Pancreatic cancer is rapidly fatal in most cases. Identifying etiologic factors and early symptoms may provide means to reduce mortality through prevention or early detection of this devastating disease. Numerous studies have examined associations between pancreatic cancer and type II diabetes with most finding a significant positive association, but the temporal relationship between the two is not clear. We examined associations between self-reported diabetes (diagnosed at age β30 y) and diabetes duration with incident pancreatic cancer in a prospective cohort of postmenopausal women in Iowa aged 55-69 years in 1986. Diabetes status and age at diagnosis were assessed at baseline and in 5 follow-up questionnaires. The analytic cohort included 36,084 women of whom 2,205 had baseline diabetes. During 22 years of follow-up, 292 incident cases of exocrine pancreatic cancer were identified through Iowa's State Health Registry, part of the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program. Deaths were ascertained through the Health Registry and the National Death Index. Proportional hazards regression was used to compute hazard ratios and their 95% confidence intervals (HR, 95% CI) for pancreatic cancer in relation to baseline diabetes. Analyses with diabetes modeled as a time dependent variable accounted for diabetes reported during follow-up. Women with a diabetes diagnosis that was followed by pancreatic cancer diagnosis in less than 2 years were considered non-diabetic in prospective analyses (n=4). All models were adjusted for age, pack-years, smoking status, education, and body mass index. A self-reported history of diabetes at baseline was associated with an increased risk of pancreatic cancer HR=1.86 (95% CI: 1.23, 2.83). When accounting for diabetes reported at baseline and during follow-up, the adjusted HR of pancreatic cancer associated with a history of diabetes was 1.94 (95% CI: 1.40, 2.69). The effect of diabetes duration on pancreas cancer risk was assessed in a nested-case control analysis. Four controls (n= 1168) were matched per case (n=292) on birth year and vital status at time of cancer diagnosis. Odds ratios (ORs) and 95% CIs were calculated using conditional logistic regression. ORs for pancreatic cancer by categories of diabetes duration were examined in multivariate models that included continuous diabetes duration as a covariate. ORs were increased in all diabetes duration categories relative to non-diabetics; though increased ORs and differences between categories were not statistically significant. Categories in years and corresponding ORs and 95% CIs were as follows: (0<Xα2): 1.86 (0.34, 10.3); (2<Xα8): 1.13 (0.13, 9.60); (Xα8) 2.83 (0.86, 9.30). Results from this prospective cohort study support the view that women with diabetes are at nearly 2 fold increased risk for pancreatic cancer and the risk does not differ with duration of diabetes.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 668. doi:1538-7445.AM2012-668