Whole Genome Sequencing Illuminates the Genetic and Biological Features Underlying the Transition of SMM to MM

Abstract 296

A well defined model of disease progression has been established in myeloma based on the transition of normal plasma cells to monoclonal gammopathy of undetermined significance (MGUS), smouldering myeloma (SMM), myeloma (MM) and finally to plasma cell leukaemia. The most likely mechanism underlying the initiation and progression of myeloma is that the primary genetic events, such as IGH translocations or hyperdiploidy confer a proliferative/survival advantage but not a truly malignant phenotype. Subsequently, these cells acquire secondary genetic aberrations that are associated with uncontrolled proliferation and invasion which manifest as bone lesions and myelosuppression. We have addressed the hypothesis that we can gain critical understandings of the mechanisms leading to disease progression, by defining the different genetic events present in the early pre-clinical stages of the disease (SMM) compared to those present in the clinical stages of the disease (MM). One of the essential components of disease progression is the acquisition of novel genetic hits and clonal selection based on the selection and expansion of the subclone most able to survive. We hypothesise that within the malignant plasma cells there is substantial clonal heterogeneity and that at different stages of the disease the extent of this is variable, with progression occurring in an oligoclonal fashion by acquisition of multiple genetic hits rather than as a result of the linear acquisition of sequential genetic hits.

In this study we utilized massively parallel sequencing to study paired plasma cells from 2 patients who had progressed from SMM to MM and compared them to the patients non-involved DNA. Patients had been diagnosed with SMM, and had a bone marrow aspirate taken, at least 18 months prior to being diagnosed with MM. Both samples were taken before the patient underwent any treatment. The 2 patients were analyzed at the two time points and compared to the germline DNA obtain from the peripheral blood sample, using whole genome sequencing to identify acquired single nucleotide variants (SNVs), indels and translocations in the SMM and MM samples. Additionally, changes acquired at the transition from SMM to MM were examined along with frequency of abnormal reads at these sites. 100 ng genomic DNA from CD138 MACsorted cells and normal white cells was sequenced using 75 bp paired-end reads on a GAIIx (Illumina) to a median depth of 32x and 98% at 1x and 84% at 20x coverage. Data were aligned to the human genome (hg19) using ELAND v2e and acquired SNVs and indels called using CASAVA 1.8.

In both samples the majority of SNVs were found in intergenic (range 64–71%) or intronic (25–31%) regions. The number of acquired SNVs within exons (including UTRs) accounted for only 0.5–1.1% of all variants. There was no statistical difference in the genomic distribution of SNVs between SMM and MM samples. Analysis of the SMM and MM samples shows that SNVs in the MM sample, which are acquired in the tumour, are all present in the SMM sample albeit at lower levels, indicating that the dominant MM clone was always present before transformation from SMM. Additionally, there are SNVs present in the SMM sample which are not found in the MM sample from the same individual. For example, variant calls can be detected at an incidence of 11–15% in the SMM sample and are undetected in the MM sample, at this sequencing depth. Acquired indels were found almost exclusively (>99%) within intergenic and intronic regions in both SMM and MM samples. As with the SNVs, all indels found in the MM samples were also found in the SMM sample from the same patient, but not all indels found in the SMM sample were found in the MM sample.

The results presented indicate that during the progression from SMM to MM there is an oligoclonal selection process which results in the emergence of a dominant clone, and other sub-clones containing non-advantageous passenger mutations are not selected. Any of these SMM clones can clearly and rapidly transform to MM with full malignant potential. Therefore, targeted therapeutic approaches aimed at mutations in the MM dominant clone may not be curative, but may allow the propagation of minor clones without these mutations. It is therefore imperative to determine which mutations are present in pre-malignant clones to establish a clinical course of action in treatment.