Immune-related prognostic genes signatures in the tumor microenvironment of sarcoma

Abstract

Sarcomas are a heterogeneous group of malignant mesenchymal neoplasms. This study aimed to investigate the immune-related prognostic gene signatures in the tumor microenvironment of sarcoma. The RNA-sequencing data and clinical phenotype data of 260 sarcoma samples and two normal samples were downloaded from The Cancer Genome Atla (TCGA) database. Tumor purity and immune cells infiltration were evaluated by Estimation of Stromal and Immune cells in Malignant Tumors using Expression data (ESTIMATE) deconvolution algorithm. Differentially expressed genes (DEGs) were screened in high vs. low immune score groups. Survival analysis was performed using Kaplan-Meier curve with log-rank test. Tumor infiltrating of immune cells was analyzed by Tumor Immune Estimation Resource (TIMER). High immune score and ESTIMATE score were associated with favorable prognosis. A total of 623 immune DEGs were screened. The majority of these genes (532 genes accounting for 85% of the DEGs) were up-regulated, and these genes were significantly enriched in various immune related biological processed and pathways, such as neutrophil activation, T cell activation, antigen processing and presentation. A total of 146 prognosis-related immune DEGs, and seven hub genes were identified, including B2M, HLA-DRB1, HLA-DRA, HLA-E, LCK, HLA-DPA1, and VAV1. Survival analysis showed that high expression of these genes was associated with a favorable prognosis. There were negative correlations between the expression of these hub genes and tumor purity, while positive correlations between expression of these hub genes and f infiltration levels of B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages and dendritic cells. These results help to stratify patients with different immune subtypes and help to design immunotherapy strategies for these patients in sarcoma.