Accepted Manuscript Posterior Fossa Intra-axial Tumors in Adults Rachel Grossman, MD, Zvi Ram, MD PII:

S1878-8750(15)01770-2

DOI:

10.1016/j.wneu.2015.12.066

Reference:

WNEU 3563

To appear in:

World Neurosurgery

Received Date: 30 May 2015 Revised Date:

21 December 2015

Accepted Date: 22 December 2015

Please cite this article as: Grossman R, Ram Z, Posterior Fossa Intra-axial Tumors in Adults, World Neurosurgery (2016), doi: 10.1016/j.wneu.2015.12.066. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

ACCEPTED MANUSCRIPT

Posterior Fossa Intra-axial Tumors in Adults

RI PT

Rachel Grossman, MD; Zvi Ram, MD

Department of Neurosurgery, Tel-Aviv Medical Center, affiliated to the

Correspondence to:

M AN U

SC

Sackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel

Rachel Grossman, M.D.

Department of Neurosurgery

Tel-Aviv Sourasky Medical Center

TE D

6 Weizmann St., Tel Aviv 6423906, Israel Telephone: +972-3-6974273 Fax: +972-3-6974860

AC C

EP

E-mail: [email protected]

Key words: posterior fossa, brain tumors, outcome, ependymomas, medulloblastomas, pilocytic astrocytomas, gliomas, lymphomas

2

ACCEPTED MANUSCRIPT Abstract

Background: The posterior fossa is the site of many types of tumors, and brain metastases are the most common malignancies in that location among adults. Other brain tumors, such as

RI PT

ependymomas, medulloblastomas and juvenile pilocytic astrocytomas, mostly occur during childhood and are relatively rare in the adult population. The majority of primary malignant brain tumors, such as gliomas and lymphomas, tend to be located in the supratentorial

SC

.compartment

Methods: In this review we summarized the current data including prognostic factors and

M AN U

therapeutic management and molecular data of patients with intra-axial posterior fossa tumors in adults including ependymomas, medulloblastomas pilocytic astrocytomas in the ."posterior fossa in adults

Results: We present the literature on intra-axial posterior fossa tumors in adults that relies

TE D

mainly on limited retrospective clinical studies and employing a wide range of treatment approaches that were usually based on therapies developed specifically for children or for supratentorial brain tumors

EP

Conclusions: The clinical course and surgical outcome of adult patients with intra-axial brain tumors in the posterior fossa is summarized in this review. The prognostic factors and

AC C

therapeutic management of patients with these tumors are controversial due to their rarity, their heterogeneity and the lack of sufficient data in the literature.

Review

The posterior fossa is the site of many types of tumors, and brain metastases are the most common adult malignancies in this region. Other brain tumors, such as ependymomas, medulloblastomas and juvenile pilocytic astrocytomas, mostly occur during childhood and are relatively rare in adults. The prognostic factors and therapeutic management of adult

3

ACCEPTED MANUSCRIPT patients with these tumors are controversial due to their rarity, their heterogeneity and the lack of sufficient data in the literature. In fact, most of the clinical series reported thus far were retrospective, involved pediatric populations, and included relatively small numbers of

RI PT

patients. As a result, the optimal treatment of posterior fossa tumors in adults remains a matter of debate. Attempts have been made to identify prognostic factors for those patients, but mostly have not been conclusively established. In this review, we characterize the clinical

SC

course and surgical outcome of adults diagnosed with an intra-axial brain tumor in the

M AN U

posterior fossa.

Ependymomas

Ependymoma, a relatively rare tumor in adults, accounts for only 2%-5% of all intracranial tumors, but they are the fourth most common brain tumor in children. About 30% of pediatric

TE D

ependymomas are diagnosed in children younger than three years of age (1). Ependymomas arise from ependymal cells and mostly occurs in the posterior fossa (2, 3). They are classified as subependymomas (WHO grade I), ependymomas (WHO grade II) and anaplastic

EP

ependymomas (WHO grade III). Posterior fossa ependymomas comprise 50% of adult ependymoma cases (1, 4). Despite the histological similarity of ependymomas, they are very

