REVIEW ARTICLE

published: 22 July 2014

doi: 10.3389/fonc.2014.00176

Pediatric medulloblastoma – update on molecular

classification driving targeted therapies
Ruth-Mary DeSouza 1 , Benjamin R. T. Jones 2 , Stephen P. Lowis 3 and Kathreena M. Kurian 4 *
1
Neurosurgery, King’s College Hospital London, London, UK
2
Royal Bolton Hospital, Bolton, UK
3
BMT/Oncology, University of Bristol, Bristol, UK
4
Brain Tumour Group, Institute of Clinical Neuroscience, University of Bristol, Bristol, UK

Edited by: As advances in the molecular and genetic profiling of pediatric medulloblastoma evolve,
Arnab Chakravarti, The Ohio State
associations with prognosis and treatment are found (prognostic and predictive biomark-
University Medical Center, USA
ers) and research is directed at molecular therapies. Medulloblastoma typically affects
Reviewed by:
Kamalakannan Palanichamy, The Ohio young patients, where the implications of any treatment on the developing brain must be
State University Medical Center, USA carefully considered. The aim of this article is to provide a clear comprehensible update
Timothy A. Chan, Memorial Sloan on the role molecular profiling and subgroups in pediatric medulloblastoma as it is likely
Kettering Cancer Center, USA
to contribute significantly toward prognostication. Knowledge of this classification is of
*Correspondence:
particular interest because there are new molecular therapies targeting the Shh subgroup
Kathreena M. Kurian, Department of
Neuropathology, Southmead of medulloblastomas.
Hospital, Southmead Road, Bristol Keywords: pediatric, medulloblastoma, molecular, therapies, classification
BS10 5NB, UK
e-mail: [email protected]

INTRODUCTION intra-operatively and on the presence of metastases (10). The

Brain tumors are the most common solid malignancies in chil- Chang system is no longer used, although elements of it form
dren, and among these medulloblastoma is the most frequent (1). the current clinical risk stratification of medulloblastoma (10).
The incidence of medulloblastoma is higher in males and higher Currently, medulloblastoma is classified clinically into high risk
in early childhood, with almost half occurring before the age of and standard (average) risk, which is summarized in Table 1. The
5 (1). At least 75% of childhood medulloblastomas arise in the factors contributing to this classification are solely clinical – age,
cerebellar vermis, and project into the fourth ventricle, with the metastases, and resection (3). Age is a key factor, which may reflect
remainder involving the cerebellar hemispheres (see Figure 1) (1). in part the aggressive natural history of tumors in the under-
Spread may be via CSF and present in up to one-third of cases at three age group and also reflect the limitations and side effects of
presentation (2). therapy (3).
The clinical features of medulloblastoma, as with other poste-
rior fossa pathology, can be difficult to detect initially in young HISTOLOGICAL CLASSIFICATION
children, sometimes leading to a delayed diagnosis (3). Symptoms The World Health Organisation (WHO) classification system 2007
include headache, general malaise, failure to feed, vomiting, clum- for medulloblastomas uses histology to classify medulloblastomas,
siness, and other presentations that mimic common and benign which can be considered three major groups including the classic
childhood pathologies seen in primary care (4). Typically, the subtype; desmoplastic/nodular/medulloblastoma with extensive
treatment strategies for medulloblastoma are threefold: maximal nodularity (MBEN) subtypes; and large cell/anaplastic medul-
safe resection (plus/minus CSF diversion), neuraxis radiotherapy, loblastoma subtypes (Figures 1 and 2). Classic medulloblastoma
and chemotherapy (4). represents the most common histological subtype (66%) (1), and
Survival in children with medulloblastoma has improved over is composed of sheets of densely packed small round blue cells
the last 20 years, and the quality of life in medulloblastoma sur- (basophilic) with a high nuclear to cytoplasmic ratio, mitotic and
vivors has been evaluated in terms of physical and non-physical apoptotic activity, and may occur in the midline (1).
(5). Physical impairments include neurological deficits, secondary Desmoplastic/nodular medulloblastomas/MBEN (15%) typ-
malignancy, and endocrine dysfunction, whereas non-physical ically carry a favorable prognosis, and may arise laterally in
deficits include cognitive difficulties and psychological and social a cerebellar hemisphere (1, 6). Desmoplastic medulloblastomas
problems (6). The effect of these problems can be far reach- also comprise small round blue tumor cells, but typically harbor
ing, affecting employment and family life (7–9). Many long-term reticulin-free “pale islands” within a reticulin-rich stroma, which
sequelae are secondary to radiotherapy and one of the goals of are often immunopositive for synaptophysin indicating neuronal
modern therapy is to minimize or avoid radiotherapy (7–9). differentiation (1, 4, 6). Anaplastic medulloblastomas (15%) are
characterized by marked nuclear pleomorphism, nuclear mold-
CLINICAL CLASSIFICATION ing, and cell–cell wrapping (1) and the large cell variant (2–4%)
Medulloblastoma was classified clinically by Chang in 1969 displays a monomorphous population of large cells whose nuclei
based on the size and invasiveness of the tumor as determined exhibit prominent nucleoli (1, 7). Both variants are characterized

www.frontiersin.org July 2014 | Volume 4 | Article 176 | 1

DeSouza et al. Pediatric medulloblastoma

FIGURE 1 | MRI of head showing sagittal and horizontal views. Sagittal

view shows a midline posterior fossa medulloblastoma with intermediate
signal intensity. There is an obstruction to the flow of CSF, marked
hydrocephalus, and edema. Horizontal view shows a homogenous
enhancing medulloblastoma arising from the right cerebellar hemisphere
with displacement of the vermis.

