Characterization of osteosarcoma subtypes mediated by macrophage-related genes and creation and validation of a risk score system to quantitatively assess the prognosis of osteosarcoma and reflect the tumor microenvironment

BACKGROUND: Macrophages are the main immune components in the microenvironment of osteosarcoma. The treatment strategy centered on macrophages has become a hot topic to improve cancer treatment. However, the research on the role of macrophages in the treatment of osteosarcoma is still in its infancy. METHODS: The data of osteosarcoma samples were downloaded from the Therapeutically Applicable Research to Generate Effective Treatments (TARGET) and GSE21257 datasets, and the macrophage enrichment fraction of osteosarcoma samples in TARGET was calculated by single-sample gene set enrichment analysis (ssGSEA) method to screen macrophage-related genes for consensus clustering. Differential expression analysis, univariable Cox, and least absolute shrinkage and selection operator (LASSO) regression were conducted to select reliable predictors and create a risk score system. The GSE21257 dataset was used as a verification set to verify the accuracy of risk score system. RESULTS: We identified 2 osteosarcoma clusters mediated by 22 macrophage score-related genes, namely cluster 1 (C1) and cluster 2 (C2). Compared with C2, C1 had a significant advantage in prognosis, and the degree of immune cell infiltration in tumor microenvironment (TME) was significantly higher, the expression of immune checkpoint molecules was significantly enhanced, and the Tumor Immune Dysfunction and Exclusion (TIDE) score was also significantly down-regulated. A robust risk score system was presented and validated, which demonstrated accuracy and independence in assessing the risk of death of osteosarcoma. The risk score system could also monitor TME infiltration in osteosarcoma samples and showed a close relationship with osteosarcoma biology, including metastasis and immunity. CONCLUSIONS: We identified 2 types of clusters mediated by macrophage-related genes and helped to analyze the cluster suitable for immunotherapy. A new prognostic risk score system was created to quantitatively evaluate the prognosis and TME of osteosarcoma, and to provide a new entry point for the design of personalized treatment.