Fax (713) 794-4297
Novel therapies for myelodysplastic syndromes†
Article first published online: 14 JUN 2004
Copyright © 2004 American Cancer Society
Volume 101, Issue 2, pages 226–241, 15 July 2004
How to Cite
Faderl, S. and Kantarjian, H. M. (2004), Novel therapies for myelodysplastic syndromes. Cancer, 101: 226–241. doi: 10.1002/cncr.20381
Dr. Armand Keating was Guest Editor for this article.
- Issue published online: 2 JUL 2004
- Article first published online: 14 JUN 2004
- Manuscript Accepted: 23 APR 2004
- Manuscript Received: 17 MAR 2004
- acute myeloid leukemia;
- chronic myelomonocytic leukemia;
- refractory anemia;
- myelodysplastic syndrome;
- targeted therapies
- Top of page
- CLASSIFICATION AND PROGNOSTIC FACTORS
- BIOLOGY OF MDS
The assessment of patients with myelodysplastic syndromes (MDS) and the choice of therapies remain challenging. New therapies are now emerging after the identification of molecular targets that result in improvement of hematologic parameters and may hold promise for the prevention of disease progression.
A review of the English literature was performed that included original articles and related reviews from MEDLINE (PubMed) and abstracts based on published meeting material.
MDS is a heterogeneous group of disorders. Although current classification and prognostic schemes have proven valid to define subgroups, they are insufficient to take into consideration the significant biologic diversity of MDS. New molecular targets are identified as the mosaic of pathophysiologic pathways in MDS is being unraveled. Novel and targeted therapeutic agents, such as the inhibition of farnesyl transferases and receptor tyrosine kinases, more potent thalidomide analogs, and arsenic trioxide, have shown encouraging results and may offer durable benefit to patients with MDS.
Although progress has been made in the understanding of clinical manifestations and some of the molecular pathways underlying ineffective hematopoiesis and leukemic transformation in MDS, intensive clinical and laboratory research continues to 1) identify further relevant pathophysiologic pathways, 2) better define MDS subgroups, and 3) develop new drugs based on a clearer understanding of disease biology. Cancer 2004. © 2004 American Cancer Society.
Myelodysplastic syndromes (MDS) are clonal hematopoietic stem cell disorders in which the common denominator is dysplasia in one or more hematopoietic cell lines.1 MDS have been referred to as refractory anemia, preleukemia, or smoldering leukemia, terminologies that reflect a tendency of MDS to evolve into acute myeloid leukemia (AML). However, restricting the view of MDS to a purely preleukemic condition ignores important differences: 1) The clinical picture and prognosis for patients with MDS are defined more by cytopenias, and most patients will not progress to AML; and 2) distinct biologic differences identify MDS as an entity separate from AML.2 Therapeutic decisions in patients with MDS are complex. Current classifications and prognostic systems do not take into account the considerable clinical heterogeneity of MDS, nor is there any approved therapy. Supportive care remains unsatisfactory, because patients continue to be exposed to the inherent complications of worsening cytopenias and leukemic transformation. Recent years have witnessed an evolution in our understanding of pathophysiologic pathways in MDS. At the same time, many novel and targeted therapies are being investigated in clinical trials, offering patients the prospect of sustained benefit, and changing the natural course of the disease.3 With further molecular dissection of MDS and development of new and targeted drugs, hopefully, significant progress soon will contribute to improved survival and quality of life for patients with MDS.
- Top of page
- CLASSIFICATION AND PROGNOSTIC FACTORS
- BIOLOGY OF MDS
The median age of patients with MDS at diagnosis is between 60–70 years. A slight male predominance is typical. Approximately 12,000 new cases of MDS are diagnosed each year in the U.S., although this figure may underestimate the true incidence of MDS.4
Although numerous risk factors have been identified (male gender, smoking, hair dyes, pesticides, herbicides, organic chemicals, heavy metals, benzene, and other solvents), a clear association with any of these predisposing factors is reported to be found in only 20–30% of patients (secondary MDS).5–9 Treatment-related MDS (t-MDS) after exposure to bone marrow-damaging agents such as chemotherapy drugs or ionizing radiation has received the most attention.10–12 Secondary MDS and t-MDS differ from primary MDS, in that the age of onset is approximately 10 years earlier, dysplasia is more prominent, cytopenias are more severe, disease progression is more rapid, and outcomes are worse.13
Disease manifestations (fatigue, pallor, infections, bleeding) are related mainly to the severity of blood cytopenias. Lymphadenopathy and hepatosplenomegaly are infrequent. Patients with chronic myelomonocytic leukemia (CMML) are more likely to present with leukocytosis and extramedullary involvement.
MDS blasts usually are myeloid in origin, and only rare cases exhibit B-lineage lymphoid (CD19 or CD10) or mixed-lineage morphologies. Immunophenotypic clustering of blasts by flow cytometry may become a useful tool for diagnosis and determination of response to therapy.14, 15 Bone marrow samples in MDS are typically hypercellular. Albeit rare, the diagnosis of hypocellular MDS is important because, in these patients, a significant overlap with aplastic anemia may exist and immunomodulatory strategies may be beneficial.
Other conditions may be associated with dysplastic changes (e.g., vitamin B12 and folic acid deficiency, malnutrition, human immunodeficiency virus [HIV]-related disorders, chronic inflammation, liver disorders, metastatic disease to the bone marrow, and other hematologic malignancies) and are to be excluded by appropriate tests and sometimes a period of observation.