FIGURE 2 | Graph showing the age distribution for different subgroups

of medulloblastoma adapted from Ref. (8).
Table 1 | Established prognostic variables accepted by the North
American Children’s Oncology Group (COG) and the SIOP
(International Society of Pediatric Oncology) Group. clinical outcome (8). For a detailed comprehensive review on the
molecular subgroups of medulloblastoma, see the consensus paper
Risk classification Characteristics
by Taylor et al. (5).
Standard-risk tumor ≥3 years of age without evidence of metastatic
spread and having ≤1.5 cm2 (maximum cross-
Wnt MEDULLOBLASTOMAS
sectional area) of residual disease after surgery
Wnt tumors are thought to be the rarest subgroup of medul-
loblastoma, accounting for 11% (9), but they have probably been
High-risk tumor ≥3 years of age with evidence of CSF spread the most studied and have a very good long-term prognosis with
(M1–M3) and/or those with less complete overall survivals reaching 90% (18) (Figure 4). Wnt tumors also
resection (≥1.5 cm2 ) or <3 years of age at diagnosis show a specific age distribution being almost absent in infants
(aged <4 years) (see Figure 2) but predominantly affecting chil-
dren with a peak incidence of 10–12 years (see Figure 2) (9).
by a very high proliferative activity, abundant apoptosis, and a Wingless (Wnt) is a family of growth factor receptors that are
much poorer prognosis (1, 11, 12). involved in embryogenesis and also in cell–cell control mecha-
The majority of medulloblastomas exhibit neuronal differen- nisms (9). Wnt tumors are thought to arise from mossy-fiber
tiation in the form of immunoreactivity to synaptophysin and neuron precursors, which may be involved in the formation of
some also display focal glial differentiation (Glial fibrillary acidic synapses in the developing cerebellum (19). The majority of Wnt
protein (GFAP) immunopositivity) (1, 7, 8). Rare examples show medulloblastomas show classic histology, however rarely, they are
myogenic differentiation (medullomyoblastoma) or melanotic phenotypically large cell/anaplastic (1) and may remarkably retain
differentiation (1, 7, 8). their relatively good prognosis with this phenotype (14). Molecu-
lar analysis of sporadic Wnt medulloblastomas commonly shows
MOLECULAR SUBGROUPS CTNNB1 mutations, which encode β-catenin (see Figure 1) (14).
More recently a consensus conference in Boston in 2010 sup- Moreover, germline mutations of the Wnt pathway inhibitor APC
ported classification of four main subgroups of medulloblastomas predispose to Turcot syndrome in which medulloblastomas may
based on the molecular profiling (7, 8, 13–17) (Figure 3). The occur (7). Other less common mutations are found in sporadic
Wnt and Shh groups were named after the predominant sig- medulloblastomas, including APC, AXIN1, and AXIN2, which are
naling pathways thought to be affected in their pathogenesis. also keys to this pathway (14). A recent paper has also identi-
Less is known currently regarding the pathogenesis of groups 3 fied mutations in the RNA helicase DDX3X, which potentiates
(tending to harbor MYC amplification) and 4 (tending to have transactivation of a TCF promoter, which is further downstream
isochromosome 17q) and therefore generic names were chosen (15). Most mutations result in over-activation of the Wnt signal-
until it is better understood (8). The Shh group has become of ing pathway with increased nuclear (as opposed to cytoplasmic)
increasing interest because of the availability and temporary suc- immunohistochemical staining for β-catenin, which can be rel-
cess of small molecule inhibitors to smoothened (SMO), which atively easily identified by neuropathologists (15). Stimulation
is part of the Shh pathway. All four groups show relatively dis- of Wnt signaling results in nuclear accumulation of β-catenin
tinct variation in demographics, histology, genetic profile, and which complexes to TCF-4/lef-1 and functions in cell division

Frontiers in Oncology | Neuro-Oncology July 2014 | Volume 4 | Article 176 | 2

DeSouza et al. Pediatric medulloblastoma

FIGURE 3 | Histology showing (A) Classic medulloblastoma with nuclear β-catenin immunostaining; (B) Nodular medulloblastoma with cytoplasmic
β-catenin immunostaining; (C) Anaplastic medulloblastoma with cytoplasmic β-catenin immunostaining.

and proliferation (transcribes c-myc and cyclin D1), breakdown Shh MEDULLOBLASTOMAS
of the extracellular matrix, as well as cell–cell adhesion (20). Shh tumors are thought to account for 28% of all medul-
Interaction between the PI3K/Akt and Wnt pathways occurs loblastomas (7) (Figure 5). They have an intermediate prog-
in medulloblastomas and this appears to be crucial for tumor nosis between good prognosis Wnt tumors and poor prognosis
survival (20). group 3 tumors, and may be similar in prognosis to group 4
Wnt medulloblastomas appear to be associated with the loss of (5, 20). Shh medulloblastomas show a dichotomous age distri-
chromosome 6 and interestingly, they rarely express chromosome bution being more common in both infants (<4 years) and adults
17 aberrations which are the most common chromosomal alter- (>16 years) (see Figure 2) (20). Aberrant Shh signaling in nor-
ations detected in other medulloblastoma subgroups, particularly mal human development can cause holoprosencephaly, a disorder
groups 3 and 4 (20). Wnt medulloblastomas also have high levels which affects the midline of the face and nervous system, and there
of expression of MYC (5). A recent paper showed that mutations is an increased risk of infant medulloblastoma in Gorlin syndrome,
in CTNNB1 disrupt the normal differentiation and migration of which have germline mutations in PTCH, the Shh receptors (21).
progenitor cells on the dorsal brainstem, resulting in the accumu- The sonic hedgehog (Shh) pathway plays a key role in normal
lation of aberrant cell collections, which may relate to their midline cerebellar development where it induces proliferation of neuronal
origins (19). precursor cells in the developing cerebellum and other tissues (22).

www.frontiersin.org July 2014 | Volume 4 | Article 176 | 3

DeSouza et al. Pediatric medulloblastoma

FIGURE 4 | Schematic overviewing Wnt signaling.

