Prediction of the efficacy after the first transarterial chemoembolization in hepatocellular carcinoma using CT radiomics combined with inflammatory composite indicators

Background Hepatocellular carcinoma (HCC) ranks among the most prevalent tumors globally. Transcatheter arterial chemoembolization (TACE) serves as the standard treatment for intermediate and advanced stages of HCC. However, patient responses to TACE vary significantly. This study aims to assess the predictive value of combining CT radiomics with inflammatory composite indicators for evaluating the efficacy of initial TACE in HCC patients. Methods We included 175 patients with pathologically confirmed HCC, categorizing them into a good efficacy group (95 cases) and a poor efficacy group (80 cases). We compared radiomics features and inflammatory composite indicators between these groups. To identify independent risk factors for predicting TACE efficacy, we performed multivariate Logistic regression analysis. We developed a radiomics prediction model and a clinical prediction model based on inflammatory composite indicators. A combined prediction model was created using selected inflammatory composite indicators and radiomics features, and visualized with a nomogram. We assessed the model's predictive performance using the receiver operating characteristic (ROC) curve, and its stability and authenticity through 1,000 bootstrap resampling. The clinical benefit was evaluated using a decision curve analysis (DCA) curve. Results The multivariate logistic regression analysis revealed that platelet-to-lymphocyte ratio (PLR), maximum tumor diameter, and Radiomics score (Radscore) were independent risk factors for predicting the efficacy of the first TACE in HCC. The clinical model, based on the inflammatory composite index, achieved an AUC of 66.6 for efficacy prediction. The radiomics model, developed from radiomics features, demonstrated an AUC of 76.1. Notably, the combined prediction model, integrating both radiomics features and the inflammatory composite index, achieved an AUC of 80.4. Conclusion CT radiomics, when combined with composite inflammatory indicators, demonstrated high predictive efficacy for the first TACE treatment outcomes in HCC patients. The developed visual nomogram aids clinicians in creating personalized pre-operative treatment plans for these patients.