AC C

heterogeneous tumors. In fact, transcriptional profiling of two large independent cohorts of ependymoma reveals the existence of two demographically, transcriptionally, genetically, and clinically distinct groups of posterior fossa ependymomas (5). Group A patients are younger, have laterally located tumors with a balanced genome, and are much more likely to exhibit recurrence, metastasis at recurrence, and death compared with Group B patients (5). Loss of chromosome 22 loss was one of the most frequent genomic alterations occurring often in Group B posterior fossa ependymomas and rarely in Group A tumors (5) with upregulation of Laminin alpha-2 (LAMA2) gene in group A and Neural Epidermal Growth Factor Like-2

4

ACCEPTED MANUSCRIPT (NELL2) in group B. Moreover, group A including other biomarker known to be associated with poor outcome such as CHI3L1, TNC, VEGF, EGFR, ERRB4, BIRC5, and S100A6 (5).

The prognostic factors, as well as the pattern of recurrence among adults with ependymomas,

RI PT

remain to be defined. The lack of validated significant prognostic factors can probably be explained by the small number of patients included in reported series, the heterogeneity of the patient population in terms of age (adults and children) and anatomical location

SC

(supratentorial, infratentorial, and spinal), and the very long treatment periods. The treatment paradigm for WHO II posterior fossa ependymomas in adults continues to be controversial,

M AN U

and there are essentially no standard clinical guidelines for treatment. The largest, relatively homogenous clinical series of WHO grade II supra- and infratentorial brain ependymomas in adults comprised of 114 patients from 32 French Neurosurgical Centers who were operated between 1990 to 2004 (6). Most of the ependymomas (80.7%) were located in the posterior

TE D

fossa. Multivariate analysis revealed that overall survival (OS) was associated with the preoperative Karnofsky performance scale (KPS) and extent of surgical resection. These findings are in line with other clinical series of adults with posterior fossa ependymomas (2,

EP

7, 8). Reni et al. conducted a multicenter study of 70 adult patients with ependymomas and reported longer survival for younger patients and an infratentorial tumor location (4).

AC C

Mirzadeh et al. (8) retrospectively summarized the functional recovery of 45 patients with posterior fossa ependymomas in adults. The authors reported that 75.6% of the patients had returned to their preoperative functional status at one year. Several variables, such as large tumor volume (>30 cm3), cystic changes on preoperative magnetic resonance imaging (MRI), a hyperintense T2-weighted MR signal surrounding the resection cavity, and patients requiring CSF diversion, were identified as having worse functional and neurological outcome.

5

ACCEPTED MANUSCRIPT Obviously, the specific molecular subtype is unknown during surgery but may be of importance for the planning of post-operative adjuvant therapy. Because group B tumors are much less likely to recur, patients could be treated less aggressively than Group A patients.

RI PT

Generally, radiation therapy for ependymomas has been a matter of debate for decades. Radiotherapy was not associated with enhanced survival among the study population as a whole in one French neurosurgical cohort (6), but rather only for a subpopulation that

SC

underwent subtotal resection (STR). The same trend was reported by Guyotat et al. (2) in a multicenter retrospective study of 106 patients with infratentorial ependymomas.

M AN U

Interestingly, patients who underwent gross total resection (GTR) followed by radiation therapy had no progression free survival (PFS) or OS benefit, but there was a significant survival benefit in a subgroup of patients who were operated for STR and received postoperative radiation therapy compared to those who did not receive adjuvant treatment.

TE D

There are, however, several retrospective studies based on large and uniform cohorts of patients with infratentorial ependymomas that have shown better disease control for patients after GTR and radiotherapy compared to GTR alone (2, 6, 7). Overall, the existing literature

EP

supports a role for a maximal extent of surgery and the preoperative performance status on determining prognosis in adult low-grade ependymomas. The role of adjuvant radiation

AC C

therapy is still not clear, but it seems to impact favorably on OS and PFS in incompletely resected WHO grade II ependymomas (2). Prospective clinical trials are warranted to assess the actual impact of postoperative radiation therapy in this population.

