Introduction: The role of obesity in pancreatic cancer survival is poorly understood. This is possibly because most studies have related body mass index (BMI) at the time of surgery for pancreatic cancer to survival. However, BMI at the time of surgery does not reflect pre-cancer BMI since pancreatic cancer has a catabolic effect on weight. To have a better understanding of the role of obesity in pancreatic cancer survival, prospective studies where BMI measurements were taken before cancer diagnosis are needed. Nonetheless, these studies are lacking. Using data from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial cohort, we evaluated the associations of BMI before cancer diagnosis and pancreatic cancer survival.

Methods: Weight, height, and other health-related data were collected from study participants aged 49 to 78 years at study enrollment between November 1993 and June 2001. Among participants who developed pancreatic cancer after study enrollment, 619 (588 exocrine [429 adenocarcinoma] and 25 endocrine) had complete information on pre-diagnosis BMI. For our study, follow-up time was from the date of cancer diagnosis to death. We categorized BMI into 4 (<25 [reference], 25 to <30, 30 to <35, ≥35) kg/m2. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated using Cox proportional hazards model. The multivariable model was adjusted for age (≥59, 60-64, 65-69, ≥70), smoking status (never, former, current) history of diabetes (yes, no) and tumor stage (localized, locally advanced, metastatic).

Results: Ninety percent (529) of the participants with exocrine pancreas cancer died during the follow-up period. Median survival time was 2 months among those with the highest BMI compared to 4 months among those with BMI <25 and 25 to <30 kg/m2 and 5 months among those with BMI 30 to <35 kg/m2 but the difference was not statistically significant (p-value=0.12). We observed non-significant J-shaped associations between BMI and mortality from pancreatic cancer. Compared to those in the lowest BMI category (<25) at study enrollment, the HR for mortality among those with BMI ≥35 was 1.31 (95%CI 0.86-2.00). The results were similar in analyses stratified by sex, stage and history of diabetes. In analyses conducted using BMI at age 50, compared to those with normal BMI, the HRs for those with BMI 25 to <30 kg/m2, 30 to <35 kg/m2 and >35 kg/m2 were 0.87 (95% CI 0.71-1.07), 1.22 (95%CI 0.88-1.63), and 1.91 (95% CI=0.93-3.90), respectively, among people without diabetes.

Conclusion: Although we did not observe statistically significant associations between obesity and mortality from pancreatic cancer, our results suggest that obesity during mid-life, especially among people without diabetes may be associated with increased mortality. Because of the long latency of pancreatic cancer, prospective studies evaluating the role of BMI earlier in life are needed.

Citation Format: Adetunji T. Toriola, Su-Hsin Chang, Lara Dalidowitz, Graham A. Colditz, Rachael Stolzenberg-Solomon. Prediagnosis body mass index and mortality from pancreatic cancer. [abstract]. In: Proceedings of the Twelfth Annual AACR International Conference on Frontiers in Cancer Prevention Research; 2013 Oct 27-30; National Harbor, MD. Philadelphia (PA): AACR; Can Prev Res 2013;6(11 Suppl): Abstract nr B48.