Diabetes Status and Pancreatic Cancer Survival in the Nationwide Veterans Affairs Healthcare System.

Abstract

Long-standing type 2 diabetes is a known risk factor for developing pancreatic cancer, however, its influence on cancer-associated outcomes is understudied. To examine the associations between diabetes status and pancreatic cancer outcomes. We identified patients diagnosed with pancreatic adenocarcinoma in the national Veterans Administration System from 2010 to 2018. We classified each patient by pre-cancer diagnosis diabetes status: no diabetes, new-onset diabetes (NOD) of ≤ 3years duration, or long-standing diabetes of > 3years duration. We used Cox proportional hazards models to examine the association between diabetes status and survival. We adjusted the models for age, race, sex, body mass index, tobacco, and alcohol use, coronary artery disease, hypertension, chronic kidney disease, year of cancer diagnosis, and cancer stage and treatment. We identified 6342 patients diagnosed with pancreatic adenocarcinoma. Most had long-standing diabetes (45.7%) prior to their cancer diagnosis, 14.5% had NOD, and 39.8% had no diabetes. Patients with long-standing diabetes had 10% higher mortality risk compared to patients without diabetes after adjusting for sociodemographic factors and medical comorbidities (adjusted HR 1.10; 95% CI 1.03-1.16). This difference in mortality remained statistically significant after additionally adjusting for cancer stage and receipt of potentially curative treatment (adjusted HR 1.09; 95% CI 1.02-1.15). There was no significant difference in mortality between patients with NOD compared to those without diabetes. Long-standing but not new-onset diabetes is independently associated with increased mortality among patients with pancreatic cancer. This information has implication for prognostication and risk stratification among pancreatic cancer patients.