Single-Cell Genetic Analysis Reveals The Genetic Composition Of Founder Clones, Phylogenetic Patterns Of Branching and Parallel Evolution, and Clonal Fluctuations Following Patient Treatment In Multiple Myeloma

Multiple Myeloma (MM) is a proliferation of aberrant plasma cells in the bone marrow. Ig translocations or hyperdiploidy are MM initiating events present in all clonal cells in contrast to other secondary lesions involved in progression (e.g. RASmutations), which may be subclonal. We and others have recently described the presence of different tumour clonal populations—a phenomenon called intra-tumour heterogeneity—in MM, yet the phylogenetic relationships of the tumour subclones remains to be fully elucidated.

To describe the intra-tumour heterogeneity and tumour phylogenies in a series of t(11;14) MM patients characterised by KRAS/NRAS/BRAF mutations, we combined whole-exome sequencing and single-cell genetic analysis. A novel approach for single-cell multiplex-qPCR analysis using nano-fluidic arrays (Fluidigm) was followed in 100-250 single-cells per sample. The t(11;14) breakpoint was defined using Ig regions-targeted massively-parallel sequencing. Taqman assays specific for detection of the t(11;14) breakpoints and for mutations in selected genes were custom-made designed. Copy number assays for chromosomal regions of interest were also used. This strategy allowed us: first, to report the presence of the t(11;14), mutations and copy number aberrations (gains/losses) at the single-cell level; second, to define subclonal populations; and third, to delineate the most plausible sequence of events for each case. An additional series of 14 MM patients with paired-samples at presentation/relapse was included for the study of the effect of treatment in clonal architecture. Lastly, to analyse the engraftment-ability of subclonal populations, we injected 1x10⁶ CD138⁺ cells from one of the patient-samples at diagnosis into the tibia of NOD/SCIDyc^(null)-mice and compared the engrafted-myeloma with the paired presentation-relapse samples.

We demonstrate that MM is comprised of 2-6 clones, related through a linear (3/7 cases) or a branching (4/7) phylogeny. The t(11;14) was seen in 91-100% of tumour cells, supporting its aetiological role as a MM initiating event. For the first time in MM, we describe a parallel evolutionary pattern in two samples that carried double hits in KRAS or KRAS/NRAS. These mutations were acquired separately in two divergent clonal lineages, which were derived from the same ancestor but evolved independently. We suggest that RAS is a true driver mutation in MM as such alterations seem to provide clonal advantages for myeloma subclones in the bone marrow environment. We also report the concomitant acquisition of RAS mutations and IRF4p.K123R mutation in 2% (9/453) screened MM patients. This finding suggests a collaborative effect provided by the mutations in both pathways to trigger myeloma development.

We examine the ability of subclonal populations to survive patient treatment by the analysis of paired presentation-relapse samples. Not only did the intra-tumour heterogeneity shift in the transition to relapse meaning that subclonal populations fluctuated during treatment, but also new clones emerged from early and late subclones formerly described at diagnosis. To note, some of these new subclones acquired mutations in KRAS/BRAF during transition, likely leading to relapse. In parallel, we tested the ability to recapitulate the disease in NOD/SCIDyc^(null)-mice. Engrafted-myelomas also had clonal fluctuations with 1/3 clones shown at patient-diagnosis neither found at relapse, nor in the engrafted-MM. This missing clone was similarly outcompeted by the other clones during patient treatment and xeno-transplantation. Altogether, these results suggest that subclonal populations have different abilities to survive treatment or xeno-transplantation and hence, to lead to relapse or reconstitution of myeloma by the generation of new clonal subpopulations.

We confirm the existence of distinct MM subclones that are related through a linear or a branching phylogeny, with examples of parallel evolution for alteration of the RAS/MAPK pathway. Myeloma subclones are subject of a selection process involving clonal extinction and clonal tides, similar to the theory of Natural Selection by Charles Darwin. In conclusion, intra-tumour heterogeneity is an elementary foundation for Darwinian selection underlying both disease progression and the development of treatment resistance in MM.