Pilocytic Astrocytoma Pilocytic astrocytomas (PA) are classified by the WHO as a grade I astrocytic tumors (9). They represent up to 40% of brain tumors in childhood (10), but are rarely found in adults for whom the annual incidence is 4.8 per million (11). Pilocytic astrocytomas in adults tend to be

6

ACCEPTED MANUSCRIPT located in the supratentorial compartment and have unfavorable prognosis. The most

common genomic abnormalities found in sporadic PAs are BRAF fusion genes (12) appeared mostly in the cerebellar PAs and are less frequent at other sites (13). There are limited data on

RI PT

the long-term outcome of patients with PA in the posterior fossa. While some studies suggest that adult patients with PA will have a benign course after initial surgical resection (14), other studies suggest the opposite (15). Ellis et al. (14) reported a cohort of 20 adult patients with

SC

PA for whom there was a 30% recurrence rate. Three out of four patients who were operated for tumor recurrence showed a malignant transformation of their tumor. These results are

M AN U

similar to the findings reported by Stüer et al. (15) who summarized the clinical course of 44 adult patients with PA who were operated in a single institution. The authors also reported a 30% tumor recurrence or disease progression. Specifically, 12 patients had cerebellar PA and seven had brainstem PA: 66% of the former underwent GTR while only 33% of the

TE D

brainstem PA could be completely removed. Adjuvant radiotherapy (54-58 Gy) was given in all cases of primary or secondary anaplastic PA, and four of them received concomitant chemotherapy consisting of a combination of procarbazine, lomustine, and vincristine (PCV).

EP

One of the latter patients who received concomitant chemotherapy, was also treated with temozolomide. Performance status before and after surgery, as measured by KPS, was

AC C

strongly associated with enhanced OS. Complete surgical resection seems to be curative in all patients with PA that are WHO grade I. Nineteen patients reported by Stüer et al. (15) underwent GTR, and all but one (who had been diagnosed as having an anaplastic PA WHO III and experienced tumor recurrence) had no tumor recurrence over a median follow-up period of 55 months. A recently reported series by Wade et al. (16) included ten patients who were operated for tumor resection. Eight patients underwent GTR and two patients underwent STR. After a follow-up period of up to 334 months, only those patients who underwent STR

7

ACCEPTED MANUSCRIPT had tumor recurrence. Interestingly, the recurrence rate was not correlated to the tumor proliferation index, as assessed by Ki67. There are limited data and no randomized studies to guide treatment policy for PA,

RI PT

especially with respect to postsurgical management. Patients are usually either observed expectantly or offered up-front adjuvant treatment. The use of BRAF inhibitors can

hypothetically be treatment option for PA. The benefit of this treatment has yet to be

SC

analyzed in large clinical studies (17). It is accepted to observe patients after GTR, reserving radiation therapy for salvage. Thirteen patients in one series of 30 adults with PA were found

M AN U

to have infratentorial tumors (18). Four patients who had undergone STR and one patient who was post-GTR were treated with radiation therapy. The authors concluded that postoperative radiation therapy had no effect on OS.

TE D

Medulloblastomas

Medulloblastoma is one of the most common brain tumors in children, representing up to 30% of all pediatric CNS tumors. They are rare in adults and account for 1-3% of all

EP

adult primary brain tumors (19, 20). For that reason, most of the literature on medulloblastomas in adults relies on retrospective analyses over several decades, and include

AC C

a wide range of treatment protocols that are usually based on therapies developed for children. It has recently became clear that medulloblastomas in adults have distinct molecular, radiological and clinical properties (20-24). Unlike pediatric medulloblastomas that tend to be located in the vermis, adult tumors tend to involve the cerebellar hemispheres (38). Another difference from pediatric tumors is the greater prevalence of late recurrence. Riffaud et al. (25) reported a recurrence rate of 41% with 4.2 years median time to recurrence (range 0.7 to 18 years). In a phase II prospective study, Brandes et al. (26) reported that the risk for recurrence in adults with medulloblastomas increased markedly after 7 years of

8

ACCEPTED MANUSCRIPT follow-up in low-risk patients. These results suggest that adults with medulloblastomas have

an ongoing lifetime risk for tumor recurrence and can therefore not be considered cured, even after several years of remission (27). Recent advances in molecular biology have allowed the

RI PT

identification of four distinct molecular subgroups. Two well characterized subgroups of medulloblastoma with peculiar pathways: wingless (WNT) subtype, sonic hedgehog (SHH), and the remaining two subgroups groups 3 and 4 (21). Each subgroup have different

