O-188 Decoding uterine leiomyoma tumorigenesis using single-cell transcriptomics and single-cell proteomics

Abstract Study question Can single-cell transcriptomics and proteomics contribute to a better understanding uterine of leiomyoma tumorigenesis? Summary answer We demonstrate the significant involvement of the MAPK, PI3K-Akt, and proteoglycan pathways in smooth muscle, endothelial and perivascular cells in leiomyoma tumorigenesis. What is known already Uterine leiomyomas (LM), also known as fibroids, are benign tumors of the uterus that arise from the myometrium. Previous studies have shown cellular heterogeneity in both myometrium and uterine leiomyomas, although cell spatial location within the tissue has not been shown. Further, no data describing single-cell-level proteomic differences and key pathways involved in uterine LM tumorigenesis have yet been reported. Study design, size, duration A prospective, observational, and biomedical study of cohorts was conducted at Hospital La Fe (Valencia, Spain) for one year. Single-cell RNAseq (scRNA-seq; n = 16) and single-cell proteomic (scP; n = 16) analyses were performed on eight sample pairs of LM and matched myometrium (MM), to generate a high-resolution transcriptomic and proteomic map decoupled from cell type, state, and spatial location. Participants/materials, setting, methods After obtaining informed consent, LM and MM samples were collected from eight patients between 35-50 years undergoing hysterectomies. Part of the samples were preserved in paraffin for spatial transcriptomics using VISIUM (10x Genomics). While the remaining tissues were dissociated into single-cell suspensions and subjected to Chromium Controller and Orbitrap Eclipse Tribid mass spectrometry for scRNA-seq and scP, respectively. All data were analyzed using publicly available R/Python tools. Main results and the role of chance After restrictive quality control filtering, we analyzed a total of 52,599 and 5,909 cells by scRNA-seq and scP, respectively. While LM and MM possessed similarities in terms of cellular composition, they displayed differential expression of genes and proteins across all the cell populations studied, particularly in smooth muscle, endothelial and perivascular cells. In LM samples, these cell populations displayed impaired MAPK signaling, which acts as a signal integrator for growth factors, estrogen, and vitamin D. We also observed alterations in the PI3K-Akt and proteoglycan pathways in LM smooth muscle and perivascular clusters, which relate to cell proliferation and tumor growth. Additionally, we encountered a subset of consistently dysregulated genes in all LM populations, which may suggest the existence of a shared tumorigenic pathway independent of cell type. Spatial transcriptomics further demonstrated relationships between cells and their relative locations within the tissue, which we validated by immunofluorescence. Together, our results highlight the relevance of specific cell populations in LM tumorigenesis. Limitations, reasons for caution This study involved a sample cohort limited to Caucasian women; therefore, further studies including more patients, and addressing racial disparities will help to generalize these findings to a broader population. Wider implications of the findings Our work describes an unprecedented transcriptomic and proteomic analysis of LM and MM at single-cell resolution, which supports a novel understanding of myometrial tumorigenesis. Trial registration number NCT04214457