National Trends in the Management of Small Localized Non-Functioning Pancreatic Neuroendocrine Tumors

Abstract

Introduction: Controversy exits regarding the role of resection in small non-functioning pancreatic neuroendocrine tumors (NF-PNETs). Our objective was to characterize national trends in the management of small NF-PNETs and elucidate factors associated with treatment selection and survival. Methods: The 2004-2015 National Cancer Database was queried for patients with small (<2cm, cT1) NF-PNETs and clinically negative nodes (cN0). Patients were stratified by observation versus surgical resection (pancreaticoduodenectomy (WP) or distal pancreatectomy (DP)). Results: Of 2,421 patients, 1,717 (71%) underwent pancreatectomy (DP 69.9%; WP 30.1%). Despite a majority undergoing surgery, rates of pancreatectomy decreased significantly over time, with rates of WP (slope -2.44, p=0.03) decreasing at a greater magnitude than DP (slope -1.40, p=0.04). A significant decrease in pancreatectomy was noted at academic centers (p=0.008), but not non-academic centers (p=0.08). Independent predictors of resection included younger age (OR 0.96; CI 0.95-0.97), private insurance (OR 0.69; CI 0.55-0.87), tumor location within the body/tail (OR 2.0; CI 1.76-2.73), and tumor size 1-2cm (OR 1.81; CI 1.45-2.26). Kaplan Meier analysis demonstrated a survival advantage associated with resection of tumors 1-2cm in size (p<0.001), but not for tumors <1cm (p=0.16). While surgery remained independently prognostic for survival on Cox analysis (HR 2.95; CI 1.87-4.65), tumor size was no longer independently prognostic (p=0.78). Instead, tumor grade represented the single most important prognostic factor for death (HR 5.14; CI 1.83-14.42). Conclusion: Evidence-based guidelines are gradually being implemented across institution types. Tumor grade and other socio-demographic factors, rather than tumor size, appear to be the primary determinants of survival.