甲胎蛋白阳性肝硬化肝细胞癌特异性生存期预后模型的构建:一项回顾性研究  

Construction of a specific survival prognosis model for alpha-fetoprotein-positive hepatocellular carcinoma in cirrhosis:a retrospective study

作　　者：姜树森 姚红兵 谭李军 王雪尧 韩欣萌 向桢 JIANG Shusen;YAO Hongbing;TAN Lijun;WANG Xueyao;HAN Xinmeng;XIANG Zhen(Department of Hepatobiliary and Pancreatic Surgery,The Second Affiliated Hospital of Guilin Medical University,Guilin 541199,Guangxi,China;Department of Nursing,The Second Affiliated Hospital of Guilin Medical University,Guilin 541199,Guangxi,China;Health Examination Centre,Affiliated Hospital of Guilin Medical University,Guilin 541004,Guangxi,China)

机构地区：[1]桂林医学院第二附属医院肝胆胰外科,广西桂林541199 [2]桂林医学院第二附属医院护理部,广西桂林541199 [3]桂林医学院附属医院健康体检中心,广西桂林541004

出　　处：《右江医学》2025年第3期204-213,共10页Chinese Youjiang Medical Journal

基　　金：广西壮族自治区医疗卫生适宜技术开发与推广应用基金资助项目(S2021008,S2024076);广西壮族自治区医疗卫生重点培育学科建设基金资助项目(桂卫科教发[2021]8号);广西壮族自治区卫生健康委自筹经费科研课题(Z-C20231010)。

摘　　要：目的构建预测甲胎蛋白阳性肝硬化肝细胞癌(AHCC)患者癌症特异性生存期(CSS)的预后模型。方法回顾性分析SEER数据库2010—2015年间866例AHCC患者的数据。患者以7∶3的比例随机分为训练集和验证集。先后采用单因素及多因素Cox回归分析识别与CSS相关的独立预后因素,并以此为基础构建预后模型和可视化列线图。应用受试者工作特征(ROC)曲线、曲线下面积(AUC)、校准曲线以及临床决策曲线(DCA)评估模型性能。利用X-tile软件对患者进行风险分层,并通过Kaplan-Meier法和Log-rank检验评估不同风险组的生存差异。结果组织学分级、肿瘤大小、T分期、N分期、M分期、手术和淋巴结清扫是AHCC患者CSS的独立预后因素。ROC曲线显示预后模型预测效果良好。训练集中,预后模型预测AHCC患者1年、3年和5年CSS的AUC分别为0.838(95%CI:0.802~0.874)、0.854(95%CI:0.824~0.884)和0.854(95%CI:0.824~0.884);验证集对应的AUC分别为0.836(95%CI:0.783~0.889)、0.864(95%CI:0.819~0.909)和0.868(95%CI:0.825~0.911)。校准曲线显示预测值与实际观察值高度一致,DCA证明模型具有良好的临床应用价值。根据列线图得分,成功将患者分为低风险组(得分<190分)、中风险组(得分为190~264分)及高风险组(得分>264分),组间预后差异有统计学意义(P<0.001)。结论本研究开发的预后模型在预测AHCC患者CSS方面表现出较高的准确性和可靠性,为个体化临床决策提供重要参考。Objective To construct a prognostic model for predicting cancer-specific survival(CSS)in patients with alpha-fetoprotein-positive cirrhotic hepatocellular carcinoma in cirrhosis(AHCC).Methods A retrospective analysis was conducted on the data of 866 AHCC patients in the SEER database from 2010 to 2015.Patients were randomly divided into training cohort and validation cohort in a 7∶3 ratio.Univariate and multivariate Cox regression analyses were successively conducted to identify independent prognostic factors associated with CSS,and based on this,a prognostic model and a nomogram were constructed.The performance of the model was evaluated by receiver operating characteristic(ROC)curves,area under the curve(AUC),calibration curves,and decision curve analysis(DCA).The patients were stratified into different risk groups by X-tile software,and the survival differences among various risk groups were assessed by Kaplan-Meier method and Log-rank test.Results Histological grade,tumor size,T stage,N stage,M stage,surgery,and lymph node dissection were identified as independent prognostic factors for CSS in AHCC patients.The ROC curves demonstrated good predictive performance of the model.In the training set,the AUC values of the prognostic model for predicting 1-year,3-year,and 5-year CSS in AHCC patients were 0.838(95%CI:0.802–0.874),0.854(95%CI:0.824–0.884),and 0.854(95%CI:0.824–0.884),respectively.In the validation set,the corresponding AUC values were 0.836(95%CI:0.783–0.889),0.864(95%CI:0.819–0.909),and 0.868(95%CI:0.825–0.911),respectively.Calibration curves showed excellent agreement between predicted and observed values,and DCA proved that the model had good clinical application value.According to the column chart scores,patients were successfully divided into low-risk group(<190 points),medium risk group(190–264 points),and high-risk group(>264 points),and there was statistically significant difference in prognosis between the groups(P<0.001).Conclusion The prognostic model developed in this study

关 键 词：肝细胞癌 癌症特异性生存期 肝硬化 列线图 

分 类 号：R735.7[医药卫生—肿瘤]