Christine Lasset5, Frederique Berger1, Bernard Asselain1, Olivier Lantz1, Marc-Henri Stern1 and Jean-Yves Pierga1.
1
Institut Curie, Paris, France; 2Institut Curie, Saint Cloud, France; 3Gustave Roussy, Villejuif, France; 4Centre Eugne Marquis,
Rennes, France and 5Centre Lon Brard, Lyon, France.
Body: Background:
Circulating tumor DNA (ctDNA) has been proposed as a biological surrogate of the repertoire of molecular aberrations in cancer
patients. The genetic information derived from the analysis of ctDNA may be therefore employed as biomarker for diagnosis,
prognostication, therapy response monitoring and assessment of genetic mechanisms of resistance. We have initiated a
prospective study in a population of BRCA1-mutation carrier at high risk of either relapse and/or new cancer growth. The CirCA01
study is based on the combination of (i) on the clinical side, the systematic inactivation of the TP53 gene by mutations in
BRCA1-related cancers, whatever the histological type and the organ of origin, and (ii) on the bench side, an original assay that
allows to detect any TP53 mutation in plasma with exquisite sensitivity.
Methods:
CirCA01 is a national prospective study, opened in several cancer centers in France (Institut Curie, Gustave Roussy, Centre Lon
Brard, Centre Eugne Marquis) and funded by the French Ministry of Health (PHRCK1369250N). Inclusion criteria are: patients
with no evidence of any invasive tumor mass at inclusion (clinical and, if any, radiological exams), carriers of known germline
BRCA1 deleterious mutation (a personal history of cancer being not mandatory), age 30 years for patient with previous history
of cancer or age 40 years for patient without previous history of cancer, written informed consent. Main exclusion criteria are:
patient presenting with invasive tumor masses (e.g. stage IV cancer or localized cancer not yet surgically removed), carriers of
germline BRCA1 variant of unknown significance, carriers of germline BRCA2 deleterious mutation or variant, individuals with a
low risk of BRCA1-related tumor growth, i.e. women who underwent prophylactic bilateral mastectomy, annexectomy or
controlateral prophylactic mastectomy after breast cancer. 200 BRCA1 mutation carriers will be enrolled and followed up
regularly, as per national guideline for high-risk patients. Fresh plasma samples will be collected at each follow-up visit to the
hospital and will be used to detect any mutation in TP53 exons and flanking non-translated regions by a "digital NGS"-based
technique. (NCT registration ongoing).
Results: Correlation with patients outcome will allow to report whether TP53 ctDNA detection may be used as (i) a new screening
test that can be repeated easily in BRCA1-carriers with no evidence of tumor growth (ii) a new diagnostic test in BRCA1-carriers
who present a clinical or radiological abnormality (typically ACR class 3 mammograms): (A) diagnosis of the malignant vs benign
nature of the abnormality, based on TP53 mutation levels, (B) differential diagnosis of a new cancer growth vs relapse of a
previously treated cancer based on TP53 mutation characterization (different vs similar TP53 mutations).

Title: Phase II study on radiofrequency ablation in stage 0 and I breast cancer without extensive intraductal components
Shigeru Imoto1, Shinji Nagamine2, Shunichi Ito3, Hitoshi Tsuda4, Masayuki Yoshida5, Mitsuhiro Tozaki6, Satoshi Morita7 and
Takayuki Ueno1. 1School of Medicine Kyorin University, Japan; 2Okinawa Red Cross Hospital, Japan; 3Rinku General Medical
Center, Osaka, Izumisano, Japan; 4National Defense Medical College, Japan; 5National Cancer Center Hospital, Japan; 6Kameda
Medical Center, Japan and 7Kyoto University Graduate School of Medicine, Kyoto, Japan.
Body: Background: Screening mammography makes it possible to identify small size of breast cancer (BC) and minimize surgical
management. Previously we reported about a multi-center cohort study on radiofrequency ablation (RFA) in early breast cancer
(ASCO2012 #1119). Although various devices and ablation procedure were attempted, 5-years recurrence-free survival of
ipsilateral breast tumor was 96% in 425 cases of T1BC treated with RFA alone (the mean follow-up, 50 months). To validate
complete pathological ablation in BC, we started a non-randomized phase II study in 2013 (UMIN000013836). Eligibility criteria:
Unilateral BC patients with stage 0 (TisN0M0) or I (T1N0M0 or T1N1miM0) are eligible. Written informed consent is obtained.
Tumor diameter of 2cm or less should be diagnosed by ultrasound and MR mammography. BC with diffuse calcification,
extensive intraductal components, or multiple tumors is excluded. Methods: RFA is performed by Cool-tip RF ablation system
(Covidien, USA). The needle electrode is inserted into the center of referent tumor under ultrasound guidance. The first ablation is
started at an initial electrical power level of 5W for 1 min and the power is increased at 10W for 1 min. After then, it is increased in
steps of 10W from every minute until rapid elevation of tissue impedance. The second ablation is allowed by physicians
discretion. One month later, ablated tissue is collected by core needle biopsy or vacuum-assisted breast biopsy. Cell viability is
examined by central review of independent pathologists using tumor specimens stained with hematoxylineosin and nicotinamide
adenine dinucleotide (NADH) diaphorase. In case of complete ablation, breast irradiation and adjuvant therapy will be performed.
In case of incomplete ablation, partial mastectomy should be recommended. Aims: The primary endpoint is complete ablation
rate. The secondary endpoints are deformity after RFA, relapse-free survival and overall survival for 10 years. Two-step design is
used for statistical evaluation. Finally, 32 patients will be needed. Present accrual: As of April 2014, 16 patients were enrolled.
One patient was ineligible because of macrometastasis in a sentinel node. Of 13 eligible patients who underwent pathological
examination, one patient had viable cancer cells with NADH diaphorase staining. We also investigate optimal histological
examination of cell viability and adequate image diagnosis after RFA.

Title: Will chest wall and regional nodal radiotherapy post mastectomy or the addition of regional nodal radiotherapy to breast
radiotherapy post lumpectomy reduce the rate of invasive cancer events in patients with positive axillary nodes who convert to
ypN0 after neoadjuvant chemotherapy? NSABP B-51/RTOG 1304 a phase III trial
Eleftherios P Mamounas1,2, Hanna Bandos1,3, Julia R White4,5, Thomas B Julian1,6, Atif J Khan1,7, Simona F Shaitelman4,8, Mylin A
Torres1,9, Susan A McCloskey1,10, Frank A Vicini1,11, Patricia A Ganz1,10, Soonmyung Paik1,12, Nilendu Gupta4,13, Joseph P
Costantino1,3, Walter J Curran, Jr4,9 and Norman Wolmark1,6. 1National Surgical Adjuvant Breast and Bowel Project (NSABP),
Pittsburgh, PA; 2UF Health Cancer Center at Orlando Health; 3University of Pittsburgh Graduate School of Public Health, NSABP
Biostatistical Center, PA; 4Radiation Therapy Oncology Group; 5Ohio State University; 6Allegheny Cancer Center at Allegheny
General Hospital; 7Rutgers Cancer Institute of New Jersey, New Brunswick, NJ; 8University of Texas MD Anderson Cancer
Center, Houston, TX; 9Winship Cancer Institute of Emory University, Atlanta, GA; 10University of California, Los Angeles, CA;
11
Saint Joseph Mercy Oakland (21st Century Oncology); 12Severance BioMedical Science Institute and Yonsei University College
of Medicine and 13Ohio State University, James Cancer Hospital.
Body: Background
This phase III randomized post-neoadjuvant chemotherapy trial will evaluate if chest wall and regional nodal XRT (CWRNRT)
after mastectomy or whole breast irradiation (WBI) with RNRT after breast-conserving surgery significantly reduces the rate of
events for invasive breast cancer recurrence-free interval (IBCR-FI) in patients who present with histologically positive axillary
nodes but become histologically negative axillary nodes after neoadjuvant chemotherapy. Secondary aims are OS, LRR-FI, DRFI,
DFS-DCIS, and second primary cancer. HYPOTHESIS: Can We Use Tumor and Nodal Response to Neoadjuvant Chemotherapy
in Order to Individualize the Use of L-R XRT?
Correlative science will examine the effect of RT by tumor subtype, molecular predictors of outcome for patients with residual
disease, and the development of predictors of degree of reduction in loco-regional recurrence.
Methods
Eligible patients with clinical T1-3, N1 breast cancer with pathologic axillary nodal involvement (positive FNA or core needle
biopsy) must complete 12 weeks of neoadjuvant chemotherapy (anthracycline and/or taxane-based regimen). HER2-positive
patients must receive neoadjuvant trastuzumab or other anti-HER2 therapy. After neoadjuvant chemotherapy either
breast-conserving surgery or mastectomy will be performed. At the time of surgery, all removed axillary nodes must be
histologically free from cancer. 3 or more histologically negative sentinel nodes are acceptable to determine axillary nodal
involvement. ER/PR and HER-2 neu status before neoadjuvant chemotherapy is required. All patients will receive additional
required systemic therapy.
Site radiation credentialing with a facility questionnaire and case benchmarking is required. Randomization for mastectomy
patients will be to no CWRNRT or CWRNRT and for breast-conserving surgery patients to WBI or WBI RNRT.
Statistical Considerations
1636 patients will be enrolled over 5 years with definitive analysis at 7.5 years. The study is powered at 80% to test the main
hypothesis that RT reduces the annual hazard rate of events for IBCR-FI by 35% for an absolute risk reduction in the 5-year
cumulative rate of 4.6%. Analysis will be on intent-to-treat with 3 formal interim analyses at 43, 86, and 129 events, with a 4th/final
analysis at 172 events. Current accrual is 37. (as of 6-10-14)
736 enrolled patients will be evaluated with targeted patient-reported outcome instruments focusing on the effect of RT. Patient
assessments will be prior to randomization and then at 3, 6, 12, and 24 months.
Contact Information
Study protocol information can be found under the protocol-specific web page on the CTSU member web site
https://www.ctsu.org. For protocol-specific questions contact NRG Oncology Pittsburgh Clinical Coordinating Department.
Phone: 1-800-477-7227. Email: ccd@nsabp.org. All investigators will enroll patients by accessing OPEN at https://open.ctsu.org
or from the OPEN tab on the CTSU members side of the web site.
Support: NCI PHS U10-CA-12027, -69651, -37377, -69974, and -2166.

Title: HIOB trial - Hypofractionated whole-breast irradiation preceded by intraoperative radiotherapy with electrons as anticipated
boost
Roland Reitsamer1, Gerd Fastner2, Michael Kopp2 and Felix Sedlmayer2. 1University Hospital Salzburg, Breast Center Salzburg,
Paracelsus Medical University, Salzburg, Austria and 2University Hospital Salzburg, Paracelsus Medical University, Salzburg,
Austria.
Body: ClinicalTrials.gov Identifier: NCT01343459
Brief background discussion:
The commonly accepted standard fractionation schedule for whole breast irradiation (WBI) consists of 25fractionations with single
fractional doses of 2 Gy, resulting in cumulative doses of 50 Gy to the whole breast. Hypofractionation trials from UK and Canada
demonstrated a similar outcome in local control and cosmesis compared to conventional fractionation. The combination of IOERT
boost and hypofractionation was not yet evaluated.
Trial design:
Prospective one-armed multi-center trial - ISIORT 01
Eligible patients are treated with hypofractionated whole breast irradiation (WBI) of 40.5 Gy in 2.7 Gy per fraction for 15 days,
preceded by an intraoperative electron boost (IOERT) to the tumor bed of 10 Gy (90% isodose) during breast conserving surgery.
Inclusion criteria:
Invasive breast cancer
Age 35 years and older
T1/T2, N1/N2, G1 - G3
Hormone receptor and Her 2 status: no limitations
Exclusion criteria:
Non-invasive breast cancer
Age less than 35 years
T3/T4, N2/N3
Neoadjuvant chemotherapy
Aim of the study:
To assess the effectiveness of IOERT in combination with hypofractionated WBI in terms of in-breast tumor control and cosmetic
outcome, by matching or exceeding the best published results for annual LR rates in 3 different age groups in reference to an
upper limit (exceeding = inferiority) and a lower limit (undershooting = superiority) which were estimated on the basis of the
existing literature as follows:
Age 35 40 years:
Upper limit (best published evidence): 1.2 % (EORTC 22881- Boost trial)
Lower limit (commonly expected dimension): 0.72 % (CONSORT trial)
Age 41 50 years:
Lower limit (commonly expected dimension): 0.72 % (CONSORT trial)
Age over 50 years:
Lower limit (commonly expected dimension): 0.4 % (START B trial)
Primary endpoint:
Proof of superiority or iso-effectiveness.
Secondary endpoints:
Acute toxicity (CTC- toxicity scoring system) and late toxicity (LENT SOMA scoring system), cosmetic results (Van Limbergen Score), DFS, OS.
Sequential Probability Ratio Test (SPRT):

As a consequence of best published analyses for this patient selection, stopping rules for annual local recurrence rates are
defined as follows:
Age group 35-40: H0: p1 <= 0.72% vs H1: p1 > 0.72% with p2 = 4.0%
Age group 41-50: H0: p1 <= 0.72% vs H1: p1 > 0.72% with p2 = 2.4%
Age group 51+: H0: p1 <= 0.4% vs H1: p1 > 0.4% with p2 = 1.4%
H0-Hypothesis means no superiority to standard regimen.
H1-Hypothesis means superiority of HIOB against best published evidence.
Present accrual and target accrual:
Start date: January 2011
Estimated enrollment: 1500 patients
542 patients recruited by March 2014
10 international institutions recruiting.

Title: Fulvestrant alone versus fulvestrant and everolimus versus fulvestrant, everolimus and anastrozole: A Phase III
randomized, placebo-controlled trial in postmenopausal patients with hormone-receptor-positive stage IV breast cancer:
SWOG-Clinical Trials Initiative (CTI)* S1222
George Somlo1, William Barlow2, Halle Moore3, Julie Gralow4, Anne Schott5, Daniel Hayes6, Peter Kuhn7, James Hicks8, Danika
Lew9, Debu Tripathy10 and Gabriel Hortobagyl11. 1City of Hope Cancer Center, Duarte, CA; 2SWOG Statistical Center, Seattle,
WA; 3Cleveland Clinic Foundation, Cleveland, OH; 4University of Washington, Seattle, WA; 5University of Michigan, Ann Arbor,
MI; 6University of Michigan, Ann Arbor, MI; 7University of Southern California, Los Angeles, CA; 8Cold Spring Harbor Laboratory,
Cold Spring Harbor, NY; 9SWOG Statistical Center, Seattle, WA; 10MD Anderson Cancer Center, Houston, TX and 11MD
Anderson Cancer Center, Houston, TX.
Body: BACKGROUND: The median survival for women with hormone-receptor-positive (HR+) stage IV metastatic breast cancer
(MBC) is 36 months. Strategies to improve outcome for postmenopausal women, include combinations of aromatase inhibitors
(AIs) with either the estrogen receptor downregulator fulvestrant, or with the PI3kinase/AKT/mTOR signal inhibitor everolimus, as
this pathway has been implicated as a mediator of resistance to endocrine therapies. However,fulvestrant 500 mg has not been
tested in combination with everolimus alone, or in combination with everolimus and an AI.
SPECIFIC AIMS/TRIAL DESIGN: S1222 (NCT02137837) is a randomized phase III double-blinded, placebo-controlled clinical
trial and is currently accruing. Patients receive fulvestrant (500 mg IM every 2 weeks for 3 doses, then monthly) on all arms,
everolimus 10 PO daily (Arms 2 and 3, placebo Arm 1), and anastrozole 1 mg daily (Arm 3), or placebo (Arms 2 and 1). The
co-primary objectives are to compare progression-free survival (PFS) between fulvestrant and everolimus (Arm 2) vs. fulvestrant
(Arm 1), and fulvestrant, everolimus, and anastrozole (Arm 3) vs. fulvestrant (Arm 1). Additional objectives include comparison of
PFS between Arms 2 and 3, and overall survival (OS), response and clinical benefit rates, toxicities, feasibility, compliance, and
OS among the study arms.
Translational research goals are to test molecular determinants of response in circulating tumor cells (CTCs), assess a previously
piloted CTC-Endocrine Therapy Index (CTC ETI), perform CTC-Next Generation Sequencing Analysis (CTC-NGS), and NGS on
cancer tissue and germ line DNA.
ELIGIBILITY CRITERIA: Postmenopausal women with histologically confirmed de novo or newly relapsed stage IV MBC, with
HER2-negative and HR-positive (>1% positive nuclear staining) features, with either measurable, or evaluable disease are
eligible. Previous neoadjuvant/adjuvant therapy allowed, but adjuvant anti-HR therapy must have been completed 12 months
prior to enrollment.
STATISTICAL METHODS/TARGET ACCRUAL: This is a parallel randomized design with equal allocation to the three arms: (1)
fulvestrant + placebo for everolimus + placebo for anastrozole; (2) fulvestrant + everolimus + placebo for anastrozole; (3)
fulvestrant, everolimus, and anastrozole. PFS is the primary outcome. There are two coprimary hypotheses: (1) Arm 2 versus
Arm 1; and (2) Arm 3 versus Arm 1, each will be tested at = 0.025 (2-sided) .Arm 2 versus Arm 3 is a secondary comparison.
Sample size computations are based on accrual of 33 patients per month for 24 months resulting in a computed accrual total of
792. We estimate that the final analysis will occur 24 months after the last patient is accrued, an average followup of 36 months at
the time of the final analysis with an expected trial duration of 4 years.
*SWOG-CTI manages the non-federally funded components of SWOG under The Hope Foundation. The study is supported by
AstraZeneca and Novartis Pharmaceuticals Corp.

Title: A phase 2 study of neoadjuvant goserelin and letrozole for premenopausal women with estrogen receptor positive HER2
negative stage 2 and 3 breast cancer
Andre Mattar1, Rodrigo Goncalves2, Matthew J Ellis2 and Roberto Hegg1. 1Hospital Perola Byington, So Paulo, SP, Brazil and
2
Washington University, School of Medicine, St Louis, MO.
Body: Background: More than half of premenopausal women with breast cancer have tumors that are estrogen receptor (ER)
positive. Despite this, there has been limited utilization of neoadjuvant endocrine therapy in that setting.
Trial Design: A Phase 2 study in premenopausal women (Age>18) with tumor size>2cm and locally advanced tumor, ER positive
(Allred 6-8), HER2 negative, breast cancer who are not considered candidates for breast conservation surgery (BCS) or are
borderline for BCS. Patients will be treated with a LHRH-A in combination with letrozole for a period of 4 weeks. At the end of 4
weeks, patients will undergo biopsy of the tumor for Ki67 assessment along with a serum estradiol (E2) level to confirm adequate
hormone suppression.
Patients will then fall into one of 3 groups:
1. Postmenopausal (E2<10pg/ml) and Ki6710%
These patients will remain on letrozole for 16-18 weeks followed by definitive surgery. Patients with a PEPI-0 and continued
estradiol suppression will be offered surgical oophorectomy, continued letrozole and no chemotherapy (CT).
2. Postmenopausal (E2<10pg/ml) and Ki67>10%
These patients will be designated endocrine therapy resistant and be offered neoadjuvant CT. Patients who decline neoadjuvant
CT should be offered immediate surgery.
3. Premenopausal (E210pg/ml)
There is a failure of the LHRH-A to fully suppress estradiol levels at one month, treatment decisions will be individualized.
Specific aims
1. To determine the pathological complete response (pCR) rate to CT in patients with a Ki67>10% at one month, despite a fully
suppressed estradiol level.
2. To determine the PEPI-0 rate in patients whose estradiol is fully suppressed
at both 4 weeks and 16 weeks and whose 4 week tumor Ki67<10%.
3. In patients with a PEPI-0 tumor to determine the acceptability of management with surgical oophorectomy with continued oral
letrozole and no CT.
Statistical methods:The study design was chosen assuming the expectation for the efficacy of neoadjuvant CT in premenopausal
women with ER+ breast cancer is low, with a pCR rate in the range of 5%.Our hypothesis is that patients with tumors resistant to
endocrine therapy will be enriched for sensitivity to CT to a level typical of patients with ER negative disease (20%). Thus, the trial
was designed so that at a 0.1 significance level, there would be a 90% chance of detecting a pCR rate >20% when the true pCR
rate >5%. A 90% binomial confidence interval (CI) for the true pCR rate will be constructed. We anticipate approximately 20% of
patients will have a 4 week Ki67 > 10%. A sample of 235 eligible patients will be required to obtain 35 with a 4-week post
treatment Ki67 value >10 % and estradiol level <10 pg/ml who are willing to switch to neoadjuvant CT.
We anticipate that 75% of patients will have a 4-week Ki67 <10% along with estradiol <10pg/ml. Based upon the sample size
needed, there would be 175 patients who complete 16 weeks of neoadjuvant endocrine therapy. We estimate at least 20% of the
patients with a Ki67 <10% at 4 weeks will be in the PEPI 0, Stage 0/1 group after surgery. A 90% binomial CI will be constructed
for the proportion of these who choose to forego CT.
Accrual: Open June 2014. Target=235.

