BST2 is an immune-related prognostic biomarker for glioma

Abstract

Abstract Background Glioma is the most common brain cancer. Research has indicated that the tumor microenvironment (TME) plays an important role in the proliferation, invasion, and treatment response of tumors. However, the role of TME in glioma remains unclear. Here, we try to identify a TME-related gene in glioma that can predict prognosis. Methods The transcriptome data and corresponding clinical data of 701 glioma samples were downloaded from the TCGA database. The transcriptome data of normal samples were obtained from the Genotype-Tissue Expression (GTEx) database. R software and R packages were used to perform statistical analysis, calculation of TME scores, survival analysis, Cox regression analysis, and functional enrichment analyses. GSEA software was used to perform Gene Set Enrichment Analysis (GSEA). Protein-protein interaction (PPI) network analysis was performed using Cytoscape software. The protein expression level of BST2 was detected by immunohistochemistry. Results By gene differential expression analysis based on TME scores, we obtained two sets of differentially expressed genes (DEGs), respectively, followed by intersection analysis, we obtained shared DEGs. By univariate Cox regression analysis and PPI network analysis of the shared DEGs, we obtained a prognostic gene set and hub gene set, respectively, followed by intersection analysis, we obtained three hub genes associated with prognosis, including BST2, CCL2, and RSAD2. Subsequent analyses were focused on BST2. Compared with normal samples, BST2 expression was higher in glioma samples. Moreover, BST2 expression was positively correlated with pathological grades and was negatively correlated with overall survival time (OS). Consistently, a validation cohort of 42 glioma patients further verified the upregulation of BST2 and its influence on prognosis. Mechanistically, the result of GSEA indicated that BST2 might be involved in regulating tumor immunity. By differential analysis of tumor-infiltrating immune cells (TICs) contents between high- and low-BST2 expression groups, and correlation analyses between the expression level of BST2 and TICs contents, we obtained 7 key TICs, and 5 of them were significantly associated with OS, especially regulatory T cells and M2 macrophages. Conclusions The TME-related gene BST2 could promote the progression of glioma via regulating the composition of TICs, and thus might be a novel prognostic biomarker.