TCGA And Single-cell Sequencing-based Screening Of Glycolysis-related Prognostic Markers For EGFR Wild-type Lung Adenocarcinoma

Objective:Lung cancer is the most common cancer in China and the leading cause of cancer-related death in the country.The complex etiology and pathogenesis of lung cancer pose challenges in clinical diagnosis and treatment,including low five-year survival rates,unclear mechanisms of disease development,and limited clinical treatment options for patients with wild-type EGFR lung cancer.Therefore,exploring the pathogenesis and tumor progression mechanisms of wild-type EGFR lung cancer,identifying new biomarkers and potential therapeutic targets,discovering more combination therapy regimens and personalized precision treatment,and improving patient survival and quality of life are of utmost importance.This study used single-cell sequencing technology based on next-generation sequencing and transcriptome sequencing data analysis to explore the interaction between tumor cells and their surrounding complex tumor microenvironment,conduct a detailed analysis of tumor occurrence and development at the molecular level.The study aims to identify prognostic markers for wild-type EGFR lung adenocarcinoma,establish an effective clinical prognostic model,and explore relevant molecular regulatory mechanisms affecting lung adenocarcinoma prognosis,providing a theoretical basis for the pathogenesis research and future clinical diagnosis and treatment of lung adenocarcinoma.Research Methods:Differential expression analysis and survival analysis were employed based on published transcriptome sequencing and single-cell RNA sequencing data to identify genes that were associated with the prognosis of EGFR wild-type lung adenocarcinoma.LASSO regression was performed to establish risk score model.Univariate and multivariate Cox regression was used to discover the prognosis-related influencing factors and establish survival prediction model combined with the clinical information of patients.Model performance is evaluated by calculating the C-index,plotting calibration curves and clinical decision curves.Next,tumor cell subpopulations associated with prognostic phenotypes were identified by the Scissor algorithm.The biological processes represented by prognostic markers were studied utilizing gene set enrichment analysis,functional annotation,metabolic scoring,etc.Cell communication analysis and protein-protein interaction network were used to further study the gene regulatory relationship between candidate prognostic markers and the related molecular biological mechanisms affecting prognosis.Immunohistochemical staining was performed to verify the expression of potential prognostic biomarkers in EGFR wild-type lung adenocarcinoma tissues.Results:Nine differentially expressed genes(BIRC3,PABPC1,ENO1,PKM,B4GALT1,DCBLD2,MYLIP,NTHL1,CD74)were screened out by differential expression analysis.The risk score signature composed of them was found to be an independent prognostic factor for lung adenocarcinoma(P<0.001).A comprehensive prediction model incorporating the risk score and the corresponding clinical characteristics was constructed to predict the overall survival of lung cancer patients utilizing multivariate Cox proportional hazards regression.The results showed that the age(HR:1.048,95%CI:1.022-1.073,P=0.001),AJCC TNM stage(Stage II HR:2.294,95%CI:1.253-4.198,P=0.007;Stage III/IV HR:3.595,95%CI:2.037-6.346,P<0.001)and risk score(HR:1.267,95%CI:1.175-1.366,P<0.001)were prognostic risk factors in patients with lung adenocarcinoma.The model showed good predictive performance(C-index=0.81).Gene enrichment analysis and metabolic scoring revealed significantly enhanced glycolytic metabolic activity in poor prognosis-associated lung adenocarcinoma tumor subpopulations.Cell communication analysis and protein-protein interaction network results showed that there were frequent cellular interactions between cancer-associated fibroblasts and tumor cells,and the GAS6/AXL receptor-ligand pair between them was significantly activated.There is a network of interactions between the PI3K/AKT,HIF-1αpathways,and glycolytic enzymes ENO1 and PKM in lung adenocarcinoma tumor cells,which is associated with altered glycolytic activity and poor prognosis in lung adenocarcinoma.IHC staining results showed that compared with adjacent normal tissues,the expression of glycolysis-related enzymes ENO1(χ~2=5.304,P=0.021)and PKM(χ~2=10.167,P=0.001)were significantly increased in EGFR wild-type lung adenocarcinoma tissues.Conclusions:1.A prognostic model integrated gene expression and clinical information was established for EGFR wild-type lung adenocarcinoma;2.The activation of glycolysis in tumor cells was associated with poor prognosis of lung adenocarcinoma;3.The expression of glycolysis-related enzymes ENO1 and PKM was significantly increased in lung adenocarcinoma tissues compared with adjacent normal tissues.