Construction of an immune prediction model for osteosarcoma based on coagulation-related genes

Abstract

Abstract Objectives The prognostic outcome of osteosarcoma, as the most common primary malignancy in children and adolescents, has not improved better with the development of modern medical care, and the aim of this study was to investigate the role of the coagulation system in the diagnosis and development of osteosarcoma. Methods TRGET and GEO databases were used to acquire clinical information and matching RNA data from osteosarcoma patients. To find novel molecular groupings based on coagulation systems, shared clustering was used. TIMER, SSGSEA, CIBERSORT, QUANTISEQ, XCELL, EPIC, and MCPCOUNTER analyses were used to identify the immunological status of the identified subgroups and tumor immune microenvironment (TIME). To understand the underlying processes, functional studies such as GO, KEGG, and protein-protein interaction (PPI) network analysis were used. Prognostic risk models were built using the LASSO technique and multivariate Cox regression analysis. Results The survival rates of the two molecular groupings were considerably different. large immunological scores, poor tumor purity, a large number of immune infiltrating cells, and a reasonably good immune status were all related with a better prognosis. According to GO and KEGG analyses, DEGs between the two groupings were primarily enriched in immunological and extracellular matrix-related pathways. Risk models based on coagulation system-related genes (CRGs) show promise in predicting osteosarcoma survival. A nomogram that combines risk models and clinical data may reliably predict the prognosis of individuals with osteosarcoma. Conclusion In patients with osteosarcoma, the expression of genes associated to the coagulation system is strongly related to the immunological milieu and can be utilized to correctly predict the prognosis of osteosarcoma.