FIGURE 5 | Schematic overviewing Shh signaling.

The Shh ligand is normally secreted by Purkinje neurons and pro- indicating that there may be a common therapeutic target (26,
motes formation of the external germinal layer from the upper 28–31). Shh have high levels of expression of MYCN (5) Similarly,
rhombic lip (21), and Shh tumors are thought to arise from the both of these subtypes show over-expression of genes involved in
granule neuron precursor cells (23). Notch and Platelet-derived Growth Factor (PDGF) signaling (9).
Desmoplastic/nodular and MBEN are almost exclusively asso- Shh medulloblastomas appear to be almost exclusively associated
ciated with Shh pathway activation, although pathway activa- with deletions of chromosome 9q, which is also the location of
tion is also observed in classic and large cell/anaplastic tumors the PTCH1 gene (9q22) (8). While Shh tumors have been largely
(24, 25). Molecular analysis of sporadic medulloblastomas com- identified using transcriptional profiling, immunohistochemistry
monly shows Patched-1 (PTCH1) mutations, although mutations using, SFRP1, GLI1, and GAB1 have been proposed (5, 18).
in SMO and Suppressor of Fused (SUFU) have been described
(23–25). All mutations result in over-activation of the Shh signal- GROUP 3 MEDULLOBLASTOMAS
ing pathway. Binding of Shh to its receptor PTCH1 relieves tonic Group 3 tumors account for 28% of all medulloblastomas, and
inhibition of SMO and allows release of the Gli family of tran- conceptually it may be convenient to consider them as being
scription factors from inhibitory protein complexes, such as SUFU associated with MYC amplification (not MYCN) but not exclu-
(26). Activation of the hedgehog pathway leads to an increase sively (5, 7, 16, 32). They are currently detected by transcriptional
in Snail protein expression and a resultant decrease in cell–cell profiling, although immunohistochemistry for NPR3 has been
adhesion (26). Hedgehog signaling also appears to be a crucial proposed (7, 16, 32). They are associated with the worst progno-
regulator of angiogenesis and thus metastasis (27). Interestingly, sis of all the subgroups and are frequently metastatic (8). Group
there seems to be some overlap between Wnt and Shh signaling 3 tumors are found in infants and children but very rarely in

