Development and Validation of a Prediction Model for Hepatitis B Virus-Related Hepatocellular Carcinoma Patients Receiving Postoperative Adjuvant Transarterial Chemoembolization.

ABSTRACT

Background: Hepatocellular carcinoma (HCC) patients who are at significant risk of tumor recurrence and mortality can benefit from postoperative adjuvant transarterial chemoembolization (PA-TACE). However, the benefits of PA-TACE remain unclear. Herein, we aimed to develop a model for predicting the prognosis of HBV-related patients who undergo PA-TACE and endeavored to guide individualized clinical treatment. Methods: We included 432 HBV-related patients who underwent PA-TACE after curative resection were included. The dataset was divided into a training set (n=216) and an internal validation set (n=216). For identifying independent risk factors, the least absolute shrinkage and selection operator and univariate and multivariate Cox analyses were performed. We derived a prognostic model from the training set that was internally validated. The concordance index (C-index), receiver operating characteristic (ROC) curve, calibration curve, and risk stratification were used to evaluate the performance of the nomogram. Results: Patients undergoing PA-TACE had significantly longer overall survival (OS) than those who did not undergo PA-TACE. Age, albumin levels, macrovascular invasion, tumor size, and, stages of Barcelona Clinic Liver Cancer were identified as independent risk variables and concluded into the nomogram to predict the OS of HBV-related patients who received PA-TACE. The nomogram's C-index values OS were 0.710 and 0.652 in the training and internal validation sets, respectively. Both time-dependent AUC and the calibration curve showed good discrimination and model fitness. The risk score -0.12 was kept as the cut-off value that would accurately divide patients into high-risk and low-risk groups; furthermore, the Kaplan-Meier curve showed a high discriminative ability of the model. Conclusion: We developed a predictive model. comprising a formula and nomogram to predict the OS and provide risk stratification for HBV-related patients undergoing PA-TACE, which could contribute to suitable treatment options for this patient population.