Radiotherapy and Oncology xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Radiotherapy and Oncology

journal homepage: www.thegreenjournal.com

Original article

Ten years results of the Canadian breast intensity modulated radiation

therapy (IMRT) randomized controlled trial
Jean-Philippe Pignol a,⇑, Pauline Truong b, Eileen Rakovitch c, Margriet G. Sattler a, Timothy J. Whelan d,
Ivo A. Olivotto e
a
Erasmus MC Cancer Institute, Radiation Oncology Department, Rotterdam, The Netherlands; b Vancouver Island Cancer Centre, Radiation Oncology Department, Victoria;
c
Sunnybrook Health Sciences Centre, Institute for Clinical Evaluative Sciences and Radiation Oncology Department, Toronto; d Hamilton Health Sciences, Juravinski Cancer Centre,
Hamilton; and e Tom Baker Cancer Centre, Division of Radiation Oncology, Calgary, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Background and purpose: We report the long-term outcomes in patients enrolled in a multicenter ran-
Received 11 June 2016 domized controlled trial comparing Intensity Modulated Radiation Therapy (IMRT) with standard wedge
Received in revised form 20 August 2016 radiotherapy.
Accepted 29 August 2016
Materials and methods: Trial participants were assessed to compare long-term side effects between treat-
Available online xxxx
ment arms. The primary endpoint was chronic breast pain assessed by trained observers blinded to treat-
ment allocation. Secondary endpoints included cosmesis and quality of life measures.
Keywords:
Results: Median follow-up time was 9.8 years and 241 patients were available for assessment. There was
Breast
IMRT
no significant difference in chronic pain between treatment arms (OR = 0.74, range 0.432–1.271). There
Clinical trial were also no differences for the secondary endpoints. Univariate and multivariate analyses identified
young age (p = 0.013) and pain during RT (p < 0.001) to be associated with chronic pain. Acute moist
desquamation was associated with late subcutaneous fibrosis (p = 0.003) and telangiectasia (p = 0.039).
Pain during RT was associated with a long-term poorer self-assessed cosmetic outcome (p < 0.001) and
quality of life (p < 0.001).
Conclusions: Breast IMRT cannot be recommended for all patients to reduce long-term side effects.
However, late toxicities were significantly correlated with acute side effects, which are increased in
patients having poor dose distribution. Breast IMRT may hence be useful for selected patients.
Ó 2016 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology xxx (2016) xxx–xxx

With the advent of screening mammography, the majority of distribution of the dose through the target tissue, or a better avoid-
breast cancers are diagnosed at an early stage amenable to breast ance of underlying critical structures. In 2008, our group published
conserving therapy [1]. Those patients have excellent outcomes, the results of a multicenter randomized trial demonstrating that
so the goals of treatment are to ensure local control, breast preser- breast IMRT reduced acute toxicities compared to standard RT
vation, and to minimize acute and long-term side effects. Based on using a 2D wedge technique for missing tissue compensation [9].
large randomized trials and meta-analyses, the standard treatment The rate of moist desquamation was reduced from 47.8% with stan-
for early stage breast cancer includes breast-conserving surgery dard wedge RT to 31.2% with breast IMRT (p = 0.002). Moist
followed by whole breast adjuvant radiotherapy (RT) [2–5]. desquamation was significantly associated with pain (p = 0.002)
Radiotherapy can lead to acute toxicities, including moist and reduced health-related quality of life (p = 0.003). Although
desquamation and breast pain, and long-term toxicities including important those symptoms are spontaneously resolving, so it is
skin and soft tissue fibrosis, cosmetic changes and chronic pain important to assess if IMRT may also reduce the long-term effects
[6–8]. Over the last 20 years there have been many advances in of RT on the patient and her breast to justify adopting the tech-
the technical delivery of RT. Intensity Modulated RT (IMRT) per- nique [10].
mits modulation of the RT beam to ensure either a more uniform In the initial study a secondary objective was to determine if
increased acute skin toxicity translated into increased late skin
toxicity, and the current study reports late toxicity outcomes in
⇑ Corresponding author at: Erasmus MC Cancer Institute, Radiation Oncology patients enrolled in the Canadian multicenter randomized con-
Department, 301 Groene Hilledijk, Rotterdam, NL 3008 AE, The Netherlands. trolled trial, 8–10 years after the initial surgery [9].
E-mail address: j.p.pignol@erasmusmc.nl (J.-P. Pignol).

