Abstract
Background: Lung adenocarcinoma ranks as the second most widespread form of cancer globally, accompanied by a significant mortality rate. Several studies have shown that T cell exhaustion is associated with immunotherapy of tumours. Consequently, it is essential to comprehend the possible impact of T cell exhaustion on the tumor microenvironment. The purpose of this research was to create a TEX-based model that would use single-cell RNA-seq (scRNA-seq) and bulk-RNA sequencing to explore new possibilities for assessing the prognosis and immunotherapeutic response of LUAD patients. Methods: RNA-seq data from LUAD patients was downloaded from the Cancer Genome Atlas (TCGA) database and the National Center for Biotechnology Information (GEO). 10X scRNA sequencing data, as reported by Bischoff P et al., was utilized for down-sampling clustering and subgroup identification using TSNE. TEX-associated genes were identified through gene set variance analysis (GSVA) and weighted gene correlation network analysis (WGCNA). We utilized LASSO-Cox analysis to establish predicted TEX features. External validation was conducted in GSE31210 and GSE30219 cohorts. Immunotherapeutic response was assessed in IMvigor210, GSE78220, GSE35640 and GSE100797 cohorts. Furthermore, we investigated differences in mutational profiles and immune microenvironment between various risk groups. We then screened TEXRS key regulatory genes using ROC diagnostic curves and KM survival curves. Finally, we verified the differential expression of key regulatory genes through RT-qPCR. Results: Nine TEX genes were identified as highly predictive of LUAD prognosis and strongly correlated with disease outcome. Univariate and multivariate analysis revealed that patients in the low-risk group had significantly better overall survival rates compared with those in the high-risk group, highlighting the model's ability to independently predict LUAD prognosis. Our analysis revealed significant variation in the biological function, mutational landscape, and immune cell infiltration within the tumor microenvironment of both high-risk and low-risk groups. Additionally, immunotherapy was found to have a significant impact on both groups, indicating strong predictive efficacy of the model. Conclusions: The TEX model showed good predictive performance and provided a new perspective for evaluating the efficacy of preimmunization, which provides a new strategy for the future treatment of lung adenocarcinoma.