Patient-derived osteosarcoma tumoroid model reveals functional phenotypic diversity with implications for drug responses

Abstract Osteosarcoma (OS) is the most common primary bone malignancy affecting children and adolescents, for which survival has not improved in more than four decades. The lack of accurate OS preclinical models hinders the understanding of tumor heterogeneity and its interaction with the surrounding extracellular matrix (ECM), limiting the discovery of predictive biomarkers and the development of effective therapies. Four 2D preclinical models from OS patients were established and characterized for their functional differences in comparison to OS cell lines for their growth, cellular phenotypic attributes, osteogenic differentiation capabilities and metabolic responses to growth factors and BMP-2. Molecular and cellular profiling revealed intra-tumoral heterogeneities that were very distinct from the endorsed OS cell lines. The OS patient-derived (PD) cells also displayed differential sensitivity to Doxorubicin and Cisplatin and resistance against Methotrexate. Subsequently, the 3D PDTs (Patient- derived tumoroids) models were developed by self-aggregating spheroids with and without Matrigel® ECM matrix. These PDTs models were screened for selective ECM and bone-specific gene markers, revealing dynamic differences between 2D and 3D models with and without ECM, with heightened dysregulation observed in 3D systems. The drug response variances observed among 2D OS cells and 3D tumoroids model within Matrigel® highlights the need for optimized platforms for in vitro personalized drug screening. Thus, our findings support the screening of preclinical PD OS models for phenotypic profiling and elucidating ECM contributions to drug responses and pathophysiology.