Integrating Genomics and Metabolomics to Develop Predictive Models of Prostate Cancer in Multiethnic Men - PROJECT SUMMARY/ABSTRACT Prostate cancer (PCa) is the second leading cause of cancer death among American men, with men of African ancestry have the highest PCa incidence and mortality rates. While the causes of this notable health disparity are unknown, there is evidence of genetic contributions and it is likely that environmental factors contribute as well. However, most PCa research has focused on men of European descent, particularly genome-wide association studies (GWAS), which has resulted in polygenic models having poorer predictive value in non- Europeans. The overarching goal of this research is to identify genomic and metabolomic factors that contribute to PCa risk in multiethnic men. To facilitate the construction of a genome-wide multiethnic polygenic risk score (PRS), in the K99 phase of this research, Dr. Darst will develop a novel variant selection algorithm to identify informative variants among genome-wide data (Aim 1). She will then construct a multiethnic genome-wide PRS using a large multiethnic PCa sample (PRACTICAL, N=237,380) (Aim 2). This PRS will be developed on overall PCa (Aim 1A), aggressive PCa (Aim 2B), and age of PCa onset (Aim 2C) and is expected to lead to improved predictive value compared to a PRS of ~150 known variants and to a PRS developed using Europeans only. Dr. Darst's expertise in genetic epidemiology will be complemented with additional training in cancer epidemiology, advanced statistical genomics, and cancer health disparities received through coursework, seminars, conferences, and guidance provided by her expert mentoring team. In the R00 phase, Dr. Darst will initiate a new line of research the applies integrative techniques to investigate combined genomic and metabolomic factors influencing PCa risk. This will build directly upon the research and training received in the K99 period. She will identify genetically-regulated metabolites that could be causally associated with PCa in multiethnic populations (Aim 3). This will require developing a large multiethnic metabolomics imputation panel and using subsequent meta-analysis summary statistics to develop genome-wide multiethnic PRS for each investigated metabolite. These PRS will be used to impute metabolomics into the PRACTICAL consortium. A metabolome- wide association study will then be performed to identify imputed, or genetically-regulated, metabolite levels that are predictive of overall PCa or aggressive PCa. Using a subset of 1,186 African American men from the Multiethnic Cohort (MEC) and the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial, Dr. Darst will investigate metabolites that could mediate the effects of genetic factors, including the PRS from Aim 2, on overall and aggressive PCa (Aim 4a). She will also use integrative techniques to identify genomic and metabolomic factors that distinguish subgroups of individuals with high PCa risk (Aim 4b). Results are expected to substantially improve our ability to detect high risk individuals who would benefit from earlier or more intensive PCa screening across multiethnic populations and provide novel biological mechanisms to target for preventative measures, likely reducing PCa mortality and the number of indolent PCa cases treated unnecessarily.
