Abstract
Digestive system cancers (DSCs) present a global health challenge. This study aimed to investigate the causal relationships between cathepsins and DSCs using Mendelian Randomization (MR) analysis. Single Nucleotide Polymorphisms (SNPs) associated with nine cathepsin types (cathepsins B, E, F, G, H, O, S, L2, Z) were obtained from the INTERVAL study (3301 individuals). Genetic association data for DSCs were obtained from a large-scale GWAS conducted by Sakaue et al. (Nat Genet 53: 1415-1424, 2021), while SNPs associated with cathepsin expression were sourced from the INTERVAL study. (esophageal cancer: 998 cases, 475,308 controls; gastric cancer: 1,029 cases, 475,087 controls; colorectal cancer: 6,581 cases, 463,421 controls; hepatic cancer: 379 cases, 475,259 controls; pancreatic cancer: 1,196 cases, 475,049 controls). Inverse variance weighted (IVW), MR-Egger, and weighted median methods were adopted to conduct univariable MR (UVMR), reverse MR, and multivariable MR (MVMR) analyses to estimate causality. The UVMR analyses discovered that elevated SNP-predicted cathepsin S expression were significantly associated with increased colorectal cancer risk(IVW: p = 0.0162, odds ratio (OR) = 1.0513, 95% confidence interval (CI) = 1.0093-1.0951), whereas cathepsin H exhibited a protective role against pancreatic cancer(IVW: p = 0.0410, OR = 0.9131, 95% CI = 0.8369-0.9963). These findings were corroborated by the MVMR analyses (cathepsin S: IVW: p = 0.0040, OR = 1.0725, 95%CI = 1.0233-1.1252; cathepsin H: IVW: p = 0.0160, OR = 0.8851, 95%CI = 0.8025-0.9773). No reverse causality was found. Our findings suggested the potential causalities of cathepsins S on colorectal cancer and cathepsin H on pancreatic cancer, indicating their potential as biomarkers and therapeutic targets in DSCs. Further research is necessary to integrate these genetic associations into clinical applications.