Abstract
Background and Aim: The association of the ABO blood groups with gastric cancer is controversial and was examined in this study.

Methods: We conducted a case-control study of 2145 cases of gastric cancer and 2472 controls of Korean ethnicity to determine the risk of gastric cancer for each blood group phenotype using multivariable logistic regression models.

Results: Patient ages were 33.3 (SD = 9.76) years in controls compared to 59.5 (SD = 11.91) years in cases. The percentages of males and females were similar among cases and controls. The crude OR revealed phenotype O as a protective factor 0.63 (CI: 0.56–0.72) (P < 0.0001) compared to phenotype A 1.30 (CI: 1.15–1.47) (P < 0.0001) and AB 1.29 (CI: 1.07–1.57) (P < 0.01), which were risk factors for gastric cancer, and phenotype B 1.09 (CI: 0.94–1.24) (P = 0.27) which was neither protective nor a risk factor. The adjusted OR taking patient characteristics into account and using blood group phenotype O as a reference baseline risk showed that phenotype A (1.21; CI: 1.14–1.39) (P = 0.012) and phenotype AB (1.17; CI: 1.08-1.26) (P = 0.038) conferred an increased risk for gastric cancer compared to phenotype B (0.97; CI: 0.84–1.04) (P = 0.63), which was not associated with gastric cancer risk.

Conclusions: We revealed an association of blood group phenotypes A and AB with gastric cancer risk, emphasizing the need to further study the underlying molecular mechanism of this association with gastric cancer.

Key words: ABO blood group phenotype, Gastric carcinoma, Case-control study