CLASSIFICATION AND PROGNOSTIC FACTORS
- Top of page
- CLASSIFICATION AND PROGNOSTIC FACTORS
- BIOLOGY OF MDS
Although the French–American–British (FAB) classification system (Table 1) has been valuable for the management of patients with MDS, it has several disadvantages16, 17: 1) It does not account for other prognostic markers (e.g., karyotype); 2) some MDS entities (e.g., 5q − syndrome) are not considered; 3) the FAB groups describe disease progression through different stages of one disease (from refractory anemia [RA] or RA with ringed sideroblasts [RARS] to RA with excess blasts [RAEB], RAEB in transformation [RAEB-t], and AML) rather than different disease entities in which commonality need not be defined by the percentage of blasts; 4) patients who have trilineage cytopenias have a less favorable prognosis compared with patients who have cytopenia in one-cell lineage, a circumstance that is not considered in the RA category; 5) the group of RAEB is too heterogeneous; 6) Auer rods have been associated with a favorable prognosis rather than a poor prognosis18; 7) the inclusion of CMML as part of the MDS is problematic19; and 8) there is no separation of t-MDS from de novo MDS.
|Category||Frequency (%)||% blasts||Transformation to AML (%)||Other|
|RA||10–40||< 5||≤ 1||10–20||—|
|RARS||10–20||< 5||≤ 1||10||>15% RS|
|RAEB-t||10–30||21–29||≥ 5||50–60||Auer rods|
|CMML||10–20||≤ 20||< 5||20||> 109 monocytes|
Addressing some of these concerns, the World Health Organization (WHO) has suggested a new classification system for MDS (Fig. 1).20 RA and RARS have been divided into unilineage and multilineage dysplasias. The 5q − syndrome is now defined based on the presence of < 5% bone marrow blasts, thrombocytosis, and the characteristic morphologic findings of micromegakaryocytes. RAEB includes 2 groups, RAEB-1 (5–10% blasts) and RAEB-2 (11–19% blasts). RAEB-t has been eliminated, and patients with ≥ 20% blasts are considered to have AML.21 CMML is a separate entity characterized by either myelodysplastic or myeloproliferative features. A category of unclassifiable MDS (MDS-U) has been added. Although it is an advance over the FAB classification system, the WHO nomenclature still is based on morphology and blast counts, and whether it can capture the clinical heterogeneity of MDS remains to be seen.22
The International Prognostic Scoring System (IPSS) is based on the percent of bone marrow blasts, the presence of cytogenetic abnormalities, and the degree of cytopenia (Table 2).23 It assigns points to each of the factors and divides patients into low, intermediate-1, intermediate-2, and high-risk groups with corresponding median survival times of 5.7 years, 3.5 years, 1.2 years, and 0.4 years, respectively. The IPSS was developed in newly diagnosed, untreated patients; survival was calculated from diagnosis. Therefore, it may not apply well to other groups of patients (e.g., previously treated patients or patients who are referred to tertiary care centers).24
|Variable||IPSS score value|
|Bone marrow blasts (%)|
|IPSS risk group|
|Five-yr survival rate (%)|
|Progression to AML (%)c|
BIOLOGY OF MDS
- Top of page
- CLASSIFICATION AND PROGNOSTIC FACTORS
- BIOLOGY OF MDS
MDS is a clonal disorder of hematopoietic stem cells that results in excessive apoptosis, as reflected by the degree of dysplasia and proliferation and loss of differentiation of hematopoietic progenitor cells.1, 25–27 Cytogenetic abnormalities occur in 30–80% of patients with primary MDS and in 80–100% of patients with t-MDS.28, 29 Deletions of the long arm of chromosomes 5 (5q −), 7 (7q −), and 20 (20q −) are common. Subgroups of MDS can be defined based on specific cytogenetic abnormalities (Table 3). Chromosome 5 abnormalities are of particular interest, because several genes involved in hematopoiesis are located there.30 Interstitial deletions of the long arm of chromosome 5 between bands q12 and q32 as the sole karyotypic abnormality describe the 5q − syndrome (Table 4).31, 32 The 5q − syndrome has been associated with prolonged survival and the absence of transformation to AML. Lower levels of apoptosis in 5q − syndrome have been described and may explain the milder clinical course of this MDS subtype.33 In few patients with CMML, balanced reciprocal translocations have been demonstrated involving the platelet-derived growth factor receptor β (PDGFRβ) gene (Table 5).34 Irrespective of the fusion partner, these translocations result in constitutively abnormal activation of the tyrosine kinase receptor function of the PDGFRβ. Thus, detection of these fusion proteins offers additional targets for therapeutic agents, such as imatinib mesylate, a potent inhibitor of the PDGFRβ tyrosine kinase.
|Cytogenetic abnormality||Clinicopathologic features|
|Deletion 5q31||RA (5q − syndrome)|
|Deletion 17p||Pseudopelger-type neutrophils and cytoplasmic vacuoles|
|Monosomy 7||Neutrophil dysfunction|
|Abnormal 3q26 (EVI-1)||Thrombocytosis plus dysplastic megakaryocytes|
|5q33 translocations (PDGFRβ)||CMML-Eos (WHO)|
|Deletion 20q||RA, favorable prognosis|
|5q− as the sole cytogenetic abnormality|
|Mostly younger women|
|Normal or increased platelet count|
|Macrocytosis, eythroid hypoplasia, monolobated megakaryocytes|
|Indolent course, favorable prognosis (AML transformation, 5–10%; median survival, > 5 yrs)|
|Responsive to hematopoietic growth factors|
|Transfusion needs (watch iron overload)|
The initial stages of MDS are defined by excessive apoptosis of progenitor cells thereby leading to ineffective hematopoiesis, which is counterbalanced by an increased rate of proliferation of hematopoietic elements. Overproduction of proapoptotic cytokines, such as tumor necrosis factor α (TNF-α) and soluble Fas ligand, may contribute to excessive apoptotic cell death in MDS.35, 36 Increased TNF-α production appears to be restricted to earlier MDS phases (RA and RARS) and is not observed in patients with high-grade MDS or AML. Hematopoiesis may be compromised by numerous, often ill-defined injuries to the bone marrow space. Dysregulation of the immune response has been implicated and provides the rationale for the use of immunosuppressive agents in MDS, such as antithymocyte globulins (ATG), cyclosporine A, or steroids.37–39 During proliferation, and perhaps through genomic instability, additional molecular abnormalities (bcl-2 overexpression, p53 mutations, Ras mutations and dysfunction, hypermethylation of p15INK4b) shift the balance from excessive apoptosis toward maturation arrest and unchecked proliferation of hematopoietic progenitor cells, a process that predominates in more advanced MDS phases (RAEB and RAEB-t).40, 41
- Top of page
- CLASSIFICATION AND PROGNOSTIC FACTORS
- BIOLOGY OF MDS
Although it is unsatisfactory for most patients, supportive care still is considered the mainstay of management for patients with MDS. Supportive care measures include transfusion of blood products, antibiotics, and hematopoietic growth factors. Trials of vitamins (D, A, retinoids), differentiating agents (hexamethylene bisacetamide, sodium phenylbutyrate), amifostine (a cytoprotective sulfhydryl compound with in vitro hematopoietic-stimulatory activity), or androgens may be beneficial in occasional patients but generally should not be recommended, because their overall success rate is low.