Frontiers in Oncology | Neuro-Oncology July 2014 | Volume 4 | Article 176 | 4

DeSouza et al. Pediatric medulloblastoma

adults (see Figure 2) (5). Similar to group 4 tumors, relatively lit- varied with high-risk groups receiving higher-dose multimodal
tle is known about the molecular pathogenesis of group 3 tumors chemotherapy protocols in addition to craniospinal radiation
and they are grouped according to similar transcription pro- (2, 37–40). There are substantial concerns, however, over the
files (7). Group 3 medulloblastomas are mostly classic or large long-term neurocognitive sequelae of whole brain radiation on
cell/anaplastic morphology (5). MYC amplification appears to be the developing brain meaning that in patients younger than the
highly associated with group 3 tumors and is associated with a age of 3 years (or sometimes as old as 7 years) craniospinal radi-
worse prognosis (14). It has been proposed that group 3 medul- ation is often delayed or eliminated (3, 38–44). Many trials are
loblastomas are further categorized in to 3α and 3β based on the currently being undertaken, which are aimed at optimizing the
expression of Myc (5) with 3α tumors have MYC amplification doses and drugs used in chemotherapy regimes in children to
and hence carry a worse prognosis (5) and 3β not over-expressing achieve maximum effect, however, these will not be covered in this
MYC and have a similar prognosis to group 4 tumors – underlying review.
the fluid nature of these classifications at present (5). There is a
possible role of the developmentally regulated transcription factor TARGETING THE Wnt PATHWAY
OTX2 in the development of groups 3 and 4 medulloblastomas In contrast to the Shh pathway, relatively few drugs have been
(33). Interestingly, OTX2 had been shown to transcriptionally up- developed, which specifically target the Wnt pathway. The reason
regulate the oncogene Myc (33). 26% of group 3 tumors have for this may be the inherent challenges of targeting the Wnt path-
isochromosome 17q, however, group 3 tumors are much more way (20, 45–47). While the Frizzled receptor would make a possi-
likely than group 4 tumors to show gain of chromosome 1q and/or ble target, the majority of mutations in medulloblastomas occur
loss of chromosome 5q and chromosome 10q (5). downstream of this by mutations in CTNNB1, which encodes
for β-catenin (20, 48, 49). Recently, a group found a naturally
GROUP 4 MEDULLOBLASTOMAS occurring compound in beetles termed cantharidin (derivative
Group 4 medulloblastomas are thought to be the most common norcantharidin) which blocked Wnt signaling in vitro and reduced
“typical” subgroup of medulloblastoma, accounting for around the size of intracranial tumors in a mouse model in vivo (50).
34% (5), and can be thought of conceptually as being associated Cantharidin and norcantharidin are known to inhibit protein
with isochromosome 17q (5). Group 4 medulloblastomas rarely phosphatases 1 and 2A (PP1 and PP2A) (51). It is thought that
affect infants (0–3 years) and mainly affect children, with a peak PP2A is required for Wnt mediated β-catenin stabilization down-
age of 10 years (see Figure 2) (5). They are also currently detected stream of the Wnt ligand (51). This represents a possible model for
by transcriptional profiling, although immunohistochemistry for the development of synthetic derivatives although more research
KCNA has been proposed and is awaiting validation (5). is needed. This may also represent a possible treatment for other
Although they frequently metastasize, they still have an inter- types of medulloblastoma because activated Wnt signals interact
mediate prognosis, compared with the poor prognosis of group with other signaling pathways (20, 26, 45, 49).
3 (4, 13, 16, 18). The vast majority of group 4 medulloblastomas
have a classic histology, although less frequently they can have a TARGETING THE SONIC HEDGEHOG PATHWAY
large cell/anaplastic morphology (5). Almost two-thirds of group Since the discovery of a naturally occurring hedgehog pathway
4 medulloblastomas have an isochromosome 17q (i17q) though inhibitor cyclopamine, a number of cyclopamine derivatives have
occasionally isolated 17p deletions are seen (7, 8). Isochromosome been developed with increased potency and bioavailability (22,
17q and 17p mutations are also observed in some group 3 medul- 52–57). Similar to cyclopamine, these drugs act by inhibition of
loblastomas though less frequently (5). Group 4 medulloblastomas SMO (23, 57, 58). The most studied of these analogs is Vismodegib
are associated with CDK6 and MYCN amplification but minimal (GDC-0449), however, much of the research have been conducted
MYC over-expression (8). Interestingly, chromosome X loss is seen in basal cell carcinomas (BCCs) rather than medulloblastomas
in 80% of females with group 4 medulloblastomas (8) Groups 3 because an increased prevalence of BCCs and a similar molecu-
and 4 medulloblastomas have recently been shown to have EZH2 lar pathogenesis (23). Unfortunately, a number of tumors treated
and KDM6A alterations which are involved in histone methylation with Vismodegib later acquired resistance by de novo mutations,
(specifically H3K27) (34, 35). specifically D473H point mutations (23). A new drug, Saridegib
Other histone methylases/acteylases, such as HDAC5, HDAC9, (IPI-926), showed decreased drug resistance in mouse models, and
MLL2, and MLL3, have also been found to be over-expressed may be of future interest (28). Other SMO inhibitors have been
in medulloblastomas, but in these studies the authors did not developed but many of these have not entered clinical trials (23–25,
investigate their prevalence in individual subgroups (34, 36). Inter- 28, 29, 57–61).
estingly, the HDAC5 gene locus is located on chromosome 17q, Interestingly, new in vitro studies have highlighted a role for
which is commonly amplified in group 4 tumors (34, 36). arsenic compounds in the treatment of hedgehog-driven cancers
through a different method to other drugs (29). Arsenic com-
THERAPEUTIC TARGETING OF MEDULLOBLASTOMAS pounds appear to disrupt tumorigenesis by targeting GLI, which
Current treatment strategies for medulloblastoma are developed are hedgehog signaling pathway components downstream of SMO
based on the risk stratification and age of the patient (see Table 1). and PTCH1 and so may be useful in tumors resistant to treatment
In all subgroups of patients, surgery is first line treatment, which with SMO inhibitors (29). Two other drugs, HPI 1 and 4, have also
aims for maximal tumor resection. Postsurgical treatment is then been found to inhibit proliferation of cerebellar granule neuron

