A retrospective cohort study of pancreatic neuroendocrine tumors at single institution over 15 years: New proposal for low- and high-grade groups, validation of a nomogram for prognosis, and novel follow-up strategy for liver metastases.

Purpose: Pancreatic neuroendocrine tumors (PNETs) exhibit various tumor behaviors and symptoms that are difficult for physicians to stage and predict prognosis. We assess prognostic factors combined with staging classifications to optimal the models and try to improve follow-up strategy to monitor liver metastases after surgery.
Methods: Patients with PNETs treated at Sun Yat-sen Memorial Hospital between 2000 and 2015 were recruited. Patients were regrouped on the basis of functional status and mitotic rates. Nomograms to predict the prognostic values of classifications (AJCC, ENETS, and WHO) were constructed; the accuracy of the nomograms were quantified by the C-index and calibration plots.
Results: We identified 78 PNETs patients with pathological reports. Correlations with OS in univariate analysis included nonfunctional status (P = 0.002), CgA>200 ng/ml (P < 0.001), Ki-67 (3-20%, P = 0.014; >20%, P < 0.001), and mitotic rate (3-20/10HPF, P = 0.011; >20/10HPF, P < 0.001). By multivariate analysis, nonfunctional status and mitotic rate maintained significance (P = 0.039; 3-20/10HPF, P = 0.015; >20/10HPF, P = 0.007). Evaluating the new proposed system, the difference in OS between low- and high-groups was statistically significant (P = 0.001). The C-index of the regrouped nomograms were higher than that of premise ones (AJCC cohort, 0.605 v 0.576, P < 0.01; ENETS cohort, 0.73 v 0.691, P < 0.01; WHO cohort, 0.678 v 0.603, P < 0.01).
Conclusion: An prognostic model based on mitotic rates and functional status correlates strongly with survival. PNETs should return visits every 2 months for the first half years, and every 3 months as followed until 2 years after surgery.
(Copyright © 2016 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.)