Guidelines for therapy have been proposed (Fig. 2).42 In general, all patients should be encouraged to participate in clinical trials. Response criteria for clinical trials in MDS differ in some respects from response criteria for acute leukemias (Fig. 3), acknowledging that improvement of hematologic indices is a valuable treatment goal in these patients.43
Hematopoietic growth factors
Erythropoietin (Epo) and granulocyte-colony stimulating factor (G-CSF) may improve anemia and reduce transfusion requirements in up to 40% of patients. Responses are more favorable in patients who have RA and RARS compared with patients who have RAEB, are durable in a substantial number of patients with low-risk MDS, and increase the patients' quality of life without adverse effects on long-term outcome.44, 45 In a clinical decision model for treating anemia in MDS with combinations of Epo and G-CSF, the best responses are achieved in patients who require < 2 units of packed red cell transfusions per month and with low serum Epo levels (< 100 mU/mL) (Fig. 4).46 Combinations of Epo with all-transretinoic acid (ATRA) have shown activity that results not only in erythroid but also in neutrophil and platelet responses.47 Combinations of erythropoietic proteins and thalidomide have resulted in a high incidence of thromboembolic complications and, thus, should be used with caution.48 Darbepoietin is a modified form of Epo with prolonged serum half-life and increased in vivo biologic activity, necessitating less frequent dosing.49 Its impact on MDS treatment needs to be established.
Although G-CSF and granulocyte–macrophage-colony stimulating factor improve neutropenia in 70–80% of patients, there have been no data showing that they reduce infectious episodes, prolong survival, or influence the rate of transformation to AML. Their use appears reasonable in patients with MDS and recurrent neutropenic febrile episodes, but they are not recommended for chronic prophylactic use alone in patients with MDS.50, 51
Interleukin-11 (IL-11) is a megakaryocytic growth factor that attenuates thrombocytopenia and reduces the need for platelet transfusions after myelosuppressive chemotherapy in patients with nonmyeloid malignancies. A recent report indicated platelet responses in 38% of patients with MDS who received low doses of IL-11 (10 μg/kg daily).52 Pegylated recombinant human megakaryocyte growth and development factor (MGDF) stimulates megakaryopoiesis in vitro and increases platelet counts in healthy volunteers and in patients with cancer after chemotherapy. In a Phase I/II study, MGDF was given intravenously daily for 14 days to 16 patients with aplastic anemia (AA) and to 21 patients with MDS (RA and RARS). Increases in platelet counts were observed at 5–6 weeks after therapy in higher dose cohorts.53
Because early-stage MDS shares some pathobiologic characteristics of acquired AA, immunosuppressive therapies (ATG or antilymphocyte globulins [ALG], cyclosporine A, steroids) may be beneficial.54 Immunosuppressive therapy in unselected patients with MDS can lead to sustained increases in red blood cell, neutrophil, and platelet production in 15–30% of patients.38, 55, 56 Expression of D-related human leukemic antigen 15 (HLA-DR15; a serologic split of HLA-DR2 that is over represented in MDS, similar to AA), younger age, and shorter duration of red cell transfusion dependence have been identified in multivariate analysis as pretreatment characteristics that correlate with response to immunosuppressive therapy.57 A paroxysmal nocturnal hemoglobinuria (PNH)-type phenotype has been demonstrated in nearly 20% of patients with RA. These patients also have a greater prevalence of the HLA-DR15 allele. Presence of the PNH phenotype has been associated with a greater probability of response to cyclosporine and ATG therapy.58, 59 Addition of cyclosporine and/or steroids to ATG does not appear to improve response rates any further.38 Problems with serum sickness and other toxicities may be significant with ATG and ALG, especially in older patients, emphasizing the need to select patients carefully for these treatment modalities.60
Farnesyl transferase inhibitors
Interest in the ras family of oncoproteins as a target in hematologic malignancies has been derived from the observation that mutations of these proteins are relatively common in leukemias. Activating point mutations of ras occur in up to 20% of unselected patients with MDS, but they are more frequent in patients with CMML.61 Ras proteins are at the center of a network of intracellular signaling pathways that couple signals of activated growth factor receptors to downstream mitogenic effectors, resulting in the activation of transcription factors and the expression of protooncogenes. The rate-liming step in the posttranslational activation of ras is a prenylation reaction of the carboxy-terminal consensus sequences by protein farnesyl transferases (FT) and other enzymes.
FT inhibitors (FTIs) have emerged as a new class of signaling inhibitors for many of the hematologic malignancies.62 The two orally available, nonpeptidomimetic FTIs tipifarnib (R115777) and lonafarnib (SCH66336) are the leading compounds in clinical trials of MDS and AML to date.63 In a Phase I trial of 21 patients (2 patients with RA, 7 patients with RAEB, 2 patients with RAEB-t, and 10 patients with CMML), tipifarnib was administered twice daily on a 3-week on and 1-week off schedule for 8 weeks.64 The starting dose was 300 mg orally twice daily. The dose-limiting toxicity occurred at 900 mg twice daily (fatigue and confusion). The maximum tolerated dose was 400 mg twice daily. The most frequent side effect was myelosuppression. Of 20 evaluable patients, objective responses were noted in 6 patients (30%) and included 3 patients who had hematologic improvement, 2 patients who had partial responses (PRs), and 1 patient who had a complete response (CR). Only two of the six responders had ras mutations. In a subsequent Phase II study, 28 patients with all subtypes of MDS were treated with tipifarnib at doses of 600 mg orally twice daily in cycles of 4 weeks followed by 2 weeks of rest.65 Three patients responded, including two CRs and one PR. All responders were patients with high-risk MDS. However, toxicities (myelosuppression, constitutional symptoms, neurotoxicity, and skin reactions) necessitated dose reductions or discontinuation of therapy in 11 of 27 patients (41%), suggesting a dose of 300 mg orally twice daily as more appropriate for future use.
