PTPS-11ATYPICAL TERATOID/RHABDOID TUMOUR IS AN EPIGENETICALLY HETEROGENEOUS DISEASE CHARACTERIZED BY SUBGROUP SPECIFIC SUPER-ENHANCERS

Atypical teratoid/rhabdoid tumor (ATRT) is one of the most common brain tumors in infants or young children. Although the prognosis of ATRT patients is poor, some patients respond very well to current treatments, suggesting inter-tumor molecular heterogeneity. To investigate this, we have analyzed a large cohort of 152 ATRTs genetically and epigenetically. Three distinct molecular subgroups of ATRTs, associated with differences in patient's age, tumor location and type of SMARCB1 mutations, were identified using DNA-methylation and gene expression analyses which we have termed ATRT-TYR, ATRT-SHH, ATRT-MYC after genes and pathways that were highly overexpressed in the subgroups. Whole genome DNA- or RNA-sequencing found no other additional recurrent mutations explaining the differences between subgroups. However, whole genome bisulfite-sequencing and H3K27ac ChipSequencing of primary tumors revealed clear epigenetic differences and showed that ATRT genomes in two of the three subgroups were overall hypermethylated: Large partially methylated domains (PMDs) were only frequently present in subgroups ATRT-SHH and ATRT-MYC and genes encompassed by these regulatory elements were often downregulated in these subgroups. By statistical testing, we derived subgroup specific enhancers from the H3K27ac ChipSequencing data that characterize the three subgroups and developed a model for enhancer-gene regulation. Among the subgroup specifically regulated genes were key components of the SHH pathway for ATRT-SHH (such as GLI2) and members of the melanogenesis pathway (such as the transcription factor MITF) for subgroup ATRT-TYR. Having identified MITF as a central regulatory hub in this subgroup, we performed ChipSequencing for this molecule and found that genes previously implicated in ATRT biology (such as BMP4 and CCND1) were important targets of this transcription factor. In summary, by examining various levels of data, we could find clear epigenetic evidence for three ATRT subgroups and identified regulatory molecules for the subgroups that may be used as a target in subsequent clinical studies.