SC

cytogenetic, mutational and gene expression signatures, demographics, histology and

prognosis (28, 29). 1) WNT is the least common subgroup, with better prognosis that account

M AN U

for10% of all medulloblastomas (15% in adults) (23). Typically WNT medulloblastomas occur in children over three years of age, and rarely found in adults. The most common genetic alteration WNT medulloblastomas, present in 90% of the cases, is B-catenin 1 gene mutation (CTNNB1) which leads to enhanced activation of MYC and MYCN oncogenes and

TE D

subsequent increase in cellular proliferation (30). Other frequently altered genes are DDX3X (50%), SMARCA4(26%), MLL2 (12%) and TP53 (12%) and loss of chromosome 6q and 17p (54). 2) The SHH subgroup, consisting of 30% of all patients with medulloblastoma,

EP

mainly comprises adults (60%) (23), and infants. The prognosis is intermediate for both age groups. The most commonly found genetic mutations are PTCH1 (28%) and suppression of

AC C

the fusedhomologue (SUFU). Other frequently mutated genes are TP53 (13%), MLL2 (13%), MYCN (8%), LDB1 (7%), GLI2(5%) (29). 3) Group 3 has the worst prognosis, with high tendency to send metastases, very rare in adults and more common in children and infants. The most common genetic alteration in this subgroup is proto-oncogene MYC mutation/amplification and also by an elevated genomic instability with frequent chromosome 1p gains, 10q and 5q loss and the presence of iso chromosome 17 (29). 4) Group 4 is the largest subgroup accounting for 35%of all medulloblastomas with intermediate

9

ACCEPTED MANUSCRIPT

prognosis (29). Mostly occurs in adults (20–25% of all adult medulloblastomas) and children (35%). Group 4 is characterized by MYCN and CDK6 amplifications (29). The initial treatment paradigm for adults with newly-diagnosed medulloblastomas

RI PT

includes maximal safe resection followed by postoperative radiotherapy (31) and possibly adjuvant chemotherapy (24, 27). There is growing evidence that correlates patient outcome with extent of tumor resection (32, 33). Postoperative radiation therapy of 36-40 Gy to the

SC

spinal axis followed by a boost of 54-58 Gy to the posterior fossa is given in most cases. This treatment is associated with a 5-year OS rate of 58-84% (32). Unlike medulloblastoma in

M AN U

childhood, the role of chemotherapy in adults is not well defined and is usually given to highrisk patients. Moreover, there is no consensus regarding the specific regimen to be used. Treatment of patients with disease recurrence is challenging, and there is currently no clear protocol. Although no randomized controlled studies have assessed the role of

TE D

reoperation, retrospective clinical series suggest that, when feasible, patients should undergo repeat resection of the recurrent tumor followed by radiation therapy, high-dose

EP

chemotherapy and stem cell transplantation (34).

Cerebellar Hemangioblastoma

AC C

Hemangioblastomas are benign vascular tumors, classified as WHO grade I tumor, and composed of stromal cells and abundant capillaries (35). Hemangioblastomas can be found in any part of the CNS, but they are most frequently located in the posterior fossa (36). These tumors constitute 2% of all brain tumors, and represent up to 10% of all posterior fossa tumors in adults. Approximately 25% of hemangioblastomas are associated with Von HippelLindau disease (VHL), an autosomal dominant inherited disease characterized by the development of hemangioblastomas of the CNS and retina, clear cell renal carcinoma, pheochromocytoma, pancreatic tumors and endolymphatic sac tumors. The rest of the

10

ACCEPTED MANUSCRIPT hemangioblastomas occur sporadically. The histological appearance of sporadic hemangioblastomas or VHL-associated hemangioblastomas is identical. Previous clinical studies tended to mix sporadic hemangioblastomas with VHL-associated ones. Notably,

RI PT

whereas sporadic hemangioblastomas are solitary in most cases and usually occur in adults, hemangioblastomas in patients with VHL disease tend to be multiple, evolve rapidly, and occur in younger patients (mean age of 29 years). Several clinical series have estimated that

SC

multiple hemangioblastomas will develop in 60%-90% of patients with VHL disease (37, 38), and it is important to bear in mind that systemic manifestation of VHL carries a major

M AN U

deleterious impact on the quality of life of the patients (39).

