Myelodysplastic Syndromes (MDS) Represent Myelodysplastic Myeloid Clonal Hemopathies with Relatively Hetero- Geneous Spectrums of Presentation

Home , Erythroid dysplasia, Myelodysplastic syndrome

838

NCCN Overview The myelodysplastic syndromes (MDS) represent Myelodysplastic myeloid clonal hemopathies with relatively hetero- geneous spectrums of presentation. The major clini- Syndromes cal problems in these disorders are morbidities caused by patients’ cytopenias and the potential for MDS to Clinical Practice Guidelines in Oncology evolve into acute myeloid leukemia (AML). In the Peter L. Greenberg, MD; Eyal Attar, MD; John M. Bennett, MD; general population, the incidence rate of MDS is ap- Clara D. Bloomfield, MD; Uma Borate, MD; proximately 5 per 100,000 people per year.1,2 MDS is Carlos M. De Castro, MD; H. Joachim Deeg, MD; rare among children/adolescents and young adults, Olga Frankfurt, MD; Karin Gaensler, MD; Guillermo Garcia-Manero, MD; Steven D. Gore, MD; with an incidence rate of 0.2 per 100,000 people David Head, MD; Rami Komrokji, MD; Lori J. Maness, MD; per year in those younger than 40 years; however, Michael Millenson, MD; Margaret R. O’Donnell, MD; among individuals older than 70 years, the incidence Paul J. Shami, MD; Brady L. Stein, MD, MHS; Richard M. Stone, MD; James E. Thompson, MD; rate increases to approximately 26 per 100,000 peo- Peter Westervelt, MD, PhD; Benton Wheeler, MD; ple, and increases further to 48 per 100,000 people Dorothy A. Shead, MS; and Maoko Naganuma, MSc among those aged 80 years and older.1

Abstract Please Note The myelodysplastic syndromes (MDS) represent a heteroge- The NCCN Clinical Practice Guidelines in Oncology neous group of clonal hematopoietic disorders characterized (NCCN Guidelines®) are a statement of consensus of the by cytopenias, dysplasia in one or more myeloid lineages, and authors regarding their views of currently accepted ap- the potential for development of acute myeloid leukemia. These disorders primarily affect older adults. The NCCN Clini- proaches to treatment. Any clinician seeking to apply or ® cal Practice Guidelines in Oncology for MDS provide recom- consult the NCCN Guidelines is expected to use inde- mendations on the diagnostic evaluation and classification pendent medical judgment in the context of individual of MDS, risk evaluation according to established prognostic clinical circumstances to determine any patient’s care or assessment tools (including the new revised International treatment. The National Comprehensive Cancer Net- Prognostic Scoring System), treatment options according to work® (NCCN®) makes no representation or warranties risk categories, and management of related anemia. (JNCCN of any kind regarding their content, use, or application 2013;11:838–874) and disclaims any responsibility for their applications or NCCN Categories of Evidence and Consensus use in any way. The full NCCN Guidelines for Myelo- Category 1: Based upon high-level evidence, there is uni- dysplastic Syndromes are not printed in this issue of form NCCN consensus that the intervention is appropri- JNCCN but can be accessed online at NCCN.org. ate. © National Comprehensive Cancer Network, Inc. Category 2A: Based upon lower-level evidence, there is 2013, All rights reserved. The NCCN Guidelines and the uniform NCCN consensus that the intervention is appro- illustrations herein may not be reproduced in any form priate. without the express written permission of NCCN. Category 2B: Based upon lower-level evidence, there is Disclosures for the NCCN Myelodysplastic NCCN consensus that the intervention is appropriate. Syndromes Panel Category 3: Based upon any level of evidence, there is major NCCN disagreement that the intervention is ap- At the beginning of each NCCN Guidelines panel meeting, panel propriate. members review all potential conflicts of interest. NCCN, in keep- ing with its commitment to public transparency, publishes these All recommendations are category 2A unless otherwise disclosures for panel members, staff, and NCCN itself. noted. Individual disclosures for the NCCN Myelodysplastic Syndromes Clinical trials: NCCN believes that the best management for Panel members can be found on page 874. (The most recent any cancer patient is in a clinical trial. Participation in clinical version of these guidelines and accompanying disclosures are trials is especially encouraged. available on the NCCN Web site at NCCN.org.)

These guidelines are also available on the Internet. For the latest update, visit NCCN.org.

Managing MDS is complicated by the generally in MDS. A more complete version of the discussion advanced age of the patients (median age at diagno- text for the NCCN Guidelines for MDS is available sis is 70–75 years),2 the attendant nonhematologic at NCCN.org, which also includes pediatric MDS, comorbidities, and the older patients’ relative inabil- molecular abnormalities in MDS, comorbidity indi- ity to tolerate certain intensive forms of therapy. In ces, and management of iron overload. addition, when the illness progresses to AML, these patients experience lower response rates to standard 3 therapy than those with de novo AML. Diagnostic Classification The multidisciplinary panel of MDS experts for the NCCN Clinical Practice Guidelines in Oncol- Initial evaluation of patients with suspected MDS ogy (NCCN Guidelines) meets annually to update requires careful assessment of their peripheral blood recommendations on standard approaches to the smear and blood counts, their marrow morphology, diagnosis and treatment of MDS in adults. These the duration of their abnormal blood counts, other recommendations are based on a review of recent potential causes for their cytopenias, and concomitant clinical evidence that has led to important advances illnesses. The NCCN Guidelines for MDS include the in treatment or has yielded new information on bio- WHO classification system for diagnostic evaluations. logical factors that may have prognostic significance In 2008, a revision of the WHO classification Text continues on p. 850

NCCN Myelodysplastic Syndromes Rami Komrokji, MD Panel Members Moffitt Cancer Center Lori J. Maness, MD‡ *Peter L. Greenberg, MD‡ UNMC Eppley Cancer Center at Stanford Cancer Institute The Nebraska Medical Center Eyal Attar, MD†‡ Michael Millenson, MDÞ‡ Massachusetts General Hospital Cancer Center Fox Chase Cancer Center John M. Bennett, MDÞ† Margaret R. O’Donnell, MD‡ Consultant City of Hope Comprehensive Cancer Center Clara D. Bloomfield, MD† Paul J. Shami, MD‡ The Ohio State University Comprehensive Cancer Center – Huntsman Cancer Institute at the University of Utah James Cancer Hospital and Solove Research Institute Brady L. Stein, MD, MHSÞ‡ Uma Borate, MD Robert H. Lurie Comprehensive Cancer Center of University of Alabama at Birmingham Northwestern University Comprehensive Cancer Center Richard M. Stone, MD‡ Carlos M. De Castro, MD† Dana-Farber/Brigham and Women’s Cancer Center Duke Cancer Institute H. Joachim Deeg, MD†‡ James E. Thompson, MD†‡ Fred Hutchinson Cancer Research Center/ Roswell Park Cancer Institute Seattle Cancer Care Alliance Peter Westervelt, MD, PhD† Olga Frankfurt, MD‡ Siteman Cancer Center at Barnes-Jewish Hospital and Robert H. Lurie Comprehensive Cancer Center of Washington University School of Medicine Northwestern University *Benton Wheeler, MD Karin Gaensler, MD‡ St. Jude Children’s Research Hospital/ UCSF Helen Diller Family Comprehensive Cancer Center The University of Tennessee Health Science Center Guillermo Garcia-Manero, MD‡ NCCN Staff: Maoko Naganuma, MSc, and The University of Texas MD Anderson Cancer Center Dorothy A. Shead, MS Steven D. Gore, MD†‡ KEY: The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins *Writing Committee Member *David Head, MD≠ Specialties: †Medical Oncology; ‡Hematology/Hematology Vanderbilt-Ingram Cancer Center Oncology; ÞInternal Medicine; ≠Pathology

Myelodysplastic Syndromes, Version 2.2014

INITIAL EVALUATION ADDITIONAL TESTING CLASSIFICATION

MDS See Classification Systems (MDS-3 and MDS-5)

Required: Helpful in Some Clinical Situations: H&P Consider flow cytometry (FCM) for MDS diagnostic aidtd o assess possible CBC, platelets, differential, reticulocyte count large granular lymphocytic (LGL) diseaseaefnd to evaluate for PNH clone Examination of peripheral smear HLAtyping if hematopoietic stem cell transplant (HSCT) candidateg Diagnosis of MDS SeeAdditional Bone marrow aspiration with iron stain + biopsy + Consider HLA-DR15 typingh Cytopenia(s), established based Testing: Helpful in Consider observation to cytogenetics by standard karyotyping HLAtyping if indicated for platelet support suspect on morphologic Some Clinical document indolent course vs. Serum erythropoietin (prior to RBC transfusion) HIV testing if clinically indicated marked progression of severe myelodysplasiaa and clinical Situations (facing RBC folate, serum B Evaluate CMML patients for 5q31-33 translocations and/or PDGFRgene cytopenia or increase in blasts 12 criteriab,c page) rearrangementsi Serum ferritin, iron, total iron-binding capacity (TIBC) Consider molecular testing for JAK2 mutation in patients with Documentation of transfusion history thrombocytosis TSH (thyroid stimulating hormone) to rule out Consider additional genetic screening for patients with familial cytopenias, hypothyroidism particularly for younger patientsj Consider evaluation of copper deficiency

AML (See NCCN Guidelines for AML*)

*To view the most recent version of these guidelines, visit NCCN.org.

dSee Recommendations for Flow Cytometry (MDS-A; available online, in these guidelines, at NCCN.org) and discussion. eMarrow or peripheral blood cell FCM may be assayed and T-cell gene rearrangement studies may be conducted if LGLs are detected in the peripheral blood. Chan WC, Foucar K, Morice WG, Catovsky D. T-cell large granular lymphocytic leukemia. In: Swerdlow SH, Campo E, Harris NL, et al., eds. WHO classification of tumours of haematopoietic and lymphoid tissues. 4th ed. Lyon, France: IARC; 2008:272-273. fFCM analysis of granulocytes and monocytes with FLAER (fluorescent aerolysin) and at least one GPI-anchored protein to assess the presence of a PNH clone. Borowitz MJ, Craig FE, Digiuseppe JA, et al. Guidelines for the diagnosis and monitoring of paroxysmal nocturnal hemoglobinuria and related aMDS is also suspected in the presence of acquired MDS-related cytogenetic abnormalities and in the unexpected increase in blasts or dysplasia. disorders by flow cytometry. Cytometry B Clin Cytom 2010;78:211-230. bConfirm diagnosis of MDS according to FAB or WHO criteria for classification with application of IPSS. See Classification Systems (MDS-3 and MDS-5). gFamily HLA: evaluation to include all full siblings; unrelated evaluation to include high-resolution allele level typing for HLA-A, -B, -C, -DR, and -DQ. The percentage of marrow myeloblasts based on morphologic assessment (aspirate smears preferred) should be reported. Flow cytometric estimation of hTo assist determination of patient’s potential responsiveness to immunosuppressive therapy. blast percentage should not be used as a substitute for morphology in this context. In expert hands, expanded flow cytometry may be a useful adjunct for iCMML patients with this abnormality may respond well to tyrosine kinase inhibitors (TKIs) such as imatinib mesylate. diagnosis in difficult cases (see discussion). jTo assess possible Fanconi anemia or dyskeratosis congenita (DKC). Shortened telomere length has been associated with diseases of bone marrow failure, cPatients with significant cytopenias and karyotypes t(8;21), t(15;17), and/or inv(16) or variants should be considered to haveAML. (See NCCN Clinical including inherited disorders such as DKC, particularly in the presence of mutations in the telomerase complex genes. Telomere length can be measured by Practice Guidelines in Oncology [NCCN Guidelines] forAcute Myeloid Leukemia; to view the most recent version of these guidelines, visit NCCN.org). fluorescence in situ hybridization (FISH) assays using leukocyte samples ()see discussion .

MDS-1 MDS-2

Clinical trials: NCCN believes that the best management of any cancer patient is in a clinical trial. Participation in clinical trials is especially encouraged. All recommendations are category 2A unless otherwise indicated.

Myelodysplastic Syndromes, Version 2.2014

INITIAL EVALUATION ADDITIONAL TESTING CLASSIFICATION

MDS See Classification Systems (MDS-3 and MDS-5)

AML (See NCCN Guidelines for AML*)

*To view the most recent version of these guidelines, visit NCCN.org.

MDS-1 MDS-2

Version 2.2014, 05-21-13 ©2013 National Comprehensive Cancer Network, Inc. All rights reserved. The NCCN Guidelines® and this illustration may not be reproduced in any form without the express written permission of NCCN®.

Myelodysplastic Syndromes, Version 2.2014

CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 1 of 4) CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 2 of 4)

m k,l 2008 WHO Classification of MDS Myelodysplastic/Myeloproliferative Neoplasms (MDS/MPN) WHO Classificationn AMLWith Myelodysplasia-Related Changesq WHO Classificationr SubtypeBlood Marrow SubtypeBlood Bone marrow 1. AMLpost MDS or MDS/MPN 2. AML with an MDS-related cytogenetic abnormality Refractory cytopenia with unilineage dysplasia Chronic Dysplasia in 1 Single or bicytopenia Dysplasia in 10% of one cell line, <5% blasts >1 x 109/L monocytes, ุ 3. AML with multilineage dysplasia (RCUD)m myelomonocytic <5% blasts hematopoietic line, leukemia-1 (CMML-1) Refractory anemia with ringed sideroblasts 15% of erythroid precursors w/ring sideroblasts, <10% blasts Anemia, no blasts (RARS) erythroid dysplasia only, <5% blasts Dysplasia in 1 >1 x 109/L monocytes, ุ Refractory cytopenia with multilineage Cytopenia(s), Dysplasia in 10% of cells in 2 hematopoietic CMML-2 5%-19% blasts hematopoietic line, 9 10%-19% blasts dysplasia (RCMD) <1 x 10 /L monocytes lineages, ± 15% ring sideroblasts, <5% blasts or Auer rods or Auer rods Cytopenia(s), Refractory anemia with excess blasts-1 Unilineage or multilineage dysplasia, 2%-4% blasts, <1 x 109/L Atypical chronic myeloid WBC >13 x 109/L, (RAEB-1) No Auer rods, 5%-9% blasts Hypercellular, monocytes leukemia (CML), BCR- neutrophil precursors <20% blasts Cytopenia(s), ABL1-negative >10%, <20% blasts Refractory anemia with excess blasts-2 Unilineage or multilineage dysplasia 5%-19% blasts, <1 x 109/L (RAEB-2) Auer rods, ± 10%-19% blasts monocytes Juvenile 9 >1 x 10 /L monocytes, 9 myelomonocytic o >1 x 10 /L monocytes, Myelodysplastic syndrome, unclassified Unilineage dysplasia or no dysplasia but <20% blasts Cytopenias leukemia (JMML) (MDS-U) characteristic MDS cytogenetics, <5% blasts Dysplasia + MDS/MPN, Dysplasia + Anemia, platelets normal or Unilineage erythroid dysplasia, isolated del(5q), myeloproliferative MDS associated with isolated del(5q) unclassifiable myeloproliferative increased <5% blasts featuresp, ('overlap syndrome') features No prior MDS or MPN Refractory anemia with excess blasts in Cytopenias, 5%-19% blastsMultilineage dysplasia, Auer rods ±, 20%-30% blasts transformation (RAEB-T)l

nRefer to Tables 4.01 (p. 76), 4.02 (p. 80), 4.03 (p. 82), and 4.04 (p. 85) of 2008 WHO Classification: OraziA, Bennet JM, Germing U, et al, Myelodysplastic/myeloproliferative neoplasms. In: Swerdlow S, Campo E, Harris NL, et al., eds. World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: IARC Press, 2008:76-86. oPh-negative plus ุ2 features: Hb F, PB immature myeloid cells, WBC >10 x 109/L, clonal chromosomal abnormality, GM-CSF hypersensitivity in vitro. kRefer to Table 5.01 (p. 89) of 2008 WHO Classification: Swerdlow SH, Campo E, Harris NL, et al. World Health Organization Classification of Tumours of pExamples include thrombocytosis, leukocytosis, and splenomegaly. In addition, RARS-T has been suggested as a provisional MDS/MPN entity for individuals Haematopoietic and Lymphoid Tissue. Lyon, France: IARC; 2008. with RARS and platelet counts 450,000 x 109/L (Vardiman JW, Bennett JM, Bain BJ, et al. Myelodysplastic/myeloproliferative neoplasm, unclassifiable. In: lIn the 2008 WHO classification, RAEB-T (as previously classified by the FAB group, Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the Swerdlow S, Campo E, Harris NL, et al., eds. World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, classification of the myelodysplastic syndromes. Br J Haematol 1982;51:189-199) is classified as AML with myelodysplasia-related changes and may be France: IARC Press; 2008:85-86). more akin to MDS than AML. Refer to Arber DA, Brunning RD, OraziA, et al. Acute myeloid leukaemia with myelodysplasia-related changes. In Chapter 6. qGreater than 20% blasts in PB or marrow. Some cases with 20%-29% blasts, especially if arising from MDS, may be slowly progressive and may behave Acute Myeloid Leukemia and Related Precursor Neoplasms. In: Swerdlow S, Campo E, Harris NL, et al., eds. World Health Organization Classification of more similarly to MDS (RAEB-T by FAB classification) than to overt AML. Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, France: IARC; 2008:124-126. rArber DA, Brunning RD, OraziA, et al.Acute myeloid leukaemia with myelodysplasia-related changes. In Chapter 6. Acute Myeloid Leukemia and Related mThis category encompasses refractory anemia (RA), refractory neutropenia (RN), and refractory thrombocytopenia (RT). Cases of RN and RT were Precursor Neoplasms. In: Swerdlow S, Campo E, Harris NL, et al., eds. World Health Organization Classification of Tumours of Haematopoietic and previously classified as MDS, unclassified. Lymphoid Tissues. 4th ed. Lyon, France: IARC; 2008:124-126.