In a Phase I/II study of 67 patients with advanced MDS (RAEB and RAEB-t; n = 32 patients) and CMML (n = 35 patients), lonafarnib was given continuously at doses of 200–200 mg orally twice daily.66 Diarrhea (26%), fatigue (17%), anorexia (12%), and nausea (9%) were the most common Grade 3 and 4 toxicities and required discontinuation of therapy in 17 patients (26%). Responses occurred in 12 of 42 evaluable patients (29%) and included a CR in 2 patients and hematologic improvement in 10 patients. Optimization of dose and schedule of the FTIs, identification of molecular targets and correlations with response, as well as combinations of FTIs with other agents will be of interest for future trials.
No responses after imatinib have been demonstrated in unselected patients with AML or MDS.67 In a small subset of patients with CMML, a reciprocal translocation transposes the PDGFRβ gene on chromosome 5q33 to various fusion partners on other chromosomes, leading to constitutive activation of the tyrosine kinase function of PDGFRβ (Table 5).34 These receptor tyrosine kinase (RTK) fusion genes have proven responsive to inhibitors of PDGFRβ activity like imatinib. In six such patients reported to date, imatinib induced durable hematologic and cytogenetic responses.68–70
Interactions between the bone marrow stromal microenvironment and the MDS clone are essential for clonal expansion and modulation of the level of apoptosis of bone marrow progenitor cells. Angiogenesis has emerged as a key biologic process, the dysregulation of which has been established as a prognostic factor not only in patients with solid tumors but also in patients with hematologic malignancies.71, 72 Abnormal secretion of several angiogenic cytokines and growth factors has been demonstrated in vitro by AML blasts. Increased microvessel density, the pathologic correlate for solid tumor vascularization in leukemias, has been shown in MDS bone marrow samples. Vascular endothelial growth factor (VEGF) has emerged as a particularly important angiogenic factor, although others (basic fibroblast growth factor [bFGF], angiogenin, TNF-α, transforming growth factor [TGF]-α, TGF-β, and angiogenin) most likely play a role as well.73 VEGF receptors (VEGFRs) are expressed on endothelial cells, hematopoietic stem cell, megarkaryocytes, and platelets. At least two types of RTKs become phosphorylated and, hence, are activated upon binding to VEGF, VEGFR-1 (Flt-1), and VEGFR-2 (Flk/Kdr). Based on a number of preclinical investigations that confirmed the pathophyiologic role of angiogenesis in leukemias and MDS, a number of small molecule inhibitors of angiogenic agents and receptors have been developed and have entered clinical trials.74
Thalidomide has significant antitumor activity in patients who have hematologic malignancies with increased bone marrow angiogenesis and has been shown to inhibit bFGF and VEGF-induced angiogenesis.75 Although thalidomide frequently is considered the prototype antiangiogenic agent, it has additional properties, including 1) immunomodulation (thalidomide stimulates CD8-positive T lymphocytes, leads to a shift from Th1 to Th2 responses, and inhibits T-cell proliferation) and 2) anticytokine effects (thalidomide inhibits production of TNF-α and other cytokines).75, 76
In what to our knowledge is the largest Phase II study published to date, 83 patients with MDS (36 patients with RA, 13 patients with RARS, 24 patients with RAEB, 6 patients with RAEB-t, and 4 patients with CMML) received oral thalidomide at doses between 100 mg and 400 mg daily.77 Among 51 evaluable patients, the majority took thalidomide at doses of between 150 mg and 200 mg daily. Doses of up to 400 mg were tolerated poorly and were maintained for only a short time. In an intent-to-treat analysis, 16 patients (19%), including 9 patients with RA, 5 patients with RARS, and 2 patients with RAEB, demonstrated hematologic improvement. Ten previously transfusion-dependent patients became transfusion-independent. Responders had lower pretherapy blast counts, higher pretherapy platelet counts, and a lower duration of pretherapy platelet transfusions. The median interval to erythroid response was 16 weeks (range, 12–20 weeks). In a multicenter National Cancer Institute-Cancer Trials Evaluation Program-sponsored trial of thalidomide in 73 patients with MDS (43 patients with IPSS scores ≤ intermediate-1 and 37 patients with IPSS scores > intermediate-2), doses ranged from 200 mg to 1000 mg daily.78 Six patients achieved hematologic improvement, and 1 patient achieved a PR, for an overall response rate of 21%. Only 32 patients could receive at least 3 courses (12 weeks of therapy), and most patients had to be taken off study because of toxicities. Although these and other smaller trials have confirmed the erythropoietic response of thalidomide, careful patient selection and choice of appropriate dose levels that can be sustained are important. A multicenter, placebo-controlled, Phase III trial of low-dose thalidomide patients with in MDS is in progress.
Analogs of thalidomide have been designed to increase efficacy and to decrease toxicity. The new compounds fall into one of two classes: immunomodulatory drugs and selective cytokine-inhibitory drugs.75, 79 CC5013 (Revimid™) is an immunomodulatory drug that currently is being tested in MDS. Compared with its parent compound, it inhibits TNF-α with a 10,000-fold increased potency. It resembles thalidomide in its ability to inhibit other cytokines, including IL-1β, IL-6, and IL-12. Costimulation of T cells is enhanced significantly. It also has been shown that CC5013 increases heterotypic adhesion of hematopoietic progenitors to bone marrow stroma, sensitizes cells to death receptor-induced apoptosis, and inhibits trophic signals to angiogenic factors in hematopoietic and stromal cells. In a Phase I–II study in 33 patients with MDS (17 patients with RA, 9 patients with RARS, 6 patients with RAEB, and 1 patient with RAEB-t), oral CC5013 was given at 1 of 3 dose schedules: 10 mg daily, 25 mg daily, and 10 mg daily for 21 days every 4 weeks.80 Twenty-nine patients (88%) reportedly showed responses. Twenty-one patients (64%) experienced an erythroid response, including major responses in 19 patients. Erythroid responses were greatest in patients with RA (82%), in patients who had low/intermediate-1 IPSS scores (71%), and in patients with a del(5q) abnormality (91%). In addition, among 17 patients who had chromosomal abnormalities, major cytogenetic responses were seen in 11 patients (65%). Myelosuppression was common but was dose-dependent, favoring the 10-mg daily dose at a 3-weeks-on/1-week-off schedule. Correlative studies demonstrated near complete suppression of plasma VEGF in treated patients and reductions of bone marrow microvessel density in responding patients. U.S. Food and Drug Administration pivotal trials with CC5013 in patients with early-stage MDS and transfusion dependence and in patients with MDS who have 5q − abnormalities are ongoing.
