(P 04) THE CELL ATLAS OF CLEAR CELL SARCOMA

Objective: We hypothesize that combined sequencing of single nuclei extracted from a well characterised cohort of clear cell sarcomas combined with bulk DNA, RNA and methylation analysis will help identify the normal cell components along with subpopulations of tumour cells. This information could open the opportunity to identification of therapeutic vulnerabilities.

Methods: Fresh frozen samples of clear cell sarcoma were sourced from the Royal National Orthopaedic Hospital NHS trust, London, UK and processed for multi-omic analysis. Nuclei were extracted from frozen tissue using a Tris Tween based dissociation and run through the 3 prime 10X genomics RNA sequencing pipeline. 5 primary cases and 5 metastatic cases were sequenced, with 2 cases having matched primary and metastatic regions. Using the R package Seurat, data was quality controlled, clustered, and annotated based on their RNA signatures. These clusters were then compared with other sequencing and histological data to produce a robust set of identities for the cells annotated in the data. Data for other tumour cell types were accessed via the GEO repository and a cohort of normal cell types arising from the tendon were made available by the Human Tissue atlas group based in Oxford, UK. This research was supported by the generous philanthropic gift to UCL by the Edward Showler Foundation. https://www.edwardshowlerfoundation.com/

Results: Cell annotations confirm the neural crest and melanocytic program that is active in the tumour cells and associates with expression of SOX10, MITF and PMEL. Clustering revealed a minimum of three distinct populations of tumour cells in each sample which are underpinned by proliferation and cell differentiation trajectories. These correlate closely with the clonal and sub-clonal copy number gains seen in the whole genome sequencing. The immune contingent of the tumour is very low and predominantly comprised of macrophages with small clusters of T-cells.

Conclusion: Initial findings from our data look promising and have revealed novel insights into genomic and transcriptional heterogeneity in CCS. Analysis of longitudinal data and comprehensive characterisation of the differences between primary and metastatic tumours could boost our understanding of the evolution patterns in this cancer. The immune-excluded nature of the tumour suggests that conventional immune-checkpoint therapy on its own may not be a rational approach and that combination therapies or other forms of immune therapies such as CAR-T cell therapy may be worth pursuing.