RESEARCH—HUMAN—CLINICAL STUDIES

Shingo Fujio, MD, PhD‡ § || #

Tareq A. Juratli, MD‡ §
Kazunori Arita, MD, PhD||
Hirofumi Hirano, MD, PhD||
Yushi Nagano, MD|| #
Tomoko Takajo, BS||
Koji Yoshimoto, MD, PhD|| #
A Clinical Rule for Preoperative Prediction of BRAF
Ivanna V. Bihun, BS‡
Alexander B. Kaplan, BA‡ Mutation Status in Craniopharyngiomas
Naema Nayyar, BS‡
Alexandria L. Fink, BS§
Mia S. Bertalan, BS‡
Shilpa S. Tummala, BS§ BACKGROUND: Papillary craniopharyngiomas are characterized by BRAFV600E mutations.
William T. Curry, Jr, MD§ Targeted therapy can elicit a dramatic radiographic regression of these tumors. Therefore,
Pamela S. Jones, MD, MS, MPH§
prediction of BRAF mutation status before definitive surgery could enable neoadjuvant

Downloaded from https://academic.oup.com/neurosurgery/article/85/2/204/5210774 by guest on 04 September 2020

Maria Martinez-Lage, MD¶
Daniel P. Cahill, MD, PhD§∗ treatment strategies.
Fred G. Barker, II, MD§∗
Priscilla K. Brastianos, MD‡∗ OBJECTIVE: To establish preoperative prediction criteria to identify patients with a BRAF
mutant craniopharyngioma.
‡
Divisions of Neuro-Oncology and He- METHODS: Sixty-four patients with craniopharyngioma were included in this study. We
matology/Oncology, Departments of
Medicine and Neurology, Massachu- determined BRAF mutation status by targeted sequencing. After scoring interobserver
setts General Hospital Cancer Center,
Harvard Medical School, Boston, Ma- variability between presurgical clinical data and radiographic features, we established a
ssachusetts; § Department of Neuros- diagnostic rule for BRAF mutation in our discovery cohort. We then validated the rule in an
urgery, Massachusetts General Hos-
pital, Harvard Medical School, Boston, independent cohort.
Massachusetts; ¶ Department of Path- RESULTS: The BRAFV600E mutation was detected in 12 of 42 patients in the discovery cohort.
ology, Massachusetts General Hos-
pital, Harvard Medical School, Boston, There were no patients under age 18 with BRAF mutation. Calcification was rare in tumors
Massachusetts; || Department of Neu- with BRAF mutation (P < .001), and 92% of them were supradiaphragmatic in location.
rosurgery, Graduate School of Medical
and Dental Sciences, Kagoshima Uni- Combining these 3 features—older than 18 years, absence of calcification, and supradi-
versity, Kagoshima, Japan; # Pituitary aphragmatic tumor location—we established a rule for predicting BRAF mutation. In cases
Disorders Center, Kagoshima Unive-
rsity Hospital, Kagoshima, Japan where all 3 criteria were fulfilled, the sensitivity and specificity for the presence of BRAF
mutation were 83% and 93%, respectively. In the validation cohort (n = 22), the sensitivity
∗
These authors contributed equally to was 100% and specificity was 89%.
this work.
CONCLUSION: We propose predictive criteria for a BRAF mutation in craniopharyngioma
Correspondence: using preoperative clinical and radiographic data. This rule may be useful in identifying
Priscilla K. Brastianos, MD, patients who could potentially benefit from neoadjuvant BRAF V600E -targeted systemic
Stephen E. and Catherine Pappas Center
therapies.
for Neuro-Oncology,
Divisions of Neuro-Oncology and KEY WORDS: BRAF V600E mutation, Clinical rule, Craniopharyngioma, Papillary, Prediction
Hematology/Oncology,
Departments of Medicine and Neurology, Neurosurgery 85:204–210, 2019 DOI:10.1093/neuros/nyy569 www.neurosurgery-online.com
Massachusetts General Hospital Cancer
Center,
Harvard Medical School,