Title: A pilot and phase II study of entinostat and anastrozole/tamoxifen in women with triple negative breast cancer (TNBC) to
evaluate biomarkers and surrogates for response
Saranya Chumsri1, Gauri Sabnis1, Nancy Tait1, Jane Lewis1, Emily Bellavance1, Susan Kesmodel1, Steven Feigenberg1, Katherine
RH Tkaczuk1, Peter Ordentlich3, Martin J Edelman2 and Angela H Brodie1. 1University of Maryland Greenebaum Cancer Center,
Baltimore, MD; 2University of New Mexico Cancer Center, Albuquerque, NM and 3Syndax Pharmaceuticals, Inc, Waltham, MA.
Body: Background: Triple negative breast cancer (TNBC) is an aggressive form of breast cancer that lack ER, PR, and HER2.
TNBC is inherently resistant to endocrine therapy (ET) such as aromatase inhibitor (AI) and tamoxifen. Previous studies
suggested that the loss of ER/PR expression in TNBC is due to epigenetic silencing. We have demonstrated histone deacetylase
inhibitor (HDACi), entinostat, can induce expression of functional ER and render TNBC sensitive to ET both in vitro and in vivo
(Sabnis et al. Cancer Research 2011). Furthermore, the combination of entinostat and AI can induce prolonged tumor regression
and significantly reduce lung metastasis in vivo.
Trial Design: This is a single arm phase I/II study of entinostat in combination with ET. There are 2 cohorts in this study: cohort 1
operable stage I-II TNBC, cohort 2 metastatic and unresectable locally advanced TNBC. For ET, anastrozole will be used in
postmenopausal women and tamoxifen in premenopausal women. The treatment regimen is: entinostat started on day 1 followed
by ET on day 4. In cohort 1, paraffin embedded tissue from core needle biopsy and surgical specimens will be collected. In cohort
2, biopsies before and after 15-29 days of treatment are required.
Eligibility Criteria: Female age 18 with ECOG PS 2 and TNBC defined as ER and PR < 1% with no HER2 overexpression.
Cohort 1 includes pts with operable stage I-II and cohort 2 includes pts with metastatic and unresectable locally advanced with
2 prior lines of chemotherapy who have accessible tumors for biopsies, excluding bone only metastasis.
Specific Aims: To determine ER expression as well as the percentage change in proliferation index (Ki67) after treatment.
Moreover, safety and tolerability of the combination given either in neoadjuvant or metastatic setting will be evaluated. Response
is defined as 50% reduction in Ki67 AND 1% ER-positivity OR pathologic complete response in the post-treatment surgical
specimens.
Statistical Methods: For the pilot phase, a 3+3 cohort design is employed. The Simons two-stage design is used in the phase II.
The total sample size for the phase II is 32 with 12 and 20 pts in the first and second stage respectively. The study will be
terminated if there is no response among the first 12 pts. If 1 response in these 12 pts is observed, then 20 more pts will be
accrued. The combination will be considered promising, if 4 out of 32 pts have a response. This design yields 87% power. The
probability of early stopping and declaring that this combination has no sufficient activity is 0.54, if the true success rate is 5% and
0.07 if the true response rate is 20%. If the true proportion of pts with Ki67 reduction combined with ER up-regulation is 0.20, the
probability of concluding that the drug has sufficient activity is 0.87 and 0.07 if the true proportion is 0.05. To allow about 10%
inevaluability, the phase II trial will accrue 35 eligible pts. Therefore, the total target accrual is 41 pts.
To date, there are a total of 7 pts enrolled. Accrual is currently ongoing. Please contact ntait@umm.edu for further information.

Title: Alliance A011106: ALTernate approaches for clinical stage II or III Estrogen Receptor positive breast cancer NeoAdjuvant
TrEatment (ALTERNATE) in postmenopausal women: A phase III study
Cynthia X Ma1, Vera Suman2, A Marilyn Leitch3, Souzan Sanati1, Katherine DeSchryver1, Gary W Unzeitig4, Paul Haluska5, Mark
Watson1, Olwen Hahn6, Jo Anne Zujewski7, Kelly Hunt8, Eric P Winer9, Cliff A Hudis10 and Matthew J Ellis1. 1Washington
University School of Medicine, St Louis, MO; 2Alliance Statistics and Data Center, Mayo Clinic, Rochester, MN; 3University of
Texas Southwestern Medical Center, Dallas, TX; 4Doctor's Hospital of Laredo, Laredo, TX; 5Mayo Clinic, Rochester, MN;
6
University of Chicago, Chicago, IL; 7National Cancer Institute, Bethesda, MD; 8MD Anderson Cancer Center, Houston, TX;
9
Dana-Farber Cancer Institute, Boston, MA and 10Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Neoadjuvant endocrine therapy (ET) in patients (pts) with locally advanced estrogen receptor positive (ER+) breast cancer
(BC) improves breast conservation rate, and importantly offers an opportunity for individualized assessment of tumor
responsiveness to ET to guide subsequent treatment. Previous neoadjuvant ET trials demonstrated that pathologic tumor size
(pT), axillary lymph node status (N), and tumor Ki67 value at surgery predicted risk of relapse. Preoperative endocrine prognostic
index (PEPI) was therefore developed to assign risk scores based on these factors. Modified PEPI 0, defined by pT1-2 N0 Ki67<
2.7%, was associated with extremely low risk of recurrence without adjuvant chemotherapy. In addition, high Ki67 post 2-4 weeks
(wks) of neoadjuvant ET identified resistant tumors with poor outcome.
ALTERNATE trial is a three-arm phase III neoadjuvant/adjuvant ET trial designed to achieve two primary objectives. The first is to
prospectively validate that modified PEPI 0 predicts > 95% relapse-free survival (RFS). The second is to determine whether
fulvestrant (F), or fulvestrant (F) plus anastrozole (A), is superior to A in inducing a higher rate of modified PEPI 0. Secondary
objectives include assessing RFS for pts with endocrine resistant tumor, defined by Ki67 > 10% at 4 or 12 wk, disease
progression, or modified PEPI non-0, and pathologic responses of resistant tumors to neoadjuvant chemotherapy. Correlative
studies include degrees of Ki67 suppression and ER level post treatment. In addition, ongoing research sequencing studies of
DNA and RNA will be performed to contrast sensitive vs resistant tumors.
During the first phase of the trial, 1200 pts are randomized 1:1:1 to the F, A or F/A. This provides an 82% chance, 1-sided alpha
0.025 chi-square test to detect at least 10% difference in modified PEPI 0 rate comparing F or F/A with A. While waiting for result
of analysis, the A arm will continue for the 2nd phase enrollment. Only the F-containing arm(s) superior over A will be continued to
the 2nd phase.
During the 2nd phase, an additional 540 pts in each arm is estimated to obtain 317 pts with PEPI score 0. This will have a 90%
chance, with a one-sided alpha=0.025 nonparametric Brookmeyer-Crowley type one sample survival test, rejecting that 5 year
RFS rate is 95%. The maximum sample size is 2820 pts.
Eligible pts include postmenopausal women with newly diagnosed clinical stage II or III ER+ (Allred score 6-8) HER2- BC. Tumor
biopsy is required at baseline, 4 wk, and surgery for central Ki67 analysis. Treatment decision is individualized based on 4 wk
Ki67 and modified PEPI score. Pts with Ki67 > 10% at 4 wk are switched to chemotherapy. Pts with a modified PEPI score of 0
are recommended not to receive chemotherapy, but continue the assigned ET for 1.5 years followed by 3 years of A.
This trial is currently accruing pts through CTSU. This research is supported in part by NCI BIQSFP, and grants from Breast
Cancer Research Foundation, Genentech, and the Investigator-Sponsored Study Program of AstraZeneca. The Clinical
Trials.gov Identifier: NCT01953588.

Title: Prospective multicenter study evaluating the effect of impaired tamoxifen metabolization on efficacy in breast cancer
patients receiving tamoxifen in the neo-adjuvant or metastatic setting - The CYPTAM-BRUT 2 trial
Kathleen Van Asten1, Lynn Jongen1, Anne-Sophie Dieudonn1, Anneleen Lintermans1, Chantal Blomme2, Olivier Brouckaert2,
Diether Lambrechts3, Hans Wildiers4, Marie-Rose Christiaens5, Dirk Timmerman2, Ben Van Calster6, Jan Decloedt7, Patrick
Berteloot8, Didier Verhoeven9, Markus Joerger10, Khalil Zaman11, Vincent Dezentj12 and Patrick Neven2. 1KU Leuven, Oncology,
Leuven, Belgium; 2University Hospitals Leuven, Leuven, Belgium; 3KU Leuven, Laboratory of Translational Genetics, Leuven,
Belgium; 4University Hospitals Leuven, Leuven, Belgium; 5University Hospitals Leuven, Leuven, Belgium; 6KU Leuven,
Development and Regeneration, Leuven, Belgium; 7AZ Sint-Blasius, Gynecology, Dendermonde, Belgium; 8AZ Sint-Maarten,
Gynecology and Obstetrics, Duffel, Belgium; 9AZ Klina, Medical Oncology, Brasschaat, Belgium; 10Cantonal Hospital, St-Gallen,
Switzerland; 11University Hospital CHUV, Breast Center, Lausanne, Switzerland and 12Leiden University Medical Center, Leiden,
Netherlands.
Body: Background
Tamoxifen is commonly used to treat and prevent hormone receptor positive breast cancers. This drug is metabolized into more
active metabolites by liver enzymes such as cytochrome P450 (CYP) enzymes. Endoxifen is considered to be the principal active
metabolite of tamoxifen. As CYP enzymes are highly polymorphic in humans, endoxifen plasma levels are modulated by the
patients genotype. It, however, is not yet clear if lowered endoxifen plasma levels have an effect on tamoxifen efficacy. This is
the first prospective study where the association between endoxifen plasma concentrations, multiple CYP-genotypes and clinical
outcome in postmenopausal patients treated with tamoxifen is investigated.
Trial Design
CYPTAM-BRUT 2 is a prospective multi-center open label, single-arm, non-randomized observational study. Postmenopausal
women with measurable, estrogen receptor positive breast cancer receiving tamoxifen as neo-adjuvant or as first-line metastatic
treatment are included in this study. The objective treatment response and clinical benefit are observed to investigate the efficacy
of 20 mg tamoxifen daily. Patients are allowed to have started tamoxifen before inclusion but not more than three months.
Further, if more than twelve months have passed after completion of the adjuvant therapy prior endocrine therapy in the adjuvant
setting is allowed. Patients receiving neo-adjuvant tamoxifen will be assessed no more than four months after starting with
tamoxifen.
The primary endpoint is a statistical association between steady-state endoxifen plasma concentrations and the objective
response rate (ORR) after 3-6 months of tamoxifen, under the assumption that the relationship is linear with an odds ratio (OR) of
1.49 per 10 nmol/L. Using available data on endoxifen concentrations, this OR is chosen to reflect an improvement from 10%
ORR in the lowest endoxifen quartile to 30% in the highest endoxifen quartile when the overall ORR is around 18%. To have 90%
power at a 5% significance level, 180 patients have to be included into the study. The main secondary study endpoint is the
relation between endoxifen plasma concentrations and clinical benefit (CR+PR+SD at 6 months). The study has to include 270
patients to detect a statistically significant association with endoxifen with 90% power at a 5% significance level, assuming an OR
of 1.28 per 10 nmol/L. This OR is chosen to reflect an improvement of clinical benefit at 6 months from 30% in the lowest
endoxifen quartile to 50% in the highest endoxifen quartile (overall clinical benefit around 39%). For both endpoints the RECIST
criteria are used. Other endpoints are progression-free survival, tolerability of tamoxifen treatment and the association between
CYP2D6 genotype and clinical outcome.
Patient accrualPatients from 22 participating centers in Belgium and Switzerland are included in this trial. In May 2014, the
predefined sample size of 270 patients was reached. Follow-up of the last patients will continue until all required data are
obtained (i.e blood samples and response evaluation).

Title: E2112: A randomized phase III trial of endocrine therapy plus entinostat/placebo in patients with hormone receptor-positive
advanced breast cancer
Roisin M Connolly1, Fengmin Zhao2, Kathy D Miller3, Amye J Tevaarwerk4, Lynne I Wagner5, Min-Jung Lee6, Judy Murray1, Robert
Gray2, Richard L Piekarz7, Jo Anne A Zujewski7 and Joseph A Sparano8. 1Sidney Kimmel Comprehensive Cancer Center at Johns
Hopkins, Baltimore, MD; 2Dana-Farber Cancer Institute, Boston, MA; 3Indiana University Melvin and Bren Simon Cancer Center,
Indianapolis, IN; 4University of Wisconsin Carbone Cancer Center, Madison, WI; 5Northwestern University Feinberg School of
Medicine, Chicago, IL; 6National Cancer Institute, Bethesda, MD and 7Cancer Therapy Evaluation Program (CTEP), National
Cancer Institute, Bethesda, MD.
Body: Background:
Epigenetic alterations in the genome, including abnormal DNA methylation and histone hypoacetylation, initiate and promote
cancerous changes. A potential mechanism of resistance to endocrine therapy in breast cancer involves changes in gene
expression secondary to epigenetic modifications, which might be modulated with the use of histone deacetylase (HDAC)
inhibitors such as entinostat. Results from ENCORE 301, a phase II study evaluating the addition of entinostat to the steroidal
aromatase inhibitor (AI) exemestane in patients with hormone receptor (HR)-positive advanced breast cancer who had
experienced disease progression on a non-steroidal AI (NSAI), showed a significant improvement in progression-free survival
(PFS), the primary endpoint, and overall survival (OS), an exploratory endpoint. Based on the OS results, entinostat has been
designated a Breakthrough Therapy by the FDA when used in combination with exemestane in HR-positive advanced breast
cancer.
Trial design:
E2112 is a multicenter randomized double-blind placebo-controlled phase III study (NCT02115282) enrolling patients with
advanced HR-positive, HER2-negative breast cancer who have experienced disease progression on a NSAI (n=600). Patients
will receive exemestane 25mg po daily and entinostat/placebo 5mg po every week (28 day cycle). Staging every 12 weeks.
Eligibility Criteria:
Postmenopausal women and men ( 18 years), ECOG performance status 0-1, locally advanced/metastatic invasive
adenocarcinoma of the breast: ER/PR-positive, HER2-negative, measurable or non-measurable (20% cap) disease. Disease
progression after NSAI use in the metastatic setting OR relapse while on or within 12 months of end of adjuvant NSAI therapy.
Prior CDK inhibitor or everolimus permitted, but not exemestane or fulvestrant. One prior chemo permitted in metastatic setting.
Specific Aims:
Both PFS (defined by central review) and OS are primary endpoints, and the study is designed to show an improvement in either
PFS or OS. Secondary endpoints include: Safety and tolerability, objective response rate, changes in lysine acetylation status in
peripheral blood mononuclear cells, patient-reported symptom burden and treatment toxicities, adherence.
One-sided type 1 error 0.025 split between two hypotheses tests: 0.001 for PFS test and 0.024 for OS. PFS is tested in the first
360 pts, 88.5% power to detect 42% reduction in the hazard of PFS failure (median PFS 4.1 to 7.1 months); OS is tested in all
600 pts, 80% power to detect 25% reduction in the hazard of death (median OS 22 to 29.3 months)
Present and Targeted Accrual: This study was activated in March 2014 and accrual is anticipated to complete in 40 months.
Contact Person: Dr. Roisin Connolly, Email: rconnol2@jhmi.edu
Funding: ECOG-ACRIN Cancer Research Group (ECOG-ACRIN), Cancer Therapy Evaluation Program (CTEP) at the National
Cancer Institute, and Syndax Pharmaceuticals.

Title: A phase II study with orteronel as monotherapy in patients with androgen receptor (AR) expressing metastatic breast
cancer (MBC)
Denise A Yardley1, Suzanne F Jones2, Nancy W Peacock3, Mythili Shastry2, Robyn R Young4, Andre M Kallab5 and Howard A
Burris III1. 1Sarah Cannon Research Institute and Tennessee Oncology, PLLC, Nashville, TN; 2Sarah Cannon Research Institute,
Nashville, TN; 3Tennessee Oncology, PLLC, Nashville, TN; 4Center for Cancer and Blood Disorders, Fort Worth, TX and
5
Northeast Georgia Medical Center, Gainesville, GA.
Body: Background: Androgen Receptor (AR) signaling is a new target present in and under evaluation in all breast cancer
subtypes. AR expression is noted in 70%-90% of primary breast tumors, up to 75% of all breast cancer metastases and is
associated with resistance to endocrine therapy. Orteronel, an oral, selective, nonsteroidal inhibitor of androgen synthesis, is
being developed as an endocrine therapy for hormone-sensitive cancers. In preclinical studies, orteronel suppresses sex
hormone levels in blood and hormone-dependent malignant tissue. This study will evaluate the safety and efficacy of orteronel in
the treatment of AR+ MBC.
Study Objectives: This phase II trial is designed to determine the response rate and disease control rate (CR+PR+SD at 6 mo)
following treatment with orteronel in pts with AR+ MBC. Two groups will be evaluated: Cohort 1 will include pts with AR+, triple
negative (TN) (ER-/PR-/HER2-) MBC, and Cohort 2 will include pts with AR+, ER+ and/or PR+ MBC (may be HER2+ or HER2-).
Secondary objectives include evaluation of PFS and OS, safety and tolerability, as well as evaluation of changes in serum levels
of total and free testosterone, sex hormone binding globulins (SHBG), adrenocorticotropic hormone (ACTH),
dehydroepiandrosterone sulfate (DHEA-S), cortisol and estradiol during orteronel treatment. As an exploratory objective, archived
tumor tissue will be analyzed for the presence of phosphatidylinositol 3-kinase (PIK3CA) mutations and loss of phosphatase and
tensin homolog (PTEN).
Eligibility: Pts 18 years with AR+ (10% staining by immunohistochemistry) MBC, either TN or ER+ and/or PR+, are eligible.
TNBC pts must have had 1-3 prior chemotherapy regimens in the advanced setting and ER+ and/or PR+ pts must have had 1-3
prior hormonal therapies and 1 chemotherapy regimen in the advanced setting. HER2+ pts must have received 2 lines of
HER2-directed therapies. Additional eligibility criteria include: ECOG PS 0-2; adequate bone marrow and organ function,
including left ventricular ejection fraction of 50%.
Trial Design: Six pts will be enrolled and treated with orteronel in the lead-in phase to confirm safety and tolerability. In the
absence of any safety concerns, subsequent pts will be enrolled to either Cohort 1 or Cohort 2 (described above). Cohort 1 will
contain 31 pts with AR+ TNBC and Cohort 2 will contain 55 pts with AR+ ER+ and/or PR+ MBC. All pts will receive 300 mg
orteronel PO BID over a 4 week cycle. Pts will be evaluated with CT scans every 2 cycles and treatment will continue until
disease progression or unacceptable toxicity. Response rate and disease control rate will be presented as the point estimate
along with 95% confidence interals calculated using both asymptotic normal approximation and exact binomial methods. Simons
two-stage design will be applied using alpha=0.10 and power=0.80 for each cohort. Blood samples will be collected at baseline,
on Cycle 2 Day 1, Cycle 4 Day 1, and at the end of treatment to evaluate serum total and free testosterone, SHBG, ACTH,
DHEA-S, cortisol, and estradiol levels. Archival tumor samples will be collected for exploratory PTEN and PIK3CA biomarker
evaluations. This trial is currently enrolling and has a total accrual goal of 86 pts.