www.frontiersin.org July 2014 | Volume 4 | Article 176 | 5

DeSouza et al. Pediatric medulloblastoma

precursors, which expressed an oncogenic form of SMO that was 11. Haberler C, Slavc I, Czech T, Gelpi E, Heinzl H, Budka H, et al. Histopatho-
resistant to cyclopamine (57). logical prognostic factors in medulloblastoma: high expression of survivin
is related to unfavourable outcome. Eur J Cancer (2006) 42(17):2996–3003.
TARGETING GROUPS 3 AND 4 MEDULLOBLASTOMAS doi:10.1016/j.ejca.2006.05.038
12. Giangaspero F, Wellek S, Masuoka J, Gessi M, Kleihues P, Ohgaki H. Stratification
Owing to the relative paucity of information on the underlying
of medulloblastoma on the basis of histopathological grading. Acta Neuropathol
pathophysiology of groups 3 and 4 medulloblastomas, no specific (2006) 112(1):5–12. doi:10.1007/s00401-006-0064-x
treatments have been developed to target them as yet. The histone 13. Robinson G, Parker M, Kranenburg TA, Lu C, Chen X, Ding L, et al.
methylases EZH2 and KDM6A represent possible future targets Novel mutations target distinct subgroups of medulloblastoma. Nature (2012)
of these subgroups as they appear to be exclusively expressed in 488(7409):43–8. doi:10.1038/nature11213
14. Ellison DW, Kocak M, Dalton J, Megahed H, Lusher ME, Ryan SL, et al. Defin-
groups 3 and 4 medulloblastomas (34, 35, 61). Similarly, there may ition of disease-risk stratification groups in childhood medulloblastoma using
be a role for histone deacetylase inhibitors, considering the possi- combined clinical, pathologic, and molecular variables. J Clin Oncol (2011)
ble amplification of HDAC5 in association with i17q (35). Some 29:1400–7. doi:10.1200/JCO.2010.30.2810
preclinical studies with histone deacetylase inhibitors have already 15. Pugh TJ, Weeraratne SD, Archer TC, Pomeranz Krummel DA, Auclair D, Bochic-
shown some promise in medulloblastomas in vitro, however, these chio J, et al. Medulloblastoma exome sequencing uncovers subtype-specific
somatic mutations. Nat Lett. Available from: http://www.nature.com/nature/
were not associated with molecular profiling of the tumors and journal/vaop/ncurrent/full/nature11329.html
identification of subgroups (35). 16. MAGIC – Medulloblastoma Advanced. Genomics International Consortium
Stratifying and Targeting Pediatric Medulloblastoma through Genomics. Available
CONCLUSION from: http://www.bcgsc.ca/project/magic
The identification of different molecular pathways involved in the 17. Cho YJ, Tsherniak A, Tamayo P, Santagata S, Ligon A, Greulich H, et al.
pathogenesis of medulloblastomas provides exciting new thera- Integrative genomic analysis of medulloblastoma identifies a molecular sub-
group that drives poor clinical outcome. J Clin Oncol (2011) 29:1424–30.
peutic targets for the development of new drugs with reduced side
doi:10.1200/JCO.2010.28.5148
effects (62–70). Although this classification provides a simplified 18. Ellison DW, Dalton J, Kocak M, Nicholson SL, Fraga C, Neale G, et al. Medul-
molecular schema for subdividing tumors, it does not take into loblastoma: clinicopathological correlates of SHH, WNT, and non-SHH/WNT
account emerging work around molecular heterogeneity within molecular subgroups. Acta Neuropathol (2011) 121(3):381–96. doi:10.1007/
tumors, which may become increasingly important as different s00401-011-0800-8
19. Gibson P, Tong Y, Robinson G, Thompson MC, Currle DS, Eden C, et al. Sub-
molecular therapies enter clinical use. Moreover, in the future a types of medulloblastoma have distinct developmental origins. Nature (2010)
systems biology approach may be relevant when considering these 468:1095–9. doi:10.1038/nature09587
complex and intercommunicating pathways. 20. Baryawno N, Sveinbjörnsson B, Eksborg S, Chen CS, Kogner P,
Johnsen JI. Small-molecule inhibitors of phosphatidylinositol 3-kinase/Akt
REFERENCES signaling inhibit Wnt/beta-catenin pathway cross-talk and suppress medul-
1. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, editors. WHO Classification of loblastoma growth. Cancer Res (2010) 70(1):266–76. doi:10.1158/0008-5472.
Tumours of the Central Nervous System. Lyon: IARC (2007). CAN-09-0578
2. Pizer BL, Clifford SC. The potential impact of tumour biology on improved clin- 21. Chiang C, Litingtung Y, Lee E, Young KE, Corden JL, Westphal H, et al. Cyclopia
ical practice formedulloblastoma: progress towards biologically driven clinical and defective axial patterning in mice lacking Sonic hedgehog gene function.
trials. Br J Neurosurg (2009) 23(4):364–75. doi:10.1080/02688690903121807 Nature (1996) 383:407–13. doi:10.1038/383407a0
3. Wilne S, Collier J, Kennedy C, Jenkins A, Grout J, Mackie S, et al. Progression 22. Mimeault M, Batra SK. Frequent deregulations in the hedgehog signaling
from first symptom to diagnosis in childhood brain tumors. Eur J Pediatr (2012) network and cross-talks with the epidermal growth factor receptor pathway
171(1):87–93. doi:10.1007/s00431-011-1485-7 involved in cancer progression and targeted therapies. Pharmacol Rev (2010)
4. Lamont JM, McManamy CS, Pearson AD, Clifford SC, Ellison DW. Combined 62(3):497–524. doi:10.1124/pr.109.002329
histopathological and molecular cytogenetic stratification of medulloblastoma 23. Von Hoff DD, LoRusso PM, Rudin CM, Reddy JC, Yauch RL, Tibes R, et al. Inhi-
patients. Clin Cancer Res (2004) 10:5482–93. doi:10.1158/1078-0432.CCR-03- bition of the hedgehog pathway in advanced basal-cell carcinoma. N Engl J Med
0721 (2009) 361(12):1164–72. doi:10.1056/NEJMoa0905360
5. Taylor MD, Northcott PA, Korshunov A, Remke M, Cho YJ, Clifford SC, et al. 24. Rudin CM, Hann CL, Laterra J, Yauch RL, Callahan CA, Fu L, et al. Treatment
Molecular subgroups of medulloblastoma: the current consensus. Acta Neu- of medulloblastoma with hedgehog pathway inhibitor GDC-0449. N Engl J Med
ropathol (2012) 123:465–72. doi:10.1007/s00401-011-0922-z (2009) 361(12):1173–8. doi:10.1056/NEJMoa0902903
6. Giangaspero F, Perilongo G, Fondelli MP, Brisigotti M, Carollo C, Burnelli R, 25. Yauch RL, Dijkgraaf GJ, Alicke B, Januario T, Ahn CP, Holcomb T,
et al. Medulloblastoma with extensive nodularity: a variant with favorable prog- et al. Smoothened mutation confers resistance to a Hedgehog pathway
nosis. J Neurosurg (1999) 91:971–7. doi:10.3171/jns.1999.91.6.0971 inhibitor in medulloblastoma. Science (2009) 326:572–4. doi:10.1126/science.
7. Kool M, Korshunov A, Remke M, Jones DT, Schlanstein M, Northcott PA, et al. 1179386
Molecular subgroups of medulloblastoma: an international meta-analysis of 26. Li X, Deng W, Lobo-Ruppert SM, Ruppert JM. Gli1 acts through Snail
transcriptome, genetic aberrations, and clinical data of WNT, SHH, Group and E-cadherin to promote nuclear signalling by β-catenin. Oncogene (2007)
3, and Group 4 medulloblastomas. Acta Neuropathol (2012) 123:473–84. 26:4489–98. doi:10.1038/sj.onc.1210241
doi:10.1007/s00401-012-0958-8 27. Katoh Y, Katoh M. Hedgehog target genes: mechanisms of carcinogenesis
8. Northcott PA, Korshunov A, Witt H, Hielscher T, Eberhart CG, Mack S, et al. induced by aberrant hedgehog signaling activation. Curr Mol Med (2009)
Medulloblastoma comprises four distinct molecular variants. J Clin Oncol 9(7):873–86. doi:10.2174/156652409789105570
(2011) 29:1408–14. doi:10.1200/JCO.2009.27.4324 28. Lee MJ, Hatton BA, Villavicencio EH, Khanna PC, Friedman SD, Ditzler S, et al.
9. Kool M, Koster J, Bunt J, Hasselt NE, Lakeman A, van SluisP, et al. Integrated Hedgehog pathway inhibitor saridegib (IPI-926) increases lifespan in a mouse
genomics identifies five medulloblastoma subtypes with distinct genetic pro- medulloblastoma model. Proc Natl Acad Sci U S A (2012) 109(20):7859–64.
files, pathway signatures and clinicopathological features. PLoS One (2008) doi:10.1073/pnas.1114718109
3(8):e3088. doi:10.1371/journal.pone.0003088 29. Beauchamp EM, Ringer L, Bulut G, Sajwan KP, Hall MD, Lee YC, et al.
10. Chang CH, Housepian EM, Herbert C Jr. An operative staging system and a Arsenic trioxide inhibits human cancer cell growth and tumor development in
megavoltage radiotherapeutic technique for cerebellar medulloblastas. Radiol- mice by blocking Hedgehog/GLI pathway. J Clin Invest (2011) 121(1):148–60.
ogy (1969) 93(6):1351–9. doi:10.1172/JCI42874

