Inference of genomic lesions from single-cell RNAseq in myeloma improves functional intra- and inter-clonal analysis.

Smoldering Multiple Myeloma (SMM) is an asymptomatic plasma cell (PC) neoplasm that may evolve with variable frequency into multiple myeloma (MM). SMM is initiated by chromosomal translocations involving the IgH locus or by hyperdiploidy and evolves through acquisition of additional genetic lesions. In this scenario, we aimed at establishing a reliable analysis pipeline to infer genomic lesions from transcriptomic analysis, by combining single-cell RNA sequencing (scRNA-seq) with B-cell receptor sequencing and copy- number abnormality (CNA) analysis to identify clonal PCs at the genetic level along their specific transcriptional landscape. We profiled 20,465 bone marrow (BM) PCs derived from five SMM/MM patients and unbiasedly identified clonal and polyclonal plasma cells. Hyperdiploidy, t(11;14) and t(6;14) were identified at the scRNA level by analysis of chimeric reads. Subclone functional analysis was improved by combining transcriptome with CNA analysis. As examples, we illustrate the different functional properties of a light chain escape subclone in SMM, and of different B-cell and PC subclones in a patient affected by Wäldenstrom Macroglobulinemia and SMM. Overall, our data provide a proof of principle for inference of clinically relevant genotypic data from scRNAseq, which in turn will refine functional annotation of the clonal architecture of PC dyscrasias.