影像组学联合模型对中晚期肝癌患者TACE联合靶向及免疫治疗疗效的预测-东南大学学报医学版

影像组学联合模型对中晚期肝癌患者TACE联合靶向及免疫治疗疗效的预测

单位：1. 江苏省泗阳中医院介入科, 江苏宿迁 223700;
2. 东南大学附属中大医院介入与血管外科, 江苏南京 210009

分类号：R735.7

出版年·卷·期（页码）：2026·45·第一期（111-118）

摘要：

目的:基于影像组学方法构建能够早期预测中晚期肝癌患者肝动脉化疗栓塞术(TACE)联合靶向及免疫治疗应答的模型。方法:回顾性纳入2018年1月至2022年12月就诊于东南大学附属中大医院介入与血管外科行TACE联合靶向及免疫治疗的中晚期肝癌患者的资料。按照2∶1比例将队列随机分为训练集和验证集。通过多步骤特征筛选,采用XGBoost算法构建影像组学预测模型;基于与治疗应答相关的临床变量(体能状态评分、巴塞罗那肝癌临床分期、Child-Pugh 肝功能分级、是否存在血管侵犯、是否存在肝外转移、基线甲胎蛋白水平以及基于 up-to-seven 标准评估的肿瘤负荷),采用多元逻辑回归建立临床预测模型;进一步融合影像组学特征标签与临床变量,构建影像-临床联合预测模型。本研究的主要预测终点为治疗的客观缓解率。结果:共纳入中晚期肝癌患者158例患者,其中训练集106例,验证集52例。总体人群以乙肝相关肝癌为主(142例,89.9%),肿瘤负荷较大[104例患者(65.8%)的肿瘤负荷超过up-to-seven标准]且分期较晚[114例患者(72.2%)为BCLC C期]。经过特征降维和筛选后,最终保留8个影像组学特征用于构建影像组学标签值,影像组学标签值为客观缓解率(ORR)的独立预测因子(P<0.001)。训练集中,影像组学模型、临床模型及联合模型的曲线下面积(AUC)分别为0.85、0.68和0.89;验证集AUC分别为0.80、0.70和0.84。联合模型表现出最佳的预测性能,优于单独的影像组学模型和临床模型。结论:对于中晚期肝癌患者,在使用TACE联合靶向及免疫治疗前,基于影像组学与临床的联合模型具有良好的疗效预测价值。

Objective: To construct a model for early prediction of therapeutic response to transarterial chemoembolization(TACE) combined with targeted and immunotherapy in patients with intermediate and advanced hepatocellular carcinoma(HCC) based on radiomics methods. Methods: Clinical data of patients with intermediate and advanced HCC who underwent TACE combined with targeted and immunotherapy in the Department of Interventional and Vascular Surgery, Zhongda Hospital Affiliated to Southeast University from January 2018 to December 2022 were retrospectively collected. The cohort was randomly divided into a training set and a validation set at a ratio of 2∶1. A radiomics prediction model was established using the XGBoost algorithm after multi-step feature selection. A clinical prediction model was constructed via multivariate Logistic regression based on clinical variables associated with therapeutic response, including Eastern Cooperative Oncology Group(ECOG) performance status score, Barcelona Clinic Liver Cancer(BCLC) stage, Child-Pugh liver function grade, presence of vascular invasion, presence of extrahepatic metastasis, baseline alpha-fetoprotein(AFP) level, and tumor burden evaluated by the up-to-seven criteria. A radiomics-clinical combined prediction model was further developed by integrating radiomics signature and clinical variables. The primary predictive endpoint of this study was the objective response rate(ORR) of treatment. Results: A total of 158 patients with intermediate and advanced HCC were enrolled, including 106 cases in the training set and 52 cases in the validation set. The majority of the overall population had hepatitis B virus(HBV)-related HCC(142 cases, 89.9%), with high tumor burden [tumor burden exceeding the up-to-seven criteria in 104 patients(65.8%)] and advanced disease stage [114 patients(72.2%) at BCLC stage C]. After feature dimensionality reduction and selection, 8 radiomics features were finally retained to construct the radiomics signature, which was identified as independent predictors of ORR(P<0.001). In the training set, the area under the curve(AUC) values of the radiomics model, clinical model and combined model were 0.85, 0.68 and 0.89, respectively; the corresponding AUC values in the validation set were 0.80, 0.70 and 0.84. The combined model exhibited the optimal predictive performance, which was superior to the radiomics model and clinical model alone. Conclusion: For patients with intermediate and advanced HCC, the radiomics-clinical combined model has good value in predicting the efficacy of TACE combined with targeted and immunotherapy.

