Rapid Diagnosis of Medulloblastoma Molecular Subgroups

2011;17:1883-1894. Published OnlineFirst February 16, 2011.Clin Cancer Res Ed C. Schwalbe, Janet C. Lindsey, Debbie Straughton, et al. Rapid Diagnosis of Medulloblastoma Molecular Subgroups

Updated version

10.1158/1078-0432.CCR-10-2210doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2011/03/31/1078-0432.CCR-10-2210.DC1.html

Access the most recent supplemental material at:

Cited Articles

http://clincancerres.aacrjournals.org/content/17/7/1883.full.html#ref-list-1

This article cites by 36 articles, 16 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/17/7/1883.full.html#related-urls

This article has been cited by 8 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

[email protected]

To request permission to re-use all or part of this article, contact the AACR Publications Department at

on October 1, 2014. © 2011 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 16, 2011; DOI: 10.1158/1078-0432.CCR-10-2210



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-10-2210

http://clincancerres.aacrjournals.org/content/suppl/2011/03/31/1078-0432.CCR-10-2210.DC1.html

http://clincancerres.aacrjournals.org/content/17/7/1883.full.html#ref-list-1

http://clincancerres.aacrjournals.org/content/17/7/1883.full.html#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

mailto:[email protected]

http://clincancerres.aacrjournals.org/


Imaging, Diagnosis, Prognosis

Rapid Diagnosis of Medulloblastoma Molecular Subgroups

Ed C. Schwalbe1, Janet C. Lindsey1, Debbie Straughton1, Twala L. Hogg2, Michael Cole1,Hisham Megahed1, Sarra L. Ryan1, Meryl E. Lusher1, Michael D. Taylor4, Richard J. Gilbertson2,David W. Ellison3, Simon Bailey1, and Steven C. Clifford1

AbstractPurpose:Microarray studies indicate medulloblastoma comprises distinct molecular disease subgroups,

which offer potential for improved clinical management.

Experimental Design: Minimal mRNA expression signatures diagnostic for the Wnt/Wingless (WNT)

and Sonic Hedgehog (SHH) subgroups were developed, validated, and used to assign subgroup affiliation

in 173 tumors from four independent cohorts, alongside a systematic investigation of subgroup clinical

and molecular characteristics.

Results:WNT tumors [12% (21/173)] were diagnosed >5 years of age (peak, 10 years), displayed classic

histology, CTNNB1 mutation (19/20), and associated chromosome 6 loss, and have previously been

associated with favorable prognosis. SHH cases [24% (42/173)] predominated in infants (<3 years) and

showed an age-dependent relationship to desmoplastic/nodular pathology; all infant desmoplastic/

nodular cases (previously associated with a good outcome) were SHH-positive, but these relationships

broke down in noninfants. PTCH1 mutations were common [34% (11/32)], but PTCH1 exon1c

hypermethylation, chromosome 9q and REN (KCTD11) genetic loss were not SHH associated, and

SMO or SUFU mutation, PTCH1 exon1a or SUFU hypermethylation did not play a role, indicating novel

activating mechanisms in the majority of SHH cases. SHH tumors were associated with an absence of

COL1A2 methylation. WNT/SHH-independent medulloblastomas [64% (110/173)] showed all histolo-

gies, peaked at 3 and 6 years, and were exclusively associated with chromosome 17p loss.

Conclusions: Medulloblastoma subgroups are characterized by distinct genomic, epigenomic and

clinicopathologic features, and clinical outcomes. Validated array-independent gene expression assays for

the rapid assessment of subgroup affiliation in small biopsies provide a basis for their routine clinical

application, in strategies including molecular disease-risk stratification and delivery of targeted therapeu-

tics. Clin Cancer Res; 17(7); 1883–94. �2011 AACR.

Introduction

Medulloblastoma, a primitive neuro-ectodermal tumorof the cerebellum, is the most common malignant braintumor of childhood. Five-year overall survival rates haveincreased over recent years to 70% to 80% for standard-riskpatients. However, for high-risk patients (infants <3 years,

cases with metastatic disease at diagnosis or incompletesurgical resection), current treatments only cure around40% to 60% of cases. In addition, many survivors exhibitlong-term therapy-associated late effects. The developmentof novel targeted treatments, alongside refined patientstratification, will be essential to increase survival ratesand reduce adverse sequelae (1). Improvements in under-standing of themolecular basis of medulloblastoma will befundamental to such advances.

The constitutive activation of developmental signalingpathways plays a key role in medulloblastoma pathogen-esis, and pathway components represent the major muta-tional targets identified in the disease to date. The SonicHedgehog (SHH) pathway plays an essential role in normalcerebellar development, is activated by PTCH1mutation inaround 10% of human primary medulloblastomas, andpromotesmedulloblastomadevelopment inmousemodelsof the disease (2–4). Similarly, mutations in components ofthe canonicalWnt/Wingless (WNT) signaling pathway havebeen described in up to 20% of cases (5–7). Importantly,these pathways appear to have therapeutic significance;WNT-active cases are associated with a favorable prognosis

Authors' Affiliations: 1Northern Institute for Cancer Research, NewcastleUniversity, Newcastle upon Tyne, UK; Departments of 2DevelopmentalNeurobiology and 3Pathology, St. Jude Children's Research Hospital,Memphis, Tennessee; 4Division of Neurosurgery, Arthur and Sonia LabattBrain Tumour Research Centre, The Hospital for Sick Children, Toronto,Ontario, Canada

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Steven C. Clifford, Northern Institute for CancerResearch, Newcastle University, Sir James Spence Institute Level 5, RoyalVictoria Infirmary, Newcastle upon Tyne NE1 4LP, UK. Phone: 44-191-2821319; Fax: 44-191-2821326; E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-10-2210

�2011 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 1883




(>90% overall survival; refs 8 and 9), whereas small mole-cule inhibitors of the SHH pathway show preclinical andearly-clinical activity against the disease (10, 11).

Recent array-based genome-wide genomic and transcrip-tomic investigations in medulloblastoma have identifieddistinct molecular disease subgroups, which are distin-guished by their gene expression profiles, and displayrelated clinical disease features. Two disease groupings,characterized respectively by activation and mutation ofthe WNT and SHH signaling pathways, are consistentlysupported by these studies (12, 13). The WNT subgroup isbest documented, and is distinguished by nuclear b-cateninimmunostaining, CTNNB1mutations, and chromosome 6loss (5, 13–15), alongside its associated favorable prog-nosis (8, 9). The SHH subgroup is, however, less wellcharacterized; PTCH1 mutations are only identified in asubset of SHH cases, indicating a role for other activatingmechanisms and correlates. A series of putative mechan-isms of SHH activation [e.g., PTCH1 hypermethylation,SUFU/SMO mutation, and REN (KCDT11) genetic loss]have been reported in medulloblastoma (16–19), but anyrelationships to the SHH subgroup remain to be estab-lished. Moreover, SHH subgroup clinical features requireclarification; SHH activation has been associated with thedesmoplastic/nodular (DN) disease histologic variant insome studies, but not others, and associations with infantcases have also been postulated (12, 17, 20, 21).