MDS-3 MDS-4

Myelodysplastic Syndromes, Version 2.2014

CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 1 of 4) CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 2 of 4) m k,l 2008 WHO Classification of MDS Myelodysplastic/Myeloproliferative Neoplasms (MDS/MPN) WHO Classificationn AMLWith Myelodysplasia-Related Changesq WHO Classificationr SubtypeBlood Marrow SubtypeBlood Bone marrow 1. AMLpost MDS or MDS/MPN 2. AML with an MDS-related cytogenetic abnormality Refractory cytopenia with unilineage dysplasia Chronic Dysplasia in 1 Single or bicytopenia Dysplasia in 10% of one cell line, <5% blasts >1 x 109/L monocytes, ุ 3. AML with multilineage dysplasia (RCUD)m myelomonocytic <5% blasts hematopoietic line, leukemia-1 (CMML-1) Refractory anemia with ringed sideroblasts 15% of erythroid precursors w/ring sideroblasts, <10% blasts Anemia, no blasts (RARS) erythroid dysplasia only, <5% blasts Dysplasia in 1 >1 x 109/L monocytes, ุ Refractory cytopenia with multilineage Cytopenia(s), Dysplasia in 10% of cells in 2 hematopoietic CMML-2 5%-19% blasts hematopoietic line, 9 10%-19% blasts dysplasia (RCMD) <1 x 10 /L monocytes lineages, ± 15% ring sideroblasts, <5% blasts or Auer rods or Auer rods Cytopenia(s), Refractory anemia with excess blasts-1 Unilineage or multilineage dysplasia, 2%-4% blasts, <1 x 109/L Atypical chronic myeloid WBC >13 x 109/L, (RAEB-1) No Auer rods, 5%-9% blasts Hypercellular, monocytes leukemia (CML), BCR- neutrophil precursors <20% blasts Cytopenia(s), ABL1-negative >10%, <20% blasts Refractory anemia with excess blasts-2 Unilineage or multilineage dysplasia 5%-19% blasts, <1 x 109/L (RAEB-2) Auer rods, ± 10%-19% blasts monocytes Juvenile 9 >1 x 10 /L monocytes, 9 myelomonocytic o >1 x 10 /L monocytes, Myelodysplastic syndrome, unclassified Unilineage dysplasia or no dysplasia but <20% blasts Cytopenias leukemia (JMML) (MDS-U) characteristic MDS cytogenetics, <5% blasts Dysplasia + MDS/MPN, Dysplasia + Anemia, platelets normal or Unilineage erythroid dysplasia, isolated del(5q), myeloproliferative MDS associated with isolated del(5q) unclassifiable myeloproliferative increased <5% blasts featuresp, ('overlap syndrome') features No prior MDS or MPN Refractory anemia with excess blasts in Cytopenias, 5%-19% blastsMultilineage dysplasia, Auer rods ±, 20%-30% blasts transformation (RAEB-T)l

MDS-3 MDS-4

Myelodysplastic Syndromes, Version 2.2014

CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 3 of 4) CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 4 of 4)

International Prognostic Scoring System (IPSS)s,t Revised International Prognostic Scoring System (IPSS-R)x WHO-Based Prognostic Scoring System (WPSS)z

Survivaland AMLevolution Score value Variable Variable scores Score value Prognostic 00.5 11.5 234 021 3 variable Prognostic variable 010.5 .0 1.5 2.0 RCUD, RARS, MDS with Very Very WHO category RCMD RAEB-1RAEB-2 Cytogeneticsy –Good – IntermediatePoor isolated deletion (5q) Marrow blasts (%)u <5 5-10 —11-20 21-30 good poor v Marrow Karyotype Good IntermediatePoor --- Karyotypev Good IntermediatePoor ื2 –>2-<5 –5-10>10 – blasts (%) Severe anemia (hemoglobin <9 Cytopeniaw 0/1 2/3 Hemogloblin 10 –8-<10 <8 ––– ุ g/dL in men or Absent Present ------<8 g/dL in Platelets ุ100 50-<100 <50– ––– women) Median 25%AML ANC ุ0.8 <0.8 – –– –– IPSS survival (y) in progression (y) risk category Overall the absence in the absence (% IPSS pop.) score of therapy of therapy Median 25%AML survival (y) in progression (y) LOW (33) 0 5.7 9.4 IPSS-R Risk category Overall the absence of in the absence WPSS risk Sum of individual variable scores (% IPSS-R pop.) score therapy of therapy INT-1 (38) 0.5-1.0 3.5 3.3 Very low 0 VERYLOW (19) ≥15. 8.8Not reached INT-2 (22) 1.5-2.0 1.1 1.1 Low 1 LOW (38) >1.5-3.0 5.3 10.8 Intermediate 2 HIGH (7) ≥2.5 0.4 0.2 INT (20) >3.0-4.5 3.0 3.2 High 3-4

HIGH (13) >4.5-6.0 1.6 1.4 Very high 5-6

VERYHIGH (10) >6.0 0.8 0.7

sIPSS should be used for initial prognostic and planning purposes. WPSS wCytopenias: neutrophil count <1800/mcL, platelets <100,000/mcL, permits dynamic estimation of prognosis at multiple time points during the hemoglobin <10 g/dL. course of MDS. xGreenberg PL,Tuechler H, Schanz J, et al. Revised International Prognostic tGreenberg P, Cox C, LeBeau M, et al. International scoring system for Scoring System (IPSS-R) for myelodysplastic syndromes. Blood evaluating prognosis in myelodysplastic syndromes [published correction 2012;120:2454-2465. appears in Blood 1998;2091:1100]. Blood 1997;89:2079-2088; ©The Web sites for accessing the IPSS-R calculator tool: http://www.ipss-r.com or American Society of Hematology. http://mds-foundation.org/calculator/index.php.Amobile App for the uPatients with 20%-30% blasts may be considered to have MDS (FAB) or calculator tool is also available for smartphones at MDS IPSS-R. AML(WHO). yCytogenetic risks: Very good = -Y, del(11q); Good = Normal, del(5q), v Cytogenetics: Good = normal, -Y alone, del(5q) alone, del(20q) alone; del(12p), del(20q), double including del(5q); Intermediate = del(7q), +8, x Poor = complex (3 abnormalities) or chromosome 7 anomalies; +19, i(17q), any other single or double independent clones; Poor = -7, Cytogenetics: Good = normal, -Y alone, del(5q) alone, del(20q) alone; Poor = complex (3 abnormalities) or chromosome 7 anomalies; Intermediate = other inv(3)/t(3q)/del(3q), double including -7/del(7q), complex: 3 abnormalities; abnormalities. [This excludes karyotypes t(8;21), inv16, and t(15;17), which are considered to be AMLand not MDS.] Intermediate = other abnormalities. [This excludes karyotypes t(8;21), z inv16, and t(15;17), which are considered to be AML and not MDS.] Very poor = Complex: >3 abnormalities. Malcovati L, Della Porta MG, Strupp C, et al. Impact of the degree of anemia on the outcome of patients with myelodysplastic syndrome and its integration into the WHO classification-based Prognostic Scoring System (WPSS). Haematologica 2011;96:1433-1440.

MDS-5 MDS-6

Myelodysplastic Syndromes, Version 2.2014

CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 3 of 4) CLASSIFICATION SYSTEMS FOR DE NOVO MDS (page 4 of 4)

International Prognostic Scoring System (IPSS)s,t Revised International Prognostic Scoring System (IPSS-R)x WHO-Based Prognostic Scoring System (WPSS)z

Survivaland AMLevolution Score value Variable Variable scores Score value Prognostic 00.5 11.5 234 021 3 variable Prognostic variable 010.5 .0 1.5 2.0 RCUD, RARS, MDS with Very Very WHO category RCMD RAEB-1RAEB-2 Cytogeneticsy –Good – IntermediatePoor isolated deletion (5q) Marrow blasts (%)u <5 5-10 —11-20 21-30 good poor v Marrow Karyotype Good IntermediatePoor --- Karyotypev Good IntermediatePoor ื2 –>2-<5 –5-10>10 – blasts (%) Severe anemia (hemoglobin <9 Cytopeniaw 0/1 2/3 Hemogloblin –8-<10 <8 ––– ุ10 g/dL in men or Absent Present ------<8 g/dL in Platelets ุ100 50-<100 <50– ––– women) Median 25%AML ANC ุ0.8 <0.8 – –– –– IPSS survival (y) in progression (y) risk category Overall the absence in the absence (% IPSS pop.) score of therapy of therapy Median 25%AML survival (y) in progression (y) LOW (33) 0 5.7 9.4 IPSS-R Risk category Overall the absence of in the absence WPSS risk Sum of individual variable scores (% IPSS-R pop.) score therapy of therapy INT-1 (38) 0.5-1.0 3.5 3.3 Very low 0 VERYLOW (19) ≥15. 8.8Not reached INT-2 (22) 1.5-2.0 1.1 1.1 Low 1 LOW (38) >1.5-3.0 5.3 10.8 Intermediate 2 HIGH (7) ≥2.5 0.4 0.2 INT (20) >3.0-4.5 3.0 3.2 High 3-4

HIGH (13) >4.5-6.0 1.6 1.4 Very high 5-6

VERYHIGH (10) >6.0 0.8 0.7

MDS-5 MDS-6

Myelodysplastic Syndromes, Version 2.2014

aa TREATMENT PROGNOSTIC CATEGORYaa TREATMENT PROGNOSTIC CATEGORY

IPSS: Low/Intermediate-1 IPSS: Low/Intermediate-1 bb IPSS-R: Very Low, Low, Intermediatebb IPSS-R: Very Low, Low, Intermediate WPSS: Very Low, Low, Intermediate WPSS: Very Low, Low, Intermediate del(5q) ± other Seefacing page cytogenetic abnormalities del(5q) ± other cytogenetic No abnormalities (for symptomatic Lenalidomidegg responsedd or Follow pathway below anemia only) intolerance

Symptomatic Serum EPO anemia Seefacing page 500 mU/mL Epoetin alfa (rHu EPO) No Follow Antithymocyte Serum EPO ±G-CSFhh No Good probability responsedd appropriate globulin (ATG),jj 500 mU/mL or responsedd to respond to ISTee or pathway Clinically No del(5q) ± other cyclosporinA significant Darbepoetin alfa intolerance below Supportive carecc cytogenetic abnormalities cytopenia(s) ±G-CSFhh as an adjunct to or increased treatment Serum EPO marrow Seefacing page >500 mU/mL Symptomatic blasts anemia with no del(5q) No Good probability to ATG,jj responsedd or Follow appropriate Azacitidine/decitabine Clinical trial respond to ISTee cyclosporinA intolerance pathway below Clinically relevant or or thrombocytopenia Disease Immunosuppressive Consider allo-HSCT or neutropeniaor progression/ therapy (IST)ee for selected IPSS increased marrow No responsedd Serum EPO or intermediate-1 blasts >500 mU/mL Clinical trial patientsff Azacitidine/decitabine Clinical trial or No or Poor probability to Consider lenalidomide responsedd or Consider allo-HSCT for respond to ISTii or intolerance selected Intermediate-1 Clinical trial patientsff

aaPresence of comorbidities should also be considered for evaluation of prognosis (see references 128-133 in the discussion) ffIPSS Intermediate-1, IPSS-R and WPSS Intermediate patients with severe aaPresence of comorbidities should also be considered for evaluation of prognosis (see references 128-133 in the discussion). bbIPSS-R Intermediate patients may be managed as very low/low cytopenias would also be considered candidates for HSCT (hematopoietic stem bbIPSS-R Intermediate patients may be managed as very low/low risk or high/very high risk depending on additional prognostic factors, such as age, risk or high/very high risk depending on additional prognostic cell transplant): allogeneic-matched sibling transplant including standard and performance status, serum ferritin levels, and serum LDH levels. If patients initially are managed as lower risk but fail to respond, move to higher risk factors, such as age, performance status, serum ferritin levels, and reduced-intensity preparative approaches or matched unrelated donor (MUD). management strategies. serum LDH levels. If patients initially are managed as lower risk but ggExcept for patients with low neutrophil counts or low platelet counts. ccSee Supportive Care (MDS-B; available online, in these guidelines, at NCCN.org). fail to respond, move to higher risk management strategies. Recommended initial dose is: 10 mg/d for 21 out of 28 days monthly for 2-4 ddResponse should be evaluated based on IWG criteria: Cheson BD, Greenberg PL, Bennett JM, et al. Clinical application and proposal for modification of ddResponse should be evaluated based on IWG criteria: Cheson BD, months to assess response (see discussion). Alternative option to lenalidomide the International Working Group (IWG) response criteria in myelodysplasia. Blood 2006;108:419-425. Greenberg PL, Bennett JM, et al. Clinical application and proposal may include an initial trial of ESAs in patients with serum EPO 500 mU/mL. eePatients generally aged 60 y, and with 5% marrow blasts, or those with hypocellular marrows, HLA-DR15 positivity, or PNH clone positivity. for modification of the International Working Group (IWG) response hhSee dosing of hematopoietic cytokines (MDS-10). ffIPSS Intermediate-1, IPSS-R and WPSS Intermediate patients with severe cytopenias would also be considered candidates for HSCT (hematopoietic stem criteria in myelodysplasia. Blood 2006;108:419-425. iiPatients lack features listed in footnote“dd”. cell transplant): allogeneic-matched sibling transplant including standard and reduced-intensity preparative approaches or matched unrelated donor eePatients aged 60 y, or those with hypocellular marrows, HLA- jjBoth equine and rabbit ATG have been used in patients with MDS (see (MUD). DR15 positivity, or PNH clone positivity. discussion).

MDS-7 MDS-8

Myelodysplastic Syndromes, Version 2.2014

aa TREATMENT PROGNOSTIC CATEGORYaa TREATMENT PROGNOSTIC CATEGORY

MDS-7 MDS-8

Myelodysplastic Syndromes, Version 2.2014

PROGNOSTIC CATEGORYaa TREATMENT EVALUATION OF RELATEDANEMIA TREATMENT OF SYMPTOMATIC ANEMIA FOLLOW-UP

Continue lenalidomide, IPSS: Intermediate-2, Highee tt Response decrease dose to tolerance IPSS-R: Intermediate,kk High, Very High del(5q) ± other WPSS: High, Very High Lenalidomidegg cytogenetic Azacitidine/decitabine abnormalities See IPSS: Low/Intermediate-1 Allo- If No Yes or rr WPSS: Very Low, Low, HSCTmm,nn relapse response Clinical trial Intermediate (MDS-8) Transplant Responsedd Continue High-intensity candidate therapy and Continue EPO, Response, candidatecc,ll Donor Responsett decrease dose rHu EPO decrease available Azacitidine (preferred) (category 1)/decitabineoo to tolerance 40,000-60,000 U dose to or Serum EPO 1-3 x/wk High-intensity chemotherapypp 500 mU/mL tolerance No subcutaneous or H&P Ring CBC, platelets, or Clinical trial sideroblasts qq differential, Darbepoetin alfa <15% 150-300 mcg/wk reticulocyte count Treat coexisting subcutaneous Consider No Clinical trial Examination of No No dd causes adding response or peripheral smear responsess response cc Replace iron, G-CSF Azacitidine (preferred) (category 1)/decitabineoo or relapse Supportive care Bone marrow (despite (See MDS-8) Not high-intensity folate, Bif 1-2 mcg/kg or aspiration with 12 adequate therapy candidatecc,ll needed 1-3 x/wk Clinical trial iron stain + iron stores) RBC subcutaneous biopsy transfusions + cytogenetics rHu EPO (leukoreduced) Serum EPO level 40,000-60,000 U Supportive 1-3 x/wk tt Consider HLA- cc Response Decrease dose to tolerance care subcutaneous + DR15 typing Serum EPO G-CSF 1-2 mcg/kg Rule out 500 mU/mL 1-3 x/wk coexisting Ring subcutaneous causes sideroblasts or 15% qq Darbepoetin alfa See IPSS: Low/Intermediate-1 150-300 mcg/wk No WPSS: Very Low, Low, subcutaneous + responsess Intermediate (MDS-8) G-CSF

aaPresence of comorbidities should also be considered for evaluation of prognosis (see references 128-133 in the discussion). ccSee Supportive Care (MDS-B; available online, in these guidelines, at NCCN.org). ddResponse should be evaluated based on IWG criteria: Cheson BD, Greenberg PL, Bennett JM, et al. Clinical application and proposal for modification of Serum EPO See Serum EPO >500 mU/mL(MDS-8) the International Working Group (IWG) response criteria in myelodysplasia. Blood 2006;108:419-425. >500 mU/mL ffIPSS Intermediate-1, IPSS-R and WPSS Intermediate patients with severe cytopenias would also be considered candidates for HSCT: allogeneic-matched sibling transplant including standard and reduced-intensity preparative approaches or matched unrelated donor (MUD). kkIPSS-R Intermediate patients may be managed as very low/low risk or high/very high risk depending on additional prognostic factors, such as age, performance status, serum ferritin levels, and serum LDH levels. llBased on age, performance status, major comorbid conditions, psychosocial status, patient preference, and availability of caregiver. Patients may be taken ccSee Supportive Care (MDS-B; available online, in these guidelines, rrLack of 1.5 g/dL rise in Hb or decreased RBC transfusion requirement by 3- immediately to transplant, or bridging therapy should be used to decrease marrow blasts to an acceptable level before transplant. at NCCN.org). 4 months of treatment. mmAzacitidine, decitabine, or other therapy may also be used as a bridge to transplant while awaiting donor availability. However, these agents should not ggExcept for patients with low neutrophil counts or low platelet counts. ssLack of 1.5 g/dL rise in Hb or decreased RBC transfusion requirement by 6- be used to delay available HSCT. Recommended initial dose is: 10 mg/d for 21 out of 28 days monthly for 8 weeks of treatment. nnHSCT: allogeneic-matched sibling, including standard and reduced-intensity preparative approaches or MUD. 2-4 months to assess response (see discussion). Alternative option to ttTarget hemoglobin range 10-12 g/dL; not to exceed 12 g/dL. ooAlthough the response rates are similar for both drugs, survival benefit from a phase lll randomized trial is reported for azacitidine and not for decitabine. lenalidomide may include an initial trial of ESAs in patients with serum ppHigh-intensity chemotherapy: EPO 500 mU/mL. qq • Clinical trials with investigational therapy (preferred), or In some institutions, darbepoetin alfa has been administered using doses Standard induction therapy if investigational protocol is unavailable or if it is used as a bridge to HSCT. up to 500 mcg weekly; also, note that darbepoetin alfa, 300 mcg every • other week is equivalent to 150 mcg weekly.

