Identification of a Gene Signature Closely Related to Immunosuppressive Tumor Microenvironment Predicting Prognosis of Patients in EGFR Mutant Lung Adenocarcinoma

Abstract Background: Lung adenocarcinomas (LUAD) harboring epidermal growth factor receptor (EGFR) mutations generally are unable to benefit from immune checkpoint inhibitors (ICIs), due to an immunosuppressive tumor microenvironment (TME) and a lower tumor mutation burden (TMB). Currently, there has been no gene signature that can comprehensively evaluate the TME and predict the prognosis of EGFR mutant LUAD patients. Methods: Using the cancer genome atlas (TCGA) database of EGFR mutant LUAD based on the immune score derived from the ESTIMATE algorithm, we screened the differential immune-related genes with prognostic value and compared the TMB profiles. Gene ontology (GO) and Kyoto encyclopedia of gene and genomic (KEGG) enrichment analysis were used to analyze the potential functions. The least absolute shrinkage and selectionator operator (LASSO) cox regression model was applied to identify a gene signature and constructed risk model. Kaplan-Meier survival and receiver operating characteristic (ROC) analysis were used to evaluate the prognostic value of the gene signature. CIBERSORT was used to evaluate the abundance of immune cells infiltration.Results: We screened 396 the differential immune-related genes based on immune score, whose potential functions were significantly related to T cell differentiation, immune response, cell cycle and cell proliferation. The disparities of TMB profile could be found between the high and low immune score group. Then, we identified a three-gene signature, including B and T lymphocyte attenuator (BTLA), BUB1 mitotic checkpoint serine/threonine kinase B (BUB1B) and centromere protein E (CENPE). The three-gene signature could well identify at-risk patients of EGFR-mutant LUAD patients in the training and validating set, and the high-risk patients were related to shorter overall survival (OS) (p=0.0053 and p=0.035). The immune activity of B cells and macrophages were higher in the low-risk group, in contrast the immune activity of Natural Killer (NK) cells and T cells were higher in the high-risk group. Conclusions: The three-gene signature closely related to immunosuppressive TME could predict risk prognosis of patients in EGFR mutant LUAD.