Testing for Causal Association Between Serum Urate, Gout, and Prostatic Cancer in European Males

We conducted a two-sample Mendelian randomization (MR) study including only European men to test for a causal relationship between serum urate (SU), gout, and prostate cancer. Using genome-wide association (GWAS) data, we generated instrumental variables (IVs) associated with gout and urate. These included 20 single nucleotide polymorphisms (SNPs) associated with gout but not urate (non-hyperuricemia compartment of gout) and four SNPs from loci containing urate transporter genes for an IV representing urate levels. MR methods included the inverse-variance weighted (IVW) method, MR-Egger regression, and the weighted median method. The non-hyperuricemia compartment of gout IV showed a causal effect of gout on prostate cancer (weighted median: p = 0.01). In contrast, the SU IV showed no evidence of a causal effect of SU on prostate cancer (IVW: p = 0.83; weighted median: p = 0.97). MR-Egger showed no evidence of horizontal pleiotropy (gout: p = 0.33; urate: p = 0.80). Loci contributing most strongly to the non-hyperuricemia causal effect included three genes: IL1R1, IL1RN, and SLC30A5. There was no evidence of a causal relationship between prostate cancer and gout or SU. In conclusion, MR analysis in a European male population found evidence of a causal relationship between the non-hyperuricemia compartment of gout and prostate cancer. Implication of the IL1R1 and IL1RN genes directly implicates the gouty inflammation pathway in prostate cancer.