MDS-9 MDS-10

Myelodysplastic Syndromes, Version 2.2014

PROGNOSTIC CATEGORYaa TREATMENT EVALUATION OF RELATEDANEMIA TREATMENT OF SYMPTOMATIC ANEMIA FOLLOW-UP

MDS-9 MDS-10

Myelodysplastic Syndromes Text continued from p. 839 incorporated new scientific and clinical informa- consistent with MDS, and thrombocytosis (platelet tion and refined diagnostic criteria for previously counts ≥450 × 109/L).13 The morphology of RARS- described neoplasms; it also introduced newly rec- T is characterized by RARS features (no blasts in ognized disease entities.4 A new subtype in the MDS peripheral blood, dysplastic erythroid proliferation, classification is refractory cytopenia with unilineage ring sideroblasts ≥15% of erythroid precursors, and dysplasia (RCUD), which includes refractory ane- <5% blasts in marrow), with proliferation of large mia (RA; unilineage erythroid dysplasia), refractory atypical megakaryocytes similar to those seen in es- neutropenia (RN; unilineage dysgranulopoiesis), sential thrombocythemia or primary myelofibrosis; and refractory thrombocytopenia (RT; unilineage up to 60% of RARS-T cases have the JAK2 V617F dysmegakaryocytopoiesis). RN and RT were previ- mutation or MPL W515K/L mutation.13 ously classified as MDS unclassifiable.5 A review ar- The WHO classification excludes patients with ticle discusses the major changes and the rationale RAEB in transformation (RAEB-T) from MDS behind the changes in the 2008 WHO classification (proposing that AML should now include patients of MDS and AML evolving from MDS.6 with ≥20% marrow blasts, rather than the previously Other categories within the WHO classifica- used cutoff of ≥30%). However, MDS not only are tion include refractory cytopenia with multilineage related to blast quantitation but also possess a dif- dysplasia (RCMD) with or without ring sideroblasts, fering pace of disease related to distinctive biologic RA with ringed sideroblasts (RARS), RA with ex- features that differ from de novo AML.14,15 In addi- cess of blasts (RAEB) cases separated into those with tion, therapeutic responses generally differ between less than 10% marrow blasts (RAEB-1), and those these patient groups. with 10% or more marrow blasts (RAEB-2), 5q dele- The 2008 WHO classifications have helped tion [del(5q)] syndrome, and MDS unclassified (with clarify the clinical differences between the patients MDS cytogenetics, with or without unilineage dys- with FAB RAEB-T and those with AML.16 The cur- plasia). The del(5q) syndrome, recognized by WHO rent WHO classification lists the entity “AML with as a separate MDS category, includes patients with an myelodysplasia-related changes,” which encompass- isolated 5q31–33 deletion and marrow showing less es patients with AML post-MDS, AML with mul- than 5% blasts, often with thrombocytosis.7–9 This tilineage dysplasia, and AML with MDS-associated disorder generally has a relatively good prognosis10 cytogenetic abnormalities.16 According to the 2008 and is highly responsive to lenalidomide therapy.11 WHO classification, some patients with AML with The category myelodysplastic/myeloprolifera- myelodysplasia-related changes having 20% to 29% tive neoplasms (MDS/MPN), includes chronic my- marrow blasts, especially those arising from MDS elomonocytic leukemia-1 and -2 (CMML-1 and (considered RAEB-T by the FAB classification), CMML-2); atypical chronic myelogenous leukemia may have disease that behaves more similar to MDS (CML), BCR-ABL1–negative; and juvenile myelo- than to AML. monocytic leukemia (JMML) as disorders having The decision to treat patients having marrow overlapping dysplastic and proliferative features, and blasts in the range of 20% to 30% with intensive the MDS/MPN unclassifiable group.12 The distinc- AML therapy is thus complex and should be individ- tion between CMML-1 and CMML-2 is based on ualized. The clinician should consider factors such as the percentage of blasts plus monocytes in peripheral age, antecedent factors, cytogenetics, comorbidities, blood and bone marrow. CMML had been categorized pace of disease, performance status, and patient’s by French-American-British (FAB) classification as goal for treatment. The NCCN MDS Panel currently MDS, and by the International MDS Risk Analy- endorses reporting and using the WHO classification sis Workshop (IMRAW) as proliferative type (WBC system. However, as indicated in the algorithm (see count ≥12,000/mm3; a myeloproliferative disorder MDS-3, page 842), patients with RAEB-T (with [MPD]) or nonproliferative type (dysplastic MDS).10 20%–30% blasts AND a stable clinical course for at The MDS/MPN unclassifiable group includes the least 2 months) should be considered as either MDS provisional entity, RARS associated with marked or AML. These patients were previously included in thrombocytosis (RARS-T), which includes cases and benefited from many therapeutic trials for MDS, that present with clinical and morphologic features and should continue to be eligible for MDS-type

Myelodysplastic Syndromes therapy. Thus, the NCCN MDS Panel recommends cytopenias; prior infections or bleeding episodes; using the WHO classification but having the RAEB- and number of transfusions. Concomitant medica- T subgroup considered as either MDS or AML. Stud- tions and comorbid conditions require careful as- ies have provided evidence supporting the use of the sessment. Because MDS are relatively indolent dis- WHO proposals.17–21 orders, blood count stability is used to distinguish AML evolving from MDS (AML-MDS) is often MDS from evolving AML. Other possible causes for more resistant to standard cytotoxic chemotherapy patients’ cytopenias also require careful evaluation. than is de novo AML, which arises without ante- In addition to establishing current blood and re- cedent hematologic disorder. Patients with high-risk ticulocyte counts, clinicians need a peripheral blood MDS or AML-MDS, and some elderly patients with smear evaluation to determine the degree of dysplasia AML may have a more indolent course in terms of and, thus, potentially dysfunctional cells. Bone mar- short-term progression compared with patients with row aspiration with Prussian blue stain for iron and standard presentations of de novo AML. Separate biopsy are needed to evaluate the degree of hema- protocols for treating patients with a standard pre- topoietic cell maturation abnormalities and relative sentation of de novo AML and these other presenta- proportions, percentage of marrow blasts, marrow tions seem appropriate (see NCCN Guidelines for cellularity, presence or absence of ringed sideroblasts Acute Myeloid Leukemia; to view the most recent (and presence of iron per se), and fibrosis. Marrow version of these guidelines, visit NCCN.org). cytogenetics (using standard karyotyping methods) To help provide consistency in diagnostic guide- should be obtained because they are of major impor- lines of MDS, an international consensus working tance for prognosis. group recommended that minimal diagnostic criteria Other useful screening laboratory studies include for this disease include required diagnostic prereq- serum erythropoietin (sEpo), vitamin B12, red blood uisites: stable cytopenia (for at least 6 months un- cell (RBC) folate levels, and serum ferritin. Serum less accompanied by a specific karyotype or bilineage ferritin levels may be nonspecific, particularly in the dysplasia, in which case only 2 months of stable cy- face of inflammatory conditions such as rheumatoid topenias are needed) and the exclusion of other po- arthritis, and therefore obtaining the serum iron lev- tential disorders as a primary reason for dysplasia or/ els and total iron binding capacity along with serum and cytopenia. In addition, the diagnosis of MDS re- ferritin may be helpful. Because hypothyroidism and quires at least 1 of 3 MDS-related (decisive) criteria: other thyroid disorders can lead to anemia, patients 1) dysplasia (≥10% in ≥1 of the 3 major bone marrow should also be evaluated for levels of thyroid-stimu- lineages), 2) a blast cell count of 5% to 19%, and 3) a lating hormone.23 specific MDS-associated karyotype, such as del(5q), If patients require platelet transfusions for severe del(20q), +8, or –7/del(7q). Furthermore, several thrombocytopenia, HLA typing (A, B) may be help- co-criteria help confirm the diagnosis of MDS, in- ful. For hematopoietic stem cell transplant (HSCT) cluding studies with flow cytometry, bone marrow candidates, the cytomegalovirus (CMV) status and histology and immunohistochemistry, or molecular full HLA typing (A, B, C, DR, and DQ) of the pa- markers (to detect or exclude abnormal CD34 anti- tient and potential donors are needed. Bone mar- genic expression, fibrosis, dysplastic megakaryocytes, row flow cytometry for assessing the percentage of atypical localization of immature progenitors, and CD34+ cells (blast cells are usually CD34+), and HIV myeloid clonality).22 screening if clinically indicated, may also be valu- able in some clinical situations. However, estimates of blast percentage derived from flow cytometry do Initial Evaluation not provide the same prognostic information as that Several types of evaluations are needed to determine derived from morphologic evaluation. Accordingly, the clinical status of patients with MDS. Under- data from flow cytometry should not be used in lieu standing of clinical status is necessary to determine of the determination of morphologic blast percentage diagnostic and prognostic categorization and decide by an experienced hematopathologist. The screen- on treatment options. Clinical history should in- ing for paroxysmal nocturnal hemoglobinuria (PNH) clude the timing, severity, and tempo of abnormal and HLA-DR15 is potentially useful for determining

Myelodysplastic Syndromes

which patients may be more responsive to immuno- rancies have also been described in erythroid cells, suppressive therapy, particularly in young patients erythroid analysis is not provided by most flow cy- with normal cytogenetics and hypoplastic MDS24,25 tometry laboratories. The European LeukemiaNET (see “Prognostic Stratification,” facing page). PNH recently developed a flow cytometric score based on is a rare acquired hematopoietic stem cell disorder reproducible parameters using CD34 and CD45 as arising from mutations in the PIGA gene, resulting markers to help diagnose MDS.35 The scoring system in defective synthesis of the glycophosphatidylinosi- was developed using multicenter retrospective data tol (GPI) anchor, which in turn leads to deficiency from patients with low-grade MDS (defined as <5% of proteins that are normally linked to the cell mem- marrow blasts; n=417) and those with nonclonal brane of blood cells via a GPI anchor.26–28 Deficiency cytopenias as controls (n=380). This population of in GPI-anchored proteins such as those involved in patients was chosen for the study because low-grade complement inhibition (eg, CD-55, CD-59) leads to MDS often lack specific diagnostic markers (eg, ring complement sensitivity of RBCs and subsequent he- sideroblasts, clonal cytogenetic abnormalities), and molysis.26 Flow cytometry is the established method therefore may be difficult to diagnose based on mor- for detecting GPI anchor–deficient cells for the di- phology alone. Bone marrow samples from patients agnosis of PNH. Recent data show that fluorescent with MDS showed different flow cytometric patterns aerolysin (FLAER), a protein that specifically binds compared with samples from patients with nonclonal to GPI anchors, was a reliable marker for detecting cytopenias with regard to the following: increased GPI anchor–deficient clones among granulocytes or CD34+ myeloblast-related cluster size (defined by monocytes, with high sensitivity.29 For evaluation of wider distribution of CD45 expression and greater PNH clone, multiparameter flow cytometry analysis side scatter characteristics [SSC]), decreased CD34+ of granulocytes and monocytes using FLAER and at B-progenitor cluster size (defined by relatively low least one GPI-anchored protein are recommended.26,29 CD45 expression and low SSC), aberrant myeloblast Although evidence for a minor PNH clone may be CD45 expression (based on lymphocyte-to-myelo- present in approximately 20% of patients with MDS, blast CD45 ratio), and decreased granulocyte SSC usually no evidence of PNH-related hemolysis is seen value (based on granulocyte-to-lymphocyte SSC ra- in these patients. tio).35 These 4 parameters were included in a logis- Bone marrow biopsy staining for reticulin is tic regression model, and a weighted score (derived helpful for evaluating the presence and degree of from regression coefficients) was assigned to each pa- bone marrow fibrosis. rameter; the sum of the scores provided the overall Discrete from basic flow cytometric evaluation at flow cytometric score for each sample, with a score of presentation for characterizing blasts and evaluating 2 or higher defined as the threshold for MDS diagno- lymphoid populations, expanded flow cytometry may sis.35 Using this flow cytometric score in the learning be a useful adjunct for diagnosing MDS in difficult cohort, a correct diagnosis of MDS was made with cases. In expert hands (both in terms of technical so- 70% sensitivity and 93% specificity. Among patients phistication and interpretation), flow cytometry may with MDS with no specific markers of dysplasia, 65% demonstrate abnormal differentiation patterns or were correctly identified. The positive and negative aberrant antigen expression in myeloid or progeni- predictive values were 92% and 74%, respectively. tor cells, which may help confirm diagnosis of MDS, These outcomes were confirmed in the validation exclude differential diagnostic possibilities, and, in cohort, which showed 69% sensitivity and 92% some patients, provide prognostic information.30–34 specificity.35 This flow cytometric scoring system To achieve these purposes, the flow analysis should demonstrated high diagnostic power in differentiat- use appropriate antibody combinations with 4-chan- ing low-grade MDS from nonclonal cytopenias, and nel fluorescence instrumentation.30–34 Multiple aber- may be particularly useful in establishing a diagnosis rancies should be present for diagnosing MDS, be- when traditional diagnostic methods are indeter- cause single aberrancies are not infrequent in normal minate. Further independent validation studies are populations. For follow-up studies, antibody combi- warranted to determine the utility of this method. nations may be tailored to detect specific abnormali- Because of the associated expense, requirements ties found in the initial evaluation. Although aber- for both technical and interpretational expertise,

Myelodysplastic Syndromes and need for greater consensus on specific antibody and/or AML.45 The mutations showed an autoso- combinations and procedures that are most informa- mal dominant pattern of inheritance, and affected tive and cost-effective, flow cytometric assays should individuals with this familial form of MDS/AML be performed by experienced laboratories and used had poor outcomes in the absence of allogeneic stem in general practice only when diagnosis is uncertain cell transplant.45 Identification of familial MDS is with traditional approaches (eg, blood counts, mor- of clinical importance because it is associated with phology, cytogenetics, increased blasts). Flow cytom- chromosomal fragility, and these patients may there- etry studies may also be used to assess the possibil- fore respond differently to hypomethylating agents; ity of large granular lymphocytic (LGL) disease, if more importantly, family members may not be eli- relevant as indicated by LGLs being present in the gible as donors for allogeneic HSCT. peripheral blood.36 Determination of platelet-derived growth fac- Additional genetic screening should be con- tor receptor beta (PDGFRβ) gene rearrangements sidered for patients with familial cytopenias, which is helpful for evaluating patients with CMML/MPD will help evaluate for Fanconi anemia or dyskerato- 5q31–33 translocations. The activation of this gene sis congenita (DC). Shortened telomere length has encoding a receptor tyrosine kinase for PDGFRβ been associated with diseases of bone marrow fail- has been shown in some of these patients.46,47 Data ure, including inherited disorders such as DC, par- have indicated that patients with CMML/MPD with ticularly in the presence of mutations in the DKC1, these PDGFRβ fusion genes may respond well to TERT, or TERC genes that encode for components treatment with the tyrosine kinase inhibitor ima- of the telomere complex.37,38 Telomere length can be tinib mesylate.48–50 measured by fluorescence in situ hybridization assays The frequency of activating mutations of the using leukocyte (or leukocyte subset) samples.37,39 tyrosine kinase known as Janus kinase 2 (JAK2) in Other genetic lesions, such as those occurring in the MDS and de novo AML is lower compared with that RUNX1 or GATA2 gene, have been implicated in in myeloproliferative disorders.51 If thrombocytosis familial cases of MDS and other myeloid malignan- is encountered in patients with MDS, screening for cies. Lesions within the RUNX1 gene (mutations, JAK2 mutations may be helpful; a positive result is deletions, or translocations) have been identified consistent with the presence of a myeloproliferative as a cause of a relatively rare autosomal dominant component of this disorder.52 familial platelet disorder that predisposes to my- Recent flow cytometric studies suggest the po- eloid malignancies.40,41 In affected families with the tential efficacy of this methodology for characterizing RUNX1 lesions, the incidence of MDS/AML is MDS marrow blast cells and as an aid for assessing high, ranging from 20% to 60%; the median age of prognosis.53,54 However, because of the nonstandard- onset is 33 years.42 This familial platelet disorder is ized nature of these analyses, further investigations are characterized by the presence of thrombocytopenia, warranted before their routine use can be suggested. and a tendency for mild to moderate bleeding gener- Reports have shown that copper deficiency can ally present from childhood; however, some affected mimic many of the peripheral blood and marrow individuals may not display these clinical character- findings seen in MDS.55–57 Thus, in certain instances, istics.42 Different types of genetic lesions in RUNX1 assessment of copper and ceruloplasmin levels may account for the variable phenotypes associated with be indicated as part of the initial diagnostic workup familial platelet disorder between different families. of suspected MDS. Clinical features associated with Cryptic genetic lesions in RUNX1 have been report- copper deficiency include vacuolation of myeloid ed in some patients with Fanconi anemia and MDS/ and/or erythroid precursors,55–57 prior gastrointestinal 43 55,56 56,58 AML. The GATA2 gene codes for a transcription surgery, and a history of vitamin B12 deficiency. factor involved in gene regulation during the development and differentiation of hematopoietic cells, and its expression was shown to correlate with severe Prognostic Stratification dysplasia in patients with primary MDS.44 Recent- Despite its value for diagnostic categorization of pa- ly, heritable mutations in GATA2 were identified tients with MDS, the prognostic limitations of the in families with highly penetrant early-onset MDS FAB classification have become apparent, with vari-

Myelodysplastic Syndromes

able clinical outcomes within the FAB subgroups. ties (≥3 chromosome anomalies) or chromosome 7 The morphologic features contributing to this vari- anomalies had relatively poor prognoses (16%). The ability include the wide range of marrow blast per- remaining patients were intermediate in outcome centages for patients with RAEB (5%–20%) and (14%). Of the patients in the “complex” category, CMML (1%–20%); lack of inclusion of critical bio- most had chromosome 5 or 7 abnormalities in addi- logic determinants, such as marrow cytogenetics; tion to other anomalies. and the degree and number of morbidity-associated To develop the IPSS for MDS, relative risk scores cytopenias. These well-perceived problems for cate- for each significant variable (marrow blast percent- gorizing patients with MDS have led to the develop- age, cytogenetic subgroup, and number of cytope- ment of additional risk-based stratification systems.59 nias) were generated.10 Through combining the risk Prognostic Scoring Systems scores for the 3 major variables, patients were strati- The International Prognostic Scoring System (IPSS) fied into 4 distinctive risk groups in terms of both for primary MDS emerged from deliberations of the survival and AML evolution: low, intermediate-1 IMRAW.10 Compared with previously used systems, (INT-1), intermediate-2 (INT-2), and high. When the risk-based IPSS has markedly improved prognos- either cytopenias or cytogenetic subtypes were omit- tic stratification of MDS cases. In this analysis, cy- ted from the classification, discrimination among the togenetic, morphologic, and clinical data were com- 4 subgroups was much less precise. Both for survival bined and collated from a relatively large group of and AML evolution, the IPSS showed statistically MDS cases that had been included in previously re- greater prognostic discriminating power than earlier 10 ported prognostic studies.10,59 FAB morphologic cri- classification methods, including the FAB system. teria were used to establish the diagnoses of MDS. In Recent data have indicated that additional addition, relative stability of peripheral blood counts clinical variables are additive to the IPSS regarding for 4 to 6 weeks was needed to exclude other pos- prognosis for patients with MDS. The WHO classifi- sible causes for the cytopenias, such as drugs, other cation–based prognostic scoring system (WPSS) in- diseases, or incipient evolution to AML. CMML was corporates the WHO morphologic categories, IPSS subdivided into proliferative and nonproliferative cytogenetic categories, and patients’ need or lack subtypes. Patients with proliferative-type CMML of dependence on RBC transfusion.61 This system (WBC >12,000/mcL) were excluded from this anal- showed that the requirement for RBC transfusions is ysis.10 Patients with nonproliferative CMML (WBC a negative prognostic factor for patients in the lower- ≤12,000/mcL and other features of MDS) were in- risk MDS categories. In addition, depth of anemia cluded in the analysis.60 per se has additive and negative prognostic import Significant independent variables for determin- for the intermediate IPSS categories.62 Compared ing outcome for both survival and AML evolution with the 4 groups defined by the IPSS, the WPSS were found to be marrow blast percentage, number classifies patients into 5 risk groups, differing in both of cytopenias, and cytogenetic subgroup (good, in- survival and risk of AML: very low, low, intermedi- termediate, poor). Patients with the chromosome ate, high, and very high. Malcovati et al61 initially anomalies t(8;21) or inv16 are considered to have reported on the usefulness of WPSS, and subsequent AML and not MDS, regardless of the blast count. studies have confirmed the findings.63–65 The ini- Age was also a critical variable for survival, although tial WPSS was recently refined to address the no- not for AML evolution. The percentage of marrow tion that the requirement for RBC transfusion may blasts was divided into 4 categories: 1) less than 5%; be somewhat subjective. In the refined WPSS, the 2) 5% to 10%; 3) 11% to 20%; and 4) 21% to 30%. measure of the degree of anemia based on transfusion Cytopenias were defined for the IPSS as hav- dependency is replaced by the presence (or absence) ing hemoglobin level less than 10 g/dL, an absolute of severe anemia, defined as hemoglobin levels less neutrophil count (ANC) less than 1800/mcL, and than 9 g/dL for men and less than 8 g/dL for wom- platelet count less than 100,000/mcL. Patients with en.66 This approach allows for an objective assess- normal marrow karyotypes, del(5q) alone, del(20q) ment of anemia, while maintaining the prognostic alone, and -Y alone had relatively good prognoses implications of the 5 risk categories defined in the (70%), whereas those with complex abnormali- original WPSS (as mentioned earlier).66 Currently,

