Uncovering somatic genetic drivers in prostate cancer through comprehensive genome-wide analysis

Abstract

Given that hereditary prostate cancer (PCa) accounts for only a small fraction of PCa phenotypes, there is still a substantial journey ahead in exploring the somatic genetic drivers contributing to sporadic PCa. The expression quantitative trait loci (eQTLs) data were sourced from the GTEx dataset for prostate-specific genes, and the summary statistic information was collected for 5854 genes. Genetic associations with PCa were extracted from three well-established consortiums: the UK Biobank (9131 cases and 173,493 controls), the PRACTICAL study (79,148 cases and 61,106 controls), and the FinnGen cohort (13,216 cases and 119,948 controls). To prioritize potential causal targets, additional analysis, including the protein–protein interaction (PPI), The Cancer Genome Atlas (TCGA) dataset, and the single-cell-type expression analysis, was performed. Generally, a total of 150 common significant genes with the same causal association with PCa were identified. Out of the 150 genes examined, 67.33% (101/150) were found to have protein-coding functions, while only 30.67% (46/150) of these genes had prior mentions in the scientific literature. Notably, the analysis of the TCGA dataset showed that only 44.67% (67/150) of the genes produced consistent results with the Mendelian randomization (MR) analysis. Furthermore, the evaluation of single-cell RNA-seq data and colocalization analysis singled out MSMB as a critical gene associated with the occurrence of PCa. We pinpointed a range of prostate-specific genes that display causal associations with the onset of PCa. Among these, the MSMB gene emerged as a pivotal factor linked to PCa, demonstrating robust consistency across all four assessments, including the MR, TCGA dataset, single-cell RNA-seq data, and colocalization analysis. These findings provided fresh perspectives on the pathogenesis of PCa and presented potential targets for drug development.