MDS: unraveling the mystery

David A. Rizzieri

In this issue of Blood, Sekeres et al report a phase 2 study of lenalidomide and azacitidine for higher risk myelodysplastic syndromes (MDS).1 High overall response and duration of response supports the concept of combination, multitargeted approaches rather than traditional chemotherapeutics for this disease.

MDS affect approximately 75 per 100 000 persons in the US over the age of 65 and is widely recognized as clonal disorders of diverse pathogenicities leading to bone marrow failure, cytopenias, infection and bleeding risks, leukemic transformation, and early death.2 Until recently, that was essentially where our understanding ended, prompting some to adapt Churchill's adage that it is “… a riddle, wrapped in a mystery, inside an enigma; but perhaps there is a key….” Recent advances in molecular techniques can be well applied to this problem, helping us to unravel this mystery. Although conventional karyotyping may detect abnormalities in only approximately 50% of subjects, single nucleotide polymorphism arrays and comparative genomic hybridization can identify abnormalities in approximately 80%. An integrated look into the overlap of the multiple abnormalities identified in MDS has recently been presented by Lindsley and Ebert.3 In those with 5q deletion, the haplo-insufficiency for this ribosomal protein gene leads to an abnormally elevated p53 drive and is associated with the increased apoptosis noted in this disorder. In addition, altered miRNA expression from that region triggers signaling cascades affecting the toll receptor pathway and nuclear factor-KB activation, leading to competitive growth advantage of the dysplastic cells. Numerous mutations have recently been identified that regulate methylation networks as well, including DNMT3A, IDH, TET2, and others, noting the effect changes have promoting the disease. Genome sequencing has also revealed recurrent somatic mutations in the spliceosome, with evidence that this machinery is involved in the epigenetic regulation of gene expression as well. Understanding the combinations and permutations of abnormalities in RNA splicing, epigenetic regulation, DNA repair, and kinase signaling may lead to new and better therapies for these heterogeneous disorders.

This improved understanding of abnormal molecular pathway regulation in MDS has led to initial forays into targeted therapies, resulting in 2 classes being approved for this disease: the imids and hypo-methylating agents. Lenalidomide inhibits haplo-deficient phosphatases and releases progenitors from p53 arrest.4 Additional mechanisms include anti-angiogenesis via inhibition of bFGF-, VEGF-, and TNF-α–induced endothelial cell migration. Further, lenalidomide has multiple immunomodulatory effects involving stimulation of T-cell proliferation including natural killer (NK) cell number and function, and the production of multiple cytokines.5,6 As a single agent it results in significant improvements, particularly in patients with chromosome 5q31.1 abnormalities. List and colleagues reported an 83% response rate in those with chromosome 5q31.1 abnormalities and 57% in those with normal karyotype, with 12% response in those with other karyotype changes.7 Of the 20 with abnormal karyotypes identified, 50% attained a complete cytogenetic remission, which appeared to be durable.7 Similarly, hypo-methylating agent benefits include transfusion independence, encouraging complete remission rates, and even improved overall survival in phase 3 prospective studies. Kantarjian and colleagues showed in a phase 3 study an overall response rate of 16% and another 47% hematopoietic improvement in those treated with 5-azacytidine compared with 7% with supportive care.8 There was a median of 22 months until leukemic transformation or death in the treatment group versus 7 months in the supportive care arm as well (P < .0001). Similar benefits have also been reported for azacytidine.8 These effects are mediated in part through depletion of methylcytosine resulting in reversal of hypermethylation of CpG islands in the promoter regions of certain genes, leading to the reversal of epigenetic silencing.

Studies effectively combining multiple targeted therapies given concurrently, as provided here by Sekeres et al, have been lacking. The rationale for such a study is clear as those with higher risk MDS seem to retain the multiple abnormal mechanisms present from the low-risk state from which they often evolved. Thus, using therapies that attack different mechanistic abnormalities may allow for synergistic benefits. This study treated 36 patients with azacytidine (75 mg/m2 per day × 5 days) and lenalidomide (10 mg per day × 21 days [28-day cycle]), resulting in a 72% overall response rate (44% complete response) with a median duration of response over 17 months. Infections, cytopenias, and constitutional symptoms in the 10% range were the primary toxicities encountered. Building on recent information elucidating the role of multiple molecular abnormalities in MDS, Sekeres et al correlated their clinical work with a comprehensive genetic analysis of their subjects. Their data suggest patients with abnormalities in the methylation pathways have a higher rate of complete remissions with this combination therapy than would otherwise be expected. This type of molecular characterization of responders/nonresponders is what is needed as we increase our targeted therapy options and seek to personalize care choices in future studies.

Success of this combination targeted therapy approach is exciting and encouraging, although obstacles remain. While we continue to focus on maximum tolerated dosages for combination strategies, there is growing evidence for the relevance of both scheduling and ratio-metric dosing to optimize tumor kill. Because blood-borne cancers are not common, support for clinical studies to allow for rapid accrual is important but often lacking, particularly if the targeted therapies are made by different companies. Interpretation and extension of the clinical results is dependent upon completion of the important hypothesis-driven correlative science as noted in this report. Yet, increasingly, this funding is difficult to secure in the current research environment. The spiraling costs of health care may soon impact the planning of such novel combinations as well. Wang et al have recently noted the significant improvement in survival provided by hypo-methylating agents in this disease is associated with a dramatic increase in costs of care, indicating comparative effectiveness research may have a growing role in our future assessments of targeted therapy efficacy as well.9 For these reasons, while it is an exciting time for cancer research, translating the significant advances in our understanding of the molecular pathogenesis of diseases to improved combination targeted therapies in the clinical arena will remain a challenge for our current generation of clinical trialists.


  • Conflict-of-interest disclosure: The author reports research support, advisory board participation, and speaker bureau support from Celgene, and research support from Supergen. ■