Frontiers in Oncology | Neuro-Oncology July 2014 | Volume 4 | Article 176 | 6

DeSouza et al. Pediatric medulloblastoma

30. Leary SE, Zhou T, Holmes E, Geyer JR, Miller DC. Histology predicts a developing cerebellum. Development (2012) 139(10):1724–33. doi:10.1242/dev.
favorable outcome in young children with desmoplastic medulloblastoma: a 050104
report from the children’s oncology group. Cancer (2011) 117(14):3262–7. 48. Silva RD, Marie SK, Uno M, Matushita H, Wakamatsu A, Rosemberg S, et al.
doi:10.1002/cncr.25856 CTNNB1, AXIN1 and APC expression analysis of different medulloblastoma
31. Rieken S, Gaiser T, Mohr A, Welzel T, Witt O, Kulozik AE, et al. Outcome and variants. Clinics (Sao Paulo) (2013) 68(2):167–72. doi:10.6061/clinics/2013(02)
prognostic factors of desmoplastic medulloblastoma treated within a multi- OA08
disciplinary treatment concept. BMC Cancer (2010) 10:450. doi:10.1186/1471- 49. Rogers HA, Sousa S, Salto C, Arenas E, Coyle B, Grundy RG. WNT/β-catenin
2407-10-450 pathway activation in Myc immortalised cerebellar progenitor cells inhibits neu-
32. Eberhart CG, Kratz J, Wang Y, Summers K, Stearns D, Cohen K, et al. Histopatho- ronal differentiation and generates tumors resembling medulloblastoma. Br
logical and molecular prognostic markers in medulloblastoma: c-myc, N-myc, J Cancer (2012) 107(7):1144–52. doi:10.1038/bjc.2012.377
TrkC, and anaplasia. J Neuropathol Exp Neurol (2004) 63(5):441–9. 50. Cimmino F, Scoppettuolo MN, Carotenuto M, De Antonellis P, Dato VD, De
33. Adamson DC, Shi Q, Wortham M, Northcott PA, Di C, Duncan CG, et al. OTX is Vita G, et al. Norcantharidin impairs medulloblastoma growth by inhibition of
critical for the maintenance and progression of classic medulloblastoma. Cancer Wnt/β-catenin signalling. J Neurooncol (2012) 106:59–70. doi:10.1007/s11060-
Res (2010) 70(1):181–91. doi:10.1158/0008-5472.CAN-09-2331 011-0645-y
34. Milde T, Oehme I, Korshunov A, Kopp-Schneider A, Remke M, Northcott P, 51. Yang J, Wu J, Tan C, Klein PS. PP2A:B56epsilon is required for Wnt/beta-
et al. HDAC5 and HDAC9 in medulloblastoma: novel markers for risk strati- catenin signaling during embryonic development. Development (2003)
fication and role in tumour cell growth. Clin Cancer Res (2010) 16:3240–52. 130(23):5569–78. doi:10.1242/dev.00762
doi:10.1158/1078-0432.CCR-10-0395 52. Lee EY, Ji H, Ouyang Z, Zhou B, Ma W, Vokes SA, et al. Hedgehog pathway-
35. Furchert SE, Lanvers-Kaminsky C, Juürgens H, Jung M, Loidl A, Frühwald MC. regulated gene networks in cerebellum development and tumorigenesis. Proc
Inhibitors of histone deacetylases as potential therapeutic tools for high-risk Natl Acad Sci U S A (2010) 107(21):9736–41. doi:10.1073/pnas.1004602107
embryonal tumors of the nervous system of childhood. Int J Cancer (2007) 53. Haldipur P, Bharti U, Govindan S, Sarkar C, Iyengar S, Gressens P, et al. Expres-
120:1787–94. doi:10.1002/ijc.22401 sion of Sonic hedgehog during cell proliferation in the human cerebellum. Stem
36. Parsons DW, Li M, Zhang X, Jones S, Leary RJ, Lin JC, et al. The genetic land- Cells Dev (2012) 21(7):1059–68. doi:10.1089/scd.2011.0206
scape of the childhood cancer medulloblastoma. Science (2011) 331:435–9. 54. Pritchard JI, Olson JM. Methylation of PTCH1, the Patched-1 gene, in a panel
doi:10.1126/science.1198056 of primary medulloblastomas. Cancer Genet Cytogenet (2008) 180(1):47–50.
37. Yuh GE, Loredo LN, Yonemoto LT, Bush DA, Shahnazi K, Preston W, et al. doi:10.1016/j.cancergencyto.2007.09.008
Reducing toxicity from craniospinal irradiation: using proton beams to treat 55. Lee Y, Kawagoe R, Sasai K, Li Y, Russell HR, Curran T, et al. Loss of suppressor-
medulloblastoma in young children. Cancer J (2004) 10(6):386–90. doi:10.1097/ of-fused function promotes tumorigenesis. Oncogene (2007) 26(44):6442–7.
00130404-200411000-00009 doi:10.1038/sj.onc.1210467
38. Taylor RE, Bailey CC, Robinson KJ, Weston CL, Walker DA, Ellison D, et al. 56. Brugières L, Remenieras A, Pierron G, Varlet P, Forget S, Byrde V, et al.
Outcome for patients with metastatic (M2-3) medulloblastoma treated with High frequency of germline SUFU mutations in children with desmoplas-
SIOP/UKCCSG PNET-3 chemotherapy. Eur J Cancer (2005) 41(5):727–34. tic/nodular medulloblastoma younger than 3 years of age. J Clin Oncol (2012)
doi:10.1016/j.ejca.2004.12.017 30(17):2087–93. doi:10.1200/JCO.2011.38.7258
39. Maddrey AM, Bergeron JA, Lombardo ER, McDonald NK, Mulne AF, Baren- 57. Taipale J, Chen JK, Cooper MK, Wang B, Mann RK, Milenkovic L, et al.