C
55 Fruit Street, raniopharyngiomas are rare primary brain deficits, as well as the pituitary gland and the
Boston, MA 02114. tumors derived from the embryonic hypothalamus, leading to pituitary hormone
E-mail: PBRASTIANOS@mgh.harvard.edu remnants of Rathke’s pouch.1,2 While insufficiency, disorders of regulation of body
histologically benign, the clinical course experi- weight, memory loss, and psychiatric sequelae.1-4
Received, June 5, 2018.
Accepted, November 11, 2018. enced by patients with craniopharyngiomas is Surgical resection and adjuvant radiation have
Published Online, November 27, 2018. often neurologically devastating. Located in served as the most common forms of treatment,
the midline parasellar region, these tumors but the rate of complications, such as endocrine
C Congress of Neurological Surgeons
compress nearby critical structures, including deficits, cranial nerve injuries, and hypothalamic
2018.
the optic nerves, causing ophthalmological disorder, is relatively high due to the sensitive
This is an Open Access article distributed location of these tumors and the challenging
under the terms of the Creative
Commons Attribution-NonCommercial-
process of complete surgical removal.2,3
NoDerivs licence ABBREVIATIONS: ACP, adamantinomatous cranio- Craniopharyngiomas are divided in 2 main
(http://creativecommons.org/licenses/ pharyngioma; CT, computed tomography; MRI, histological subtypes: adamantinomatous cranio-
by-nc-nd/4.0/), which permits magnetic resonance imaging; PCP, papillary cranio-
pharyngioma (ACP) and papillary craniopharyn-
non-commercial reproduction and pharyngioma; WT, wild type
distribution of the work, in any medium,
gioma (PCP). While ACP follows a bimodal
provided the original work is not altered Neurosurgery Speaks! Audio abstracts available for this age distribution with peaks in the range of 5
or transformed in any way, and that the article at www.neurosurgery-online.com. to 15 yr and in the fourth to sixth decades
work is properly cited. For commercial
re-use, please contact
journals.permissions@oup.com

204 | VOLUME 85 | NUMBER 2 | AUGUST 2019 www.neurosurgery-online.com

 PREDICTION OF BRAF MUTATIONS IN CRANIOPHARYNGIOMA

of life, PCP occurs predominantly in adults, typically between

the ages of 40 and 55 yr.1 ACPs are characterized by alterations in
the Wnt pathway, mainly involving the central regulatory gene
CTNNB1 that encodes β-catenin,5-10 whereas the majority of
PCPs are driven by a canonical V600E mutation in the BRAF
gene, which constitutively activates the mitogen-activated protein
kinase signaling pathway.5-10 Dramatic therapeutic responses
following treatment of patients harboring BRAF V600E mutations
have been observed with BRAF inhibitors.11-14 Currently, BRAF
V600E
can only be detected via postoperative tissue analysis, but

Downloaded from https://academic.oup.com/neurosurgery/article/85/2/204/5210774 by guest on 04 September 2020

preoperative prediction of this mutation could enable BRAF
V600E
-targeted inhibitor treatment prior to a surgical procedure.
Such a neoadjuvant strategy could reduce tumor burden prior to
resection, potentially reducing operative morbidity.
To facilitate such a strategy, we conducted a multicenter study
to derive diagnostic criteria from patient demographic infor-
mation and preoperative images, with an aim to establish a
clinical predictive rule for the presence of BRAF V600E mutation
in patients with craniopharyngiomas.

METHODS
Description of Patient Cohorts
With Institutional Review Board approval, clinical and radiographic
data were collected from patients with histologically confirmed cranio-
pharyngioma. Patients in the discovery cohort were recruited between
January 2000 and April 2017. The validation cohort consisted of an
independent set of patients with surgical intervention between April
2002 and September 2016. The requirement for informed consent was
waived due to the retrospective nature of the study.

Tissue Samples and DNA Extraction

DNA was purified using AllPrep DNA/RNA formalin-fixed paraffin-
embedded (FFPE) kit for FFPE tissue (Qiagen, Germantown, Maryland)
following the manufacturer’s instructions. To ensure adequate tumor
histology and content, hematoxylin and eosin slides were reviewed from
each FFPE block before DNA extraction. A senior neuropathologist
independently reviewed tumor samples. The DNA was then quantified
using PicoGreenR
dye (Invitrogen, Carlsbad, California).

