Cancer subtype identification using somatic mutation data

Abstract

BACKGROUNDWith the onset of next generation sequencing technologies, we have made great progress in identifying recurrent mutational drivers of cancer. As cancer tissues are now frequently screened for specific sets of mutations, a large amount of samples has become available for analysis. Classification of patients with similar mutation profiles may help identifying subgroups of patients who might benefit from specific types of treatment. However, classification based on somatic mutations is challenging due to the sparseness and heterogeneity of the data.\n\nMETHODSHere, we describe a new method to de-sparsify somatic mutation data using biological pathways. We applied this method to 23 cancer types from The Cancer Genome Atlas, including samples from 5, 805 primary tumors.\n\nRESULTSWe show that, for most cancer types, de-sparsified mutation data associates with phenotypic data. We identify poor prognostic subtypes in three cancer types, which are associated with mutations in signal transduction pathways for which targeted treatment options are available. We identify subtype-drug associations for 14 additional subtypes. Finally, we perform a pan-cancer subtyping analysis and identify nine pan-cancer subtypes, which associate with mutations in four overarching sets of biological pathways.\n\nCONCLUSIONSThis study is an important step towards understanding mutational patterns in cancer.