VEGF RTK inhibitors.
A number of small-molecule inhibitors of VEGF RTK inhibitors have been developed. The Sugen compounds SU5416 and SU11248 have been disappointing, with low response rates and, in some cases, dose-limiting nonhematologic organ toxicities that precluded further development of the latter compound.81, 82 PTK787 is an orally available inhibitor of VEGFR kinases.83 In a Phase I study of 19 patients with advanced MDS (n = 12 patients) and poor-prognosis AML (n = 7 patients), 2 of 12 patients with MDS had stable disease, but no objective responses were observed. Dose-limiting toxicities at doses of 1000 mg twice daily were nausea and emesis.
Other agents in development.
Bevacizumab is a recombinant, anti-VEGF, humanized, monoclonal antibody. It has been shown that antibody neutralization of VEGF also inhibits the bone marrow production of TNF-α, a major inhibitory cytokine that contributes to ineffective hematopoiesis in MDS. In a Phase II study in MDS, bevacizumab was given intravenously at a dose of 10 mg/kg every 2 weeks for 4 months with the possibility of escalating the dose to 15 mg/kg for an additional 2 months.84 Preliminary results in 10 of 15 patients (7 patients with RA, 3 patients with RARS, 4 patients with RAEB, and 1 patient with CMML) showed major erythroid responses in 2 patients that did not occur until after at least 4 months of therapy. Notable toxicities to date have included hypertension and fatigue. Three patients had to be taken off that study early because of cyotpenias or disease progression.
Matrix metalloproteinases (MMPs) are a family of zinc-dependent endopeptidases that degrade structural components of the extracellular matrix and, thus, disrupt pathways of stromal adhesion signals to hematopoietic progenitors as well. Limited experience is available for the orally available MMP inhibitor AG3340.85
TNF-α is a pleiotropic cytokine with potent hematopoietic inhibitory activity. It is composed of 2 subunits, p55 and p75, of which p55 transduces the inhibitory effects of TNF-α on hematopoietic stem cells.86 TNF-α levels are elevated in patients with early-stage MDS and has become a target for novel therapy approaches.87
Pentoxifylline-based combinations (ciprofloxacin, ciprofloxacin with amifostine and dexamethasone) have been ineffective.88 Success with the soluble, recombinant TNF-receptor fusion protein etanercept has been modest in MDS, although it has excellent activity in inflammatory arthropathies.89–91 Infliximab is a chimeric monoclonal antibody that neutralizes TNF-α by inhibiting its binding to both p55 and p75 receptors. Experience of infliximab in MDS is limited, although hematologic improvement after therapy has been noted in up to 20% of patients treated.92, 93
Arsenic trioxide (As2O3) has multiple mechanisms of action, including induction of apoptosis and tumor cell differentiation, abrogation of cellular proliferation, and inhibition of angiogenesis. The impact of As2O3 on multiple tumor pathways has spurned its investigation in other hematologic malignancies, including MDS.94
In a Phase II study in the U.S., arsenic trioxide was administered at an intravenous dose of 0.25 mg/kg daily for 5 days per week for 2 weeks, followed by 2 weeks off.95 Efficacy and safety data have been reported for 40 patients (6 patients with RA, 8 patients with RARS, 14 patients with RAEB, 10 patients with RAEB-t, and 2 patients with CMML). Hematologic responses were observed in 7 of 28 evaluable patients (25%), including 4 major and 2 minor erythroid responses and 1 major platelet response. Responses were seen first 2–6 months after the start of treatment. As2O3 was tolerated well. No QT cycle length (QTc) prolongations have been reported. A different dose schedule was used in a European Phase I–II study: A 1-hour intravenous infusion loading dose of 0.30 mg/kg daily for 5 days was followed by maintenance with 0.2 mg/kg daily twice per week for > 15 weeks.96 Data on 62 patients (6 patients with RA, 6 patients with RARS, 38 patients with RAEB, 10 patients with RAEB-t, and 2 patients with CMML) have been reported. Hematologic responses occurred in 13 of 50 evaluable patients (26%) and took approximately 2–6 months to develop. The responses included eight major erythroid responses. Four of 13 patients became transfusion-independent while on study. QTc prolongation was noted in one patient. Alternative dosing schedules of As2O3, combinations of As2O3 with other agents (e.g., thalidomide), and novel preparations (such as oral tetra arsenic tetrasulfide; As4S4) are being explored.
Global and gene-specific DNA methylation is frequent in many solid tumors and hematologic cancers, including MDS, in which the cell cycle inhibitor p15INK4b frequently is hypermethylated.97, 98 The association of aberrant DNA methylation with tumor resistance and disease progression has generated interest in the use of agents that possibly may restore normal gene transcription by reversing hypermethylation.
