Construct a Prognostic Model for Patients with Lung Adenocarcinoma by Using Glycolysis-related LncRNA
Objective To construct a prognostic model of lung adenocarcinoma patients by using glycolysis-related LncRNA,and to help predict the efficacy of individualized drugs and disease recurrence.Methods The TCGA and GSEA databases were used to screen the expression data of lncRNA related to glycolysis in lung adenocarcinoma.The prognostic model was constructed by LASSO and Cox regression analysis.The receiver operating characteristic curve(ROC)was drawn and calibrated.The clinicopathological features and risk scores were integrated to construct a nomogram.The relationship between immune cell distribution,immune-related molecules and drug sensitivity and risk score was analyzed.Results Four effective glycolysis gene sets(BioCarta,Hallmark,KEGG,REACTOME and WP)were selected from the GSEA database,and 1025 glycolystic-related lncRNAs were obtained by combining with the expression data of lncRNAs in the TCGA data.A total of 186 glycolytic-related lncRNAs were differentially expressed between tumor and normal tissues by differential analysis,and 19 prognostic related lncRNAs were obtained by univariate COX and LASSO regression analysis.A prediction model consisting of 12 lncRNAs was obtained by Cox proportional hazard regression analysis.The ACU value of the model suggested that the prediction performance was good,and the AUC of 1,3 and 5 years survival time were 0.711,0.713 and 0.699,respectively.The patients with lung adenocarcinoma could be divided into high and low risk groups,and the difference of overall survival(OS)between the two groups was statistically significant(P＜0.05).Univariate and multivariate Cox analysis showed that risk score could be used as an independent prognostic indicator for the survival of lung adenocarcinoma,and the risk score predicted better than other clinicopathologic features.In addition,there were statistically significant differences in risk scores between genders,T,N,M,and Stage(P＜0.05).Risk scores and histograms constructed with clinicopathological features improved prognostic ability at 1,3,and 5 years(AUC at 1,3,and 5 years survival time was 0.741,0.750,and 0.715,respectively).There were statistically significant differences in immune microenvironment between the high and low risk groups,showing that most immune cells were positively correlated with the low risk score.Drug sensitivity analysis suggested that there were significant differences in drug sensitivity of mitomycin C,paclitaxel,rapamycin,docetaxel and erotinib between the two groups.Conclusion The prognosis model of lung adenocarcinoma constructed by glycolysis-related lncRNA can effectively and accurately predict the prognosis of patients with lung adenocarcinoma,which has certain clinical significance.
万方数据
维普