Identification of these medulloblastoma molecular sub-groups has significant potential to (i) improve clinicalmanagement, through molecular disease-risk stratification

strategies and the identification of patients who couldbenefit from SHH and WNT targeted molecular therapeu-tics; and (ii) provide a basis for biological investigations tofurther understand disease molecular pathogenesis and itstherapeutic applications. However, subgroup identificationhas so far relied on advanced genomics (i.e., microarray)technologies, which are relatively expensive and have useddifferent gene expression data collection and analysismethodologies. The development of robust biomarkersand assays for subgroup detection, which can be routinelytested in clinical material across multiple treatment centersand are informative in small tumor biopsies, will thereforebe essential for their future study and any clinical applica-tion. Moreover, studies in modestly sized cohorts havelimited comprehensive investigation of subgroup clinicalcharacteristics and their molecular basis.

In this study,we report thedevelopment andvalidationofminimal mRNA expression signatures, which are robust forthe routine diagnosis of the SHH and WNT subgroups inclinical material and can be applied rapidly and cost-effectively using array-independent methodologies. Usingthese signatures, we show that equivalent disease subgroupsare detected in 4 independent medulloblastoma cohorts,independent of gene expression assay used. We then usethese signatures to assign subgroup affiliation in this largecombined cohort of 173 medulloblastomas and use thesedata as the basis for comprehensive investigations to definethe clinical and molecular characteristics of each subgroup,and their utility for improved disease management.

Materials and Methods

Tissue samplesA representative cohort of 55 medulloblastoma samples

was analyzed, comprising 39 classic, 5 large cell/anaplastic(LCA), and 11 tumors of the DN histopathologic subtype(22), 21 female and 34 male cases, 11 infants (�3 years),41 children (>3–15 years), and 3 adults (�16 years). DNA(n¼ 55) and RNA (n¼ 39) were extracted from these snap-frozen tissues using standard methods. A panel of consti-tutional DNA samples from 100 normal individuals wasobtained from the North Cumbria Community GeneticsProject in the United Kingdom. Research Ethics Committeeapproval has been obtained for the collection, storage, andbiological study of all material.

Mutation screeningAll coding exons of the PTCH1 and SMO genes were PCR

amplified using the primers and conditions shown inSupplementary Table S1. Mutation screening methods forthe SUFU gene have been described previously (23). Muta-tion screening was performed by analysis of PCR productsfor heteroduplex formation, before and after "spiking" withequal amounts of control wild-type DNA using denaturinghigh performance liquid chromatography (Wave DNAFragment Analysis System, Transgenomic) according tothe manufacturer’s instructions. Products detected as con-taining a heteroduplex were sequenced directly on an ABI

Translational Relevance

Understanding the molecular basis of medulloblas-toma will be essential to improve clinical outcomesthrough guidance of molecular risk-adapted adjuvanttherapies and delivery of molecularly targeted agents.Expression microarray studies indicate the existence ofdiscrete medulloblastoma molecular subgroups asso-ciated with activation of specific developmental signal-ing pathways (i.e., SHH, WNT). However, anytranslational utility will require definition of subgroupclinical and molecular correlates in large cohorts, along-side biomarkers and assays for routine subgroup assign-ment before adjuvant therapy selection. We reportdevelopment of diagnostic gene expression signatures,which can be applied rapidly and cost-effectively insmall biopsies, using array-independent methods, toassign subgroup status. Using these signatures in 173medulloblastomas, we demonstrate that disease sub-groups are robust and reproducible and have distinctclinical, molecular, and outcome characteristics oftherapeutic importance. These findings provide strongrationale and methodologies to support subgroupassessment as a basis for future medulloblastoma clin-ical trials and biological investigations.

Schwalbe et al.

Clin Cancer Res; 17(7) April 1, 2011 Clinical Cancer Research1884




377 sequencer (Applied Biosystems). In reported studiesincluding our own, denaturing high-performance liquidchromatography has been reported to identify >90% ofsequence variants (23). Mutation analysis of the CTNNB1and APC genes was performed as previously described (8).

Analysis of promoter methylation statusTwo promoter-associated CpG islands of the PTCH1

gene, spanning exons 1a and 1c (16), were identifiedand characterized using the Emboss CpGPlot website(http://www.ebi.ac.uk/emboss/cpgplot/): 1a methylationstatus was determined by methylation-specific PCR(MSP), and 1c by bisulphite sequencing using previouslypublished primers (16). A CpG island of the SUFU genewas also identified, and its methylation status analyzed byMSP (24). COL1A2 methylation status was assessed bybisulphite sequencing, and has previously been reported(25). Bisulphite treatment, methylated, and unmethylatedcontrols have been described previously (23). Primers andconditions for analysis of PTCH1 CpG island 1a are shownin Supplementary Table S2. Methylation status was desig-nated as methylated or unmethylated, as previouslydescribed (26). For loci assessed by MSP, any sampleshowing a visible PCR product using primers specific forthe methylated sequence was classed as showing evidenceof methylation (i.e., methylated). For loci assessed bybisulphite sequencing, the relative peak intensities at eachCpG residue were determined. Samples in which themethylated peak represented >25% of the total peak heightin greater than 25% of the analyzed CpG sites were classedas showing evidence of methylation (i.e., methylated).

Loss of hetererozygosity analysisLoss of hetererozygosity (LOH) of chromosome regions

9q22.3 and the p-arm of chromosome 17 was analyzedusing polymorphic microsatellite markers D9S1689,D9S1816, D9S287, D9S1809, D9S1786, D9S1851, andD17S2196, D17S936, D17S969, D17S974, D17S1866,D17S1308, respectively (www.ncbi.nlm.nih.gov/genome/sts), as previously described (27). LOH of chromosome 6was analyzed using previously described markers andmethods (14).

A multigene mRNA expression signature to identifySHH and WNT pathway activated tumorsA 13-gene multiplex mRNA expression assay (GeXP;

Beckman Coulter; ref. 28) was developed and used to testtumors for membership of the SHH or WNT medulloblas-toma subgroups in 39 cases. Detailed clinical and patho-logic data for individual cases are summarized in Table 1.Two previously reported independent medulloblastomaexpression microarray data sets (12, 13) were used todesign SHH and WNT subgroup signatures. Data wererespectively downloaded from the St Jude Research website(http: / /www.stjuderesearch.org / site / data /medulloblas-toma) and the Gene Expression Omnibus (GEO; ref. 29).Data were rma normalized (30) using R and Bioconductor.Using t tests, we identified probes differential for the WNT

(group A) or SHH (group B) subgroups, defined by Kooland colleagues (12), which were significant in both datasets. Previous work had validated 3 SHH-associated genes(GLI1, PTCH2, and SFRP1) and 2 WNT-associated genes(DKK2, WIF1) by quantitative reverse transcriptase real-time PCR (13) and these genes were also considered forinclusion in the signatures. Putative signature genes wereassessed for assay suitability, and assays were subsequentlyperformed using 50 ng total RNA per replicate, according tomanufacturer’s instructions. To eliminate the possibility ofdetecting genomic DNA, PCR products were designedacross exon boundaries; products were also designed wherepossible to overlap the Affymetrix probes from which theywere derived. For genes with multiple transcripts, ampli-cons were designed that detected all known transcripts. Testgene expression was normalized to 28S rRNA, and resultsdisplayed are means based on independent assessment intriplicate. The presence of SHH or WNT expression signa-tures was used to identify sample subgroup membership(see next section). Signature gene selection is summarizedin Supplementary Table S3, and PCR primer sequences arelisted in Supplementary Table S4.

