Comment on “PARP-1 rs1136410 Polymorphism and Gastrointestinal Cancer Risk: A Meta-Analysis of Cancer-Type and Ethnic-Specific Associations”

Dear Editor,

We read with interest the meta-analysis by Naseri et al.¹, examining the PARP-1 rs1136410 (Val762Ala) polymorphism and the risk of colorectal cancer (CRC) and gastric cancer (GC). The authors highlight the importance of population-specific genetic backgrounds; we agree and believe that these findings merit discussion in the context of recent poly adenosine diphosphate poly [(adenosine diphosphate-ribose) polymerase inhibitor (PARPi)] trials in gastrointestinal (GI) malignancies^{2, 3, 4}.

In CRC, the allele-level association appears to differ by ancestry. The C allele was protective in Caucasian cohorts [odds ratio (OR)=0.487] but was associated with a markedly increased risk in Asian cohorts (OR=5.785)¹. Such opposite-direction effects argue against one-size-fits-all genetic screening and support ancestry-informed risk stratification, including consideration of population-specific effect sizes when developing polygenic risk scores.

The meta-analysis included 3,591 patients and 5,433 controls, and performed sensitivity analyses to assess the robustness¹. However, substantial between-study heterogeneity, particularly in the overall CRC and GC analyses, and the inclusion of studies deviating from Hardy-Weinberg equilibrium warrant cautious interpretation of pooled estimates. Heterogeneity can reflect genuine differences between populations, but it may also stem from methodological variations, such as control selection, genotyping platform, and study design. Therefore, prespecified stratified analyses, including ethnicity-based strata, are particularly informative.

Regarding GC, rs1136410 was associated with an increased risk in the overall analyses (C vs T, OR=2.012; CT vs TT, OR=1.359; CC vs TT, OR=1.766)¹. Because Val762Ala lies within the catalytic domain of PARP-1, functional variation could plausibly influence DNA repair capacity and treatment sensitivity. Nevertheless, any pharmacogenomic role of germline rs1136410 should be considered hypothesis-generating, pending functional validation and prospective correlation with tumor biology and PARPi outcomes.

We would appreciate clarification on the direction of some genetic-model comparisons reported in Table 2. For GC, the ORs for CT vs TT and CC vs TT indicate increased risk in the heterozygous and homozygous C-carrier genotypes, respectively. However, the dominant comparison (CC+CT vs TT) is reported as OR=0.649, whereas the recessive comparison (CC vs CT+TT) is reported as OR=1.541¹. These results appear directionally inconsistent and approximately reciprocal; if the comparison labels were inverted (TT vs CC+CT), the estimates would align with the contrasts for individual genotypes.

From a translational standpoint, multiple trials have not demonstrated a clinically meaningful benefit from PARPis in unselected GI cancers. In advanced CRC, the phase III LYNK-003 maintenance study evaluating olaparib, with or without bevacizumab, did not improve outcomes compared with standard maintenance². In advanced GC, the Phase III GOLD trial of olaparib plus paclitaxel did not demonstrate a significant survival benefit in the overall population³. In addition, a phase II study of single-agent niraparib in homologous recombination deficiency-positive metastatic esophageal/gastroesophageal junction/proximal gastric adenocarcinoma was terminated early for lack of efficacy⁴. These data underscore that germline risk variants alone are unlikely to guide PARPi use without additional functional validation and assessment within the tumor context.

In summary, Naseri et al.¹ provide a useful synthesis supporting ethnicity-dependent associations between PARP-1 rs1136410 and GI cancer risk. Addressing the apparent directionality issue in genotype-model reporting and conducting functional studies of Val762Ala may refine clinical interpretation and inform future pharmacogenomic investigations.