Title: A phase II open label study of everolimus in combination with anti-estrogen therapy in hormone receptor-positive
HER2-negative advanced breast cancer
Denise A Yardley1, Mythili Shastry2, Laura M DeBusk2, Howard A Burris III1 and John D Hainsworth1. 1Sarah Cannon Research
Institute and Tennessee Oncology, PLLC, Nashville, TN and 2Sarah Cannon Research Institute, Nashville, TN.
Body: Background: Hormone receptor positive (HR+) breast cancer, defined as estrogen receptor and/or progesterone receptor
positive, accounts for 70% of invasive breast cancers. While most HR+ tumors initially respond to anti-estrogen therapy,
resistance, linked to crosstalk between cell signaling and signal transduction pathways, subsequently develops in almost 100%.
Preclinical data has shown that the PI3K/Akt/mammalian target of rapamycin (mTOR) pathway is activated in long-term
estrogen-deprived and aromatase inhibitor-resistant breast cancer cells. It is hypothesized that in this group of endocrine-resistant
patients (pts), resistance to anti-estrogen therapy is driven by the PI3K/Akt/mTOR pathway and hence the inhibition of this
pathway may reverse resistance. This study will evaluate the efficacy of adding everolimus (an inhibitor of mTOR) to
anti-estrogen therapy in pts with HR+ metastatic breast cancer (MBC) who have progressed on anti-estrogen therapy.
Study Objectives: This multi-center, open-label Phase II trial is designed to determine the efficacy (primary endpoint is PFS),
safety and tolerability of everolimus in combination with anti-estrogen therapy in pts with HR+, HER2-negative MBC with disease
progression on prior anti-estrogen therapy. An exploratory objective of this study is to evaluate the prognostic value of the
VeriStrat assay in this population. VeriStrat testing has been used to predict treatment efficacy in clinical trials of non small cell
lung cancer, head and neck squamous carcinoma, and MBC.
Eligibility: Pts 18 years with HR+, HER2-negative breast cancer that is either unresectable, locally recurrent, or metastatic are
eligible. Pts must have progressed on anti-estrogen therapy (recurrence while on/within 12 months of treatment for early stage
breast cancer, or progressed while on/within 1 month of treatment for advanced/MBC), but anti-estrogen therapy is not required to
be the last therapy prior to enrollment. No washout for anti-estrogen therapy is required, and 1 previous chemotherapy regimen
is allowed. Post menopausal or pre-/ peri-menopausal women on tamoxifen are eligible and ovarian function suppression is
permitted. Prior treatment with a mTOR inhibitor is not allowed. Additional eligibility requirements include: measurable or
evaluable disease, ECOG 2, adequate bone marrow and organ function and an adequate lipid profile (fasting serum cholesterol
300 mg/dL or 7.75 mmol/L and fasting triglyceride 2.5 x ULN).
Trial Design: Everolimus will be administered at a dose of 10 mg PO daily in combination with the FDA prescribed doses of the
most recent anti-estrogen therapy on which the pt progressed. Treatment cycles will be 4 weeks, with response evaluations by
CT scans every 2 cycles. Pts will be treated until disease progression or unacceptable toxicity occurs. With an alpha=0.5, a
sample size of 42 will provide 80% power to detect improvement in median PFS. Blood samples for the VeriStrat assay will be
collected at baseline, on Cycle 3 Day 1, at disease progression and at the first follow-up visit post disease progression. Archival
tumor tissue samples will be collected for comprehensive genomic profiling. This trial has a total accrual goal of 46 pts.


Title: RAD1901, a novel tissue-selective estrogen receptor degrader (SERD) demonstrates estrogen receptor engagement in a
phase 1 clinical study
Gary Hattersley1, Frank David1, Alan Harris1, Marcie Clarkin1, Kate Banks1, Greg Williams1, Andor Glaudemans2, Janine
Doorduin2, Michel Koole2, Erik de Vries2 and Rich Lyttle1. 1Radius Health, Waltham, MA and 2University Medical Center
Groningen, Groningen, Netherlands.
Body: Despite advances in the treatment of metastatic breast cancer through modulation of estrogen receptor (ER) activity, after
initial efficacy, the use of hormonal therapies is frequently followed by the development of de novo or acquired endocrine
resistance. There therefore exists a potentially significant opportunity for new agents that can overcome endocrine resistance.
One mechanism by which this could be achieved is through degradation of the estrogen receptor and thereby elimination of
estrogen receptor mediated signaling. RAD1901 is a novel, non-steroidal small molecule that selectively binds and degrades the
ER that is currently being evaluated for the treatment of metastatic breast cancer. RAD1901 has demonstrated good tissue
selectivity in preclinical models, its does not stimulate the uterine endometrium, and it protects against bone loss in an
ovariectomy-induced osteopenia rat model. In addition, we believe that RAD1901 has the ability to cross the blood-brain barrier.
In vitro, treatment of tamoxifen-sensitive and resistant human breast cancer cell lines with RAD1901 resulted in a potent
degradation of the ER (IC50 1.6nM in MCF7 cells) and inhibition of both basal and estradiol-stimulated proliferation. In a mouse
xenograft tumor model with MCF7, oral dosing with RAD1901 resulted in a significant decrease in estradiol-stimulated tumor
growth. An ongoing phase 1 clinical study in healthy volunteers is being conducted to evaluate the tolerability, safety and
pharmacokinetics of RAD1901 at escalating doses. This study is also using 18F-estradiol positron emission tomography
(FES-PET) to provide a pharmacodynamic assessment of estrogen receptor engagement/turnover. Following 6-days of daily
treatment with RAD1901, at doses that were well tolerated, a complete suppression of FES-PET signal was observed, with
standardized uptake values (SUV) comparable to background tissues. The relationship between PK and PD will be presented. To
date, the maximum tolerated dose of RAD1901 has not been determined. In conclusion, RAD1901 is a novel SERD that we
believe has the potential to be used for the treatment of hormone driven and hormone resistant metastatic breast cancers,
including breast cancer brain metastasis. The proposed phase 1b study design in metastatic breast cancer will be outlined.

Title: tnAcity: A phase II/III trial of nab-paclitaxel (nab-P) plus either gemcitabine (Gem) or carboplatin (Carbo) vs Gem/Carbo as
first-line treatment for patients with triple-negative metastatic breast cancer (TNMBC)
Denise A Yardley1,2, Robert E Coleman3, Pierfranco Conte4, Joyce O'Shaughnessy5, Javier Cortes6, Stefan Glck7, Adam
Brufsky8, Jean-Marc A Nabholtz9, Li Li10, JulieAnn Miller10, Debora Barton10, Nadia Harbeck11 and on bahalf of the tnAcity
Investigators. 1Sarah Cannon Research Institute, Nashville, TN; 2Tennessee Oncology, PLLC, Nashville, TN; 3Weston Park
Hospital, Sheffield Cancer Research Centre, Sheffield, United Kingdom; 4University of Padova, Padova, Italy; 5US Oncology,
Dallas, TX; 6University Hospital Vall d'Hebron, Barcelona, Spain; 7University of Miami, Miami, FL; 8University of Pittsburgh Medical
Center, Pittsburgh, PA; 9Centre de Lutte Contre le Cancer d'Auvergne, Clermont Ferrand, France; 10Celgene Corporation,
Summit, NJ and 11Breast Center, University of Munich, Munich, Germany.
Body: Background: TNMBC has an aggressive clinical course and poor prognosis. In phase II trials, nab-Pbased therapy has
demonstrated efficacy and tolerability as first-line therapy, in patients with MBC. The international triple-negative Albumin-bound
paclitaxel combination treatment study (tnAcity) will evaluate 2 nab-P regimens (with either Gem or Carbo) and compare the
selected regimen with Gem/Carbo as first-line treatment for TNMBC.
Trial design: All patients must have pathologically confirmed TNBC (TN defined as < 1% estrogen receptor and progesterone
receptor expression by immunohistochemistry (IHC) and 0 - 1+ human epidermal growth factor receptor 2 expression by IHC or
confirmed negative by fluorescence in situ hybridization) with measurable metastatic disease. Eligibility criteria include no prior
cytotoxic therapy for MBC, prior adjuvant or neoadjuvant anthracycline use unless not indicated by physician or other appropriate
regimens were administered, and no history or current evidence of brain metastasis. Prior neoadjuvant or adjuvant chemotherapy
must have been completed 6 months before randomization ( 12 months if using taxane-, Gem-, or platinum-containing
regimen). Patients will be enrolled internationally in North America, Europe, Latin America, and Australia. In the phase II portion,
240 patients will be randomized to receive nab-P 125 mg/m2 plus Gem 1000 mg/m2 (arm A), nab-P 125 mg/m2 plus Carbo area
under the curve (AUC) 2 (arm B), or the control regimen of Gem 1000 mg/m2 plus Carbo AUC 2 (arm C), all given intravenously
on days 1 and 8 of a 21-day cycle. The phase II portion will identify the nab-P combination to be evaluated in the phase III portion
based on efficacy and safety (Table 1). In the phase III portion, 550 patients will receive the selected nab-P regimen (A or B) or
Gem/Carbo. The primary endpoint of the phase III portion is progression-free survival (PFS) by independent radiological
assessment; secondary endpoints include overall response rate (ORR), overall survival (OS), disease control rate, duration of
response, and safety (Table 1). Monotherapy is allowed in the event of hypersensitivity or toxicity to 1 of the treatment
components. The trial design provides 90% power to detect a hazard ratio of 0.70 for PFS with a 2-sided 5% significant level.
Current enrollment is 52 patients. Clinical trial NCT01881230.
Trial Design
Parameter
Phase II
Phase III
n = 240
n = 550
1:1:1
1:1
Randomization
Stratification
Disease-free interval (DFI): 1 y vs > 1 y
DFI: 1 y vs > 1 y
Prior neo/adjuvant taxane (yes vs no)
Key Endpoints
Primary
PFS by investigator assessment
PFS by independent assessment
Secondary
ORR by investigator assessment
ORR
Percentage of patients to initiate cycle 6
OS
OS
PFS by investigator assessment
Safety
Disease control rate
Duration of response
Safety
Exploratory
Time to second-line therapy or death
Time to second-line therapy or death
Circulating tumor cells
Quality of life
Tumor biomarkers
Healthcare resource utilization

Tumor biomarkers

Title: Phase I/II clinical trial on the combination of carboplatin, eribulin and E7449
Virginia Kaklamani1, Ruth o'Regan2, Kari Wisinski3, Kent Hoskins4, James Wade5 and William Gradishar1. 1Northwestern
University, Chicago, IL; 2Emory University, Atlanta, GA; 3University of Wisconsin, Madison, WI; 4UIC, Chicago, IL and
5
CancerCare Specialists of Central Illinois, Dacatur, IL.
Body: The combination of carboplatin and eribulin has been found to be effective in triple negative breast cancer patients. A
recently conducted neoadjuvant clinical trial showed a pathologic complete response (pCR) of 45% with four cycle of carboplatin
and eribulin. Recent preclinical data from EISAI suggest that there is synergy between carboplatin, eribulin and E7449 (an oral
PARP inhibitor) in BRCA deficient and PTEN mutated xenograft models.
Loss of Heterozygosity (LOH) is DNA alterations which develop through double strand DNA breaks. The homologous
recombination (HRD) Score is calculated by the number of LOH regions greater 15Mb in length and less than an entire
chromosome present in the genome. A high HRD score is present in BRCA deficient tumors and predicts for sensitivity to
platinum-based chemotherapy. We propose a phase I/II clinical trial on the combination of carboplatin, eribulin and E7449. As
part of the phase I clinical trial HRD will be performed and correlated to response to therapy. In the phase II clinical trial women
with BRCA related breast or ovarian cancer, triple negative breast or ovarian cancer with high HRD score or PTEN deficient will
be enrolled at a dose level found during the phase I clinical trial. Tissue will be requested to test for PARP inhibition. The primary
endpoint of the phase II clinical trial is response rate. A total of 30 patients will be enrolled to the phase II clinical trial.

Title: MERIBEL study: Efficacy of eribulin in first line of taxane-resistant patients with HER2 negative metastatic breast cancer
Vanessa Ortega1, Mara Martnez2, Antonio Antn3, Isabel Garau4, Lourdes Calvo5, Esperanza Blanco6, Yolanda Fernndez7,
Jos Juan Illarramendi8, Mirta Garca9, Antonio Llombart10 and Javier Corts10. 1Vall d'Hebron Institute of Oncology, Barcelona,
Spain; 2Hospital del Mar, Barcelona, Spain; 3Hospital Universitario Miguel Servet, Zaragoza, Spain; 4Hospital Son Llatzer, Palma
de Mallorca, Spain; 5Complejo Hospitalario de A Corua, A Corua, Spain; 6Hospital Universitario Infanta Cristina, Badajoz,
Spain; 7Hospital Central de Asturias, Oviedo, Spain; 8Hospital General de Navarra, Pamplona, Spain; 9Hospital Insular de las
Palmas, Las Palmas de Gran Canaria, Spain and 10Medica Scientia Innovation Research, Barcelona, Spain.
Body: Brief background discussion:
Eribulin is a nontaxane microtubule dynamics inhibitor. It has shown a significant benefit in overall survival in patients with
metastatic breast cancer (MBC) that underwent treatments in multiple previous lines (Cortes et al. 2009). However, little is known
about the activity of the drug in patients with taxane-resistant early advanced disease. This trial is designed to determine the
safety and efficacy of Eribulin as single first line agent in the treatment of patients with HER2(-) MBC deemed resistant to
taxanes.
Trial design:
This study is a Phase IIa, multicenter, open-label, single-arm in patients with HER2 (-) metastatic breast cancer clinical trial. The
scheme consists in the administration of 1,23 mg/m2 of Eribulin (equivalent to 1.4mg/m2 eribulin mesylate) as single agent the
days 1 and 8 of each 21 days of the cycle, until progression of illness, unacceptable toxicity level or investigators criteria. The
study was approved by the competent authorities and Ethics committees from 14 participating centers in Spain and Portugal.
The principal inclusion and exclusion criteria were: (1) Patients with MBC that have not received prior chemotherapy for
Advanced disease; (2) Previous chemotherapy for early stages (I-IIIB) including taxanes (paclitaxel and/or docetaxel) for at least
4 cycles; (3) Less than 36 months* between the last cycle with taxanes and metastatic relapse; (4) Evaluable disease under
RECIST criteria; (5) Absence of HER2 over-expression (IHQ of 0/1 or negative FISH/CISH); (6) Absence of neurotoxicity > G1.
(*) Amended to 48 months (Llombart et al. ASCO 2013).
Specific aims:
The main objective is to determine the efficacy of Eribulin as single first line agent in the treatment of patients with HER2-negative
MBC and taxanes resistant criteria. Secondary objectives are: Estimate the tolerance to the treatment; assess the clinical
response of the study treatment by RECIST criteria; correlate hormone receptor status (+/-) with clinical response and tolerability;
identify the relationship between presence of visceral disease and response to therapy.
The primary endpoint is time to progression (TTP). Secondary endpoints include the safety of treatment, clinical benefit rate,
objective response, progression-free survival and overall survival.
A total of 60 patients were predefined for the primary outcome. The sample size is based on a one-arm survival design. We test
the null hypothesis that true TTP is 3.7 versus the alternative hypothesis (H1) that TTP is 5.5 months. An interim analysis with 30
patients completed has been planned. The study will be stopped for futility or efficacy if conditional power is less than 20% or
more than 80%, respectively. Type I and II error assumed were 5% one-sided and 16%, respectively.
Present accrual:
Currently, 16 women have been recruited for the trial since study start in July 2013. The expected end of accrual will be on March
2015.
Contact information for people with a specific interest in the trial:
Maria Campos (maria.campos@medsir.org)
Scientific director
Medica Scientia Innovation Research (MedSIR ARO).


Title: Metronomic eribulin in metastatic breast cancer
Pavani Chalasani2, Livingston B Robert2, Thomas A Rado3, Vijayakrishna K Gadi1, Thomas D Kummet4, Jennifer M Specht1,
Alison Stopeck2 and Hannah M Linden1. 1University of Washington, Seattle Cancer Care Alliance, Seattle, WA; 2University of
Arizona Cancer Center, Tucson, AZ; 3Columbia Basin Hematology and Oncology, Kennewick, WA and 4Olympic Medical Cancer
Center, Sequim, WA.
Body: Introduction: Breast cancer is a common and curable malignancy. In the setting of late or early recurrence, improvement
in therapy is indicated as overall survival (OS) has changed little over the past few decades. Eribulin mesylate is a non-taxane
microtubule dynamics inhibitor FDA approved in salvage metastatic breast cancer (MBC) treatment based on improvements in
OS. Eribulin mesylate is a synthetic analogue of halichondrin B, a product isolated from the marine sponge Halichondria okadai.
Anti-tubulin therapy is an established cornerstone of chemotherapeutics for MBC. Work at our center and others has shown a
favorable therapeutic index for weekly metronomic vinorelbine. The history of other effective and tolerable agents, including
paclitaxel, capecitabine, and vinorelbine, suggests that a low dose metronomic schedule of eribulin could be effective, well
tolerated and result in longer time to progression (TTP), and higher quality of life during therapy. In phase I studies, Eribulin
showed activity at doses well below the maximally tolerated 1.4mg/m2 given 2/3 weeks, suggesting a lower metronomic schedule
would not compromise efficacy. A lower dose metronomic schedule will allow responding patients to remain on treatment,
resulting in longer TTP and greater use of the drug in practice.
Study Design: This is an open-label, multi-center, phase II study of eribulin for patients with MBC. HER2 positive patients may
enroll with concomitant use of trastuzumab. Patients will receive eribulin 0.9 mg/m2 administered intravenously over 2 to 5
minutes on Days 1 8 and 15 of a 28-day cycle. Treatment will be continued until disease progression, unacceptable toxicity or
withdrawal of patient consent.
Eligibility Criteria: Patients with MBC whose disease has progressed following 1-6 prior regimens with prior exposure to a
taxane, ECOG PS of 0 2, measurable disease per RECIST 1.1, normal marrow and organ function.
Patients with known CNS metastases must have stable disease off steroids after treatment with surgery or radiation. Exclusion
criteria: mild hepatic or renal impairment, grade > 2 peripheral neuropathy, significant cardiovascular impairment or other severe
or uncontrolled medical disease or psychiatric or neurologic disorder.
Specific Aims: The primary aim is to increase progression free survival (PFS). We hypothesize that metronomic dosing of
eribulin will result in PFS of 4-6 months.
Secondary aims include decreasing the frequency of alopecia to less than 50%, grade 3 or 4 neutropenia to less than 30% and
sensory neuropathy (all grades) to less than 25% of subjects enrolled.
An exploratory aim is to assess the role of circulating endothelial cell precursors (CEPs) and apoptotic circulating endothelial cells
in predicting early response to treatment.
Statistical Methods: PFS will be measured as the time from study enrollment until the earliest date of disease progression or
death. The sample size of n=60 has 99% power for the lower bound of the confidence interval to be greater than 2.2 months (the
median PFS in the physicians discretion arm of the EMBRACE trial), assuming that the true median PFS is 4 months,
comparable to or an incremental improvement over the median PFS for the eribulin arm of the EMBRACE trial (3.7 months).
Currently 12 of 60 expected enrollments have occurred.