http://dx.doi.org/10.1016/j.radonc.2016.08.021
0167-8140/Ó 2016 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: Pignol J-P et al. Ten years results of the Canadian breast intensity modulated radiation therapy (IMRT) randomized con-
trolled trial. Radiother Oncol (2016), http://dx.doi.org/10.1016/j.radonc.2016.08.021
2 Breast IMRT clinical trial 10 years results

Methods breast cancer rating system, that uses four categories of skin sur-
face involvement (none, <1 cm2, 1–4 cm2 and >4 cm2) [17]. Quality
Randomization and masking of life was evaluated using the EORTC QLQ30 general module and
the BR23 module self-assessment questionnaires [18,19].
Patients were randomized to receive a breast RT 50 Gy in 25
Local recurrence-free survival was defined as the absence of
fractions, using two opposing beams with either standard RT using
invasive or in situ recurrent cancer in the ipsilateral breast. Disease
a 2D-wedge technique for missing-tissue compensation or breast
free-survival was defined as the absence of any cancer events in
IMRT in a multicenter, double-blinded, randomized controlled
the surviving patients. Overall survival was defined as the absence
trial. Either inverse-planned or IMRT or forward-planned field-in-
of death from any cause at the latest follow-up time.
field techniques were used and randomization was blocked 1:1
according to breast size and the delivery of an additional boost
dose to the surgical bed. Details of the treatment techniques were Sample size
previously reported [9]. Dosimetry quality assurance measures
included the capture and analysis of DVHs and standard portal The primary endpoint of chronic breast pain was estimated
imaging. from available literature as approximately 43% among patients
All surviving patients enrolled in the initial trial were contacted treated with breast conserving surgery followed by adjuvant radio-
and offered participation in the long-term follow-up study. therapy [7–8,20]. The sample size calculation assumed a reduction
Patients were given standardized questionnaires and scales for of a factor of 2 for chronic pain using breast IMRT, from 43% to 22%.
toxicity scoring before being examined by a radiation oncologist A sample size of 148 patients was calculated as being sufficient to
and a clinical research assistant (CRA), both blinded to the treat- detect a significant difference between those groups with a = 0.05
ment arm. and a power of 80%.

