Abstract Transarterial chemoembolization (TACE) is the first-line therapy for intermediate-stage hepatocellular carcinoma (HCC), yet treatment response exhibits marked heterogeneity and robust preoperative predictive tools are lacking. This study developed and validated a novel nomogram integrating CT-quantified body composition parameters with clinical factors to predict tumor response in 209 HCC patients after TACE. Using a multi-center retrospective design (training cohort: n = 112; internal validation: n = 46; external validation: n = 51), we extracted from adipose and muscle tissues at the L3 vertebral level to compute standardized indices. Multivariable logistic regression identified independent predictors, which were incorporated into a nomogram for risk stratification. The combined model integrating alpha-fetoprotein (AFP), tumor number, visceral adipose tissue (VATI), and adipose tissue ratio (VSR) effectively discriminated responders from non-responders, achieving high predictive accuracy across cohorts (AUC: 0.805 training, 0.829 internal validation, 0.815 external validation). Calibration and decision curve analysis confirmed superior model performance and clinical net benefit. This nomogram quantifies the prognostic impact of adipose tissue distribution, providing an actionable preoperative tool to guide personalized TACE management—including alternative or intensified therapies for high-risk patients or standard care for low-risk individuals.
Zhao et al. (Tue,) studied this question.