Integration and analysis of combined gene expressiondata sets

An additional medulloblastoma expression microarraydata set, reported by Fattet and colleagues (15), wasnormalized as described earlier. All 3 publicly availabledata sets have linked clinical (age at diagnosis, pathologysubtype, and gender) and PTCH1/CTNNB1 mutationaldata available [except PTCH1mutation data not availablefor the Fattet and colleagues (15) data set]. Expressiondata of signature genes from all 3 data sets, together withour own, were integrated. First, our GeXP expression datawere log2 transformed as for the expression microarraydata and, before joining, all data sets were scaled sepa-rately, on a per-gene basis, to give a mean of 0, withvariance of 1. Unsupervised hierarchical clustering, sup-ported by principal component analysis (PCA), was usedto assign subgroup membership status (Fig. 1), and thesedata were used in conjunction with stacked bar plots(Fig. 2) and silhouette plots (ref. 31; SupplementaryFig. S1) to assess performance of the signatures. Therewas some sample overlap between studies: 11 sampleswere assessed by expression array by Thompson andcolleagues (13) and GeXP; 3 samples were assessed byexpression array by Kool and colleagues (12) and GeXP,which enabled the comparison of subgroup assignmentin individual samples using our signatures, when evalu-ated using different gene expression assays. Clinical andgenomic correlates were also combined for the joinedcohort. Duplicate analyses were removed from all corre-lative investigations.

Statistical analysisFisher’s exact and c2 tests were used to identify relation-

ships between expression signatures and molecular andclinical features of the disease in the primary investigative

Medulloblastoma Molecular Subgroups

www.aacrjournals.org Clin Cancer Res; 17(7) April 1, 2011 1885




Tab

le1.

Assoc

iatio

nsbetwee

nmolec

ular

subgrou

psan

dmed

ulloblastom

age

nomic,ep

igen

omic,an

dclinical

disea

sefeatures

Co

rrec

ted

'p' v

alu

e'p

' val

ue

2213

1943

1432

4517

555

5431

210

73

5238

953

514

3026

4642

1612

2740

4128

2124

4420

2511

50

SH

H-

-

WN

T-

-

PT

CH

FS

FS

FS

FS

FS

9.4

x 10

-5 *

1.5

x 10

-3 *

SU

FU

MS

ns

ns

SM

On

sn

s

CT

NN

B1

MS

MS

MS

5.6

x 10

-9 *

9.0

x 10

-8 *

AP

Cn

sn

s

PT

CH

1an

sn

s

PT

CH

1c0.

03 *

ns

SU

FU

ns

ns

CO

L1A

20.

01 *

ns

69.

3 x

10-9

*1.

5 x

10-7

*

9q

0.09

**

ns

17p

(K

CD

T11

)0.

06 *

*n

s

Ag

e1.

02.

62.

93.

54.

811

.514

.216

.719

.09.

010

.010

.32.

52.

52.

62.

93.

03.

34.

14.

34.

64.

65.

05.

15.

45.

55.

76.

36.

87.

58.

58.

69.

810

.010

.212

.512

.615

.417

.0n

sn

s

Pat

ho

log

y0.

003

*0.

05*

Sex

FM

MF

FM

FM

MF

MM

MM

FM

FF

MM

FM

MM

MM

FM

MM

FF

FM

FM

MF

Fn

sn

s

M S

tag

en

sn

s

Clin

ical

fea

ture

s

Mo

lecu

lar

sub

gro

up

a

Gen

etic

mu

tati

on

b

Hyp

er-

met

hyl

atio

nc

Ch

rom

oso

mal

loss

d

SH

HW

NT

WN

T-/

SH

H-i

nd

epen

den

t

NOTE

.Cas

esaresh

ownarrang

edbysign

aturestatus

(SHH,W

NT,

WNT-/SHH-ind

epen

den

t)as

determined

inFig.

1,an

dthen

byas

cend

ingag

e,amolec

ular

subgrou

p;bmutation

status

;chy

permethy

latio

nstatus

(black

,methy

lated;white,un

methy

lated);an

ddch

romos

omal

loss

(black

,allelic;white,no

loss).Age

issh

ownin

yearsan

dca

tego

rized

into

infants(�

3ye

ars,

black

)an

dno

ninfan

ts(>3ye

ars,

white).Patho

logy

varia

ntis

indicated

byawhite

squa

re(class

ic),black

squa

re(DN)or

gray

squa

re(LCA).Gen

der

isindicated

byblac

ksq

uares(F,fem

ale)an

dwhite

squa

res(M

,male).M

stag

e0/1isindicated

bywhite

squa

resan

dM

stag

e2/3byblack

squa

res.Raw

Pan

dP

values

correc

tedfor

multip

lehy

pothe

sistestingaresh

ownforrelatio

nshipsbetwee

nmolec

ular/clinical

correlates

andsu

bgrou

pmem

bersh

ip(c2

tests,

Bon

ferron

icorrection).S

ignific

ant(*P<0.05

)an

dmargina

llysign

ifica

nt(**P¼

0.05

1–0.10

)as

sociations

aremarke

d.Diago

nally

hatche

dbox

esindicateun

available

data.

Abbreviations

:FS

,fram

eshift;MS,misse

nsemutation;

ns,no

tsign

ifica

nt.

Schwalbe et al.





(GeXP) cohort (n ¼ 39) and in the combined cohort (n ¼173). Bonferroni corrections for multiple hypothesis test-ing were applied where appropriate. Additional patient agedata were kindly provided by Dr. M. Kool (AcademicMedical Centre, Amsterdam, the Netherlands).

Results

A diagnostic multigene mRNA expression signatureassay for SHH and WNT medulloblastoma subgroupsWe first developed and validated multigene mRNA

expression signatures to facilitate routine identificationof the SHH and WNT medulloblastoma molecular dis-ease subgroups and assessment of their molecular basis,

associated biomarkers, and clinical relevance. The 8-gene SHH (BCHE, GLI1, ITIH2, MICAL1, PDLIM3,PTCH2, RAB33A, and SFRP1) and 5-gene WNT(CCDC46, DKK2, PYGL, TNC, and WIF1) subgroupsignatures developed were initially assessed by GeXPassay in our primary sample cohort (n ¼ 39) and wereunequivocal for all samples, independent of data ana-lysis method used (stacked bar plots, PCA, or unsuper-vised hierarchical clustering; Figs. 1 and 2). Validation ofthese signatures in 3 independent medulloblastomaexpression microarray data sets [Thompson and collea-gues (n ¼ 46; ref. 13); Kool and colleagues (n ¼ 62;ref. 12); and Fattet and colleagues (n ¼ 40; ref. 15)]showed the signatures could be successfully interrogated