Genetic Analysis
To detect BRAF V600E mutations, polymerase chain reaction ampli-
fication was carried out with BRAF primers (forward: 5 CAGA
CAACTGTTCAAACTGATGGGACCCAC3 , reverse: 5 TGCTTG
CTCTGATAGGAAAATGA3 ).
CTNNB1 mutations were detected using the primers:
forward: 5 AGTTGGACATGGCCATGGAA3 , reverse: 5 ACATC
CTCTTCCTCAGGATT3 .

Imaging Data Characterization

We performed a retrospective analysis of the preoperative magnetic
FIGURE 1. Axial computed tomography bone image shows a high density
resonance imaging (MRI) and computed tomography (CT; bone
in the tumor mass, indicating calcification (arrow). A and B are tumors
window) images. The axial bone CT images were evaluated for the with an adamantinomatous histology. Calcifications may occur only inside
presence of intratumoral calcification (Figure 1). Additionally, the size, the tumor, or only on the cyst wall as a small granule.
volume, location, composition, and shape of the tumors were analyzed

NEUROSURGERY VOLUME 85 | NUMBER 2 | AUGUST 2019 | 205

FUJIO ET AL

Downloaded from https://academic.oup.com/neurosurgery/article/85/2/204/5210774 by guest on 04 September 2020

FIGURE 2. Sagittal postcontrast T1-weighted magnetic resonance images of craniopharyngiomas. A and B depict craniopharyngiomas of the subpial (Sp)
type. The tumor is located above the diaphragm, and the pituitary gland can be clearly observed in the sella. C and D are tumors of the subarachnoid (Sa)
type. The tumor was observed in the suprasellar region and expanded into the sella. The pituitary gland is visible under the tumor. E and F demonstrate
infradiaphragmatic (Id) type craniopharyngiomas. The tumor is mainly located in the sella. The pituitary gland is not clearly visible or compressed by the
tumor. The arrow indicates the tumor. The arrowhead indicates the pituitary gland.

on the MR images. The tumor volume variable was calculated using Whitney U test and Fisher’s exact test, according to the characteristics
the ABC/2 formula, and the tumor size variable was defined as the of the dataset. Inter-reviewer agreement was evaluated using the Kappa
maximum measurable diameter on axial MR images. The tumor location statistic (κ = 0-0.40, poor; κ = 0.41-0.60, moderate; κ = 0.61-0.80,
variable was categorized using an adapted scale modified from Pan good; κ = 0.81-1.00, excellent). A 2-sided P-value < .05 was considered
et al. 15 For simplification, we combined tumors in the subpial (Sp) to be statistically significant.
and subarachnoid (Sa) groups into one category that reflects a supra-
diaphragmatic origin (Figure 2). Using this approach, we classified the
craniopharyngiomas in our study into 2 subgroups: supradiaphragmatic
and infradiaphragmatic. The tumor composition was categorized as solid
RESULTS
or cystic, as previously described by Yue et al.10 In brief, if the solid Clinical, Pathological, and Genetic Alterations
component represented more than 50% of the volume, the tumor was
defined as solid, and if not, it was defined as cystic. Tumor shape was The discovery cohort consisted of 42 primary craniopharyn-
classified as spherical, lobular, or irregular. Two neurosurgeons who were gioma, 15 of which were papillary and the remaining 27
blinded to the mutational status of the patients reviewed the images and adamantinomatous. Targeted sequencing revealed 12/42 tumors
provided an assessment of interobserver variability. (28.6%) with a BRAFV600E mutation, all with a histologically
confirmed PCP (80% of all PCP vs none of ACP, P < .001;
Statistical Analysis Table 1). Consistent with previous reports, BRAF V600E and
Statistical analyses were performed using StatFlex version 6.0 CTNNB1 mutations were mutually exclusive. The median age at
(Artech Co Ltd, Osaka, Japan) and OpenEpi (http://www.openepi. diagnosis in the BRAF-mutant group was higher than in the BRAF
com/Menu/OE_Menu.htm). The data were analyzed using the Mann- wild-type (WT) group (52.0 vs 29.0 yr, P = .04). The youngest

206 | VOLUME 85 | NUMBER 2 | AUGUST 2019 www.neurosurgery-online.com

 PREDICTION OF BRAF MUTATIONS IN CRANIOPHARYNGIOMA

TABLE 1. Patients Demographics and Tumor Features in the Discovery and Validation Cohorts

Discovery cohort Validation cohort

BRAF V600E BRAF WTa BRAF V600E BRAF WT
mutation (n = 12) (n = 30) P-value mutation (n = 4) (n = 18) P-value

