Advances in the Management of Myelodysplastic Syndrome Proceedings from ASH 2004

Advances in the Management of Myelodysplastic Syndrome: Proceedings from ASH 2004

Advances in the Management of Myelodysplastic Syndrome: Proceedings from ASH 2004December 4-7, 2004San Diego, CAHagop M. Kantarjian, M.D., University of Texas MD Anderson Cancer Center


Myelodysplastic syndrome (MDS) has evolved from a disease for which no treatment options were available to one where the supportive growth factors erythropoietin (EPO) and granulocyte colony stimulating factors (G-CSF) are improving quality of life, and one agent, 5-azacytidine (Vidaza), has modified the course of MDS. 5-azacytidine is now the first FDA-approved treatment for MDS. The availability of these options has resulted in an increase in the documented incidence of MDS from 12,000 cases per year in the U.S. to more than 20,000 cases. This is because more bone marrow studies are now performed in elderly patients with mild cytopenias to reach a precise diagnosis, which may document MDS and allow early interventions. Of several proposed classifications (FAB, WHO, IPSS)

[1],[2],[3]the International Prognostic Scoring System is now commonly used. It divides patients, based on the percent of marrow blasts, karyotype, and degree of cytopenias into low, intermediate 1, intermediate 2 and high-risk groups, with median survival of 5.6, 3.1, 1.2 and 0.4 years, respectively. (Tables 1-3)1 Of note, the IPSS excludes CMML and secondary MDS, and applies only to newly diagnosed MDS. When previously treated MDS cases, referred to tertiary centers following failure of standard options, are considered, patient outcomes, even within lower risk groups are worse. (Table 3)


Table 1: Scoring System Classification

* Good, Normal, -Y, del(5q), del(20q); Poor, complex (> 3 abnormalities) or chromosome 7 anomalies; Intermediate, other abnormalities.

  • Definitions of cytopenias: hemoglobin < 10 g/dl, granulocytes < 1.5 x 109/L, platelets < 100 x 109/L

Table 2: International Prognostic Scoring System Classification

*IPSS risk applies to only to newly diagnosed MDS

Table 3: Prognosis using IPSS by Prior Therapy

In general, lower risk MDS refers to the RA and RA-RS FAB categories, or to low and intermediate 1 IPSS risk groups. Higher risk MDS refers to RAEB and RAEBT FAB categories or to Intermediate-1 and high-risk IPSS groups. A simple categorization considers high-risk MDS when the percent of blasts is > 10%. Lower risk MDS constitutes 60-70% of newly diagnosed MDS and 40-50% of treated MDS referred to specialized centers. In them, the initial approaches include observation; transfusions; growth factors; immune therapies (e.g. ATG, steroids, cyclosporine), particularly in HLADr2/15 positive or CD55/CD59 negative cases; 5-azacytidine (more for intermediate 1 risk or worse), or low intensity investigational strategies including hypomethylating agents (e.g. Dacogen), histone deacetylate inhibitors (e.g. valproic acid, depsipeptide, SAHA), farnesyl transferase inhibitors (FTI; e.g. tipifarnib, lonafarnib), and angiogenesis TNFa/cytokine suppression (e.g. thalidomide, CC5013 [Revlimid]). Revlimid, an Immunomodulatory Inhibitory Derivative (IMID) of thalidomide, appears to benefit transfusion dependent low-risk MDS or low-risk MDS with chromosome 5q31 abnormalities (5q syndrome, other 5q31 disorders).

This report summarizes the recent advances in the management of MDS as highlighted at the American Society of Hematology (ASH) December 2004 meeting, as well as in recently published studies. Some of the data presented include the ASH meeting updates of the published abstracts.

Growth Factors

Aranesp® (darbepoetin alfa) is a long-acting erythropoiesis-stimulating agent that has an increased serum half-life compared to epoetin alfa. At ASH, Mannone, et al. updated their results in patients with lower risk MDS treated with Aranesp® (glycosylated form of EPO, longer T1/2) 300 mcg weekly x 12; GCSF 150 mcg TIW was added after 4 months, for another 4 months, if no response to Aranesp® was observed.