Figure 1. Diagnostic WNT andSHH subgroup gene expressionsignatures recognize equivalentmolecular subgroups acrossmultiple medulloblastomacohorts. Data from 4 independentdata sets are shown [A, primaryinvestigation cohort; B, Kool andcolleagues (n ¼ 62; ref. 12); C,Thompson and colleagues(n ¼ 46; ref. 13); and D, Fattet andcolleagues (n ¼ 40; ref. 15)]. WNTand SHH subgroup expressionsignature positivity demonstratesclose concordance with (i)underlying molecular defects inthe respective pathways and (ii)discrete sample clusters identifiedon independent clustering of themost variable probes in each arraydata set (Supplementary Fig. S2).Each panel (A–D) showshierarchical clustering of signaturegenes (WNT subgroup, red; SHHsubgroup, blue; WNT-/SHH-independent, gray). Mutationalinformation for CTNNB1 andPTCH1 is shown (colored boxes,mutation; gray checked boxes,missing data). Array clusterswhich show concordance with thedetected SHH andWNT subgroupsignatures (derived by clusteringthe most variable probes of eachwhole array data set) are shown[purple, SHH subgroup; orange,WNT subgroup (SupplementaryFig. S2)]. Biplots show PCA foreach signature geneset. Arrowsshow projections of expressionaxes for each gene (SHH signaturegenes, blue; WNT signaturegenes, red; WNT-/SHH-independent cases, gray).






by hierarchical cluster analysis and PCA, were diagnosticin all cases, independent of cohort or gene expressionassay used (Fig. 1), and showed close consistency withstacked bar plot data (Fig. 2). Signature positivity wasconcordant with the disease subgroups apparent afterindependent clustering of the most variable probeswithin each entire array data set, and correctly classifieddisease subgroup (i.e., as SHH, WNT, or WNT/SHHindependent) in 99% (146/148) of cases overall(Fig. 1). Data sets similarly showed complete concor-dance in subgroup assignment, using all analyticalmethods, of the 14 cases (4 SHH, 2 WNT, and 8WNT/SHH independent cases) where signatures wereassessed in parallel by microarray and GeXP assays.

Integration of cohorts and gene expression data sets:molecular subgroup incidence

Based on our validated gene expression signatures anddata analysis methods, the molecular subgroup data fromall 4 cohorts could be combined for meta-analysis with themutational and clinical data which were consistentlyreported for all studies. A total cohort of 173 cases wasassessed in this analysis (Supplementary Table S5). Clinicaldemographics available for the combined cohort wereconsistent with previously reported groupwide estimatesfor medulloblastoma (1); male:female ratio (1.42:1),histology [115 classic (68%), 39 (23%) DN, 16 (9%)LCA, 3 cases with data not available; pathologic classifica-tions were those reported in the original publications

Figure 2. Identification of SHH and WNT subgroup medulloblastomas using diagnostic expression signatures. Data are illustrated from the 4 independentdata sets using stacked bar plots [A, primary investigation cohort; B, Kool and colleagues (n¼ 62; ref. 12); C, Thompson and colleagues (n¼ 46; ref. 13); andD, Fattet and colleagues (n ¼ 40; ref. 15)]. WNT signature genes (red) and SHH signature genes (blue) in combination clearly define subgroup membership.Vertical dashed lines delineate sample groups positive for WNT and SHH signatures by hierarchical clustering and PCA analysis in Fig. 1. Right-hand panelindicates stacked order of genes for each signature. Before generation of bar plots, expression data from each cohort were scaled on a per genebasis to a mean of zero and a variance of 1. Samples expressing all or most signature genes at above average levels will show bars of greater positivemagnitude.

Schwalbe et al.





(12, 13, 15)], and age at diagnosis [median 6.45, range 0.3–35.3 years, 34 (20%) infant, 138 (80%) noninfant cases, 1case with data not available]. An overall incidence of 21(12%) WNT and 42 (24%) SHH subgroup cases wasobserved. Subclassification of the remainder of cases basedon their transcriptomic patterns has not produced consis-tent findings in previous studies (12, 13). RemainingWNT-/SHH-independent medulloblastomas were there-fore classified as a single group, representing 110 (64%)cases.

Genetic and epigenetic mechanisms of SHH and WNTactivationGenetic mutations in PTCH1 and CTNNB1 were exclu-

sively detected in SHH and WNT subgroup cases, respec-tively, across all 4 cohorts, where data were available[combined cohort, P ¼ 5.3 � 10�8 and P ¼ 0 (c2 test),respectively; Fig. 1], further validating the fidelity of thegene expression signatures developed. CTNNB1 mutationwas the primary mechanism and correlate ofWNT pathwayactivation; in 20 cases where combined expression andmutation data were available, all except 1 WNT subgroupcase (19/20; 95%) had concurrent CTNNB1 mutation.Consistent with this, no APC mutations were found inthe 55 cases tested within our primary cohort.PTCH1 mutation was a major mechanism of SHH path-

way activation; 34% (11/32) of SHH subgroup cases inves-tigated harbored a PTCH1 mutation (Fig. 1). We thereforenext undertook a systematic investigation of alternativegenetic mechanisms of pathway activation in the remain-ing majority of SHH cases in our primary cohort. Muta-tional analysis encompassed all pathway genes in whichmutations have previously been reported; the completecoding sequences of PTCH1 [including exon 1B, whichhas been shown to code for protein (32), but has not beenanalyzed in previous studies (3, 20, 21)], SUFU andSMO [which has only been previously been assessed formutation at specific residues (18)]. In addition to PTCH1truncating mutations (n ¼ 5), only 1 further potentiallypathogenic missense SUFU mutation was identified, withno evidence of SMO mutation found (SupplementaryTable S6). Additional nonpathogenic variants discovered(e.g., polymorphisms or intronic changes) are summarizedin Supplementary Table S7. Allelic loss of chromosome17p, targeting REN (KCTD11) at 17p13.2, has also beenpreviously associated with SHH activation in medulloblas-toma (19, 33), and was observed in 24% (9/37) of cases.Epigenetic mechanisms of pathway activation were addi-

tionally investigated as an alternative to genetic determi-nants. Two predicted promoter-associated CpG islandsspanning exons 1a and 1c of PTCH1, and a promoter-associated CpG island within SUFU, were identified andinvestigated for evidence of DNA hypermethylation, whichmay lead to epigenetic transcriptional silencing. No evi-dence of DNA methylation of the PTCH1 exon 1a-asso-ciated or the SUFU CpG island was seen in any tumoranalyzed (n ¼ 39; Table 1), indicating no major role indisease development. Mixed patterns of methylation of the

PTCH1 exon 1c CpG island were observed in 12 of 27 oftumors successfully analyzed (44%; Table 1), suggesting apotential role for this epigenetic mechanism.

Unlike PTCH1 mutations, other defects identified[PTCH1 exon 1c methylation, SUFU missense mutation,and 17p allelic loss (REN; KCTD11)] were not associatedwith the SHH subgroup (Table 1), indicating that any rolethese mechanisms may play in medulloblastoma develop-ment is independent of the SHH pathway.

Genomic biomarkers of SHH and WNT pathwayactivation

We next investigated selected medulloblastoma chro-mosomal abnormalities (chromosome 6, 9q, and 17ploss) and epigenomic defects (COL1A2 status) of biolo-gical and/or prognostic significance (1, 25), for theirassociations with the SHH and WNT disease subgroups,and each other, to assess any utility as biomarkers ofpathway activation.

