Abstract Introduction: The “obesity paradox,” in which higher body mass index (BMI) appears to confer a survival advantage despite its role in carcinogenesis, has been reported in several solid tumors, but its relevance across racial groups in patients with pancreatic cancer is unclear. We performed a retrospective database study using TriNetx, a multinational deidentified dataset, to identify patients with pancreatic cancer and to examine associations between BMI, race/ethnicity, survival, and post operative complications. Methods: Using the TriNetx database, we identified patients with pancreatic cancer and stratified the cohort by BMI at time of diagnosis: underweight (UW) (BMI 18.5 kg/m2), healthy weight (HW) (18.5-24.9 kg/m2), and overweight (OW) (≥25.0 kg/m2). Subgroups were created of Black, Hispanic, and White patients within each BMI cohort to evaluate for racial/ethnic differences in BMI associated outcomes. The primary outcome was overall survival with secondary outcomes of post-operative complications. Propensity score matching was performed for age, sex, stage, comorbidities, and type of surgical resection performed. Kaplan-Meier curves and multivariable Cox models were used to evaluate survival across BMI categories overall and within racial strata. Multivariable logistic regression was used to evaluate associations between BMI category and postoperative complications. Results: We identified 247,150 patients with pancreatic cancer, including 51,057 (87%) OW, 3241 (6%) HW, and 3815 (7%) UW patients. In propensity matched models, BMI category was found to be associated with overall survival. Compared with HW, OW was associated with a longer 5-year median survival (895 vs 493) and reduced hazard of death (HR = 0.76, 95% CI (0.70-0.82)), while UW was associated with worse survival than HW (HR = 1.24, 95% CI 0.74-0.87)). The magnitude and direction of these associations differed by race/ethnicity with OW Black (HR = 0.69, 95% CI (0.59-0.85)) and OW Hispanic patients (HR = 0.88, 95% CI (0.62-1.25)) deriving more substantial survival benefit over HW when compared to their OW White counterparts (HR = 0.80, 95% CI (0.73-0.86)). Among propensity matched surgical patients, OW was associated with increased risk intraabdominal abscess (10.2% vs 5.1%, RR = 2, p = 0.02), and increased odds of AKI (19.7% vs 12.9%, RR = 1.53, p = 0.026) compared with HW individuals. No significant difference in postoperative pancreatic leak was identified between the OW and HW cohorts. (11.5% vs 11.8%, RR = 0.97, p = 0.89). Conclusions: In this large, real-world cohort of patients with pancreatic cancer, BMI and race were significant determinants of survival and postoperative morbidity, with patterns consistent with an obesity paradox. These data may refine risk stratification and motivate studies of adiposity and pancreatic cancer biology across diverse populations. Citation Format: Jacob Lambdin, Charlotte Burch, William Royster, Adam Khader, Leopoldo Fernandez, Raphael Louie, Ricardo Bello, Jose G. Trevino. The obesity paradox in pancreatic cancer abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 6448.
Lambdin et al. (Fri,) studied this question.