Discordant risk factors between pancreatic neuroendocrine neoplasms and pancreatic ductal adenocarcinoma

Abstract

Pancreatic neuroendocrine neoplasm (panNEN) is a rare malignancy and the second most common type of pancreatic cancer after pancreatic ductal adenocarcinoma (PDAC), but its etiology is poorly understood. We investigated whether the risk factors of panNEN are concordant with those known for PDAC. We performed the largest case-control study to date on panNENs, comprising 927 sporadic nonfunctional panNEN cases and 1807 frequency-matched controls, using data from the Mayo Clinic Biospecimen Resource for Pancreas Research. We assessed associations for obesity, first-degree family history of pancreatic cancer, cigarette smoking, overall type II diabetes mellitus (T2DM), new-onset T2DM (<1 year before panNEN diagnosis), longstanding T2DM (≥5 years), alcohol intake and aspirin use. Multivariable logistic regression was used to calculate odds ratios and 95% confidence intervals (CIs). Our results show that overall T2DM (OR = 1.71, 95% CI: 1.37-2.14) and new-onset T2DM (OR = 2.65, 95% CI: 1.92-3.69) are associated with higher odds of panNEN, but not longstanding T2DM (OR = 1.29, 95% CI: 0.94-1.75). A non-significant elevated odds of panNEN was observed among participants with a positive family history of pancreatic cancer (OR = 1.44, 95% CI: 0.96-2.14). Alcohol use was inversely related to panNEN (OR = 0.52, 95% CI: 0.42-0.66, ever-vs-never). No association was observed for smoking, obesity or aspirin use. These findings indicate that overall T2DM and new-onset T2DM are associated with higher odds of panNEN. Unlike PDAC, alcohol use was inversely related to panNEN, and we found no associations for cigarette smoking, obesity or aspirin use. These results indicate differences in the risk factor profiles of panNEN and PDAC.

Keywords: NEC; nets; neuroendocrine cancer; panNEN; panNET; pancreatic neuroendocrine neoplasms; risk factors.

Figure 1

Flowchart of sample selection for cases and controls. The flowchart shows reasons for the exclusion of panNEN cases from the initial sample of 1,298 patients. We frequency-matched 927 panNEN cases to 1,807 cancer-free controls based on age (5-year intervals), sex and region of residence (Midwest, other) for final analyses. Abbreviations: panNEN, pancreatic neuroendocrine neoplasms; NOS, not otherwise specified.