BACKGROUND: Few comprehensive studies have examined the associations of the ABO blood group with survival outcomes for patients with resected pancreatic cancer, overall and by adjuvant chemotherapy regimens. METHODS: This multicenter study enrolled 1153 patients with resected pancreatic cancer. The hazard ratios (HRs) for disease-free and pancreatic cancer-specific survival were calculated with adjustment for potential confounders, including KRAS mutation and CDKN2A (p16), TP53, and SMAD4 expression, using the Cox proportional hazards regression model. Blood group antigen expression in tumors was immunohistochemically assessed. RESULTS: The ABO blood group was not associated with disease-free or pancreatic cancer-specific survival (P > 0.90). For pancreatic cancer-specific survival, blood groups A, B, and AB had multivariable HRs of 0.97 (95% confidence interval [CI], 0.81-1.15), 1.03 (95% CI, 0.83-1.26), and 0.99 (95% CI, 0.76-1.30), respectively (vs. O). The associations between ABO blood group and disease-free and pancreatic cancer-specific survival differed according to the adjuvant chemotherapy regimens (P = 0.011 and 0.008, respectively). For the patients without chemotherapy, the HRs for disease-free survival were 0.99 (95% CI, 0.69-1.41) for blood group A, 1.65 (95% CI, 1.09-2.48) for blood group B, and 1.79 (95% CI, 1.01-3.17) for blood group AB, (vs. O). For the patients receiving S-1-based chemotherapy, blood group AB (vs. O) exhibited a reverse association (HR, 0.63; 95% CI, 0.39-1.00). Similar interactions were observed when blood group antigen expression in tumors was analyzed. CONCLUSIONS: The ABO blood group is not a prognostic biomarker in resected pancreatic cancer overall but may predict the effectiveness of adjuvant chemotherapy.