Genetic associations between SGLT2 inhibition, DPP4 inhibition or GLP1R agonism and prostate cancer risk: a two-sample Mendelian randomisation study

Background: Epidemiological studies have linked the use of the anti-diabetic medications, sodium-glucose co-transporter-2 inhibitors (SGLT2I), dipeptidyl peptidase-4 inhibitors (DPP4I) and glucagon-like peptide-1 receptor agonists (GLP1RA), with prostate cancer risk. However, these studies cannot infer causality. Methods: This was a two-sample Mendelian randomization (MR) using genome-wide association study data designed to identify causal relationships between SGLT2I, DPP4I or GLP1RA and prostate cancer. Genetic associations with HbA1c and risk of prostate cancer were extracted from IEU Open-GWAS Project database with GWAS id ukb-d-30750_irnt (UK Biobank cohort) and ebi-a-GCST006085 (European Molecular Biology Laboratory's European Bioinformatics Institute cohort), respectively. The two GWAS datasets chosen were obtained from individuals of European ancestry to minimise potential bias from population stratification. The encoding genes targeted by SGLT2I, DPP4I and GLP1RA were SGC5A2, DPP4 and GLP1R, located in Chr16: 31494323-31502181, Chr2: 162848755-162930904 and Chr6: 39016557-39059079, respectively. Results: A total of 31, 2 and 5 single nucleotide variants (SNVs) were used for SGC5A2, DPP4 and GLP1R. Our MR analysis results supported a causal relationship between genetic variation in SLC5A2 and DPP4 and reduced risk of prostate cancer at the Bonferroni-corrected threshold, with odds ratios (OR) [95% confidence intervals] of 0.47 [0.38-0.58] and 0.35 [0.24-0.53], but not for GLP1R (OR: 1.39 [0.93-2.07]). Sensitivity analyses by the leave-one-out method did not significantly alter the OR for SGLT2I. Conclusions: The two-sample MR analysis found that SGLT2 and DPP4 inhibition, but not GLP1R agonism, was associated with lower risks of developing prostate cancer.