The radiological appearance of a hemangioblastoma is classified into four types: without associated cysts, with intratumoral associated cysts, with associated peritumoral cysts and those associated with both peri- and intratumoral cysts (40). Peritumoral cysts tend to be

TE D

associated with a more symptomatic clinical course depending on the rate of cyst growth and volume. The mechanism of peritumoral cyst formation is different from the one associated with intratumoral cyst formation. Specifically, an intratumoral cyst results from tumor

EP

necrosis, whereas a peritumoral cyst is the result of surrounding interstitial pressure, increased tumor vascular permeability and tumor surrounding edema (38, 40).

AC C

In general, resection of hemangioblastomas in VHL patients is indicated when the lesion causes symptoms. The treatment for both sporadic and VHL-associated CNS hemangioblastomas is resection. Due to their abundant vascularity, surgical resection of hemangioblastomas may result in significant intraoperative blood loss and preoperative vascular embolization has been rarely reported (41). Resection of symptomatic hemangioblastomas improves or stabilizes the patient’s symptoms and overall outcome. The recurrence rate after removal of sporadic or VHL-associated hemangioblastomas range between 20 to 33% (42). Analysis of 40 CNS hemangioblastomas in 20 patients with VHL

11

ACCEPTED MANUSCRIPT treated with radiosurgery showed no progression in 33% of the patients six years after treatment. This, however, was similar to the natural history of untreated patients with hemangioblastomas (38). Tumor removal leads to resolution of the peritumoral cysts, and

RI PT

removal of the cyst wall is not required in every case. In fact, resection of the tumor will cause the cyst to collapse. However, it is important to carefully inspect the cyst wall intraoperatively in order to detect small tumor foci. Overall, the treatment paradigm of

M AN U

therapies, especially for tumors that may not progress.

SC

symptomatic hemangioblastomas is surgical resection and the avoidance of unnecessary

Gliomas

Gliomas comprise the most common primary tumor in the brain. However, they are most frequently located supratentorially. Few clinical series with limited numbers of patients

TE D

(mostly with infratentorial gliomas) have been reported [25, 26]. Although glioblastoma, a WHO grade IV glioma, is the most common primary brain tumor, cerebellar glioblastomas are rare and comprise only 0.4–3.4% of all intracranial glioblastomas. The largest clinical

EP

series of patients with cerebellar glioblastomas used clinical data from the population-based Surveillance, Epidemiology and End Results (SEER) national database between 1973 to 2009

AC C

(43). Attempted have been made to identify factors associated with survival in this patient population and compared them to those of patients with supratentorial glioblastomas. It appeared that cerebellar glioblastomas tended to occur in younger patients, had smaller tumor size and a shorter OS, with a median survival of eight months. Moreover, the cerebellar location of a glioblastoma independently predicted poor survival compared to other locations (43). In another single-center clinical series of 45 adults with cerebellar glioblastomas, the authors found that preoperative tumor volume, extent of tumor resection, brainstem invasion and further postoperative neuro-oncological treatment were all significantly associated with

12

ACCEPTED MANUSCRIPT

enhanced survival (44). Tsung et al. (45) conducted a retrospective review of 21 patients with cerebellar glioblastoma and also noted that this distinct patient population tended to be young (mean age of 39 years at diagnosis) and that postoperative chemotherapy was associated with

RI PT

extended survival. Interestingly, the extent of resection made no impact on either OS or PFS. This may reflect the small cohort of patients who presented with a mixture of true cerebellar glioma with tumors with brain stem involvement and even with tumors that caused

SC

leptomeningeal spread. Strauss et al. (46) reported a unique subpopulation of 16 patients who were diagnosed with supratentorial gliomas and developed late noncontiguous infratentorial

M AN U

extension of their tumors. In that report, another two patients had concurrent disease in the posterior fossa at the time of diagnosis of their primary supratentorial tumor. All 18 patients had hyperintense lesions on MRI FLAIR-weighted images that were observed as being adjacent or around the fourth ventricle causing a mass effect and displacement of the

TE D

ventricle. Twelve of those patients had high-grade gliomas, and the other six had low-grade gliomas. Most of those posterior fossa tumors were not biopsied, and it is not known whether malignant transformation had occurred. Overall, it was not clear whether this progression

