基于铜死亡相关LncRNA的肝细胞癌预后预测模型构建及验证  

Construction and validation of a prognostic prediction model for hepatocellular carcinoma based on cuproptosis-related LncRNA

作　　者：李京城 张永芳 胡婵 刘玲 彭倩 桑圣刚[1] 曾江正[3] LI Jingcheng;ZHANG Yongfang;HU Chan;LIU Ling;PENG Qian;SANG Shenggang;ZENG Jiangzheng(Clinical Laboratory,The First Affiliated Hospital of Hainan Medical University,Haikou 570100,China;不详)

机构地区：[1]海南医学院第一附属医院检验科,海口570100 [2]海南医学院第一附属医院病理科 [3]海南医学院第一附属医院肿瘤内科

出　　处：《山东医药》2022年第31期5-10,共6页Shandong Medical Journal

基　　金：国家自然科学基金资助项目(82160556)。

摘　　要：目的 基于铜死亡相关长链非编码RNA(LncRNA)建立肝细胞癌(HCC)预后预测模型,并进行效能验证。方法 从癌症基因图谱(TCGA)数据库获取肝细胞癌的RNA表达数据和临床数据,从文献中获取铜死亡相关基因集,相关性分析筛选铜死亡相关的LncRNA,Lasso回归和Cox回归筛选有临床预后意义的LncRNA并构建预后预测模型,通过生存分析、受试者工作特征曲线、校准曲线评估该模型的预测效能。结果 共纳入6个具有独立预后意义的LncRNA(PICSAR、AC026356.1、MKLN1-AS、AL928921.1、LINC00479、AC026412.3),并成功构建预后预测模型。分别对整个数据集、测试集、训练集按高、低风险分组进行生存分析,结果表明,在3个数据集中高风险组的总生存期均短于低风险组,整个数据集中高风险组的无进展生存期短于低风险组(P均<0.05);随着风险评分的增加,样本死亡的数量也在增加,在测试集与整个数据集中也验证了该结论。该风险模型预测1、3、5年生存率的曲线下面积分别为0.765、0.664、0.649;与TNM分期(0.667)、年龄(0.538)、性别(0.512)、分化程度(0.507)相比,风险评分模型的曲线下面积(0.765)最大。单因素和多因素Cox回归分析表明,TNM分期、风险评分是独立预后因素;绘制列线图,一致性指数为0.679(95%CI:0.652~0.705),结合校准曲线,表明预测的生存率与实际观测到的生存率基本一致。结论 根据6个铜死亡相关LncRNA成功构建了HCC患者的预后预测模型,经验证该模型预测效能较高。Objective To establish a prognostic prediction model for hepatocellular carcinoma(HCC) patients based on cuproptosis-related long non-coding RNA(LncRNA) and to validate its efficacy.Methods We obtained RNA expression data and clinical data of HCC from The Cancer Genome Atlas(TCGA).Getting cuproptosis-related gene sets from literatures.Correlation analysis was used to screen cuproptosis-related LncRNAs.We used Lasso regression and Cox regression to screen LncRNAs with clinical prognostic significance and constructed a prognosis prediction model.The predictive efficacy of the model was evaluated by survival analysis,receiver operating characteristic(ROC) curve,and calibration curve.Results A total of 6 LncRNAs with independent prognostic significance(PICSAR,AC026356.1,MKLN1-AS,AL928921.1,LINC00479,and AC026412.3) were included to construct the prognostic prediction model.The survival analysis of the whole data set,test set and training set was carried out according to the high and low risk groups.The results showed that in three data sets,the overall survival time of the high risk group was shorter than that of the low risk group,and in the whole data set,the progression-free survival time of the high risk group was also shorter than that of the low risk group(both P<0.05).The number of sample deaths increased with the increase of risk score,and this conclusion was verified in the test set and the whole data set.The area under the curve of the risk model in predicting 1-,3-and 5-year survival rates was 0.765,0.664,and 0.649,respectively;compared with the other clinical factors [TNM stage(0.667),age(0.538),sex(0.512),and differentiation(0.507)],the area under the curve of the risk score model(0.765) was the largest.Univariate and multivariate Cox regression analysis showed that TNM stage and risk score were independent prognostic factors.According to the nomogram,its consistency index was 0.679(95% CI:0.652-0.705).Combined with the calibration curve,the predicted survival rate was basically consistent with the obser

关 键 词：肝细胞癌 铜死亡相关基因 长链非编码RNA 预后预测模型 

分 类 号：R735.7[医药卫生—肿瘤]