参考文献：

[1] HAN B,ZHENG R,ZENG H,et al.Cancer incidence and mortality in China,2022[J].J Natl Cancer Cent,2024,4(1):47-53.
[2] ZHU H,LI H,HUANG M,et al.Transarterial chemoembolization with PD-(L)1 inhibitors plus molecular targeted therapies for hepatocellular carcinoma(CHANCE001)[J].Signal Transduct Target Ther,2023,8(1):58.
[3] MONTASSER A,BEAUFRERE A,CAUCHY F,et al.Transarterial chemoembolisation enhances programmed death-1 and programmed death-ligand 1 expression in hepatocellular carcinoma[J].Histopathology,2021,79(1):36-46.
[4] BEN K N,SEIDENSTICKER M,RICKE J,et al.Atezolizumab and bevacizumab with transarterial chemoembolization in hepatocellular carcinoma:the DEMAND trial protocol[J].Future Oncol,2022,18(12):1423-1435.
[5] LAMBIN P,LEIJENAAR R T H,DEIST T M,et al.Radiomics:the bridge between medical imaging and personalized medicine[J].Nat Rev Clin Oncol,2017,14(12):749-762.
[6] SEGAL E,SIRLIN C B,OOI C,et al.Decoding global gene expression programs in liver cancer by noninvasive imaging[J].Nat Biotechnol,2007,25(6):675-680.
[7] SONG W,YU X,GUO D,et al.MRI-based radiomics:associations with the recurrence-free survival of patients with hepatocellular carcinoma treated with conventional transcatheter arterial chemoembolization[J].J Magn Reson Imaging,2020,52(2):461-473.
[8] CHEN M,CAO J,HU J,et al.Clinical-radiomic analysis for pretreatment prediction of objective response to first transarterial chemoembolization in hepatocellular carcinoma[J].Liver Cancer,2021,10(1):38-51.
[9] JIN Z,CHEN L,ZHONG B,et al.Machine-learning analysis of contrast-enhanced computed tomography radiomics predicts patients with hepatocellular carcinoma who are unsuitable for initial transarterial chemoembolization monotherapy:a multicenter study[J].Transl Oncol,2021,14(4):101034.
[10] KUDO M,REN Z,GUO Y,et al.Transarterial chemoembolisation combined with lenvatinib plus pembrolizumab versus dual placebo for unresectable,non-metastatic hepatocellular carcinoma(LEAP-012):a multicentre,randomised,double-blind,phase 3 study[J].Lancet,2025,405(10474):203-215.
[11] SANGRO B,KUDO M,ERINJERI J P,et al.Durvalumab with or without bevacizumab with transarterial chemoembolisation in hepatocellular carcinoma(EMERALD-1):a multiregional,randomised,double-blind,placebo-controlled,phase 3 study[J].Lancet,2025,405(10474):216-232.
[12] YANG F,XU G,HUANG J,et al.Transarterial chemoembolization combined with immune checkpoint inhibitors and tyrosine kinase inhibitors for unresectable hepatocellular carcinoma:efficacy and systemic immune response[J].Front Immunol,2022,13:847601.
[13] PINATO D J,MURRAY S M,FORNER A,et al.Trans-arterial chemoembolization as a loco-regional inducer of immunogenic cell death in hepatocellular carcinoma:implications for immunotherapy[J].J Immunother Cancer,2021,9(9):e00311.
[14] JIN Z,CHEN J,ZHU X,et al.Immune checkpoint inhibitors and anti-vascular endothelial growth factor antibody/tyrosine kinase inhibitors with or without transarterial chemoembolization as first-line treatment for advanced hepatocellular carcinoma(CHANCE2201):a target trial emulation study[J].EClinicalMedicine,2024,72:102622.
[15] 朱方,魏娟,赵慧慧,等.仑伐替尼和卡瑞利珠单抗联合TACE术治疗晚期或不可切除肝细胞癌临床疗效[J].东南大学学报(医学版),2022,41(4):557-561.
[16] LOSIC B,CRAIG A J,VILLACORTA M C,et al.Intratumoral heterogeneity and clonal evolution in liver cancer[J].Nat Commun,2020,11(1):291.
[17] RIOS A C.Resolving the spatial heterogeneity of cancer in 3D[J].Nat Rev Cancer,2022,22(10):548-549.
[18] BERA K,BRAMAN N,GUPTA A,et al.Predicting cancer outcomes with radiomics and artificial intelligence in radiology[J].Nat Rev Clin Oncol,2022,19(2):132-146.
[19] SHI Z,HUANG X,CHENG Z,et al.MRI-based quantification of intratumoral heterogeneity for predicting treatment response to neoadjuvant chemotherapy in breast cancer[J].Radiology,2023,308(1):e222830.
[20] BO Z,CHEN B,ZHAO Z,et al.Prediction of response to lenvatinib monotherapy for unresectable hepatocellular carcinoma by machine learning radiomics:a multicenter cohort study[J].Clin Cancer Res,2023,29(9):1730-1740.
[21] KONG C,ZHAO Z,CHEN W,et al.Prediction of tumor response via a pretreatment MRI radiomics-based nomogram in HCC treated with TACE[J].Eur Radiol,2021,31(10):7500-7511.
[22] WANG D,ZHANG L,SUN Z,et al.A radiomics signature associated with underlying gene expression pattern for the prediction of prognosis and treatment response in hepatocellular carcinoma[J].Eur J Radiol,2023,167:111086.
[23] HUA Y,SUN Z,XIAO Y,et al.Pretreatment CT-based machine learning radiomics model predicts response in unresectable hepatocellular carcinoma treated with lenvatinib plus PD-1 inhibitors and interventional therapy[J/OL].J Immunother Cancer,2024,12(7):e008953.doi:10.1136/jitc-2024.008953.
[24] MENG X,WANG Y,JU S,et al.Radiomics analysis on multiphase contrast-enhanced CT:a survival prediction tool in patients with hepatocellular carcinoma undergoing transarterial chemoembolization[J].Front Oncol,2020,10:1196.
[25] LIU K,ZHENG X,LU D,et al.A multi-institutional study to predict the benefits of DEB-TACE and molecular targeted agent sequential therapy in unresectable hepatocellular carcinoma using a radiological-clinical nomogram[J].Radiol Med,2024,129(1):14-28.
[26] DING G,LI K.A CT-based clinical-radiomics nomogram for predicting the overall survival to TACE combined with camrelizumab and apatinib in patients with advanced hepatocellular carcinoma[J].Acad Radiol,2025,32(4):1993-2004.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】

提示：您还未登录，请登录！点此登录