Abstract / Summary
Abstract Background Disentangling the role of micronutrients in gastric cancer (GC) is challenging and results are inconsistent to date. We investigated associations of dietary intake of B vitamins with GC in a pooled analysis of international studies. Methods We included orifinal data from 13 case-control studies of GC from the Stomach Cancer Pooling Project. Data were harmonized and study-specific quartiles of dietary intake of the B1 (thiamine), B2 (riboflavin), B3 (niacin) and B6 (pyridoxine) were created. We calculated odds ratios (OR) of associations between each B vitamin with GC and the corresponding 95% confidence intervals (CI) through logistic regression models adjusted for major risk factors. We then pooled study-specific ORs using random effect meta-analysis. Stratified analyses were conducted by sex, histologic type and anatomical site. Secondary analyses were further adjusted for Helicobacter pylori (Hp) infection, alcohol, meat and vegetables and fruit intake. Results The study population included 5252 GC cases and 11,397 controls. We observed the following associations for one-quartile increase of dietary vitamin intake: B1, OR = 0.93 (95% CI = 0.88–0.97); B2, OR=0.97 (95% CI = 0.93–1.02); B3, OR = 0.95 (95% CI = 0.90–1.00); B6, OR = 0.86 (95% CI = 0.82–0.91); we observed an interaction between B1 and sex, with beneficil effect in men for the highest quartile of intake ( p = 0.009). Main results were confirmed when further adjusting for Hp. When accounting for alcohol, meat and vegetables and fruit intakes, the associations were weakened or no more apparent. No heterogeneity was found between cardia and non-cardia GC; B2 was directly associated with diffuse but not intestinal GC ( p = 0.03). Conclusions Dietary intake of B1 and B6 vitamins might reduce the risk of GC. Studies with homogeneous method of micronutrients assessment and adjustments are needed to confirm our results.