222P A hybrid text-knowledge graph retrieval-augmented generation system for clinical decision support in soft tissue sarcoma

conducted, and target genes were identified for further investigation.For the validation of the proteomic hits and translation into functional vulnerabilities, patient-derived synovial sarcoma cell models were established and characterized. Results:The proteomic analysis of the different stages showed in a GSEA that the primary tumors were enriched in respiratory electron transport and TCA cycle.For the relapse group there was an upregulation of keratan sulfate metabolism and downregulation of the translation and for the metastasis, the extracellular matrix was downregulated.These findings were supported by the WGCNA analysis where the most important modules for primary tumors were enriched in TCA cycle and in the metastasis negatively enriched for extracellular matrix.Based on the analysis of the differentially expressed individual proteins, potential drug candidates were selected which will be tested in the established patient-derived cell models to correlate the proteomic expression with functional vulnerabilities. Conclusions:The proteomic analysis of the longitudinal synovial sarcoma cohort highlighted several pathways for further investigation such as the respiratory electron transport or the extracellular matrix.How these proteomic differences translate into drug sensitivities will be evaluated with newly established patient-derived synovial sarcoma cell models.