(P 282) AN EMBRYONIC ZEBRAFISH XENOGRAFT MODEL OF RHABDOMYOSARCOMA REVEALS GENOTYPE DEPENDENT TUMOUR VASCULARISATION DYNAMICS AND RESPONSE TO ANTI-ANGIOGENIC RECEPTOR TYROSINE KINASE INHIBITORS

Objective: Rhabdomyosarcoma (RMS) is a highly aggressive soft tissue sarcoma, predominantly affecting children and adolescents. Treatment includes intensive chemotherapy, surgery and radiotherapy and outcome has remained largely unchanged over the last decades with 5-year survival still < 30% for some cohorts of patients. Better therapeutic regimens including novel agents are urgently needed.
Multi receptor tyrosine kinase inhibitors (MRTKIs) are an important class of agents, many of which are licensed for adult indications and explored in the clinic in sarcoma patients including children and young adults. The MRTKI Regorafenib is currently assessed in the relapse setting in patients with RMS in the Frontline and Relapsed RhabdoMyoSarcoma (FaR-RMS) trial (NCT04625907). Strikingly, biomarkers of response for these agents have yet to be identified.
MRTKIs act against not only the cancer cell, but also the supporting stroma, particularly the vasculature. This interaction is poorly modelled in standard cell culture and in vivo mouse xenograft studies. The early embryonic zebrafish is translucent and provides an opportunity to assess this interaction with high throughput in vivo imaging.
We have, therefore, developed a novel embryonic zebrafish xenograft model of RMS where we can directly visualise tumour and vasculature development and assess the effects of MRTKIs.

Methods: FLK-1-GFP (blood vessel reporter) transgenic zebrafish embryos were inoculated in the yolk with fluorescently labelled cells from 8 different RMS cell lines (fusion-positive (FP): Rh4, Rh30, Rh41, RMS-01, fusion-negative (FN): RD, JR1, CT10) and a healthy control line (HFF1)), at 2-day post-fertilization and incubated at 34C for up to 3 days. Tumour and vessel development were assessed using a Cytation-5 microscope. Images were analysed using custom FIJI pipelines to generate multifaceted metrics of tumour development and vascularisation. Tumour growth assessments were validated through cell counting. Three clinically relevant MRTKIs (Regorafenib, Infigratinib and Axitinib) with different target receptor profiles were used in this study. To identify the appropriate treatment dose, non-tumour bearing embryo zebrafish were exposed to a range of drug doses (0-10µM) to identify the highest dose where no overt side effects on embryo development are observed.

Results: All 7 RMS cell lines engrafted in the embryo zebrafish with a tumour take rate of >90%, compared with 0% for HFF1 control cells. Tumour burden assessment by fluorescent imaging indicated growth during the observation period for all models. Direct cell counts, performed for RD and Rh4, correlated with imaging data (r=0.85) and confirmed high viability (RD = 86.7%; Rh4 = 84.1%). Inoculated RMS tumours induced both, neo-vascularisation towards the tumour (60%-98% of cases) and increased density of proximal vessel beds. Correlation analysis revealed RMS subtype specific differences in vessel induction, with neo-vascularisation correlating with tumour burden in FN lines (r=0.58-0.8) and proximal vessel bed complexity correlating with tumour burden in FP lines (r=0.43-0.61). Tumour and associated vessel treatment responses were assessed for the MRTKis Regorafenib, Infigratinib and Axitinib at the maximum tolerated doses of 0.1 µM, 0.1 µM and 0.05 µM, respectively. This revealed significant heterogeneity in treatment response, but also primary site of drug action. This was most apparent between FN RD and FP Rh4 cells in response to Infigratinib, where the drug achieved an 80% reduction in Rh4 tumour burden compared to 35% in RD, but only a 65% reduction in neo-vessel induction compared to 90% in RD. The observation of inter tumour cell differences in tumour growth versus vascularisation inhibition suggests additional value of our model for response prediction to MRTKIs.

Conclusion: We have developed a novel RMS xenograft model, utilising the unique imaging properties of the embryonic zebrafish to profile the initial tumour-stroma interactions in developing RMS tumours. This model allows assessment of tumour growth, vascularisation initiation and therapeutic responses to clinically relevant MRTKIs. Significant heterogeneity was observed and correlative analysis with the presence of autocrine signalling factors and expression of receptor tyrosine kinases on RMS cells is ongoing. Future work will include the assessment of patient-derived tumour cells (ideally from ongoing clinical trials), alongside comparative analysis with standard preclinical models to define the value of this model in preclinical drug development and identify biomarkers of response to MRTKIs.