Prediction of the Efficacy of PA-TACE in Patients with Clearly Defined Hepatocellular Carcinoma by Imaging Combined with Serology: A Retrospective Cohort Study with a Two-Center Follow-Up of up to 12 Years.

OBJECTIVE: The purpose of this study was to determine which patients with hepatocellular carcinoma (HCC) actually benefit from Postoperative adjuvant Transarterial chemoembolization (PA-TACE) and to develop a reliable predictive model for tailored treatment choices. METHODS: A total of 150 HCC patients received PA-TACE treatment between 2012 and 2020 were included in a retrospective dual-center analysis, of whom 100 were allocated to the training cohort and 50 to the external validation cohort. Radiomics feature extraction, liver segmentation, and neovascularization evaluation were performed using preoperative contrast-enhanced CT scans. In addition, perioperative serological markers were assessed. To find independent predictive markers linked to survival outcomes, univariate and multivariate Cox regression analyses were used. RESULTS: Radiomics features, neovascularization, and the albumin (ALB) ratio were independently correlated with overall and progression-free survival. The combined model integrating these factors exhibited superior predictive accuracy, which was further confirmed by external validation. CONCLUSIONS: The suggested dual-center nomogram shows excellent practical relevance in assessing the effectiveness of PA-TACE and improving postoperative care techniques. It integrates CT radiomics and serological markers, with overall survival as the major outcome.