Background: The objective of this study was to estimate probabilities of overall survival (OS) and cancer-specific survival (CSS) in patients with pancreatic head adenocarcinoma after surgery. In addition, we attempted to build nomograms to predict prognosis of these patients. Methods: Patients diagnosed with surgically resected pancreatic head adenocarcinoma between 2004 and 2014 were selected for the study from the Surveillance, Epidemiology, and End Results (SEER) database. Nomograms were established for estimating 1-, 2- and 3-year OS and CSS based on Cox regression model and Fine and Grey's model. The performance of the nomogram was measured by concordance index (C-index) and the area under receiver operating characteristic (ROC) curve (AUC). Results: A total of 2374 patients were retrospectively collected from the SEER database. The discrimination of nomogram for OS prediction was superior to that of the Tumor-Node-Metastasis (TNM) 7th or 8th edition stage systems (C-index = 0.640, 95% CI, 0.618 - 0.662 vs 0.573, 95% CI, 0.554 - 0.593, P < 0.001; 0.640, 95% CI, 0.618 - 0.662 vs 0.596, 95% CI, 0.586 - 0.607, P < 0.001, respectively). The comparisons of values of AUC showed that the established nomograms displayed better discrimination power than TNM 7th or 8th stage systems for predicting both OS and CSS. Conclusions: The nomograms which could predict 1-, 2- and 3-year OS and CSS were established in this study. Our nomograms showed a relatively good performance and could be served as an effective tool for prognostic evaluation of patients with pancreatic head adenocarcinoma after surgery.
Keywords: Cancer-specific survival; Competing risk analysis; Nomogram; Overall survival; Pancreatic head adenocarcinoma.