EP

represented secondary expansion of the primary disease (multifocal) or if there actually had been multiple primary tumors (multicentric). It was clear, however, that secondary posterior

AC C

fossa involvement was associated with poor prognosis. Others

Intracranial dermoid cysts are congenital lesions that account for approximately 0.3% of all intracranial lesions (47) occurs most commonly in the posterior fossa (48, 49).These cysts are encapsulated by stratified squamous epithelium and contain dermal derivatives such as hair follicles, sebaceous glands, and sweat glands (47). These lesions typically grow slowly, with most not becoming symptomatic until the 3rd decade of life. Growth of these cysts occurs due to the accumulation of desquamated cell debris from their capsule. These cysts are most

13

ACCEPTED MANUSCRIPT commonly located in the midline and often the cyst's capsule is tightly attached to the

surrounding parenchyma, nerves or vasculature and is difficult to remove (48-51). They can present with CNS infection and cause elevated ICP and also hydrocephalus. Gross total

RI PT

resection when safely feasible, is the mainstay of treatment to prevent further infection episodes.

Epidermoid cysts are benign slow growing congenital neoplasms of the CNS; contain waxy

SC

keratinous material that originated from retained ectodermal remnants (52). They account for approximately 0.5-1% of all intracranial tumors and develop from aberrant ectodermal

M AN U

embryonic tissue in neural groove at the first month of embryonal development. Epidermoid cysts are most commonly located in the cerebellopontine angle (CPA) and para-sellar region. In a recent retrospective analysis of 50 cases of posterior fossa epidermoid cysts surgically treated over a decade the most common presenting symptoms were associated with trigeminal

TE D

neuralgia, hearing loss, gait ataxia and raised intra-cranial pressure. The recurrence rate was much higher among those who underwent STR compared to GTR (53).

EP

Future Perspective

This review summarizes the current data on intra-axial tumors in the posterior fossa in adults.

AC C

The clinical course, as well as the prognostic factors and therapeutic management of this relatively rare and heterogeneous group of patients, still remain inconclusive. Emphasis is placed on the need for obtaining more data on the clinical course of this unique group of patients. This review is intended to help familiarize the readers with these interesting and rare pathologies.

14

ACCEPTED MANUSCRIPT Table 1: Clinical, histopathological and molecular summary of posterior fossa tumors in adults. Pathology

Incidence

Prognostic

Histopathology

Molecular data

Subependymoma, WHO I

Group A: younger, laterally

Ependymomas

2-5%

KPS, EOR,

Ependymoma, WHO II

RI PT

factors

Anaplasticependymoma, WHO III

located tumors, worse prognosis. Upregulation of LAMA2 gene.

Pilocytic astrocytomas

3%

KPS, EOR

M AN U

SC

Group B: Loss of chromosome.

PA

22,upregulation of NELL2 gene. BRAF mutation (>70%)

Anaplastic PA

(PA) 1-3%

EOR

WHO IV:

WNT: rare in adults. CTNNB1

Classic

gene mutation. Better

Nodular desmoplastic

prognosis.

Large cell/anaplastic (LCA)

SHH: mainly in adults,

AC C

EP

TE D

Medulloblastomas

Cerebellar

Group3: worst prognosis, MYC mutation/amplification

mostly in adults. MYCN and CDK6 amplification.

10%

4.4%

mutation.

Group4: intermediate prognosis,

WHO I

von Hippel-Lindau (VHL) germline mutation

hemangioblastomas Gliomas

intermediate prognosis.PTCH1

Age

Low grade glioma, WHO II

IDH mutation, Loss of

KPS

Anaplastic astrocytoma, WHO III

expression mutations of ATRX,

EOR

Glioblastoma, WHO IV

EGFR amplification, TERT mutation, PTEN mutation

15

ACCEPTED MANUSCRIPT References 1.

Korshunov A, Golanov A, Sycheva R, Timirgaz V. 2004. Cancer 100: 1230-7

2.

Guyotat J, Metellus P, Giorgi R, Barrie M, Jouvet A, et al. 2009. Acta Neurochir (Wien) 151: 947-60 McGuire CS, Sainani KL, Fisher PG. 2009. J Neurosurg 110: 725-9

4.

Reni M, Brandes AA, Vavassori V, Cavallo G, Casagrande F, et al. 2004. Cancer 100: 1221-

RI PT

3.

