Platelet count and gastric cancer susceptibility: A Mendelian randomization study

Abstract

The objective of this study was to ascertain the potential causal linkage between platelet (PLT) counts and the incidence of gastric cancer (GC). This study employed a 2-sample Mendelian randomization (MR) approach, utilizing the inverse variance weighting, weighted median, and MR-Egger regression methodologies. The publicly accessible summary statistics dataset from the genome-wide association study pertaining to individuals of European ancestry (n = 145,648) was employed as the foundational resource for the exposure variable. Concomitantly, the non-cancer disease codes for GC (n = 6563), derived from individuals within the UK Biosample Bank, were utilized as the outcome measure. A set of 132 single-nucleotide polymorphisms exhibiting genome-wide significance were selected as instrumental variables, drawn from the genome-wide association studies focused on PLT counts. The application of the weighted median methodology yielded indications suggesting the possible absence of a causal relationship between PLT counts and GC (beta = 0.139, SE = 0.079, P = .077). Contrarily, the implementation of the inverse variance weighting technique produced results indicative of a potential causal relationship between PLT counts and GC (beta = 0.128, SE = 0.049, P = .009). The assessment of Cochran Q test and the scrutiny of funnel plots unveiled no discernible indications of heterogeneity or asymmetry, thus signifying the absence of directional pleiotropy. The outcomes derived from the MR analysis lend credence to the hypothesis that there exists a plausible causal relationship between erythrocyte pressure and an elevated susceptibility to gastric cancer.