Atrt-01. Modelling Atypical Teratoid/ Rhabdoid Tumour and Healthy Brain Crosstalk

Abstract BACKGROUND Atypical Teratoid/Rhabdoid Tumours (ATRT) are aggressive paediatric brain tumours which mostly occur in the first three years of life and which rapidly recur after surgical resection. There has been debate over the use of cytotoxic consolidation therapy due to the severity of off target effects associated with chemo/radiotherapy use in young patients. Astrocytes have previously been shown to support progression and therapy resistance of other brain tumours; however, their role in ATRT pathophysiology is yet to be elucidated. METHODS To better understand how adjacent “healthy” brain tissue may be complicit in ATRT recurrence, we have developed a 3D in vitro model in which ATRT cells are co-cultured with astrocytes within a bespoke decellularized human cerebellum extracellular matrix (ECM) hydrogel. RESULTS Immunofluorescence confirmed that human cerebellar ECM hydrogels retain human brain ECM components (hyaluronic acid, laminin, and fibronectin). ATRT-astrocyte co-culture spheroids were formed within the ECM hydrogel and observed to display a distinctive morphology relative to monoculture controls, with no difference in overall viability. Co-cultured tumour cells were successfully harvested from the model and isolated from astrocyte populations using fluorescence activated cell sorting. In both 2D and 3D culture, we have shown that astrocytes labelled with CM-DiI transfer this cell tracker dye to ATRT-MYC (BT12, BT16) and ATRT-TYR (BT37) cells, potentially indicating intercellular trafficking. Furthermore, in 2D culture ATRT-MYC (BT12 and BT16) cells form actin rich cell-cell processes with astrocytes when co-cultured in vitro. CONCLUSIONS This work represents a comprehensively characterised model of ATRT interaction with the healthy brain. Future work will analyse changes in ATRT cell transcriptomes when astrocytes are present in the model. By understanding how the presence of astrocytes in the tumour microenvironment may promote tumour survival and progression, we aim to use this data to identify candidate therapeutics to inhibit crosstalk and improve patient outcomes.