| Objective: Lung cancer is one of the most common malignant tumors in the world today,and both the morbidity and mortality of patients are increasing year by year.Lung adenocarcinoma(LUAD)is the most common histological type of lung cancer,accounting for approximately 50%-70% of lung cancers removed by surgery.Immune cells and infiltrating stromal cells are the main components of normal cells in tumor tissue and play an important role in tumor biology.Previous studies have shown that the infiltration degree of tumor microenvironment cells,immune cells and stromal cells plays a vital role in prognosis.However,many current studies mainly focus on the influence of single gene on the prognosis of lung adenocarcinoma.Therefore,this study aims to establish a multi-gene survival risk assessment model related to the microenvironment of lung adenocarcinoma,so as to effectively predict the prognosis of lung adenocarcinoma patients.Methods: In this study,the Estimate algorithm was used to calculate the immune and stromal scores of patients with lung adenocarcinoma in the TCGA(Tumor Genome Atlas)database,and the relationship between the immune/stromal scores and patient subtype,stage,and prognosis was analyzed.According to their immune/stromal score,they were divided into high and low groups.The genes with biomarker characteristics were screened by Cox regression analysis for survival analysis.The differentially expressed genes of the high and low groups were analyzed,and the GO(gene ontology)enrichment analysis and KEGG(Kyoto Encyclopedia of Genes and Genomes)analysis of the effective target genes was carried out through the G: profiler database,and the PPI network was obtained through the STRING.These genes were verified by GEO(Comprehensive Gene Expression Database)dataset,and qRT-PCR,Western Blot,and IHC were used for validation.The association between prognostic gene expression and immune infiltration was evaluated using the TIMER database.A specific model for risk grading was developed based on the best prognostic characteristic genes identified in the current analysis.The ability of the model to predict patient outcomes was verified using the Kaplan-Meier curve.A nomogram for overall survival prediction was further constructed by integrating risk grading based on immune scores with clinicopathological risk factors.Finally,the LUAD mutation data was downloaded from the TCGA database,the relationship between the high and low risk groups and the gene mutation frequency of LUAD patients was analyzed,and the tumor mutation burden(TMB)of each sample was calculated.Results: 1.Immune and stromal scores of 495 patients with lung adenocarcinoma in TCGA database were calculated,and the relationship between immune and stromal scores and pathological parameters and prognosis of lung adenocarcinoma was statistically analyzed.The order of immune/stromal scores of LUAD subtypes from high to low was Squamoid > Bronchioid > Magnoid.Patients with Stage III + IV had lower immune and stromal scores,but stromal scores’ difference between patients with Stage I+ II and Stage III + IV was not statistically significant.Patients in the low immune score group had a higher median overall survival time(P=0.0322).Compared with patients with a higher stromal score,patients with a lower stromal score also had a shorter median overall survival time,but the difference was not statistically significant.2.Based on the comparison of immune scores,744 genes in the high group were up-regulated and 43 genes were down-regulated.In the low stromal score group,2688 genes were up-regulated and 253 genes were down-regulated in the high score group.Among the differential expression genes between high/low immune score group,84 differential expression genes showed significant survival differences.3.Functional enrichment analysis was performed on 84 genes,and the GO category with the largest number was immune system process,KEGG pathway analysis shows that differentially expressed genes are involved in signaling pathways such as primary immunodeficiency.PPI network showed that several key genes of immune response occupied the center of the module,including WDFY4,CD52,PLCB2,etc.4.Gene expression data of 84 LUAD patients from GEO database were downloaded and analyzed.A total of four genes were found to be significantly associated with prognosis.5.Real-time quantitative polymerase chain reaction(qRT-PCR),Western Blot and Immunohistochemistry(IHC)results showed that the expressions of DSE,EMR2 and GFPT2 in lung adenocarcinoma tissues were significantly higher than those in adjacent tissues,while the expression of FLT3 was significantly lower.6.Timer database analysis showed that DSE expression was positively correlated with infiltration of neutrophilsand dendritic cells.The expression of EMR2 was positively correlated with the infiltration of neutrophils and dendritic cells.The expression of FLT3 was positively correlated with the infiltration of B cells,CD4+ T cells,neutrophils and dendritic cells.GFPT2 expression was positively correlated with neutrophil infiltration.7.The entire cohort was divided into high-and low-risk groups according to the risk score,and the higher the risk score,the shorter the survival time(P = 0.0022).The C index of the nomogram obtained by integrating the characteristics of the 4 genes and other clinical features was 0.715,which was effective in detecting the 1-year,3-year and5-year clinical survival rate.8.TP53,TTN and MUC16 mutated more frequently in the high-risk group,and the TMB of the high-risk group was higher than that of the low-risk group.Conclusion: 1.Immune score is correlated with prognosis,overall survival,subtype and stage of lung adenocarcinoma patients.2.Prognostic differentially expressed genes based on immune score are involved in the biological process related to the development of lung adenocarcinoma.3.DSE,EMR2,GPT2 and FLT3 were correlated with the prognosis of patients with lung adenocarcinoma.Gene markers based on immune score can be used as a tool for prognostic stratification of lung adenocarcinoma. |
PDF Full Text Request