Abstract
Evidence of a causal link between sclerostin-lowering interventions and ischemic stroke remains uncertain due to the limitations of previous observational studies and the post hoc analyses of randomized trials. Therefore, this study aims to investigate the causal effect of plasma sclerostin abundance on incident ischemic stroke and its subtypes. To do this, we conducted a drug-target Mendelian randomization analysis. First, we established a primary hypothesis holding that genetically perturbed plasma sclerostin levels are causally associated with any ischemic stroke (AIS). We then established exploratory hypotheses to test whether plasma sclerostin levels could affect the risk of subtypes of several types of ischemic stroke, in this case cardiometabolic stroke, large artery stroke (LAS), and small vessel stroke. For phenotypes that showed a statistically significant effect, we conducted Bayesian colocalization analyses to test for confounding by linkage disequilibrium. As a result, we identified a single cis-acting genetic instrument perturbing plasma sclerostin levels. A significant effect of reduced sclerostin levels on increased odds for AIS was observed (odds ratio per 1-unit decrease in sclerostin levels: 1.32; 95% confidence interval: 1.03-1.68; P = .027). Additionally, among the 3 subtypes of ischemic stroke, the odds for LAS were significantly affected by sclerostin levels under a Bonferroni-adjusted significance level (odds ratio: 2.71; 95% confidence interval: 1.31-5.61; P = .007). No statistical evidence of confounding by linkage disequilibrium was observed with AIS (PP.H3 = 4.13 × 10-3) or LAS (PP.H3 = .0104). In conclusion, reduced plasma sclerostin levels could increase the risk of ischemic stroke, particularly LAS, suggesting the importance of monitoring cerebrovascular risk in patients receiving sclerostin-reducing therapies.