Application of the pancreatic adenocarcinoma staging system to pancreatic neuroendocrine tumors

Background: The American Joint Committee on Cancer (AJCC) 6(th) edition staging system for pancreatic adenocarcinoma specifically excludes pancreatic neuroendocrine tumors (PNETs), and a widely accepted staging classification does not exist. Our objective was to evaluate the feasibility of applying the AJCC pancreatic adenocarcinoma staging system to PNETs. Study Design: Patients with PNETs were identified from the National Cancer Data Base (1985 to 2004). Overall survival was estimated using the Kaplan-Meier method and compared using the logrank test. Cox proportional hazards modeling was used to evaluate the effect of tumor size, nodal status, and distant metastases on survival. Results: Of 4,793 patients with PNETs, 93.6% were of islet cell origin, and 6.4% were carcinoid tumors; 1,815 (37.9%) underwent resection. Overall 5-year survival rates were 29.2% for all patients, 55.4% for resected patients, and 15.6% for unresected patients. The AJCC staging system provided good prognostic survival discrimination between stage groups for resected patients (p < 0.0001). When comparing outcomes to those of patients with pancreatic adenocarcinoma, the estimated median survival was significantly better for resected patients with PNETs (60 versus 13 months, p < 0.0001). Distant metastasis was the only independent predictor of survival (p < 0.0001). Conclusions: When applied to PNETs, the AJCC staging system for pancreatic adenocarcinoma provides survival discrimination by stage for surgical and nonsurgical patients. Survival rates are better for PNETs than for pancreatic adenocarcinoma, but the staging system can effectively stratify patients with PNETs.