Integrated multi-omics analyses reveal causal insights into the molecular landscape of urologic cancers

Background Urologic cancers continue to present an ongoing challenge to human health, underscoring the necessity for a systematic exploration of their underlying mechanisms. Hence, we employed a multi-omics Mendelian randomization (MR) approach to infer potential pathogenic genes and unravel the associated mechanisms. Methods Genome-wide association study (GWAS) summary statistics on urologic cancers, along with expression quantitative trait loci (QTLs), DNA methylation QTLs, and protein QTLs were employed for summary data-based MR (SMR), colocalization, and transcriptome-wide association study (TWAS) analysis to identify putative causal genes. Additionally, metabolites GWAS summary statistics from both blood and urine were analyzed by two-sample MR method to uncover potential relationships with the risk of urologic cancers. Results A total of 15 genes from blood were identified related to prostate cancer (PCa) risk using SMR, colocalization, and TWAS analysis, with C10orf32, MARVELD1, UHRF1BP1, and POLI were being replicated in tissue SMR analysis. Additionally, 15 genes were prioritized as potential causal genes with their methylation regulatory, encompassing both well-established genes (NSUN4, FAAH, and SIK2) as well as novel candidates, such as HAAO, and UHRF1BP1. Furthermore, four metabolites, including phenyllactate, N-lactoyl phenylalanine, indolelactate, and adenosine 3',5'-cyclic monophosphate, were positively associated with PCa risk. For bladder cancer, its risk was associated with the expression variation of GSTM1, BTNL8, and BTNL9 and PSCA methylation. Conclusion This integrative approach not only enhances our understanding of the complex molecular landscape underlying urologic cancers but also opens up new avenues for future precision medicine and personalized treatment strategies.