Abstract 2720: Unbiased identification of RNA splicing-defined patient populations in lung adenocarcinoma

Abstract

Abstract Aberrant alternative splicing is known to directly contribute to tumor progression in lung adenocarcinoma (LUAD). While the action of individual splicing variants as cancer drivers have been characterized in LUAD, the global splicing landscape and how that corresponds to patient prognosis is poorly understood. Using whole transcriptome RNA-Seq data provided from The Cancer Genome Atlas (TCGA), we applied a new computational pipeline, OncoSplice, to identify potentially novel prognostic disease splicing subtypes and elucidate the biological mechanisms underlying these subtypes. OncoSplice identified over 30 novel splicing-defined subtypes of LUAD, several of which correlate with worse patient survival. Additionally, we identified patient clusters that were significantly enriched in mutations for RNA binding proteins, such as Cap Methyltransferase 2 (CMTR2) and RNA Binding Motif 10 (RBM10). More specifically, mutations in the splicing factor RBM10 were found to be enriched but not requisite for the dominant poor prognosis splicing subtype in LUAD. In non-mutant patients within this subset, RBM10 gene expression downregulation was induced, mimicking the mutation phenotype. Hence, we find alternative non-genetic pathways for impacting splicing in LUAD that leads to poor prognosis through disruption of canonical splicing. Citation Format: Audrey Crowther. Unbiased identification of RNA splicing-defined patient populations in lung adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2720.