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.

Tumors of the same histologic type often comprise clinically and molecularly distinct subgroups; however, the etiology of these subgroups is unknown. Here, we report that histologically identical, but genetically distinct, ependymomas exhibit patterns of gene expression that recapitulate those of radial glia cells in the corresponding region of the central nervous system. Cancer stem cells isolated from ependymomas displayed a radial glia phenotype and formed tumors when orthotopically transplanted in mice. These findings identify restricted populations of radial glia cells as candidate stem cells of the different subgroups of ependymoma, and they support a general hypothesis that subgroups of the same histologic tumor type are generated by different populations of progenitor cells in the tissues of origin.

Glioblastomas (GBMs) are aggressive brain tumors characterized by extensive inter- and intratumor heterogeneity. Patient-derived models, such as organoids and explants, have recently emerged as useful models to study such heterogeneity, although the extent to which they can recapitulate GBM genomic features remains unclear. Here, we analyze bulk exome and single-cell genome and transcriptome profiles of 12 IDH wild-type GBMs, including two recurrent tumors, and of patient-derived explants (PDEs) and gliomasphere (GS) lines derived from these tumors. We find that PDEs are genetically similar to, and variably retain gene expression characteristics of, their parent tumors. Notably, PDEs appear to exhibit similar levels of transcriptional heterogeneity compared with their parent tumors, whereas GS lines tend to be enriched for cells in a more uniform transcriptional state. The approaches and datasets introduced here will provide a valuable resource to help guide experiments using GBM-derived models, especially in the context of studying cellular heterogeneity.

Biallelic inactivation of SMARCB1, encoding a member of the SWI/SNF chromatin remodeling complex, is the hallmark genetic aberration of atypical teratoid rhabdoid tumors (ATRT). Here, we report how loss of SMARCB1 affects the epigenome in these tumors. Using chromatin immunoprecipitation sequencing (ChIP-seq) on primary tumors for a series of active and repressive histone marks, we identified the chromatin states differentially represented in ATRTs compared with other brain tumors and non-neoplastic brain. Re-expression of SMARCB1 in ATRT cell lines enabled confirmation of our genome-wide findings for the chromatin states. Additional generation of ChIP-seq data for SWI/SNF and Polycomb group proteins and the transcriptional repressor protein REST determined differential dependencies of SWI/SNF and Polycomb complexes in regulation of diverse gene sets in ATRTs.

Ependymal tumors across age groups are currently classified and graded solely by histopathology. It is, however, commonly accepted that this classification scheme has limited clinical utility based on its lack of reproducibility in predicting patients' outcome. We aimed at establishing a uniform molecular classification using DNA methylation profiling. Nine molecular subgroups were identified in a large cohort of 500 tumors, 3 in each anatomical compartment of the CNS, spine, posterior fossa, supratentorial. Two supratentorial subgroups are characterized by prototypic fusion genes involving RELA and YAP1, respectively. Regarding clinical associations, the molecular classification proposed herein outperforms the current histopathological classification and thus might serve as a basis for the next World Health Organization classification of CNS tumors.

We recently reported that atypical teratoid rhabdoid tumors (ATRTs) comprise at least two transcriptional subtypes with different clinical outcomes; however, the mechanisms underlying therapeutic heterogeneity remained unclear. In this study, we analyzed 191 primary ATRTs and 10 ATRT cell lines to define the genomic and epigenomic landscape of ATRTs and identify subgroup-specific therapeutic targets. We found ATRTs segregated into three epigenetic subgroups with distinct genomic profiles, SMARCB1 genotypes, and chromatin landscape that correlated with differential cellular responses to a panel of signaling and epigenetic inhibitors. Significantly, we discovered that differential methylation of a PDGFRB-associated enhancer confers specific sensitivity of group 2 ATRT cells to dasatinib and nilotinib, and suggest that these are promising therapies for this highly lethal ATRT subtype.

Glioblastoma (GBM) is a brain tumor that carries a dismal prognosis and displays considerable heterogeneity. We have recently identified recurrent H3F3A mutations affecting two critical amino acids (K27 and G34) of histone H3.3 in one-third of pediatric GBM. Here, we show that each H3F3A mutation defines an epigenetic subgroup of GBM with a distinct global methylation pattern, and that they are mutually exclusive with IDH1 mutations, which characterize a third mutation-defined subgroup. Three further epigenetic subgroups were enriched for hallmark genetic events of adult GBM and/or established transcriptomic signatures. We also demonstrate that the two H3F3A mutations give rise to GBMs in separate anatomic compartments, with differential regulation of transcription factors OLIG1, OLIG2, and FOXG1, possibly reflecting different cellular origins.