berg PD, et al. Neuropsychological performance and quality of life of 10 year Effects of oncogenic mutations in Smoothened and Patched can be reversed
survivors of childhood medulloblastoma. J Neurooncol (2005) 72(3):245–53. by cyclopamine. Nature (2000) 406(6799):1005–9. doi:10.1038/35023008
doi:10.1007/s11060-004-3009-z 58. Drenkhahn SK, Jackson GA, Slusarz A, Starkey NJ, Lubahn DB. Inhibition of
40. von Hoff K, Hinkes B, Gerber NU, Deinlein F, Mittler U, Urban C, et al. Long- Hedgehog/Gli signaling by botanicals: a review of compounds with poten-
term outcome and clinical prognostic factors in children with medulloblastoma tial hedgehog pathway inhibitory activities. Curr Cancer Drug Targets (2013)
treated in the prospective randomised multicentre trial HIT’91. Eur J Cancer 13(5):580–95. doi:10.2174/15680096113139990003
(2009) 45(7):1209–17. doi:10.1016/j.ejca.2009.01.015 59. Del Carratore R, Carpi A, Beffy P, Lubrano V, Giorgetti L, Maserti BE, et al.
41. Frange P, Alapetite C, Gaboriaud G, Bours D, Zucker JM, Zerah M, et al. Itraconazole inhibits HMEC-1 angiogenesis. Biomed Pharmacother (2012)
From childhood to adulthood: long-term outcome of medulloblastoma patients. 66(4):312–7. doi:10.1016/j.biopha.2011.11.004
The Institut Curie experience (1980–2000). J Neurooncol (2009) 95(2):271–9. 60. Dubuc AM, Remke M, Korshunov A, Northcott PA, Zhan SH, Mendez-Lago M,
doi:10.1007/s11060-009-9927-z et al. Aberrant patterns of H3K4 and H3K27 histone lysine methylation occur
42. Ribi K, Relly C, Landolt MA, Alber FD, Boltshauser E, Grotzer MA. Outcome across subgroups in medulloblastoma. Acta Neuropathol (2013) 125(3):373–84.
of medulloblastoma in children: long-term complications and quality of life. doi:10.1007/s00401-012-1070-9
Neuropediatrics (2005) 36(6):357–65. doi:10.1055/s-2005-872880 61. Spiller SE, Ditzler SH, Pullar BJ, Olson JM. Response of preclinical medul-
43. Gajjar A, Chintagumpala M, Ashley D, Kellie S, Kun LE, Merchant TE, et al. loblastoma models to combination therapy with 13-cis retinoic acid and
Risk-adapted craniospinal radiotherapy followed by high-dose chemother- suberoylanilide hydroxamic acid (SAHA). J Neurooncol (2008) 87:133–41.
apy and stem-cell rescue in children with newly diagnosed medulloblas- doi:10.1007/s11060-007-9505-1
toma (St Jude Medulloblastoma-96): long-term results from a prospective, 62. Hyman JM, Firestone AJ, Heine VM, Zhao Y, Ocasio CA, Han K, et al.
multicentre trial. Lancet Oncol (2006) 7(10):813–20. doi:10.1016/S1470- Small-molecule inhibitors reveal multiple strategiesfor Hedgehog pathway
2045(06)70867-1 blockade. Proc Natl Acad Sci U S A (2009) 106(33):14132–7. doi:10.1073/pnas.
44. von Bueren AO, von Hoff K, Pietsch T, Gerber NU, Warmuth-Metz M, Dein- 0907134106
lein F, et al. Treatment of young children with localized medulloblastoma by 63. Results from A Prospective Trial of Proton Radiotherapy for Medulloblastoma:
chemotherapy alone: results of the prospective, multicenter trial HIT 2000 con- Clinical Outcome Including Hearing and Neurocognitive. Available from: http:
firming the prognostic impact of histology. Neuro Oncol (2011) 13(6):669–79. //www.oncolink.org/conferences/article.cfm?id=2185
doi:10.1093/neuonc/nor025 64. Packer RJ, Macdonald T, Vezina G, Keating R, Santi M. Medulloblastoma and
45. Polakis P. Drugging Wnt signalling in cancer. EMBO J (2012) 31:2737–46. primitive neuroectodermal tumors. Handb Clin Neurol (2012) 105:529–48.
doi:10.1038/emboj.2012.126 doi:10.1016/B978-0-444-53502-3.00007-0
46. Lorenz A, Deutschmann M, Ahlfeld J, Prix C, Koch A, Smits R, et al. Severe alter- 65. Clifford SC, Lusher ME, Lindsey JC, Langdon JA, Gilbertson RJ, Straughton D,
ations of cerebellar cortical development after constitutive activation of Wnt et al. Wnt/Wingless pathway activation and chromosome 6 loss characterize a
signaling in granule neuron precursors. Mol Cell Biol (2011) 31(16):3326–38. distinct molecular sub-group of medulloblastomas associated with a favorable
doi:10.1128/MCB.05718-11 prognosis. Cell Cycle (2006) 5(22):2666–70. doi:10.4161/cc.5.22.3446
47. Pei Y, Brun SN, Markant SL, Lento W, Gibson P, Taketo MM, et al. WNT sig- 66. Attard TM, Giglio P, Koppula S, Snyder C, Lynch HT. Brain tumors in
naling increases proliferation and impairs differentiation of stem cells in the individuals with familial adenomatous polyposis: a cancer registry experience

