SUMMARY While molecular subgrouping has revolutionized medulloblastoma classification, the extent of heterogeneity within subgroups is unknown. Similarity network fusion (SNF) applied to genome-wide DNA methylation and gene expression data across 763 primary samples identifies very homogeneous clusters of patients, supporting the presence of medulloblastoma subtypes. After integration of somatic copy-number alterations, and clinical features specific to each cluster, we identify 12 different subtypes of medulloblastoma. Integrative analysis using SNF further delineates group 3 from group 4 medulloblastoma, which is not as readily apparent through analyses of individual data types. Two clear subtypes of infants with Sonic Hedgehog medulloblastoma with disparate outcomes and biology are identified. Medulloblastoma subtypes identified through integrative clustering have important implications for stratification of future clinical trials.

A B S T R A C T PurposeReports detailing the prognostic impact of TP53 mutations in medulloblastoma offer conflicting conclusions. We resolve this issue through the inclusion of molecular subgroup profiles. Patients and MethodsWe determined subgroup affiliation, TP53 mutation status, and clinical outcome in a discovery cohort of 397 medulloblastomas. We subsequently validated our results on an independent cohort of 156 medulloblastomas. ResultsTP53 mutations are enriched in wingless (WNT; 16%) and sonic hedgehog (SHH; 21%) medulloblastomas and are virtually absent in subgroups 3 and 4 tumors (P Ͻ .001). Patients with SHH/TP53 mutant tumors are almost exclusively between ages 5 and 18 years, dramatically different from the general SHH distribution (P Ͻ .001). Children with SHH/TP53 mutant tumors harbor 56% germline TP53 mutations, which are not observed in children with WNT/TP53 mutant tumors. Five-year overall survival (OS; Ϯ SE) was 41% Ϯ 9% and 81% Ϯ 5% for patients with SHH medulloblastomas with and without TP53 mutations, respectively (P Ͻ .001). Furthermore, TP53 mutations accounted for 72% of deaths in children older than 5 years with SHH medulloblastomas. In contrast, 5-year OS rates were 90% Ϯ 9% and 97% Ϯ 3% for patients with WNT tumors with and without TP53 mutations (P ϭ .21). Multivariate analysis revealed that TP53 status was the most important risk factor for SHH medulloblastoma. Survival rates in the validation cohort mimicked the discovery results, revealing that poor survival of TP53 mutations is restricted to patients with SHH medulloblastomas (P ϭ .012) and not WNT tumors. ConclusionSubgroup-specific analysis reconciles prior conflicting publications and confirms that TP53 mutations are enriched among SHH medulloblastomas, in which they portend poor outcome and account for a large proportion of treatment failures in these patients.

The development of targeted anti-cancer therapies through the study of cancer genomes is intended to increase survival rates and decrease treatment-related toxicity. We treated a transposon–driven, functional genomic mouse model of medulloblastoma with ‘humanized’ in vivo therapy (microneurosurgical tumour resection followed by multi-fractionated, image-guided radiotherapy). Genetic events in recurrent murine medulloblastoma exhibit a very poor overlap with those in matched murine diagnostic samples (<5%). Whole-genome sequencing of 33 pairs of human diagnostic and post-therapy medulloblastomas demonstrated substantial genetic divergence of the dominant clone after therapy (<12% diagnostic events were retained at recurrence). In both mice and humans, the dominant clone at recurrence arose through clonal selection of a pre-existing minor clone present at diagnosis. Targeted therapy is unlikely to be effective in the absence of the target, therefore our results offer a simple, proximal, and remediable explanation for the failure of prior clinical trials of targeted therapy.