[5]Eligibility required marrow blasts < 10%, hemoglobin < 10 g/dl and erythropoietin levels < 500 mu/ml. Fifty-three patients were evaluable. The median age was 77 years. FAB diagnosis was RA in 17, RARS in 20, RAEB 1 (< 10% blasts) in 14, and CMML in 2. Only 5% had unfavorable cytogenetics. By IPSS, 38% were low, and 51% intermediate-1 risk. Overall, 34 of 53 patients (64%) had erythroid response, which was major in 26. Among 8 patients in whom GCSF was added to Aranesp®, 2 (25%) responded. Response by FAB was: RA 67%, RA-RS 42%, and RAEB 1 67%. This study suggests a high level of erythroid response with darbopoetin in lower risk MDS.

In a study by Miller et al., 105 patients with RA (n = 40), RA-RS (n = 36), or RAEB (n= 29) were randomized to 1) EPO 150 mcg/kg daily with later addition of G-CSF 1 mcg/kg daily if no response, versus 2) supportive care, with crossover to EPO allowed after 4 months if transfusion needs increased by 50% or more. < /SPAN> [6](Table 4) The overall response rate was 35% in 53 patients receiving EPO/GCSF, and 9% in 56 patients on supportive care. Response rates were: 1) 30% among 23 patients who crossed over from supportive care to EPO; 2) 22% among 27 patients in whom GCSF was added to EPO; and 3) 60% among 10 patients in whom EPO was increased from 150 to 300 mcg/kg. The median survival was 53 months for responders and 26 months for non-responders (p = 0.009).

Table 4: Erythropoietin alone or with GCSF versus supportive care in MDS

Hypomethylating Agents and Histone Deacetylase Inhibitors

In a randomized study in MDS, 5-azacytidine (Vidaza) had previously shown superiority to supportive care, with significantly higher CR (7% vs. 0%) and overall response rates (63% vs. 7%), longer time to transformation or death (median 22 versus 12 months; p < 0.01), a trend for better survival (median 18 versus 14 months, P = 0.1), particularly when compared to patients who did not cross over from supportive care to 5-azacytidine (by landmark analysis).< /SPAN > [7]Of note, 35% of patients had intermediate 2 or high-risk MDS; 15% had prior therapies; and the median MDS duration was 2 months. The median number of 5-azaycytidine courses was 7. 5-azacytidine was also reported to improve quality of life, and to improve transfusion requirements in about 30% of patients (Vidaza brochure). This resulted in its FDA approval in May 2004 for the treatment of MDS. Of interest, Shadduck et al. treated 15 patients with AML with 5-azacytidine 75 mg/m2 SQ daily X 7 every 4 weeks.

[8]Their median age was 68 years (range 44 to 80 years). Overall, 4 achieved CR and 4 had PR, for an overall response in 8/15 patients (53%). As with MDS, the median time to CR was 3 courses (range 2 to 5), the median duration of response was 8 months, and the median survival with stable disease or better response was 14+ months.

Dacogen (decitabine) is another hypomethylating agent with activity in MDS, AML, CML and sickle cell disease. In a randomized study by Saba, et al., 170 patients with MDS were randomized to Dacogen 15 mg/m2 over 3 hours Q 8 hours x 3 days (45 mg/m2 daily x 3 = 135 mg/m2 courses) every 6 weeks for up to 10 cycles (n = 89), or supportive care (n = 81). < /SPAN>[9]By IPSS, 70% had intermediate 2 or high-risk MDS; 27% had prior therapy; and 14% had secondary MDS. The median MDS duration was 6 months. Unfortunately, the median number of courses delivered was only 3. Overall, Dacogen resulted in a higher overall response rate (CR + PR 17% versus 0%; p < 0.001), a trend for longer time to event (AML or death) overall, significantly longer times to event in intermediate 2-high-risk MDS and in treatment-naïve patients, and in significantly improved transfusion independence rates. (Table 5) Transfusion independence was reduced from 70% pretreatment with Dacogen to 22% post 6 cycles. In the supportive care group, transfusion dependence was 59% pretreatment and 57% 6-9 months later.

Table 5: Dacogen versus supportive care in MDS

Recognizing the limitations of the prior study in relation to number of courses delivered, timely and frequent delivery of Dacogen therapy prior to judging a response, and the schedules used, Kantarjian et al. conducted a 3-arm randomized study in higher risk MDS, in an attempt to improve the results.