Title: Phase II single arm study of liposomal doxorubicin (D), bevacizumab (A), and temsirolimus (T) for treatment of triple
negative early breast cancer refractory to standard neoadjuvant chemotherapy
Rashmi K Murthy1, Naoto T Ueno1, William F Symmans2, Jennifer K Litton1, Banu K Arun1, Nuhad K Ibrahim1, Ricardo H Alvarez1,
Michael Z Gilcrease2, Vicente Valero1, Debasish Tripathy1 and Stacy L Moulder1. 1University of Texas MD Anderson Cancer
Center, Houston, TX and 2University of Texas MD Anderson Cancer Center, Houston, TX.
Body: Background: Historically, triple negative breast cancers (TNBCs) were categorized into an ill-defined group that lacked
estrogen receptors, progesterone receptors, and overexpression of human epidermal growth factor receptor 2. When TNBCs are
treated with standard neoadjuvant chemotherapy (NAC), a pathologic complete response (pCR) is achieved in only 30-50% of the
time. In an effort to improve outcomes for chemotherapy - refractory disease, emerging subtypes of TNBC have been identified
through advances in molecular profiling leading to the discovery of distinct molecular aberrations which may be targeted. For
example, mesenchymal-like TNBCs are a subset with aberrations that activate the PI3K/AKT/mTOR axis. Similar to
mesenchymal-like, metaplastic breast cancers are commonly triple negative, refractory to standard therapy, and also harbor a
high rate of molecular aberrations that lead to activation of the PI3K pathway. As such, metaplastic tumors may act as a
morphologically identified surrogate of response to evaluate the activity of targeted therapy regimens in mesenchymal-like
TNBC. Based upon favorable results noted in a phase I trial of liposomal doxorubicin, bevacizumab, and the mTOR inhibitor
temsirolimus (DAT) in metastatic metaplastic tumors, we will conduct this single arm phase II trial with neoadjuvant DAT for
chemotherapy-refractory mesenchymal-like TNBC.
Trial Design: At our institution, women with TNBC who are receiving NAC will be enrolled onto a triaging protocol, where they will
be randomized in a 2:1 fashion to be evaluated by a microarray-based predictive signature to determine chemotherapy sensitivity
and categorized as chemotherapy-sensitive (CS) or chemotherapy-insensitive (CI). Subsequently, patients will be treated with
standard anthracycline-based chemotherapy (AC or FAC) for 4 cycles. Patients with CI tumors with <80% decrease in tumor size
and those with CS tumors with <10% decrease in tumor size will be deemed non-responders. Mesenchymal-like non-responders
will be recommended for this single arm phase II study of DAT. Eligible patients will receive D at a dose of 30mg/m2 IV on day 1,
A at 15 mg/kg IV on Day 1, and T at 25mg IV on Days 1, 8, and 15 for 4 cycles.
Eligibility Criteria: Female; > 18 years of age; newly diagnosed stage I-III triple negative breast cancer; refractory to standard
NAC; LVEF > 50% by Echocardiogram or MUGA; ECOG PS 0-1
Statistical Methods: The primary endpoint is pCR following therapy with DAT. Secondary objectives will include recurrence-free
survival and overall survival. Using a Simon optimal two-stage design with alpha = beta = 10%, and then setting the threshold for
an acceptable pCR rate at 20%, we would enroll 12 patients into the first stage. If we see 0 patients with pCR, we would stop the
study after the first stage. If we see at least one patient with a pCR we will continue to enroll 25 more patients for a total of 37
patients. We would declare the treatment worthy of further study if we see at least 4 patients with pCR out of the 37 patients. This
design has a 54% probability of early termination after the first stage if the true pCR probability is 5%.
Target Accrual: 37.

Title: A phase II study of S-1 (14 days administration followed by 7 days rest) for metastatic breast cancer
Akira Hirano1, Akinori Hattori1, Kaoru Ogura1, Hiroaki Inoue1, Fumie Okubo1, Mari Kamimura1, Shiho Sakaguchi1, Jun Kinoshita1,
Reiko Miyamoto2, Norie Jibiki2, Yoshihiko Naritaka1 and Tadao Shimizu1. 1Tokyo Women's Medical University, Medical Center
East, Tokyo, Japan and 2Tokyo Women's Medical University, Yachiyo Medical Center, Yachiyo, Japan.
Body: Background
S-1, an orally administered fluorinated pyrimidine (FP), improves overall survival in the adjuvant setting of gastric and pancreatic
cancer. It is also effective for metastatic breast cancer (Yamamoto et al. Anticancer Res 30:3827-32, 2010). Although S-1 is
usually administered for 28 days followed by 14 days rest, severe adverse events are often observed by Day 15.
By altering the schedule into 14 days administration followed by 7 days rest, discontinuation of treatment can often be avoided in
daily practice. This schedule maintains the same dose intensity.
Equally, an anti-tumor effect and fewer adverse events are observed in 14 days administration followed by 7 days rest than in 28
days administration followed by 14 days rest for the treatment of squamous cell carcinoma of the head and neck (Tsukuda et al.
Br J Cancer 93:884-889, 2005).
In order to reduce the occurrence and severity of adverse events of S-1, we paid attention to dose reduction according to the
renal function of patients. S-1 is comprised of tegafur, 5-chloro-2, 4-dihydroxy-pyridine(CDHP) and potassium oxonate. CDHP
enhances the serum concentration of 5-FU by inhibiting dihydropyrimidine dehydrogenase (DPD), and it is excreted by the
kidneys. Therefore, an accumulation of DPD in patients with renal dysfunction escalates the serum concentration of 5-FU and
induces more severe adverse events. Thus, the dosage of S-1 should be reduced according to renal function.
Trial design
This is a phase II trial to evaluate the efficacy and toxicity of S-1 (14 days administration followed by 7 days rest) for metastatic
breast cancer. S-1 is administered orally twice daily at a dose of 40-60 mg depending on the BSA and creatinine-clearance from
Days 1 to 14, which is followed by 7 days rest. This treatment is repeated until disease progression.
(Trial registration :UMIN 000014096)
Eligibility criteria
Patients with histologically confirmed breast cancer, and recurrent and /or metastatic disease are included. Patients with
HER2-positive (IHC:3+ or FISH+) tumor are excluded. Patients within 2 lines of previous chemotherapy for metastatic disease are
allowed. Eligible patients are aged between 20 years and 80 years with a performance status of 0 to 2 and adequate organ
function.
Specific aims
The primary endpoint is the response rate, and the secondary endpoints are disease control rate, progression-free survival, time
to failure, toxicities and feasibility.
Statistical methods
The sample size was calculated using the Simon method, with a type I error of 5% and a study power of 90%. The unpromising
response rate was 15% and the promising rate was 42%. The required number of patients was estimated to be 23.
Target accrual
Accrual of 25 patients was planned to produce a minimum of 23 evaluable patients. Patient accrual started in June 2014.

Title: No Show

Title: SGN-LIV1A: A phase 1 trial evaluating a novel antibody-drug conjugate in patients with LIV-1-positive breast cancer
Andres Forero1, Howard Burris III2, Patricia LoRusso3, Jennifer Specht4, Kathy Miller5, Monica Mita6, Minetta C Liu7, Shanu Modi8,
Lajos Pusztai9, Django Sussman10 and Ana Kostic10. 1University of Alabama, Birmingham, AL; 2Sarah Cannon Research Institute,
Nashville, TN; 3Karmanos Cancer Institute, Detroit, MI; 4Seattle Cancer Care Alliance, University of Washington, Seattle, WA;
5
Indiana University Simon Cancer Center, Indianapolis, IN; 6Cedars-Sinai Medical Center, Los Angeles, CA; 7Mayo Clinic,
Rochester, MN; 8Memorial Sloan Kettering Cancer Center, New York, NY; 9Yale University School of Medicine, New Haven, CT
and 10Seattle Genetics, Inc, Bothell, WA.
Body: Background
First identified as an estrogen-inducible gene in a breast cancer cell line, LIV-1 is a multispan transmembrane protein of the
solute-carrier family 39 with putative zinc transporter and metalloproteinase activity. As a downstream target of STAT3, it
promotes the epithelial-to-mesenchymal transition that is important in the malignant progression to metastasis. LIV-1 is expressed
in a number of cancers with the highest prevalence and level of expression in breast, prostate, and melanoma. Additionally, LIV-1
has been linked with malignant progression to metastasis and associated with lymph node involvement in breast cancer. Normal
tissue expression is predominantly limited to hormonally regulated tissues, including breast and prostate. In metastatic breast
cancer tissue samples, the current validated immunohistochemistry assay for LIV-1 has detected moderate to high expression in
82% of samples tested, with 88% positivity in hormone receptor-positive tumors and 73% in triple-negative tumors.
SGN-LIV1A is an antibody-drug conjugate (ADC) composed of a humanized anti-LIV-1 monoclonal antibody conjugated to the
microtubule-disrupting agent monomethyl auristatin E (MMAE) via a protease-cleavable linker. In vitro, SGN-LIV1A shows
target-specific internalization and cytotoxic activity against LIV-1-positive neoplastic cell lines. Significant dose-dependent tumor
regression was demonstrated in mouse xenograft models.
Methods
The primary objective of this first-in-human phase 1, open label, multicenter study is to evaluate the safety and tolerability, and to
identify the maximum tolerated dose of SGN-LIV1A using a 3+3 dose-escalation study design. Pharmacokinetics,
immunogenicity, and antitumor activity will also be evaluated (ClinicalTrials.gov #NCT01969643). Enrollment to this US-based
trial began in late 2013.
Eligible patients are adult females who have hormone receptor-positive/HER2-negative or triple-negative metastatic breast
cancer. Tumor tissue must be positive for expression of LIV-1 per central assessment. Patients must have received at least 2
prior cytotoxic regimens in the metastatic setting and have measurable disease per RECIST v1.1. Pre-existing neuropathy
Grade 2 is not permitted.
SGN-LIV1A is administered intravenously every 3 weeks at protocol-defined doses starting at 0.5 mg/kg. Patients who achieve an
objective response or stable disease per RECIST v1.1 are eligible to continue treatment until disease progression. At the
completion of dose escalation, expansion cohorts may be opened to enroll patients with specific breast cancer subtypes to further
define safety and antitumor activity. Tumor biopsies are being obtained at baseline and after one cycle of treatment to evaluate
the role of LIV-1 expression, target saturation, and other tumor-specific measurements in the antitumor activity of SGN-LIV1A.

Title: A prospective study of glycoprotein 88 GP-88 blood test in healthy women with Gail model risk <=1.66 undergoing
screening for breast cancer (BC) with mammography (MM)
Katherine HR Tkaczuk1, Cristina Campassi1, Susan Kesmodel1, Emily Bellavance1, John Olson1, Elizabeth Nichols1, Steven J
Feigenberg1, Binbin Yue2, David Hicks2 and Ginette Serrero2. 1University of Maryland Greenebaum Cancer Center, Baltimore, MD
and 2A & G Pharmaceutical, Columbia, MD.
Body: Background: In the US, the majority of BC is diagnosed by screening XRAY mammography (MM) but up to 20% of BC
are undetected by MM. The development of reliable, blood-based BC screening test to increase the sensitivity and specificity of
currently existing BC screening methods such as MM.
Rationale: GP88 is expressed & secreted by BC cells & is not expressed by normal mammary epithelial cells. Two retrospective
randomized multi-site trials (one training study & one validating study with about 300 cases each) have demonstrated that
elevated GP88 expression in ER+ IDC was statistically correlated with a 4-fold increase in the risk of 5-yr disease recurrence.
Multivariate analysis showed that GP88 as a risk predictor was independent from PR expression, tumor size, grade, lymph node
status & disease stage. The quantitative GP88 EIA developed to determine the amount of GP88 in biological fluids was
developed at A&G. The assay is highly specific for GP88 & both sensitive & linear over a wide dynamic range, i.e. detection of
GP88 concentrations from 0.1 to 20ng/ml. A baseline (28.4 5 ng/ml) was established for healthy volunteers (HV). In BC pts a
statistically significant increase of serum GP88 was seen for early stage pts (40.7 16 ng/ml; p=0.007). Additionally, a
stratification of pts according to their clinical outcomes shows that pts having no evidence of disease (NED) have serum GP88
levels within the range of HV. These data suggest that pts with breast tumors express & secrete high levels of GP88.
Objectives: 1. To determine prospectively GP-88 blood levels in HV at average risk of developing BC who are undergoing
screening MM & in women with biopsy-confirmed BC. 2. To establish the statistical distribution of GP88 serum levels in these
women by baseline BIRAD classification (1-6). 3. To determine if baseline GP88 level is predictive of change in BIRADS
classification from baseline to 12-mos follow-up. 4. To determine if baseline GP88 level is predictive of the appearance of BC at
12 mos follow-up in HV who were cancer-free at baseline.
Inclusion Criteria: Female, aged >=40 yrs old, presenting to UM Breast Center for screening MM or for diagnostic MM or
diagnostic workup and/or biopsy due to BIRADS 0 MM at an outside facility <= to 12 wks; Gail model assessment of risk of
developing BC<=1.66.
Study procedures: Serum levels of GP88 in subjects with average BC risk factors will be measured prospectively at baseline;
3-6 mos & 6-12 mos. & correlated with BIRADS reading of the screening MM, BIRADS 1 or 2 (benign), BIRADS 3 (short term MM
follow-up) & BIRADS 4 or 5 (suspicious and tissue diagnosis needed immediately). GP88 blood level will be correlated with
pathologic results of breast biopsies performed on subjects with suspicious BIRADS (4 & 5) MM & final pathologically confirmed
diagnosis of BC as BIRADS 6.
Statistical Considerations The total number of pts will be 725 & screened up to 1400 subjects. Study is UM IRB approved & just
started accruing. Funding is provided by Maryland Industry Partnership Grant (MIPS)& Avon Grant No. 02-2013-018. Contact:
Ntait@umm.edu.

Title: Testing BRCA 1/2 mutation using next generation sequencing (BRCANGS)
Hyung Seok Park1, Seo Jin Park1, Sanghwa Kim1, Jegyu Ryu1, Jee Ye Kim1, Seho Park1, Seung Il Kim1, Joo Hyuk Sohn1 and
Jong Rak Choi1. 1Yonsei University College of Medicine, Seoul, Seodaemoongu, Korea.
Body: Testing BRCA 1/2 mutation is important for patients with breast cancer, and Sanger sequencing is a standard method to
identify BRCA 1/2 mutation. Next generation sequencing (NGS) is a high-throughput parallel sequencing that can provide genetic
information with high accuracy. NGS is a faster and cost-effective method to detect gene mutations compared to Sanger
sequencing. This study is a prospective non-randomized obsevational study which evaluates the clinical role of NGS testing for
BRCA 1/2 compared to Sanger sequencing. Eligibility criteria are wome aged 19 to 80 years, patients with breast or ovarian
cancer history in 2nd degree family members, male breast cancer, bilateral breast cancer, patient with breast cancer under 40
year of age, simultaneous breast and ovarian cancer, patients with epithelial ovarian cancer, and breast cancer with other
simultaneous extramammary malignancy. Primary endopont of the study is the overall accuracy of NGS compared to Sanger
sequencing. Seconary endponts are sensitivity, specifcity, false negative, and false positive rates of NGS compared to Sanger
method. Target accrual of the pilot study is 12 patients, and futher extension of patients accrual is scheduled after the anlaysis of
the pilot study. This study is approved by institutial review board of Severance Hospital. The study is conducted and planned by
Prof. HS Park, Yonsei University College of Medicine, Seoul, Korea and supported by Korea Breast Cancer Foundation.
ClinicalTrials.gov identifier is NCT02151747.

Title: Brain metastases in breast cancer network Germany (BMBC, GBG 79): Multicentric, retro- and prospective collection of
patient data and biomaterial from breast cancer patients with brain metastases
Isabell Witzel1, Elena Laakmann1, Tanja Fehm2, Volker Mbus3, Gunter vonMinckwitz4, Julia Kaiser4, Sibylle Loibl4 and Volkmar
Mller1. 1University Medical Center Hamburg-Eppendorf, Hamburg, Germany; 2Translational Research Board of the
Arbeitsgemeinschaft fr Gynkologische Onkologie (AGO-Trafo), Germany; 3Breast Study Group of the Arbeitsgemeinschaft fr
Gynkologische Onkologie (AGO-B), Germany and 4German Breast Group, Neu-Isenburg, Hessen, Germany.
Body: Background: The incidence of brain metastases in breast cancer patients is rising and has become a major clinical
challenge in the last years. So far, limited therapeutic options and insights into the biology of brain metastases exist. Most reports
include patients with brain metastases from different tumor entities and few studies so far analyzed exclusively patient data or
tumor samples from breast cancer patients. In this context, an open, multicenter registry and biobank should be helpful for
analysis of a large number of breast cancer patients with brain metastases. Therefore, we initiated the Brain Metastases in Breast
Cancer Network Germany (BMBC; GBG79).
Trial design and eligibility criteria: Patients with brain metastases from breast cancer are eligible. Registration of patient data is
allowed retrospectively and prospectively. To enable an easy access, an internet-based database was chosen ("MedCodes" by
the German Breast Group).
Specific aims: Documented data cover incidence, symptoms, number and localization of brain metastases, histopathology,
imaging methods, outcome, therapy, and quality of life measurement. Tissue of brain metastases and primary tumors will be
collected for translational research projects. Planned analyses include treatment patterns in Germany, patient outcome, as well as
validation of prognostic scoring systems in a multicenter setting and in the context of new targeted therapies. Translational
research projects investigate the impact of glycosylation, resistance mechanisms against HER2-targeted therapies, role of the
blood brain barrier, evaluation of markers of radioresistance and specific genomic alterations associated with brain tropism of
breast cancer cells. The registry and biobank should be also open for cooperation with other groups.
Present accrual and target accrual: The accrual target is app. 1000 patients in the database and app. 400 tissues of brain
metastases from German centers. The study was opened for documentation in April 2014 with more than 50 participating centers
as of June 2014.

