Causal inference between rheumatoid arthritis and prostate cancer

Abstract

It is unclear if the association between rheumatoid arthritis (RA) and a higher risk of prostate cancer (Pca) reflects a causal relationship. We conducted a meta-analysis and used the Mendelian randomization method (MR) to evaluate the association between RA and Pca risk. A meta-analysis and subgroup analysis of the incidence of Pca in patients with RA was conducted. To determine whether genetically elevated RA levels were causally linked to Pca, two MR samples were employed. To eliminate gender-related bias, we conducted a stratified analysis of the GWAS data for RA by gender, specifically including 140,254 males. Additional MR analysis was also performed to determine potential confounding factors influencing the association between genetically susceptible RA and Pca. In total, 409,950 participants were enrolled in 20 trials to investigate the Pca risk in patients with RA. The meta-analysis suggested that RA was unrelated to the Pca risk (SIR = 1.072, 95% CI, 0.883-1.261). However, a subgroup analysis showed that low smoking rates might increase the Pca risk in patients with RA by 24%. The MR analysis showed that increased genetic susceptibility to RA was related to a high Pca risk (OR = 36.20, 95%CI = 1.24-1053.12, P = 0.037). The causality estimation of MR-Egger, Weighted mode, Simple mode, and Weighted median method were similar in direction and magnitude. Although our meta-analysis found no correlation between RA and Pca risk, MR analyses supported a causal relationship between genetic susceptibility to RA and increased prostate risk. Early attention to Pca risk in patients with RA may be important for improving prognosis and mortality in such patients. Further research is needed to determine the etiology of RA attributed to Pca and its underlying mechanisms.

Keywords: Genetic variants; Genome-wide association studies; Mendelian randomization; Prostate cancer risk; Rheumatoid arthritis.