Single cell RNA sequencing of dormant myeloma cells from bone

Multiple myeloma is a neoplasm of plasma cells which develops in the skeleton. The bone microenvironment supports myeloma growth and long-term survival of dormant myeloma cells. Despite targeted treatments, patients relapse and survival is poor. Dormant cells contribute to disease relapse and represent potential therapeutic targets. However, the molecular pathways that control dormancy are unclear. We developed a method to identify dormant cells in vivo and have used single cell RNA sequencing (scRNA-seq) approaches to define the transcriptome profile of these cells.To identify dormant myeloma cells in the skeleton, 5TGM1 murine myeloma cells expressing eGFP were labelled with a fluorescent membrane dye, (Vybrant DiD). Dividing cells share DiD with daughter cells which DiD is diluted and activated cells become DiDneg, whereas, non-dividing dormant cells retain DiD (DiDhigh). DiD-labelled cells were injected (i.v.) into C57BLKalwRij mice. After 21 days, individual DiDhigh and DiDneg cells were sorted into 384 well plates by FACS for scRNA-seq and sequenced at >2x106 reads/cell. 5TGM1 cells were validated by the reconstruction of the unique clonal idiotypic sequence and the expression of specific Ig genes. A mean of ~4000 unique transcripts were identified with over 900 genes up-regulated in DiDhigh cells compared to DiDneg cells (fold change >2). Functional annotation of differentially expressed genes confirmed the dormant status of DiDhigh cells with down-regulation of metabolic pathways and up-regulation of immune system-related pathways. We identified the receptor tyrsone kinase, Axl, as a potential therapeutic target. FACS analysis confirmed expression in dormant cells and analysis of CD138+/CD38+ plasma cells from myeloma patients identified a small population of cells express AXL. Inhibition of Axl with Cabozantinib or BMS-777607 decreased dormant cell numbers and increased tumour burden in myeloma bearing mice suggesting Axl is functionally important in dormancy retention. Moreover, survival analysis in independent patient cohorts correlated the dormancy signature expressed in dormant myeloma cells with improved survival.These data demonstrate that dormant myeloma cells persist in the skeleton, have a unique transcript profile and this knowledge can be used to predict survival and identify new targets that alter the behaviour of myeloma disease in the skeleton.