9

Witt H, Mack SC, Ryzhova M, Bender S, Sill M, et al. 2011. Cancer Cell 20: 143-57

6.

Metellus P, Barrie M, Figarella-Branger D, Chinot O, Giorgi R, et al. 2007. Brain 130: 1338-

SC

5.

M AN U

49 7.

Metellus P, Guyotat J, Chinot O, Durand A, Barrie M, et al. 2010. Neuro Oncol 12: 976-84

8.

Mirzadeh Z, Bina R, Kusne Y, Coons SW, Spetzler RF, Sanai N. 2014. J Neurosurg 120: 1063-8

Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, et al. 2007. Acta Neuropathol 114: 97-109

TE D

9.

Schneider JH, Jr., Raffel C, McComb JG. 1992. Neurosurgery 30: 58-62; discussion -3

11.

Burkhard C, Di Patre PL, Schuler D, Schuler G, Yasargil MG, et al. 2003. J Neurosurg 98: 1170-4

13. 14.

Deshmukh H, Yu J, Shaik J, MacDonald TJ, Perry A, et al. 2011. BMC Med Genomics 4: 57

AC C

12.

EP

10.

Collins VP, Jones DT, Giannini C. 2015. Acta Neuropathol 129: 775-88 Ellis JA, Waziri A, Balmaceda C, Canoll P, Bruce JN, Sisti MB. 2009. J Neurooncol 95: 37782

15.

Stuer C, Vilz B, Majores M, Becker A, Schramm J, Simon M. 2007. Cancer 110: 2799-808

16.

Wade A, Hayhurst C, Amato-Watkins A, Lammie A, Leach P. 2013. Acta Neurochir (Wien) 155: 1431-5

17.

Gessi M, Engels AC, Lambert S, Rothamel T, von Hornstein S, et al. 2015. Neuropathol Appl Neurobiol

16

ACCEPTED MANUSCRIPT 18.

Ishkanian A, Laperriere NJ, Xu W, Millar BA, Payne D, et al. 2011. Cancer 117: 4070-9

19.

Abacioglu U, Uzel O, Sengoz M, Turkan S, Ober A. 2002. Int J Radiat Oncol Biol Phys 54: 855-60 Zhang N, Ouyang T, Kang H, Long W, Thomas B, Zhu S. 2015. J Neurooncol 124: 255-64

21.

Brandes AA, Bartolotti M, Marucci G, Ghimenton C, Agati R, et al. 2015. Crit Rev Oncol

RI PT

20.

Hematol 94: 348-59 22.

Korshunov A, Remke M, Werft W, Benner A, Ryzhova M, et al. 2010. J Clin Oncol 28:

23.

SC

3054-60

Remke M, Hielscher T, Northcott PA, Witt H, Ryzhova M, et al. 2011. J Clin Oncol 29:

24.

M AN U

2717-23

Vigneron C, Entz-Werle N, Lutz P, Spiegel A, Jannier S, et al. 2015. Cancer Radiother 19: 347-57; quiz 58-9, 62

Riffaud L, Saikali S, Leray E, Hamlat A, Haegelen C, et al. 2009. J Neurosurg 111: 478-87

26.

Brandes AA, Franceschi E, Tosoni A, Blatt V, Ermani M. 2007. Cancer 110: 2035-41

27.

von Bueren AO, Friedrich C, von Hoff K, Kwiecien R, Muller K, et al. 2015. Eur J Cancer 51: 2434-43

TE D

25.

Cho YJ, Tsherniak A, Tamayo P, Santagata S, Ligon A, et al. 2011. J Clin Oncol 29: 1424-30

29.

Northcott PA, Korshunov A, Witt H, Hielscher T, Eberhart CG, et al. 2011. J Clin Oncol 29: 1408-14

Kongkham PN, Northcott PA, Croul SE, Smith CA, Taylor MD, Rutka JT. 2010. Oncogene

AC C

30.

EP

28.

29: 3017-24

31.

Buglione M, Ghirardelli P, Triggiani L, Pedretti S, Pasinetti N, et al. 2015. Rep Pract Oncol Radiother 20: 284-91

32.

