Serum EpCAM or PECAM levels and risk of ischemic stroke: a two-sample Mendelian randomization study

Abstract

Background and purpose Elevated serum Epithelial cell adhesion molecule (EpCAM) or Platelet endothelial cell adhesion molecule (PECAM) are associated with ischemic stroke (IS), but the causality remains unclear. A two-sample Mendelian randomization (MR) study was performed to examine the causal effect of serum EpCAM or PECAM levels on the risk of IS subtypes. Methods Seven independent single-nucleotide polymorphisms (SNPs) related to serum EpCAM levels were identified as instrumental variables (IVs) from a genome-wide association study (GWAS) of 1,323 European subjects. 46 independent SNPs related to serum PECAM levels were identified as IVs from a GWAS of 21,758 European subjects. GWAS for IS subtypes included 410,484 (large artery stroke), 198,048 (small vessel stroke) and 413,304 (cardioembolic stroke) European individuals, respectively. All GWAS were obtained from the ieu open GWAS project. Mendelian Randomization (MR) analysis was conducted using the inverse variance-weighted (IVW) method, weighted median, MR–Egger, and maximum likelihood methods. Results No significant causal association was observed for EpCAM levels with any of three IS subtypes. Main IVW MR analysis indicated that serum PECAM levels were negatively related to the incidence of large artery stroke (LAS) and small vessel stroke (SVS) but not to cardioembolic stroke (CES). However, MR Egger method indicated a significant negative correlation between PECAM levels and CES risk. Sensitivity analyses showed the SNPs of PECAM exhibited pronounced horizontal pleiotropy in the CES GWAS. Therefore, serum PECAM levels are negatively correlated with CES risk significantly. In addition, no heterogeneity or outliers were found by MR-Egger and IVW in Cochran’s Q statistic. Finally, Leave-one-out analysis showed that no individual SNP substantially drove the causal association of serum PECAM levels on risk of IS subtypes. Conclusions High serum EpCAM levels show no clear causal relationship with the risk of LAS, SVS, and CES. But high serum PECAM levels were causally associated with decreased risks of LAS, SVS and CES.