Development and validation of prognostic nomograms for single large and huge hepatocellular carcinoma after curative resection

Abstract

AimThe prediction model of postoperative survival for single large and huge hepatocellular carcinoma (SLH-HCC, diameter > 5.0 cm) without portal vein tumour thrombus has not been well established. This study aimed to develop novel nomograms to predict postoperative recurrence and survival of these patients. MethodsData from 2469 patients with SLH-HCC who underwent curative resection from January 2005 to December 2015 in China were retrospectively collected. Specifically, nomograms of recurrence-free survival (RFS) and overall survival (OS) using data from a training cohort were developed with the Cox regression model (n = 1012). The modes were verified in an internal validation cohort (n = 338) and an external cohort comprising four tertiary institutions (n = 1119). ResultsThe nomograms of RFS and OS based on tumour clinicopathologic features (diameter, differentiation, microvascular invasion, α-fetoprotein), operative factors (preoperative transcatheter arterial chemoembolisation therapy, scope of liver resection and intraoperative blood transfusion), underlying liver function (albumin-bilirubin grade) and systemic inflammatory or immune status (neutrophil-to-lymphocyte ratio) achieved high C-indexes of 0.85 (95% confidence interval [CI], 0.79–0.91) and 0.86 (95% CI, 0.79–0.93) in the training cohort, respectively, which were significantly higher than those of the five conventional HCC staging systems (0.62–0.73 for RFS, 0.63–0.75 for OS). The nomograms were validated in the internal cohort (0.83 for RFS, 0.84 for OS) and external cohort (0.87 for RFS, 0.88 for OS) and had well-fitted calibration curves. Our nomograms accurately stratified patients with SLH-HCC into low-, intermediate- and high-risk groups of postsurgical recurrence and mortality. ConclusionsThe two nomograms achieved optimal prediction for postsurgical recurrence and OS for patients with SLH-HCC after curative resection.