Nomograms for survival prediction in patients undergoing liver resection for hepatitis B virus related early stage hepatocellular carcinoma

Background: The long-term outcomes of patients who underwent liver resection (LR) for early-stage hepatitis B virus (HBV)-related hepatocellular carcinomas (HCCs) are difficult to predict. This study aimed to develop two nomograms to predict postoperative disease-free survival (DFS) and overall survival (OS), respectively. Methods: Data on a primary cohort of 1328 patients who underwent LR for HBV-related HCCs within Milan criteria at the Eastern Hepatobiliary Surgery Hospital (EHBH) from 2000 to 2006 were used to develop the nomograms by the Cox regression analyses. An internal validation cohort of 442 patients operated from 2006 to 2011 at the EHBH and an external validation cohort of 474 patients operated from 2007 to 2009 at the Zhongshan Hospital were used for validation studies. Discrimination and calibration were measured using concordance index (C-index), calibration plots and Kaplan-Meier curves. Results: The independent predictors of DFS or OS which included tumour stage factors, biomarker and HBV-DNA level were respectively incorporated into the two nomograms. In the primary cohort, the C-indexes of the models in predicting DFS and OS were 0.76 (95% confidence interval: 0.75-0.78) and 0.79 (0.77-0.81), respectively. The calibration curves fitted well. Both nomograms accurately stratify patients into four distinct incremental prognostic subgroups. The C-indexes of the nomogram for OS prediction was significantly higher than those of the six conventional staging systems (0.65-0.71, all P < 0.001). These results were verified by the internal and external validations. Conclusion: The proposed nomograms showed good prognostication for patients with early HBV-related HCCs after hepatectomy. (C) 2016 Elsevier Ltd. All rights reserved.