Background
Several scoring systems are currently used to predict prognosis of hepatocellular carcinoma (HCC), but none of them integrates liver function, systemic inflammation, and tumor characteristics in a unified model. The current study aimed to develop and validate a novel prognostic score that integrates liver function, systemic inflammation, and tumor characteristics in a unified model to predict the prognosis of HCC after curative resection.
Methods
Patients with HCC who underwent curative liver resection were included in a training set (n = 1027). Multivariate Cox regression was performed to determine the risk factors for a poor prognosis. A prognostic score was developed by assigning points for risk factors in proportion to beta coefficients in a Cox multivariable model. Predictive performance and distinction ability of the prognostic score were further evaluated in two independent validation cohorts treated with either curative resection (n = 281) or transarterial chemoembolization (TACE) (n = 404) and compared with 16 other models.
Results
The prognostic predictive system, named the function-inflammation-burden-alpha-fetoprotein (FIBA) score, was derived by assigning points for six independent predictors including albumin, total bilirubin, lymphocyte count, diameter of the largest tumor, number of tumors, and alpha-fetoprotein (AFP). The FIBA score showed an outperformed predictive value compared with other systems in both training and validation cohorts by giving the highest C-index, likelihood ratio chi-square values, and Wald test values as well as the lowest Akaike information criterion.
Conclusion
The FIBA score can be used to stratify HCC patients treated with curative resection. Meanwhile, the FIBA score performs well against other prognostic scoring systems and is potentially broadly applicable to a TACE-treated patient cohort.
