Genetics
Mendelian randomisation finds no causal link between Lp(a) and type 2 diabetes despite an inverse observational correlation, in 563,420 participants (Arch Med Sci 2024)
Original title: Association between Lp(a) and T2D: a Mendelian randomization study
Using two-sample Mendelian randomisation with data from the UK Biobank and FinnGen consortia (563,420 participants), the authors tested whether genetically predicted Lp(a) causally influences type 2 diabetes (T2D) risk, a relationship previously observed to be inversely correlated. The inverse-variance-weighted analysis found no statistically significant association between genetically predicted Lp(a) and T2D (P=0.362). The findings suggest that the inverse relationship between Lp(a) and T2D seen in observational studies is not causal, which has implications for interpreting Lp(a)-lowering therapy expected effect, or lack thereof, on diabetes risk.
Original abstract
Introduction: Blood lipoprotein(a) (Lp(a)) levels have been observed to be inversely correlated with type 2 diabetes (T2D). In this Mendelian randomization (MR) study, the causal impact of genetically predicted Lp(a) on T2D was assessed.
Methods: A two-sample MR analysis was conducted. Data were obtained from UK Biobank and FinnGen consortia. Primary analysis was based on an inverse-variance-weighted mean (IVM) approach.
Results: No statistically significant association between the genetically predicted levels of Lp(a) and T2D was detected (p = 0.362) in IVM analysis involving data of 563,420 patients.
Conclusions: Genetically predicted Lp(a) concentration does not appear to be causally related to the risk of T2D.
Summary written by lp-a.org from the published abstract; figures as published. Page updated 18 August 2026. Methods.