To our knowledge, the majority of experience to date has been accumulated with the DNA methyltransferase inhibitors 5-azacytidine [5-Azacytidine (Vidaza®) has been approved by the FDA for therapy in MDS on May 20, 2004] and 5-aza-2′-deoxycytidine (decitabine). In a Phase III study, the Cancer and Leukemia Group B compared observation (n = 92 patients) with 5-azacytidine 75 mg/m2 subcutaneously daily for 7 days every 4 weeks (n = 99 patients) in patients with MDS.99 Responses occurred in 63% of 5-azacytidine-treated patients (CR, 6%; PR, 10%; hematologic improvement, 47%) compared with 7% in the observation arm. The median time to leukemia transformation (21 months vs. 13 months; P < 0.01), median survival (20 months vs. 14 months; P = 0.10), and quality of life were superior in patients who were treated with 5-azacytidine.99, 100
5-Azacytidine is a prodrug for decitabine, because it undergoes conversion to decitabine before DNA incorporation and irreversible binding to DNA methyltransferases. Decitabine is a more potent hypomethylating agent than 5-azacytidine.101 It has demonstrated activity against a number of myeloid malignancies at higher doses but was associated with severe and prolonged myelosuppression.102, 103 In vitro data have suggested that decitabine exerts a dual effect on normal and neoplastic cells. At higher doses, it is cytotoxic; at lower doses, its cytotoxicity is minimal, and the treated cells exhibit marked reduction of DNA methyltransferase activity, reduced overall and gene-specific DNA methylation, and reactivation of silenced genes, including tumor-suppressor genes. In a multicenter, Phase II study, Wijermans et al. treated 66 elderly patients (median age, 66 years) with decitabine 45 mg/m2 daily (15mg/m2 intravenously over 4 hours every 8 hours) for 3 days every 6 weeks.104 Twenty-five patients had high IPSS risk scores. Overall, 49% of patients responded (CR, 20%; PR, 4%; hematologic improvement, 24%), and the response rate was 64% in the high-risk group. The median response duration was 7.4 months, and survival was 22 months from the time of diagnosis and 15 months from the start of therapy. Myelosuppression was common, and the treatment-related mortality rate was 7%. Among 53 patients who had received at least 1 course of decitabine, 35 patients (66%) experienced increased platelet counts, which exceeded 100 × 109/L in 19 patients (36%). In a recent update of 169 patients (median age, 70 years; 71 patients in the high-risk group according to IPSS scores) who received decitabine at doses of 45–50 mg/m2 daily for 3 days every 6 weeks, the overall response rate was 49% (51% among high-risk patients).105 The median response duration was 9.5 months, and the median survival was 15 months. Responses were independent of age. Hospitalization and quality of life were improved.105
Methylation status of p15 and p16 prior to and after therapy with decitabine was analyzed.106 Hypermethylation of the p15 gene was detected in 15 of 23 patients (65%), whereas the p16 region was unmethylated in all samples. Among 12 patients with p15 hypermethylation, demethylated epigenotypes and reestablishment of normal p15 protein expression were observed in 9 patients (75%). However, responses to decitabine also occurred without p15 hypermethylation, suggesting that p15 is only 1 molecular target for demethylation or that decitabine may induce anti-MDS activity through mechanisms other than hypomethylation.
Issa et al. studied a lower dose, prolonged exposure schedule of decitabine, trying to establish a minimally biologic effective dose of decitabine, thus avoiding cytotoxicity and myelosuppression.107 A dose of 15 mg/m2 intravenously over 1 hour daily for 10 days induced most responses with no additional benefit from further increases in dose or treatment duration. Among 50 patients with recurrent and refractory hematologic malignancies, responses occurred in 32% of patients (CRs, 18%; objective responses, 14%). Among patients with MDS, 4 of 7 patients who were treated (57%) achieved objective responses, including a CR and hematologic improvement in one patient each. Some responses were slow to occur, with eventual recovery of normal hematopoiesis at 5–6 weeks, suggesting a noncytotoxic mode of action.
Silencing of gene transcription is a complex process of chromatin remodeling. Deacetylation of histones has been associated with transcriptionally inactive chromatin. Various histone deacetylase inhibitors currently are under investigation in clinical trials, including suberoylanilide hydroxamic acid, depsipeptide (FK228), MS-275, and valproic acid.108, 109 Future trials will focus on confirming the optimal dose and schedule of DNA methyltransferase inhibitors; combination studies of these agents with histone deacetylase inhibitors; and other combinations, such as with retinoids.
Other Novel Agents
Other novel agents that are undergoing clinical trials in MDS include antisense oligonucleotides to inactivate p53 RNA; the plant alkaloid homoharringtonine, which has activity in AML and chronic myeloid leukemia; small-molecule inhibitors of fms-like tyrosine kinase 3, including the proteasome inhibitor PS-341 (bortezomib) and TLK199 (Table 6).110–113 TLK199 is a novel glutathione analog that has demonstrated in vitro myelostimulant activity in human bone marrow cultures. In addition to the inhibition of glutathione S-transferases, it has been shown that TLK199 activates intracellular pathways of the microtubule-associated protein kinase cascade.114 In a Phase I dose-escalation study, TLKL199 was administered as an intravenous infusion over 60 minutes daily (50–400 mg/m2) × 5 days every 2 weeks for up to 8 cycles.115 Four of eight evaluable patients in that study showed hematologic improvements (one patient with RA, two patients with RAEB, and one patient with RAEB-t). Six of eight patients experienced an increase in bone marrow progenitor colonies after therapy.
|Class of drug||Examples|
|Thalidomide analogs: IMIDs||CC5013 (Revimid)|
|FTIs||R115777 (Tipifarnib), SCH66336 (Lonafarnib)|
|Histone deacetylase inhibitors||SAHA, valproic acid, MS275|
|Proteasome inhibitors||PS-341 (Bortezomib)|
|Anti-VEGF monoclonal antibodies||Bevacizumab (Avastin)|
|VEGFR tyrosine kinase inhibitors||SU5416, PTK787|
|Soluble recombinant TNF receptor fusion protein||Etanercept|
|Anti-TNF monoclonal antibody||Infliximab|
|Flt-3 inhibitors||CEP701, PKC412|
|Glutathione analog inhibitors of GST||TLK199|
Hematopoietic stem cell transplantation (SCT) and AML-type chemotherapy are the only modalities to date that have been associated with a measurable cure fraction in patients with MDS. However, they have an attendant increased risk of treatment-related mortality and morbidity. Thus, they usually are reserved for patients with high-risk MDS, because their prognosis is as unfavorable as in AML. Age and performance status have to be considered in the treatment decision. High-intensity therapy is most appropriate for patients age < 60 years and who have a good performance status.