Eligibility Analysis

All patients who received their assigned treatment on trial were The comparisons of patient characteristics and outcomes
eligible for the current follow-up study. The trial accrued women between the two treatment arms were performed using Chi-
with a confirmed diagnosis of invasive carcinoma or ductal carci- square tests for categorical variables and Wilcoxon’s rank sum
noma in situ, with zero to three positive nodes, treated with breast tests for continuous variables. An evaluation of the factors associ-
conserving surgery. All patients on trial were eligible to receive ated with chronic pain defined as a pain grade 1 or higher of the
adjuvant radiotherapy to the breast alone, without radiation to NCI CTCAE scale, poorer cosmetic outcomes or quality of life was
the regional nodes. Patients were excluded if they had bilateral also performed using univariate and multivariate analyses. Uni-
breast cancers, a post-operative wound infection requiring antibi- variate analysis included Spearman’s rank order statistics for dis-
otics, previous radiation to the same breast, connective tissue dis- crete variables, and Wilcoxon’s rank sum tests or Spearman’s
orders, or were pregnant. correlation statistics for continuous variables. Significant factors
The institutional research ethics board of the Sunnybrook Health with a p value below 0.05 in univariate analysis were incorporated
Sciences Centre and the British Columbia Cancer Agency approved into multivariate models. Kaplan–Meier survival curves and Log-
the initial and follow-up study protocols and all patients provided Rank tests were used to evaluate the survival outcomes between
written informed consent. The study was done in accordance with treatment arms.
the Code of Ethics of the World Medical Association (Declaration
of Helsinki) for experiments involving humans and was registered
at the National Institute of Health (www.clinicaltrials.gov Role of the funding source
#NCT01803139). The Canadian Institutes of Health Research fund the trial after
anonymous peer review. It had no other role in study design, col-
lection, analyses or interpretation of data.
Study endpoints and evaluation
The primary endpoint was chronic breast pain, quantitatively
Results
graded by the blinded CRA using the Common Terminology Criteria
for Adverse Events Version 4.0 (CTCAE 4.0) [11]. This is a four-point
Study cohort
scale, ranking pain on its impact on the instrumental and self-care
activities of daily life. Pain was also quantitatively reported by Of 358 patients enrolled in the initial trial, twelve were
study subjects using a Visual Analog Scale ranging from 0 (no pain) excluded since they did not receive the allocated treatment. Of
to 10 (extreme pain) to determine the worst pain experienced over the remaining 346 patients, 30 had died, 14 declined participation,
the preceding six months [12], and qualitatively using the short 9 were lost to follow-up, 6 had mastectomy and were ineligible,
form (SF) of the McGill pain questionnaire [13]. and 46 agreed to participate but did not arrive at the scheduled
Secondary endpoints included breast cosmesis, quality of life, appointment and declined further contact. Ninety-one patients
and local recurrence-free, disease-free, and overall survival. Cos- accepted assessment by phone or by completing questionnaires
metic outcomes were self-assessed by the patients using the Breast but declined traveling to the cancer center for a physical examina-
Cancer Treatment Outcome Scale (BCTOS) questionnaire [14]. To tion. A total of 241 out of 316 potentially eligible patients (76%)
improve consistency, radiation oncologist and a trained CRA per- were available for the analysis of the primary endpoint, 125 in
formed a joint consensus cosmetic assessment using the EORTC the breast IMRT arm and 116 in the standard RT arm (Fig. 1).
breast cancer rating system, which incorporates assessment of The median follow-up time from surgery was 9.8 years, range
breast size or shape, nipple location, skin pigmentation, edema, 8.1–11.1 years. Patient and treatment characteristics are summa-
visibility of scar, and overall cosmetic appearance [15]. The pres- rized in Table 1. There was no statistically significant difference
ence of sub-cutaneous induration was evaluated using the RTOG/ between the two arms There were significant (p < 0.001) differ-
EORTC Late Radiation Morbidity Scoring Scheme scale [16]. Telang- ences in the RT dosimetry characteristics, reflecting the better dose
iectasia was scored using Bentzen’s scale, included in the EORTC distribution homogeneity provided by breast IMRT.

Please cite this article in press as: Pignol J-P et al. Ten years results of the Canadian breast intensity modulated radiation therapy (IMRT) randomized con-
trolled trial. Radiother Oncol (2016), http://dx.doi.org/10.1016/j.radonc.2016.08.021
 J.-P. Pignol et al. / Radiotherapy and Oncology xxx (2016) xxx–xxx 3

Assessed for eligibility in the initial study (n= 358)

Randomized

Breast IMRT (n=181) Allocation Standard Wedge RT (n=177)

- Received IMRT (n=176) - Received standard RT (n=170)
- Did not receive allocated - Did not receive allocated
intervention(n=5) intervention (n=7)

- Declined participation (n=5)

Follow-Up - Declined participation (n=9)
- Delayed mastectomy (n=3) study - Delayed mastectomy (n=3)
- Died (n=14) - Died (n=16)
- Lost to follow-up (n=3) - Lost to follow-up (n=6)
- Failed to attend scheduled - Failed to attend scheduled
assessment (n=26) assessment(n=20)

Analyzed (n=125) Analysis Analyzed (n=116)

Fig. 1. Flow chart of study subjects.

