肺腺癌中胆固醇代谢亚型的探索与生存预测模型构建  

Exploration of Cholesterol Metabolism Subtypes and Construction of Survival Prediction Models in Lung Adenocarcinoma

作　　者：李敬儒 曹邦伟[1] LI Jingru;CAO Bangwei(Dept.of Oncology,Beijing Friendship Hospital,Capital Medical University,Beijing 100050,China)

机构地区：[1]首都医科大学附属北京友谊医院肿瘤科,北京100050

出　　处：《中国医院用药评价与分析》2024年第12期1409-1412,共4页Evaluation and Analysis of Drug-use in Hospitals of China

基　　金：国家自然科学基金资助项目(No.82173056)。

摘　　要：目的:通过生物信息学方法分析胆固醇代谢相关基因在肺腺癌中的表达变化,并基于这些基因构建一个预后风险模型,以期为肺腺癌的诊断和药物治疗提供新的思路与靶点。方法:从癌症基因组图谱计划(TCGA)中获取600例肺腺癌样本的临床生存数据及RNA测序数据,通过差异表达分析鉴定在肿瘤和正常组织中表达不同的基因,并从分子特征数据库筛选胆固醇代谢相关基因。利用Boruta算法和多因素Cox比例风险模型构建胆固醇代谢评分模型,并在GEO数据库的GSE30219数据集中进行验证。结果:在TCGA数据库中,共鉴定出28个胆固醇代谢相关的差异表达基因。通过Boruta算法筛选出对肺腺癌患者总生存期有显著影响的18个基因,构建了胆固醇代谢评分模型。该模型在训练集和验证集中均表现出良好的预测能力,Kaplan-Meier生存曲线分析显示低代谢评分组的生存显著优于高代谢评分组。结论:本研究构建的胆固醇代谢评分模型在肺腺癌预后评估中具有显著的预测能力,揭示了胆固醇代谢在肺腺癌中的重要作用,为预后评估和靶向药物治疗提供了新的研究方向。OBJECTIVE:To analyze the expression changes of cholesterol metabolism-related genes in lung adenocarcinoma through bioinformatics methods,and to construct a prognostic risk model based on these genes,in order to provide new thoughts and targets for the diagnosis and drug treatment of lung adenocarcinoma.METHODS:Clinical survival data and RNA sequencing data of 600 lung adenocarcinoma samples were obtained from the Cancer Genome Atlas(TCGA)project,genes differentially expressed in tumors and normal tissues were identified by differential expression analysis,and cholesterol metabolism-related genes were screened from the molecular signature database.The cholesterol metabolism scoring model was constructed by using the Boruta algorithm and the multifactorial Cox proportional hazard model,which were validated in the GSE30219 dataset of the GEO database.RESULTS:A total of 28 cholesterol metabolism-related differentially expressed genes were identified in the TCGA database.The Boruta algorithm screened out 18 genes that significantly affect the overall survival of lung adenocarcinoma patients,and constructed a cholesterol metabolism scoring model.This model showed good predictive ability in both the training set and validation set,and Kaplan-Meier survival curve analysis showed that the survival of the low metabolism score group was significantly better than that of the high metabolism score group.CONCLUSIONS:The cholesterol metabolism scoring model constructed in this study has significant predictive ability in the prognostic evaluation of lung adenocarcinoma,which reveals the important role of cholesterol metabolism in lung adenocarcinoma,provides a new research direction for prognostic evaluation and targeted drug therapy.

关 键 词：肺腺癌 胆固醇代谢 Boruta算法 COX比例风险模型 

分 类 号：R979.1[医药卫生—药品]