Title: No Show

Title: A two-cohort, phase II, cardiac safety study of pertuzumab, trastuzumab, and neoadjuvant anthracycline-based
chemotherapy in patients with HER2-positive, locally advanced, inflammatory, or early breast cancer
Sandra M Swain1, Michael S Ewer2, Hannah Douthwaite3, Tania Szado4, Maeve Waldron-Lynch3 and Chau Dang5. 1MedStar
Washington Hospital Center, Washington, DC; 2University of Texas MD Anderson Cancer Center, Houston, TX; 3Roche Products
Limited, Welwyn, United Kingdom; 4Genentech, San Francisco, CA and 5Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background: In HER2-positive early breast cancer (BC), neoadjuvant pertuzumab (P) + trastuzumab (T) + docetaxel (D)
significantly increased pathologic complete response (pCR) rates compared with T+D (Gianni L, et al. Lancet Oncol 2012); high
pCR rates were also seen with P+T in combination with anthracycline- and non-anthracycline-based regimens (Schneeweiss A,
et al. Ann Oncol 2013). Neoadjuvant P+T + chemotherapy was well tolerated, with little increased toxicity associated with the
addition of P. Although there was a higher rate of cardiac adverse events (AEs), left ventricular systolic dysfunction (LVSD), and
left ventricular ejection fraction (LVEF) declines in P+T-containing arms compared with arms without P, these events were
generally asymptomatic, reversible, and patients (pts) recovered with no sequelae. No data are available for neoadjuvant P-based
regimens that include paclitaxel (Ptx) and dose-dense doxorubicin plus cyclophosphamide (ddAC), despite these being widely
used in (neo)adjuvant BC therapy. This study will evaluate the cardiac safety of two anthracycline-based neoadjuvant regimens in
combination with P+T. Following surgery, all pts will receive adjuvant P+T.
Trial design: In this non-randomized, open-label, phase II study, two parallel cohorts of pts receive different P-based
neoadjuvant treatment regimens. In Cohort A, pts receive four 2-weekly cycles of ddAC, followed by 12 doses of weekly Ptx, plus
four 3-weekly (q3w) cycles of P+T from the start of Ptx. In Cohort B, pts receive four q3w cycles of 5-fluorouracil, epirubicin, and
cyclophosphamide, followed by four q3w cycles of D, plus four q3w cycles of P+T from the start of D. Following surgery, both
cohorts receive 13 cycles of P+T.
Eligibility: Pts with HER2-positive, locally advanced, inflammatory, or early BC (size >2 cm or >5 mm and node-positive) are
eligible if they have a baseline LVEF 55% and an ECOG performance status 1. Among the exclusion criteria are prior incisional
biopsy or excision of the primary tumor, prior systemic or radiation therapy for cancer, and prior chemopreventive agents. Pts are
also excluded if they have poorly controlled hypertension, angina requiring medication, a history of congestive heart failure,
serious or uncontrolled cardiac arrhythmia requiring treatment, or myocardial infarction within 6 months prior to enrollment.
Aims: The primary objective is cardiac safety during the neoadjuvant period, assessed by incidence of symptomatic LVSD and
LVEF changes. Secondary objectives include cardiac safety in the adjuvant and post-treatment periods, incidence of all AEs, pCR
rate, event-free survival, invasive disease-free survival, and overall survival.
Statistical methods: No statistical hypothesis testing will be performed. Safety and efficacy results will be summarized
descriptively. A sample size of 200 pts per cohort is expected to provide sufficient data to evaluate the cardiac safety of each
regimen with acceptable precision (exact Clopper-Pearson 95% CIs) around the expected rates.
Accrual: The first patient in is planned for August 2014.

Title: A Phase 3 randomized, double-blind trial comparing PF-05280014 + docetaxel and carboplatin vs. trastuzumab + docetaxel
and carboplatin for neoadjuvant treatment of operable HER2+ breast cancer
Ira Jacobs1, Jennifer Coiro1, Fiona Hilton2, John Orazem2, Richat Abbas3 and Charles Zacharchuk4. 1Oncology Biosimilars, Pfizer,
New York, NY; 2Clinical Statistics, Pfizer, New York, NY; 3GEPB, Pfizer, Collegeville, PA and 4Pfizer, Cambridge, MA.
Body: Background: PF-05280014 is being developed as a potential biosimilar to trastuzumab. PF-05280014 demonstrated
similarity to trastuzumab in nonclinical evaluations. In a Phase I trial in healthy subjects, pharmacokinetic (PK) characteristics and
safety profile of PF-05280014 were similar to those of trastuzumab. The goal of this Phase 3 trial is to demonstrate that the
efficacy and safety of PF-05280014 + docetaxel and carboplatin are similar to those of trastuzumab sourced from the EU
(trastuzumab-EU) + docetaxel and carboplatin in the neoadjuvant treatment of women with HER2-positive operable breast
cancer.
Trial design: In this randomized, double-blind trial, subjects will be randomized (1:1) with stratification by primary tumor size (<5
cm or 5 cm), estrogen receptor (ER) status, and progesterone receptor (PR) status to PF-05280014 + docetaxel and carboplatin
or trastuzumab-EU + docetaxel and carboplatin. PF-05280014 or trastuzumab (8 mg/kg in Cycle 1; 6 mg/kg thereafter over 90
min) will be administered followed by docetaxel (75 mg/m2) and carboplatin (target area under the curve [AUC]: 6 mg/mL/min; 30to 60-minute infusion) every 3 weeks for 6 treatment cycles. The primary objective is to compare the percentages of patients with
Cycle 5 Ctrough (trastuzumab serum trough concentration) >20 g/mL in the neoadjuvant setting. Secondary objectives include
measures of tumor control, safety, immunogenicity, PK, and to explore the relationship between drug exposure and pathologic
complete response (pCR).
Eligibility criteria: Female subjects with known ER and PR status 18 years with confirmed HER2 overexpressing breast cancer
and a plan for definitive surgical resection and neoadjuvant chemotherapy, Eastern Cooperative Oncology Group status 0 or 1,
normal left ventricular ejection fraction and normal laboratory values are eligible. Key exclusion criteria are bilateral or
inflammatory breast cancer; prior treatment, including chemotherapy, endocrine therapy, biologic therapy, radiation or surgery
(except diagnostic biopsy); other concomitant active malignancy or history of malignancy in the past 5 years or presence of
known distant metastases. All subjects must provide informed consent.
Specific aims: The goal of this Phase 3 trial is to demonstrate that PF-05280014 in combination with docetaxel and carboplatin
has similarity in PK (trough level) and comparable efficacy and safety versus trastuzumab-EU + docetaxel and carboplatin in
subjects with operable HER2-positive breast cancer in the neoadjuvant setting.
Statistical methods: This study tests whether percentage of subjects with steady state (Cycle 5) Ctrough >20 g/mL of
PF-05280014 is similar to that of trastuzumab-EU, using a noninferiority margin of -12.5% tested with =0.025 (one-sided).
Assuming the percentages of subjects reaching steady state is 95% with trastuzumab-EU and 93% with PF-05280014, 188
subjects (94/arm) will be needed to achieve 85% power.
Target accrual: 220 subjects.

Title: An open-label, multicentre, phase IIIb study with intravenous administration of pertuzumab, subcutaneous trastuzumab, and
a taxane in patients with HER2-positive metastatic breast cancer (SAPPHIRE)
Natasha Woodward1, Richard H De Boer2, Andrew Redfern3, Vita Von Neumann-Cosel4, Ronelle M Heath5 and Jane Beith6.
1
Mater Hospital Cancer Services, Brisbane, Queensland, Australia; 2Royal Melbourne Hospital, Parkville, VIC, Australia; 3Royal
Perth Hospital, Perth, WA, Australia; 4Roche Products, Pty. Limited, Dee Why, NSW, Australia; 5Novotech, (Australia) Pty.
Limited, Pyrmont, NSW, Australia and 6Chris O'Brien Lifehouse, Camperdown, NSW, Australia.
Body: Background: The primary goals of treatment for patients (pts) with metastatic breast cancer (mBC) are maximising
survival and preserving the quality of life. Intravenous (IV) trastuzumab has proven clinical benefits in pts with human epidermal
growth factor receptor 2 (HER2)-positive mBC. Pertuzumab also targets HER2 through an independent epitope to that of
trastuzumab. Addition of pertuzumab to the established combination of trastuzumab and docetaxel has shown improved efficacy
with acceptable toxicity in mBC (Swain et al. Lancet Oncol 2013;14:461-71); they are considered standard of care. Subcutaneous
(SC) and IV trastuzumab formulations have shown comparable efficacy but SC administration is preferred by pts for reducing
duration of clinic visits (Pivot, et al. Lancet Oncol 2013;14: 962970). The combination of IV pertuzumab and SC trastuzumab has
not been studied. The aim is to assess safety, tolerability, and efficacy of combining IV pertuzumab with SC trastuzumab and a
taxane, as first-line therapy in pts with HER2-positive mBC.
Trial Design: SAPPHIRE is an open-label, multicentre, Phase IIIb study. Pts will receive IV pertuzumab every 3 weeks with a
loading dose of 840 mg and subsequent doses at 420 mg combined with SC trastuzumab at 600 mg/5 mL every 3 weeks and a
taxane (docetaxel, paclitaxel, or nab-paclitaxel; regimen determined by the investigator). Treatment will continue until disease
progression, unacceptable toxicity, or pts withdraw consent, whichever occurs first. The study is expected to run for 42 months.
Eligibility: Pts aged 18 years old with histologically or cytologically confirmed HER2-positive [immunohistochemistry (IHC)
positive at 3+ or in situ hybridisation-positive (ISH+)] mBC with at least one measurable lesion and/or non-measurable disease
according RECIST version 1.1 and ECOG performance status (PS) 0-2 are eligible.
Specific Aims: The primary objective is to assess the safety and tolerability of combining IV pertuzumab with SC trastuzumab
and investigators choice taxane chemotherapy. The secondary objectives are to assess the efficacy of the first-line combination
of pertuzumab, trastuzumab, and taxane chemotherapy and second-line treatments for mBC after disease progression.
Statistical Methods: Primary safety analyses will report incidence and severity of adverse events (AEs)/serious AEs, AEs
leading to premature discontinuation of study treatment and evidence of cardiac dysfunction. Secondary safety analyses will
include exposure and duration of study treatment, ECOG PS and laboratory data. Secondary efficacy analyses will report the
overall response rate, progression-free survival, event-free survival and overall survival. Continuous data will be summarised
using mean, median, range, standard deviation, and standard error. Discrete data will be summarised using frequency counts and
percentages. Time-to-event analyses will be based on Kaplan-Meier methodology.
Accrual: The planned 50 pts will be recruited from 13 Australian Centres. The study started in December 2013 with 24 pts now
enrolled. Clinicaltrials.gov#NCT02019277.

Title: A phase 2 study of eribulin in combination with pertuzumab and trastuzumab for advanced or recurrent human epidermal
growth factor receptor 2 (HER2)-positive breast cancer (SONG-02)
Hajime Abe1, Shunji Kamigaki2, Yoshifumi Komoike3, Nobuki Matsunami4, Yoshiaki Nakano5, Kenji Tezuka6, Junji Tsurutani3, Jun
Yamamura2 and Keiichi Yamazaki1. 1Bell Land General Hospital, Sakai, Japan; 2Sakai City Hospital, Sakai, Japan; 3Kinki
University Faculty of Medicine, Osakasayama, Japan; 4Osaka Rosai Hospital, Sakai, Japan; 5Kaizuka City Hospital, Kaizuka,
Japan and 6NHO Kinki-Chuo Chest Medical Center, Sakai, Japan.
Body: Background: Pertuzumab, an anti-HER2 humanized monoclonal antibody that inhibits receptor dimerization, has a
mechanism of action that is complementary to that of trastuzumab, and the combination of pertuzumab plus trastuzumab plus
taxanes, when used as first-line treatment for HER2-positive metastatic breast cancer, significantly prolonged progression-free
survival (PFS). However, in the second and later line treatment of HER2-positive advanced or recurrent breast cancer, it has not
settled whether it should be treated with pertuzumab plus trastuzumab plus chemotherapy or with trastuzumab plus
chemotherapy. Eribulin mesylate is a non-taxane microtubule dynamics inhibitor that has been proved to prolong the overall
survival of advanced or recurrent breast cancer patients compared with the treatment of physician's choice. The benefit of eribulin
in combination with trastuzumab for patients with locally recurrent or metastatic HER2-positive breast cancer has been reported.
However, the efficacy and safety of eribulin in combination with pertuzumab and trastuzumab for advanced or recurrent
HER2-positive breast cancer patients has not been reported. The purpose of this study is to evaluate the efficacy and safety of
eribulin in combination with pertuzumab and trastuzumab as second and later line therapy for patients with advanced or recurrent
HER2-positive breast cancer.
Trial Design: This is a multicenter single arm phase 2 study for advanced or recurrent HER2-positive breast cancer patients who
have experienced progression with anti-HER2 therapy. Patients received eribulin mesylate 1.4mg/m2 administered via
intravenous (IV) infusion over 2 to 5 minutes on Days 1 and 8 each 21-day cycle and pertuzumab 840mg/kg IV and trastuzumab
8mg/kg IV over 90 minutes on Day 1 of Cycle 1. Thereafter eribulin mesylate 1.4mg/m2 and pertuzumab 420mg/kg and
trastuzumab 6mg/kg was infused each 21-day cycle until disease progression or the appearance of toxic effects that could not be
effectively managed. The primary endpoint is PFS of the combination therapy, based on local assessment of response using
RECIST 1.1 criteria. Secondary endpoints are overall response rate (ORR), safety and tolerability. In addition, we examine PFS
and safety according to the most recent treatment regimen. The study was conducted in accordance with the Declaration of
Helsinki (2008), and the protocol and informed consent forms were submitted for approval to institutional review boards by the
primary investigators. All patients provided written informed consent before undergoing any study-related procedures
Statistical Method: All efficacy analyses were based primarily on the full analysis set (FAS), which included all patients who
received over 1 dose(s) of study treatment. The PFS and ORR were calculated 95% confidence intervals (CIs). Treatment of 39
evaluable patients with the identified phase 2 doses will detect this difference with a power of 80% and alpha=5% (one-sided
test). Accounting for a 10% invaluable rate and lead-in patients, a total of 43 patients will be enrolled on the study. Clinical trial
information UMIN000014107.

Title: Phase I, open-label study evaluating the safety and tolerability of LJM716, BYL719 and trastuzumab in patients with
metastatic HER2+ breast cancer
Payal D Shah1, Sarat Chandarlapaty1, Gary Ulaner1, Stephen Zamora1, Valentina Sterlin1, Alexia Iasonos1, Maura N Dickler1,
Mary Ellen Moynahan1, Clifford A Hudis1, Jose Baselga1 and Shanu Modi1. 1Memorial Sloan Kettering Cancer Center, New York,
NY.
Body: Background: Breast cancers that overexpress the human epidermal growth factor receptor 2 (HER2+) are driven by
HER2/HER3/PI3K signaling. Single agent therapy against this pathway is only modestly effective due to redundant mechanisms
of pathway activation and regulatory feedback loops. Drug combinations designed to directly inhibit multiple nodes simultaneously
have proven synergistic in preclinical models. Single agents and doublets targeting HER2, HER3, and PI3K have thus far proven
safe in clinical testing. Methods: We are conducting a phase I clinical trial of LJM716, a fully human monoclonal antibody against
HER3, plus BYL719, a small molecule alpha-specific PI3K inhibitor, and trastuzumab, an approved human monoclonal antibody
against HER2, in patients (pts) with HER2+ metastatic breast cancer (MBC). There will be dose-escalation and dose-expansion
phases, with correlative serum and tumor tissue studies. The primary objective is to determine the MTD of BYL719 with fixed
dose LJM716 and trastuzumab in pts with HER2+, PIK3CA mutated MBC; determination of MTD will utilize the Continual
Reassessment Method (CRM). LJM716 will be administered at 20mg/kg and trastuzumab at 2mg/kg, both intravenously weekly;
starting dose of BYL719 will be 250mg orally daily, with dose escalation to MTD. Secondary objectives are to describe safety
(CTCAE 4.0) and preliminary efficacy (RECIST v1.1) of the combination and to describe immunogenicity of LJM716. Exploratory
objectives are to assess pre-treatment (tx) genomic alterations predictive of response or resistance to the tx; to assess proteomic
pharmacodynamics markers of effective HER2, HER3 and PI3K inhibition, and to correlate cell-free DNA mutant allele fraction
with clinical response. For these correlative studies, all pts will undergo serial plasma collection for cell-free DNA, and 10 pts will
undergo pre-tx, on-tx, and at-progression biopsies. Eligible pts for dose-escalation and expansion are required to have ECOG
0-1, HER2+ MBC (immunohistochemistry 3+ or in-situ hybridization 2.0), and prior treatment with pertuzumab and
ado-trastuzumab emtansine. For dose-escalation, pts may have measurable or non-measurable disease and somatic PIK3CA
mutations are required. For dose-expansion, pts must have measurable disease and there will be two cohorts: one for wild-type
and the other for mutant PIK3CA tumors. Accrual beginning June 2014 will include an anticipated 18 pts for dose-escalation and
30 pts for the expansion. Contact modis@mskcc.org for further information.

Title: A phase 2 single-arm study to evaluate the clinical activity, safety and tolerability of enzalutamide (ENZA) with trastuzumab
in patients with advanced human growth factor receptor 2 (HER2)-positive breast cancer
Maureen E Trudeau1, Eric Winer2, Joyce L Steinberg3, Maggie Liosatos3, Srinivasu Poondru3, Michael Dydo3, Yun Zhou3, Michael
Westphal3, Amy C Peterson4 and William Gradishar5. 1Sunnybrook Health Sciences Center, Toronto, Canada; 2Dana-Farber
Cancer Institute, Boston, MA; 3Astellas Inc, Northbrook, IL; 4Medivation Inc, San Francisco, CA and 5Northwestern University,
Chicago, IL.
Body: Background
Androgen receptor (AR) expression is observed in 60% of patients (pts) with HER2+ breast cancer1. In vitro, ENZA inhibits
proliferation of AR+/HER2+ cell lines and enhances the activity of trastuzumab. ENZA also inhibits proliferation of trastuzumab
resistant HER2 positive cells2.
Methods
Any amount of AR expression (local or central) is allowed, submission of tissue is mandatory. Patients must have measurable
disease per RECIST 1.1, and have received 1 to 4 prior lines of anti-HER2 therapy in the advanced / metastatic setting. Brain
imaging is required to exclude patients with CNS metastases. Pts with a seizure history are excluded.
Women with metastatic or locally advanced HER2+, AR+, and ER-/PgR- breast cancer will receive daily ENZA (160 mg)
continuously and trastuzumab (6 mg/kg) administered every 21 days, until disease progression (NCT02091960). The primary
endpoint (EP) is clinical benefit rate (CBR) where benefit is defined as complete or partial response (CR or PR) or stable disease
(SD) 24 weeks according to RECIST 1.1 criteria. Additional EPs include safety and tolerability, and the relationship between AR
expression and ENZA activity. If the CBR is 3 in 21 evaluable pts, the sample size will increase to 66 pts.
The primary EP will be analyzed in pts with centrally confirmed AR expression (10% nuclear staining by IHC), who have
received at least one dose of ENZA, and have 1 post-baseline tumor assessment. The null hypothesis (H0), that the true CBR is
10%, will be tested against a 1-sided alternative. This Simons two-stage design yields 90% power when the true response rate is
25% with a 1-sided type 1 error rate of 5%. Enrollment is expected to continue through 2016.
References:
1. Collins LC, Cole KS, Marotti JD, Hu R, Schnitt SJ, Tamimi RM. Androgen receptor expression in breast cancer in relation to
molecular phenotype: results from the Nurses' Health Study. Mod Pathol. 2011;24(7):924-31.
2. J. Richer, AACR Advances in Breast Cancer, San Diego, CA, October 2013.

