Operative resection in early stage pancreatic neuroendocrine tumors in the United States: Are we over-or undertreating patients?

Background: Many current guidelines recommend nonoperative management for pancreatic neuroendocrine tumors <2 cm. The objective of this study was to evaluate the utilization and outcomes of resection for these pancreatic neuroendocrine tumors in the United States. Methods: Using the National Cancer Database (2004-2014), 3,243 cases of T1 (<= 2.0 cm) pancreatic neuroendocrine tumors were identified. Additional patient and tumor characteristics were examined. Multivariate models were used to identify factors that predicted resection and to assess patient survival after resection. Results: 75% of pancreatic neuroendocrine tumors measuring 0 to 1.0 cm and 80% of pancreatic neuroendocrine tumors measuring >1.0 and <= 2.0 cm were resected. Eighty-four pancreatic neuroendocrine tumors were functional, of which 82% were resected. Variables influencing resection included positive lymph nodes, tumor in body or tail of pancreas, well or moderately differentiated tumors, and resection at academic medical centers (odds ratio 1.5-4.9). When controlling for other variables, patients with pancreatic neuroendocrine tumors 1 to 2 cm who underwent resection had a prolonged 5-year survival rate (hazard ratio 0.51, confidence interval 0.34-0.75) when compared with those who did not undergo resection. This survival benefit of resection was not found for pancreatic neuroendocrine tumors 0 to 1 cm (hazard ratio = 0.63, confidence interval 0.36-1.11). Conclusions: Contrary to many current recommendations, most patients with pancreatic neuroendocrine tumors <= 2.0 cm undergo surgical resection in the United States. A survival benefit was found for resection of pancreatic neuroendocrine tumors 1 to 2 cm, suggesting that current recommendations should perhaps be revised. (C) 2019 Elsevier Inc. All rights reserved.