Padovani L, Sunyach MP, Perol D, Mercier C, Alapetite C, et al. 2007. Int J Radiat Oncol Biol Phys 68: 433-40

33.

Roldan G, Brasher P, Vecil G, Senger D, Rewcastle B, et al. 2008. Can J Neurol Sci 35: 2105

17

ACCEPTED MANUSCRIPT 34.

Kostaras X, Easaw JC. 2013. J Neurooncol 115: 1-8

35.

Kleihues P, Burger PC, Scheithauer BW. 1993. Brain Pathol 3: 255-68

36.

Hasso AN, Bell SA, Tadmor R. 1994. Neuroimaging Clin N Am 4: 849-70

37.

Filling-Katz MR, Choyke PL, Oldfield E, Charnas L, Patronas NJ, et al. 1991. Neurology 41:

38.

RI PT

41-6

Lonser RR, Butman JA, Huntoon K, Asthagiri AR, Wu T, et al. 2014. J Neurosurg 120: 1055-62

Lammens CR, Bleiker EM, Verhoef S, Hes FJ, Ausems MG, et al. 2010. Clin Genet 77: 483-

SC

39.

91

Jagannathan J, Lonser RR, Smith R, DeVroom HL, Oldfield EH. 2008. J Neurosurg 108:

M AN U

40.

210-22 41.

Standard SC, Ahuja A, Livingston K, Guterman LR, Hopkins LN. 1994. Surg Neurol 41: 405-10

42.

Conway JE, Chou D, Clatterbuck RE, Brem H, Long DM, Rigamonti D. 2001. Neurosurgery

43.

TE D

48: 55-62; discussion -3

Adams H, Chaichana KL, Avendano J, Liu B, Raza SM, Quinones-Hinojosa A. 2013. World Neurosurg 80: e237-43

Weber DC, Miller RC, Villa S, Hanssens P, Baumert BG, et al. 2006. Int J Radiat Oncol Biol

EP

44.

Phys 66: 179-86

Tsung AJ, Prabhu SS, Lei X, Chern JJ, Benjamin Bekele N, Shonka NA. 2011. J Neurooncol

AC C

45.

105: 555-62

46.

Strauss I, Jonas-Kimchi T, Bokstein F, Blumenthal D, Roth J, et al. 2013. J Neurooncol 115: 401-9

47.

Prabhu K, Daniel RT, Mani S, Chacko AG. 2009. Surg Neurol 72: 717-21; discussion 21

48.

Chu W, Feng H, Zhu G, Ye X, Lin J. 2008. Surg Neurol 70: 531-5

49.

Orakcioglu B, Halatsch ME, Fortunati M, Unterberg A, Yonekawa Y. 2008. Acta Neurochir (Wien) 150: 1227-34; discussion 34

18

ACCEPTED MANUSCRIPT 50.

Karagoz Guzey F, Bas NS, Sencer A, Emel E, Hamamcioglu MK, et al. 2007. Pediatr Neurosurg 43: 323-6

51.

Vinchon M, Soto-Ares G, Assaker R, Belbachir F, Dhellemmes P. 2001. Pediatr Neurosurg 34: 255-63 Chon KH, Lee JM, Koh EJ, Choi HY. 2012. J Korean Neurosurg Soc 52: 148-51

53.

Gopalakrishnan CV, Ansari KA, Nair S, Menon G. 2014. Clin Neurol Neurosurg 117: 93-9

AC C

EP

TE D

M AN U

SC

RI PT

52.

ACCEPTED MANUSCRIPT

WHO=world health organization LAMA2=Laminin alpha-2 NELL2=Neural Epidermal Growth Factor Like-2

RI PT

OS=overall survival KPS=Karnofsky performance scale MRI= magnetic resonance imaging

SC

STR= subtotal resection PFS=progression free survival

M AN U

PA= Pilocytic astrocytomas PCV= procarbazine, lomustine, and vincristine WNT=wingless SHH=subtype, sonic hedgehog CTNNB1=B-catenin 1 gene mutation

TE D

VHL= Von Hippel-Lindau

SEER= Surveillance, Epidemiology and End Results

AC C

EP

CPA= cerebellopontine angle

ACCEPTED MANUSCRIPT Conflict of interests

AC C

EP

TE D

M AN U

SC

RI PT

The authors report no conflict of interest