[10]All patients received the same dose per course (Dacogen 100 mg/m2/course), but were randomized, initially equally (1:1:1), and later (post Patient 60), in a Bayesian design, to Dacogen: 1) 20 mg/m2 IV over 1 hour daily x 5, 2) 10 mg/m2 SQ BID x 5 days, or 3) 10 mg/m2 IV over 1 hour daily x 10. Key elements to the study were: 1) delivery of courses every 4 weeks regardless of counts as long as there was persistent disease and no significant toxicities (courses were delayed only if there was no evidence of disease in a hypoplastic marrow or if severe, usually myelosuppression-associated complications [e.g. infections or bleeding] occurred); 2) allowing for time to recovery only after every 3 courses; and 3) persisting with therapy for at least 3 courses before evaluating response. Fifty-three patients have so far been treated. Their median age was 64 years (range 39-90). Significant cytopenias prior to therapy were present in 45% to 81%. By IPSS, 32% had intermediate 1, 33% intermediate 2, 19% high risk, and 15% had CMML. Unfavorable cytogenetics were present in 57%, and secondary MDS in 21%; 90% had prior therapies. Overall, among 52 patients currently evaluable, 18 patients achieved CR (35%), 4 had PR and 43/52 patients (83%) improved by the International Working Group response criteria (only major responses included). No serious extramedullary drug-related events occurred. In particular, there were no severe incidences of nausea or vomiting and no severe local inflammation, reaction, or pain with the subcutaneous schedule. Twenty-five percent of courses resulted in hospitalizations for myelosuppression-associated events (fever, infection). There were only 2 deaths (4%) in the first 3 months of therapy. Compared with a historical group of patients with MDS receiving intensive chemotherapy (2000 2004), Dacogen has so far been associated with better survival (p = 0.02), probably because of the lower incidence of early (6-week) mortality (2% versus 21%). The study is ongoing and will accrue a total of 90 patients.

Kuendgen, et al. treated 18 patients with MDS with valproic acid (to target serum concentrations of 50-100 mcg/ml), with or without ATRA.

[11]They noted 1 PR, 4 erythroid hematologic improvements (2 major, 2 minor), 1 major neutrophil response, and 2 major platelet responses, for a total response in 8/18 patients (40%). Combinations of hypomethylating agents with histone deacetylase inhibitors (e.g. Dacogen + valproic acid) are ongoing.


Farnesyl Transferase Inhibitors (FTIS)

Following several encouraging pilot trials of tipifarnib (Zarnestra; R115777) in MDS, Kuzrock, et al. conducted a multi-institutional study of tipifarnib 300 mg orally BID x 3 weeks every 4 weeks.

[13]In the ASH update, 82 patients were reported, with a median age of 67 years. By IPSS, 17% had intermediate 1, 39% had intermediate 2, and 44% had high-risk MDS. Thirty-seven percent had prior therapies. Overall, 7 patients achieved CR, 4 had CR with low platelets, 2 had PR, and 13 had hematologic improvement, for an overall response in 26/82 patients (32%). The median time to AML transformation was 14.1 months and to AML or death was 6.4 months. The median survival was 11.9 months. Severe side effects included myelosuppression in 24-37%, and fatigue, nausea, diarrhea or rashes in less than 5%.

Thalidomide and Lenalidomide (Revlimid, CC5013)

Thalidomide has previously shown modest activity in MDS, with a response rate of 10% to 20%. Bouscary, et al. updated their experience using thalidomide 100 to 400 mg orally daily in 82 patients with MDS.

[14]Twenty-one patients had low, 44 had intermediate 1, 7 had intermediate 2, and 7 had unknown IPSS risk. After 3 months, 47 of 72 patients continued therapy for 3 months and only one (2%) had a response. Thirty-three patients continued therapy beyond 3 months: among them, 6 had a major erythroid response and 8 a minor erythroid response, for a response rate of 28% (14 of 50 evaluable). Neutrophil and platelet responses were uncommon. The authors conclude that thalidomide should be used in low doses (50 to 100 mg daily) and only in patients with low-risk MDS/refractory anemia with red cell transfusion dependence.

The data of Lenalidomide (Revlimid) has been updated in several meetings with favorable results. In the original pilot study by List, et al, 45 patients with lower risk MDS received Lenalidomide 10-25 mg orally daily.

