Causal associations between CD40/CD40L and aortic diseases: A mendelian randomization study.

Abstract

Background: CD40 and CD40L have been reported as associated with aortic dissection (AD) and aortic aneurysm (AA), but the causality of the associations has not been established yet. Methods: We conducted a two-sample Mendelian randomization (MR) study to assess the causal inference between CD40/CD40L and aortic diseases including AD and AA. The instrumental variables (IVs) for CD40 and CD40L were selected from a high-quality protein quantitative trait loci dataset released by a genomic study involving 30,931 individuals of European ancestry. The genome-wide association studies summary statistics for AD and AA were from the FinnGen Release 7, with 288638 controls for all outcomes of interests, 680 cases for AD and 6,092 cases for AA, also from European ancestry. For AA subtypes, there were 5,881 cases of thoracic AA (TAA) and 2,434 cases of abdominal AA (AAA) respectively. Inverse-variance weighted and Wald ratio were applied for calculating causal estimates. Horizontal pleiotropy and heterogeneity were assessed using MR-Egger regression analysis and Cochran Q test, respectively. Leave-one-out analyses were further performed. Results: Three single-nucleotide polymorphisms (SNPs) for CD40 and one SNP for CD40L were selected as IVs. We found genetic proxied CD40 levels inversely associated with the risk of AD (odds ratio [OR]: 0.777, 95% confidence interval [CI]: 0.618-0.978, p = 0.031) and AA (OR: 0.905, 95% CI: 0.837-0.978, p = 0.012), consistent across TAA (both p < 0.050). There were trends of increased risks of AD and AA in the presence of CD40L while not reaching statistical significance. No significant horizontal pleiotropy or heterogeneity was observed. Conclusion: Our MR study provides evidence supporting the causal association between CD40 and the reduced risks of both AD and AA.