【摘要】 目的 基于代谢基因的生物信息学构建肺腺癌预后模型及验证。方法 获取癌症基因组图谱(TCGA)数据库和基因表达数据集(GEO)肺腺癌相关数据，套索(LASSO)回归构建多基因预后模型并计算风险值(RS)。单因素、多因素Cox独立预后分析，通过受试者工作特征(ROC)曲线评价模型的ROC曲线下面积(AUC)并进行生存分析。构建列线图评价模型的可行性，通过基因集富集分析(GSEA)进行代谢基因功能富集分析。肿瘤免疫评估资源(TIMER)数据库分析患者RS与免疫细胞浸润以及与免疫检查点分子表达的相关性。结果 运用TCGA数据库基于18个代谢相关基因构建肺腺癌预后模型，RS可以作为独立的预后因子。ROC曲线下面积为0.713。生存分析显示，与高风险组相比，低风险组总体生存率更高，预后模型与免疫细胞的浸润以及与免疫检查点分子的表达有关。结论 代谢相关基因肺腺癌预后模型的RS是独立预后因子，模型具有较高的预后判断价值。更多还原

【Abstract】 Objective To construct and validate a prognostic model for lung adenocarcinoma based on bioinformatics of metabolic genes. Methods Lung adenocarcinoma-related data from The Cancer Genome Atlas(TCGA) database and gene expression omnibus(GEO) were acquired, and LASSO regression was used to construct multi-gene prognostic models and calculate risk-score(RS). Univariate and multivariate Cox independent prognostic analysis was performed. The area under receiver operating characteristic(ROC) curve(AUC) of the model was evaluated by ROC curve and survival analysis was performed. Nomogram were constructed to evaluate the feasibility of the model, and metabolic gene functional enrichment analysis was performed by GSEA. Tumor immune estimation resource(TIMER) database was used to analyze the correlation of patients RS with immune cell infiltration and with the expression of immune checkpoint molecules. Results The TCGA database was used to construct a prognostic model for lung adenocarcinoma based on 18 metabolism-related genes, and RS was used as an independent prognostic factor. The area under the ROC curve was 0.713. Survival analysis showed that overall survival was higher in the low-risk group compared to the high-risk group, and the prognostic model was associated with infiltration of immune cells and with the expression of immune checkpoint molecules. Conclusion RS is an independent prognostic factor in the prognostic model of lung adenocarcinoma with metabolic genes, suggesting a high prognostic value of this model.更多还原