Evaluation of the current treatment strategies for pancreatic neuroendocrine tumors: a propensity score matched analysis

Abstract

Purpose: The management of pancreatic neuroendocrine tumors (PNET) varies between observation(O), pancreatic resection(PR) and enucleation(E). Currently, size, grade and location are used to determine which treatment strategy may be employed. We sought to evaluate each strategy and further clarify the role for surgery. Methods: Utilizing the National Cancer Database we identified patients with PNET and stratified based upon size < 1cm, 1-2cm and >2cm. Propensity score matching was performed by age, Charleson-Deyo score, and grade. Survival analyses was performed using the Kaplan-Meier method. A p< 0.05 was considered significant. Results: We identified 17,921 patients(< 1cm, 1214, 1-2cm, 4325, and >2cm, 12,382) with a median age of 61.5 (18-90). Tumors < 1cm and well differentiated(WD) the median and 5-year survival in the O group was not reached (NR)(77%) vs 142.6 month(87%) in the surgery groups, p< 0.04; in the 1-2 cm WD group 95.7 months and 60% vs NR and 94%, p< 0.001. Similarly in the PD tumors < 1cm the median and survival was 32.9 months and 24% in the O vs NR and 81%, p< 001; in the 1-2 cm group 14.8 months and 19% vs NR and 73%, p< 0.001. There were no differences in survival between PR or E, p=0.09. Conclusions: While observation is acceptable for the management of < 1cm WD PNET, we found an improvement in survival in the patients undergoing surgery. Enucleation and PR did not differ in overall survival. Surgery for PNET should be considered as the first line treatment of these patients.