(P 417) MULTI-OMIC CHARACTERISATION OF MOLECULAR HETEROGENEITY AND NEOADJUVANT THERAPY RESPONSE IN SYNOVIAL SARCOMAS

Objective: Surgery is the mainstay for curative treatment of localised synovial sarcomas (SS). However, approximately 50% of patients develop recurrent disease. Neoadjuvant therapy (NAT) can improve surgical outcomes and prevent recurrence, however the biological mechanisms underlying response and resistance to NAT remains unknown. Utilising paired pre- and post-treatment specimens, we conducted an integrated proteomic and transcriptomic analysis to 1. Assess the intrinsic molecular heterogeneity within SS which could explain the range of clinical responses seen in patients and 2. Determine the molecular changes associated with NAT.

Methods: This is a single institution retrospective cohort comprising of 34 patients: 16 received NAT (n=30 specimens) and 18 did not receive NAT (treatment naive) (n=32 specimens) as a control group (Fig 1A). NAT samples comprised of biopsy (pre-treatment) and surgical (post-treatment) specimens. There was no intervening treatment between the biopsy and surgical specimen in the treatment naïve group. Paired samples refer to those cases for which matched biopsy and its corresponding excision specimen were available for profiling.
Specimens were subjected to proteomics and transcriptomics analysis. Details of specimens analysed are shown in Fig 1A. Molecular subgroups within SS were identified through consensus clustering of transcriptomic data from baseline biopsy specimens (n=24). Biological features associated with consensus clusters were determined by overrepresentation analysis (ORA) and through transcriptomic immune cell deconvolution using Microenvironment Cell Populations (MCP)-counter.
To elucidate molecular changes induced by NAT, Mfuzz soft clustering was employed to compare proteomic and transcriptomic expression patterns in paired samples. This led to the identification of subsets of proteins and genes that are associated with treatment. Altered genes and proteins were subjected to ORA.

Results: Consensus clustering of the transcriptomic data of baseline samples identified two clusters of distinct molecular profiles (p < 0.05) (Fig 1B). ORA showed that Cluster 2 had an enrichment of interferon gamma signalling (Fig 1C) indicating a relatively “immune hot” molecular subgroup. This was further substantiated by immune deconvolution analysis where neutrophils, cytotoxic lymphocytes, B lineage and monocytic lineage cells showed higher levels in Cluster 2 vs Cluster 1.
For paired samples that received NAT, soft clustering of proteomics data identified 4 groups with distinct patterns (Groups A-D), each associated with proteins that changed in post-treatment compared to pre-treatment samples. As an exemplar, we highlight a subset of proteins in Group A (Fig 1D). These proteins (n=148 proteins) were downregulated in post-treatment vs pre-treatment specimens (Fig 1D left). While there was also a decrease in median expression of these proteins in some paired specimens from the treatment-naïve patients, only in the patients who received NAT was this difference statistically significant (p=0.0156 in NAT vs p=0.195 in treatment naïve). ORA showed an enrichment of ribosome and protein translation pathways in Group A, the proteosome complex in Group B and the complement cascade in Group D. A similar analysis in the transcriptomic data identified neuroactive ligand receptor interactions being enriched following NAT.

Conclusion: We show that SS is comprised of two intrinsic molecular subgroups, one which has elevated immune signalling. Moreover, we have shown that there is a consistent group of genes, proteins and biological pathways that are significantly altered in response to NAT but not in treatment naïve paired specimens. While our cohort is small and our findings hypothesis generating, this study provides important insights into the biological mechanisms associated with NAT which has potential for future applications as candidate prognostic biomarkers and identification of new drug targets for NAT.