Age (years) 52 (18-70) 29 (0.8-77) .04b 68 (31-77) 36.5 (9-83) .16b

Sex .07c .09c
Male 8 9 0 11
Female 4 21 4 7
Imaging data

Downloaded from https://academic.oup.com/neurosurgery/article/85/2/204/5210774 by guest on 04 September 2020

Location .27c .50c
Supradiaphragmatic 11 21 4 12
Infradiaphragmatic 1 9 0 6
Tumor volume (cc) 7.8 (1.2-12.6) 8.7 (2.4-376) .37b n/a n/a
Maximum diameter (mm) 30 (15-36) 31 (20-96) .34b n/a n/a
Tumor composition .43c n/a n/a
Solid 4 5
Cystic 8 25
Calcification <.001c .01c
Yes 1 27 0 15
No 11 3 4 3
Pathological subtype <.001c <.001c
Papillary 12 3 4 0
Adamantinomatous 0 27 0 18
CTNNB1 genotype <.001c .01c
CTNNB1 mutated 0 21 0 15
CTNNB1 WT 12 9 4 3
n/a: not applicable.
a
Wild type.
b
Mann-Whitney U test.
c
Fisher’s exact test.
Data are presented as median (range) unless otherwise specified.

patient with a BRAF mutant tumor in our cohort was 18 yr old. In (κ = 0.06, P = .35), we did not take this parameter into further
the BRAF-WT group, 70% of the patients harbored a CTNNB1 consideration.
mutation. All tumors with a detectable CTNNB1 mutation had
adamantinomatous histology.
BRAF Mutation Probability Criteria
Based on our findings, supradiaphragmatic tumor location,
Radiographic Evaluation absence of calcification in the CT scan, and an age older than
We did not observe significant differences between BRAF-WT 18 yr were associated with the presence of BRAF mutation in
and BRAF-mutant tumors in terms of tumor volume, maximum craniopharyngiomas. We combined the 3 parameters to develop
tumor size, or tumor composition (solid vs cystic). However, a classification rule to identify craniopharyngiomas with BRAF
calcifications were significantly more common in BRAF-WT mutations. For this rule, all 3 criteria must be fulfilled in order to
tumors (27/30, 90%) than in BRAF-mutant tumors (1/12, 8.3%, classify a tumor as likely BRAF mutant.
P < .001). In addition, most of the BRAF-mutant tumors (11/12, Using this classification in our discovery cohort rendered 83%
91.7%) had a supradiaphragmatic location vs 70% of the BRAF- sensitivity and 93% specificity in predicting BRAF mutations
WT tumors (odds ratio: 4.71, 95% confidence interval: 0.53- (Table 3).
42.17, P = .27). To independently validate this rule, we tested these 3 selected
These radiographic findings in the discovery cohort are criteria in a separate cohort (Table 1) of 22 craniopharyngioma
summarized in Table 1. Furthermore, we evaluated the inter- patients (11 females and 11 males). Four tumors had a BRAF
V600E
reviewer consensus with the kappa statistic (Table 2). The inter- mutation, with a papillary histology in all 4 cases. Of the
reviewer agreement for tumor locations and for calcification were 18 ACPs, 15 (83.3%) harbored a CTNNB1 mutation—3 tumors
largely consistent (κ = 0.66 and 0.90, respectively). On the other were wild-type on assays for both mutations. The prediction
hand, since the inter-reviewer consensus for tumor shape was low sensitivity and specificity for a BRAF mutation in the validation

NEUROSURGERY VOLUME 85 | NUMBER 2 | AUGUST 2019 | 207

FUJIO ET AL

TABLE 2. Inter-reviewer Agreement on Multiple Imaging Tumor Features in the Discovery and Validation Cohorts

Discovery cohort Validation cohort

Category κ value P-value κ value P-value

Tumor location 0.66 (0.36 to 0.96) <.001 0.89 (0.48 to 1.00) <.001
Calcification 0.90 (0.60 to 1.00) <.001 0.90 (0.48 to 1.00) <.001
Tumor composition 0.66 (0.37 to 0.94) <.001 n/a
Tumor shape 0.06 (–0.06 to 0.18) .35 n/a
n/a: not applicable.

