Abstract
Background: Osteosarcoma (OS) is a highly aggressive bone malignancy with limited treatment options and poor prognosis. This exploratory study aimed to identify molecular subtypes of early-stage, treatment-naive OS to guide precise therapeutic strategies. Methods: We analyzed RNA-seq data obtained from tumor tissues from 102 OS patients using a non-negative matrix factorization algorithm (NMF) to classify the tumors into three subtypes: S1, S2, and S3. Differential gene expression was evaluated using DESeq2, followed by functional enrichment analysis with clusterProfiler and CancerHallmarks. The tumor microenvironment was assessed through ESTIMATE and CIBERSORT, and drug sensitivity was predicted using OncoPredict. SAOS-2 and MG63 cells, representing the S1 subtype, were used in the viability essays to determine the effect of hesperidin, a natural phenolic compound noted for its anti-cancer potential, alone and in combination with doxorubicin and 5-fluorouracil. Results: This study revealed three OS subtypes: S1 was enriched in cell cycle regulation, vesicular transport, and RNA metabolism while S2 and S3 were enriched in pathways related to extracellular matrix organization and protein translation, respectively. S1 displayed high tumor purity, significant chemoresistance, and overexpression of KIF20 A, correlating with poor prognosis. AURKB, a hesperidin target, was implicated in S1 pathogenesis. In vitro, hesperidin significantly reduced the viability of SAOS-2 and MG63 cells and enhanced doxorubicin efficacy. Conclusions: Our findings support the molecular subclassification of OS, emphasizing subtype-specific mechanisms of tumor progression and chemoresistance, with hesperidin offering potential as a therapeutic adjunct for high-risk OS patients.