Stem-02. syngn: a novel epigenetic platform for patient-specific drug-matching in glioblastoma

Abstract The alteration of the glioblastoma genome by epigenetic mechanisms that share functions with normal developmental processes, such as self-renewal and fate specification of NSC, is a key piece of evidence that links brain cancer pathogenesis with dysregulated stem cell functions. A patient-specific comparison of glioblastoma cells with NSC, a putative cell of origin of at least a proportion of these tumours, is not feasible as patient-matched endogenous NSC are not surgically accessible and all epigenetic studies in glioblastoma have so far compared epigenetic changes of different tumours with each other or to comparators obtained from foetal brains or an unrelated donor. We reasoned that availability of syngeneic GSC and NSC pairs would allow to identify crucial epigenetic differences on a patient-specific basis and would provide essential therapeutic contrast to define disease-and patient-intrinsic biomarkers of drug response that are less confounded by germline variation. We have derived GSC from IDH-wildtype glioblastoma and harnessed state-of-the-art stem cell technologies to generate patient-matched fibroblast-derived EPSC, which were induced to NSC (iNSC). We demonstrate that integrated analysis of the transcriptome and DNA methylome of GSC/iNSC pairs identifies druggable target genes (PTGER4, ALDH3B1, NTRK2) in a proportion of patients and we validate this patient-specific prediction of drug response at pre-clinical level in 3D synGLICO and in in vivo models. Integration of small RNA regulation to the pipeline allows refined in silico prediction of drug responses on a patient-specific basis. Most recently we have added assessment of chromatin remodelling (H3K4me3, H3K27ac, H3K3me3, H3K27me3 and ATACSeq) to the pipeline and shown that redistribution of selected histone marks as well as shifts in chromatin states across the genome identifies known and novel druggable regulatory mechanisms in GSC, which are specific of the neoplastic context. In conclusion, SYNGN is an epigenetic platform which can identify patient-specific druggable targets in glioblastoma.