Downloaded from https://academic.oup.com/neurosurgery/article/85/2/204/5210774 by guest on 04 September 2020

TABLE 3. Assessment of Predictive Criteria in the Discovery Cohort

Number of matched parameters

Result 3 ≥2 ≥1

True-positive (cases) 10 12 12
False-positive (cases) 2 14 27
True-negative (cases) 28 16 3
False-negative (cases) 2 0 0
Sensitivity (95% CIa ) 0.83 (0.55-0.95) 1.00 (0.76-1.00) 1.00 (0.76-1.00)
Specificity (95% CI) 0.93 (0.79-0.98) 0.53 (0.36-0.70) 0.10 (0.03-0.26)
PPVb (95% CI) 0.83 (0.55-0.95) 0.46 (0.29-0.65) 0.31 (0.19-0.47)
NPVc (95% CI) 0.93 (0.79-0.98) 1.00 (0.81-1.00) 1.00 (0.44-1.00)
Positive LRd (95% CI) 12.49 (4.51-34.64) 2.14 (1.86-2.47) 1.11 (1.00-1.20)
Negative LR (95% CI) 0.18 (0.07-0.48) 0 0
a
Confidence interval.
b
Positive predictive value.
c
Negative predictive value.
d
Likelihood ratio.

cohort were 100% and 89%, respectively (Table 4). Additionally, juvant treatment for patients with craniopharyngioma. Such an
the inter-reviewer agreements were excellent (Table 2). approach would require a noninvasive prediction algorithm of
BRAF mutation status in these tumors. Ideally, the components
DISCUSSION of a prediction rule should be consistent with well-established
clinical characteristics of ACP and PCP.
Craniopharyngiomas can be clinically challenging tumors For example, although there are no reports relating BRAF
due to their location and the tendency to recur.2,3 Histori- mutation and calcification, it is well known that calcification is
cally, neurosurgeons have endeavored to achieve local control one of the defining characteristics of ACP.1,20,21 Lee et al reported
of these tumors through open surgical resections.16,17 More calcification in the CT scans in 83.3% ACPs and in none of the
recently, endoscopic surgery has gained popularity as a less PCPs.21 In our study, only 6% of the tumors from patients with
invasive procedure.18,19 Regardless of surgical technique, it is BRAF mutation showed calcification, while 88% of the tumors
difficult to avoid complications from radical resection while safely from BRAF-WT patients showed calcification. Therefore, the
achieving effective removal, due to the close proximity of critical presence of calcification serves as a negative predictor for BRAF
neurological structures.3 Given these surgical challenges, the mutation in craniopharyngiomas.
approach to treat BRAF mutant craniopharyngioma with BRAF It is thought that PCPs have a suprasellar origin.15,20 This
inhibitors is being investigated in an ongoing national Alliance- proposal is supported by the fact that over 90% of the
sponsored clinical trial (Alliance A071601) and has the potential BRAF-mutant craniopharyngioma arise in a supradiaphragmatic
to drive a transformative change in management strategy. BRAF location (Table 1).10,22 In keeping with this developmental logic,
inhibitor monotherapy or a combination of BRAF and mitogen- we find that the suprasellar/supradiaphragmatic tumor location
activated protein kinase (MEK) inhibitors can elicit a marked can indeed serve as a strong predictor for the presence of BRAF
regression in tumor volume,11-14 raising the possibility of neoad- mutations. The correlation between the molecular subtypes of

208 | VOLUME 85 | NUMBER 2 | AUGUST 2019 www.neurosurgery-online.com

 PREDICTION OF BRAF MUTATIONS IN CRANIOPHARYNGIOMA

TABLE 4. Assessment of Predictive Criteria in the Validation Cohort

Number of matched parameters

Result 3 ≥2 ≥1

True-positive (cases) 4 4 4
False-positive (cases) 2 12 17
True-negative (cases) 16 6 1
False-negative (cases) 0 0 0
Sensitivity (95% CIa ) 1.00 (0.51-1.00) 1.00 (0.51-1.00) 1.00 (0.51-1.00)
Specificity (95% CI) 0.89 (0.67-0.97) 0.33 (0.16-0.56) 0.06 (0.01-0.26)

Downloaded from https://academic.oup.com/neurosurgery/article/85/2/204/5210774 by guest on 04 September 2020

