Plasma proteome-wide Mendelian randomization reveals multi-ancestry drug targets for gastric cancer

Background Gastric cancer (GC) exhibits marked epidemiological differences between European (EUR) and East Asian (EAS) populations, with significantly higher incidence rates in EAS. Circulating proteins represent promising drug targets; however, most proteomic studies have focused primarily on EUR ancestry, leaving EAS-specific targets largely underexplored. This study aims to identify ancestry-specific plasma protein targets for GC using Mendelian randomization (MR). Methods We employed Mendelian randomization (MR) and colocalization approaches to assess the putative causal effects of plasma proteins on GC risk across diverse populations. Specifically, we examined proteins in EUR and EAS populations. Subsequent single-cell RNA sequencing and molecular docking were conducted to identify cellular expression patterns and potential therapeutic compounds. Finally, experiments were conducted to verify the newly discovered drug target. Results Our analyses revealed four significant protein targets in the EUR population (SLURP1, ANGPTL3, NME4, ANXA10) and nine in the EAS population (ICAM5, SMOC1, PSCA, SATB1, SCRG1, PTPRB, ISLR2, NCR3LG1, SELE) associated with GC susceptibility. These drug targets are mainly expressed in epithelial cells, and Pit mucous cells in gastric tumor tissue. Experimental validation indicated significant downregulation of NME4 and upregulation of ICAM5 in GC cell lines and tissue samples, suggesting potential tumor-suppressive and oncogenic functions, respectively. Moreover, molecular docking identified seven repurposable drugs, with digoxin demonstrating cross-ancestry therapeutic potential. Conclusion Our findings delineate 13 ancestry-specific protein targets, including five novel candidates, thereby enhancing our understanding of GC pathogenesis. This study underscores the importance of ancestral diversity in informing drug development strategies for GC. Future research should focus on validating these targets and exploring their functional mechanisms further to translate these findings into clinical applications.