Intensive chemotherapy in MDS produces remission rates from 40% to 60%.116–119 With appropriate supportive care measures and prophylactic antibiotics, the mortality rate generally is below 10%. Factors that have a positive influence on outcomes with intensive chemotherapy include age < 50 years, normal karyotype, and FAB diagnosis of RAEB-t. Although remission rates are comparable to the results in AML, CR duration is short, and < 10% of patients can be expected to be alive and disease free beyond 2 years. In a multivariate analysis, Beran et al. analyzed the outcome of 394 newly diagnosed patients with high-risk MDS who were treated with various chemotherapy combinations (idararubicin plus cytarabine, fludarabine plus cytarabine, fludarabine plus cytarabine and idarubicin, topotecan plus cytarabine, and cyclophosphamide plus cytarabine and topotecan).120 The overall CR rate was 58%. Response was associated with karyotype, performance status, age, duration of antecedent hematologic disorder, and treatment in laminar air-flow rooms. There was no difference in the CR rate based on the regimen used. Topotecan-based regimens had the lowest induction mortality, especially in patients age > 65 years. The topotecan plus cytarabine regimen also was particularly safe and effective in patients with RAEB. Overall survival was similar to that of patients who wee treated with idarubicin and cytarabine. Thus, improvements in outcome may come not from more intensified therapy but from innovative postremission strategies (targeted therapies, immunomodulation, anticytokines), new drugs (e.g., novel nucleoside analogs, such as clofarabine or troxacitabine; novel topoisomerase I inhibitors, such as 9-aminocamptothecin, 9-nitocamptothecin, and DX8951F), new formulations of older drugs (liposomal or polyethylene glycol preparations), or different dose schedules.
Stem cell transplantation
SCT remains applicable to only a small subset of patients with MDS because of age restrictions, concomitant medical conditions, and donor availability. Results from several large centers indicated disease-free survival rates of 30–50%.121, 122 Results were better in younger patients, in patients with low-risk MDS, and in patients who underwent transplantation within 1 year from the date of diagnosis. Failure was due primarily to transplantation-associated mortality in patients with low-risk MDS and to disease recurrence in patients with high-risk MDS. In patients with MDS who underwent allogeneic SCT, long-term follow-up studies showed 3-year survival rates of 23%, similar to patients who received intensive chemotherapy. In an update of the International Bone Marrow Transplant Registry, 452 patients who underwent HLA-identical sibling transplantations were reviewed.123 The median patient age was 38 years, and most patients (60%) had high-risk MDS. Overall survival at 3 years was 42%. According to that review, survival was more favorable among younger patients and among patients who had platelet counts > 100 × 109/L. The incidence of recurrence was higher in patients who had high percentages of bone marrow blasts at the time of transplantation, high IPSS scores, and T-cell depleted transplantations. When patients were evaluated according to IPSS scores, the diseases-free survival rate was 60% in the low-risk group, 36% in the intermediate-1 risk group, and 28% in the intermediate-2 risk group.121 This was compared with 5-year survival rates of 55%, 35%, and 7%, respectively, for unselected patients who did not undergo SCT. Thus, SCT mostly benefited patients with high-risk MDS, but the appropriate timing and optimal bone marrow ablation regimen remain disputed.
Results from matched, unrelated donor programs still lag behind the results from matched, related sibling transplantation programs. In an update from the National Marrow Donor Program in MDS, the 2-year survival rate was 29%, the transplantation-related mortality rate was 54%, and the recurrence probability was 14%.124 Better survival rates (3-year recurrence-free survival rate, 59%) have been reported by the Seattle group using conditioning regimens with targeted busulfan and cyclophosphamide.125
Recently, several groups reported on their experience with nonmyeloablative SCT (NST).126–130 Although differences in patient populations, preparative regimens, and graft-versus-host disease prophylaxis as well as donor source (related vs. unrelated) have to be considered, overall survival of up to 30% at 2 years and disease free survival rates of 55% at 1 year have been achieved in most studies. The main cause for treatment failure in patients who underwent NST was disease recurrence, which was greater compared with patients who underwent allogeneic SCT, especially in patients who had advanced-stage disease. Overall, the preliminary experience suggests that NST may become a valuable alternative for older patients or for patients who are at high risk for complications after undergoing standard allogeneic SCT.
Autologous SCT (ASCT) is limited to patients who have achieved a CR, can be harvested, and are candidates for the procedure. ASCT after successful induction chemotherapy may increase the proportion of long-term survivors, thus improving CR duration in some patients with MDS, particularly in younger patients in remission. Results for older patients and for transplantation patients who are in recurrence or subsequent remissions are unsatisfactory. The recurrence rate remains high.131, 132
- Top of page
- CLASSIFICATION AND PROGNOSTIC FACTORS
- BIOLOGY OF MDS
Therapeutic nihilism in MDS no longer is justified. Many novel and targeted agents have been introduced into clinical trials in MDS in recent years, some with encouraging results. Further development of rational therapies and improvements in the outcome of patients with MDS eventually will be possible only through a comprehensive understanding of the pathophysiology of the disease. Further attempts at classification and identification of risk groups will become more relevant clinically as we gain a better understanding of the biology of this heterogeneous disorder. In particular, improving the MDS classification may help identify more appropriate therapies for low-risk versus high-risk groups and a better definition of the status of SCT.
- Top of page
- CLASSIFICATION AND PROGNOSTIC FACTORS
- BIOLOGY OF MDS
- 7Atomic bomb radiation increases the risk of MDS. Leuk Res. 2001; 25: S13., , , et al.
- 42[No authors listed.] NCCN practice guidelines for the myelodysplastic syndromes. National Comprehensive Cancer Network. Oncology (Huntingt). 1998; 12(11A): 53–80.
- 46A validated decision model for treating the anaemia of myelodysplastic syndromes with erythropoietin + granulocyte colony-stimulating factor: significant effects on quality of life. Br J Haematol. 2003; 120: 1037–1046., , , et al.
- 53Pegylated recombinant human megakaryocyte growth and development factor (PEG-rHuMDGF) increased platelet counts (plt) in patients (pts) with aplastic anemia (AA) and myelodysplastic syndrome (MDS). Blood. 2000; 96: 296a., , , et al.
- 55ATG and cyclosporin A as combination therapy in myelodysplastic syndromes RA and RAEB. Blood. 2002; 100: 98a., , , et al.
- 56Antithymocyte globulin (ATG)-based therapy in patients with myelodysplastic syndromes (MDS). Blood. 2002; 100: 794a., , , et al.
- 66Continuous oral lonafarnib (Sarasar™) for the treatment of patients with myelodysplastic syndrome. Blood. 2003; 102: 421a., , , et al.
- 70Response to STI571 in chronic myelomonocytic leukemia with platelet derived growth factor beta receptor involvement: a new case report. Haematologica. 2003; 88: ECR18., , , et al.
