Wei Ran, Liu Zelong, Ju Lingjie, Zuo Mengxuan, Yao Wang, Li Wang, Fu Yan, Liu Wendao, Li Chengzhi, Wu Peihong, Han Jianjun, Zhang Yaojun, Tu Jianfei, Ren Junhong, An Chao, Peng Zhenwei
Department of Gastrointestinal Surgery, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou, China.
Department of Radiation Oncology, Cancer Center, The First Affiliated Hospital, Sun Yat-sen University, 58 Zhongshan Road, Guangzhou, China 510080.
Radiol Imaging Cancer. 2025 Sep;7(5):e250034. doi: 10.1148/rycan.250034.
Purpose To develop and test a machine learning (ML)-based model that integrates preoperative variables for prediction of advanced-stage progression (ASP) after transarterial chemoembolization (TACE). Materials and Methods This multicenter retrospective study (ResearchRegistry.com identifier no. researchregistry9425) included patients with intermediate-stage hepatocellular carcinoma (HCC) who underwent TACE at seven hospitals from June 2008 to December 2022. Thirty-four preoperative clinical and CT imaging variables were input into six ML-based models for prediction of ASP, and model performances were compared. Furthermore, the best-performing ML model was compared with the major staging systems, and its utility in performing post-TACE therapies was assessed. The performances of the models were compared by using area under the receiver operating characteristic curve (AUC) with DeLong test. Kaplan-Meier survival curves were compared using the log-rank test. Results A total of 2333 eligible patients (mean age, 54 years ± 12 [SD]; 2051 male patients) were categorized into the training set ( = 1026), the internal test set ( = 257), and the external test set ( = 1050). ASP was found in 8.4% (86 of 1026), 8.2% (21 of 257), and 6.7% (70 of 1050) of patients in the three datasets, respectively. Among all ML models, the Categorical Gradient Boosting (CatBoost) model yielded the highest AUC: 0.97 (95% CI: 0.95, >0.99) for the training set, 0.94 (95% CI: 0.92, 0.97) for the internal test set, and 0.93 (95% CI: 0.90, 0.95) for the external test set. Furthermore, it yielded better discriminatory ability with higher concordance indexes than the five staging systems (all < .001). The time-dependent AUC of the CatBoost model was also higher than that of the clinical staging systems at various time points (all < .001). Moreover, post-TACE systemic therapy improved progression-free survival and overall survival for patients in the high-risk group (both < .001) but not in the low-risk group. Conclusion The CatBoost model demonstrated higher predictive performance compared with existing staging systems in predicting ASP after TACE in patients with intermediate-stage HCC. This model effectively stratified patients by risk level and identified those who benefited from post-TACE systemic therapy. Liver, Oncology, Transarterial Chemoembolization, Hepatocellular Carcinoma, Advanced-stage Progression, Machine Learning, Risk Differentiation ResearchRegistry.com identifier no. researchregistry9425 © RSNA, 2025 See also commentary by Rouzbahani in this issue.
目的 开发并测试一种基于机器学习（ML）的模型，该模型整合术前变量以预测经动脉化疗栓塞术（TACE）后晚期进展（ASP）。材料与方法 这项多中心回顾性研究（ResearchRegistry.com标识符：researchregistry9425）纳入了2008年6月至2022年12月在七家医院接受TACE的中期肝细胞癌（HCC）患者。将34个术前临床和CT成像变量输入六个基于ML的模型以预测ASP，并比较模型性能。此外，将性能最佳的ML模型与主要分期系统进行比较，并评估其在TACE后治疗中的效用。使用受试者操作特征曲线下面积（AUC）和DeLong检验比较模型性能。使用对数秩检验比较Kaplan-Meier生存曲线。结果 总共2333例符合条件的患者（平均年龄54岁±12[标准差]；2051例男性患者）被分为训练集（n = 1026）、内部测试集（n = 257）和外部测试集（n = 1050）。在三个数据集中，分别有8.4%（1026例中的86例）、8.2%（257例中的21例）和6.7%（1050例中的70例）的患者出现ASP。在所有ML模型中，分类梯度提升（CatBoost）模型的AUC最高：训练集为0.97（95%CI：0.95，>0.99），内部测试集为0.94（95%CI：0.92，0.97），外部测试集为0.93（95%CI：0.90，0.95）。此外，与五个分期系统相比，它具有更好的鉴别能力和更高的一致性指数（均P <.001）。CatBoost模型的时间依赖性AUC在各个时间点也高于临床分期系统（均P <.001）。此外，TACE后全身治疗改善了高危组患者的无进展生存期和总生存期（均P <.001），但低危组未改善。结论 与现有分期系统相比，CatBoost模型在预测中期HCC患者TACE后ASP方面表现出更高的预测性能。该模型有效地按风险水平对患者进行分层，并识别出从TACE后全身治疗中获益的患者。肝脏、肿瘤学、经动脉化疗栓塞术、肝细胞癌、晚期进展、机器学习、风险分化 ResearchRegistry.com标识符：researchregistry9425 © RSNA，2025 另见本期Rouzbahani的评论。
