M2 TAM-related risk signature based on single-cell and bulk RNA sequencing to evaluate tumor immune microenvironment and predict prognosis in glioma

Abstract

Abstract Background Glioma is the most common malignant brain tumor, and the treatment effect is still not satisfactory. The tumor microenvironment (TME) is crucial in the incidence and development of tumors. Previous research has shown that TAMs are a vital ingredient of the tumor microenvironment and relate to tumor development, but their roles in glioma remain a mystery. Results In this study, we combined the TCGA dataset and GEO single-cell dataset to obtain 58 M2 TAM-related genes. Use univariate Cox regression analysis and LASSO-Cox regression analysis to screen out ten prognostic-related genes and construct a prognostic signature. The CCGA dataset was used to validate the prognostic signature. TCGA and CCGA cohorts were divided into two groups based on the prognostic signature. The AUCs at 1, 3, and 5 years in the TCGA cohort were 0.81, 0.91 and0.90, and 0.67, 0.72 and 0.77 in the CCGA cohort, respectively. In addition, we developed a highly reliable nomogram based on prognostic signatures and clinical characteristics. According to Functional Enrichment Analysis, the differential genes between the two groups were enriched in immune-related pathways. The tumor immune microenvironment showed immune cell infiltration and immune responses are more pronounced in the high-risk group. Drug sensitivity prediction identifies twelve drugs sensitive to high-risk groups, with Bortezomib having the lowest IC50. Conclusion In summary, M2 TAM-related risk signature will assist clinical prognosis prediction and personalized treatment of glioma patients.