Myelodysplastic Syndromes whether the WPSS offers an improvement over the risk categories (very low, low, intermediate, high), IPSS is a matter of ongoing debate. Based on the cur- but with only 1 patient categorized as very high risk. rent available data, the NCCN MDS Panel included The IPSS-R was significantly predictive of survival the WPSS in the current version of the treatment outcomes in both the subgroups of IPSS low and algorithm with a category 2B designation. INT-1. Within the IPSS low-risk group, median sur- Most recently, a revised IPSS (IPSS-R) was vival based on the IPSS-R risk categories was 118.8 developed that also defines 5 risk groups (very low, months for very low, 65.9 months for low, and 58.9 low, intermediate, high, and very high) versus the months for intermediate (P<.001). Within the IPSS 4 groups defined in the initial IPSS.67 The IPSS-R, INT-1 risk group, median survival based on the IP- which was derived from an analysis of a large data- SS-R risk categories was 113.7 months for very low, set from multiple international institutions, refined 60.3 months for low, 30.5 months for intermediate, the original IPSS by incorporating the following into and 21.2 months for high risk (P<.001).73 In addi- the prognostic model: more detailed cytogenetic tion, within the IPSS INT-1 risk group (but not for subgroups, separate subgroups within the “marrow IPSS low-risk), IPSS-R was significantly predictive blasts less than 5%” group, and depth of cytopenias of the 3-year rate of AML evolution.73 Thus, in this defined, with cutoffs for hemoglobin, platelet, and analysis, the IPSS-R seemed to provide prognostic neutrophil counts. In the IPSS-R, the cytogenetic refinement within the IPSS INT-1 group, with a subgroups comprise 5 risk groups (vs 3 in the original large proportion of patients (511 of 1096 those with IPSS) based on the recently published cytogenetic IPSS INT-1) identified as having poorer prognosis scoring system for MDS.68 Other parameters includ- (median survival, 21–30 months). This study also ing age, performance status, serum ferritin, lactate applied the refined WPSS to further stratify the IPSS dehydrogenase (LDH), and ß2-microglobulin pro- low and INT-1 risk groups, and was able to identify vided additional prognostic information for survival a group of patients (refined WPSS high-risk group) outcomes but not for AML evolution; age as an addi- within the IPSS INT-1 group who had poorer prog- tional factor was more prognostic among lower-risk nosis (185 of 1096 IPSS INT-1 patients; median sur- versus higher-risk groups.67 The predictive value of vival 24.1 months). However, the IPSS-R identified the IPSS-R was recently validated in several inde- a larger proportion of patients with poor-risk IPSS pendent studies based on registry data, including in INT-1 than the refined WPSS (47% vs 17%).73 Out- studies that evaluated outcomes for patients treated comes of risk stratification using the MD Anderson with hypomethylating agents.69–74 In a multiregional Lower-Risk Prognostic Scoring System are discussed study of registry data of patients with MDS from in the following paragraph. In a retrospective da- Italy (N=646), significant differences in outcomes tabase analysis of patients with MDS from a single among the IPSS-R risk categories were found for institution (N=1029), median overall survivals ac- overall survival, AML evolution, and progression- cording to IPSS-R risk categories was 82 months for free survival (latter defined as leukemic evolution or very-low, 57 months for low, 41 months for interme- death from any cause).71 Notably, in this cohort, the diate, 24 months for high, and 14 months for very- predictive power (based on Harrell’s C statistics) of high risk groups (P<.005).72 The median follow-up in IPSS-R was found to be greater than IPSS, WPSS, this study was 68 months. IPSS-R was also predictive and refined WPSS for the 3 outcome measures men- of survival outcomes among the patients who under- tioned earlier. The investigators acknowledged the went therapy with hypomethylating agents (n=618). short follow-up (median, 17 months) of the study A significant survival benefit with hypomethylating cohort.71 In a retrospective analysis of data from pa- agents was shown only for the group of patients with tients with lower-risk MDS (IPSS low or INT-1) in very-high-risk IPSS-R (median 16 vs 7 months with a large multicenter registry (N=2410) from Spain, no hypomethylating therapy; P<.005). In addition, the IPSS-R could identify 3 risk categories (very significantly longer overall survival with allogeneic low, low, intermediate) within the IPSS low-risk HSCT was only observed for patients with high (me- group, with none of the patients categorized as IPSS- dian, 42 vs 21 months without HSCT; P=.004) and R high or very high.73 Within the IPSS INT-1 risk very high (median, 31 vs 12 months without HSCT; group, the IPSS-R further stratified patients into 4 P<.005) risks.72 The IPSS-R may therefore provide a

Myelodysplastic Syndromes

tool for therapeutic decision-making, although pro- IPSS low-risk group, median survival using the LR- spective studies are needed to confirm these findings. PSS risk categories was 130.3 months for category 1 Further evaluation is warranted and ongoing regard- (low-risk), 69.7 months for category 2 (intermedi- ing the utility of the IPSS-R in both the settings of ate-risk), and 58.4 months for category 3 (high-risk; clinical trials and routine clinical practice. P<.001); the corresponding median survival values The Lower-Risk Prognostic Scoring System for the patients with IPSS INT-1 risk were 115.2, (LR-PSS) developed by investigators at MD Ander- 51.3, and 24.1 months, respectively (P<.001). The son Cancer Center is another prognostic model used LR-PSS identified an important proportion of pa- in the evaluation of MDS, and was designed to help tients (334 of 1096 patients; 30.5%) within the IPSS identify patients with lower-risk disease (IPSS low or INT-1 risk group who had poorer prognosis (median INT-1) who may have poor prognosis.75 The prognos- survival 24 months). In addition, within the IPSS tic model was developed using clinical and laboratory INT-1 risk group (but not for IPSS low-risk), LR- data from patients with IPSS low (n=250) and INT- PSS was significantly predictive of the rate of AML 1 (n=606) MDS. Factors associated with decreased evolution at 3 years.73 In another analysis using data survival were identified and a prognostic model con- from a cohort of patients with lower-risk MDS from structed based on results of multivariate Cox regres- 2 centers (N=664), median survival according to sion analysis. The final model included the following the LR-PSS risk categories was 91.4 months for cat- factors that were independent predictors for survival egory 1, 35.6 months for category 2, and 22 months outcomes: unfavorable cytogenetics, older age (≥60 for category 3.78 Using data from the same cohort of years), decreased hemoglobin (<10 g/dL), decreased patients, median survival according to the IPSS-R platelet counts (<50 × 109/L or 50–200 × 109/L), and risk groups was 91.4 months for IPSS-R very good, higher percentage of bone marrow blasts (≥4%).75 35.9 months for good, and 27.8 months for the com- Important to note, however, is that the cytogenetic bined intermediate/high/very-high risk groups. Both categories in this system were derived from the pre- of these prognostic scoring systems were significantly viously defined IPSS categories rather than from predictive of survival outcomes. The predictive pow- the more-refined IPSS-R. Each of these factors were er (based on Harrell’s C statistics) of LR-PSS and given a weighted score, and the sum of the scores IPSS-R was 0.64 and 0.63, respectively.78 (ranging from 0–7 points) were used to generate 3 risk categories; a score of 0 to 2 points was assigned to category 1, a score of 3 or 4 was category 2, and a Therapeutic Options score of 5 to 7 was category 3. Using this scoring sys- The patient’s IPSS risk category is used in initial tem, median survival was 80.3 months for category planning of therapeutic options because it provides 1, 26.6 months for category 2, and 14.2 months for a risk-based patient evaluation (category 2A). In ad- category 3; the 4-year survival rates were 65%, 33%, dition, factors such as the patient’s age, performance and 7%, respectively. The scoring system allowed for status, and presence of comorbidities are critical further stratification into these 3 risk categories for determinants because they have a major influence both the subgroup of patients with IPSS low-risk and on the patient’s ability to tolerate certain intensive IPSS INT-1 risk disease.75 The LR-PSS could there- treatments. The WPSS provides dynamic estimation fore be a useful tool to identify patients with lower of prognosis at any time during the course of MDS. risk disease who may have poorer prognosis and may In patients who were only recently evaluated, require earlier treatment. The prognostic value of the determining the relative stability of the patient’s LR-PSS has been validated in several independent blood counts over several months is important to as- studies.73,76–78 In a retrospective analysis of data from sess disease progression, including incipient transfor- lower risk MDS (IPSS Low or INT-1) patients in the mation to AML. In addition, this assessment permits multicenter Spanish registry (N=2410), the LR-PSS determination of other possible causes for cytope- was able to further stratify these lower risk patients nias. The patient’s preference for a specific approach into 3 risk categories.73 The LR-PSS was significantly is also important in deciding treatment options. The predictive of survival outcomes in the subgroups of therapeutic options for MDS include supportive patients with IPSS low and INT-1 risk. Within the care, low-intensity therapy, high-intensity therapy,

Myelodysplastic Syndromes and/or participation in a clinical trial. In evaluating symptomatic anemia as needed (generally leukocyte- results of therapeutic trials, the panel found it impor- reduced) or platelet transfusions for bleeding events; tant for studies to use the standardized International however, platelet transfusions should not be used Working Group (IWG) response criteria.79–81 routinely in patients with thrombocytopenia in the For the MDS therapeutic algorithm, all patients absence of bleeding. Based on differing institutional should undergo relevant supportive care. After that, policies, the panel reached nonuniform consensus re- the panel has proposed initially stratifying patients garding the necessity for routine irradiation of blood with clinically significant cytopenias into 2 major products used in patients with MDS; however, the risk groups: 1) relatively lower-risk patients (who are panel agreed that all directed-donor and transfused in the IPSS low and INT-1 category; IPSS-R very low, products for potential stem cell transplant patients low, and intermediate categories; or WPSS very low, should be irradiated. Additionally, CMV-negative low, and intermediate categories); and 2) higher-risk blood products are recommended whenever possible patients (who are in the IPSS INT-2/high categories; for CMV-negative recipients. Aminocaproic acid or IPSS-R intermediate, high, very high categories; or other antifibrinolytic agents may be considered for WPSS high, very high categories). Patients in the bleeding episodes refractory to platelet transfusions IPSS-R intermediate category may be managed as or for profound thrombocytopenia. very low/low risk or high/very high risk depending on evaluation of additional prognostic factors, such as age, performance status, serum ferritin levels, Hematopoietic Cytokines and serum LDH levels.67 In addition, patients with Hematopoietic cytokine support should be consid- intermediate-risk whose disease does not respond to ered for refractory symptomatic cytopenias.82 For therapy for lower-risk disease would be eligible to re- example, recombinant human granulocyte colony- ceive therapy for higher-risk MDS. stimulating factor (G-CSF) or granulocyte-monocyte Based on IWG response criteria, hematologic colony-stimulating factor (GM-CSF) treatment improvement is the major therapeutic aim for pa- could be considered for neutropenic patients with tients in the lower-risk group, whereas alteration of MDS with recurrent or resistant bacterial infections. the disease natural history is viewed as paramount for Erythropoiesis-stimulating agents (ESAs), such those in the higher risk group. Cytogenetic response as recombinant human erythropoietin (Epo) or dar- and quality-of-life (QOL) parameters are also im- bepoetin, with or without G-CSF, have been evalu- portant outcomes to assess. The algorithms outline ated for the treatment of symptomatic anemia in pa- management of primary MDS only. Most patients tients with MDS.83–87 In a phase II study in patients with therapy-related MDS have poorer prognoses with MDS (RA, RARS and RAEB; N=50), Epo than those with primary MDS, including a substan- combined with G-CSF (n=47 evaluable) resulted in tial proportion with poor-risk cytogenetics. These hematologic response in 38% of patients (complete patients are generally managed as having higher-risk response in 21%).83 Epo and G-CSF seemed to have disease. synergistic activity. Lower serum Epo levels (<500 Supportive Care mU/mL) and lower pretreatment RBC transfusion Currently, the standard of care in the community requirement (<2 units per month) were associated for MDS management includes supportive care (see with higher response rate; response rates were not MDS-B [available online, in these guidelines, at significantly different across IPSS risk groups.83 Me- NCCN.org] and appropriate NCCN Guidelines for dian survival (N=71, including patients from a prior Supportive Care; see list of NCCN Guidelines for study) was 26 months. Among patients with low-risk Supportive Care, available online, at NCCN.org). IPSS, median survival had not been reached at 5 This entails observation, clinical monitoring, psy- years, and the 5-year survival rate was 68%. Median chosocial support, and QOL assessment. Major efforts survivals among INT-1 and INT-2 risk groups were should be directed toward addressing the relevant 27 and 14 months, respectively. AML progression QOL domains (eg, physical, functional, emotional, occurred in 28% of patients overall during the obser- spiritual, social) that adversely affect the patient. vation period; the frequencies of AML progression Supportive care should include RBC transfusions for in the low-, INT-1, INT-2, and high-risk groups were

Myelodysplastic Syndromes

12%, 21%, 45%, and 100%, respectively. Among re- This study also showed improvements in QOL pa- sponding patients who received maintenance treat- rameters among responding patients.85 Collectively, ment with Epo and G-CSF, the median duration of these studies suggest that ESAs may provide clinical response was 24 months.83 A subsequent analysis of benefit in patients with lower-risk MDS with symp- long-term outcomes with Epo combined with G-CSF tomatic anemia. Limited data are available on the (given 12–18 weeks, followed by maintenance in re- effectiveness of ESAs in the treatment of anemia in sponders) in patients with MDS (n=121; patients lower-risk patients with del(5q). Epo has been shown from 3 phase II Nordic trials) reported a hematologic to promote the growth of cytogenetically normal response rate of 39%, with a median duration of re- cells isolated from patients with del(5q), while hav- sponse of 23 months.84 Long-term outcomes of these ing minimal proliferative effects on MDS progeni- patients were compared with outcomes from untreat- tor cells from these patients in vitro.88 Retrospective ed patients (n=237) as controls. Based on multivari- studies from the French group reported hematologic ate Cox regression analysis, treatment with Epo plus response rates of 46% to 64%, with a median dura- G-CSF was associated with significantly improved tion of response of 11 months (and mean duration of survival outcomes (hazard ratio, 0.61; 95% CI, 0.44– 13–14 months) among patients with del(5q) treated 0.83; P=.002). An exploratory analysis revealed that with ESAs, with or without G-CSF.86,87 Duration of the association between treatment and survival was response in these patients was significantly decreased significant only for the IPSS low-risk group. In ad- compared with that in patients without del(5q) dition, the survival benefit with treatment was ob- (mean duration, 25–27 months).86 Based on multi- served only in patients requiring fewer than 2 units variate analysis, del(5q) was a significant predictor of RBC transfusions per month. No significant asso- of shorter response duration with treatment.87 The ciation was found between treatment and frequency use of ESAs to treat symptomatic anemia is discussed of AML progression.84 Similar findings were reported further in “Therapy for Lower Risk Patients” and in a study from the French group, which analyzed “Evaluation and Treatment of Related Anemia” on outcomes with ESAs (epoetin or darbepoetin), with pages 863 and 865, respectively. or without G-CSF, in patients with MDS with ane- Severe thrombocytopenia is associated with mia (N=403).87 Based on IWG 2000 criteria, the he- increased risks for bleeding events, and is currently matologic response rate was 62%, with a median du- managed with platelet transfusions. The mecha- ration of response of 20 months. Based on IWG 2006 nism of thrombocytopenia in patients with MDS criteria, the corresponding results were 50% and 24 may be attributed to decreased platelet production months, respectively. IPSS low/INT-1 risk was as- (possibly related to regulatory pathways involving sociated with significantly higher response rates and the production and/or metabolism of endogenous longer response durations. In a comparison of out- thrombopoietin [TPO]) and increased destruction comes (in the subset of low/INT-1 risk with anemia) of bone marrow megakaryocytes or circulating plate- between treated patients (n=284) and a historical lets.89,90 Endogenous TPO levels have been reported cohort of untreated patients (n=225), multivariate to be increased among patients with MDS compared analysis showed a significant association between with those of healthy individuals.90 At the same treatment with ESAs and survival outcomes. The time, TPO receptor (c-mpl) sites per platelet seem frequency of AML progression was similar between to be decreased among patients with MDS compared the cohorts.87 In a phase II study evaluating darbepo- with healthy subjects. Among patients with MDS, etin (given every 2 weeks for 12 weeks), with or with- those with RA seemed to have the highest TPO out G-CSF (added at 12 weeks in nonresponders), levels compared with those with RAEB or RAEB-T, in patients with MDS with lower-risk IPSS with whereas the number of TPO receptor sites remained anemia (and with serum Epo levels <500 mU/mL), similar across subtypes.90 hematologic response rate was 48% at 12 weeks and In addition, studies have reported that high 56% at 24 weeks.85 Median duration of response was endogenous TPO levels correlated with decreased not reached, with a median follow-up of 52 months. platelet counts in patients with RA, but not in pa- The 3-year cumulative incidence of AML progres- tients with RAEB or RAEB-T.90,91 This observation sion was 14.5% and the 3-year survival rate was 70%. suggests that the regulatory pathway for endogenous

Myelodysplastic Syndromes

TPO may be further disrupted in patients with RAEB with supportive care in patients with MDS (N=191; or RAEB-T, potentially because of overexpression of previously untreated in 83%; all IPSS risk groups), TPO receptors in blasts that may lead to inadequate hematologic responses occurred in 60% of patients TPO response.90,91 Several studies are investigat- in the AzaC arm (7% complete response, 16% par- ing the role of the thrombopoietin receptor agonist tial response, 37% hematologic improvement) com- romiplostim in the treatment of thrombocytope- pared with a 5% hematologic improvement (and no nia in patients with lower-risk MDS.92–96 Phase I/II responses) in those receiving supportive care.104 The studies with romiplostim showed promising rates of median time to AML progression or death was sig- platelet response (46%–65%) in patients with lower- nificantly prolonged with AzaC compared with sup- risk MDS.93,95 Recent randomized placebo-controlled portive care (21 vs 13 months; P=.007). Addition- studies in patients treated for lower-risk MDS have ally, the time to progression to AML or death was reported beneficial effects of romiplostim in terms improved in patients who received AzaC earlier in of decreased bleeding events and reduced needs for the course of disease, suggesting that the drug pro- platelet transfusions in patients receiving hypometh- longed the duration of stable disease. Subsequently, ylating agents,92,94 and decreased frequency of dose Silverman et al105 provided a summary of 3 studies of reductions or delays in those receiving lenalidomide AzaC in a total of 306 patients with high-risk MDS. therapy.96 Eltrombopag is another thrombopoietin In this analysis, which included patients receiving receptor agonist that has been shown to increase either subcutaneous or intravenous delivery of the normal megakaryopoiesis in vitro in bone marrow drug (75 mg/m2/d for 7 days every 28 days), com- cells isolated from patients with MDS.97,98 Ongo- plete remissions were seen in 10% to 17% of patients ing phase I and II clinical trials are investigating the treated with AzaC, partial remissions were rare, and activity and safety of this agent in the treatment of 23% to 36% of patients had hematologic improve- thrombocytopenia in patients with MDS. Concerns ment. Of the responses, 90% were seen by cycle 6 for potential proliferation of leukemic blasts in re- and the median number of cycles to first response was sponse to exogenous TPO were raised in early in vi- 3.105 The authors concluded that AzaC provided im- tro studies in the past, particularly for patients with portant clinical benefits for patients with high-risk 99,100 high-risk MDS. Results from ongoing clinical MDS. Results from a phase III randomized trial in trials with the TPO mimetics mentioned earlier will patients (N=358) with higher-risk MDS (IPSS INT- help to elucidate the risks for leukemic transforma- 1, 5%; INT-2, 41%; high-risk, 47%) showed that tions in patients with MDS. Neither romiplostim AzaC was superior to conventional care (standard nor eltrombopag is currently approved for use in pa- chemotherapy or supportive care) in terms of overall tients with MDS. survival.101 AzaC was associated with a significantly Low-Intensity Therapy longer median survival compared with conventional Low-intensity therapy includes the use of care (24.5 vs 15 months; hazard ratio, 0.58; 95% CI, low-intensity chemotherapy or biologic response 0.43–0.77; P=.0001), thus providing support for the modifiers. Although this type of treatment is mainly use of this agent in patients with higher-risk disease. provided in the outpatient setting, certain treat- AzaC therapy should be considered for treating ments may require supportive care or occasional hos- patients with MDS with progressing or relatively pitalization (eg, to treat infections). high-risk disease. It is FDA-approved for the treatment of patients with MDS and is generally administered at a dose of 75 mg/m2/d subcutaneously for 7 Hypomethylating Agents days every 28 days for at least 4 to 6 courses. Treat- As a form of relatively low-intensity chemotherapy, ment courses may need to be extended further or randomized phase III trials have shown that the DNA may be used as a bridging therapy to more definitive methyltransferase inhibitor (DMTI) hypomethyl- therapy (eg, HSCT for patients whose marrow blast ating agents 5-azacytidine (AzaC) and decitabine counts require lowering before that procedure). Al- (5-aza-2′-deoxycytidine) decrease the risk of leuke- though the optimal duration of therapy with AzaC mic transformation, and, in a portion of the patients, has not been defined, some data suggest that contin- improve survival.101–104 In one trial comparing AzaC uation of AzaC beyond first response may improve