Title: MotHER: A US registry for women with breast cancer who have received trastuzumab, pertuzumab in combination with
trastuzumab, or trastuzumab emtansine (T-DM1) during pregnancy or within 7 months prior to conception
Vikki Brown1, Richard K Miller2, Laura Chu3, Cheryl Chow3, Caroline Trudeau3 and Elizabeth B Andrews4. 1INC Research,
Raleigh, NC; 2University of Rochester School of Medicine and Dentistry, Rochester, NY; 3Genentech, San Francisco, CA and
4
RTI International, Research Triangle Park, NC.
Body: Background: Trastuzumab (Herceptin) and pertuzumab (Perjeta) are HER2-targeted monoclonal antibodies, and T-DM1
(Kadcyla) is a HER2-targeted antibodydrug conjugate. These molecules are classified as FDA Pregnancy Category D, indicating
evidence of fetal harm; there have been postmarketing reports of oligohydramnios in women who received trastuzumab during
pregnancy; in non-clinical studies of pertuzumab, oligohydramnios, delayed renal development, and embryo-fetal death occurred
in pregnant cynomolgus monkeys. Furthermore, maytansine, the parent molecule of DM1, has demonstrated embryotoxicity in
mice. The MotHER pregnancy registry was established in 2008 as an FDA postmarketing commitment to evaluate the effects of
trastuzumab on pregnancy outcome. Pertuzumab and T-DM1 were included in the registry following their respective FDA
approvals in 2012 and 2013.
Study design: MotHER is a US pregnancy registry and uses a prospective, observational cohort design. Enrollment is voluntary
and can be initiated by the patient or her healthcare provider. Women are followed up until pregnancy outcome; infants are
followed up through the first year of life.
Eligibility criteria: Pregnant women 18 years of age and resident in the US are eligible if they have received at least one dose
of trastuzumab, pertuzumab in combination with trastuzumab, or T-DM1 for breast cancer during pregnancy or within 7 months
prior to conception. Enrollment must be initiated before pregnancy outcome is known.
Specific aims: The primary outcome measures are pregnancy outcomes and the incidence of pregnancy complications, such as
oligohydramnios, congenital anomalies, and fetal/infant functional deficits. Information on potential birth defects, noted either at
birth or during pediatric follow-up, is evaluated and classified by a birth defect evaluator/clinical geneticist.
Statistical methods: The study is descriptive and not designed for formal hypothesis testing; event rates with 95% confidence
intervals will be calculated.
Accrual: There is no pre-specified sample size. Data collection is ongoing, from 2009 to 2019 for trastuzumab, from 2012 to 2022
for pertuzumab plus trastuzumab, and from 2013 to 2023 for T-DM1. Fourteen women have enrolled as of January 31, 2014.

Title: A phase II, randomized study of T-DM1 versus T-DM1 plus short induction with docetaxel in first line treatment for locally
advanced or metastatic HER2+ breast cancer (SOLTI-1203)
Antonio Llombart1, Javier Corts2, Eva Ciruelos3, Xavier Gonzlez4, Lorena de la Pea5, Patricia Villagrasa5, Elena Bernedo6,
Katya Moreno6, Susanne Meya6 and Jose Baselga7. 1Hospital Universitari Arnau de Vilanova de Valencia, Valencia, Spain;
2
Hospital Universitari Vall d'Hebron, Barcelona, Spain; 3Hospital Universitario 12 de Octubre, Madrid, Spain; 4Hospital Universitari
Quirn Dexeus, Barcelona, Spain; 5SOLTI Breast Cancer Research Group, Barcelona, Spain; 6Hoffmann-La Roche, Ltd, Basel,
Switzerland and 7Memorial Sloan Kettering Cancer Center, New York, NY.
Body: Background: This study is founded upon the hypothesis that the addition of a short induction phase with docetaxel to the
treatment with T-DM1 can improve efficacy with respect to T-DM1 administered as a single agent, in patients with advanced
HER2 positive breast cancer who have not received previous treatment for the advanced disease. The rational for this approach
is based on the results of the phase II randomized study of T-DM1 versus docetaxel trastuzumab, in which the administration of
T-DM1 to the patients resulted in a statistically significant and clinically relevant improved progression free survival (PFS). One
important observation with respect to this comparative study is that, in the T-DM1 arm, a higher number of patients progressed
during the first six months than in the docetaxel trastuzumab arm. The existence of T-DM1 resistant patients, who progress during
the first 6 months, highlights the need to explore new combinations with cytotoxic agents to induce an early response.
Study design: This is a phase II, prospective, randomized, open label, research study with two groups designed to assess the
efficacy of T-DM1 together with short induction with docetaxel (SI docetaxel) versus T-DM1monotherapy treatment, in patients
recently diagnosed with locally advanced or metastatic, progressive or recurrent, HER2+ breast cancer, who have not received
previous chemotherapy for the advanced disease. The randomization will be carried out at a 1:1 ratio to one of the following two
groups: 1.Monotherapy treatment regimen with T-DM1 at a dose of 3.6 mg/kg every 3 weeks until treatment finalization, 2.T-DM1
at a dose of 3.6 mg/kg on Day 1 and every 3 weeks until treatment finalization, together with docetaxel 60 mg/mg on Day 1 and
every 3 weeks for a total of 4 treatment cycles.
Eligibility criteria: Female adults recently diagnosed with progressive or recurrent, locally advanced or metastatic HER2+ breast
cancer, who have not received previous chemotherapy for the advanced disease.
Endpoints: The primary endpoint is to compare the early efficacy of the combination of T-DM1 together with SI docetaxel versus
monotherapy treatment with T-DM1 measured as the PFS rate at 4 months. Secondary endpoints include comparison of the
ORR, CBR, DR and OS of the combination versus monotherapy treatment Exploratory endpoints consist to assess if the PAM50
intrinsic subtypes (HER2-enriched in comparison with the rest) predict benefit of T-DM1 plus SI docetaxel or T-DM1 as
monotherapy treatment, in terms of ORR and PFS.
Statistical methods: The considerations for the sample size are based on obtaining a statistical power of 80 % to be able to
detect a difference in the rate of PFS at 4 months of 8% (83% in group A and 91% in group B), with a bilateral significance level of
5% and supposing a patient withdrawal rate of 10%. According to these calculations, the sample size is estimated at 236 patients.
The analysis of the primary endpoint will be based on the Kaplan Meier estimate analysis.
Target accrual: Approximately 40 European sites from Spain, Portugal and Austria will be participating. The FPI is planned in
October 2014.

Title: Combination immunotherapy with trastuzumab and the HER2 vaccine E75 (nelipepimut-S) in high-risk HER2+ breast
cancer patients to prevent recurrence
Beth A Mittendorf2, Erika J Schneble1, Nuhad K Ibrahim2, Julia M Greene1, John S Berry1, Alfred F Trappey1, Guy T Clifton1,
Jarrod P Holmes4, Sathibalan Ponniah3 and George E Peoples1. 1San Antonio Military Medical Center, San Antonio, TX; 2MD
Anderson Cancer Center, Houston, TX; 3Cancer Vaccine Development Lab, Uniformed Services of Health Sciences, Bethesda,
MD and 4Redwood Regional Medical Group, Santa Rosa, CA.
Body: Background: In an adjuvant phase II trial, the HER2-derived nelipepimut-S (E75) + GM-CSF vaccine (Neuvax) has been
shown to reduce breast cancer recurrence. Preclinical testing of the combination of trastuzumab (Tz) and nelipepimut-S has
shown synergistic cytolysis against HER2 expressing cancer cells. In pilot phase II data in HER2+ patients (pts), 55 pts dosed
with CD8-eliciting HER2 derived peptide vaccines sequentially after treatment with Tz resulted in no recurrences at 36 months
median follow-up compared with a 16% recurrence rate in 34 randomized controls treated with Tz without vaccine (p=.012).
Based on these data, we have designed a trial to evaluate the ability of the combination of Tz and the E75 vaccine concurrently to
prevent recurrence in pts with high-risk, HER2+ breast cancer.
Trial Design: This study will be a multicenter, prospective, randomized, single-blinded, phase II trial evaluating adjuvant Tz +
NeuVax (E75+GM-CSF) vs. Tz + GM-CSF alone in high-risk HER2+ (IHC 3+ and/or FISH >2.2) breast cancer pts. High-risk pts
include: 1) those that did not achieve a pathologic complete response (pCR) after neoadjuvant chemotherapy and HER2targeted therapy or 2) those treated with upfront surgery that are node positive (> 4+ LN or 1-3+ LN if hormone receptor
negative). Pts must be HLA-A2 /A3+ to be eligible (E75 is HLA-A2/A3-restricted) with ECOG performance status 0-1. Pts will be
enrolled after completing standard of care multi-modal therapy but prior to the 3rd dose of Tz maintenance therapy (monotherapy).
Pts will be randomized 1:1 to receive either NeuVax or GM-CSF alone which will be administered as six monthly intradermal
inoculations concurrently with Tz therapy. Pts will then receive four booster inoculations of either NeuVax or GM-CSF every 6
months. The primary efficacy endpoint is to compare disease-free survival (DFS) between treatment arms. Secondary objectives
will include evaluation of local and systemic toxicity, distant recurrence free survival, and in vivo/in vitro immunologic responses.
From previously published experience with Tz, we expect a recurrence rate of 20% in Tz (plus GM-CSF) treated pts and
anticipate that the combination of Tz with E75+GM-CSF will reduce this recurrence rate to 5%. In order to show statistical
difference between these recurrence rates, we plan to enroll 50 pts per treatment arm (100 total) with a type I error rate of 5% and
80% power to detect the primary endpoint. Trial accrual is anticipated to begin in September of 2014, with a two year period for
trial enrollment followed by a three year follow-up period.
Conclusion: We hypothesize that combination adjuvant immunotherapy with Tz and NeuVax will result in a greater reduction in
breast cancer recurrence than Tz therapy alone. We have designed a prospective, randomized, single-blinded, phase II trial
evaluating the efficacy of this immunotherapy combination in high-risk HER2+ breast cancer pts to test this hypothesis.
Contact Information: This trial is funded by a DoD grant to EAM with matching funds from Galena Biopharma and is being
conducted with the assistance of the academic CRO, Cancer InCITe, LLC.

Title: HERmione: A Phase 2, randomized, open label trial comparing MM-302 plus trastuzumab with physicians choice of
chemotherapy plus trastuzumab, in anthracycline naive HER2-positive, locally advanced/metastatic breast cancer, previously
treated with pertuzumab and trastuzumab emtansine (T-DM1)
Kathy Miller1, Javier Cortes2, Sara A Hurvitz3, Ian E Krop4, Debu Tripathy5, Sunil Verma6, Kaveh Riahi7 and Denise A Yardley8.
1
Indiana University Melvin and Bren Simon Cancer Center, Indianapolis, IN; 2Vall d'Hebron Institute of Oncology, Barcelona,
Spain; 3University of California, Los Angeles, CA; 4Dana-Farber Cancer Institute, Boston, MA; 5MD Anderson Cancer Center,
Houston, TX; 6Sunnybrook Odette Cancer Centre, Toronto, ON, Canada; 7Merrimack Pharmaceuticals, Inc, Cambridge, MA and
8
Sarah Cannon Research Institute and Tennessee Oncology, PLLC, Nashville, TN.
Body: Background: Despite improvements in treatment with newly approved HER2-targeted therapies such as pertuzumab and
trastuzumab emtansine (T-DM1), HER2-positive metastatic breast cancer (MBC) remains a serious and life-threatening disease.
MM-302 is an antibody drug conjugated liposomal doxorubicin targeted to HER2. The present Phase 2 study was developed
based upon data from a recently reported Phase 1 study (SABCS 2013, Munster, et al., #P4-12-29) in 47 heavily pretreated
HER2-positive MBC patients treated with MM-302 alone or in combination with trastuzumab, with more benefit being seen in
patients without prior anthracycline exposure. In that study, MM-302 appeared to have an acceptable safety profile and,
importantly, no clinically significant declines in left ventricular ejection fraction (LVEF) were observed.
Trial design: HERmione is a randomized Phase 2, two-arm, open-label trial designed to evaluate if MM-302 can address an
unmet medical need in treating anthracycline naive patients with HER2-positive metastatic breast cancer who have previously
received pertuzumab and T-DM1 in the locally advanced or metastatic setting. Patients are randomly assigned 1:1 to MM-302 (30
mg/m2, IV q3w) plus trastuzumab (6 mg/kg, IV q3w) or a chemotherapy of physicians choice (chosen from vinorelbine,
capecitabine, or gemcitabine) plus trastuzumab (6 mg/kg, IV q3w). Patients will be stratified according to geographic region,
presence of visceral disease, and 4 prior chemotherapy-containing regimens for metastatic disease.
Eligibility criteria: HER2-positive locally advanced or MBC (HER2 confirmed by central lab); prior trastuzumab in any setting, prior
pertuzumab and T-DM1 exposure in the incurable locally advanced or metastatic setting; unlimited prior lines of therapy; ECOG
0-1, and LVEF 50%. Patients with prior anthracycline exposure in any setting are excluded. Central nervous system metastases
are permitted, if stable and without symptoms or steroids for 4 weeks after treatment. Patients with significant cardiac dysfunction
or risk of cardiac dysfunction are excluded.
Specific aims: The primary efficacy endpoint is independently assessed progression free survival (PFS) with tumor response
evaluated based on RECIST version 1.1. Secondary endpoints include Investigator Assessed PFS, Overall Survival (OS), 6 and
12 month OS rate, Time to Treatment Failure, Objective Response Rate, Duration of Response, Clinical Benefit Rate, safety and
toxicity, Patient Related Outcomes and pharmacokinetic exposure. Safety assessments include all adverse events, abnormal
laboratory values, cardiac events, and LVEF decline.
Statistical methods: A total sample size of 250 subjects will be enrolled to observe 191 PFS events for 90% power to detect a
Hazard Ratio of 0.625. The MM-302 arm will be compared to the control arm on the primary endpoint of PFS using a stratified
log-rank test at a one-sided 0.025 level.
Accrual status: Recruitment is planned to start in July 2014 at approximately 60 sites in the United States and Western Europe.

Title: IMPACT: IMaging PAtients for Cancer drug selecTion Metastatic breast cancer (MBC)
Frederike Bensch1, Adrienne Brouwers1, Andor Glaudemans1, Johan de Jong1, Erik de Vries1, Winette van de Graaf2, Eline Boon2,
Wim Oyen2, Lioe-Fee de Geus-Oei2, Eric Visser2, Erik van Helden3, Willemien Menke-van der Hoeven van Oordt3, Henk Verheul3,
Otto Hoekstra3, Jim Janssen3, Marc Huisman3, Sjoerd Elias4, Carl Moons4, Liesbeth de Vries1 and Carolien Schrder1. 1University
Medical Center Groninge, Netherlands; 2Radboud University Medical Center Nijmegen, Netherlands; 3VU Medical Center,
Amsterdam, Noord-Holland, Netherlands and 4University Medical Center, Netherlands.
Body: Background: Therapy of newly identified MBC is largely based on estrogen receptor (ER) and human epidermal growth
factor receptor 2 (HER2) status. Optimal receptor information should be up-to-date and preferably from the whole body, given
receptor conversion over time and intra-patient tumor heterogeneity. Novel molecular imaging by means of 18F-fluoroestradiol
(FES)- and 89Zr-trastuzumab-PET/CT is a non-invasive, patient friendly way to obtain such information. Comprehensive
prospective data comparing novel molecular imaging, metastasis biopsy and blood biomarkers, are needed to assess clinical
utility for optimal therapy guidance and response prediction.
Trial design: The IMPACT-MBC trial (NCT01957332), is a multicenter prospective cohort study, supported by the Dutch Cancer
Society-Alpe dHuZes, in which n=200 newly diagnosed MBC patients will be entered. Prior to start of treatment patients will
undergo i) standard MBC work up including bone scan, diagnostic CT and 18F-fluorodeoxyglucose(FDG)-PET/CT, ii) a metastasis
biopsy, for standard (immuno)pathology and DNA sequencing, iii) 18F-FES- and 89Zr-trastuzumab-PET/CT to assess whole-body
metastatic ER and HER2 status, and iv) blood sampling (CTCs, ctDNA, germline DNA, 89Zr-radioactivity measurements).
Treatment advice will be based on standard work up and experimental PET scans. Tumor response is assessed by a 2 week
18F-FDG-PET/CT (experimental) and an 8 week diagnostic CT (standard; primary outcome).
Eligibility criteria: All newly diagnosed non-rapidly progressive MBC patients with measurable or clinical evaluable (bone only)
disease can be enrolled, regardless of primary tumor ER and HER2 status. Patients should be eligible for systemic therapy, but
not require immediate start of chemotherapy. A histological biopsy of a metastatic lesion should be safely obtainable. Excluded
are pregnant or lactating women and patients with a prior allergic reaction to immunoglobulins.
Specific aims: i) To assess the (added) clinical utility of 18F-FES- and 89Zr-trastuzumab-PET/CT, in the setting of MBC at first
presentation, in relation to other diagnostics, ii) to assess the relation of experimental 18F-FES-, 89Zr-trastuzumab- and 2 week
18F-FDG-PET/CT with (progression free) survival and iii) to assess the cost-effectiveness of the experimental PET/CT scans.
Statistical methods: IMPACT-MBC aims to model the predictive value of several tests (novel molecular imaging, biopsy and
blood biomarkers) in combination, by means of multivariable regression-model based techniques, combined with state-of-the-art
methods for estimating the added value of novel tests to existing information (e.g. NRI, IDI). All these analyses will be employed
both on (predicting responsiveness on) a patient- and metastasis level.
Present accrual and target accrual: The IMPACT-MBC trial was opened for accrual at the University Medical Center (UMC)
Groningen, in August 2013. Accrual rate is as anticipated 2-3 patients/month/center. The two other participating centers, Radboud
MC Nijmegen and VUmc Amsterdam recently opened. It is anticipated that accrual of patients will be finalized in 2016.

