Association Between GLP-1 Receptor Agonists, Other Antidiabetic Drugs, and Diabetic Microvascular Complications: A Mendelian Randomization Analysis
Rijing Liu, Shiqi Tang, Siyu Li, Wengen Zhu, Hong QiuIntroduction/Objective:
To assess the causal effects of major antidiabetic drug classes on the risks of diabetic retinopathy (DR), diabetic nephropathy (DN), and diabetic peripheral neuropathy (DPN) using Mendelian randomization (MR) with genetic proxies for drug targets.
Methods:
Target genes were selected for metformin (PRKAB1, ETFDH), insulin and insulin analogs (INSR), thiazolidinediones (PPARG), sulfonylureas (KCNJ11/ABCC8), DPP- 4 inhibitors (DPP4), GLP1R agonists (GLP1R), and SGLT2 inhibitors (SLC5A2). Cis-e- QTL instruments were obtained from the eQTLGen Consortium. Two-sample MR analyses were performed using FinnGen GWAS summary statistics for DR, DN, and DPN. HbA1c levels and diabetes risk were included as positive controls. Sensitivity analyses included MR-Egger regression, MR-PRESSO, and the weighted median method. Colocalization analysis was performed to assess whether the observed associations were supported by shared genetic signals.
Results:
Genetically proxied KCNJ11 inhibition was associated with reduced risks of DR (OR = 0.806, FDR = 4.110 × 10−6) and DN (OR = 0.762, FDR = 6.046 × 10−3). Genetically proxied PRKAB1 activation reduced the risks of DR (OR = 0.969, FDR = 1.355 × 10−2) and DN (OR = 0.936, FDR = 1.603 × 10−2). Genetically proxied GLP1R activation was associated with a lower risk of DPN (OR = 0.573, FDR = 1.176 × 10−2), but not with DR or DN. Colocalization analyses supported the presence of shared causal variants.
Discussion:
These findings suggest that the potential protective effects of antidiabetic drug targets may differ across specific diabetic microvascular complications. The observed target- specific and complication-specific associations support a target-complication matching framework and may provide genetic evidence for more individualized therapeutic strategies in diabetes management.
Conclusion:
Genetic evidence suggests that perturbation of KCNJ11 and PRKAB1 protects against DR and DN, whereas GLP1R activation reduces the risk of DPN. Further studies are warranted to validate these findings and determine their therapeutic relevance.