PO-061 Exploring heritable predisposition to paediatric rhabdomyosarcomas

Abstract

Introduction Rhabdomyosarcomas (RMS) are the most prevalent form of paediatric soft tissue sarcomas. They arise from skeletal muscle progenitor cells and are divided predominantly into two subtypes, embryonic and alveolar, with relatively poor prognosis. The alveolar subtype appears to be driven by an oncogenic fusion gene protein, typically but not exclusively FOXO1-PAX3 or FOXO1-PAX7. Fusion negative tumours better define the embryonal subtype, with occasional RMS cases where both subtypes histologically appear to coincide. Previously published work on somatic RMS variants associate embryonal RMS with hits in the RAS/AKT pathway (KRAS, NRAS, HRAS, NF1, PIK3CA, PTEN) alongside other cancer-related genes (TP53, FGFR4, FBXW7, BCOR). In this study we generate new whole exome sequencing data and integrate with previously published datasets to study germline predisposition to RMS. Material and methods The study includes 3 published datasets (Shern 2014 – 120 samples; Seki 2015 – 16 samples and Zhang 2015 – 30 samples) in addition to our in-house set (36 tumour-normal pairs of both RMS subtypes, collected in collaboration with the Children’s Cancer and Leukaemia Group tissue bank). Our primary samples were processed using Illumina’s Nextera Rapid Capture kit and sequenced on HiSeq NGS machines. The combined analysis cohort consists of 202 cases of paediatric RMS. Results and discussions A customised bioinformatics pipeline was used for integration of the studied datasets: from raw fastqs through BWA-MEM alignment and GATK-HC joint variant calling. Germline variants passing several quality control filters were pruned for a MAF Conclusion We explored the potential for germline predisposition to RMS as genetic inheritance is common in a number of paediatric cancers. The relative rarity of paediatric RMS prevents sufficiently powered single-centre studies and requires the consolidation of all available data to make meaningful conclusions. We employ this approach to identify new candidate predisposition genes. Expanding our understanding of genes predisposing to rhabdomyosarcomas, could assist earlier diagnosis and therapy for these children, ultimately improving their survival.