Abstract
Purpose
To evaluate causal effects of lipid traits and lipid-lowering drug targets on the risk of type 1 diabetes (T1D) and its complications.
Methods
Our study conducted two-sample and drug-target Mendelian randomization (MR) to assess the genetic association of lipid traits and lipid-lowering drug targets with the type 1 diabetes risk, respectively. For significant lipid-modifying drug targets, data for expressions in tissues and colocalization provided extra evidence for causality. We also explored underlying mechanisms through mediation MR.
Results
The two-sample MR analyses detected no causal association between lipid traits and T1D. In the drug-target MR analyses, ANGPTL3 inhibitor was associated with a decreased risk of T1D (OR = 0.668, 95% CI: 0.511–0.874, P = 3.21*10− 3), of which BMI mediated 5.71% of the total effect. This was validated through multiple sensitivity analyses, replication dataset and tissue sample data. Moreover, ANGPTL3 inhibitor was also found to reduce the risk of diabetic kidney diseases. Although HMGCR inhibitor reduced the risk of T1D in the primary dataset, it was not validated in the replication dataset, and HMGCR inhibitor showed adverse effects on diabetic retinopathy and neuropathy.
Conclusion
Circulating lipids are not causally associated with the risk of T1D. ANGPTL3 inhibitor, a novel lipid-lowering drug, may be a promising candidate for treating T1D and its renal complication, with BMI probably mediating part of the effect.