(P 89) TARGETED THERAPY OF MET, TYROSINE KINASE RECEPTOR, IN PEDIATRIC HIGH-GRADE OSTEOSARCOMAS : A NEW WAY TO TREAT PATIENTS ?

Objective: Osteosarcoma (OS) is the most frequent cause of bone cancers in adolescents and young adults with a strong tendency to metastasize. The overall survival at 5 years remained stable since decades around 70% and is the witness of no major improvement in its therapeutic management limiting the prognosis of the OS patients. The genomic, epigenomic, transcriptomic and proteomic analyses have demonstrated the molecular complexity of this sarcoma, but have not yet identified promising surrogate and/or theranostic biomarker(s) amenable to new therapies. By the past, our team identified in French OS94 protocol cohort a three-gene signature involving the deregulation of osteoblastic dedifferentiation signaling, comprising notably MET (tyrosine kinase receptor for Hepatocyte Growth Factor (HGF)) gene deregulation. This membrane receptor is promoting OS proliferation, but also OS invasive phenotype and tumorigenicity. It was in multiple publications linked significantly to metastatic onset and/or poor survivals. Based on this background and on the recent development of 2D patient-derived cell cultures and 3D-tumoroid derivations, the primary objective of this study was to test several types of MET inhibitors going from ultra-selected ones (e.g., onartuzumab and savolitinib) to panTKR inhibitor (e.g., cabozantinib) and be able to select the top one inhibitor to understand its genomic and metabolomics impacts.

Methods: To investigate this objective, we used already developed 2D and 3D simple and complex patient-derived models mixing OS cells and macrophages. Based, first, on a confluence (IncuCyte® technology) and tumoroid size/invasion assessments, we studied and selected using this panel of TKR inhibitors (TKRi) the best candidate for a prolonged and definitive inhibition. To further and accurately understand the TKRi effect on OS cells, RNAsequencing analyses, as well as metabolomics approaches (HR-MAS metabolite assessment and SeaHorse (Cell Mito Stress test) were proposed to investigate the influence of the targeted treatment on cell-cell interactions and adaptation.

Results: Using this selection of TKRis, we were able to demonstrate that only savolitinib and cabozantinib had an anti-proliferative effect and this impact was reduced to the models amplifying the expression of MET gene. To check the influence on invasion criteria, the use of tumoroid models was showing a potential antimigratory effect observed mostly on the number of evading cells and the distance covered by the cells from the spheroid that was significantly higher with savolitinib. Our RNAsequencing data are supporting the fact that cell-cell interactions are particularly inhibited by savolitinib in presence of macrophages, as well as glucose rewiring metabolism. The metabolic adaptation promoting the maintenance of glutamine and glucose dependencies after treatment was confirmed by metabolomics results, affording a sustainable demonstration of less adaptive OS cells to its microenvironment after MET inhibition.

Conclusion: We demonstrated mostly with a selective-MET gene inhibitor a significant antiproliferative effect and a stop of invasive OS cell capacities based on cell-cell interaction pathways associated to a metabolic maintenance of nutrient dependencies.