Myelodysplastic Syndromes

remission quality. Secondary analysis from the afore- 13% having hematologic improvement. The median mentioned phase III randomized trial of AzaC in pa- duration of response was 10 months. The probability tients with higher-risk MDS101 showed that among of progression to AML or death was 1.68-fold greater patients responding to AzaC, response quality was for patients receiving supportive care than for those improved in 48% with continued therapy; although receiving decitabine. Based on results of this study most responding patients achieved a first response and 3 supportive phase II trials,113 the FDA also ap- by 6 cycles of therapy, up to 12 cycles was required proved the drug for treating MDS. for most responders to attain a best response.106 In In a recent phase III randomized trial, decitabine this study, the median number of cycles from first re- was compared with best supportive care in older pa- sponse to best response was 3 to 3.5 cycles, and re- tients aged 60 years or older (N=233; median age, sponding patients received a median of 8 additional 70 years; range, 60–90 years) with higher-risk MDS cycles (range, 0–27 cycles) beyond first response.106 (IPSS INT-1, 7%; INT-2, 55%; high-risk, 38%) not An alternative 5-day schedule of AzaC has been eligible for intensive therapy.103 Median progres- evaluated, both as a subcutaneous regimen (includ- sion-free survival was significantly improved with ing the 5-2-2 schedule [75 mg/m2/d subcutaneously decitabine compared with supportive care (6.6 vs for 5 days followed by 2 days of no treatment, then 3.0 months; hazard ratio, 0.68; 95% CI, 0.52–0.88; 75 mg/m2/d for 2 days, every 28 days] and the 5-day P=.004) and the risk of AML progression at 1 year schedule [75 mg/m2/d subcutaneously for 5 days ev- was significantly reduced with decitabine (22% vs ery 28 days])107 and as an intravenous regimen (75 33%; P=.036). However, no significant differences mg/m2/d intravenously for 5 days every 28 days).108 were observed between decitabine and supportive Although response rates with the 5-day regimens care for the primary end point of overall survival seemed to be similar to those for the approved 7-day (10.0 vs 8.5 months, respectively) or for median dosing schedule,107,108 survival benefit with AzaC has AML-free survival (8.8 vs 6.1 months, respective- only been shown using the 7-day schedule. ly).103 In the decitabine arm, complete and partial Similarly, the other DMTI hypomethylating responses were observed in 13% and 6% of patients, agent, decitabine, given intravenously and admin- respectively, with hematologic improvement in an istered with a regimen that requires hospitalization, additional 15%; in the supportive care arm, hemato- has also shown encouraging results in patients with logic improvement was seen in 2% of patients (with higher-risk MDS. Because the treatment regimen no hematologic responses). In addition, decitabine was generally associated with low-intensity–type was associated with significant improvements in toxicities, it is also considered to be “low-intensity patient-reported QOL measures (as assessed by the therapy.” In earlier phase II studies, the drug resulted EORTC QOL Questionnaire C30) for the dimen- in cytogenetic conversion in approximately 30% of sions of fatigue and physical functioning.103 patients,109 with an overall response rate of 49% and Alternate dosing regimens using lower doses of a 64% response rate in patients with a high-risk IPSS decitabine administered in an outpatient setting are score.110 The results of these studies were similar to currently being evaluated. In 2007, Kantarjian et those for AzaC.111,112 al114 provided an update of their results in 115 pa- The results of a phase III randomized trial of tients with higher-risk MDS using alternative and decitabine (15 mg/m2 intravenous infusion over 3 lower-dose decitabine treatment regimens. Patients hours every 8 hours [ie, 45 mg/m2/d] on 3 consecutive received 1 of 3 different schedules of decitabine, days every 6 weeks for up to 10 cycles) versus sup- including both subcutaneous and intravenous ad- portive care in adult patients (N=170) with primary ministration, and a mean of 7 courses of therapy. and secondary MDS with IPSS INT-1 (31%), INT-2 Responses were improved with this longer duration (44%), and high-risk (26%) disease indicated high- of therapy. Overall, 80 patients (70%) experienced er response rates, longer remission duration, longer a response, with 40 patients (35%) experiencing a time to AML progression, and greater survival ben- complete response and 40 (35%) a partial response. efit in the INT-2 and high-risk groups.102,111 Overall The median remission duration was 20 months and response rate (complete response plus partial re- the median survival time was 22 months. sponse) with decitabine was 17%, with an additional Kantarjian et al115 also compared the 3 different

Myelodysplastic Syndromes schedules of decitabine in a study randomizing 95 time needed to further reduce the blast count, patients with MDS or CMML to either 20 mg/m2/d improve the patient’s performance status, or intravenously for 5 days; 20 mg/m2/d subcutaneously identify a donor). In these circumstances, the for 5 days; or 10 mg/m2/d intravenously for 10 days. drugs may be used as bridging therapy for that The 5-day intravenous schedule was considered the procedure. optimal schedule, with a complete response rate in • Those who experience relapse after allogeneic this arm of 39% compared with 21% in the 5-day HSCT. subcutaneous arm and 24% in the 10-day intrave- In addition, hypomethylating agents are appro- nous arm (P<.05). priate options for patients with IPSS low- or INT-1– Several retrospective studies have evaluated the risk or IPSS-R very low- or low-risk disease without role of cytoreductive therapy with hypomethylat- symptomatic anemia, or with symptomatic anemia ing agents before allogeneic HSCT (with both my- and elevated serum Epo levels who are not expected eloablative and reduced-intensity conditioning regimens).116–119 Results suggest that hypomethylating to experience a response to (or who relapsed after) agents may provide a feasible alternative to induc- immunosuppressive therapy. tion chemotherapy regimens before transplant, and may serve as a bridge to allogeneic HSCT. Biologic Response Modifiers and Currently, AzaC and decitabine are considered Immunosuppressive Therapy to be therapeutically relatively similar, although the improved survival of higher-risk patients treated The currently available nonchemotherapy with AzaC compared with controls seen in a phase low-intensity agents (biologic response modifiers) III trial, as indicated earlier, supports the preferred include antithymocyte globulin (ATG), cyclospo- use of AzaC in this setting. “Failure to respond to rine, thalidomide, lenalidomide, anti–tumor necro- hypomethylating agents” is considered if a lack of sis factor receptor fusion protein, and vitamin D complete response, partial response, or hematologic analogues, all of which have shown some efficacy in 3,120–125 improvement is seen; frank progression to AML oc- phase I and II trials. curs, particularly with loss of control (proliferation) Use of anti-immune–type therapy with ATG 122,123 of peripheral counts; or excess toxicity precludes with or without cyclosporine has been shown continuation of therapy. The minimum number of in several studies to be most efficacious in patients courses before the treatment is considered a failure with MDS with HLA-DR15 histocompatibility– should be 4 to 6 courses. As discussed earlier, the type, marrow hypoplasia, normal cytogenetics, low- optimal duration of therapy with hypomethylating risk disease, and evidence of a PNH clone.24,25 Re- agents has not been well defined and no consensus searchers from the NIH updated their analysis of exists. The NCCN MDS Panel generally feels that 129 patients treated with immunosuppressive ther- treatment should be continued in the presence of an apy involving equine ATG and cyclosporine alone ongoing response and no toxicities, in which case or in combination.126 This study showed markedly modifications should be made to the dosing frequen- improved response rates in younger (≤60 years of cy for individual patients. age) and IPSS INT-1 patients and those with high Because data have predominantly indicated al- response probability characteristics, as indicated by tered natural history and decreased evolution to their prior criteria (HLA-DR15+, age, and number AML in patients who experience response to treat- of transfusions).126 ment, the major candidates for these drugs are those Both equine and rabbit ATG are available in with IPSS INT-2 or high-risk MDS or IPSS-R inter- the United States for immunosuppressive therapy. A mediate-, high-, or very-high-risk MDS, such as: randomized study from the NIH compared the ac- • Those who are not candidates for high-intensity tivity of equine versus rabbit ATG, combined with therapy. cyclosporine, in previously untreated patients with • Those who are potential candidates for alloge- severe aplastic anemia (N=120) who were not eli- neic HSCT but for whom delay in receipt of gible for transplant.127,128 This study showed that in that procedure is anticipated (eg, because of this patient population, rabbit ATG was inferior to

Myelodysplastic Syndromes

equine ATG, as shown by the lower 6-month he- fusion-dependent anemia.132 The primary end point matologic response rate (primary end point, 37% of RBC transfusion independence for 26 weeks or vs 68%; P<.001) and higher number of deaths (14 more was achieved in a significantly greater propor- vs 4 patients), resulting in decreased survival rates tion of patients treated with lenalidomide, 5 mg or among patients treated with rabbit ATG. The 3-year 10 mg, versus those treated with placebo (43% vs survival rate was also significantly inferior with rab- 56% vs 6%, respectively; P<.001 for both lenalido- bit ATG compared with equine ATG (76% vs 96%; mide groups vs placebo). Among patients achieving P=.04).128 The 3-year cumulative incidence of re- RBC transfusion independence with lenalidomide, lapse was not significantly different between treat- onset of erythroid response was rapid, with 86% of ment groups (11% vs 28%, respectively).127 Within patients experiencing response onset within the first the setting of MDS, however, only limited data are 2 cycles (49% in cycle 1).132 Among patients treated available regarding the comparative effectiveness of with lenalidomide with baseline serum Epo levels the 2 ATG formulations. In a relatively small phase greater than 500 mU/mL, the 10-mg dose resulted II study in patients with MDS (N=35; primarily RA in significantly higher rates of RBC transfusion in- subtype), both equine and rabbit ATG were shown dependence compared with the 5-mg dose (76% vs to be feasible and active.129 33%; P=.004). Cytogenetic response rates were sig- Lenalidomide (a thalidomide analog) is an im- nificantly higher for the lenalidomide 5- or 10-mg munomodulating agent that has shown activity in arms compared with placebo (25% vs 50% vs 0%, patients with lower-risk MDS.11,130 Beneficial results respectively; P<.001 for both lenalidomide groups vs have been particularly evident for patients with placebo; P = not significant between the lenalido- del(5q) chromosomal abnormalities.11,130,131 In a mul- mide dose groups), and complete response rates were ticenter phase II trial of lenalidomide, given at a dose observed in 16% and 29% of patients in the lenalid- of 10 mg/d for 21 days every 4 weeks or 10 mg/d to omide 5- and 10-mg arms, respectively. Median time 148 patients with MDS with del(5q) and RBC trans- to AML progression has not yet been reached in the fusion–dependent anemia, with or without addition- lenalidomide treatment arms. No significant differ- al cytogenetic abnormalities, hematologic response ences were observed in median overall survival be- was rapid (median time to response, 4.6 weeks; range, tween the lenalidomide 5 mg, 10 mg, and placebo 1–49 weeks) and sustained.11 RBC transfusion inde- groups (≥35.5 vs 44.5 vs 42 months, respectively). pendence (assessed at 24 weeks) occurred in 67% of The most common grade 3 or 4 adverse events were patients, and in 69% of those with IPSS low/INT-1 myelosuppression and deep vein thrombosis (DVT). risk (n=120).11 Cytogenetic responses were achieved Grade 3 or 4 neutropenia was reported in 74%, 75%, in 62 of 85 evaluable patients (73%); 45% had a com- and 15% of patients in the lenalidomide 5 mg, 10 plete cytogenetic response. The most common grade mg, and placebo arms, respectively; thrombocytope- 3 or 4 adverse events included myelosuppression nia occurred in 33%, 41%, and 1.5%, respectively. (neutropenia in 55%; thrombocytopenia in 44%), Grade 3 or 4 DVT occurred in 4 patients (6%) in the which often required treatment interruption or dose lenalidomide 10-mg arm, and 1 patient each in the reduction. Thus, careful monitoring of blood counts lenalidomide 5-mg (1%) and placebo (1.5%) arms.132 during the treatment period is mandatory when using A recent comparative analysis evaluated out- this agent, particularly in patients with renal dysfunc- comes with lenalidomide (based on data from the tion (because of the drug’s renal route of excretion). 2 aforementioned trials; n=295) compared with no The FDA approved lenalidomide for treating pa- treatment (based on data from untreated patients in tients with IPSS low-/INT-1–risk MDS with del(5q) a multicenter registry; n=125) in patients with IPSS and transfusion-dependent anemia, with or without low-/INT-1–risk MDS with del(5q) RBC transfu- additional cytogenetic abnormalities. sion–dependent anemia.133 Untreated patients from A recent randomized controlled phase III trial the registry had received best supportive care, includ- compared the activity of lenalidomide (5 mg/d for ing RBC transfusion, iron chelation therapy, and/or 28 days or 10 mg/d for 21 days of a 28-day cycle) ESAs. The 2-year cumulative incidence of AML pro- versus placebo in 205 patients with lower-risk MDS gression was 7% with lenalidomide and 12% in the (IPSS low- and INT-1 risks) with del(5q) and trans- untreated cohort; the corresponding 5-year rate was

Myelodysplastic Syndromes

23% and 20%, respectively. Thus, the median time in the context of clinical trials. Comparative studies to AML progression has not been reached in either have not shown benefit between the different inten- cohort. Lenalidomide was not a significant factor for sive chemotherapy regimens (including idarubicin-, AML progression in either univariate or multivariate cytarabine-, fludarabine-, and topotecan-based regi- analyses. The 2-year overall survival probability was mens) in MDS.136 90% with lenalidomide and 74% in the untreated A high degree of multidrug resistance occurs in cohort; the corresponding 5-year probability was marrow hematopoietic precursors in patients with 54% and 40.5%, respectively. The median overall advanced MDS,137 with associated decreased re- survival was 5.2 and 3.8 years, respectively (P=.755; sponses and shorter response durations associated Kaplan-Meier plot with left truncation to adjust with many standard treatment regimens of induction for differences in timing of study entry between co- chemotherapy. Thus, chemotherapeutic agents used horts).133 Based on multivariate analysis using Cox to treat “resistant-type” AML and agents that modu- proportional hazard models (also with left trunca- late this resistance are now being evaluated for treat- tion), lenalidomide was associated with significantly ing patients with advanced MDS. Although several decreased risk of death compared with no treatment studies using modulators of multidrug resistance (hazard ratio, 0.597; 95% CI, 0.399–0.894; P=.012). showed positive results in this setting,138,139 others did Other independent factors associated with decreased not.140 Further clinical trials evaluating other modu- risk of death were female sex, higher hemoglobin lators of multidrug resistance are ongoing. levels, and higher platelet counts. Conversely, in- Allogeneic HSCT from an HLA-matched sib- dependent factors associated with increased risk of ling donor is a preferred approach for treating a se- death included older age and greater RBC transfu- lect group of patients with MDS, particularly those sion burden.133 with high-risk disease.141–148 Matched nonmyeloab- A phase II study evaluated lenalidomide treat- lative transplant regimens and matched unrelated ment in 214 transfusion-dependent patients with donor stem cell transplants are becoming options at low- or INT-1–risk MDS without the del(5q).134 Re- some centers to treat these patients.149–157 In certain sults showed 26% of the non-del(5q) patients (56 of investigative settings, autologous bone marrow or 214) experienced transfusion independence after a peripheral blood stem cell transplantation is being median of 4.8 weeks of treatment. This transfusion considered.158 Whether transplants should be per- independence continued for a median duration of formed before or after patients experience remission 41 weeks, and the median rise in hemoglobin was following induction chemotherapy has not been pro- 3.2 g/dL (range, 1.0–9.8 g/dL). A 50% greater reduc- spectively established.159 Comparative clinical trials tion in transfusion requirement was noted in an ad- are needed to determine these points. ditional 37 patients (17%), yielding an overall rate of hematologic improvement of 43%. The most common grade 3 or 4 adverse events were neutropenia Recommended Treatment Approaches (30%) and thrombocytopenia (25%). Further evalu- Therapy for Lower-Risk Patients (IPSS Low and ation in more-extended clinical trials is needed to de- INT-1; IPSS-R Very Low, Low, and Intermediate; or termine the efficacy of this drug and other agents for WPSS Very Low, Low, and Intermediate) patients with non-del(5q) MDS. The NCCN MDS Regarding the algorithm for therapeutic options for Panel recommends lenalidomide be considered for the lower-risk patients with clinically significant cy- treating non-del(5q) patients with symptomatic ane- topenias or increased bone marrow blasts, the NCCN mia that did not respond to initial therapy. MDS Panel recommends stratifying these patients High-Intensity Therapy into several groups. Those with del(5q) chromosom- High-intensity therapy includes intensive induction al abnormalities and symptomatic anemia should re- chemotherapy, or HSCT.3,135 Although these ap- ceive lenalidomide. The recommended dose in this proaches have the potential to change the natural setting is 10 mg once daily for 21 days, every 28 days; history of the disease, they also have an attendant response should be assessed 2 to 4 months after ini- greater risk of regimen-related morbidity and mortal- tiation of treatment. However, lenalidomide should ity. The panel recommends giving these treatments be avoided in patients with clinically significant de-