[15]Responses, mostly erythroid, were encouraging. (Table 6)

Table 6: Response to Revlimid in low risk MDS

6A: Hematologic Responses

6B: Cytogenetic Responses

Erythroid response was noted in 24 patients (53%; 67% of 36 evaluable patients), and was major in 21 patients. Rates of major platelet response was 17% (1/6 with pretreatment platelets < 100 x 109/L) and of granulocyte response 20% (2/10 with pretreatment granulocytes < 109/L). Impressively, 10 of 11 (91%) patients with 5q31 abnormality had erythroid response, and 9 of them (82%) had a complete cytogenetic response. Responses were durable (48+ weeks; range 13+ to 101+ weeks). This resulted in two FDA pivotal trials of Revlimid 10 mg orally daily, later daily x 21 every month, in a) low risk MDS with transfusion dependence (n=215), and b) low risk MDS with 5q abnormality with transfusion dependence (n=148). The results, updated at the ASH 2004 MDS educational session, reported the following in the low risk MDS with 5q abnormality: 1) confirmation of eligibility in three-quarters of patients, 2) erythroid response in two-thirds of patients, 3) better response rates in patients with isolated 5q abnormality 4) cytogenetic responses in 75%, which were complete in 50%, and 5) durability of hematologic responses.15 Final updates of these encouraging findings may result in an FDA approval of Revlimid for low risk MDS with 5q abnormality and transfusion dependence.

Intensive Chemotherapy and Allogeneic Transplant

Intensive chemotherapy is associated with CR rates of 40% to 60% in high-risk MDS, but also with a mortality rate of 20%. Response duration of survival remain poor. At ASH 2004, Knipp et al updated their results with intensive chemotherapy (idarubicin + ara-C or TAD induction followed by 1 consolidation or 2-year maintenance) in 158 patients aged > 60 years who had AML or high-risk MDS.

[16]The median age was 67 years (range 60 to 78 years). Cytogenetics were available in 144 patients: 76 had normal karyotype and 60 had an abnormal karyotype, which was complex in 31. Overall, CR was achieved in 94/158 patients (60%) and there were 16 induction deaths (10%). The CR rates were 65% with a normal karyotype and 59% with non-complex karyotype, versus 39% with a complex karyotype (p = 0.001). The median survivals were 18, 6, and 4 months, respectively (p < 0.001). Twenty additional patients died in CR from infections during consolidation, resulting in an overall death rate of 23% (36 of 158 patients) during induction-consolidations. The authors conclude that alternative, non-intensive strategies may be indicated in elderly patients with high-risk MDS and complex karyotypes.

Results of allogeneic stem cell transplant (SCT) in MDS remain poor and are limited by patient age, comorbid conditions and donor availability. The timing of SCT in MDS is also controversial since the potential for long-term, event-free survival in 30% to 40% may be offset by the high early and 1-year mortality (about 30%). Using a Markov modeling, Cutler et al. evaluated the benefit risk of allogeneic SCT versus standard therapy in early or late MDS, by whether SCT is performed at MDS diagnosis, 2 years into MDS, or at the time of transformation.

[17]The results suggested that the best survival overall was obtained when patients underwent allogeneic SCT early in higher risk MDS, but later in lower risk MDS.17 (Table 7) For example, the estimated median survivals in lower risk MDS were 4.7 to 7.2 years if SCT was performed later versus 4.6 to 6.5 years if it was done earlier. In higher risk MDS, the opposite was true: the estimated median survivals were 3.2 to 4.9 years if SCT was performed earlier versus 2.7 to 3.2 years if it was done later.

Table 7: Life expectancy estimates (in years) for allogeneic SCT in MDS by IPPS risk and by whether it was performed immediately, 2 years into MDS, or at progression

Several studies have updated the results of non-ablative allogeneic SCT in MDS.

[18],[19],[20]In all studies, the general trends with non-ablative sibling or unrelated SCT were: 1) it could be performed safely in older age groups (average 5 to 10 years older), 2) it was associated with less acute GVHD, and with a lower early (100-day and 1-year) mortality rate and with lower rates of severe organ complications, but 3) it resulted in similar rates of chronic GVHD and long-term outcome. Thus, non-ablative allogeneic SCT is gaining popularity in the context of MDS therapy.


[1]Greenberg P, Cox C, LeBeau MM, et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood. 1997;89:2079-2088.

