Local Genetic Correlations and Pleiotropy Reveal Shared Genetic Architecture Between COVID-19 Phenotypes and Prostate Cancer in European Populations

TL;DR

This study explores the shared genetic links between prostate cancer and different outcomes of COVID-19, finding common genes and regions but no direct cause-effect relationship.

Contribution

The study is the first to comprehensively analyze shared genetic architecture between prostate cancer and multiple COVID-19 phenotypes using large-scale GWAS data.

Findings

Significant local genetic correlations were found on chromosomes 1, 7, and 14 between prostate cancer and at least one COVID-19 phenotype.

Twenty-two independent SNPs and eight shared genes were identified, primarily in respiratory, neurological, and reproductive tissues.

No causal relationship was found between prostate cancer and any of the studied COVID-19 phenotypes.

Abstract

Background: While a link between coronavirus disease 2019 (COVID-19) and prostate cancer (PrCa) has been observed in clinical settings, the shared underlying genetic influences remain unclear.

Methods: Leveraging summary statistics from the hitherto largest genome-wide association studies (GWASs) of European-ancestry populations, we performed the first comprehensive genome-wide cross-trait analysis to investigate the shared genetic architecture, pleiotropy, and potential causal relationships between three COVID-19 phenotypes and PrCa.

Results: We found no evidence of significant genome-wide genetic correlations between COVID-19 phenotypes and PrCa (P > 0.05). However, after partitioning the whole genome into 2353 independent regions, we observed significant local genetic correlations at chromosome 1 (chr1), chr7, and chr14 for PrCa with at least one COVID-19 phenotype (P < 0.05/2353). Cross-trait meta-analysis identified 22 independent single nucleotide polymorphisms (SNPs) shared between PrCa and at least one COVID-19 phenotype, totalling 25 associations, including 2 with infection, 14 with hospitalization, and 9 with critical illness. Transcriptome-wide association study (TWAS) revealed eight distinct shared genes (CCHCR1, TCF19, ADAM15, HLA-C, CYP21A1P, HCP5, ATF6B, and HLA-DQB2), predominantly enriched in tissues of the respiratory, neurological, and reproductive systems. Bidirectional Mendelian randomization (MR) demonstrated no causal association between COVID-19 phenotypes and PrCa.

Conclusions: Using a multi-layered analytical framework, our study provides novel insights into the shared genetic bases between COVID-19 phenotypes and PrCa, supported by significant local genetic correlations, pleiotropic SNPs, and shared genes. These findings highlight common biological mechanisms rather than direct causal relationships, suggesting the limited benefits of additional PrCa screening in COVID-19 survivors. Furthermore, the identified genes represent promising targets for future mechanistic research and clinical interventions, warranting further validation.

Figures