肺腺癌中糖酵解代谢特征预测预后及指导治疗

Characterization of the glycolytic metabolism in lung adenocarcinoma to prodict prognosis and guide clinical therapy

目的:有氧糖酵解作为肿瘤独特的代谢特征,其作用机制仍未被完全阐明。本研究探索糖酵解代谢在肺腺癌(lung adenocarcinoma,LUAD)及其免疫微环境中的作用,以寻求糖酵解代谢与LUAD预后和免疫治疗效果的关系。方法:从癌症基因组图谱(The Cancer Genome Atlas,TCGA)和基因表达综合(Gene Expression Omnibus,GEO)数据库下载LUAD患者的基因表达数据和临床信息,采用单变量、最小绝对收缩和选择算子(least absolute shrinkage and selection operator,LASSO)Cox回归分析构建糖酵解代谢相关基因的LUAD预后评估模型。结果:在TCGA队列中,高危评分患者的预后更差,在GEO队列中也证实了此模型的可靠性。利用统计学工具获得的由糖酵解相关基因组成的预后模型和列线图模型为预测预后提供了一种新方法,并在15对LUAD和正常组织中验证了风险标签中的15个基因,它们在LUAD组织中表达异常。此外,风险评分与LUAD免疫微环境相关,经模型评估发现高危评分患者更适合接受免疫治疗。结论:本研究构建了与糖酵解代谢相关的15个基因组成的预后模型,该模型可预测患者对免疫治疗的敏感性,可用来指导临床实践,促进肿瘤的个体化治疗。

Objective:Aerobic glycolysis is a typical metabolic feature of tumors,but the mechanisms of its action remain unclear.This article is dedicated to exploring the role of glycolytic metabolism in lung adenocarcinoma(LUAD)and its immune microenvironment to seek potential diagnostic and therapeutic strategies for LUAD and to investigate the relationship among glycolytic metabolism,LUAD prognosis,and immunotherapy.Methods:Gene expression data and clinical information of LUAD patients were downloaded from The Cancer Genome Atlas(TCGA)and Gene Expression Omnibus(GEO)databases.A LUAD prognosis risk scoring model related to genes for glycolysis was constructed by using univariate,Least Absolute Shrinkage and Selection Operator(LASSO)Cox Regression Analysis.Results:Patients with high-risk scores in the TCGA cohort had a worse prognosis,and the reliability was consistently confirmed in the GEO cohort.The prognosis model and the nomogram model composed of glycolysis-related genes obtained by using statistical tools provideed a new method for predicting prognosis.The conclusion was verified in paired tumor and normal tissues collected from 15 patients with LUAD,as observed by 15 genes serving as risk markers were abnormally expressed in LUAD tissues compared to normal tissues.In addition,the risk score was related to the immune microenvironment and model assessment indicated that patients with high-risk scores were more suitable for immunotherapy.Conclusion:This study created a new prognosis model composed of 15 genes related to glycolysis,which can also predict patient sensitive to immunotherapy.It can be used to guide clinical practice and promote the personalized tumor treatment.