标题：基于TCGA数据库肺腺癌衰老相关基因预后模型的建立与验证

      作者：邢龙 1，2，吴双丽 3，吴铁成 1，徐敬宣 1，李保健 1，田平 1，李醒亚 2    邢龙 1，2，吴双丽 3，吴铁成 1，徐敬宣 1，李保健 1，田平 1，李醒亚 21.青岛滨海学院附属医院肿瘤科，山东 青岛 266500；2.郑州大学第一附属医院肿瘤科，河南 郑州 450000；3.青岛滨海学院附属医院特检科，山东 青岛 266500

      卷次： 2024年35卷10期

      【摘要】 目的 探究肺腺癌(LUAD)差异表达的衰老相关基因(ARGs)，并构建LUAD衰老相关基因的预后模型。方法 LUAD的 RNA高通量转录组数据下载于癌症基因组图谱(TCGA)数据库，Aging Atlas数据库获取ARGs，并基于LUAD中衰老相关差异表达基因构建LUAD预后模型并验证，进一步构建列线图及校准曲线探究模型在临床中的应用价值。结果 通过韦恩图去交集共获得80个衰老相关差异表达基因。通过单因素、多因素Cox回归分析筛选得到 7个具有独立预后意义的差异表达ARGs，对筛选出的 7个基因进行LASSO回归分析并构建LUAD预后模型。构建Kaplan-Meier生存曲线显示，与低风险组相比，高风险组与较差的OS显著相关，差异具有统计学意义(P<0.05)。1、3和 5年受试者操作特性(ROC)曲线下面积(AUC)分别为 0.706、0.725、0.642，提示风险模型有较高的灵敏度和特异度。单因素及多因素Cox回归显示，风险评分为LUAD独立预后因素。进一步基于风险评分构建列线图及校准曲线，校准曲线列线图全局的一致性为 0.729，说明该模型预测结果与实际情况有较高的准确度。结论 基于差异表达的ARGs构建的风险模型可作为LUAD的一种预后特征，为LUAD患者个体化治疗提供参考。
      【关键词】 衰老基因；肺腺癌；预后模型；TCGA数据库；生物信息学
      【中图分类号】 R734.2 【文献标识码】 A 【文章编号】 1003—6350（2024）10—1374—06

Establishment and validation of prognostic model for aging-related genes of lung adenocarcinoma based onTCGA database.
XING Long 1,2, WU Shuang-li 3, WU Tie-cheng 1, XU Jing-xuan 1, LI Bao-jian1, TIAN ping 1, LIXing-ya 2. 1. Department of Oncology, Affiliated Hospital of Qingdao Binhai University, Qingdao 266500, Shandong,CHINA; 2. Department of Oncology, the First Affiliated Hospital of Zhengzhou University, Zhengzhou 450000, Henan,CHINA; 3. Department of Special Examination, Affiliated Hospital of Qingdao Binhai University, Qingdao 266500,Shandong, CHINA
【Abstract】 Objective To explore the aging-related genes (ARGs) of lung adenocarcinoma (LUAD) and con-struct the prognostic model for LUAD age-related genes. Methods The high-throughput RNA transcriptome data ofLUAD were downloaded from The Cancer Genome Atlas (TCGA) database, and ARGs were obtained from the AgingAtlas database. The LUAD prognostic model was constructed based on age-related differentially expressed genes inLUAD and then verified. The clinical application value of the model was further explored by constructing the nomogramand calibration curve. Results A total of 80 differentially expressed genes related to aging were obtained through Venndiagram intersection. Seven differentially expressed ARGs with independent prognostic significance were screened byunivariate and multivariate Cox regression analysis. The seven genes were screened by LASSO regression analysis, andthe LUAD prognostic model was constructed. Kaplan-Meier survival curve showed that compared with the low-riskgroup, the high-risk group was significantly correlated with poor OS, and the difference was statistically significant (P<0.05). The area under (AUC) the receiver operating characteristic (ROC) curve for 1, 3, and 5 years was 0.706, 0.725, and0.642, respectively, which suggested that the risk model had high sensitivity and specificity. Univariate and multivariate Cox re-gression showed that risk scores were independent prognostic factors for LUAD. The C-index of the calibration curve was0.729, indicating that the prediction results of the model had high accuracy with the actual situation. Conclusion The riskmodel constructed based on differentially expressed ARGs in this study can be used as a prognostic feature of LUAD, whichprovides reference for the individualized treatment of LUAD patients.
      【Key words】 Aging-related genes; Lung adenocarcinoma; Prognostic model; TCGA database; Bioinformatics