A refined prediction model for survival in hepatocellular carcinoma patients treated with transarterial chemoembolization.

ABSTRACT

Background/Aims: Transarterial chemoembolization (TACE) is widely performed as a major treatment for hepatocellular carcinoma (HCC) patients, and there is a need to stratify patients for whom the most benefit from the treatment. This study aimed to develop a refined prediction model for overall survival (OS) in patients undergoing TACE as a first-line treatment in a large cohort and validate its performance. Methods: A total of 2,632 patients with HCC of Barcelona Clinic Liver Cancer stage A or B who underwent TACE between 2008 and 2017 were enrolled. The patients were randomly assigned to a training cohort (n = 1,304) or a validation cohort (n = 1,328). Independent predictors of OS were used to develop a prediction model. Results: The median age of patients in the entire cohort was 63 years, with the majority having hepatitis B virus (56.6%) and being classified as Child-Pugh class A (82.4%). We developed a new prognostic model, called the TACE-prognostic (TP) score, based on tumor burden (sum of the largest tumor diameter and tumor number), alpha-fetoprotein, and Albumin-Bilirubin grade. Patients were classified into five risk groups according to TP scores, with median survival significantly differentiated in both training and validation cohorts (P < 0.001). The new model consistently outperformed other currently available models in both the training and validation cohorts. Conclusion: This newly developed TP scoring system has the potential to be a useful tool in identifying ideal candidates of TACE and predicting OS with favorable performance and discrimination. However, further external validation is needed to confirm its effectiveness.