Five glycosylation-related gene signatures predict the prognostic risks of lung adenocarcinoma

Abstract

This study aimed to identify glycosylation-related genes associated with lung adenocarcinoma (LUAD) prognosis through comprehensive bioinformatic analysis. Glycosylation-related genes were identified from the Human Gene Nomenclature Committee, and LUAD prognostic genes were screened from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO)-GSE68465 datasets. Glycosylation risk score (GLRS) was calculated to predict LUAD prognostic risk. Samples were grouped into GLRS-high and GLRS-low and compared. The Tumor Immune Dysfunction and Exclusion (TIDE) score was computed to assess the antitumor immune escape possibility after immunotherapy. From 213 glycosylation-related genes, five gene signatures served as prognostic LUAD predictors using univariate and stepwise Cox regression analyses. GLRS-based models were constructed using TCGA and GSE68465 samples; their sensitivity and specificity in predicting LUAD prognosis were confirmed. GLRS was an independent LUAD prognostic factor and contributed to the nomogram to predict patient survival. High GLRS was associated with advanced tumor stage and higher mutation frequencies, estimate scores, and TIDE scores. GLRS-high and GLRS-low patients differed in immune cell infiltration and epithelial-mesenchymal transition (EMT)-related gene expression. Thus, we propose five glycosylation-related gene signatures to predict overall survival and prognostic risks of LUAD. Their regulatory roles may be related to immune invasion, immunotherapy response, mutation, and EMT.