Response to Letter to the Editor Regarding “PARP-1 rs1136410 Polymorphism and Gastrointestinal Cancer Risk: A Meta-Analysis of Cancer-Type and Ethnic-Specific Associations”

We sincerely thank the correspondents for their thoughtful comments on our meta-analysis examining the association between PARP-1 rs1136410 (Val762Ala) and gastrointestinal cancer risk, and for the opportunity to clarify several important points 1 .We agree that the ancestry-dependent associations observed in our study are of particular importance.In colorectal cancer, the C allele showed a protective effect in Caucasian cohorts but was associated with a markedly increased risk in Asian populations, consistent with previous meta-analyses reporting elevated cancer susceptibility associated with rs1136410 in East Asian populations 2,3 .These effects in opposite directions strongly support the need for ancestry-informed interpretation of germline variants and underscore why we prespecified ethnicity-stratified analyses rather than relying solely on overall pooled estimates [1][2][3] .Regarding between-study heterogeneity and deviations from Hardy-Weinberg equilibrium (HWE), we acknowledge these concerns but note that in multi-ethnic genetic meta-analyses, substantial heterogeneity often reflects genuine biological and population-level differences rather than methodological flaws alone 2,3 .In our study, heterogeneity decreased within ethnicity-specific strata, supporting this interpretation.Furthermore, our approach of performing sensitivity analyses that exclude HWE-violating studies follows the recommended practice in genetic meta-analyses 4 .Importantly, when we excluded HWE-deviating studies in analyses of gastric cancer (GC), the associations in key models (C vs. T and CC vs. TT) remained statistically