PPVb (95% CI) 0.67 (0.30-0.90) 0.25 (0.10-0.50) 0.19 (0.08-0.40)
NPVc (95% CI) 1.00 (0.81-1.00) 1.00 (0.61-1.00) 1.00 (0.21-1.00)
Positive LRd (95% CI) 9.00 (3.38-23.98) 1.50 (1.27-1.77) 1.06 (0.94-1.19)
Negative LR 0 0 0
a
Confidence interval.
b
Positive predictive value.
c
Negative predictive value.
d
Likelihood ratio.

craniopharyngiomas and tumor location may be explained by the 3 preoperative clinical factors: age above 18 yr, a supradiaphrag-
embryonic development of these tumors.20 matic tumor location, and absence of intratumoral calcification.
Finally, our findings on age are fully consistent with prior Importantly, these factors are straightforward to determine,
reports; when combining 6 studies with a total of 117 patients and have high interobserver agreement. We envision that the
harboring a BRAFV600E mutation, including those in our prediction of BRAF mutation prior to surgical intervention will
study,5-9,22 only 2 BRAF-mutant craniopharyngiomas were found facilitate the design of neoadjuvant treatments in patients with
in pediatric patients (9 and 12 yr old).5-9,22 Consequently, it craniopharyngioma.
follows that an age older than 18 yr at diagnosis is a predictive
factor for the presence of a BRAF mutation in craniopharyn- Disclosures
giomas. This work is supported by the Burroughs Wellcome Fund Career Award (to
D.P.C.) and the Damon Runyon Research Foundation (to P.K.B.). Dr Brastianos
Limitations received honoraria from Merck and Genentech, was a consultant for Lilly, Merck,
and Angiochem, and received research funding (to MGH) from Merck. The
An important limitation of our study is the small number authors have no personal, financial, or institutional interest in any of the drugs,
of criteria that can be assessed without risk of over-fitting. We materials, or devices described in this article.
have attempted to mitigate this limitation, by including an
independent validation cohort; nevertheless, our criteria should
be further validated in larger cohorts. It is conceivable that other REFERENCES
preoperative factors may prove helpful in the identification of 1. Buslei R, Rushing EJ, Giangaspero F, Paulus W, Burger PC, Santagat S. Cranio-
BRAF-mutant craniopharyngioma. For example, in a recent single pharyngioma. In: Louis DN, Ohgaki H, Wiestler OD, Cavenee WK eds. WHO
institution study, Yue et al reported that the BRAF mutation Classification of Tumours of the Central Nervous System, rev 4th edn. Lyon: IARC;
may be predicted based on MRI features of tumors, such as 2016: 324-328.
2. Yamini B, Narayanan M. Craniopharyngiomas: an update. Expert Rev Anticancer
location, shape, composition, enhancement pattern, and pituitary Ther. 2006;6(suppl 9):S85-S92.
stalk thickness—diagnostic criteria known to correlate with the 3. Jeswani S, Nuño M, Wu A, et al. Comparative analysis of outcomes following
BRAFV600E mutation.10 In our findings, however, there was a craniotomy and expanded endoscopic endonasal transsphenoidal resection of
poor interobserver agreement (Kappa statistic) with regard to craniopharyngioma and related tumors: a single-institution study. J Neurosurg.
2016;124(3):627-638.
the shape of the tumor, and thus we omitted it. Further studies 4. Shin JL, Asa SL, Woodhouse LJ, Smyth HS, Ezzat S. Cystic lesions
will be necessary to reach a consensus regarding the optimal of the pituitary: clinicopathological features distinguishing craniopharyngioma,
predictors for treatment with BRAF inhibitors in surgically naïve Rathke’s cleft cyst, and arachnoid cyst. J Clin Endocrinol Metab. 1999;84(11):3972-
3982.
patients. 5. Brastianos PK, Taylor-Weiner A, Manley PE, et al. Exome sequencing identifies
BRAF mutations in papillary craniopharyngiomas. Nat Genet. 2014;46(2):161-
165.
CONCLUSION 6. Hölsken A, Sill M, Merkle J, et al. Adamantinomatous and papillary
craniopharyngiomas are characterized by distinct epigenomic as well as
We propose a clinical rule with a high sensitivity and speci- mutational and transcriptomic profiles. Acta Neuropathol Commun. 2016;4
ficity in BRAF mutation prediction in craniopharyngiomas, using (1):20.

