Construction of a prognostic model for lung adenocarcinoma based on membrane-tension-related genes

BACKGROUND: Lung adenocarcinoma (LUAD), which is the most common type of non-small cell lung cancer (NSCLC), is one of the most aggressive and fatal tumors. Therefore, the identification of key biomarkers affecting prognosis is important to improving the prognosis of patients with LUAD. Cell membranes have long been understood; however, few studies have focused on the role of membrane tension in LUAD. The present study aimed to construct a prognostic model associated with membrane-tension-related genes (MRGs) and explore its prognostic value in LUAD patients. METHODS: RNA sequencing data and the corresponding clinical characteristics data of LUAD were obtained from The Cancer Genome Atlas (TCGA) database. Five membrane-tension prognosis-related genes (5-MRG) were screened by univariate and multifactorial COX regression and least absolute shrinkage and selection operator (LASSO) regression analyses. The data were then divided into testing, training, and all groups to build a prognostic model, and Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), copy number variations (CNV), tumor mutation burden (TMB), and tumor microenvironment (TME) analyses were performed to explore the potential mechanisms of MRGs. Finally, single-cell data from the GSE200972 dataset in the Gene Expression Omnibus (GEO) database were obtained to determine the distribution of prognostic MRGs. RESULTS: Construction and validation of the prognostic risk models were conducted using 5-MRG in the trial, test, and all data sets. Patients in the low-risk group had a better prognosis than those in the high-risk group, and the Kaplan-Meier survival curve and receiver operating characteristic curve (ROC) confirmed that the model had a better predictive value for LUAD patients. GO and KEGG analyses of differential genes in the high- and low-risk groups were significantly enriched in immune-related pathways. Immune checkpoint (ICP) differential genes differed significantly in the high- and low-risk groups. By analyzing the single-cell sequencing data, the cells were divided into nine subpopulations and cell subpopulation localization through 5-MRG. CONCLUSIONS: The results of this study suggest that a prognostic model based on prognosis-associated MRGs can be used to predict the prognosis of LUAD patients. Therefore, prognosis-related MRGs could be potential prognostic biomarkers and therapeutic targets.