Document resource
Background Population-based Helicobacter pylori (H.pylori) screen-and-treat is a public health priority for gastric cancer (GC) prevention. However, the preventive effect varies across individuals, and the determinants of this variability remain insufficiently understood. Understanding such heterogeneity is important as large-scale eradication efforts expand globally. We investigated genetic variants related to GC development and H.pylori infection to characterize genetically determined heterogeneous treatment effects (HTEs) of H.pylori eradication for GC prevention.Methods Two randomized trials were analyzed, including the Mass Intervention Trial in Linqu, Shandong Province (MITS; ChiCTR-TRC-10000979; n=1740) and the Shandong Intervention Trial (SIT; NCT00155389; n=1811). Genotype-specific conditional average treatment effects (CATEs) were estimated using causal machine learning approaches for variants modifying H.pylori treatment effects in both trials, and were integrated into a polygenic treatment benefit score (PTBS) to predict preventive benefit for GC by H.pylori eradication. We compared a benefit-based strategy prioritizing individuals predicted to derive greater benefit for GC prevention, with a risk-based strategy defined by a polygenic risk score, and assessed whether benefit-based subgroups exhibited differential residual GC risk post-eradication.Results Eleven variants were identified to modify the preventive efficacy of H.pylori treatment, with their CATEs aggregated into the PTBS. In MITS, targeting the high-treatment-benefit subgroup reduced GC risk more substantially (hazard ratio[HR]=0.44, 95% confidence interval[CI]:0.35-0.54) than targeting the high-risk subgroup (HR=0.73, 95%CI:0.59-0.91, P-for-heterogeneity<0.001). Similarly, in SIT, the high-treatment-benefit subgroup showed a lower GC risk post-treatment (HR=0.25, 95%CI:0.08-0.76) than the high-risk subgroup (HR=0.58, 95% CI:0.37-0.93). PTBS stratified GC risk post-eradication (MITS: HR=0.65, 95%CI:0.52-0.81; SIT: HR=0.27, 95%CI:0.13-0.56, for high-treatment-benefit individuals) and was associated with slower precancerous lesion progression (IDDF2026-ABS-0386 Figure 1).Conclusions Host genetic factors modulate the preventive efficacy of H.pylori eradication. Incorporating genotype-specific HTEs provides a framework for enhancing the precision of population-based screen-and-treat programs and supports post-eradication risk management by identifying individuals with residual risk who may require more intensive surveillance.Abstract IDDF2026-ABS-0386 Figure 1