Myelodysplastic Syndromes

crease in neutrophil counts or platelet counts; in the Data from the phase III randomized trial of AzaC previously discussed phase III trial with lenalidomide compared with conventional care showed signifi- in patients with del(5q), those with low neutrophils cantly higher rates of major platelet improvement (<500/mcL) or platelet counts (<25,000/mcL) were with AzaC compared with conventional care (33% excluded.132 An alternative option in patients with vs 14%; P=.0003); it should be noted, however, that del(5q) and symptomatic anemia may include an the rates for major neutrophil improvements were initial trial of ESAs in patients who have serum Epo similar between AzaC and the control arm (19% vs levels of 500 mU/mL or less. 18%), and that the study was conducted in patients Other patients with symptomatic anemia are with higher-risk MDS.101 Patients who do not re- categorized based on their serum Epo levels. Those spond to hypomethylating agents should be consid- with levels of 500 mU/mL or less should be treated ered for treatment with immunosuppressive therapy, with ESAs (recombinant human Epo or darbepo- a clinical trial, or allogeneic HSCT. etin) with or without G-CSF (see “Evaluation and Careful monitoring for disease progression and Treatment of Related Anemia,” facing page). Non- consideration of patient preferences play major roles responders should be considered for immunosup- in the timing and decision to embark on treatment pressive therapy (with ATG or cyclosporine) if they for lower- or higher-risk disease. have a high likelihood of experiencing a response to Therapy for Higher-Risk Patients (IPSS INT-2 and this therapy. In lower-risk MDS, the most appropri- High; IPSS-R Intermediate, High, and Very High; ate candidates for immunosuppressive therapy in- or WPSS High and Very High) clude patients who are aged 60 years or younger, are Treatment for higher-risk patients depends on whether HLA-DR15–positive, have a PNH-positive clone, or they are believed to be candidates for intensive thera- have 5% or less marrow blasts or hypocellular mar- py (eg, allogeneic HSCT or intensive chemotherapy). row. Alternatively, or in patients who experience no Clinical features relevant for this determination include response to immunosuppressive therapy, treatment the patient’s age, performance status, absence of major with AzaC, decitabine, or lenalidomide should be comorbid conditions, psychosocial status, patient prefer- considered. Patients with no response to hypometh- ence, and availability of a suitable donor and caregiver. ylating agents or lenalidomide in this setting should In addition, the patient’s personal preference for type of be considered for participation in a clinical trial with therapy needs particular consideration. Supportive care other relevant agents, or for allogeneic HSCT (see should be provided for all patients. “Allogeneic HSCT,” opposite column). Intensive Therapy: Allogeneic HSCT: The potential Patients with anemia whose serum Epo levels for patients to undergo allogeneic HSCT depends on are greater than 500 mU/mL should be evaluated several factors, including patient age, performance to determine whether they have a good probability status, major comorbid conditions, psychosocial sta- of response to immunosuppressive therapy. Nonre- tus, availability of a caregiver, IPSS or WPSS score, sponders to immunosuppressive therapy would be and the availability of a suitable donor. For patients considered for treatment with AzaC, decitabine, or a who are transplant candidates, the first choice of do- clinical trial. Patients with serum Epo levels greater nor has remained an HLA-matched sibling, although than 500 mU/mL who have a low probability of re- results with HLA-matched unrelated donors have im- sponding to immunosuppressive therapy should be proved to levels comparable to those obtained with considered for treatment with AzaC, decitabine, HLA-matched siblings. With the increasing use of or lenalidomide. Others or nonresponders to these cord blood or HLA haploidentical related donors, treatments could be considered for a clinical trial or HSCT has become a viable option for many patients. for allogeneic HSCT. Patients without symptomatic High-dose conditioning is typically used for younger anemia who have other clinically relevant cytopeni- patients, whereas the approach using reduced-/low- as (particularly clinically severe thrombocytopenia) intensity conditioning (RIC) for HSCT is generally or increased bone marrow blasts should be consid- the strategy in older individuals.160 ered for treatment with AzaC or decitabine, immu- To aid therapeutic decision-making regarding nosuppressive therapy (if there is a good probability the timing and selection of patients for HSCT, a of responding to these agents), or a clinical trial. study compared outcomes with HLA-matched sib-

Myelodysplastic Syndromes ling HSCT in patients with MDS aged 60 years or row blast count), even a partial remission may be ad- younger to the data in nontreated patients with MDS equate to permit the HSCT. For this purpose, AzaC, from the IMRAW/IPSS database.161 Using a Markov decitabine, and participation in clinical trials are also decision analysis, this investigation indicated that considered valid treatment options. IPSS INT-2 and high-risk patients aged 60 years Nonintensive Therapy: For higher-risk patients who or younger had the highest life expectancy if they are not candidates for intensive therapy, the use of underwent transplantation (from HLA-identical AzaC or decitabine or enrollment in a relevant clini- siblings) soon after diagnosis, whereas patients with cal trial should be considered. Based on the recently IPSS low-risk had the best outlook if HSCT was de- published results of the phase III trial showing supe- layed until disease progression. Patients in the INT-1 rior median survival in patients receiving AzaC com- risk group had only a slight gain in life expectancy if pared with best supportive care, the NCCN MDS HSCT was delayed, and therefore decisions should Panel made this a preferred category 1 recommenda- probably be made on an individual basis in these tion compared with decitabine. Results from another patients (eg, dependent on platelet or neutrophil recent phase III trial comparing decitabine and sup- counts).161 A study published in 2008 retrospectively portive care in higher-risk patients failed to show a evaluated the impact of the WHO classification and survival advantage, although response rates are simi- WPSS on the outcome of patients who underwent lar to those reported previously for AzaC.103,168 How- allogeneic HSCT.63 The data suggest that lower-risk ever, no trials have compared AzaC head-to-head patients (based on WPSS risk score) do very well with decitabine. with allogeneic HSCT, with a 5-year overall survival For some patients eligible for HSCT therapy re- of 80%. With increasing WPSS scores, the probabil- quiring a reduction in tumor burden, the use of AzaC ity of 5-year survival after HSCT declined progres- or decitabine may be a bridge to decrease the marrow sively to 65% (intermediate risk), 40% (high risk), blast count enough to permit the transplant. and 15% (very high risk).63 Supportive Care Only: For patients with adverse Based on data regarding RIC for transplantation clinical features or disease progression despite thera- from 2 reported series162,163 and 2 comprehensive re- py and absence of reasonable specific antitumor ther- views of this field,164,165 patient age and disease status apy, adequate supportive care should be maintained. generally dictate the type of conditioning to be used. Patients older than 55 or 60 years, particularly if they have fewer than 10% marrow myeloblasts, would Evaluation and Treatment of generally undergo HSCT after RIC; if the blast count Related Anemia is high, pre-HSCT debulking therapy is generally giv- Major morbidities of MDS include symptomatic ane- en. Younger patients, regardless of marrow blast bur- mia and associated fatigue. Much progress has been den, will generally receive high-dose conditioning. made in improving the management of this anemia. Variations on these approaches would be considered However, along with giving specific treatment for by individual transplant physicians based on these anemia related to MDS, health care providers must features and the specific regimen used at their center. identify and treat any coexisting causes of anemia. Some general recommendations are presented in the Standard assessments should be performed to review article by Deeg and Sandmaier.166 look for other causes of anemia, such as gastrointesti- Intensive Chemotherapy: For patients eligible for inten- nal bleeding, hemolysis, renal disease, and nutrition- sive therapy but lacking a stem cell donor, or for those al deficiency. If needed, iron, folate, or vitamin 12B requiring marrow blast count reduction, intensive studies should be obtained and the cause of depletion induction chemotherapy should be considered.167 Al- corrected, if possible. After excluding these causes though the response rate and durability of this treat- and providing proper treatment for them, treatment ment is lower than for standard AML, this treatment of the MDS-related anemia should be considered (particularly in clinical trials with novel agents) could further. Currently, the standard of care for patients be beneficial in a portion of the patients. For patients with symptomatic anemia is RBC transfusion sup- with a potential stem cell donor who require reduc- port (using leuko-poor products). If the patient is a tion of their tumor burden (ie, to decrease the mar- potential HSCT candidate, the panel recommends

Myelodysplastic Syndromes

that CMV-negative (if the patient is CMV-negative istered daily or 1 to 3 times per week.83,171–173 G-CSF serologically) and irradiated transfused products be is available in single-use vials or prefilled syringes considered. containing either 300 or 480 mcg and requires re- Anemia related to MDS generally presents as a frigeration. Patients may be taught to self-administer hypoproductive macrocytic anemia, often associated the drug. Again, detection of erythroid responses with suboptimal elevation of serum Epo levels.3,169 To generally occurs within 6 to 8 weeks of treatment. determine WHO subtype, iron status, and the level If no response occurs in this time frame, treatment of ring sideroblasts, bone marrow aspiration with should be considered a failure and discontinued. In iron stain, biopsy, and cytogenetics should be exam- the case of treatment failure, one should rule out and ined. Patients should also be considered for HLA- treat deficient iron stores. Clinical trials or support- DR15 typing as indicated earlier. ive care are also treatment options in this category Individuals with symptomatic anemia and of patients. A validated decision model has been del(5q) with or without other cytogenetic abnormali- developed for predicting erythroid responses to Epo ties should receive a trial of lenalidomide. An alter- plus G-CSF, based on the patient’s basal serum Epo native option to lenalidomide may include an initial level and number of previous RBC transfusions.173,175 trial of ESAs in patients with serum Epo levels of 500 Improved QOL has been demonstrated in patients mU/mL or less. Those with normal cytogenetics, less experiencing response.175 This cytokine treatment than 15% marrow ringed sideroblasts, and serum Epo is not suggested for patients with endogenous serum levels of 500 mU/mL or less may respond to Epo if Epo levels greater than 500 mU/mL because of the relatively high doses of recombinant human Epo are very low erythroid response rate to these drugs in this administered.82,170,171 The required Epo dose is 40,000 patient population. to 60,000 units given 1 to 3 times a week subcutane- Darbepoetin alfa is a longer-acting form of Epo. ously. Erythroid responses generally occur within 6 to Studies predominantly involving patients with lower- 8 weeks of treatment.83,172–174 A more prompt response risk MDS have shown a substantial proportion of ery- may be obtained by starting at the higher dose; this throid responses with the initial trials, showing response dose is much higher than that needed to treat renal rates of 40% and 60% (combined major and minor re- causes of anemia wherein marrow responsiveness sponses using IWG response criteria).176,177 Results of would be relatively normal. If a response occurs, the clinical trials in patients with MDS have suggested that recommendation is to continue this dose but attempt the overall response rates to darbepoetin are similar to to decrease it to tolerance. The literature supports or possibly higher than those to epoetin.176–179 daily or 2- to 3-times-per-week dosing. These response rates may be partly from the Iron repletion must be verified before instituting dosage used (150–300 mcg/wk, subcutaneously) or Epo or darbepoetin therapy. If no response occurs with to the fact that better-risk patients were enrolled in these agents alone, the addition of G-CSF should be studies of darbepoetin than in those of epoetin. Fea- considered. Evidence suggests that G-CSF (and, to a tures predictive of response have included relatively lesser extent, GM-CSF) has synergistic erythropoietic low basal serum Epo levels, low percentage of mar- activity when used in combination and markedly en- row blasts, and relatively few prior RBC transfusions. hances the erythroid response rates.83,171–173 This is par- In March 2007 and 2008, the FDA announced ticularly evident for patients with 15% or more ringed alerts and strengthened safety warnings for the use of sideroblasts in the marrow (and serum Epo level ≤500 ESAs. They noted that increased mortality, possible mU/mL), because the very low response rates in this tumor promotion, and thromboembolic events were subgroup to Epo or darbepoetin alone are markedly observed in patients without MDS receiving ESAs enhanced when combined with G-CSF.83,173 when dosing targeted hemoglobin levels greater than For the erythroid synergistic effect, relatively 12 g/dL (study patients had chronic kidney failure; low doses of G-CSF are needed to help normalize were receiving radiation therapy for various malig- the neutrophil count in patients who are initially nancies, including head and neck, advanced breast, neutropenic or to double the neutrophil count in lymphoid, or non–small cell lung cancers; had can- patients who are initially normal. For this purpose, cer but were not receiving chemotherapy; or were an average of 1 to 2 mcg/kg subcutaneously is admin- orthopedic surgery patients).

Myelodysplastic Syndromes

However, as indicated earlier, ESAs have been by the FDA for treating specific subtypes of MDS: used safely in large numbers of adults with MDS and lenalidomide for patients with del(5q) cytogenetic have become important for symptomatic improve- abnormalities; AzaC and decitabine for treating pa- ment of the anemia caused by this disease, often tients with higher risk or nonresponsive MDS; and with a decrease in RBC transfusion requirements. deferasirox for iron chelation of patients with iron The NCCN MDS Panel recommendations for us- overloaded MDS. However, because a substantial ing ESAs in MDS have evolved from these and more proportion of patient subsets lack effective treat- recent data. In addition, studies assessing the long- ment for their cytopenias or for altering disease term use of Epo with or without G-CSF in patients natural history, clinical trials with these and other with MDS compared with either randomized180 or novel therapeutic agents along with supportive care historical controls84,87 have shown no negative im- remain the hallmark of management for this disease. pact of this treatment on survival or AML evolution. The role of thrombopoietic cytokines for managing 84 In addition, results of the studies by Jadersten et al thrombocytopenia in MDS needs further evaluation. indicated improved survival in patients with low-risk In addition, further determination of the effects of MDS with low transfusion need treated with these these therapeutic interventions on patient QOL is 87 agents. The study by Park et al further indicated important.172,174,175,181,182 Progress toward improving improved survival and decreased AML progression management of MDS has occurred over the past few of patients with IPSS low/ INT-1 disease treated with years, and more of these advances are anticipated us- Epo/G-CSF compared with the historical control ing these guidelines as a framework for coordinating IMRAW database patients. Thus, the data do not in- comparative clinical trials. dicate a negative impact of these drugs for treatment of MDS. Given these data, the NCCN MDS Panel endorses and reiterates the prior recommendations References for ESA use in the management of symptomatic ane- 1. National Cancer Institute. SEER Cancer Statistics Review mia in MDS, with a target hemoglobin range of 10 to 1975-2009: myelodysplastic syndromes (MDS), chronic 12 g/dL; the target should not exceed 12 g/dL. myeloproliferative disorders (CMD), and chronic myelomonocytic In July 2007, the Centers for Medicare & Medic- leukemia (CMML). 2010. Available at: http://seer.cancer.gov/ csr/1975_2007/results_merged/sect_30_mds.pdf. Accessed April aid Services modified the scope of their decision re- 2013. garding use of ESAs in cancer and related neoplastic 2. Ma X, Does M, Raza A, Mayne ST. Myelodysplastic syndromes: conditions to make no national coverage determina- incidence and survival in the United States. Cancer tion (NCD) on the use of ESAs in MDS (ie, not re- 2007;109:1536–1542. stricting ESA use in MDS through the NCD). Thus, 3. Greenberg P. The myelodysplastic syndromes. In: Hoffman R, Benz E, Shattil S, et al. Hematology: Basic Principles and Practice. local Medicare contractors may continue to make 3rd ed. New York, NY: Churchill Livingstone; 2000:1106–1129. reasonable and necessary determinations on uses of 4. Swerdlow S, Campo E, Harris NL, et al. WHO Classification of ESAs that are not determined by the NCD. Tumours of Haematopoietic and Lymphoid Tissues. 4th ed. Lyon, Clinical trials with other experimental agents France: IARC; 2008. that are reportedly capable of increasing hemoglo- 5. Brunning R, Orazi A, Germing U, et al. Myelodysplastic syndromes/neoplasms. In: Swerdlow SH, Campo E, Harris NL, et bin levels should be explored in patients whose dis- al, eds. WHO Classification of Tumours of Haematopoietic and ease is not responding to standard therapy. These Lymphoid Tissues. 4th ed. Lyon, France: IARC Press; 2008:88– drugs should be used in the context of therapeutic 103. approaches for the patient’s underlying prognostic 6. Vardiman JW, Thiele J, Arber DA, et al. The 2008 revision of the risk group. WHO classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood 2009;114:937–951. 7. Brunning R, Bennett J, Flandrin G, et al. Myelodysplastic syndromes. In: Jaffe E, Harris N, Stein H, et al, eds. WHO Summary Classification of Tumours. Pathology and Genetics of These suggested practice guidelines are based on ex- Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2001:61–73. tensive evaluation of the reviewed risk-based data 8. Harris NL, Jaffe ES, Diebold J, et al. World Health Organization and indicate current approaches for managing pa- classification of neoplastic diseases of the hematopoietic and tients with MDS. Four drugs were recently approved lymphoid tissues: report of the Clinical Advisory Committee

Myelodysplastic Syndromes

meeting-Airlie House, Virginia, November 1997. J Clin Oncol 24. Dunn DE, Tanawattanacharoen P, Boccuni P, et al. Paroxysmal 1999;17:3835–3849. nocturnal hemoglobinuria cells in patients with bone marrow 9. Vardiman JW, Harris NL, Brunning RD. The World Health failure syndromes. Ann Intern Med 1999;131:401–408. Organization (WHO) classification of the myeloid neoplasms. 25. Saunthararajah Y, Nakamura R, Nam JM, et al. HLA-DR15 Blood 2002;100:2292–2302. (DR2) is overrepresented in myelodysplastic syndrome and 10. Greenberg P, Cox C, LeBeau MM, et al. International scoring aplastic anemia and predicts a response to immunosuppression in myelodysplastic syndrome. Blood 2002;100:1570–1574. system for evaluating prognosis in myelodysplastic syndromes [published correction appears in Blood 1998;91:1100]. Blood 26. Borowitz MJ, Craig FE, Digiuseppe JA, et al. Guidelines for the 1997;89:2079–2088. diagnosis and monitoring of paroxysmal nocturnal hemoglobinuria and related disorders by flow cytometry. Cytometry B Clin Cytom 11. List A, Dewald G, Bennett J, et al. Lenalidomide in the 2010;78:211–230. myelodysplastic syndrome with chromosome 5q deletion. N Engl 27. Takeda J, Miyata T, Kawagoe K, et al. Deficiency of the GPI J Med 2006;355:1456–1465. anchor caused by a somatic mutation of the PIG-A gene in 12. Vardiman JW, Bennett JM, Bain BJ, et al. Myelodysplastic/ paroxysmal nocturnal hemoglobinuria. Cell 1993;73:703–711. myeloproliferative neoplasms. In: Swerdlow SH, Campo E, Harris 28. Ware RE, Rosse WF, Howard TA. Mutations within the Piga gene NL, et al, eds. WHO Classification of Tumours of Haematopoietic in patients with paroxysmal nocturnal hemoglobinuria. Blood and Lymphoid Tissues. 4th ed. Lyon, France: IARC; 2008:75–86. 1994;83:2418–2422. 13. Vardiman JW, Bennett JM, Bain BJ, et al. Myelodysplastic/ 29. Battiwalla M, Hepgur M, Pan D, et al. Multiparameter flow myeloproliferative neoplasm, unclassifiable. In: Swerdlow SH, cytometry for the diagnosis and monitoring of small GPI-deficient Campo E, Harris NL, et al, eds. WHO classification of tumours cellular populations. Cytometry B Clin Cytom 2010;78:348–356. of haematopoietic and lymphoid tissues. 4th ed. Lyon, France: 30. Kussick SJ, Fromm JR, Rossini A, et al. Four-color flow cytometry IARC; 2008:85–86. shows strong concordance with bone marrow morphology and 14. Albitar M, Manshouri T, Shen Y, et al. Myelodysplastic syndrome cytogenetics in the evaluation for myelodysplasia. Am J Clin is not merely “preleukemia”. Blood 2002;100:791–798. Pathol 2005;124:170–181. 15. Greenberg P, Anderson J, de Witte T, et al. Problematic WHO 31. van de Loosdrecht AA, Alhan C, Bene MC, et al. Standardization reclassification of myelodysplastic syndromes. Members of the of flow cytometry in myelodysplastic syndromes: report from International MDS Study Group. J Clin Oncol 2000;18:3447– the first European LeukemiaNet working conference on flow 3452. cytometry in myelodysplastic syndromes. Haematologica 16. Arber DA, Brunning RD, Orazi A, et al. Acute myeloid leukaemia 2009;94:1124–1134. with myelodysplasia-related changes. In Chapter 6: Acute 32. Westers TM, Ireland R, Kern W, et al. Standardization of flow myeloid leukemia and related precursor neoplasms. In: Swerdlow cytometry in myelodysplastic syndromes: a report from an S, Campo E, Harris NL, et al, eds. World Health Organization international consortium and the European LeukemiaNet Classification of Tumours of Haematopoietic and Lymphoid Working Group. Leukemia 2012;26:1730–1741. Tissues, 4th ed. Lyon, France: IARC Press; 2008:124–126. 33. Wood BL. Myeloid malignancies: myelodysplastic syndromes, 17. Germing U, Gattermann N, Strupp C, et al. Validation of the myeloproliferative disorders, and acute myeloid leukemia. Clin WHO proposals for a new classification of primary myelodysplastic Lab Med 2007;27:551–575, vii. syndromes: a retrospective analysis of 1600 patients. Leuk Res 34. Wood BL, Arroz M, Barnett D, et al. 2006 Bethesda International 2000;24:983–992. Consensus recommendations on the immunophenotypic analysis of hematolymphoid neoplasia by flow cytometry: optimal reagents 18. Germing U, Strupp C, Kuendgen A, et al. Refractory anaemia and reporting for the flow cytometric diagnosis of hematopoietic with excess of blasts (RAEB): analysis of reclassification according neoplasia. Cytometry B Clin Cytom 2007;72(Suppl 1):S14–22. to the WHO proposals. Br J Haematol 2006;132:162–167. 35. Della Porta MG, Picone C, Pascutto C, et al. Multicenter 19. Germing U, Strupp C, Kuendgen A, et al. Prospective validation validation of a reproducible flow cytometric score for the diagnosis of the WHO proposals for the classification of myelodysplastic of low-grade myelodysplastic syndromes: results of a European syndromes. Haematologica 2006;91:1596–1604. LeukemiaNET study. Haematologica 2012;97:1209–1217. 20. Malcovati L, Porta MG, Pascutto C, et al. Prognostic factors and 36. Chan WC, Foucar K, Morice WG, Catovsky D. T-cell large life expectancy in myelodysplastic syndromes classified according granular lymphocytic leukemia. In: Swerdlow SH, Campo E, Harris to WHO criteria: a basis for clinical decision making. J Clin NL, et al, eds. WHO classification of tumours of haematopoietic Oncol 2005;23:7594–7603. and lymphoid tissues, 4th ed. Lyon, France: IARC; 2008:272–273. 21. Muller-Berndorff H, Haas PS, Kunzmann R, et al. Comparison 37. Du HY, Pumbo E, Ivanovich J, et al. TERC and TERT gene of five prognostic scoring systems, the French-American-British mutations in patients with bone marrow failure and the significance (FAB) and World Health Organization (WHO) classifications in of telomere length measurements. Blood 2009;113:309–316. patients with myelodysplastic syndromes: results of a single-center 38. Vulliamy TJ, Marrone A, Knight SW, et al. Mutations in analysis. Ann Hematol 2006;85:502–513. dyskeratosis congenita: their impact on telomere length and the 22. Valent P, Horny HP, Bennett JM, et al. Definitions and standards diversity of clinical presentation. Blood 2006;107:2680–2685. in the diagnosis and treatment of the myelodysplastic syndromes: 39. Alter BP, Baerlocher GM, Savage SA, et al. Very short telomere consensus statements and report from a working conference. Leuk length by flow fluorescence in situ hybridization identifies patients Res 2007;31:727–736. with dyskeratosis congenita. Blood 2007;110:1439–1447. 23. Jafarzadeh A, Poorgholami M, Izadi N, et al. Immunological 40. Michaud J, Wu F, Osato M, et al. In vitro analyses of known and hematological changes in patients with hyperthyroidism or and novel RUNX1/AML1 mutations in dominant familial hypothyroidism. Clin Invest Med 2010;33:E271–279. platelet disorder with predisposition to acute myelogenous

