Background: Multiple myeloma (MM) pathogenesis is characterized by a complex Darwinian evolution process where the progressive accumulation of genetic aberrations confers distinct proliferative advantage and clonal selection. Translocations between the immunoglobulin heavy chain (IGH) locus and known oncogenes, and multiple chromosome gains (Hyper- diploidy, HRD) are considered early drivers, being detectable also in pre- malignant stages of the disease. The IGH translocations are thought to arise from aberrant AID activity in the germinal centre; however, the pathogenesis of HRD and other aneuploidies remains unclear but are hypothesized to occur during a single failed mitotic event rather than through serial acqui- sitions over time. Aims: To investigate the order of acquisition of aneuploidies in MM, we analysed whole genomes sequencing (WGS) data from 67 tumour samples, collected at different clinical time points from a cohort of 30 patients; 11 with smoldering MM (SMM) and 56 with MM at diagnosis or later stages (median of 2 samples per patients; range 1-4). Methods: Aligned sequence files were analyzed using the published tools available at the Wellcome Trust Sanger Institute. Aneuploidies and their cancer cell fraction (CCF) was extracted by the previously published Bat- tenberg approach. Results: CCF analysis of the showed that 30% recurrent anueploidies were not fully clonal, suggesting their acquisition at different times during disease evolution. Ranking each event based on the distribution of its CCF, we observed that most chromosome gains in HRD, as well as amp1q, del1p and del13q occurred earlier compared to del17p, del14q and del8p. To increase the resolution of our analysis, we further developed a “molecular time” (MT) analysis of each clonal chromosome duplication greater than 1 Mb. The MT was calculated as the corrected ratio between the total number of clonal SNVs present on > 1 allele and all other clonal SNVs on one allele only. During a single allele duplication, all SNVs already present will be duplicated acquiring an allelic fraction ~66% and conversely all SNVs acquired on the non-duplicated allele (or on one of the two gained alleles after the duplication) will have an allelic fraction ~33%. In this analysis, we considered only those copy number alteration segments with a length > 1 Mb and with more than 50 clonal SNVs, extracted using the previously described dirichlet process. Integrating CCF and MT data we found that not all HRD gains were acquired at once, and in 11/18 (61%) cases the HRD karyotype was the result of multiple independent gains (MIG) acquired at different times. Interestingly we observed that in 12/18 (66%) patients the first HRD MIG event also involved other events, including amp1q that was therefore confirmed as an early event. Similarly, independent MIGs were also present in the MM (sub)clonal architecture evolution, with mul- tiple CNAs gains lost and/or acquired during different cancer progression phases (from SMM to MM and from MM to relapsed MM). At the extreme end of the spectrum, we found 4 patients acquiring a whole genome dupli- cation, whose occurrence only in relapsed samples suggests a potential role in relapsed/refractory stages. Summary/Conclusion: In conclusion, in this study we described for the first time the pattern of acquisition of the HRD karyotype in MM. By integrating this analysis to the timing of acquisition of mutations, we have reconstructed the life history of the 30 MM cases, highlighting dynamic changes both spontaneous and after treatment, that will impact the way we manage the disease both in SMM and in active MM.
|Titolo:||The chronological order of copy number changes in hyperdiploid multiple myeloma patients|
MAURA, FRANCESCO (Primo)
BOLLI, NICCOLO' (Secondo)
|Settore Scientifico Disciplinare:||Settore MED/15 - Malattie del Sangue|
|Data di pubblicazione:||2018|
|Appare nelle tipologie:||01 - Articolo su periodico|