www.frontiersin.org July 2014 | Volume 4 | Article 176 | 7

DeSouza et al. Pediatric medulloblastoma

and pooled case report analysis. Cancer (2007) 109(4):761–6. doi:10.1002/cncr. Conflict of Interest Statement: The authors declare that the research was conducted
22475 in the absence of any commercial or financial relationships that could be construed
67. Zakaria R, Ellenbogen J, Graham C, Pizer B, Mallucci C, Kumar R. A deci- as a potential conflict of interest.
sion analysis tool for the assessment of posterior fossa tumor surgery outcomes
in children-the “Liverpool Neurosurgical Complication Causality Assessment Received: 04 April 2014; accepted: 23 June 2014; published online: 22 July 2014.
Tool”. Childs Nerv Syst (2013) 29(8):1277–83. doi:10.1007/s00381-013-2065-0 Citation: DeSouza R-M, Jones BRT, Lowis SP and Kurian KM (2014) Pediatric medul-
68. Schwalbe EC, Lindsey JC, Straughton D, Hogg TL, Cole M, Megahed H, et al. loblastoma – update on molecular classification driving targeted therapies. Front. Oncol.
Rapid diagnosis of medulloblastoma molecular subgroups. Clin Cancer Res 4:176. doi: 10.3389/fonc.2014.00176
(2011) 17(7):1883–94. doi:10.1158/1078-0432.CCR-10-2210 This article was submitted to Neuro-Oncology, a section of the journal Frontiers in
69. Ramaswamy V, Remke M, Bouffet E, Faria CC, Perreault S, Cho YJ, et al. Recur- Oncology.
rence patterns across medulloblastoma subgroups: an integrated clinical and Copyright © 2014 DeSouza, Jones, Lowis and Kurian. This is an open-access article
molecular analysis. Lancet Oncol (2013) 14(12):1200–7. doi:10.1016/S1470- distributed under the terms of the Creative Commons Attribution License (CC BY).
2045(13)70449-2 The use, distribution or reproduction in other forums is permitted, provided the original
70. Baryawno N, Sveinbjörnsson B, Kogner P, Johnsen JI. Medulloblastoma: a dis- author(s) or licensor are credited and that the original publication in this journal is cited,
ease with disorganized developmental signaling cascades. Cell Cycle (2010) in accordance with accepted academic practice. No use, distribution or reproduction is
9(13):2548–54. doi:10.4161/cc.9.13.12170 permitted which does not comply with these terms.

Frontiers in Oncology | Neuro-Oncology July 2014 | Volume 4 | Article 176 | 8