Vessel state and immune infiltration of the angiogenesis subgroup and construction of a prediction model in osteosarcoma

Abstract

Angiogenesis has been recognized as a pivotal contributor to tumorigenesis and progression. However, the role of angiogenesis-related genes (ARGs) in vessel state, immune infiltration, and prognosis remains unknown in osteosarcoma (OS). Bulk RNA sequencing data of osteosarcoma patients were obtained from the Therapeutically Applicable Research to Generate Effective Treatments (TARGET) database, and patients were divided into two angiogenesis subgroups according to the expression of ARGs. We compared their vessel state and used two independent algorithms to evaluate the tumor microenvironment (TME) in the two subgroups. Furthermore, hub genes of differentially expressed genes (DEGs) in the two subgroups were selected to perform LASSO regression and multivariate Cox stepwise regression, and two prognostic hub genes were found. An ARG_score based on prognostic hub genes was calculated and proved to be reliable in the overall survival prediction in OS patients. Furthermore, the ARG_score was significantly associated with ARGs, immune infiltration, response to immunotherapy, and drug sensitivity. To make our prediction model perform well, clinical features were added and a highly accurate interactive nomogram was constructed. Immunohistochemistry and qRT-PCR were utilized to verify the expression of prognostic hub genes. GSE21257 from the Gene Expression Omnibus (GEO) database was used as a validation dataset to verify its robustness. In conclusion, our comprehensive analysis of angiogenesis subgroups in OS illustrated that angiogenesis may lead to different vessel states and further affect immune infiltration and prognosis of OS patients. Our findings may bring a novel perspective for the immunotherapy strategies for OS patients.