Cancer comprises a diverse group of diseases with significant morbidity and mortality. Nearly all common cancers exhibit some form of sexual dimorphism, for example in incidence, prognosis, or response to therapy. This dimorphism has been hypothesized to derive from differences between males and females in hormones, sex chromosomes, and environmental exposures; however the molecular basis of these disparities remains largely unknown. An understanding of this dimorphism is fundamental to precision medicine in cancer, and may lead to discovery of novel biomarkers, therapeutic targets, and improved outcomes. We propose to discover the molecular basis of sexual dimorphism in cancer though the following Aims: Aim 1 will characterize sexual dimorphism in gene expression and its regulation within and between tumor types of NCI's The Cancer Genome Atlas (TCGA). Leveraging TCGA, we will characterize sexual dimorphism at the transcriptome level and its potential genomic and epigenetic causes within and across cancers. Aim 2 will characterize sexual dimorphism in the heritable genetic component of cancer susceptibility across common cancers. Utilizing data from genome-wide association studies of cancer, we will test multiple genetic models that may explain sexually dimorphic traits of individual cancers. We will test a multifactorial model for genetic risk, whereby the same alleles affect both sexes, but the lower-risk sex has a higher threshold and would require more or stronger genetic risk factors to develop disease. We will test for specific risk or protective factors encoded on the X or Y chromosome that affect the sexes differentially. We will test for gene- sex interactions whereby autosomal loci act on risk in a sex-dependent manner. We will test for the presence of autosomal risk factors with different penetrance or effects in males and females due to hormonally-mediated or other sexual dimorphism acting at the gene expression level. Aim 3 will characterize sexual dimorphism in response to therapeutics and define the molecular features and mechanisms contributing to that dimorphism. Utilizing molecular and phenotypic drug response data from over 1000 cancer cell lines from the Cancer Genome Project (CGP), we will build sex-specific predictive models of drug response, that we will then apply to gene expression data from the full set of TCGA tumor samples to impute a sex-specific drug response for each TCGA sample. We will correlate imputed drug responses to TCGA tumor molecular features, with focus on those identified in Aims 1 and 2. We will functionally validate predicted sexually dimorphic response to therapy using cell models, including panels of lymphoblastoid cell lines, human hepatocytes and cancer cell lines. The results of this study will define genetic and genomic features that underlie sexual dimorphism in cancer biology, susceptibility, and response to therapeutics.

R01CA229618

2019-06-01

2024-05-31