微信公众号

官网二维码
中国癌症防治杂志

中国癌症防治杂志 ›› 2025, Vol. 17 ›› Issue (3): 289-296.doi: 10.3969/j.issn.1674-5671.2025.03.05

• 肝脏肿瘤专栏 • 上一篇    下一篇

基于生物学标志和影像学特征预测肝细胞癌转化治疗后病理完全缓解列线图模型的构建与验证

  

  1. 广西医科大学附属肿瘤医院肝胆胰腺外科；柳州市工人医院肿瘤科
  • 出版日期:2025-06-25 发布日期:2025-07-10
  • 通讯作者: 黎乐群， E-mail:Li_lequn@263.net； 吴飞翔，E-mail:wufeixiang@gxmu.edu.cn
  • 基金资助:
    国家自然科学基金项目(82360537)；广西自然科学基金项目(2025GXNSFBA069386)；区域性高发肿瘤早期防治研究教育部重点实验室项目(GKE-ZZ202309);广西医科大学附属肿瘤医院青年基金项目（2024-027）

Development and validation of a nomogram model for predicting pathological complete response following conversion therapy for hepatocellular carcinoma based on biological markers and imaging features

  • Online:2025-06-25 Published:2025-07-10

PDF (PC)

8

赞

摘要： 目的 通过整合影像学特征和生物学标志物构建一个列线图模型，以预测接受转化治疗方案的肝细胞癌(hepatocellular carcinoma，HCC)患者病理完全缓解(pathologic complete response，pCR)。方法 研究对象为2019年11月至2024年10月期间，在广西医科大学附属肿瘤医院接受经导管动脉化疗栓塞术和/或肝动脉灌注化疗，并联合靶向治疗和免疫治疗转化治疗方案，并随后接受肝切除术的HCC患者。通过单因素及多因素logistic分析，筛选pCR独立预测因素，并基于这些因素构建列线图模型。模型性能评估采用受试者工作特征曲线下面积(area under the curve，AUC)、校准曲线及决策曲线分析（decision curve analysis，DCA）。结果 在纳入的135例HCC患者中，27.4%(37/135)在接受治疗后达pCR。全身炎症反应指数(systemic inflammatory response index,SIRI)、肿瘤标志物应答、肿瘤数目以及根据改良实体肿瘤疗效评估标准(modified Response Evaluation Criteria in Solid Tumors,mRECIST)评估的肿瘤完全缓解是pCR的独立预测因素(均P<0.05)。构建的列线图模型AUC为0.925（95%CI:0.882~0.967），与甲胎蛋白（alpha⁃fetoprotein, AFP）应答（AUC=0.655）或mRECIST 完全缓解（AUC=0.785）相比，该模型的预测性能显著更优（均P<0.001）。通过1 000次自助重抽样进行的内部验证显示，列线图模型的AUC为0.918 (95%CI:0.873~0.963)，校准曲线证实模型有较好的校准效果，决策曲线分析表明该模型具有重要的临床应用价值。结论 结合SIRI、肿瘤标志物应答、肿瘤数目及mRECIST 完全缓解的列线图模型，能够有效预测HCC患者转化治疗后的pCR，为个体化手术决策提供依据。

关键词: 肝细胞癌, 转化治疗, 列线图, 病理完全缓解, 生物标志物, 影像学特征

Abstract: Objective To construct a nomogram that integraties imaging features and biomarkers to predict pathologic complete response (pCR) in patients with hepatocellular carcinoma (HCC) undergoing conversion therapy.  Methods The study cohort comprised HCC patients who received the transcatheter arterial chemoembolization (TACE) and/or hepatic arterial infusion chemotherapy (HAIC) in conjunction with targeted therapy and immunotherapy at Guangxi Medical University Cancer Hospital from November 2019 to October 2024. Independent predictors of pCR were identified through univariable and multivariable logistic regression analyses, and these predictors were utilized to develop the nomogram. The performance of the nomogram was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA).  Results Among the 135 patients with HCC, 27.4% (37/135) achieved pCR following treatment. The systemic inflammatory response index (SIRI), tumor biomarker response, tumor number, and tumor complete response as assessed by the modified Response Evaluation Criteria in Solid Tumors (mRECIST) were identified as independent predictors of pCR (all P<0.05). A nomogram model was developed with AUC of  0.925 (95%CI: 0.882-0.967), demonstrating significantly superior predictive performance compared to the alpha⁃fetoprotein  (AFP) response (AUC=0.655) or mRECIST complete response (AUC=0.785)  (both  P<0.001). Internal validation using 1,000 times bootstrap resamples resulted in an AUC of 0.918 (95%CI: 0.873-0.963) for the nomogram model. The calibration curve confirmed excellent model calibration， and DCA demonstrated significant clinical utility. Conclusions The nomogram model,  incorporating SIRI, tumor biomarker response, tumor number, and mRECIST complete response,  provides an accurate pCR prediction following HCC conversion therapy in HCC patients and may serve as a foundation for individualized surgical decision⁃making.

Key words: Hepatocellular carcinoma, Conversion therapy, Nomogram, Pathologic complete response, Biomarker, Imaging features

中图分类号: 

  • R735.7