Characterizing the critical role of metabolism in osteosarcoma based on establishing novel molecular subtypes.

Abstract

Osteosarcoma is the third most frequently diagnosed cancer among adolescents. Immunotherapy is an effective curative treatment for metastatic osteosarcoma patients. This study aimed to further reveal the significance of metabolism in tumor progression, and to categorize molecular subtypes for guiding personalized therapy. Univariate Cox regression analysis was performed to screen metabolism-related genes associated with osteosarcoma prognosis. A molecular subtyping system was developed by unsupervised consensus clustering. Survival analysis and functional analysis were used to evaluate the performance of subtyping and characterize the TME of subtypes. Stepwise Akaike information criterion (stepAIC) was employed to optimize the prognostic model. C1 and C2 subtypes showed distinct prognosis, with more favorable survival in C2 subtype. C2 subtype presented a higher immune infiltration and active anti-tumor response. Notably, C2 subtype was predicted to have better immune response to immune checkpoint blockade. In addition, a 5-gene prognostic signature with robust ability to classify patients into high-risk and low-risk groups was developed. The study revealed the critical role of metabolism in tumorigenesis by comparing the features between the two subtypes. Oncogenic pathways including epithelial mesenchymal transition (EMT), glycolysis and hypoxia may be closely involved in the correlation with metabolism. Importantly, we developed a novel subtyping system and a 5-gene signature with high potential to be applied in clinical practice.