- 78N998B: multicenter Phase II trial of thalidomide (Thal) in adult patients with myelodysplastic syndromes (MDS). Blood. 2002; 100: 96a., , , et al.
- 80Efficacy and safety of CC5013 for treatment of anemia in patients with myelodysplastic syndromes (MDS). Blood. 2003; 102: 184a., , , et al.
- 82A Phase I study of repeated oral dosing with SU11248 for the treatment of patients with acute myeloid leukemia who have failed or are not eligible for conventional chemotherapy. Blood. 2002; 100: 558a., , , et al.
- 83Phase I trial of PTK787/ZK222584, an inhibitor of vascular endothelial growth factor receptor tyrosine kinases, in acute myeloid leukemia and myelodysplastic syndrome. Blood. 2003; 102: 361a., , , et al.
- 84Phase II study of bevacizumab (anti-VEGF humanized monoclonal antibody) in patients with myelodysplastic syndrome (MDS): preliminary results. Blood. 2003; 102: 425a., , , et al.
- 85Randomized, double-blind Phase II study of the matrix metalloproteinase (MMP) inhibitor, AG3340 (Prinomastat™) in patients with myelodysplastic syndrome. Blood. 2002; 100: 789a., , , et al.
- 92Hematologic improvement in response to anti-tumor necrosis factor (TNF) therapy with Remicade® in patients with myelodysplastic syndrome (MDS). Blood. 2003; 100: 795a., , , et al.
- 95Trisenox® (arsenic trioxide) in patients with myelodysplastic syndromes (MDS): preliminary findings in a Phase 2 clinical study. Blood. 2003; 102: 423a., , , et al.
- 96Trisenox® (arsenic trioxide) in patients with myelodysplastic syndromes (MDS): preliminary results of a Phase 1/2 study. Blood. 2003; 102: 422a., , , et al.
- 105Low dose decitabine for elderly high risk MDS patients: who will respond? Blood. 2002; 100: 96a., , .
- 108Valproic acid alone or in combination with All-trans-Retinoic Acid (ATRA) for the treatment of myelodysplastic syndromes and sAML/MDS. Blood. 2003; 102: 428a., , , et al.
- 109Phase I trial of the histone deacetylase inhibitor depsipeptide (FR901228) in acute myeloid leukemia. Blood. 2002; 100: 86a., , , et al.
- 111A “first in man” study of the safety and PK/PD of an oral FLT3 inhibitor (MLN518) in patients with AML or high risk myelodysplasia. Blood. 2002; 100: 336a., , , et al.
- 112Phase I study of bortezomib (PS-341), VELCADE™), a proteasome inhibitor, in patients with refractory or relapsed acute leukemias and myelodysplastic syndromes. Blood. 2002; 100: 560a., , , et al.
- 113A randomized Phase II trial of the tyrosine kinase inhibitor PKC412 in patients (pts) with acute myeloid leukemia (AML)/high-risk myelodysplastic syndromes (MDS) characterized by wild-type (wt) or mutated FLT3. Blood. 2003; 102: 614a., , , et al.
- 114[No authors listed.] Glutathione based approaches to improving cancer treatment. Chem Biol Interact. 1998; 24: 111–112.
- 115Hematologic improvement following treatment with TLK199 (a novel glutathione analog inhibitor of GST P1-1) in myelodysplastic syndrome (MDS): interim results of a Phase 1–2a study. Blood. 2003; 102: 426a., , , et al.
- 116Intensive chemotherapy for poor prognosis myelodysplasia (MDS) and secondary acute myeloid leukemia (sAML) following MDS of more than 6 months duration. A pilot study by the Leukemia Cooperative Group of the European Organization for Research and Treatment in Cancer (EORTC-LCG). Leukemia. 1995; 9: 1805–1811., , , et al.
- 119Topotecan, cytosine arabinoside and G-CSF (TAG) versus idarubicin, cytosine arabinoside and G-CSF (IDAG) in patients with myelodysplastic syndrome (MDS) or MDS in transformation: a randomized Phase III Study. Blood. 2002; 100: 98a., , , et al.
- 122Bone marrow transplantation from HLA-identical siblings as first-line treatment in patients with myelodysplastic syndromes: early transplantation is associated with improved outcome. Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation. Bone Marrow Transplant. 1998; 21: 255–261., , , et al.
- 123Bone marrow transplantation from HLA-identical siblings as treatment for myelodysplasia. Blood. 2002; 200: 1997–2004., , , et al.
- 126Reduced-intensity allogeneic haematopoietic stem cell transplantation (HSCT) in 75 patients with myelodysplastic syndromes following conditioning with fludarabine, busulphan and alemtuzumab. Blood. 2003; 102: 78a., , , et al.
- 127AML and MDS treated with nonmyeloablative stem cell transplantation: overall and progression free survival comparable to myeloablative transplantation. Blood. 2003; 102: 79a., , , et al.
- 128Efficacy of non-myeloablative allogeneic transplant for patients with myelodysplastic syndrome (MDS) and myeloproliferative disorders (MPD) (except chronic myelogenous leukemia). Blood. 2003; 102: 185a., , , et al.
- 129Reduced-intensity conditioning (RIC) for allogeneic hematopoietic stem cell transplantation (HSCT) from HLA-identical siblings in adults with myelodysplastic syndromes (MDS): a comparison with standard myeloablative conditioning: a study of the EBMT-Chronic Leukemia Working Party (EBMT-CLWP). Blood. 2003; 102: 184a., , , et al.
- 130Morbidity and mortality with nonmyeloablative compared to myeloablative conditioning before hematopoietic cell transplantation from HLA matched related donors. Blood. 2003; 102: 77a., , , et al.
- 131Autologous stem cell transplantation for treatment related MDS/AML. A report of the Chronic Leukemia Working Party (CLWP) of the EBMT. Blood. 2003; 102: 43a., , ,et al.
- 132The impact of post-remission therapy on outcome of patients with poor-risk MDS and secondary AML treated with intensive chemotherapy with or without stem cell transplantation in a Joint Study (CRIANT) of the EORTC, EBMT, SAK, HOVON, Scandinavian, and the Gimema Leukemia Groups. Blood. 2003; 102: 185a., , , et al.