The role of 1400 plasma metabolites in gastric cancer: A bidirectional Mendelian randomization study and metabolic pathway analysis

While observational studies have illustrated correlations between plasma metabolites and gastric cancer (GC), the causal association between the 2 is still unclear. Our study aims to delineate the bidirectional relationship between plasma metabolites and GC and find potential metabolic pathways. We undertook a bidirectional 2-sample Mendelian randomization (MR) analysis to investigate the causal relationship, specificity, and direction of association between 1400 plasma metabolites and GC. The GWAS data for metabolites was obtained from a cohort of 8299 European individuals. And the GC’s GWAS data was from FinnGen Consortium with 2384 European individuals, and the GWAS catalog with 1029 European ancestry cases for validation. Causal estimates were primarily calculated by the inverse-variance weighted (IVW) method. To ensure robustness, we performed comprehensive sensitivity analyses to assess heterogeneity and address concerns regarding horizontal pleiotropy. We validated the forward relationship between metabolites and GC from another database and implemented meta-analysis. Furthermore, we conducted metabolic enrichment and pathway analysis of these causal metabolites using MetaboAnalyst5.0/6.0 with the database of Kyoto Encyclopedia of Genes and Genomes. All statistical analysis was carried out using R software. Metabolites like 2s, 3R-dihydroxybutyrate, 4-acetamidobutanoate, ferulic acid 4-sulfate and methyl indole-3-acetate was proven positively linked with the development of GC. Asparagine, glucose to maltose ratio, glycohyocholate, Gulonate levels, linoleoyl ethanolamide and Spermidine to (N(1) + N(8))-acetylspermidine ratio was proven to be negatively associated with GC. Moreover, linoleic acid , histidine, glutamine, bilirubin , Succinate to proline ratio were found to be potentially linked to the development of GC. Furthermore, our analysis identified 18 significant metabolic pathways, including Arginine and proline metabolism ( P < .009) and Valine, leucine, and isoleucine biosynthesis ( P < .031). Our findings offer evidence supporting potential casual relations between multiple plasma metabolites and GC. These findings may offer great potential for future application of these biomarkers in GC screening and clinical prevention strategies.