Myelodysplastic Syndromes

leukemia: implications for mechanisms of pathogenesis. Blood 58. Prodan CI, Bottomley SS, Vincent AS, et al. Hypocupremia 2002;99:1364–1372. associated with prior vitamin B12 deficiency. Am J Hematol 41. Song WJ, Sullivan MG, Legare RD, et al. Haploinsufficiency 2007;82:288–290. of CBFA2 causes familial thrombocytopenia with propensity to 59. Sanz GF, Sanz MA, Greenberg PL. Prognostic factors and develop acute myelogenous leukaemia. Nat Genet 1999;23:166– scoring systems in myelodysplastic syndromes. Haematologica 175. 1998;83:358–368. 42. Liew E, Owen C. Familial myelodysplastic syndromes: a review of 60. Bennett JM, Catovsky D, Daniel MT, et al. The chronic myeloid the literature. Haematologica 2011;96:1536–1542. leukaemias: guidelines for distinguishing chronic granulocytic, 43. Quentin S, Cuccuini W, Ceccaldi R, et al. Myelodysplasia and atypical chronic myeloid, and chronic myelomonocytic leukaemia. leukemia of Fanconi anemia are associated with a specific pattern Proposals by the French-American-British Cooperative of genomic abnormalities that includes cryptic RUNX1/AML1 Leukaemia Group. Br J Haematol 1994;87:746–754. lesions. Blood 2011;117:e161–170. 61. Malcovati L, Germing U, Kuendgen A, et al. Time-dependent 44. Fadilah SA, Cheong SK, Roslan H, et al. GATA-1 and GATA- prognostic scoring system for predicting survival and leukemic 2 gene expression is related to the severity of dysplasia in evolution in myelodysplastic syndromes. J Clin Oncol myelodysplastic syndrome. Leukemia 2002;16:1563–1565. 2007;25:3503–3510. 45. Hahn CN, Chong CE, Carmichael CL, et al. Heritable GATA2 62. Kao JM, McMillan A, Greenberg PL. International MDS mutations associated with familial myelodysplastic syndrome and risk analysis workshop (IMRAW)/IPSS reanalyzed: impact of acute myeloid leukemia. Nat Genet 2011;43:1012–1017. cytopenias on clinical outcomes in myelodysplastic syndromes. Am J Hematol 2008;83:765–770. 46. Baxter EJ, Kulkarni S, Vizmanos JL, et al. Novel translocations that disrupt the platelet-derived growth factor receptor beta (PDGFRB) 63. Alessandrino EP, Della Porta MG, Bacigalupo A, et al. WHO gene in BCR-ABL-negative chronic myeloproliferative disorders. classification and WPSS predict posttransplantation outcome Br J Haematol 2003;120:251–256. in patients with myelodysplastic syndrome: a study from the Gruppo Italiano Trapianto di Midollo Osseo (GITMO). Blood 47. Steer EJ, Cross NC. Myeloproliferative disorders with 2008;112:895–902. translocations of chromosome 5q31-35: role of the platelet-derived growth factor receptor Beta. Acta Haematol 2002;107:113–122. 64. Cermak J, Kacirkova P, Mikulenkova D, Michalova K. Impact of transfusion dependency on survival in patients with early 48. Apperley JF, Gardembas M, Melo JV, et al. Response to imatinib myelodysplastic syndrome without excess of blasts. Leuk Res mesylate in patients with chronic myeloproliferative diseases with 2009;33:1469–1474. rearrangements of the platelet-derived growth factor receptor beta. N Engl J Med 2002;347:481–487. 65. Park MJ, Kim HJ, Kim SH, et al. Is International Prognostic Scoring System (IPSS) still standard in predicting prognosis in 49. David M, Cross NC, Burgstaller S, et al. Durable responses to patients with myelodysplastic syndrome? External validation of the imatinib in patients with PDGFRB fusion gene-positive and WHO Classification-Based Prognostic Scoring System (WPSS) BCR-ABL-negative chronic myeloproliferative disorders. Blood and comparison with IPSS. Eur J Haematol 2008;81:364–373. 2007;109:61–64. 66. Malcovati L, Della Porta MG, Strupp C, et al. Impact of the 50. Magnusson MK, Meade KE, Nakamura R, et al. Activity of degree of anemia on the outcome of patients with myelodysplastic STI571 in chronic myelomonocytic leukemia with a platelet- syndrome and its integration into the WHO classification- derived growth factor beta receptor fusion oncogene. Blood based Prognostic Scoring System (WPSS). Haematologica 2002;100:1088–1091. 2011;96:1433–1440. 51. Steensma DP, Dewald GW, Lasho TL, et al. The JAK2 V617F 67. Greenberg PL, Tuechler H, Schanz J, et al. Revised International activating tyrosine kinase mutation is an infrequent event in Prognostic Scoring System (IPSS-R) for myelodysplastic both “atypical” myeloproliferative disorders and myelodysplastic syndromes. Blood 2012;120:2454–2465. syndromes. Blood 2005;106:1207–209. 68. Schanz J, Tuchler H, Sole F, et al. New comprehensive cytogenetic 52. Szpurka H, Tiu R, Murugesan G, et al. Refractory anemia with scoring system for primary myelodysplastic syndromes (MDS) and ringed sideroblasts associated with marked thrombocytosis oligoblastic acute myeloid leukemia after MDS derived from an (RARS-T), another myeloproliferative condition characterized international database merge. J Clin Oncol 2012;30:820–829. by JAK2 V617F mutation. Blood 2006;108:2173–2181. 69. Ades L, Lamarque M, Raynaud S, et al. Revised-IPSS (IPSS-R) is a 53. Ogata K, Nakamura K, Yokose N, et al. Clinical significance of powerful tool to evaluate the outcome of mds patient treated with phenotypic features of blasts in patients with myelodysplastic azacitidine (AZA): the Groupe Francophone Des Myelodysplasies syndrome. Blood 2002;100:3887–3896. (GFM) experience [abstract]. Blood 2012;120:Abstract 422. 54. Wells DA, Benesch M, Loken MR, et al. Myeloid and 70. Cermak J, Mikulenkova D, Brezinova J, Michalova K. A monocytic dyspoiesis as determined by flow cytometric scoring reclassification of myelodysplastic syndrome (MDS) patients of in myelodysplastic syndrome correlates with the IPSS and with RAEB-1 subgroup according to IPSS-R improves discrimination outcome after hematopoietic stem cell transplantation. Blood of high risk patients and better predicts overall survival. 2003;102:394–403. A retrospective analysis of 49 patients [abstract]. Blood 55. Gregg XT, Reddy V, Prchal JT. Copper deficiency masquerading as 2012;120:Abstract 4957. myelodysplastic syndrome. Blood 2002;100:1493–1495. 71. Messa E, Gioia D, Evangelista A, et al. High predictive value of 56. Haddad AS, Subbiah V, Lichtin AE, et al. Hypocupremia and the revised International Prognostic Scoring System (IPSS-R): an bone marrow failure. Haematologica 2008;93:e1–5. external analysis of 646 patients from a multiregional italian MDS 57. Koca E, Buyukasik Y, Cetiner D, et al. Copper deficiency with registry [abstract]. Blood 2012;120:Abstract 1702. increased hematogones mimicking refractory anemia with excess 72. Mishra A, Ali NHA, Corrales-Yepez M, et al. Validation of the blasts. Leuk Res 2008;32:495–499. revised International Prognostic Scoring System (R-IPSS) for

Myelodysplastic Syndromes

patients with myelodysplastic syndromes: therapeutic implications 87. Park S, Grabar S, Kelaidi C, et al. Predictive factors of response and [abstract]. Blood 2012;120:Abstract 2816. survival in myelodysplastic syndrome treated with erythropoietin 73. Valcarcel D, Sanz G, Ortega M, et al. Identification of poor risk and G-CSF: the GFM experience. Blood 2008;111:574–582. patients in low and intermediate-1 (Int-1) IPSS MDS with the 88. Tehranchi R, Fadeel B, Schmidt-Mende J, et al. Antiapoptotic role new IPSSR index and comparison with other prognostic indexes. of growth factors in the myelodysplastic syndromes: concordance A study by the Spanish Group of MDS (GESMD) [abstract]. between in vitro and in vivo observations. Clin Cancer Res Blood 2012;120:Abstract 702. 2005;11:6291–6299. 74. Warlick ED, Hirsch BA, Nguyen PL, et al. Comparison of 89. Houwerzijl EJ, Blom NR, van der Want JJ, et al. Increased IPSS and IPSS-R scoring in a population based myelodysplastic peripheral platelet destruction and caspase-3-independent syndromes (MDS) study [abstract]. Blood 2012;120:Abstract programmed cell death of bone marrow megakaryocytes in 3841. myelodysplastic patients. Blood 2005;105:3472–3479. 75. Garcia-Manero G, Shan J, Faderl S, et al. A prognostic score for 90. Tamura H, Ogata K, Luo S, et al. Plasma thrombopoietin (TPO) patients with lower risk myelodysplastic syndrome. Leukemia levels and expression of TPO receptor on platelets in patients with 2008;22:538–543. myelodysplastic syndromes. Br J Haematol 1998;103:778–784. 76. Bejar R, Stevenson KE, Caughey BA, et al. Validation of a 91. Zwierzina H, Rollinger-Holzinger I, Nuessler V, et al. Endogenous prognostic model and the impact of mutations in patients serum thrombopoietin concentrations in patients with with lower-risk myelodysplastic syndromes. J Clin Oncol myelodysplastic syndromes. Leukemia 1998;12:59–64. 2012;30:3376–3382. 92. Greenberg PL, Garcia-Manero G, Moore M, et al. A randomized 77. Komrokji RS, Corrales-Yepez M, Al Ali NH, et al. Validation controlled trial of romiplostim in patients with low- or of the lower risk MD Anderson prognostic scoring system for intermediate-risk myelodysplastic syndrome receiving decitabine. Leuk Lymphoma 2013;54:321–328. patients with myelodysplastic syndromes [abstract]. Blood 2012;120:Abstract 3826. 93. Kantarjian H, Fenaux P, Sekeres MA, et al. Safety and efficacy of romiplostim in patients with lower-risk myelodysplastic syndrome 78. Sekeres MA, Elson P, Tiu RV, et al. Validating the lower-risk and thrombocytopenia. J Clin Oncol 2010;28:437–444. MD Anderson prognostic scoring system (LR-PSS) and the revised International Prognostic Scoring System (IPSS-R) 94. Kantarjian HM, Giles FJ, Greenberg PL, et al. Phase 2 study for patients with myelodysplastic syndromes [abstract]. Blood of romiplostim in patients with low- or intermediate-risk 2011;118:Abstract 1720. myelodysplastic syndrome receiving azacitidine therapy. Blood 2010;116:3163–3170. 79. Cheson BD, Bennett JM, Kantarjian H, et al. Report of an international working group to standardize response criteria for 95. Sekeres MA, Kantarjian H, Fenaux P, et al. Subcutaneous or intravenous administration of romiplostim in thrombocytopenic myelodysplastic syndromes. Blood 2000;96:3671–3674. patients with lower risk myelodysplastic syndromes. Cancer 80. Cheson BD, Greenberg PL, Bennett JM, et al. Clinical application 2011;117:992–1000. and proposal for modification of the International Working Group 96. Wang ES, Lyons RM, Larson RA, et al. A randomized, double- (IWG) response criteria in myelodysplasia. Blood 2006;108:419– blind, placebo-controlled phase 2 study evaluating the efficacy 425. and safety of romiplostim treatment of patients with low 81. Greenberg P, Baer, M, Bennett, J et al. NCCN Clinical Practice or intermediate-1 risk myelodysplastic syndrome receiving Guidelines in Oncology for Myelodysplastic Syndromes. Version lenalidomide. J Hematol Oncol 2012;5:71. 1, 2001. In: The Complete Library of NCCN Guidelines [CD- 97. Mavroudi I, Pyrovolaki K, Pavlaki K, et al. Effect of the ROM]. Rockledge, PA: NCCN: 2001. nonpeptide thrombopoietin receptor agonist eltrombopag on 82. Greenberg P. The role of hemopoietic growth factors in the megakaryopoiesis of patients with lower risk myelodysplastic treatment of myelodysplastic syndromes. International Journal of syndrome. Leuk Res 2011;35:323–328. Pediatric Hematology/Oncology 1997;4:231–238. 98. Will B, Kawahara M, Luciano JP, et al. Effect of the nonpeptide 83. Hellstrom-Lindberg E, Ahlgren T, Beguin Y, et al. Treatment of thrombopoietin receptor agonist Eltrombopag on bone anemia in myelodysplastic syndromes with granulocyte colony- marrow cells from patients with acute myeloid leukemia and stimulating factor plus erythropoietin: results from a randomized myelodysplastic syndrome. Blood 2009;114:3899–3908. phase II study and long-term follow-up of 71 patients. Blood 99. Hashimoto S, Toba K, Fuse I, et al. Thrombopoietin activates 1998;92:68–75. the growth of megakaryoblasts in patients with chronic 84. Jadersten M, Malcovati L, Dybedal I, et al. Erythropoietin and myeloproliferative disorders and myelodysplastic syndrome. Eur J granulocyte-colony stimulating factor treatment associated with Haematol 2000;64:225–230. improved survival in myelodysplastic syndrome. J Clin Oncol 100. Luo SS, Ogata K, Yokose N, et al. Effect of thrombopoietin 2008;26:3607–3613. on proliferation of blasts from patients with myelodysplastic 85. Kelaidi C, Beyne-Rauzy O, Braun T, et al. High response rate syndromes. Stem Cells 2000;18:112–119. and improved exercise capacity and quality of life with a new 101. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. Efficacy of regimen of darbepoetin alfa with or without filgrastim in lower- azacitidine compared with that of conventional care regimens risk myelodysplastic syndromes: a phase II study by the GFM. Ann in the treatment of higher-risk myelodysplastic syndromes: Hematol 2013;92:621–631. a randomised, open-label, phase III study. Lancet Oncol 86. Kelaidi C, Park S, Brechignac S, et al. Treatment of myelodysplastic 2009;10:223–232. syndromes with 5q deletion before the lenalidomide era; the GFM 102. Kantarjian H, Issa JP, Rosenfeld CS, et al. Decitabine improves experience with EPO and thalidomide. Leuk Res 2008;32:1049– patient outcomes in myelodysplastic syndromes: results of a phase 1053. III randomized study. Cancer 2006;106:1794–1803.