Comparisons of outcomes between treatment arms short form of McGill pain questionnaire, and 4 measures of
reduced cosmetic outcomes, including the self-reported BCTOS
Between treatment arms there was no significant differences in
score, overall EORTC score, the occurrence of fibrosis and presence
grade 2 or higher chronic pain, as assessed by the CRA, respectively
of telangiectasia.
6.9% using a standard wedge RT and 5.6% using breast IMRT (OR
Young age and the occurrence of pain during initial RT were sig-
0.79, range 0.227–2.263, p = 0.67). There was also no difference
nificantly associated with the development of chronic pain. Acute
in grade I or higher chronic pain, 29.6% using breast IMRT and
moist desquamation was also significantly associated with CRA-
36.2% with standard RT (OR = 0.74, range 0.432–1.271, p = 0.34).
measured pain (p < 0.001). A worse BCTOS score was significantly
Similarly there was no difference between arms in the occurrence
associated with severe pain during radiotherapy (p < 0.001). The
of self-reported VAS pain score P 5 (15.7% versus 13.8% respec-
occurrence of telangiectasia was significantly associated with pain
tively) or in the VAS intensity when the pain was present
(p = 0.027) and moist desquamation (OR = 4.87, range 1.61–14.7,
(p = 0.61). Finally there was no difference in the quality of pain
p < 0.001) during initial RT. Fibrosis was significantly associated
measured by the SF-McGill questionnaire (13.5% versus 11.0% pos-
with poor radiotherapy dose distributions measured by the V110
itive score respectively).
(p = 0.001), which represents the proportion of breast receiving
Global cosmetic outcome using the EORTC cosmetic breast
an excess of 10% of the prescribed radiation dose, and also with
cancer-rating system were statistically similar between the breast
the occurrence of moist desquamation during the radiation treat-
IMRT and the standard RT arms (82.0% versus 82.7% good to excel-
ment (OR = 3.75, range 1.62–8.63, p = 0.002).
lent cosmetic outcomes respectively). There was also no significant
Analyzing the factors possibly associated with a reduction in
difference in the median patient self-reported cosmesis using the
the quality of life using the QLQ30 questionnaire, we found that
BCTOS scores (p = 0.90). Finally, the rates of telangiectasia (7.8%
a larger BMI was associated with reduced physical (p = 0.004)
versus 12.2% respectively, p = 0.68) and fibrosis (17% versus 24%
and social (p = 0.033) functioning, and an increased fatigue
respectively, p = 0.58) were not statistically different between
(p < 0.001). Pain during radiotherapy was associated with
breast IMRT and standard RT, but it is possible that the small sam-
decreased physical (p = 0.046), cognitive (p = 0.007) and social
ple size prevents the results to be significant.
(p = 0.042) functioning, and increased fatigue (p = 0.002) and
Quality of life, measured using the QLQ30 or BR23 self-
insomnia (p < 0.001).
assessment questionnaires, were also similar between treatment
A multivariate model was developed using the RT technique,
arms.
the patient’s age, acute moist desquamation, and pain during RT
as independent variables, to evaluate their correlations with
chronic pain (Table 4) and cosmetic outcomes (Table 5). Acute
Univariate and multivariate analyses moist desquamation was associated with the development of
Tables 2 and 3 summarize the univariate analysis of the factors long-term fibrosis (p = 0.003) and telangiectasia (p = 0.039). Pain
associated with 3 measures of pain, including the CRA-measured during RT was highly correlated with the nine different measures
pain, the worst VAS score over the preceding 6 months and the of chronic pain, with most p values < 0.001. Pain during RT also

Please cite this article in press as: Pignol J-P et al. Ten years results of the Canadian breast intensity modulated radiation therapy (IMRT) randomized con-
trolled trial. Radiother Oncol (2016), http://dx.doi.org/10.1016/j.radonc.2016.08.021
4 Breast IMRT clinical trial 10 years results

Table 1 Table 2
Patient and treatment characteristics between the two arms. Median values are Univariate analysis of patient and treatment factors associated with breast pain and
followed by range; absolute numbers are followed by percentages. poor cosmetic outcome at 10 years.

Patient characteristics Breast IMRT Standard Wedge p CRA-assessed Patient-assessed pain