[2]Howe RB, Porwit-MacDonald A, Wanat R, et al. The WHO classification of MDS does make a difference. Blood. 2004;103:3265-3270.

[3]Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol. 1982;51:189-199.

[4]Estey E, Keating M, Pierce S, et al. Application of the international scoring system for myelodysplasia to M. D. Anderson patients. Blood. 1997;90:2843-2844.

[5]Mannone L, Gardin C, Quarre MC, et al. For the Groupe Francais des Myelodysplasies (GFM). High response rate to Darbopoetin alfa in low risk MDS: results of a phase II study. Blood. 2004;104(suppl 1);abstract 69.

[6]Miller KB, Kim HT, Greenberg P, et al. Phase III prospective randomized trial of EPO with or without G-CSF versus supportive therapy alone in the treatment of myelodysplastic syndromes (MDS): Results of the ECOG- CLSG trial (E1996). Blood. 2004;104(suppl 1);24a abstract 70.

[7]Silverman LR, Demakos EP, Peterson BL, et al. Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J Clin Oncol. 2002;20:2429-2440.

[8]Shadduck RK, Rossetti JM, Faroun Y, et al. AML Induction Therapy with Outpatient Azacitidine. Blood. 2004;104(suppl 1):abstract 1800.

[9] Saba H, Rosenfeld C, Issa JP, et al. First report of the phase III North American trial of decitabine in advanced myelodysplastic syndrome (MDS). Blood. 2004;104(suppl 1):abstract 67.

[10]Kantarjian H, Ravandi F, OBrien S, et al. Decitabine low-dose schedule (100 mg/m2/course) in myelodysplastic syndrome (MDS). Comparison of 3 different dose schedules. Blood. 2004;104(suppl 1):abstract 1437.

[11]Kuendgen A, Strupp C, Aivado M, et al. Treatment of myelodysplastic syndromes with valproic acid alone or in combination with all-trans retinoic acid. Blood. 2004;104:1266-1269.

[12]Garcia-Manero G, Kantarjian H, Sanchez-Gonzalez B, et al. Results of a phase I/II study of the combination of 5-aza-2-deoxycytidine (DAC) and valproic acid (VPA) in patients (pts) with leukemia. Blood. 2004;104(suppl 1):abstract 263.

[13]Kurzrock R, Fenaux P, Raza A, et al. High-Risk Myelodysplastic Syndrome (MDS): First results of international phase 2 study with oral farnesyltransferase inhibitor R115777 (ZARNESTRA). Blood. 2004;104(suppl 1):abstract 68.

[14]Bouscary D, Quarre MC, Vassilief D, et al. Thalidomide for the treatment of low risk myelodysplasia. The Thal-SMD-200 Trial from the French Group of Myelodysplasia (GFM). Blood. 2004:104(suppl 1);abstract 1438 .

[15]List AF, Vardiman J, Issa JP, et al. Myelodysplastic syndromes. American Society of Hematology Education Program Book, Hematology. 2004:297-316.

[16]Knipp S, Hildebrandt B, Giagounidis AN, et al. Intensive chemotherapy is not recommended for patients with AML or high-risk MDS aged over 60 years with complex karyotype anomalies. Blood. 2004;104(suppl 1);abstract 72.

[17]Cutler CS, Lee SL, Greenberg P, et al. A decision analysis of allogeneic bone marrow transplantation for the myelodysplastic syndromes: delayed transplantation for low-risk myelodysplasia is associated with improved outcome. Blood. 104:579-585, 2004.

[18]Martino R, Van Biezen A, Iacobelli S, et al On behalf of the EBMT-CLWP MDS subcommittee. Reduced-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(suppl 1):abstract 642.

[19]Diaconescu R, Flowers C, Storer B, et al. Morbidity and mortality with nonmyeloablative compared to myeloablative conditioning before hematopoietic cell transplantation from HLA matched related donors. Blood. 2003;102(suppl 1):abstract 261.

[20]Sorror ML, Maris M, Storer B, et al. Transplant-related toxicities (TRT) and mortality following HLA-matched unrelated donor hematopoietic cell transplantation (URD-HCT) using nonmyeloablative (NM) compared to myeloablative (M) conditioning: influence of pretransplant comorbidities. Blood. 2003;102(suppl 1):abstract 262.