Myelodysplastic Syndromes

103. Lubbert M, Suciu S, Baila L, et al. Low-dose decitabine versus best posttransplantation outcome in patients with MDS. Biol Blood supportive care in elderly patients with intermediate- or high- Marrow Transplant 2012;18:1211–1218. risk myelodysplastic syndrome (MDS) ineligible for intensive 119. Lubbert M, Bertz H, Ruter B, et al. Non-intensive treatment with chemotherapy: final results of the randomized phase III study of low-dose 5-aza-2’-deoxycytidine (DAC) prior to allogeneic blood the European Organisation for Research and Treatment of Cancer SCT of older MDS/AML patients. Bone Marrow Transplant Leukemia Group and the German MDS Study Group. J Clin 2009;44:585–588. Oncol 2011;29:1987–1996. 120. Deeg HJ, Gotlib J, Beckham C, et al. Soluble TNF receptor 104. Silverman LR, Demakos EP, Peterson BL, et al. Randomized fusion protein (etanercept) for the treatment of myelodysplastic controlled trial of azacitidine in patients with the myelodysplastic syndrome: a pilot study. Leukemia 2002;16:162–164. syndrome: a study of the cancer and leukemia group B. J Clin Oncol 2002;20:2429–2440. 121. Deeg HJ, Jiang PY, Holmberg LA, et al. Hematologic responses of patients with MDS to antithymocyte globulin plus etanercept 105. Silverman LR, McKenzie DR, Peterson BL, et al. Further correlate with improved flow scores of marrow cells. Leuk Res analysis of trials with azacitidine in patients with myelodysplastic 2004;28:1177–1180. syndrome: studies 8421, 8921, and 9221 by the Cancer and Leukemia Group B. J Clin Oncol 2006;24:3895–3903. 122. Jonásová A, Neuwirtová R, Cermák J, et al. 164 Promising cyclosporin a therapy for myelodysplastic syndrome. Leukemia 106. Silverman LR, Fenaux P, Mufti GJ, et al. Continued azacitidine therapy beyond time of first response improves quality of response Research 1997;21:S53. in patients with higher-risk myelodysplastic syndromes. Cancer 123. Molldrem JJ, Caples M, Mavroudis D, et al. Antithymocyte 2011;117:2697–2702. globulin for patients with myelodysplastic syndrome. Br J 107. Lyons RM, Cosgriff TM, Modi SS, et al. Hematologic response to Haematol 1997;99:699–705. three alternative dosing schedules of azacitidine in patients with 124. Raza A, Meyer P, Dutt D, et al. Thalidomide produces transfusion myelodysplastic syndromes. J Clin Oncol 2009;27:1850–1856. independence in long-standing refractory anemias of patients 108. Martin MG, Walgren RA, Procknow E, et al. A phase II study with myelodysplastic syndromes. Blood 2001;98:958–965. of 5-day intravenous azacitidine in patients with myelodysplastic 125. Strupp C, Germing U, Aivado M, et al. Thalidomide for the syndromes. Am J Hematol 2009;84:560–564. treatment of patients with myelodysplastic syndromes. Leukemia 109. Lubbert M, Wijermans P, Kunzmann R, et al. Cytogenetic 2002;16:1–6. responses in high-risk myelodysplastic syndrome following low- 126. Sloand EM, Wu CO, Greenberg P, et al. Factors affecting dose treatment with the DNA methylation inhibitor 5-aza-2’- response and survival in patients with myelodysplasia treated with deoxycytidine. Br J Haematol 2001;114:349–357. immunosuppressive therapy. J Clin Oncol 2008;26:2505–2511. 110. Wijermans P, Lubbert M, Verhoef G, et al. Low-dose 5-aza-2’- 127. Scheinberg P, Nunez O, Weinstein B, et al. Horse versus rabbit deoxycytidine, a DNA hypomethylating agent, for the treatment antithymocyte globulin in acquired aplastic anemia. N Engl J Med of high-risk myelodysplastic syndrome: a multicenter phase II 2011;365:430–438. study in elderly patients. J Clin Oncol 2000;18:956–962. 128. Scheinberg P, Wu CO, Scheinberg P, et al. A randomized trial 111. Saba H, Rosenfeld C, Issa J, et al. First report of the phase III of horse versus rabbit antithymocyte globulin in severe acquired north American trial of decitabine in advanced myelodysplastic aplastic anemia [abstract]. Blood 2010;116:Abstract LBA-4. syndrome (MDS) [abstract]. Blood 2004;104:Abstract 67. 129. Stadler M, Germing U, Kliche KO, et al. A prospective, 112. van den Bosch J, Lubbert M, Verhoef G, Wijermans PW. The randomised, phase II study of horse antithymocyte globulin vs effects of 5-aza-2’-deoxycytidine (Decitabine) on the platelet rabbit antithymocyte globulin as immune-modulating therapy count in patients with intermediate and high-risk myelodysplastic in patients with low-risk myelodysplastic syndromes. Leukemia syndromes. Leuk Res 2004;28:785–790. 2004;18:460–465. 113. Saba H, Lubbert M, Wijermans PW. Response rates of phase 2 and 130. List A, Kurtin S, Roe DJ, et al. Efficacy of lenalidomide in phase 3 trials of decitabine (DAC) in patients with myelodysplastic myelodysplastic syndromes. N Engl J Med 2005;352:549–557. syndromes (MDS) [abstract]. Blood 2005;106:Abstract 2515. 131. Nimer SD. Clinical management of myelodysplastic syndromes 114. Kantarjian HM, O’Brien S, Shan J, et al. Update of the decitabine with interstitial deletion of chromosome 5q. J Clin Oncol experience in higher risk myelodysplastic syndrome and 2006;24:2576–2582. analysis of prognostic factors associated with outcome. Cancer 2007;109:265–273. 132. Fenaux P, Giagounidis A, Selleslag D, et al. A randomized phase 115. Kantarjian H, Oki Y, Garcia-Manero G, et al. Results of a 3 study of lenalidomide versus placebo in RBC transfusion- randomized study of 3 schedules of low-dose decitabine in higher- dependent patients with low-/intermediate-1-risk myelodysplastic risk myelodysplastic syndrome and chronic myelomonocytic syndromes with del5q. Blood 2011;118:3765–3776. leukemia. Blood 2007;109:52–57. 133. Kuendgen A, Lauseker M, List AF, et al. Lenalidomide does not 116. Damaj G, Duhamel A, Robin M, et al. Impact of azacitidine increase AML progression risk in RBC transfusion-dependent before allogeneic stem-cell transplantation for myelodysplastic patients with low- or intermediate-1-risk MDS with del(5q): a syndromes: a study by the Societe Francaise de Greffe de Moelle comparative analysis. Leukemia 2013;27:1072–1079. et de Therapie-Cellulaire and the Groupe-Francophone des 134. Raza A, Reeves JA, Feldman EJ, et al. Phase 2 study of lenalidomide Myelodysplasies. J Clin Oncol 2012;30:4533–4540. in transfusion-dependent, low-risk, and intermediate-1 risk 117. Field T, Perkins J, Huang Y, et al. 5-Azacitidine for myelodysplasia myelodysplastic syndromes with karyotypes other than deletion before allogeneic hematopoietic cell transplantation. Bone 5q. Blood 2008;111:86–93. Marrow Transplant 2010;45:255–260. 135. Tricot G, Boogaerts MA. The role of aggressive chemotherapy in 118. Gerds AT, Gooley TA, Estey EH, et al. Pretransplantation the treatment of the myelodysplastic syndromes. Br J Haematol therapy with azacitidine vs induction chemotherapy and 1986;63:477–483.

Myelodysplastic Syndromes

136. Estey EH, Thall PF, Cortes JE, et al. Comparison of idarubicin outcome analysis in 510 transplants facilitated by the National + ara-C-, fludarabine + ara-C-, and topotecan + ara-C-based Marrow Donor Program. Blood 2002;99:1943–1951. regimens in treatment of newly diagnosed acute myeloid 151. Deeg H. Optimization of transplant regimens for patients with leukemia, refractory anemia with excess blasts in transformation, myelodysplastic syndromes (MDS). In: Hematology 2005: or refractory anemia with excess blasts. Blood 2001;98:3575– American Society of Hematology Education Program Book: 3583. Washington, DC: American Society of Hematology; 2005;167– 137. Sonneveld P, van Dongen JJ, Hagemeijer A, et al. High 173. expression of the multidrug resistance P-glycoprotein in high-risk 152. Deeg HJ, Scott BL, Fang M, et al. Five-group cytogenetic myelodysplasia is associated with immature phenotype. Leukemia risk classification, monosomal karyotype, and outcome after 1993;7:963–969. hematopoietic cell transplantation for MDS or acute leukemia 138. Advani R, Saba HI, Tallman MS, et al. Treatment of refractory evolving from MDS. Blood 2012;120:1398–1408. and relapsed acute myelogenous leukemia with combination chemotherapy plus the multidrug resistance modulator PSC 833 153. de Lima M, Anagnostopoulos A, Munsell M, et al. Nonablative (Valspodar). Blood 1999;93:787–795. versus reduced-intensity conditioning regimens in the treatment of acute myeloid leukemia and high-risk myelodysplastic syndrome: 139. Wattel E, Solary E, Hecquet B, et al. Quinine improves results dose is relevant for long-term disease control after allogeneic of intensive chemotherapy (IC) in myelodysplastic syndromes hematopoietic stem cell transplantation. Blood 2004;104:865– (MDS) expressing P-glycoprotein (PGP). Updated results of a randomized study. Groupe Francais des Myelodysplasies (GFM) 872. and Groupe GOELAMS. Adv Exp Med Biol 1999;457:35–46. 154. Giralt S, Estey E, Albitar M, et al. Engraftment of allogeneic 140. Greenberg PL, Lee SJ, Advani R, et al. Mitoxantrone, etoposide, hematopoietic progenitor cells with purine analog-containing and cytarabine with or without valspodar in patients with chemotherapy: harnessing graft-versus-leukemia without relapsed or refractory acute myeloid leukemia and high-risk myeloablative therapy. Blood 1997;89:4531–4536. myelodysplastic syndrome: a phase III trial (E2995). J Clin Oncol 155. Litzow MR, Tarima S, Perez WS, et al. Allogeneic transplantation 2004;22:1078–1086. for therapy-related myelodysplastic syndrome and acute myeloid 141. Anderson JE, Appelbaum FR, Fisher LD, et al. Allogeneic bone leukemia. Blood 2010;115:1850–1857. marrow transplantation for 93 patients with myelodysplastic 156. McClune BL, Weisdorf DJ, Pedersen TL, et al. Effect of age on syndrome. Blood 1993;82:677–681. outcome of reduced-intensity hematopoietic cell transplantation 142. De Witte T, Zwaan F, Hermans J, et al. Allogeneic bone marrow for older patients with acute myeloid leukemia in first complete transplantation for secondary leukaemia and myelodysplastic remission or with myelodysplastic syndrome. J Clin Oncol syndrome: a survey by the Leukaemia Working Party of the 2010;28:1878–1887. European Bone Marrow Transplantation Group (EBMTG). Br J 157. Sorror ML, Sandmaier BM, Storer BE, et al. Comorbidity Haematol 1990;74:151–155. and disease status based risk stratification of outcomes among 143. Demuynck H, Verhoef GE, Zachee P, et al. Treatment of patients patients with acute myeloid leukemia or myelodysplasia receiving with myelodysplastic syndromes with allogeneic bone marrow allogeneic hematopoietic cell transplantation. J Clin Oncol transplantation from genotypically HLA-identical sibling and 2007;25:4246–4254. alternative donors. Bone Marrow Transplant 1996;17:745–751. 158. De Witte T, Suciu S, Verhoef G, et al. Intensive chemotherapy 144. Jurado M, Deeg HJ, Storer B, et al. Hematopoietic stem cell followed by allogeneic or autologous stem cell transplantation transplantation for advanced myelodysplastic syndrome after for patients with myelodysplastic syndromes (MDSs) and acute conditioning with busulfan and fractionated total body irradiation myeloid leukemia following MDS. Blood 2001;98:2326–2331. is associated with low relapse rate but considerable nonrelapse mortality. Biol Blood Marrow Transplant 2002;8:161–169. 159. Fukumoto JS, Greenberg PL. Management of patients with higher risk myelodysplastic syndromes. Crit Rev Oncol Hematol 145. Kerbauy DM, Chyou F, Gooley T, et al. Allogeneic hematopoietic 2005;56:179–192. cell transplantation for chronic myelomonocytic leukemia. Biol Blood Marrow Transplant 2005;11:713–720. 160. Alyea EP, Kim HT, Ho V, et al. Comparative outcome of nonmyeloablative and myeloablative allogeneic hematopoietic 146. Nevill TJ, Fung HC, Shepherd JD, et al. Cytogenetic abnormalities in primary myelodysplastic syndrome are highly predictive of cell transplantation for patients older than 50 years of age. Blood outcome after allogeneic bone marrow transplantation. Blood 2005;105:1810–1814. 1998;92:1910–1917. 161. Cutler CS, Lee SJ, Greenberg P, et al. A decision analysis of 147. Scott BL, Sandmaier BM, Storer B, et al. Myeloablative vs allogeneic bone marrow transplantation for the myelodysplastic nonmyeloablative allogeneic transplantation for patients with syndromes: delayed transplantation for low-risk myelodysplasia is myelodysplastic syndrome or acute myelogenous leukemia associated with improved outcome. Blood 2004;104:579–585. with multilineage dysplasia: a retrospective analysis. Leukemia 162. Laport GG, Sandmaier BM, Storer BE, et al. Reduced-intensity 2006;20:128–135. conditioning followed by allogeneic hematopoietic cell 148. Wallen H, Gooley TA, Deeg HJ, et al. Ablative allogeneic transplantation for adult patients with myelodysplastic syndrome hematopoietic cell transplantation in adults 60 years of age and and myeloproliferative disorders. Biol Blood Marrow Transplant older. J Clin Oncol 2005;23:3439–3446. 2008;14:246–255. 149. Anderson J, Thomas, ED, Anasetti C, et al. Unrelated donor 163. McClune B, Weisdorf D, DiPersio J, et al. Non-myeloablative marrow transplantation for myelodysplasia (MDS) and MDS- hematopoietic stem cell transplantation in older patients with related acute myeloid leukaemia. Br J Haematol 1996;93:59–67. AML and MDS: results from the Center for International Blood 150. Castro-Malaspina H, Harris RE, Gajewski J, et al. Unrelated and Marrow Transplant Research (CIBMTR) [abstract]. Blood donor marrow transplantation for myelodysplastic syndromes: 2008;112:Abstract 346.

Myelodysplastic Syndromes

164. Kindwall-Keller T, Isola LM. The evolution of hematopoietic 174. Spiriti MA, Latagliata R, Niscola P, et al. Impact of a new SCT in myelodysplastic syndrome. Bone Marrow Transplant dosing regimen of epoetin alfa on quality of life and anemia in 2009;43:597–609. patients with low-risk myelodysplastic syndrome. Ann Hematol 165. Oliansky DM, Antin JH, Bennett JM, et al. The role of cytotoxic 2005;84:167–176. therapy with hematopoietic stem cell transplantation in the 175. Hellstrom-Lindberg E, Gulbrandsen N, Lindberg G, et al. therapy of myelodysplastic syndromes: an evidence-based review. A validated decision model for treating the anaemia of Biol Blood Marrow Transplant 2009;15:137–172. myelodysplastic syndromes with erythropoietin + granulocyte 166. Deeg H, Sandmaier BM. Who is fit for allogeneic transplantation? colony-stimulating factor: significant effects on quality of life. Br J Blood 2010;116:4762–4770. Haematol 2003;120:1037–1046. 167. Beran M, Shen Y, Kantarjian H, et al. High-dose chemotherapy 176. Mannone L, Gardin C, Quarre MC, et al. High-dose darbepoetin in high-risk myelodysplastic syndrome: covariate-adjusted alpha in the treatment of anaemia of lower risk myelodysplastic comparison of five regimens. Cancer 2001;92:1999–2015. syndrome results of a phase II study. Br J Haematol 2006;133:513– 168. Wijermans P, Suciu S, Baila L, et al. Low dose decitabine versus 519. best supportive care in elderly patients with intermediate or 177. Musto P, Lanza F, Balleari E, et al. Darbepoetin alpha for the high risk MDS not eligible for intensive chemotherapy: final treatment of anaemia in low-intermediate risk myelodysplastic results of the randomized phase III study (06011) of the EORTC syndromes. Br J Haematol 2005;128:204–209. leukemia and German MDS study groups [abstract]. Blood 178. Giraldo P, Nomdedeu B, Loscertales J, et al. Darbepoetin alpha 2008;112:Abstract 226. for the treatment of anemia in patients with myelodysplastic 169. Jacobs A, Janowska-Wieczorek A, Caro J, et al. Circulating syndromes. Cancer 2006;107:2807–2816. erythropoietin in patients with myelodysplastic syndromes. Br J 179. Stasi R, Abruzzese E, Lanzetta G, et al. Darbepoetin alfa for the Haematol 1989;73:36–39. treatment of anemic patients with low- and intermediate-1-risk 170. Hellstrom-Lindberg E. Efficacy of erythropoietin in the myelodysplastic syndromes. Ann Oncol 2005;16:1921–1927. myelodysplastic syndromes: a meta-analysis of 205 patients from 180. Greenberg PL, Sun Z, Miller KB, et al. Treatment of 17 studies. Br J Haematol 1995;89:67–71. myelodysplastic syndromes patients with erythropoietin with 171. Negrin RS, Stein R, Doherty K, et al. Maintenance treatment or without granulocyte colony-stimulating factor: results of a of the anemia of myelodysplastic syndromes with recombinant prospective randomized phase III trial by the Eastern Cooperative human granulocyte colony-stimulating factor and erythropoietin: Oncology Group (E1996). Blood 2009;114:2392–2400. evidence for in vivo synergy. Blood 1996;87:4076–4081. 181. Kornblith AB, Herndon JE II, Silverman LR, et al. Impact of 172. Casadevall N, Durieux P, Dubois S, et al. Health, economic, and azacytidine on the quality of life of patients with myelodysplastic quality-of-life effects of erythropoietin and granulocyte colony- syndrome treated in a randomized phase III trial: a Cancer and stimulating factor for the treatment of myelodysplastic syndromes: Leukemia Group B study. J Clin Oncol 2002;20:2441–2452. a randomized, controlled trial. Blood 2004;104:321–327. 182. Thomas M. Health-related quality of life for those with 173. Hellstrom-Lindberg E, Negrin R, Stein R, et al. Erythroid response myelodysplastic syndrome: conceptualization, measurement and to treatment with G-CSF plus erythropoietin for the anaemia of implications. In: Greenberg PL, ed. Myelodysplastic Syndromes: patients with myelodysplastic syndromes: proposal for a predictive Clinical and Biological Advances. Cambridge, England: model. Br J Haematol 1997;99:344–351. Cambridge University Press; 2006:263–295.

Myelodysplastic Syndromes

Individual Disclosures of the NCCN Myelodysplastic Syndromes Panel Advisory Boards, Speakers Bureau, Patent, Expert Witness, or Equity, or Date Panel Member Clinical Research Support Consultant Royalty Other Completed Eyal Attar, MD None Celgene Corporation; and None None 2/7/13 Jannsen Pharmaceutica Products, LP John M. Bennett, MD None Amgen Inc.;GlaxoSmithKline None Celgene Corporation; 1/16/13 plc; Novartis AG;Onconova; andJohnson & Johnson and Pfizer Inc. Services, Inc. Clara D. Bloomfield, MD None None None None 6/10/13 Uma Borate, MD GlaxoSmithKline plc None None None 8/24/12 Carlos M. De Castro, MD Celgene Corporation Alexion Pharmaceuticals, Inc. None None 11/1/11 H. Joachim Deeg, MD None None None None 3/26/12 Olga Frankfurt, MD None None None None 1/25/12 Karin Gaensler, MD None None None None 4/10/13 Guillermo Garcia- Celgene Corporation; Merck & Co., Inc.; and None None 10/10/12 Manero, MD Merck & Co., Inc.; and Novartis AG Novartis AG Steven D. Gore, MD Celgene Corporation; Celgene Corporation Celgene None 2/1/12 and Johnson & Johnson Corporation Services, Inc. Peter L. Greenberg, MD Amgen Inc.; Amgen Inc.; and None None 1/14/13 Celgene Corporation; Novartis AG GlaxoSmithKline plc; Novartis AG; and Onconova David Head, MD None None None None 10/9/12 Rami Komrokji, MD Celgene Corporation; Celgene Corporation None None 3/20/12 GlaxoSmithKline plc; Ortho Biotech Products, L.P.; Array Biopharma Inc.; Onconova Lori J. Maness, MD Cyclacel Pharmaceuticals, None None None 3/9/12 Inc. Michael Millenson, MD None None None None 2/22/13 Margaret R. O’Donnell, National Cancer Institute None None None 5/30/13 MD Paul J. Shami, MD Eisai Inc.; and ARIAD Pharmaceuticals, None None 1/28/13 Genzyme Corporation Inc.; and Novartis AG Brady L. Stein, MD, MHS None None None None 2/11/13 Richard M. Stone, MD Celgene Corporation; ARIAD Pharmaceuticals, None None 6/3/13 Novartis AG; Inc.; Agios; and and Sunesis Seattle Genetics, Inc. Pharmaceuticals, Inc. James E. Thompson, MD None None None None 10/3/12 Peter Westervelt, MD, PhD None Celgene Corporation; and None None 11/5/12 Novartis AG Benton Wheeler, MD None None None None 2/15/13

The NCCN guidelines staff have no conflicts to disclose.