RT valuea pain
(N = 125) (N = 116) Worst SF McGill Total
VAS score
Mean age in years [range] 56.1 [25–82] 57.6 [36–79] 0.85
Factor (N = 241) (N = 176) (N = 180)
Median body mass index 27.3 [16–58] 26.7 [19–42] 0.69
[range] Age p = 0.003 p = 0.002 p = 0.002
Bra size Body mass index NS NS NS
Small (32A,B; 34A,B; 36A) 19 (15%) 23 (20%) 0.62 Breast size NS NS NS
Medium (32C; 34C; 36B,C; 67 (54%) 57 (49%) Smoking NS NS NS
38A,B,C) Diabetes NS p = 0.038 NS
Large (>38 or >C cup) 39 (31%) 36 (31%) Chemotherapy NS NS NS
Smoking 11 (8.8%) 10 (8.6%) 0.96 Hormonotherapy NS NS NS
Diabetes 4 (3.1%) 6 (5.2%) 0.44 Re-excision surgery NS NS NS
Histology Radiation technique NS NS NS
Invasive ductal cancer 93 (74%) 85/115 (74%) 0.44 V110 NS NS NS
Invasive lobular cancer 7 (6%) 11/115 (10%) Radiation boost NS NS NS
Ductal carcinoma in-situ 25 (20%) 19/115 (16%) Moist p < 0.001 NS NS
Grade (Scarff–Bloom Richardson) desquamation
I 40/123 (33%) 25/115 (22%) 0.17 Pain during RT p < 0.001 p < 0.001 p = 0.001
II 48/123 (39%) 51/115 (44%)
III 35/123 (28%) 39/115 (34%) Abbreviations: CRA = clinical research assistant; VAS = visual analog scale; SF
Estrogen receptors positive 89/109 (82%) 86/102(84%) 0.77 McGill = short form of McGill pain questionnaire; NS = non significant; V110 -
Lympho-vascular invasion 12/121 (10%) 11/115 (10%) 0.89 = relative volume receiving more than 110% of the prescribed dose; RT = radiation
Tumor size (cm) mean [range] 1.4 [0.2–7.0] 1.1 [0.1–4.0] 0.45 therapy.
Node-positive 8/113 (7%) 8/111 (7%) 0.97
Treatment characteristics by questionnaire five years or more after initial treatment [8]. In
Chemotherapy 39/125 (31%) 40/115 (35%) 0.55 this later series the chronic pain occurrence was 20% lower after
Hormone therapy 70/124 (56%) 74/114 (64%) 0.22 10 years compared to the five to ten years rate after breast surgery
Re-excision surgery 17/124 (14%) 19/115 (16%) 0.54
(p = 0.042). Out of note for the present study, baseline pain data
Sentinel node staging 31/102 (30%) 45/100 (45%) 0.06
Radiation boost 39/125 (31%) 30/116 (26%) 0.36 before radiotherapy revealed no difference between arms and a
much lower rate of pain grade I or higher compared to long term
Dosimetric characteristics
Volume of breast treated in cc 964 [215– 931 [258–2890] 0.53 data, 5.8% using IMRT and 5.5% using standard RT.
2274] Our analysis of cosmetic outcome did not identify significant
Dose maximum (%) 105 [101– 110 [105–120] <0.001 differences between treatment arms. A limitation to our ability
118]
to detect cosmetic deterioration linked to radiotherapy may be
V110 (%) 0.05 [0–9.8] 1.5 [0–12.9] <0.001
due to the absence of consistently recorded baseline cosmetic
Abbreviations: RT = radiation therapy; IMRT = intensity-modulated radiation ther- scores prior to RT and hence potential other confounding factors
apy; V110 = relative volume receiving more than 110% of the prescribed dose. including the surgery. However as the patients were randomized
a
Significance using Chi-squared or Wilcoxon’s rank-sum tests.
between arms, the impact of those factors is unlikely to bias the
results. Unlike the present study, the Cambridge and the Royal
impacted negatively on patient-assessed cosmetic outcome using Marsden Hospital randomized trials of breast IMRT have reported
the BCTOS scale. improved five-years cosmetic outcomes with breast IMRT [21–
No differences were observed between treatment arms at nine 23]. In those trials, cosmesis was assessed by rating breast appear-
years in overall survival (90.4% versus 92.6%, p = 0.47), local ance on clinical photographs and by documenting the rate of
recurrence-free survival (95.4% versus 94.1%, p = 0.55), or disease telangiectasia [21,23]. Distinct from the current trial, which
free-survival (82% versus 82.4%, p = 0.90), among women receiving included all patients who were candidates for RT to the breast
standard RT and breast IMRT, respectively. alone, the Cambridge and Royal Marsden trials restricted accrual
to patients deemed to have higher risks of developing radiotherapy
side effects. The Cambridge study selected patients with inhomo-
Discussion geneous RT dose distributions (defined as >2 cc of the breast
receiving an excess of 107% of the prescribed dose) and employed
In this analysis of 10-years outcomes among patients enrolled a hypofractionated regimen of 40 Gy in 15 fractions [21]. The Royal
in a multicenter randomized trial we were unable to demonstrate Marsden study accrued patients judged to be at higher risk of
a significant impact of the RT technique on the occurrence of radiation-induced normal tissue changes due to breast size or
chronic breast pain. Between arms, there was a negligible 1.3% dif- shape [22–23]. The improved cosmesis with breast IMRT compared
ference in chronic pain impacting on the activity of daily life (grade to standard RT in those trials may thus be attributed to patient
2 or higher) between arms, and a marginal 6% reduction of chronic selection that increased the likelihood that the improved homo-
pain grade 1 or higher, from 36.2% using standard RT to 29.6% using geneity achieved with IMRT would translate into clinically mea-
breast IMRT (OR = 0.74, range 0.432–1.271, p = 0.34). Although the sured outcome differences.
study was not powered to test for non-inferiority, a reduction of 6% Although the results of this study do not support the use of
of mainly pain grade 1 not impacting on the activity of daily life breast IMRT to reduce long-term radiation side effects for all
can be considered as not clinically important. Our data founding patients, univariate and multivariate analyses identified significant
about a third of patient experiencing chronic breast pain grade I correlations between long-term and early toxicities that can be sig-
or higher are consistent with other series. Caffo reported a rate nificantly reduced by breast IMRT. Notably the existence of acute
of 43.5% of chronic pain within five years following breast radio- toxicities, including moist desquamation and pain during or just
surgical treatment in a cohort of 529 patients [20]. Peuckman after the radiotherapy, is significantly associated with chronic pain.
reported 42% of chronic pain in a cohort 1316 patients assessed In regard to dosimetry predictive factors, investigators from the

