(P 418) GENOMIC AND IMMUNOLOGICAL PROFILING OF PRIMARY AND DISTANT METASTASES OF SOFT TISSUE SARCOMAS

Objective: Soft tissue sarcomas (STS) are a diverse group of rare tumors that exhibit metastasis in approximately 50% of patients. Unfortunately, the prognosis for metastatic STS is typically fatal, with an average survival of only 20 months. Throughout the metastatic process, tumor cells undergo mutational changes and encounter selective treatment pressures, resulting in altered molecular characteristics. However, the genetic and immunological profiles of these challenging tumors and sarcoma metastases remain poorly understood.

Methods: We performed high throughput analysis of a cohort of 70 patients with soft tissue sarcoma, each with paired primary tumors and matched metastases. The histological subtypes included in the study are undifferentiated pleomorphic sarcoma (UPS) [n = 20], leiomyosarcoma (LMS) [n = 17], liposarcoma (LPS) [n = 19], and other rare histologies [n = 14]. Our primary objective is to compare the transcriptomic profiles of primary and metastatic tumors by analyzing the RNAseq data from our patient cohort. Additionally, we employ spatial biology profiling using the GeoMx® Digital Spatial Profiler system (DSP) to perform quantitative, high-plex analysis of mRNA expression within spatially defined regions of the primary tumors and metastases. Through these analyses, we aim to identify differentially expressed genes (both upregulated and downregulated) between primary and metastatic tumors, which may shed light on candidate genes or pathways involved in the metastatic process based on co-expression transcriptional networks.

Results: Analysis of the UPS sub-cohort revealed an overexpression of inflammatory pathways and hypoxia in metastatic samples compared to paired primary (Figure 1C). Specifically, two genes, ADORA2B and SERPINE1, exhibited higher expression in metastases compared to primary tumors (Figure 1A and 1B). These genes potentially play a role in the metastatic process, immune infiltration, and angiogenesis. Additionally, we observed an overexpression of VEGF in the stromal regions of metastases (Figure 1B). Notably, these genes have been associated with worse overall survival in GEPIA database analyses (Figure 2A and 2B). Furthermore, our spatial transcriptomic profiling revealed an overexpression of TGFbeta in metastases and IDO1 in primary tumors, indicating immunological differences between the two tumor types (Figure 1B). Multiplex immunohistochemistry analyses allowed to confirm these findings.

Conclusion: This study represents the largest transcriptomic profiling of paired primary and metastatic STS tumors. Our findings provide valuable insights into genes potentially involved in the metastatic process in UPS, one of the most common and undifferentiated subtypes of STS. Results from the other cohorts and of functional experiments will be presented at the meeting.