Construction of a prognostic model for lung squamous cell carcinoma based on immune-related genes

Abstract

Abstract Lung squamous cell carcinoma (LUSC) lacks appropriate prognostic and diagnostic strategies. Available studies suggest the effectiveness of immunotherapy for LUSC, but effective molecular markers are still insufficient. We obtained mRNA expression and clinical information of LUSC samples from The Cancer Genome Atlas (TCGA) database. Enrichment levels of immune-related genes were revealed by single sample gene set enrichment analysis. Then, differentially expressed genes (DEGs) related to immunity were obtained by differential analysis. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed. In addition, Cox regression analysis combined with LASSO method was utilized to identify immune-related prognostic genes, and an immune-related prognostic model was constructed. Kaplan–Meier and receiver operating characteristic (ROC) curves were drawn to verify the accuracy of the model. Finally, a nomogram and calibration curve were drawn to predict LUSC patients’ survival. Samples were assigned into high-, medium- and low-immune groups. Compared with low- and medium-immune groups, high-immune group enriched more immune cells, with higher immune infiltration degree, and higher expression of immune checkpoints and human leukocyte antigen. DEGs were enriched in biological processes and signaling pathways related to immunity. Eleven genes (ONECUT3, MAGED4, SULT2A1, HPR, S100A5, IRS4, DPP6, FGF8, TEX38, PLAAT1 and CLEC3A) were obtained to construct an immune-related prognostic model. Riskscore served as an independent prognostic factor. Besides, the nomogram prediction model could predict disease progression in LUSC patients. The constructed risk assessment model for LUSC immune-related genes could assess LUSC patients’ prognoses with great efficacy, providing guidance for the clinical treatment of LUSC.