Please cite this article in press as: Pignol J-P et al. Ten years results of the Canadian breast intensity modulated radiation therapy (IMRT) randomized con-
trolled trial. Radiother Oncol (2016), http://dx.doi.org/10.1016/j.radonc.2016.08.021
 J.-P. Pignol et al. / Radiotherapy and Oncology xxx (2016) xxx–xxx 5

Table 3
Univariate analysis of factors associated with poor cosmetic outcome at 10 years.

BCTOS EORTC overall Fibrosis Telangiectasia

Factor (N = 157) (N = 150) (N = 150) (N = 151)
Age NS NS NS NS
Body mass index NS NS NS NS
Smoking NS NS NS NS
Diabetes NS NS NS NS
Chemotherapy NS NS NS NS
Hormone therapy NS p = 0.012 NS NS
Re-excision surgery NS NS NS NS
Radiation technique NS NS NS NS
V110 NS NS p = 0.001 NS
Radiation boost NS NS NS NS
Moist desquamation NS NS p = 0.002 p < 0.001
Pain during RT p < 0.001 NS NS p = 0.027

Abbreviations: BCTOS = Breast Cancer Treatment Outcome Scale; EORTC = European Organization for Research and Treatment of Cancer; NS = non significant; V110 = relative
volume receiving more than 110% of the prescribed dose; RT = radiation therapy.

Table 4
Multivariate analysis of factors associated with chronic breast pain or inferior cosmetic outcomes at 10 years.

Chronic pain measure Age Technique (wedge versus IMRT) Initial moist desquamation Pain during radiotherapy
CRA-assessed pain p = 0.013 NS NS p < 0.001
VAS p = 0.007 NS NS p = 0.002
SF McGill total p = 0.006 NS NS p = 0.005

Abbreviations: NS = non significant; CRA = clinical research assistant; VAS = visual analog scale; SF McGill = short form of McGill pain questionnaire.