Title: MINT I: Multi-Institutional Neo-adjuvant Therapy, MammaPrint Project I
Charles E Cox1, Peter Blumencranz2, Ruben Saez3, Robert Wesolowski4, William Dooley5, Lisette Stork-Sloots6, Femke de Snoo6,
Sarah Untch6 and Eli Avisar7. 1University of South Florida, Tampa; 2Morton Plant Hospital, Clearwater; 3Plano Cancer Institute,
Plano; 4Ohio State University; 5University of Oklahoma; 6Agendia Inc, Irvine, CA and 7Miller School of Medicine, University of
Miami, Miami, FL.
Body: Background:
Women with locally advanced breast cancer (LABC) are often treated with neo-adjuvant chemotherapy to reduce the size of the
tumor prior to surgery, to enable breast conserving surgery and to observe the clinical effect of therapy in real time. Studies have
shown that the 2527% of individuals who have a pathologic complete response (pCR) to neoadjuvant therapy have a survival
advantage of 80% in 5 years, which is double the expected survival of the remaining patients without pCR. If patients who are
likely to show a pCR could be identified prior to initiation of therapy, it would enable more informed treatment decisions patients
likely to respond would be served well by current neoadjuvant chemotherapy protocols, while those unlikely to respond may be
better suited to innovative new strategies for drug discovery [von Minckwitz et al. JCO 2006]. Genomic assays, which are widely
used to provide prognostic and predictive information in early breast cancer, have the potential to provide information on the
likelihood of a patient with LABC responding to neo-adjuvant therapy [Glck et al. BRCRT2013].
Trial design:
MINT I is a prospective study designed to test the ability of molecular profiling, as well as traditional pathologic and clinical
prognostic factors, to predict response to neo-adjuvant chemotherapy in patients with LABC. MammaPrint risk profile, BluePrint
molecular subtyping profile, TargetPrint estrogen receptor (ER), progesterone receptor (PR) and HER2 single gene readout, and
TheraPrint Research Gene Panel will be analyzed on a fresh tumor specimen using the whole genome array. Patients will receive
neo-adjuvant chemotherapy pre-specified in the protocol. Response will be measured centrally. pCR is defined as the absence of
invasive carcinoma in both the breast and axilla at microscopic examination of the resection specimen, regardless of the
presence of carcinoma in situ.
Eligibility:
The study will include women 18 years with histologically-proven invasive breast cancer T2 (3.5cm)-T4, N0M0 or T2-T4N1M0,
adequate bone marrow reserves and normal renal and hepatic function who signed an IRB approved informed consent.
Objectives:
The objectives of the study are to:
1. Determine the predictive power of MammaPrint and BluePrint for sensitivity to neo-adjuvant chemotherapy as measured by
pCR.
2. Compare TargetPrint ER, PR and HER2 with local and centralized IHC and/or CISH/FISH assessment.
3. Identify correlations between TheraPrint and response to neo-adjuvant chemotherapy.
4. Identify and/or validate predictive gene expression profiles of clinical response or resistance to neo-adjuvant chemotherapy.
5. Compare BluePrint with IHC-based subtype classification.
Standard statistical tests such as the Pearson Chi-square test will be used to characterize and evaluate the relationship between
chemoresponsiveness and gene expression patterns.
Accrual:
A total of 226 eligible patients will be enrolled from multiple institutions. To date (June 06, 2014), 103 patients have been enrolled.

Title: Prospective observational study of clinical outcomes for the NanoString technologies Prosigna test by the West German
Study Group
Rachel Wuerstlein1,2, Karl Sotlar3, Burkhard Otremba4, Oleg Gluz1,5, Enrico Pelz6, Daniel Hofmann1, Ronald E Kates1, Isabelle
Witzel7, Christian Schindlbeck8, Wolfgang Janni9, Christian Schem10, Hans Tesch11 and Nadia Harbeck1,2. 1West German Study
Group, Moenchengladbach, Germany; 2University Hospital Munich, Breast Center, CCC of LMU, Munich, Germany; 3University
Clinics Munich (LMU), Institute of Pathology, Munich, Germany; 4Oncologic Practice Oldenburg/Delmenhorst, Oldenburg,
Germany; 5Ev. Bethesda Hospital, Breast Center Niederrhein, Moenchengladbach, Germany; 6Pathology Viersen, Viersen,
Germany; 7University Medical Center Hamburg-Eppendorf, Hamburg, Germany; 8Clinics Traunstein, Women Clinics, Traunstein,
Germany; 9University Clinics Ulm, Women Clinics, Ulm, Germany; 10University Clinics Kiel, Women Clinics, Kiel, Germany and
11
Oncologic Practice, Frankfurt, Germany.
Body: Background: Current diagnostic tools for breast cancer (BC), including screening, diagnostic breast imaging and biopsy,
do not provide information about subtype classification. However, the subtype classification is crucial for the assessment of the
recurrence risk and for the choice of optimal adjuvant treatment. A vast amount of data from recent research, clinical trials, and
peer reviewed publications support the value of intrinsic subtyping based on gene expression analyses to assess prognosis and
treatment options for patients with early BC.
Trial design: The multicenter (n=8), prospective study examines whether the Prosigna test influences physician and patient
adjuvant treatment selection over and above currently used prognostic factors. It also examines the impact of the test results on
patients reported outcomes including their decisional conflict status, anxiety levels, and functional status.
Prosigna is a standardized test measuring the expression levels of 50 classifier genes in formalin-fixed, paraffin-embedded
tissue samples and provides a subtype classification based on the fundamental biology of individual patients tumor (intrinsic
subtyping), as well as a prognostic score (risk of recurrence (ROR) score) that predicts the probability of cancer recurrence over
10 years.
Eligibility criteria: Postmenopausal women with resected N0, estrogen-receptor-positive (ER) (by immunohistochemistry (IHC);
>1% of stained tumor cells is considered positive), HER2-negative (IHC and/or in-situ fluorescence hybridization by local
laboratory), early-stage invasive BC (T1-T2, pN0 (i+/mol+), M0). Patients must be able to give consent and must be eligible for
treatment with adjuvant chemotherapy (Ctx) (ECOG performance status 1).
Patients with non-invasive (e.g. Pagets disease, DCIS), ER-negative or HER2-positive BC must not be included. Metastatic
disease and contraindications for adjuvant Ctx (age, ECOG >1, significant comorbidities) are exclusion criteria. Patients unable to
complete patient reported outcome surveys will be excluded.
Specific aims: Primary endpoint is the proportion of patients whose choice of treatment is changed as a result of receiving the
Prosigna result.
Secondary endpoints include the proportion of patients whose choice of treatment is changed stratified by ROR groups (low,
intermediate, high) and who have cancer recurrence risk proximal to the cutoff points between risk strata ( 5% from cutoff point).
Investigators confidence in treatment recommendations before/after Prosigna results and change in patients decisional conflict
status, anxiety levels, and functional status, stratified by cancer recurrence risk strata, will be investigated. Rate and severity of
treatment-related adverse events stratified by whether patient received adjuvant Ctx and agreement in molecular subtyping
between Prosigna and local IHC will be evaluated.
Statistical methods: Sample size of 200 produces a one-sided 95% lower-limit CI with a distance from the sample proportion to
the lower limit that is equal to 0.05 when the sample proportion is 0.25.
Present and target accrual: A total of 200 patients will be included, current accrual is n=111 as of 3rd June 2014.

Title: ADAPT - Adjuvant Dynamic marker-Adjusted Personalized Therapy trial optimizing risk assessment and therapy response
prediction in early breast cancer
Ulrike Nitz1,2, Oleg Gluz1,2, Raquel von Schumann1,2, Daniel Hofmann1, Ronald E Kates1, Sherko Kuemmel3, Michael Braun4,
Claudia Schumacher5, Benno Nuding6, Bahriye Aktas7, Helmut Forstbauer8, Nicolai Maass9, Mahdi Rezai10, Stefan Kraemer11,
Mathias Warm12, Rachel Wuerstlein1,13 and Nadia Harbeck1,13. 1West German Study Group, Moenchengladbach, Germany; 2Ev.
Bethesda Hospital, Breast Center Niederrhein, Moenchengladbach, Germany; 3Clinics Essen-Mitte, Clinics for Senology/Breast
Center, Essen, Germany; 4Rotkreuz Clinics Munich, Women Clinics, Munich, Germany; 5St Elisabeth Hospital, Cologne, Cologne,
Germany; 6Ev. Hospital Bergisch Gladbach, Berg. Gladbach, Germany; 7University Clincs Essen, Women Clinics, Essen,
Germany; 8Practice Network Troisdorf, Troisdorf, Germany; 9University Clinics Aachen, Women Clinics, Aachen, Germany;
10
LuisenHospital Duesseldorf, Practice for Senologic Oncology, Duesseldorf, Germany; 11University Clinics Cologne, Breast
Center, Cologne, Germany; 12Clinics Cologne Holweide Hospital, Breast Center, Cologne, Germany and 13University Hospital
Munich, Breast Center, CCC of LMU, Munich, Germany.
Body: Background: Early therapy response is currently not regarded for further treatment decisions as standard of care in the
treatment of breast cancer (BC). Predictive markers for the success of a certain therapy could support the physicians choice of
adequate and beneficial therapies by simultaneous reduction of unnecessary toxicity. Proliferation makers as Ki-67 seem to be a
suitable tool, as dynamic changes of proliferation (as result of induction therapy) have been shown to be most important for
outcome of neoadjuvant chemotherapy prediction in patients with pCR in distinct BC subtypes (luminal B, TNBC, HER2+).
Methods: Trial design: ADAPT combines early assessment of prognosis by conventional markers (e.g. molecular classification,
nodal status) with dynamic measurement of proliferation changes during a 3-week induction therapy, using baseline diagnostic
core biopsy and a second biopsy after induction therapy. ADAPT consists of an umbrella trial and five different sub-trials
(HR+/HER2-, HR+/HER2+, HR-/HER2+, HR-/HER2-, Elderly) and is set up as prospective, multi-center, controlled, non-blinded,
randomized phase II/III trial.
Subtype-specific treatment across the sub-trials is highly innovative and involves the following treatment strategies:
HR+/HER2-: endocrine therapy (ET) vs. chemotherapy (4xPac q2w 4xEC q2w vs. 8xNab-Pac q1w 4xEC q2w) + ET,
depending on risk classification/early response.
HER2+/HR+: T-DM1 vs. T-DM1 + ET vs. trastuzumab + ET.
HER2+/HR-: Trastuzumab + Pertzumab Paclitaxel q1w.
TN: Nab-Paclitaxel + Gemcitabine vs. nab-Pac + Carboplatin.
Elderly: 2xMyocet + Cyclophosphamide q3w, depending on cPR/cCR or NC/toxicity the treatment will be continued for two more
cycles or changed to 6xPac q1w.
Adaptation/change in therapy regimens can be made by interim analysis after n=130 in each sub-trial.
Eligibility criteria: Histologically confirmed unilateral primary invasive BC with known HR-/HER2-status (central pathology) for
allocation to the respective sub-trial. Pts requiring chemo- or targeted (anti-HER2) therapy must have adequate laboratory values
and organ function and must have no contraindications for the planned treatment.
Primary endpoints: Evaluation of dynamic test for outcome prediction/prospective comparison of 5yr EFS in responders
(intermediate risk (RS 12-25) / good response to short-term ET in HR+/HER2- or pts with pCR in HER2+/TN BC) compared to low
risk HR+/HER2- (RS11, N0-1) pts (control group).
Statistical methods: Assumption across sub-protocols: adjuvant CTx can be spared in HR+/HER2- or pCR be achieved in
HER2+/TN in expected 1120 (HR+/HER2-) or 170 (HER2+/TN) pts, respectively. Outcome will be compared to the control group
(expected n=640 HR+/HER2- pts: low risk (by RS), i.e. no CTx). Assuming 94% 5yr survival in control group, one-sided test of
non-inferiority at 95% CI will have 80% power for survival non-inferiority margin of 3.2% (i.e. 90.8% survival).
Present and target accrual: By June 2014, 73 active sites have recruited 1820 pts for ADAPT HR+/HER2-. Target accrual is 4000
pts. 190 of 380 pts were successfully randomized for ADAPT HER2+/HR+. ADAPT HER2+/HR- has included 17 of 220 pts and
ADAPT Triple Negative has recruited 150 of 336 pts.

Title: Improving outcomes in triple-negative breast cancer (TNBC) using molecular triaging and diagnostic imaging to guide
neoadjuvant therapy
Stacy Moulder-Thompson1, Wei Yang1, Naoto T Ueno1, Joe Ensor1, Vicente Valero1, Ricardo Alvarez1, Jennifer Litton1, Rashmi
Murthy1, Nuhad Ibrahim1, Banu Arun1, Beth Mittendorf1, Kelly Hunt1, Funda Meric-Bernstam1, Helen Piwnica-Worms1, Rosalind
Candelaria1, Debu Tripathy1 and Fraser Symmans1. 1University of Texas MD Anderson Cancer Center, Houston, TX.
Body: BACKGROUND:
Known disparities exist in patients with TNBC treated using neoadjuvant chemotherapy (NACT) with 30-50% having excellent
response to treatment (pCR/RCB-I) and good survival prognosis, while 50-70% demonstrate marked residual disease (RCB-II-III)
with significantly worse prognosis. Lack of response early into NACT also indicates a low chance (5%) of achieving pCR. Thus, it
is important to develop diagnostic platforms predictive of pCR, in order to direct patients with responsive disease toward standard
NACT and non-responsive disease toward experimental therapies within clinical trials.
TRIAL DESIGN
This study will determine the impact of predicting response to NACT using both molecular and imaging diagnostics and will
determine if offering a clinical trial of selected targeted therapy will impact outcomes (as measured by pCR and RCB) in predicted
non-responsive disease. An algorithm that incorporates pre-defined genomic signatures will determine predicted sensitivity to
chemotherapy (JAMA, 2011; 305:1873-81). All patients will undergo biopsy of the primary tumor for molecular analyses, but will
then be randomized 2:1 to know these results (Arm A) versus not (Arm B). All patients will begin anthracycline-based NACT with
diagnostic imaging to assess response after 4 cycles. Patients who fit molecular/imaging criteria for non-responsive disease will
be offered a clinical trial based upon molecular profiling (Arm A) or based upon physician/patient choice (Arm B). Patients who fit
criteria for responsive disease in either arm will continue with taxane based NACT.
ELIGIBILITY CRITERIA
INCLUSION: Candidate for biopsy of the primary tumor site; tumor size > 1.5 cm diameter; TNBC by standard pathologic assays;
>18 years of age; LVEF > 50%; adequate organ and bone marrow function
EXCLUSION: Stage IV disease; history of invasive cancer within 5 years; excisional biopsy of the primary tumor; biopsy site
changes that limit response assessment; medically unfit for chemotherapy; prior anthracycline; >grade II neuropathy; Zubrod
performance status of >2; history of serious cardiac event
PRIMARY AIM
Primary Aim: to prospectively determine the impact of implementation of a research platform that includes molecular (genomic)
testing from a primary tumor biopsy to predict response, and diagnostic imaging to assess response to standard NACT in patients
with localized invasive TNBC. Secondary Aims: compare rates of clinical trial enrollment between study arms, compare DFS,
integrated "prospective-retrospective" biomarker analysis, correlative science studies to identify therapeutic targets for resistant
disease
STATISTICAL METHODS AND TARGET ACCRUAL
Success will be defined as an improvement in the rate of excellent pathologic response (pCR/RCB-I) from 50%-->64% using the
triaging platform. A maximum of 360 patients will be randomized 2:1 to the experimental arm vs. the control arm using a group
sequential design with one-sided OBrien-Fleming boundaries, with up to two equally spaced binding interim tests for both futility
and superiority and one final test, having an overall Type I error .05 and power .80 to detect a response rate improvement from a
null rate of .50 to a target value of .642.

Title: NEOPAL: A randomized phase II study comparing RCB response to neoadjuvant chemotherapy or letrozole-palbociclib in
PAM50 defined postmenopausal luminal breast cancer
Paul H Cottu1, Franois Duhoux1, Jrme Lemonnier2, Herv Bonnefoi3, Anne Vincent Salomon1, Bernard Asselain1 and Suzette
Delaloge4. 1Institut Curie, Paris, France; 2UNICANCER, Paris, French Polynesia; 3Institut Bergonie, Bordeaux, France and 4Institut
Gustave Roussy, Villejuif, France.
Body: Background: HR-positive BC heterogeneously sensitive to neoadjuvant chemotherapy (NEC), with pCR rates varying
from 2 to 20% at most. The alternative systemic primary therapy in ER+ breast cancer is neoadjuvant endocrine therapy (NET).
Clinical response with NET is roughly 50%. It has also been suggested that NET may yield a higher rate of breast conserving
surgery than NEC in ER+ disease, although pCR remains under 10% in both instances. No definite pathological criteria for
response to NET has been defined so far. Within this context, the PAM 50 signature has been evaluated on numerous series of
pts and has been shown to highly correlate with Residual Cancer Burden (RCB Symmans, JCO 2007) and pCR in patients
treated with NEC, as well as with the PEPI score in patients treated with NET (Ellis 2011). Palbociclib + Letrozole has been
shown as a very active association as first-line metastastic treatment of HR+ Her2- BC pts on both response rates and
progression free survival (Paloma-1 study). This combination could represent an interesting alternative to CT in carefully selected
Luminal BC pts.
Trial design and statistical methods: Open-label, randomized, parallel, multicenter, exploratory phase II study, comparing
sequential standard NEC (3 FEC 100 followed by 3 Docetaxel 100) and a same duration letrozole + palbociclib combination as
neoadjuvant treatment of stage II-IIIA PAM 50 defined Luminal A-Node+/Luminal B breast cancer. Medical treatment will be
followed by adequate surgery and complementary chemotherapy and radiation therapy as clinically indicated. All pts will receive
adjuvant chemotherapy. Randomizations are equally balanced between the 2 arms and stratified based on T2 versus T3-T4 and
PAM 50 luminal A vs luminal B. A Fleming 2-step statistical design will be used in the experimental arm with an intermediate
futility analysis. The target accrual is 60 pts evaluable for RCB in both arms.
Eligibility criteria: Newly diagnosed pts with ER+, Her2-negative, stage II-III breast cancer will be tested for PAM50 signature.
Only luminal A N+ or Luminal B (PAM50 ROR (Prosigna) centralized evaluation) pts will be eligible and randomized. Pts must
be post-menopausal women, aged > 18 years, bear operable unilateral invasive BC, and not be candidate or uncertain for breast
conservation.
Specific aims: The main objective of this trial is to evaluate the ability of each treatment strategy to provide RCB 0-I at surgery.
Secondary end-points are: Clinical/radiological response rates in each treatment arm (RECIST 1.1), safety (CTC-AE V4.0),
relative dose intensity of each drug in both arms, positive and negative predictive values of PAM50 ROR-defined status,
assessment of several biomarkers as potential predictors of clinical and pathological response, rates of BCS, with regard to the
initially planned surgery, all in both arms
Target accrual: This study has started in autumn 2014. 132 pts will have to be included (estimation of 10% risk of non evaluable
pts). As about 10% of PAM50 evaluable pts will be classified as non luminal, and taking into account potential technical failures,
about 180 pts will be screened.

