Abstract PR13: DNA methylation mapping and computational modeling in a large Ewing sarcoma cohort identifies principles of tumor heterogeneity and their impact on clinical phenotypes

摘要: Ewing sarcoma is an excellent model for studying the role of epigenetic deregulation and tumor heterogeneity, given its low mutation rates well-defined oncogenic driver. We have recently shown that fusion oncogene EWS-FLI1 induces widespread rewiring in proximal distal enhancers (Tomazou et al. Cell Reports 2015). In current study, we validate clinical relevance our results a large cohort primary tumors, explore prevalence, characteristics, impact heterogeneity sarcoma. used reduced representation bisulfite sequencing (RRBS) to generate genome-wide profiles DNA methylation 141 17 cell lines, 32 mesenchymal stem (MSC) samples. Deep resulted measurements average 3.5 million unique CpGs per sample with data quality (>98% conversion efficiency). addition, three tumors generated comprehensive reference epigenome maps using whole genome (WGBS) ChIP-seq seven histone marks (H3K4me3, H3K4me1, H3K27me3, H3K27ac, H3K56ac, H3K36me3, H3K9me3). show can be infer enhancer activity differences among allowing us exploit dataset systematically compare regulation correlated anticorrelated enhancers. also identified Ewing-specific patterns. For example, samples consistently higher than MSCs at AP-1 binding sites, but lower sites. To within individual developed bioinformatic algorithm quantifies disorder. Using reads containing multiple from single cells, assign scores single-nucleotide resolution. This method uses probabilistic account overall rate expected disorder levels. By evaluating likelihood assumes status CpG independent nearby CpG, identify extremely heterogeneous as well highly epigenetically conserved genomic elements. These different region types distinct patterns enrichment regulatory modes transcription factor binding. compared observed those MSC samples, several hundred additional normal are unrelated analysis stratifies patients based on heterogeneity. Our constitutes largest available resource genome-scale solid pediatric tumor. It strongly confirms reprogramming sarcoma, it starting point develop biomarkers prognosis patient stratification. study supported by Austrian National Bank (OeNB project #15714) Kapsch group (https://www.kapsch.net/). abstract presented Poster A24. Citation Format: Nathan C. Sheffield, Franck Tirode, Sandrine Grossetete-Lalami, Paul Datlinger, Andreas Schonegger, Johanna Hadler, Diana Walder, Ingeborg M. Ambros, Ana Teresa Amaral, Enrique de Alava, Katharina Schallmoser, Dirk Strunk, Beate Rinner, Bernadette Liegl-Atzwanger, Berthold Huppertz, Leithner, Uta Dirksen, Peter Olivier Delattre, Heinrich Kovar, Christoph Bock, Eleni Tomazou. mapping computational modeling identifies principles their phenotypes. [abstract]. In: Proceedings AACR Special Conference Advances Pediatric Cancer Research: From Mechanisms Models Treatment Survivorship; 2015 Nov 9-12; Fort Lauderdale, FL. Philadelphia (PA): AACR; Res 2016;76(5 Suppl):Abstract nr PR13.