have reported similar findings in retrospective analyses with

Table 5 shorter median follow-up times of five years. In a series of 416
Multivariate analysis of factors associated with inferior cosmetic outcome at 10 years. patients treated before the era of breast IMRT, Lilla et al. reported
Cosmetic Radiation technique Acute moist Pain during that telangiectasia occurred in 31.4% at 4.2 years, a markedly
measure (wedge RT versus IMRT) desquamation radiotherapy higher rate compared to the current study’s rate of 12.2% at
Fibrosis NS p = 0.003 NS 9.8 years, possibly related to either the use of older equipment
Telangiectasia NS p = 0.039 NS and techniques, or a favorable evolution of telangiectasia over time
EORTC NS NS NS [6].
cosmetic Taken together, these data suggest that the cause of chronic
BCTOS NS NS p < 0.001
pain, poorer cosmetic outcomes and reduction in the quality of life
Abbreviations: NS = non significant; EORTC = European Organization for Research are multifactorial. On one hand, those long-term side effects
and Treatment of Cancer; BCTOS = Breast Cancer Treatment Outcome Scale; appear significantly linked to moist desquamation and the corre-
RT = radiation therapy; IMRT = intensity modulated radiation therapy; Quality of
sponding acute pain experience during radiotherapy. On the other
life = Quality of Life.
hand, despite the evidence that breast IMRT improves the radiation
dose distribution and reduces moist desquamation, this technique
Dana Farber Cancer Institute reported long-term breast pain in 355 may not benefit all patients regarding long-term side effect reduc-
patients treated with whole breast RT after breast conserving sur- tion. If breast IMRT remains one practical strategy, among others,
gery [24]. In their multivariate analysis an increase in the breast to reduce acute side effects, it may be beneficial only for selected
volume receiving more than 110% of the prescribed dose (V110) patients, practically those presenting significant excess of dose of
was significantly associated with long-term breast pain radiation in the breast on standard RT dosimetry.
(p = 0.007). Out of note, we did not find an association between
various DVH parameters with moist desquamation or acute pain Conclusion
in our initial study [9], and we did not find significant correlations
between chronic pain and DVH metrics for the 241 patients Long-term breast pain, cosmesis, and quality of life were similar
included in the present study. However our sample size might be between arms. As those side effects were highly correlated with
too small to draw conclusions about who might benefit based on moist desquamation and severe acute pain during radiation treat-
patient, tumor or treatment characteristics. ment, strategies to optimize RT dose homogeneity to prevent acute
Similarly to the results of the Cambridge trial [25] there was no side effects, especially moist desquamation, are justified. Although
difference between arms for patient reported outcome measures the present study reports on a limited number of patients, our find-
(PROMs). However chronic pain was significantly associated with ings and the reports from other randomized trials suggest Breast
a poorer quality of life (p < 0.001), a finding previously reported IMRT may be useful for selected patients who have a predictable
by several authors [7–8,20]. Similarly to Mukesh’s study PROMs higher risk of acute toxicity based on the initial dose distribution
were also influenced by non-radiotherapy factors like the patient’s planning.
age or body mass index [25]. This study also confirms previously
reported findings that younger age is significantly associated with
chronic pain [8,20,26]. In regard to cosmetic outcomes, we found Conflict of interest statement
significant correlations between acute moist desquamation and
later development of telangiectasia and fibrosis. Other authors We declare no conflict of interests.

Please cite this article in press as: Pignol J-P et al. Ten years results of the Canadian breast intensity modulated radiation therapy (IMRT) randomized con-
trolled trial. Radiother Oncol (2016), http://dx.doi.org/10.1016/j.radonc.2016.08.021
6 Breast IMRT clinical trial 10 years results