We devised a high-throughput, cell-based assay to identify compounds to treat Group3 medulloblastoma (G3 MB). Mouse G3 MBs neurospheres were screened against a library of approximately 7,000 compounds including US Food and Drug Administration-approved drugs. We found that pemetrexed and gemcitabine preferentially inhibited G3 MB proliferation in vitro compared to control neurospheres and substantially inhibited G3 MB proliferation in vivo. When combined, these two drugs significantly increased survival of mice bearing cortical implants of mouse and human G3 MBs that overexpress MYC compared to each agent alone, while having little effect on mouse MBs of the sonic hedgehog subgroup. Our findings strongly suggest that combination therapy with pemetrexed and gemcitabine is a promising treatment for G3 MBs.

We discovered a high-level amplicon involving the chr19q13.41 microRNA (miRNA) cluster (C19MC) in 11/45 ( approximately 25%) primary CNS-PNET, which results in striking overexpression of miR-517c and 520g. Constitutive expression of miR-517c or 520g promotes in vitro and in vivo oncogenicity, modulates cell survival, and robustly enhances growth of untransformed human neural stem cells (hNSCs) in part by upregulating WNT pathway signaling and restricting differentiation of hNSCs. Remarkably, the C19MC amplicon, which is very rare in other brain tumors (1/263), identifies an aggressive subgroup of CNS-PNET with distinct gene-expression profiles, characteristic histology, and dismal survival. Our data implicate miR-517c and 520g as oncogenes and promising biological markers for CNS-PNET and provide important insights into oncogenic properties of the C19MC locus.

Pediatric low-grade gliomas (pLGG) are frequently driven by genetic alterations in the RAS-mitogen-activated protein kinase (RAS/MAPK) pathway yet show unexplained variability in their clinical outcome. To address this, we characterized a cohort of >1,000 clinically annotated pLGG. Eighty-four percent of cases harbored a driver alteration, while those without an identified alteration also often exhibited upregulation of the RAS/MAPK pathway. pLGG could be broadly classified based on their alteration type. Rearrangement-driven tumors were diagnosed at a younger age, enriched for WHO grade I histology, infrequently progressed, and rarely resulted in death as compared with SNV-driven tumors. Further sub-classification of clinical-molecular correlates stratified pLGG into risk categories. These data highlight the biological and clinical differences between pLGG subtypes and opens avenues for future treatment refinement.

Cancers are believed to arise from cancer stem cells (CSCs), but it is not known if these cells remain dependent upon the niche microenvironments that regulate normal stem cells. We show that endothelial cells interact closely with self-renewing brain tumor cells and secrete factors that maintain these cells in a stem cell-like state. Increasing the number of endothelial cells or blood vessels in orthotopic brain tumor xenografts expanded the fraction of self-renewing cells and accelerated the initiation and growth of tumors. Conversely, depletion of blood vessels from xenografts ablated self-renewing cells from tumors and arrested tumor growth. We propose that brain CSCs are maintained within vascular niches that are important targets for therapeutic approaches.

Nearly one-third of children with medulloblastoma, a malignant embryonal tumor of the cerebellum, succumb to their disease. Conventional response monitoring by imaging and cerebrospinal fluid (CSF) cytology remains challenging, and a marker for measurable residual disease (MRD) is lacking. Here, we show the clinical utility of CSF-derived cell-free DNA (cfDNA) as a biomarker of MRD in serial samples collected from children with medulloblastoma (123 patients, 476 samples) enrolled on a prospective trial. Using low-coverage whole-genome sequencing, tumor-associated copy-number variations in CSF-derived cfDNA are investigated as an MRD surrogate. MRD is detected at baseline in 85% and 54% of patients with metastatic and localized disease, respectively. The number of MRD-positive patients declines with therapy, yet those with persistent MRD have significantly higher risk of progression. Importantly, MRD detection precedes radiographic progression in half who relapse. Our findings advocate for the prospective assessment of CSF-derived liquid biopsies in future trials for medulloblastoma.

Embryonal tumors with multilayered rosettes (ETMRs) are highly lethal infant brain cancers with characteristic amplification of Chr19q13.41 miRNA cluster (C19MC) and enrichment of pluripotency factor LIN28A. Here we investigated C19MC oncogenic mechanisms and discovered a C19MC-LIN28A-MYCN circuit fueled by multiple complex regulatory loops including an MYCN core transcriptional network and super-enhancers resulting from long-range MYCN DNA interactions and C19MC gene fusions. Our data show that this powerful oncogenic circuit, which entraps an early neural lineage network, is potently abrogated by bromodomain inhibitor JQ1, leading to ETMR cell death.