Blind men and an elephant

In this issue of Blood, Pawlyn et al examine the prognostic implications of overlapping chromosomal abnormalities in multiple myeloma (MM), demonstrating that coexistence of hyperdiploidy does not mitigate the impact of high-risk abnormalities.¹ 

The story of the blind men and an elephant originated in the Indian subcontinent and describes a group of blind men coming to different conclusions about how an elephant looks like by feeling different parts of the animal. The parable implies that one's subjective experience may be true but is inherently limited by its failure to account for alternate possibilities or the sum total of facts. One faces a similar situation when trying to examine the genetic complexity in myeloma.^2,3  When only metaphase cytogenetics was available, an abnormality was detected in one-third of the patients, and it implied poor prognosis. With the introduction of interphase fluorescence in situ hybridization (iFISH), it became clear that nearly all myeloma cells carried 1 or more abnormalities.⁴  Common abnormalities included translocations involving the heavy chain locus on chromosome 14 and a set of recurrent partner chromosomes (immunoglobulin heavy-chain locus [IgH] translocations) or trisomies of odd-numbered chromosomes (trisomic myeloma).^2,5,6  In addition, many had other abnormalities, mostly deletions involving chromosomes 1, 13, and 17, monosomies of chromosomes 13 and 17, and amplification of chromosome 1q. Because IgH translocations or trisomies are present in the majority of patients, these are considered early events in the development of myeloma. Many of these iFISH abnormalities have been consistently associated with poor outcomes, especially t(4;14), t(14;16), t(14;20), and del(17p).^5,6  Deletion of the short arm of chromosome 17 in particular is associated with poor survival. At the same time, certain therapies can mitigate the risk associated with specific abnormalities as is the case with t(4;14) or del(17p) and bortezomib-based therapies.^7,8  The recognition that specific abnormalities may benefit from particular drugs has also led to development of risk adapted treatment algorithms.

Over the years, we have classified the common iFISH abnormalities as standard, intermediate, or high risk, based primarily on its impact on overall survival with the available therapies. However, these abnormalities, with the exception of the different IgH translocations, are not mutually exclusive and can often coexist in the same plasma cell. This leads to additional complexity in interpreting the results of iFISH tests, and only recently has there been systematic evaluation of the implications of overlapping abnormalities, especially the overlap between high-risk and standard-risk lesions. Although the presence of multiple high-risk lesions is typically associated with worse outcomes, the interpretation of coexistent standard- and high-risk abnormalities can be confusing. The study presented by Pawlyn et al demonstrated superior outcome for patients with hyperdiploidy but also showed that the poor outcome seen in patients with high-risk lesions described above, as well as in those with 1q amplification, was not altered by the presence of concurrent hyperdiploidy. These results contrast sharply with a previous report from the Mayo Clinic examining the implications of trisomies among patients with newly diagnosed myeloma treated with modern therapies.⁶  In that study, patients with high-risk lesions including del(17p) had a better survival when they coexisted with trisomies compared with the rest. The different results from the 2 studies can be related to several factors, the most important of which is the difference in the therapies used. The study presented by Pawlyn et al used induction regimens that combined cyclophosphamide with thalidomide and dexamethasone, or with vincristine, doxorubicin, and dexamethasone, whereas the Mayo Clinic study included patients treated mostly with lenalidomide- or bortezomib-based regimens. The differential impact of the regimens used is clear from the superior overall survival of the entire cohort in the Mayo study. The impact of coexistent abnormalities was also examined in a recent study by the Intergroupe Francophone du Myélome in 242 patients with t(4;14) or del(17p) abnormalities using a single nucleotide polymorphism array. As with the Mayo Clinic study, a protective effect of trisomies in patients with del(17p) was seen.⁹  Although more work is required to define the impact of the different abnormalities, it is clear that the outcomes of patients depend on a complex interplay of factors including the specific combinations of lesions, the therapeutic approaches used, and the magnitude of treatment response. Thus, although the study by Pawlyn et al found no effect of trisomies in high-risk patients using thalidomide- or doxorubicin-based regimens, 2 other studies that used lenalidomide- and/or bortezomib-based therapies found that trisomies do ameliorate the adverse prognostic effect of high-risk cytogenetics in myeloma.

Coexistence of these abnormalities raises important biological questions; specifically, the chronology of development of the 2 types of abnormalities remains unclear. The current study, by performing single cell analysis of plasma cells, suggests that development of trisomies may precede development of IgH translocations. This sequence of development of genetic abnormalities had been suggested previously by Chng et al, who performed a comprehensive analysis of karyotypes in 469 patients with hyperdiploid MM.¹⁰  However, in the current study, this assumption is based on analysis of cells from 5 patients and needs verification in a larger group of patients using similar techniques.

The challenge going forward is to develop a better understanding of the implications of the multiple abnormalities seen in MM, not only from a prognostic standpoint but also the selection of therapy. Unlike the blind men coming to different conclusions about the elephant, we need to develop methodologies that will allow us to integrate all of the available genetic information to better predict outcomes in MM. We also need to recognize that the impact of various prognostic factors will vary based on the specific therapy used, and thus generalizations are not possible if a significant change in treatment is present. The demonstration of numerous mutations in myeloma cells demonstrated by recent studies using whole genome sequencing has made this task more difficult.³  Ongoing studies in the context of large clinical trials with uniform therapies will continue to shed more light on this complex problem and contribute to improving our understanding of the disease biology.