Acknowledgement [10] Kachnic LA, Powell SN. IMRT for breast cancer – balancing outcomes, patient
selection, and resource utilization. J Natl Cancer Inst 2011;103:777–9.
[11] http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_
This clinical trial has been supported by a grant from Canadian 8.5x11.pdf, accessed on April 11th, 2016.
Institutes of Health Research #231072. [12] Breivik EK, Bjornsson GA, Skovlund E. A comparison of pain rating scales by
sampling from clinical trial data. Clin J Pain 2000;16:22–8.
[13] Dworkin RH, Turk DC, Revicki DA, et al. Development and initial validation of
an expanded and revised version of the Short-form McGill Pain Questionnaire
Appendix A. Supplementary data (SF-MPQ-2). Pain 2009;144:35–42.
[14] Stanton AL, Krishnan L, Collins CA. Form or function? Part 1. Subjective
Supplementary data associated with this article can be found, in cosmetic and functional correlates of quality of life in women treated with
breast-conserving surgical procedures and radiotherapy. Cancer
the online version, at http://dx.doi.org/10.1016/j.radonc.2016.08.
2001;91:2273–81.
021. [15] Aaronson NK, Bartelink H, van Dongen JA, van Dam FS. Evaluation of breast
conserving therapy: clinical, methodological and psychosocial perspectives.
Eur J Surg Oncol 1998;14:133–40.
References [16] RTOG/EORTC. Late effect consensus conference. RTOG/EORTC. Radiother Oncol
1995;35:5–7.
[1] Nyström L, Andersson I, Bjurstam N, Frisell J, Nordenskjöld B, Rutqvist LE. [17] Bentzen SM, Overgaard M. Relationship between early and late normal-tissue
Long-term effects of mammography screening: updated overview of the injury after postmastectomy radiotherapy. Radioth Oncol 1991;20:159–65.
Swedish randomised trials. Lancet 2002;359:909–19. [18] Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for
[2] Early Breast Cancer Trialists’ Collaborative Group (EBCTCG), Darby S, McGale P, Research and Treatment of Cancer QLQ-C30: a quality of life instrument for
et al. Effect of radiotherapy after breast-conserving surgery on 10-year use in international clinical trials in oncology. J Natl Cancer Inst
recurrence and 15-year breast cancer death: meta-analysis of individual 1993;85:365–76.
patient data for 10,801 women in 17 randomised trials. Lancet [19] Sprangers MA, Groenvold M, Arraras JI, et al. The European Organization for
2011;378:1707–16. Research and Treatment of Cancer breast cancer-specific quality-of-life
[3] Clark RM, Whelan T, Levine M, et al. Randomized clinical trial of breast questionnaire module: First results from a three-country field study. J Clin
irradiation following lumpectomy and axillary dissection for node-negative Oncol 1996;14:2756–68.
breast cancer: an update. Ontario Clinical Oncology Group. J Natl Cancer Inst [20] Caffo O, Amichetti M, Ferro A, Lucenti A, Valduga F, Galligioni E. Pain and
1996;88:1659–64. quality of life after surgery for breast cancer. Breast Cancer Res Treat
[4] Fisher B, Anderson S, Bryant J, et al. Twenty-year follow-up of a randomized 2003;80:39–48.
trial comparing total mastectomy, lumpectomy, and lumpectomy plus [21] Donovan E, Bleakley N, Denholm E, et al. Randomised trial of standard 2D
irradiation for the treatment of invasive breast cancer. N Engl J Med radiotherapy (RT) versus intensity modulated radiotherapy (IMRT) in patients
2002;347:1233–41. prescribed breast radiotherapy. Radiother Oncol 2007;82:254–64.
[5] Veronesi U, Cascinelli N, Mariani L, et al. Twenty-year follow-up of a [22] Barnett GC, Wilkinson JS, Moody AM, et al. Randomized controlled trial of
randomized study compa ring breast-conserving surgery with radical forward-planned intensity modulated radiotherapy for early breast cancer:
mastectomy for early breast cancer. N Engl J Med 2002;347:1227–32. interim results at 2 years. Int J Radiat Oncol Biol Phys 2012;82:715–23.
[6] Lilla C, Ambrosone CB, Kropp S, et al. Predictive factors of late normal tissue [23] Mukesh MB, Barnett GC, Wilkinson JS, et al. Randomized controlled trial of
complications following radiotherapy for breast cancer. Breast Ca Res Treat intensity-modulated radiotherapy for early breast cancer: 5-year results
2007;106:143–50. confirm superior overall cosmesis. J Clin Oncol 2013;31:4488–95.
[7] Amichetti M, Caffo O. Pain after quadrantectomy and radiotherapy for early- [24] Mak KS, Chen YH, Catalano PJ, et al. Dosimetric inhomogeneity predicts for
stage breast cancer: incidence, characteristics and influence on quality of life. long-term breast pain after breast-conserving therapy. Int J Radiat Oncol Biol
Oncology 2003;65:23–8. Phys 2015;93:1087–95.
[8] Peuckmann V, Ekholm O, Rasmussen NK, et al. Chronic pain and other sequelae [25] Mukesh MB, Qian W, Wilkinson JS, et al. Patient reported outcome measures
in long-term breast cancer survivors: nationwide survey in Denmark. Eur J (PROMs) following forward planned field-in field IMRT: results from the
Pain 2009;13:478–85. Cambridge Breast IMRT trial. Radiother Oncol 2014;111:270–5.
[9] Pignol JP, Olivotto I, Rakovitch E, et al. A Phase III multicentre clinical trial of [26] Lundstedt D, Gustafsson M, Steineck G, et al. Risk factors of developing long-
Breast Intensity Modulated Radiation Therapy (IMRT) to reduce skin radiation lasting breast pain after breast cancer radiotherapy. Int J Radiat Oncol Biol
side effects. J Clin Oncol 2008;26:2085–92. Phys 2012;83:71–8.

Please cite this article in press as: Pignol J-P et al. Ten years results of the Canadian breast intensity modulated radiation therapy (IMRT) randomized con-
trolled trial. Radiother Oncol (2016), http://dx.doi.org/10.1016/j.radonc.2016.08.021