MRI-based deep learning model for early TACE response prediction in HCC: multicenter validation with biological insights.
Authors
Affiliations (7)
Affiliations (7)
- Zhejiang Key Laboratory of Imaging and Interventional Medicine, Zhejiang Engineering Research Center of Interventional Medicine Engineering and Biotechnology, School of Medicine, Lishui Hospital, Zhejiang University, 289 Kuocang Road, Lishui, 323000, China.
- Department of Radiology, The Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, Guangdong, 519000, China.
- Department of Radiology, Lishui Central Hospital, the Fifth Affiliated Hospital of Wenzhou Medical University, 289 Kuocang Road, Lishui, 323000, China.
- Imaging Intervention Division, Affiliated Hangzhou First People's Hospital, School of Medicine, Westlake University, Hangzhou, 310006, China.
- Division of Hepatobiliary and Pancreatic Surgery, Hepatobiliary and Pancreatic Interventional Treatment Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, 310003, China. [email protected].
- Zhejiang Key Laboratory of Imaging and Interventional Medicine, Zhejiang Engineering Research Center of Interventional Medicine Engineering and Biotechnology, School of Medicine, Lishui Hospital, Zhejiang University, 289 Kuocang Road, Lishui, 323000, China. [email protected].
- Department of Radiology, Lishui Central Hospital, the Fifth Affiliated Hospital of Wenzhou Medical University, 289 Kuocang Road, Lishui, 323000, China. [email protected].
Abstract
Transarterial chemoembolization (TACE) remains a cornerstone treatment for hepatocellular carcinoma (HCC), yet heterogeneous treatment response poses significant clinical challenges. This multicenter study aimed to develop and validate a deep learning model that leverages pretreatment MRI to predict objective response to initial TACE, while exploring imaging-biological correlations. We utilized retrospective data from 3 institutions, which included HCC patients who underwent TACE. A deep learning algorithm (hereinafter, DLTR) was developed for predicting TACE response by comparing various deep learning algorithms. A multilayer perceptron was then employed to integrate potential clinical factors into the model (hereinafter, DLTR<sub>MLP</sub>) classifier. Performance was evaluated by the area under the receiver operating characteristic curve (AUC) in internal and external cohorts. Survival differences were assessed using log-rank test in two external test sets. RNA-sequencing data from the Cancer Image Archive (TCIA) were used to link imaging signatures to biological pathways. DLTR<sub>MLP</sub> achieved higher AUC than DLTR and clinical models in predicting TACE efficacy in two external test cohorts (AUC: 0.8 vs. 0.649, 0.648; 0.818 vs. 0.629, 0.659) and effectively stratified patients by progression-free survival (<i>P</i> = 0.035). Deep learning features correlated with 149 genes (<i>P</i> < 0.05), which were notably enriched in angiogenesis, EMT, hypoxia, and TGF-β Signalling pathways. The DLTR<sub>MLP</sub> model, combining MRI-based deep learning and clinical variables, robustly predicts TACE response and reveals imaging signatures linked to tumour proliferation biology. Its potential integration into MRI workflows could help optimize treatment decision-making for HCC. The online version contains supplementary material available at 10.1186/s12885-025-15273-8.