NEUROSURGERY VOLUME 85 | NUMBER 2 | AUGUST 2019 | 209

FUJIO ET AL

7. Malgulwar PB, Nambirajan A, Pathak P, et al. Study of β-catenin 15. Pan J, Qi S, Liu Y, et al. Growth patterns of craniopharyngiomas: clinical analysis
and BRAF alterations in adamantinomatous and papillary craniopharyngiomas: of 226 patients. J Neurosurg Pediatr. 2016;17(4):418-433.
mutation analysis with immunohistochemical correlation in 54 cases. J Neurooncol. 16. Symon L, Pell MF, Habib AH. Radical excision of craniopharyngioma by
2017;133(3):487-495. the temporal route: a review of 50 patients. Br J Neurosurg. 1991;5(6):539-
8. Marucci G, de Biase D, Zoli M, et al. Targeted BRAF and CTNNB1 next- 549.
generation sequencing allows proper classification of nonadenomatous lesions of 17. Fahlbusch R, Honegger J, Paulus W, Huk W, Buchfelder M. Surgical
the sellar region in samples with limiting amounts of lesional cells. Pituitary. treatment of craniopharyngiomas: experience with 168 patients. J Neurosurg.
2015;18(6):905-911. 1999;90(2):237-250.
9. Yoshimoto K, Hatae R, Suzuki SO, et al. High-resolution melting and 18. Komotar RJ, Starke RM, Raper DM, Anand VK, Schwartz TH.
immunohistochemical analysis efficiently detects mutually exclusive genetic alter- Endoscopic endonasal compared with microscopic transsphenoidal and open
ations of adamantinomatous and papillary craniopharyngiomas. Neuropathology. transcranial resection of craniopharyngiomas. World Neurosurg. 2012;77(2):329-
2018;38(1):3-10. 341.
10. Yue Q, Yu Y, Shi Z, et al. Prediction of BRAF mutation status of craniopharyn- 19. Moussazadeh N, Prabhu V, Bander ED, et al. Endoscopic endonasal versus open

Downloaded from https://academic.oup.com/neurosurgery/article/85/2/204/5210774 by guest on 04 September 2020

gioma using magnetic resonance imaging features. J Neurosurg. 2018;129(1): transcranial resection of craniopharyngiomas: a case-matched single-institution
27-34. analysis. Neurosurg Focus. 2016;41(6):E7.
11. Aylwin SJ, Bodi I, Beaney R. Pronounced response of papillary cranio- 20. Sartoretti-Schefer S, Wichmann W, Aguzzi A, Valavanis A. MR differentiation
pharyngioma to treatment with vemurafenib, a BRAF inhibitor. Pituitary. of adamantinous and squamous-papillary craniopharyngiomas. AJNR Am J Neuro-
2016;19(5):544-546. radiol. 1997;18(1):77-87.
12. Brastianos PK, Shankar GM, Gill CM, et al. Dramatic response of BRAF 21. Lee IH, Zan E, Bell WR, Burger PC, Sung H, Yousem DM. Cranio-
V600E mutant papillary craniopharyngioma to targeted therapy. J Natl Cancer pharyngiomas : radiological differentiation of two types. J Korean Neurosurg Soc.
Inst. 2016;108(2):djv310, doi: 10.1093/jnci/djv310. 2016;59(5):466-470.
13. Roque A, Odia Y. BRAF-V600E mutant papillary craniopharyngioma dramat- 22. Omay SB, Chen YN, Almeida JP, et al. Do craniopharyngioma molecular
ically responds to combination BRAF and MEK inhibitors. CNS Oncol. signatures correlate with clinical characteristics? J Neurosurg. 2018;128(5):1473-
2017;6(2):95-99. 1478.
14. Rostami E, Witt Nyström P, Libard S, Wikström J, Casar-Borota O, Gudjonsson
O. Recurrent papillary craniopharyngioma with BRAFV600E mutation treated
with neoadjuvant-targeted therapy. Acta Neurochir. 2017;159(11):2217- Neurosurgery Speaks! Audio abstracts available for this article at www.neurosurgery-
2221. online.com.

210 | VOLUME 85 | NUMBER 2 | AUGUST 2019 www.neurosurgery-online.com