(P 334) EXPANDING THE LANDSCAPE OF PRECLINICAL MODELS OF SYNOVIAL SARCOMA: FROM 2-DIMENSIONS TO 3-DIMENSIONS TO MICE

Objective: The prognosis of patients with metastatic synovial sarcoma (SynSa) remains very poor, with 5-year overall survival rates around 10%. It is crucial to develop more effective and subtype-specific treatment options for this malignancy. Preclinical models of SynSa are important for the exploration of novel therapies in preparation of clinical trials. The currently available models of SynSa are scarce and many of them are poorly characterized. We aimed at the creation of new and well-characterized preclinical models of SynSa, including primary cell cultures, tumoroids and patient-derived xenografts (PDX).

Methods: Fresh SynSa tissue from consenting patients was implanted subcutaneously into athymic nu/nu NMRI mice to create PDX models. Tumoroids and primary cell cultures were created from donor tissue and from PDX tissue after enzymatic and mechanical dissociation of the biological material, with exhaustive attempts to optimize media and 3D-matrix conditions to grow tumoroids. The models were characterised using immunohistochemical stainings and molecular techniques, such as fluorescence in situ hybridisation and reverse transcription polymerase chain reaction to prove the presence of the characteristic SSX::SS18 fusion. The models were considered established if they grew for > 2 passages and showed stable morphologic and molecular features as compared to the samples of origin.

Results: Between 2011 and 2024 we created 6 established SynSa PDX models by xenografing tissue from 34 donors. From 2021 onwards, two patient-derived primary cell cultures and one PDX-derived primary cell culture were successfully established. Additionally, two patient- and three PDX-derived tumoroids were created. These models preserved the essential characteristics of the tumor of origin. All models were profiled using low-coverage whole genome sequencing, and all PDX models were subject to RNA sequencing. With the exception of one patient-derived primary cell culture, all established models originated from donors that did not receive any systemic therapy before the moment of sampling.

Conclusion: We expanded the number of available preclinical models of SynSa to be used for tumor biology and drug testing experiments. Optimizing growth conditions for mesenchymal tumoroids ex vivo remains a major challenge.