AAT score based on pretreatment indicators predicts outcomes in unresectable HCC patients treated with TACE, Sintilimab, and Bevacizumab

Background Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related death in China, characterized by insidious onset and poor prognosis. Most patients are diagnosed at intermediate-advanced stages with unresectable HCC (uHCC), and the combination of transarterial chemoembolization (TACE) with targeted and immunotherapy has shown promising efficacy. This study aimed to develop a prognostic tool for uHCC patients receiving TACE combined with sintilimab and bevacizumab therapy. Methods A total of 176 uHCC patients from two institutions (March 2021 to May 2024) were included. Univariate and multivariate Cox regression identified alpha-fetoprotein (AFP), alkaline phosphatase (ALP), and tumor burden score (TBS) as independent predictors of overall survival (OS), based on which the AAT score model was constructed. The area under the receiver operating characteristic (ROC) curve in training and validation cohorts was calculated, and Kaplan-Meier analyses were performed to evaluate the model. Results Cox regression confirmed AFP, ALP, and TBS as independent OS predictors, with the AAT model achieving areas under the ROC curve of 0.813 and 0.819 in training and validation cohorts, outperforming individual factors. Kaplan-Meier analysis of OS and progression-free survival (PFS) demonstrated that the AAT score could significantly stratify patients into low (≤1.8), median (>1.8 to ≤3.0), and high (>3.0) risk groups, with distinct 2-year OS rates of 80.0%, 48.0%, and 4.0%, and PFS rates of 63.3%, 56.0%, and 4.0% in the training cohort, validated in the cohort. Conclusions The AAT model effectively stratifies OS and PFS in uHCC patients undergoing TACE plus sintilimab and bevacizumab, guiding personalized treatment decisions.