Abstract
Growth differentiation factor 15 (GDF-15) levels have been revealed as a robust biomarker for metformin use. We conducted Mendelian randomization (MR) analysis to explore the association between GDF-15 and gallstone disease to inform potential therapeutic effects targeting GDF-15. Four genetic variants associated with GDF-15 levels at p < 5 × 10(-8) were selected as instrumental variables from a genome-wide association meta-analysis including 21,758 individuals. Two-sample MR analysis was conducted using summary-level data from UK Biobank (10,520 gallstone cases and 350,674 controls) and FinnGen consortium (19,023 gallstone cases and 195,144 controls). Polygenic risk score analysis using individual-level data in UK biobank was performed to complement the MR findings by examining the non-linearity of the association. Diabetic complications were taken as positive controls to validate the therapeutic effect of targeting GDF-15. Linear and nonlinear associations between genetically predicted GDF-15 levels and gallstones were estimated with stratification by the diabetic status. In the two-sample MR analysis, the odds ratio (OR) of gallstones was 1.09 (95% confidence interval (CI), 1.03-1.15; p = 0.001) for one standard deviation increase in genetically predicted GDF-15 levels in the meta-analysis of two datasets. Polygenic risk score analysis found this association to be U-shaped (p = 0.037). The observed association was predominantly seen in nondiabetic population (OR = 1.11, 95% CI: 1.01-1.21; p = 0.003). An inverse association between genetically predicted GDF-15 levels and diabetic complications (OR = 0.77, 95% CI: 0.62-0.96; p = 0.023) was observed, validating the potential therapeutic effects of targeting GDF-15 levels. This MR study indicates that the increased risk of gallstone disease should be taken into account when considering GDF-15 as a therapeutic target for diabetic complications.