基于乳酸代谢基因的肺腺癌预后模型的建立和评价  

Establishment and evaluation of prognostic model of lung adenocarcinoma based on lactate metabolism gene

作　　者：侯俊杰 米旭光[2] 李孝男 李晓男[2] 杨影 芦小单[2] 方艳秋[2] 金宁一[1,3] HOU Junjie;MI Xuguang;LI Xiaonan;LI Xiaonan;YANG Ying;LU Xiaodan;FANG Yanqiu;JIN Ningyi(Department of Pathophysiology,School of Medical Sciences,Yanbian University,Yanji 133002,China;Department of Comprehensive Oncology,Jilin Province People’s Hospital,Changchun 130021,China;Laboratory of Molecular Virology and Immunology,Academy of Military Medical Science,Changchun 130122,China)

机构地区：[1]延边大学医学院病理生理学教研室,吉林延吉133002 [2]吉林省人民医院肿瘤综合治疗科,吉林长春130021 [3]军事医学科学院分子病毒学与免疫学实验室,吉林长春130122

出　　处：《吉林大学学报（医学版）》2022年第6期1546-1554,共9页Journal of Jilin University：Medicine Edition

基　　金：吉林省卫健委卫生与健康科技创新项目(2020J109);吉林省卫健委卫生健康科技能力提升项目(2021LC058);吉林省卫健委创新人才项目(2021Y024)。

摘　　要：目的:探讨肺腺癌(LUAD)组织中乳酸代谢相关基因以及基于乳酸代谢基因构建LUAD预后评分模型,阐明其预测LUAD预后的能力。方法:采用癌症基因图谱(TCGA)数据库筛选LUAD相关乳酸代谢基因。采用单因素和多因素Cox回归及LASSO回归分析获得关键基因并构建LUAD的乳酸代谢评分模型,采用Kaplan-Meier生存分析和受试者工作特征(ROC)曲线对模型的预测能力进行验证,TIMER法评估该模型与患者临床特征和免疫细胞浸润丰度之间的关系。结果:成功筛选出16个乳酸代谢基因并构建评分模型;生存分析,低风险组患者的总生存时间(OS)明显高于高风险组(P<0.01),且有较高的预测预后能力,ROC曲线下面积(AUC)均高于0.7;多因素Cox回归分析,23个乳酸代谢基因是LUAD患者的独立的预后基因;乳酸评分与B淋巴细胞(r=-0.326,P<0.001)、CD4 T淋巴细胞(r=-0.196,P<0.001)、CD8 T淋巴细胞(r=-0.094,P=0.036)、巨噬细胞(r=-0.198,P<0.001)和树突状细胞(r=-0.119,P=0.008)百分率呈负相关关系。结论:乳酸代谢评分模型可以很好评估LUAD预后与肿瘤微环境(TIMER)的状态,可作为LUAD预后预测的生物标志物。Objective:Objective:To investigate the genes related to lactic acid metabolism in lung adenocarcinoma(LUAD)tissue and establish the prognostic score model based on the genes related to lactic acid metabolism,and to clarify its ability to predict the prognosis of LUAD.Methods:The LUAD related lactate metabolism genes were screened by The Cancer Genome Atlas(TCGA)database.Univariate and multivariate Cox regression analysis and LASSO regression analysis were used to obtain the key genes and the lactate metabolism score model for LUAD was constructed,and the predictive ability of the model was verified by Kaplan-Meier survival analysis and receiver operating characteristic(ROC)curve,and the relationships between the model and the clinical characteristics of the patient and the abundances of immune cell infiltration were evaluated by TIMER method.Results:Sixteen lactate metabolism genes were successfully screened and the score model was constructed.The survival analysis results showed that the overall survival(OS)of the patients in low-risk group was significantly higher than that in high-risk group(P<0.01),and the area under ROC curve(AUC)was higher than 0.7;the multivariate Cox regression analysis results showed that 23 lactic acid metabolism genes were the independent prognosis genes for the LUAD patients,which could accurately assess the survival rate of LUAD patients in combination with other clinical features of the patients(P<0.01),and the lactic acid score was negatively correlated with the percentages of B lymphocytes(r=-0.326,P<0.001),CD4 T lymphocytes(r=-0.196,P<0.001),CD8 T lymphocytes(r=-0.094,P=0.036),macrophages(r=-0.198,P<0.001),and dendritic cells(r=-0.119,P=0.008).Conclusion:The lactate metabolism score model can well evaluate the prognosis of LUAD and the state of tumor microenvironment(TIMER),and can be used as a biomarker to predict the prognosis of LUAD.

关 键 词：乳酸 癌症基因图谱数据库 肺腺癌 肿瘤微环境 预后标志物 

分 类 号：R734.2[医药卫生—肿瘤]