Chromosome 6 loss, CTNNB1 mutation, and theabsence of chromosome 9 and 17 abnormalities, wereobserved in all WNT cases in our primary cohort (Table 1),consistent with previous findings (12–15). Across thecombined cohort, evidence of loss of an entire copy ofchromosome loss 6 was associated with 88% (14/16) ofWNT cases with available data (P < 3 � 10�16, Fisher’sexact); however, this relationship was not exclusive. Chro-mosome 6 loss was also detected in occasional non-WNTcases (2/145). Eight of 35 tumors tested (23%) showedevidence of genetic loss at the 9q22.3 region surroundingthe PTCH1 locus in our primary cohort (Table 1). Four of 8were in the SHH subgroup and 2 of 4 tumors with PTCH1mutations showed LOH of 9q22.3; however neither asso-ciation reached significance (P ¼ 0.09 and 0.22 respec-tively, Fisher’s exact test). A significant inverse associationbetween 17p loss and membership of the SHH and WNTsubgroups was observed; 17p losses were exclusivelyobserved in WNT-/SHH-independent cases [17p LOH in0/12 SHH or WNT cases vs. 9/25 WNT-/SHH-independentcases (P ¼ 0.02, Fisher’s exact test)]. In addition, COL1A2hypermethylation was detected in 76% (25/33) of cases; anabsence of COL1A2 methylation was significantly asso-ciated with the SHH subgroup (P ¼ 0.01, c2 test), but thisrelationship was not maintained when a correction formultiple hypothesis testing was applied (Table 1).

Medulloblastoma molecular subgroups displaydistinct clinical features

Analysis of the medulloblastoma molecular subgroupsdefined by our gene expression signatures, in all 173 casesof the combined cohort, revealed marked differences intheir clinical disease features. WNT-/SHH-independenttumors made up the majority of cases, and had their peakincidence in the 3- to 6-year age group, but were extremelyrare in the first 2 years of life (Fig. 3). In contrast, SHHsubgroup tumors had their major peak of incidence ininfancy [50% (21/42) of SHH cases were �3 years of age].The SHH subgroup tumors represented the majority of






the infant clinical group [62% (21/34) of cases �3 years ofage], but were less common in noninfant children [>3–15years; 15% (16/127)], and had a further increased inci-dence in adult cases [�16 years (45%; 5/11P); overall P ¼5.8 � 10�9, c2 test]. Almost all cases <2 years of age [92%(11/12)] were SHH-positive. WNT subgroup cases were notobserved in infants (minimum age observed, 5 years) and

had a bi-modal age distribution with major and minorpeaks at 10 and 20 years, respectively.

Significant differences were also observed in the dis-tribution of medulloblastoma histologic subtypesbetween the molecular subgroups (P ¼ 3.1 � 10�11, c2

test; Figs. 4 and 5). WNT subgroup cases exclusivelydisplayed classic histology (P ¼ 0.0003, Fisher’s exacttest), and WNT-/SHH-independent tumors were also pre-dominantly of the classic subtype, but DN and LCA caseswere also observed. Consistent with previous studies (3,12, 13, 21), SHH cases were overall strongly associatedwith DN histology (P ¼ 1.1 � 10�9, Fisher’s exact test).However, this relationship was not absolute and LCA andclassic cases were also observed in the SHH group. Mostnotably, examination of this large cohort revealed therelationship between SHH activation and DN pathologyto be age dependent (Figs. 4 and 5); DN cases made upthe majority of infant (�3 years old) SHH subgroup cases;all DN cases in this infant group displayed SHH activa-tion. DN pathology may therefore serve as a surrogatemarker of SHH activation in the infant group. In contrast,there were almost equal proportions of DN, LCA, andclassic cases in SHH-expressing noninfant cases, and themajority of noninfant DN tumors were not SHH activated(P ¼ 8.6 � 10�5, Fisher’s exact test). No significantevidence for differences in metastatic stage [WNT 6%(1/16) M stage 2/3, SHH 16% (5/32), and WNT-/SHH-independent 24% (20/82)] was observed between thedifferent expression subgroups (P ¼ 0.20, c2 test).

B

A

35

40

45

10

15

20

25

30

F

Case

den

sity

0

5

0–3 3–6 6–9 9–12 12–15 15–18 >18

Age range (y)

C

C

12

14

16

Age (y)

WNT/SHH ind

SHH n=42n=21n=109

WNT

4

6

8

10

F

0

2

5–64–53–42–31–20–1

Age range (y)

Figure 3. Medulloblastoma molecular subgroups show distinct age ofincidence distributions. Data for the WNT (gray), SHH (black), and WNT-/SHH-independent (hatched) subgroups are shown, based on a combinedcohort of 173 medulloblastomas. A, density plots show subgroup-dependent ages of incidence. Case density represents the smoothedfrequency of incidence within each of the 3 subgroups. Gray dotted line isplotted at 3 years of age. B, bar plots show age distribution of data set.C, bar plot shows age distribution of cases aged �6 years at diagnosis. F,frequency.

A. WNT B. SHH

(n=7)17%

24%(n 25)60%

CLAS

LCA

DNCLAS

(n=20)100%

(n=10)(n=25)CLAS

C. WNT/SHH

13%

independent

8%( 9) (n=14)

DN

(n=9)

CLAS

LCA

79%(n=85)

Figure 4. Molecular subgroups show relationships to medulloblastomahistologic variants. Subgroup and histologic information was available for170 of 173 cases (Supplementary Table S5). A, WNT subgroup (n ¼ 20);B, SHH subgroup (n ¼ 42); C, WNT-/SHH-independent cases (n ¼ 108;P ¼ 3.1 � 10�11, c2 test). White, classic (CLAS); gray, LCA; black DNhistologic variants.

Schwalbe et al.





Discussion

The identification of distinct medulloblastoma molecu-lar subgroups (12, 13, 15) offers significant potential toimprove our understanding of disease biology and clinicalmanagement. Here, we report the development and valida-tion of minimal diagnostic gene expression signatureswhich can be routinely applied to identify the SHH,WNT, and WNT-/SHH-independent medulloblastoma dis-ease subgroups. These gene expression signatures are robustand informative for subgroup identification in RNAextracted from snap-frozen tumor material, and usingdifferent gene expression assays. In particular, the GeXPmultiplex expression assay reported offers a number ofsignificant advantages over microarray methodologies forthe routine assignment of subgroup affiliation; analysis canbe undertaken straightforwardly, rapidly (in 1working day,compared with 2–3 days for a microarray experiment) andcost-effectively (approximately one tenth of microarrayanalysis costs) and, importantly, can be performed onsmall amounts of total RNA (150 ng, compared with500 ng to 5 mg for a typical microarray expression analysis),a common limitation when only small amounts of biopsymaterial are available. This removal of the need for rela-tively time-consuming, complex, and expensive array ana-

lysis platforms for subgroup assignment provides a strongbasis for their clinical application; these methods are fea-sible for investigation in real time acrossmultiple treatmentcenters during clinical treatment and in future clinicaltrials.

We have shown that the disease subgroups recognized bythese signatures are equivalent and consistently identifiedin 4 independent medulloblastoma cohorts, allowing theirassembly into a large combined cohort. Coupled with anextensive analysis of our novel primary cohort, this hasallowed significant insights into the underlying molecularmechanisms, associated biomarkers, and clinical charac-teristics of these molecular disease subgroups.

