An analysis of prognostic risk and immunotherapy response of glioblastoma patients based on single-cell landscape and nitrogen metabolism

Abstract

Glioblastoma (GBM) is a highly invasive brain tumor of astrocytic origin. Nitrogen metabolism plays an instrumental role in the growth and progression of various tumors, including GBM. This study intended to mine nitrogen metabolism-related biomarkers for GBM-related research of prognosis and immunotherapy. Through single-cell data analysis of GBM, we identified four cell types (Astrocytes, Macrophages, Fibroblasts, and Endothelial cells). We calculated the nitrogen metabolism scores and conducted trajectory analysis for the most abundant cells, Astrocytes, revealing 6 differentiation directions of Astrocytes, which included the main differentiation direction from cells with low nitrogen metabolism scores to cells with high nitrogen metabolism scores. Furthermore, based on the differentially expressed genes (DEGs) with high/low nitrogen metabolism scores, we constructed a 7-gene prognostic model by utilizing regression analysis. qRT-PCR analysis showed that IGFBP2, CHPF, CTSZ, UPP1, TCF12, ZBTB20 and RBP1 were all significantly up-regulated in the GBM cells. Through differential analysis, a protein-protein interaction (PPI) network, and enrichment analyses, we identified and analyzed the DEGs in the high RiskScore subgroup, revealing complex interactions among DEGs, which were mainly related to pathways such as TNF signaling pathway and NF-κB signaling pathway. By leveraging univariate analysis, survival-related genes were selected from the nitrogen metabolism-related gene sets. Clustering, survival, immune, and mutation analyses manifested that the collected nitrogen metabolism-related genes had good classification performance, presenting notable differences in survival rates, immune levels, gene mutations, and sensitivity to drugs between cluster1 and cluster2. In conclusion, the project investigated the prognosis and classification value of nitrogen metabolism-related genes in GBM from multiple perspectives, predicting the sensitivity of different subtypes of patients to immunotherapy response and drug sensitivity. These findings are expected to show new research directions for further exploration in these fields.