Title: Predicting hormonal therapy response in breast cancer using diffuse optical spectroscopic imaging (DOSI): Ongoing clinical
study
Thomas D OSullivan1, Anais Leproux1, Alice M Police1, Dorota Wisner2, Christine McLaren1, Wen-Pin Chen1, Albert E Cerussi1,
Min-Ying Su1 and Bruce J Tromberg1. 1University of California, Irvine, CA and 2University of California, San Francisco, CA.
Body: Background: The goal of this multi-site prospective study is to validate a safe, painless imaging method to measure the
change in breast density caused by hormonal chemotherapy treatments such as tamoxifen. Recent studies have demonstrated
that hormonal therapies are more effective at reducing risk in women who exhibit >10% reduction in breast density compared to
women who had little or no density change, suggesting that breast density is a predictor of tamoxifen effectiveness. Current
methods to measure breast density include MRI and mammography, however frequent applications of these modalities are
limited due to cost and x-ray exposure, respectively. Alternatively, we are testing an imaging method that uses safe near-infrared
light to measure breast tissue physiology called diffuse optical spectroscopic imaging (DOSI).
Trial Design and Eligibility: The primary aim of the study is to determine whether the percentage change in the DOSI
measurement of water correlates with the change in the MRI measurement of breast density after 18 months of treatment in the
contralateral normal breast of breast cancer patients receiving tamoxifen. Other DOSI-derived parameters such as lipid content
and hemoglobin concentration will be examined in secondary aims. Two groups of women are being recruited for the study:
Pre-menopausal subjects receiving tamoxifen (treatment group) and pre-menopausal subjects not receiving chemoprevention
agents (control group). Participants are measured with DOSI and non-contrast MRI before, and 6, 12 and 18 months after
beginning tamoxifen. Eligible subjects are pre- and peri-menopausal females older than 21 years of age who have not and do not
intend to receive chemotherapy, radiation, or surgical cancer treatment to the imaged breast, and are not pregnant or nursing.
Study sites include the University of California, Irvine and San Francisco campuses.
Statistical Methods: At a 5% significance level, the pre-determined power of the study is sufficient to detect the difference
between the treatment and control groups by measuring the percentage change of breast tissue water concentration with DOSI.
The research hypothesis is that the mean difference from baseline for tissue water concentration from DOSI will be greater for the
tamoxifen-treated group than the control group. For water concentration and for MRI breast density, a two-sided independent
sample t-test will be used to test the null hypothesis that the mean difference from baseline is the same for the tamoxifen-treated
and control groups. As a secondary analysis, a multivariate mixed effects model will be built using the observed DOSI parameter
or MRI breast density measured from each patient as the outcome variable, and predictor variables to include treatment group
and measurement time, in addition to relevant clinical and demographic variables.
Accrual Update: Out of the target accrual of 36, 11 subjects (6 treatment and 5 control group) have been enrolled to date.
Enrollment is open until 11/30/2015.

Title: PROMIS: PRospective study Of MammaPrint in breast cancer patients with an Intermediate recurrence Score (PROMIS)
Hatem Soliman1, Sarah Untch2 and Lisette Stork2. 1Moffitt Cancer Center, Tampa and 2Agendia Inc, Irvine, CA.
Body: Background:
Gene expression profiling in breast cancer offers the potential to improve prognostic accuracy, treatment choice, and health
outcomes in women diagnosed with early-stage breast cancer. Numerous gene-profiling assays are now available, which can be
applied to a single tumor specimen to provide physicians with a more complete basis for treatment decisions.
MammaPrint is 70-gene profile to estimate whether patients are at high or low risk of developing metastases within the first 10
years after curative surgery.
BluePrint is an 80-gene molecular subtyping profile that discriminates between three breast cancer subtypes: Luminal, HER2,
and Basal.
TargetPrint provides a quantitative measurement of estrogen receptor (ER), progesterone receptor (PR), and HER2.
Oncotype DX measures expression of five reference genes and 16 cancer-related genes, quantifying the risk of distant
recurrence in patients with ER+ early breast cancer who are treated with adjuvant hormonal therapy.
Trial design:
PROMIS is a prospective study that will investigate the additional value of MammaPrint, BluePrint and TargetPrint in women with
an intermediate Oncotype DX score. An initial CRF capturing baseline patient characteristics, pathology information, Oncotype
DX score and the recommended treatment plan will be completed before receiving the MammaPrint result. A second CRF
capturing the actual treatment will be completed within 4 weeks after receiving the MammaPrint result.
Eligibility: The study will include women aged 18 years with histologically proven invasive stage I-II, node negative or node
positive (N1), hormone receptor positive, HER2 negative breast cancer, who received an Oncotype DX intermediate score
(18-30) and who signed informed consent.
Objectives:
Primary objective:
Assess the impact of MammaPrint on chemotherapy + endocrine versus endocrine alone treatment decisions in lymph node
negative, hormone receptor positive, HER2 negative breast cancer patients, who received an Oncotype DX intermediate score
(18-30)
Secondary objectives:
Assess the impact of MammaPrint on chemotherapy + endocrine versus endocrine alone treatment decisions in lymph node
positive (N1), hormone receptor positive, HER2 negative breast cancer patients, who received an Oncotype DX intermediate
score (18-30)
Assess the distribution of MammaPrint Low and High Risk in patients with an intermediate recurrence score
Assess concordance of TargetPrint ER, PR and HER2 results with Oncotype DX ER, PR and Her2 and with locally assessed
IHC/FISH ER, PR and HER2
Compare clinical subtype based on IHC/FISH ER, PR, HER2 and Ki-67 (if available) with BluePrint molecular subtype
A sample size of 820 lymph node negative, hormone receptor positive, HER2 negative breast cancer patients is required to detect
a 20% overall treatment change (5% significance and 90% power). A McNemars test will be performed for the comparison of the
two proportions treated (before and after), both expressed as a percentage.
Accrual: A total of 385 out of 820 have been enrolled from multiple institutions.
Contact information: Clinicaltrials.US@agendia.com
Clinical trial registry number: NCT01617954.

Title: ASTER 70s UNICANCER phase III trial: Can a genomic prognosticator help tailoring adjuvant systemic treatment for
luminal breast carcinoma in elderly women?
Coraline Dubot1, Emmanuelle Bourbouloux2, Olivier Mir3, Sylvie Kirscher4, Olivier Rigal5, Jean-Marc Ferrero6, Herve Cure7,
Emmanuel Blot8, Djelila Allouache9, Paul Cottu10, Gilles Romieu11, Claudia Lefeuvre12, Emmanuelle Malaurie13, Nicole
Tubiana-Mathieu14, Magali Lacroix-Triki15, Florence Rollot10, Hlne Peyro-Saint-Paul16, Christine Orsini17, Franck Bonnetain18 and
Etienne Brain1. 1Institut Curie Site Saint Cloud, Saint-Cloud, France; 2Institut de Cancrologie de l'Ouest, Nantes, France;
3
Gustave Roussy Cancer Campus Grand Paris, Villejuif, France; 4Instituts de cancrologie d'Avignon, Avignon, France; 5Centre
Henri Becquerel, Rouen, France; 6Centre Antoine Lacassagne, Nice, France; 7Institut Jean Godinot, Reims, France; 8Centres
Hospitalier de Vannes, Vannes, France; 9Centre Franois Baclesse, Caen, France; 10Institut Curie Site Paris, Paris, France;
11
Institut rgional du Cancer Montpellier, Montpellier, France; 12Centre Eugne Marquis, Rennes, France; 13CHI de Crteil, Creteil,
France; 14CHU de Limoges, Limoges, France; 15Institut Universitaire du Cancer Toulouse Oncopole, Toulouse, France;
16
QIAGEN Marseille SA, Marseille, France; 17UNICANCER, Paris, France and 18CHU Besanon, Besancon, France.
Body: The benefit of adjuvant chemotherapy (CT) added to hormonal therapy (HT) compared with HT alone remains debated for
women >70 with ER+ HER2- breast cancer (BC). Selection of valid indications might be improved by the use of better
prognosticator. This trial compares the impact of both strategies on overall survival (OS) according to Genomic Grade (GG).
Following surgery, 2,000 patients (Pts) will have a GG performed centrally on FFPE specimens by RT-PCR. Those with a high
risk (high or equivocal GG) will be randomized to HT alone vs CT+HT. Pts with a low GG will be followed as an observational
cohort.
OS (all deaths) is the primary endpoint. Secondary objectives include competing events, cost-effectiveness and Q-TWiST
analysis, geriatric dimension, willingness and health-related quality of life including specific ELD15. Translational research will
focus on prognostic biomarkers and pharmacogenetics.
Statistical design: sample size based on 4-year OS benefit favouring CT (87.5 vs 80%; HR 0.60); bilateral test =0.05, =0.20;
129 events expected in 700 randomized Pts enrolled over 4 years.
From 04/12-05/14, 67 centres in France and Belgium have included 990 Pts aged 70-92.
Only 31 GG evaluations were not performed (tumour blocks not available, 14; consent withdrawal or central pathology review
discordance, 7 each; treatment choice, 3). In the main recruiting site, the study was not proposed to 20% of pre-screened Pts
mostly because of team choice (50%) and inclusion criteria (25%). Amongst those informed, 66% accepted to participate. Median
time to get GG information was 17 days (11-25) from sending tumour sample to providing the information to patient.
Of 932 cases with GG report, 374 (40%), 187 (20%) and 362 (39%) were low, equivocal and high GG respectively; 9 tests (1%)
failed for technical reasons. The proportion of high-risk tumours (high/equivocal GG 59%) is similar to that observed in general
BC populations (40% to 60%) and only 21 of high-risk cases were not randomized (consent withdrawal, 6; treatment choice, 5;
laboratory values, 4; tumour phenotype not confirmed or distant metastasis, 3 each).
With 75% of target recruitment in < 2 years, we confirm the feasibility of such innovative multicentre program in an usually
underserved population. This might help to better select adjuvant strategy in the elderly BC population and to avoid jeopardising
any benefit if stymied by uncontrolled side effects.

Title: A randomized controlled trial comparing primary tumour resection plus systemic therapy with systemic therapy alone in
metastatic breast cancer (JCOG1017 study; PRIM-BC)
Tadahiko Shien1, Hiroji Iwata2, Kenichi Nakamura3, Takayuki Kinoshita8, Fumikata Hara4, Tomomi Fujisawa5, Norikazu Masuda6,
Kenichi Inoue7, Taro Shibata3 and Haruhiko Fukuda3. 1Okayama University Hospital; 2Aichi Cancer Center Hospital, Nagoya,
Japan; 3JCOG Data Center; 4Shikoku Cancer Center; 5Gunma Prefectural Cancer Center, Ota, Gunma, Japan; 6Osaka National
Hospital; 7Saitama Cancer Center and 8National Cancer Center Hospital.
Body: A brief background discussion: The efficacy and indication of primary tumour resection for breast cancer patients with
distant metastases are under debate. There were many retrospective analysis reports indicating the survival benefit of it.
However, the first results of two randomized studies reported in SABCS 2013 could not demonstrate the survival benefit.
Nevertheless, the results were not conclusive because the systemic therapy was not uniform (e.g. molecular target therapy) and
the diagnostic procedures of metastases was different from the widely accepted guidelines.
Trial design: Our trial is being conducted to confirm the superiority, in terms of the overall survival, of surgery plus systemic
therapy over systemic therapy alone in stage IV patients who are not refractory to primary systemic therapy (PST).
Eligibility criteria: The inclusion criteria for the study are as follows: untreated patients with histologically confirmed invasive breast
cancer with one or more measurable distant metastatic lesions diagnosed by radiological examination.
Specific aims: All patients receive PST according to the ER and HER2 status of the primary breast cancer after the first
registration. After three months, the patients without disease progression are randomized to the primary tumour resection plus
systemic therapy arm or the systemic therapy alone arm. After randomization and surgery in the former arm, or after
randomization in the latter arm, the same systemic therapies are continued until progression of diseases and next appropriate
regimens are started after that.
Statistical methods: The primary endpoint is the overall survival, and the secondary endpoints are proportion of patients without
tumour progression at the metastatic sites, yearly local recurrence-free survival, proportion of local ulcer/local bleeding, yearly
primary tumour resection-free survival, adverse events of chemotherapy, operative morbidity, and serious adverse events.
Sample size for randomized patients was determined to attain at least 80% of power to detect a 6 months difference with
one-sided alpha of 0.05.
Present accrual and target accrual: The patient recruitment was started in May 2011. Enrolment of 410 patients for randomization
is planned over a 5-year recruitment period. More than two hundred twenty patients were already enrolled until May 2014.

Title: A prospective, randomized trial of sentinel lymph node biopsy versus no additional staging in patients with T1-T2 invasive
breast cancer and negative axillary ultrasound
Natalia S Tucker1, William E Gillanders1, Timothy Eberlein1, Rebecca Aft1, Julie Margenthaler1, Feng Gao1, Catherine Appleton1,
Imran Zoberi1, Ademuyiwa Foluso1 and Amy Cyr1. 1Washington University, School of Medicine, St Louis, MO.
Body: Background: The American College of Surgeons Oncology Group Z0011 prospective randomized trial demonstrated no
local control or survival advantage with more extensive axillary surgery, even in the setting of known axillary disease. These
results convincingly showed that axillary surgery provides little, if any, therapeutic benefit. Given that axillary surgery is not
associated with local control or survival benefit, the current role of sentinel lymph node (SLNB) is limited to staging the axilla (in
other words, SLNB provides staging information but is not therapeutic).
Objectives: In this randomized, controlled non-inferiority trial we aim to determine the utility of axillary ultrasound (AUS) as a
pre-operative staging modality for patients with clinically node-negative invasive breast cancer with the hope that it will be a
minimally invasive replacement for SLNB.
1. Primary Objective: To assess whether axillary recurrence rates for patients randomized to Arm 1 (no SLNB) is equivalent to
axillary recurrence rates for patients randomized to Arm 2 (SLNB).
2. Secondary Objective: To assess disease-free survival in Arm 1 vs. Arm 2.
3. Tertiary Objective: To assess overall survival in Arm 1 vs. Arm 2.
Study Design: A randomized non-inferiority trial comparing no further axillary staging versus SLNB in women with clinical T1-T2,
N0 M0 breast cancer and negative AUS. Following a negative AUS, four hundred sixty women will be randomized in a 1:1 fashion
to no further axillary staging or SLNB. Subjects will be monitored for local-regional and distant recurrence per NCCN guidelines.
We have recruited 28 study participants.
Eligibility Criteria: To be eligible the patient must be 18-y.o. or older; female; have tissue-diagnosis of invasive breast cancer;
cT1-2N0M0 cancer; negative AUS; ECOG </=2; available for follow-up; candidate for SLNB. A patient is ineligible if pregnant or
lactating; has concurrent invasive bilateral breast malignancies or multicentric disease or is considered a poor surgical candidate.
Power analysis: A power analysis was performed to determine the accrual goal. The power analysis is based on axillary
recurrence, our primary endpoint. We expect an axillary recurrence rate of 1% in patients undergoing SLNB. By assuming a
noninferiority limit of 2% difference, our sample size will allow us 80% power at 1-sided 0.1 significance level to assure such a
noninferiority.
Data analysis: Longer-term formal data analysis for the study will be performed following the intent-to-treat (ITT) principle.
Demographic and clinical characteristics of the sample, as well as efficacy, complication rates and loss to follow-up will be
summarized using descriptive statistics. The balance of demographic and baseline clinical characteristics between two arms will
be compared using t-test, Mann-Whitney rank-sum test, or Chi-square test as appropriate. The differences in OS and DFS
between treatment arms will be compared using log-rank test.

Title: The LORD trial: A randomized, non-inferiority trial, between active surveillance versus standard treatment in patients with
low risk ductal carcinoma in situ
Lotte E Elshof1, Konstantinos Tryfonidis2, Leen Slaets2, A Elise van Leeuwen-Stok3, Nicolas Dif2, Victoria P Skinner1, Claudette E
Loo1, Gonneke Warnars1, Eveline Bleiker1, Ruud M Pijnappel4, Nina Bijker5, Emiel JTh Rutgers1 and Jelle Wesseling1.
1
Netherlands Cancer Institute Antoni van Leeuwenhoek, Amsterdam, Noord-Holland, Netherlands; 2European Organisation
for Research and Treatment of Cancer, Brussels, Belgium; 3BOOG Study Center, Amsterdam, Noord-Holland, Netherlands;
4
University Medical Centre, Utrecht, Netherlands and 5Academic Medical Center, Amsterdam, Noord-Holland, Netherlands.
Body: Background
The goal in treating patients with ductal carcinoma in situ (DCIS) is to prevent the development of invasive breast cancer (iBC).
However, a substantial number of DCIS lesions will never form a health hazard, particularly if it concerns slow-growing, low-grade
DCIS. The conventional treatment of low-grade DCIS is similar to that of early-stage iBC, i.e. breast-conserving surgery (BCS)
often followed by radiotherapy (RT), or mastectomy, and possibly hormonal therapy (HT). This implies that many women might be
unnecessarily going through intensive treatment resulting in a decrease in quality of life and an increase in health care costs,
without any survival benefit.
Trial Design
LORD is a randomized, international, multicenter, phase III non-inferiority trial. Patients will be randomized between active
surveillance and standard treatment according to local policy. Active surveillance comprises monitoring by annual mammography
and treatment if there is progression to iBC or higher grade DCIS. Standard treatment encompasses BCS only, BCS + RT, or
mastectomy, possibly followed by HT, and follow-up by annual mammography.
Eligibility criteria
Women 49 years
Referral to the hospital solely based on microcalcifications detected by population-based or opportunistic screening
mammography
Unilateral, pure DCIS grade I based on multiple vacuum assisted biopsies
No prior history of iBC or DCIS
Specific Aims
Our aim is to investigate whether active surveillance of low-grade DCIS is as safe as the current standard treatment, and to study
the effects on quality of life of these women by saving them from intensive treatment. The primary end-point is time to ipsilateral
iBC (iiBC). Secondary end-points include rate of iBC at final pathology after standard treatment, radical intervention rate, and
biopsy rate during follow-up. Tissue will be collected for genomic profiling and proteomics to detect iBC risk patterns in low-grade
DCIS. Furthermore we aim to develop an online informational aid for clinicians and patients to facilitate the dissemination and
understanding of information on low-grade DCIS and its treatment.
Statistical Methods
We apply a special type of non-inferiority design, where the alternative hypothesis corresponds to "minor inferiority". We assume
that the iiBC-free rate in the standard arm at 5 years equals 97.6%. The hazard ratio under the null hypothesis is 3.4 and the
hazard ratio under the alternative hypothesis is 2.1, corresponding to iiBC-free rates in the experimental arm of 92.0% and 95.0%
respectively. With a one-sided log-rank test for non-inferiority with =0.05 and =0.2, 124 events are needed for which 1842
low-grade DCIS patients need to be randomized during an accrual period of 5.5 years, and the accrual period will be followed by
a further follow-up period of 7.5 years.
Present Accrual and Target accrual
The LORD trial is endorsed by the Dutch Breast Cancer Research Group (BOOG 2014-04) and the European Organisation for
Research and Treatment of Cancer (EORTC 1401). Participating sites will be identified mid-2014. Patient accrual is expected to
start early 2015.

Sabcs 2014 Allabstracts

Uploaded by

Copyright:

Available Formats

Sabcs 2014 Allabstracts

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sabcs 2014 Allabstracts

Uploaded by

Copyright:

Available Formats

2014 San Antonio Breast Cancer Symposium

Publication Number: S1-01

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

0.58 (0.25, 1.35)

2.99 (0.40, 22.70)

1.46 (0.64, 3.35)

3.29 (0.42, 25.80)

2.42 (1.04, 5.64)

8.22 (1.06, 63.60)

0.61 (0.41, 0.93)

0.79 (0.44, 1.41)

0.2 (0.06, 0.65)

1.18 (0.73, 1.91)

1.08 (0.98, 1.19)

1.00 (0.93, 1.07)

0.99 (0.97, 1.00)

1.24 (0.65, 2.39)

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

Median age yrs

Events/Pts 5-yr BCFI

Hazard Ratio (vs T)

HR 0.78 (95% CI, 0.60-1.02)

HR 0.65 (95% CI, 0.49-0.87)

HR 0.87 (95%CI, 0.64-1.17)

HR 0.72 (95% CI, 0.52-0.98)

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

Breast Ultrasounds for Dense Breasts

BIRAD 4 and 5 Ultrasounds

Number Cancers per 1000 Screened

2014 San Antonio Breast Cancer Symposium

2014 San Antonio Breast Cancer Symposium

BCS with RT, CM