Our systematic investigation of specific medulloblas-toma genetic and epigenetic defects in this study hasinformed their roles as determinants or correlates of themolecular subgroups identified. Consistent with previousstudies (12–14), CTNNB1mutations were identified as theprimary pathway-activating event present in almost allWNT subgroup tumors, with chromosome 6 losses alsoaffecting the majority of these cases. PTCH1 mutation wasthe major mechanistic correlate of SHH activation, identi-fied in �34% of SHH cases. SHH-associated PTCH1 muta-tions were detected both in conjunction with chromosome9q loss, and in the heterozygous state, indicating disrup-tion of a single PTCH1 allele can be sufficient to cause SHHpathway disruption in medulloblastoma. An absence ofCOL1A2 hypermethylation was also significantly asso-ciated with SHH subgroup medulloblastomas, moststrongly in infant cases. Notably, a number of the pre-viously reported determinants of SHH activation that weexamined [PTCH1 exon 1c methylation (ref. 16), SUFUmissense mutation (ref. 17), and 17p REN (KCTD11; ref.19) allelic loss] were not specifically associated with theSHH subgroup, indicating any role they may play inmedulloblastoma is SHH-independent. It is similarlyunclear whether the mixed methylation patterns observedfor PTCH1 exon 1c in this study have functional signifi-cance, as this was not assessed. In addition, other SHHpathway defects examined (PTCH1 exon 1a hypermethyla-tion), including events previously reported inmedulloblas-toma [SMO mutations (ref. 18) or SUFU truncatingmutations (ref. 17)] were not observed at all, suggestingtheir roles are either less common than previously thoughtor are restricted to limited tumor subsets less well repre-sented in our mutation screening cohort. This is likely thecase for SUFU mutations, which appear to be associatedwith germline inheritance and have their peak incidence ininfants (17, 23, 34). Further mechanisms of pathwayactivation remain to be identified in the majority ofSHH cases. Chromosome 17 defects were the only eventssignificantly correlated with the most common WNT-/SHH-independent subgroup, suggesting a role for chromo-some 17 genes in these cases. This disease subgroup,however, remains the least well characterized at the mole-cular level. Subdivision of this group has been proposed onthe basis of its transcriptomic and genomic patterns; how-ever, unlike the SHH and WNT groups, inconsistent results

A1. ≤3 years, SHH

n=21

19%

B1. ≤3 years, DN

A2. >3 years, SHH

43%

B2. >3 years, DN

61%

39%

(n=16)

(n=4)

(1)

(n=9)

29%(n=6)

(n=9)

(n=14)

76%CLAS

DN

5%

CLAS

LCADN SHHnon-

SHH

29%(n=6)

SHH

(n=16)100%

Figure 5. Associations between SHH subgroupmedulloblastomas andDNhistology are age-dependent. A, histologic variants [white, classic(CLAS); gray, LCA; black DN] show significantly different distributions(P ¼ 0.05; c2 test) in SHH subgroup cases arising in infants [�3 years atdiagnosis (n ¼ 21); A1] and noninfants [>3 years (n ¼ 21); A2]. B, infant(n ¼ 16; B1) and noninfant (n ¼ 23; B2) DN medulloblastomas showsignificantly different relationships to the SHH subgroup (P ¼ 8.6 � 10�5;Fisher's exact test; SHH subgroup, white; non-SHH, black).






have been reported from different studies (12, 13), and theidentification of specific genes and pathways associatedwith the pathogenesis of this subgroup will be critical tofuture advances in our understanding of its molecular basisand any underlying heterogeneity.

The significant associations observed between medullo-blastoma molecular subgroups and specific gene, pathway,and chromosomal defects (i) strongly support the existenceof molecularly distinct SHH and WNT subgroups, (ii)inform the contributory mechanisms involved in theirpathogenesis, and (iii) provide potential biomarkers forsubgroup identification. When assessed for suitability asprimary biomarkers, only CTNNB1mutations, which werespecifically observed in all but 1 WNT subgroup case, havesufficient sensitivity and specificity to have utility. Nuclearlocalization of b-catenin has also been widely reported as apositive marker of WNT pathway activation (5, 8, 14),although its relationship to our WNT expression signatureand CTNNB1 mutations could not be investigated in thisstudy due to tissue limitations. Similarly, COL1A2 statusmay have utility in the identification of SHH subgroupinfant desmoplastic medulloblastomas [this study and(25)], particularly in cases where biopsy limitations donot allow assessment of the DN pathologic variant. Theremainder of gene and chromosomal defects investigatedwere not suitable as primary biomarkers for positive sub-group discrimination, as a result of their (i) nonexclusivity,(ii) limitation to subsets of subgroup cases, or (iii) inversecorrelation with pathway activation. In comparison, geneexpression signatures positively identified all subgroupcases and provide an accurate diagnostic test for subgroupmembership. The genomic markers examined may there-fore provide useful secondary or confirmatory markers,when used in conjunction with these signatures. It is notpresently clear whether the expression signatures reportedtranslate into subgroup-specific protein expression. Proteinexpression markers, which are testable by immunohisto-chemistry, may have utility for routine subgroup assign-ment in the diagnostic setting, however carefulinvestigation and validation of their sensitivity, robustness,and reflection of expression array subgroup data, will beessential before their application.

The observation of 2 of 148 cases which were notconsistently classified using the presently reported signa-tures and their respective array data sets (Fig. 1) indicatespotential difficulties in the robust classification of a smallsubset (<2%) of cases. Inspection of the microarray expres-sion data for these 2 cases revealed markedly reducedexpression of pathway signature genes relative to the otherpathway positive cases in their respective cohorts, despitetheir subgroup positivity using our signatures (refs. 12, 13,15; Fig. 2). In addition, the stacked bar plot analysesrevealed 2 further SHH-positive samples (T27, K452;Fig. 2) which, although clustered consistently betweensignature and array on hierarchical cluster and PCA ana-lysis, could be classified as indeterminate based on theapplication of quantitative criteria to the individual expres-sion data in stacked bar plot analysis [i.e., cumulative

stacked bar plot expression score >8 (for WNT expressionsignature positive cases) or >4 (SHH cases)]. For such cases,additional markers of pathway activation (CTNNB1,PTCH1 mutation) could aid definitive subgroup assign-ment, and it is notable that 1 of these 2 samples alsoharbored a PTCH1 mutation, further supporting SHHsubgroup membership. Silhouette analysis supported therobust assignment of subgroup status using our signatures(Supplementary Fig. S1) but, consistent with the otheranalytical methods applied, did not support the subgroupassignment made for 3 of the 4 discrepant cases describedearlier, further highlighting difficulties in their classifica-tion. Thus, in diagnostic applications, a "nonclassified"designation could be reserved for these rare cases which donot classify consistently across all analyses performed onthe signature data, particularly where subgroup assignmentmay impact clinical or therapeutic decisions.

