(P 253) THE IMPACT OF CELL FUSION IN LEIOMYOSARCOMA ONCOGENESIS

Objective: Leiomyosarcoma (LMS) is one of the most common and aggressive type of soft tissue sarcomas. Around 50% of the patients develop metastasis translating in an overall survival rate of 12 months after the dissemination of the disease. LMS are very heterogeneous and they show highly rearranged genome with no characteristic gains or losses, the most common genetic alterations are the ones found in the majority of cancers (RB1 and TP53). We propose that cell fusion is one of the potential mechanisms contributing to tumor oncogenesis, heterogeneity and chaotic genome rearrangements. It has been recently published that there are two types of LMS, one where there is overexpression of smooth muscle related genes and another one associated to fibroblast, adipocytes and mesenchymal stem cells. The objective of this study is twofold: First, study the initiation of LMS; the genome alterations produced by the process of cell fusion compared to the ones found in patients, and potentially identify the initiating cells involved in the oncogenesis of LMS. Second, study of the progression of LMS; we have developed a clinical trial where we look for the presence of hybrid cells in patients’ peripheral blood and tumor samples to correlate the presence or absence of circulating hybrid cells (CHC) with the clinical state of the patient and establish a possible prognostic biomarker.Methods: Initiation: In this study we established four different spontaneous fusion models: 1&2) SMC + fibroblast and SMC + adipose stromal cells, for the LMS associated to fibroblast, adipocytes, and mesenchymal stem cells; 3) SMC + SMC, for the LMS expressing smooth muscle genes, and 4) between SMC + macrophages to explore the reasoning behind the high index of metastasis that we find in LMS. Progression: we analyzed TMA from LMS patients’ tumors post-surgery to search for double positive (CD45+ TAGLN+) cells and we collected blood samples (10ml) from LMS patients before and after treatment. The visualization of circulating hybrid cells (CHC) was done by immunofluorescence staining on white blood cells and confocal imaging.Results: Initiation: From the fusion model of SMC + fibroblast, we developed six tetraploid hybrid cell lines and performed comparative genome hybridization (CGH) analysis. We observed a highly rearranged genome and in one case a very specific PRKG1 deletion which is involved in smooth muscle contractile function. Based on this observation we looked for PRKG1 deletion in a cohort of 123 LMS tumors with whole genome sequencing. We observed 96 LMS tumors with at least one copy of PRKG1 deleted (78% including whole chromosome 10 loss + focal PRKG1 deletions). PRKG1 targeted deletions have been validated by PCR and Sanger sequencing and characterization is ongoing. From the fusion model SMC + SMC after the CGH analysis we observed many copy number variations. The most recurrent deletions were located in: chromosome 13 deleted in 5/9 total hybrid cell lines, chromosome 17 deleted in 8/9. Chromosomes 13 and 17 are of special interest to us because it is where RB1 and TP53 are located, two of the most frequently altered genes in LMS. Progression: Blood samples from 30 LMS patients were analyzed by immunofluorescence (IF) where we look for the presence of double positive cells CD45+ (monocyte/macrophage) and TAGLN+ (smooth muscle, antibody currently used in the diagnosis of LMS). 26/30 (86%) of the patients showed metastasis dissemination and 32% of the patients had presence of double positive cells (circulating hybrid cells). 59 LMS tumor TMA slides were also analyzed by IF where we observe presence of double positive cells in 9/59 (15.25%) of the cases.Conclusion: With our cell fusion models we are able to produce models with very similar chromosomic profiles than the ones we find in LMS patients, indicating that cell fusion could be one of the mechanisms involved in oncogenesis and giving our cellular models an advantage in the field for future research. Furthermore, we have observed possible hybrid cells or double positive cells (CD45+ TAGLN+) in both patients’ tumor and peripheral blood, giving the observation of these cells a potential interest in the development of a prognostic biomarker and cellular therapeutical targeting.