OA-28: Unraveling the heterogeneity of multiple myeloma associated with therapy resistance by single-cell RNA sequencing

Introduction: Multiple myeloma (MM) is a malignancy of clonal plasma cells with identical VDJ recombination of immunoglobulin loci (called repertoire) and extensive genome or transcriptome heterogeneity. Even though many new treatment modalities have been developed to extend patient survival, most cases remain incurable. Recent evidence suggests that intra-tumor heterogeneity underlies the therapy resistance, but its details remain largely unknown.

Methods: To understand the comprehensive architecture, we subdivided bone marrow (BM) cells from MM patients at different disease stages (n=8) into 11 fractions by flow cytometry (FCM) and performed bulk RNA sequencing (bulk RNA-seq). Using BM cells with MM sorted by the panel established based on clonal repertoire distribution, we performed single-cell RNA sequencing (scRNA-seq) coupled with VDJ targeted sequencing (scVDJ-seq) (n=11) including those at diagnosis and in remission with minimal residual disease (MRD).

Results: RNA sequencing-based profiling of repertoire revealed that tumor cells are also detected in Lineage marker– (Lin–)/ CD19–/CD38++/CD138– fraction other than the main fraction (Lin–/CD19–/CD38++/CD138+) regardless of disease stage. An analysis of scRNA-seq coupled with scVDJ-seq on Lin–/CD19–/CD38+ fraction, without CD138 gating, discriminated MM cells (20084 cells) clearly from others. By integration with gene signatures established from bulk RNA-seq of CD138+ and CD138– MM cells (n=12), the heterogeneity on the CD138 axis was also unraveled. FCM analysis confirmed that CD138– MM cells were present at a rate of less than 10% across the disease stages (n=59), which showed unique morphology, methylation of histone H3K4 and cell cycle status distinct from CD138+ cells. Primary cell culture and trajectory/velocity analysis of scRNA-seq suggested that heterogeneity along the CD138 axis might be regulated by epigenetic plasticity. Interestingly, we identified unique clusters in the CD138– population which tended to remain at high proportions after treatment. They were characterized by low expression of MHC class-I genes and BCMA as well as 198 upregulated genes, including MCL1, MALAT1, NEAT1, which have been previously implicated in treatment resistance. We systematically interrogated their dependencies in growth of MM cells using a custom CRISPR/Cas9 screening system and narrowed down several important genes. Analysis of public databases showed that high expression of a part of these genes, including RNA-binding protein 39 (RBM39), are significantly associated with poor prognosis. We confirmed that RBM39 is indispensable for the MM cells by CRISPR/Cas9-mediated knockout and pharmacological protein degradation.

Conclusions: We achieved a complete view of MM diversity that was not limited by cell surface antigens or transcriptome characteristics by using repertoire clonality. Our results highlight a novel subpopulation as a candidate therapeutic target for cure.