The combination of molecular and clinicopathologicdata from 4 independent cohorts for meta-analysis, total-ing 173 cases, has facilitated clear and significant insights tothe clinical features of the medulloblastoma molecularsubgroups, which have either not been apparent or notshown statistical significance in individual analyses of thesmaller component cohorts reported to date (12, 13, 15).The SHH (24% of cases), WNT (12%), and WNT-/SHH-independent (64%) groups show different age distribu-tions and relationships to disease histopathology. SHHsubgroup tumors peak in infancy and are intimately cor-related with DN pathology in this group, to the extent thatDN pathology may be considered as a surrogate marker forSHH activation in medulloblastomas arising in infants <3years old at diagnosis, although classic and LCA cases alsoconstitute a minority of SHH subgroup cases in this agegroup. This relationship breaks down in noninfants (�3years at diagnosis), where SHH tumors are less common,and shows equivalent proportions of DN, classic, and LCAdisease; SHH-independent DN cases are also commonlyobserved in this age group. These data strongly indicate that(i) SHH subgroup and (ii) DN tumors, arising in the infantand noninfant age groups, have different biological andclinical characteristics, and that SHH-positive DN tumorsof infancy represent a unique disease subgroup associatedwith a favorable clinical behavior (35–37), and a charac-teristic molecular pathogenesis [COL1A2 unmethylated(ref. 25)] and mutational spectrum [SUFU (refs. 17,34)]. Conversely, WNT tumors display classic pathologyand occur in noninfants. Notably, both the SHH and WNTsubgroups show at least 2 different incidence peaks in theirage distribution (both have second peaks in adults), sug-gesting additional clinical and molecular heterogeneitywithin these groups. The inclusion of cohort-wide centralpathology review in future studies may aid the furtherrefinement of the associations observed.

The lack of association between M stage and molecularsubgroups (P ¼ 0.20) is in disagreement with the previousstudy by Kool and colleagues (12),who reportedmetastatictumors being less common in WNT and SHH pathway–activated medulloblastomas. This could be due to the

Schwalbe et al.





different measurement criteria for metastasis between thestudies (this study compared M0/1 vs. M2/3, whereas Kooland colleagues compared M0 vs. M1/M2/M3). Alterna-tively, the increased numbers in this study (130 vs. 58with M stage data in the Kool and colleagues study) mayhave enabled a more accurate characterization of the rela-tionship between signaling pathway activation and metas-tasis, and future large clinically controlled studies whichinclude central review of metastatic status should be infor-mative in this regard.The identification of medulloblastoma molecular sub-

groups has significant prognostic and predictive potentialto improve therapeutic delivery and disease outlook in theclinical setting and could represent a first step in themolecular diagnostic triage of medulloblastomas to guidetreatment decisions. In addition to distinct clinical features,molecular subgroups also appear to have characteristicclinical behaviors; the favorable prognosis of WNT subtypemedulloblastomas is now established in multiple clinicalcohorts (8, 9, 15, 38) and will form the basis of treatmentreductions in forthcoming international molecularly dri-ven clinical trials (1). Combined data from this and otherstudies indicate SHH-positive DN tumors arising in infantsrepresent a similarly favorable prognosis subgroup with adistinctmolecular basis (25, 35–37). The nonavailability ofoutcome data for the 4 cohorts assessed in our meta-analysis have limited any direct assessment of survivalassociations in these cohorts. Moreover, their retrospective,heterogeneously treated nature would likely confoundsuch analyses. Robust assessment of the prognostic impactof the medulloblastoma molecular subgroups will there-fore now be essential, ideally within the context of ade-quately powered, centrally reviewed, and uniformly treatedclinical trials cohorts, to determine their utility to direct theselection of adjuvant therapy. The molecular signatures wereport will have significant utility in this regard. Molecu-

larly targeted SHH inhibitors are also currently underpreclinical and clinical development, and have shown earlyevidence of activity in medulloblastoma (10, 11). Theability to accurately diagnose the SHHmolecular subgroupwill thus be important for the targeted delivery of thesenovel agents, and our findings have identified SHH-posi-tive subgroups of medulloblastomas which would be pre-dicted to benefit most from SHH inhibition strategies.However, the SHH pathway plays a key role in normal,including cerebellar, development, and its transient inhibi-tion in young mice causes permanent defects in growthplate formation and bone structure (39). In view of suchpotential toxicities, we recommend caution in the applica-tion of our data, particularly in the infant age group whereSHH subgroup tumors predominate.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

Medulloblastomas investigated in this study include samples provided bythe UK Children’s Cancer and Leukaemia Group (CCLG) as part of CCLG-approved biological study BS-2007–04. This study was conducted withethics committee approval from Newcastle/North Tyneside REC (studyreference 07/Q0905/71).

Grant Support

This work was supported by grants from The Katie Trust (to S.C. Clifford),The Samantha Dickson Brain Tumour Trust (to S.C. Clifford), and CancerResearch UK (to S.C. Clifford and D.W. Ellison).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received August 17, 2010; revised January 6, 2011; accepted January 10,2011; published OnlineFirst February 16, 2011.

References1. Pizer BL, Clifford SC. The potential impact of tumour biology on

improved clinical practice for medulloblastoma: progress towardsbiologically driven clinical trials. Br J Neurosurg 2009;23:364–75.

2. Kenney AM, Rowitch DH. Sonic hedgehog promotes G(1) cyclinexpression and sustained cell cycle progression in mammalian neu-ronal precursors. Mol Cell Biol 2000;20:9055–67.

3. Raffel C, Jenkins RB, Frederick L, Hebrink D, Alderete B, Fults DW,et al. Sporadic medulloblastomas contain PTCH mutations. CancerRes 1997;57:842–5.

4. Goodrich LV, Milenkovic L, Higgins KM, Scott MP. Altered neural cellfates and medulloblastoma in mouse patched mutants. Science1997;277:1109–13.

5. Eberhart CG, Tihan T, Burger PC. Nuclear localization and mutation ofbeta-catenin in medulloblastomas. J Neuropathol Exp Neurol2000;59:333–7.

6. Huang H, Mahler-Araujo BM, Sankila A, Chimelli L, Yonekawa Y,Kleihues P, et al. APC mutations in sporadic medulloblastomas. Am JPathol 2000;156:433–7.

7. Dahmen RP, Koch A, Denkhaus D, Tonn JC, S€orensen N, Berthold F,et al. Deletions of AXIN1, a component of the WNT/wingless pathway,in sporadic medulloblastomas. Cancer Res 2001;61:7039–43.

8. Ellison DW, Onilude OE, Lindsey JC, Lusher ME, Weston CL, TaylorRE, et al. Beta-Catenin status predicts a favorable outcome in child-hood medulloblastoma: the United Kingdom Children's Cancer StudyGroup Brain Tumour Committee. J Clin Oncol 2005;23:7951–7.

9. Gajjar A, Chintagumpala M, Ashley D, Kellie S, Kun LE, Merchant TE,et al. Risk-adapted craniospinal radiotherapy followed by high-dosechemotherapy and stem-cell rescue in children with newly diagnosedmedulloblastoma (St Jude Medulloblastoma-96): long-term resultsfrom a prospective, multicentre trial. Lancet Oncol 2006;7:813–20.

10. Romer JT, Kimura H, Magdaleno S, Sasai K, Fuller C, Baines H, et al.Suppression of the Shh pathway using a small molecule inhibitoreliminates medulloblastoma in Ptc1(þ/-)p53(-/-) mice. Cancer Cell2004;6:229–40.

11. Rudin CM, Hann CL, Laterra J, Yauch RL, Callahan CA, Fu L, et al.Treatment of medulloblastoma with hedgehog pathway inhibitorGDC-0449. N Engl J Med 2009;361:1173–8.

12. Kool M, Koster J, Bunt J, Hasselt NE, Lakeman A, van Sluis P, et al.Integrated genomics identifies five medulloblastoma subtypes withdistinct genetic profiles, pathway signatures and clinicopathologicalfeatures. PLoS ONE 2008;3:e3088.

13. Thompson MC, Fuller C, Hogg TL, Dalton J, Finkelstein D, Lau CC,et al. Genomics identifies medulloblastoma subgroups that are






enriched for specific genetic alterations. J Clin Oncol 2006;24:1924–31.

14. Clifford SC, Lusher ME, Lindsey JC, Langdon JA, Gilbertson RJ,Straughton D, , et al. Wnt/Wingless pathway activation and chromo-some 6 loss characterize a distinct molecular sub-group of medullo-blastomas associated with a favorable prognosis. Cell Cycle2006;5:2666–70.

15. Fattet S, Haberler C, Legoix P, Varlet P, Lellouch-Tubiana A, Lair S,et al. Beta-catenin status in paediatric medulloblastomas: correlationof immunohistochemical expression with mutational status, geneticprofiles, and clinical characteristics. J Pathol 2009;218:86–94.

16. Diede SJ, Guenthoer J, Geng LN, Mahoney SE, Marotta M, Olson JM,et al. DNA methylation of developmental genes in pediatric medullo-blastomas identified by denaturation analysis of methylation differ-ences. Proc Natl Acad Sci USA 2009;107:234–9.

17. Taylor MD, Liu L, Raffel C, Hui CC, Mainprize TG, Zhang X, et al.Mutations in SUFU predispose to medulloblastoma. Nat Genet2002;31:306–10.

18. Reifenberger J, Wolter M, Weber RG, Megahed M, Ruzicka T, LichterP, et al. Missense mutations in SMOH in sporadic basal cell carci-nomas of the skin and primitive neuroectodermal tumors of the centralnervous system. Cancer Res 1998;58:1798–803.

19. Di Marcotullio L, Ferretti E, De Smaele E, Argenti B, Mincione C,Zazzeroni F, et al. REN(KCTD11) is a suppressor of Hedgehogsignaling and is deleted in human medulloblastoma. Proc Natl AcadSci U S A 2004;101:10833–8.

20. Pietsch T, Waha A, Koch A, Kraus J, Albrecht S, Tonn J, et al.Medulloblastomas of the desmoplastic variant carry mutations ofthe human homologue of Drosophila patched. Cancer Res 1997;57:2085–8.

21. Wolter M, Reifenberger J, Sommer C, Ruzicka T, Reifenberger G.Mutations in the human homologue of the Drosophila segment polar-ity gene patched (PTCH) in sporadic basal cell carcinomas of the skinand primitive neuroectodermal tumors of the central nervous system.Cancer Res 1997;57:2581–5.

22. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A,et al. The 2007 WHO classification of tumours of the central nervoussystem. Acta Neuropathol 2007;114:97–109.

23. Scott DK, Straughton D, Cole M, Bailey S, Ellison DW, Clifford SC.Identification and analysis of tumor suppressor loci at chromosome10q23.3–10q25.3 in medulloblastoma. Cell Cycle 2006;5:2381–9.

24. Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB. Methyla-tion-specific PCR: a novel PCR assay for methylation status of CpGislands. Proc Natl Acad Sci U S A 1996;93:9821–6.

25. Anderton JA, Lindsey JC, Lusher ME, Gilbertson RJ, Bailey S, EllisonDW, et al. Global analysis of the medulloblastoma epigenome iden-tifies disease-subgroup-specific inactivation of COL1A2. Neuro-oncology 2008;10:981–94.

26. Lindsey JC, Lusher ME, Anderton JA, Bailey S, Gilbertson RJ, PearsonAD, et al. Identification of tumour-specific epigenetic events in medul-

loblastoma development by hypermethylation profiling. Carcinogen-esis 2004;25:661–8.

27. Langdon JA, Lamont JM, Scott DK, Dyer S, Prebble E, Bown N, et al.Combined genome-wide allelotyping and copy number analysis iden-tify frequent genetic losses without copy number reduction in medul-loblastoma. Genes Chromosomes Cancer 2006;45:47–60.

28. Rai AJ, Kamath RM, Gerald W, Fleisher M. Analytical validation of theGeXP analyzer and design of a workflow for cancer-biomarker dis-covery using multiplexed gene-expression profiling. Anal BioanalChem 2009;393:1505–11.

29. Barrett T, Troup DB, Wilhite SE, Ledoux P, Rudnev D, Evangelista C,et al. NCBI GEO: mining tens of millions of expression profiles—database and tools update. Nucleic Acids Res 2007;35:D760–5.

30. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ,Scherf U, et al. Exploration, normalization, and summaries of highdensity oligonucleotide array probe level data. Biostatistics 2003;4:249–64.

31. Rousseeuw PJ. Silhouettes: a graphical aid to the interpretationand validation of cluster analysis. J Comput Appl Math 1987;20:53–65.

32. Kogerman P, Krause D, Rahnama F, Kogerman L, Und�en AB, Zaphir-opoulos PG, et al. Alternative first exons of PTCH1 are differentiallyregulated in vivo and may confer different functions to the PTCH1protein. Oncogene 2002;21:6007–16.

33. De Smaele E, Di Marcotullio L, Ferretti E, Screpanti I, Alesse E, GulinoA. Chromosome 17p deletion in human medulloblastoma: a missingcheckpoint in the Hedgehog pathway. Cell Cycle 2004;3:1263–6.

34. Brugieres L, Pierron G, Chompret A, Paillerets BB, Di Rocco F, VarletP, et al. Incomplete penetrance of the predisposition to medulloblas-toma associated with germ-line SUFU mutations. J Med Genet47:142–4.

35. Rutkowski S, Bode U, Deinlein F, Ottensmeier H, Warmuth-Metz M,Soerensen N, et al. Treatment of early childhood medulloblastomaby postoperative chemotherapy alone. N Engl J Med 2005;352:978–86.

36. Garre ML, Cama A, Bagnasco F, Morana G, Giangaspero F, BrisigottiM, et al. Medulloblastoma variants: age-dependent occurrence andrelation to Gorlin syndrome—a new clinical perspective. Clin CancerRes 2009;15:2463–71.

37. McManamy CS, Pears J, Weston CL, Hanzely Z, Ironside JW, TaylorRE, , et al. Nodule formation and desmoplasia in medulloblastomas-defining the nodular/desmoplastic variant and its biological behavior.Brain Pathol 2007;17:151–64.

38. Korshunov A, Remke M, Werft W, Benner A, Ryzhova M, Witt H, et al.Adult and pediatric medulloblastomas are genetically distinct andrequire different algorithms for molecular risk stratification. J ClinOncol 2010;28:3054–60.

39. Kimura H, Ng JM, Curran T. Transient inhibition of the Hedgehogpathway in young mice causes permanent defects in bone structure.Cancer Cell 2008;13:249–60.

Schwalbe et al.





Documents

Rapid Diagnosis of Medulloblastoma Molecular Subgroups