Problematic WHO Reclassification of Myelodysplastic Syndromes
2000; Lippincott Williams & Wilkins; Volume: 18; Issue: 19 Linguagem: Inglês
10.1200/jco.2000.18.19.3447
ISSN1527-7755
Autores Tópico(s)Lymphoma Diagnosis and Treatment
ResumoArticle Tools SPECIAL DEPARTMENTS Article Tools OPTIONS & TOOLS Export Citation Track Citation Add To Favorites Rights & Permissions COMPANION ARTICLES No companion articles ARTICLE CITATION DOI: 10.1200/JCO.2000.18.19.3447 Journal of Clinical Oncology - published online before print September 21, 2016 PMID: 11013289 Problematic WHO Reclassification of Myelodysplastic Syndromes Members of the International MDS Study GroupxSearch for articles by this author Members of the WHO Myeloid Disease Writing Committees and the WHO Clinical Advisory CommitteexSearch for articles by this author David HeadxDavid HeadSearch for articles by this author Show More * Members of the International MDS Study Group who contributed to this letter are as follows: Peter Greenberg, Jeanne Anderson, Theo de Witte, Elihu Estey, Pierre Fenaux, Pankaj Gupta, Terry Hamblin, Eva Hellstrom-Lindberg, Alan List, Ghulam Mufti, Radana Neuwirtova, Kazuma Ohyashiki, David Oscier, Guillermo Sanz, Miguel Sanz, and Cheryl Willman. Their affiliations are as follows: Stanford University Medical Center, Stanford, CA; University of Texas Health Sciences Center, San Antonio, TX; University Hospital St. Reboud, Nijmegen, the Netherlands; M.D. Anderson Cancer Center, Houston, TX; Services des Maladies du Sang, Lille, France; VA Medical Center, Minneapolis, MN; Royal Bournemouth Hospital, Bournemouth, England; Karolinska Institute, Huddinge, Sweden; University of Arizona Cancer Center, Tucson, AZ; King's College Hospital, London, England; Charles University Hospital, Prague, Czechoslovakia; Tokyo Medical University, Tokyo, Japan; Hospital Universitario La Fe, Valencia, Spain; and University of New Mexico Cancer Center, Albuquerque, NM.* Members of the WHO Committees who contributed to this letter are as follows: John M. Bennett, University of Rochester Cancer Center, Rochester, NY; George Flandrin, Hôpital Necker, Paris, France; Estella Matutes, Royal Marsden Hospital/Institute of Cancer Research, London, England; Clara D. Bloomfield, Ohio State University, Columbus, OH; David Head, Vanderbilt University Medical Center, Nashville, TN; Robert Pierre, University of Southern California Medical School, Los Angeles, CA; Richard D. Brunning, University of Minnesota, Minneapolis, MN; Michele Imbert, Henri Mondor Hospital/University of Paris XII, Créteil, France; and James Vardiman, University of Chicago, Chicago, IL. Vanderbilt University Medical Center Nashville, TN https://doi.org/10.1200/JCO.2000.18.19.3447 First Page Full Text PDF Figures and Tables © 2000 by American Society of Clinical OncologyjcoJ Clin OncolJournal of Clinical OncologyJCO0732-183X1527-7755American Society of Clinical OncologyjcoJ Clin OncolJournal of Clinical OncologyJCO0732-183X1527-7755American Society of Clinical OncologyResponse19102000Response19102000Reply 1:We appreciate the comments of Dr Greenberg et al regarding the classification of the myelodysplastic syndromes (MDS) proposed by the World Health Organization (WHO) Committees.1 The letter is timely, because it has engendered a re-evaluation of the proposals before their final publication, and because our response allows an opportunity to provide the rationale for some of the suggested modifications to the French-American-British (FAB) scheme. It is important to note that the WHO proposals were not instigated by a small number of individuals. Rather, they represent the consensus opinion of an international panel of nearly 100 clinicians, pathologists, and scientists who met at Airlie House, VA, in 1997, where many of the same issues raised in the letter from Greenberg et al were debated at length. Although complete unanimity was not achieved on every issue, it was the majority view that there are sufficient data in the literature to indicate that changes to the FAB classification for MDS2 and acute myeloid leukemia (AML)3 should be made to permit better identification of biologically distinct subgroups of patients. The following paragraphs address the major concerns of the International MDS Study Group (IMDSSG) and reflect the discussions of the Airlie House Conference participants.1. Elimination of the category of refractory anemia with excess blasts in transformation (RAEBT) by lowering the blast requirement for a diagnosis of AML from 30% to 20% risks losing the fundamental difference between AML and RAEBT. In their letter, the members of the IMDSSG argue that RAEBT is biologically different from AML, and that it has a more indolent pace. In part, we agree. But the comparison that must be made is whether there are significant biologic and clinical differences between RAEBT and the proposed WHO category into which such cases would be reclassified, ie, "AML with multilineage dysplasia (with or without a preceding history of MDS)." The literature indicates that RAEBT and cases of AML that arise from a previous MDS or that have a background of multilineage dysplasia share important biologic and clinical features, including poor-risk cytogenetic abnormalities, increased expression of MDR-1, and poor response to chemotherapy.4-6 Some investigators have also reported that when matched for similar disease features, patients with RAEBT and AML who are treated with identical therapy have nearly identical outcome and survival times.7-9 Furthermore, our interpretation of the data generated by the International MDS Risk Analysis Workshop is that RAEBT is not necessarily more indolent than AML; 25% of patients with ≥ 20% marrow blasts evolved to AML in 2 to 3 months, 50% in 3 months, and more than 60% within 1 year, with an overall median survival time less than 1 year.10 The sum of such data indicates to us that patients with MDS and 20% blasts in the marrow have essentially the same disease as patients with AML with multilineage dysplasia and 30% blasts in the marrow.We agree with Greenberg et al that MDS and AML cannot be separated by an arbitrary blast count, and we urge that treatment decisions, in particular, be made in the context of the clinical findings, knowledge of the cytogenetic and molecular genetic data, and the rate of disease progression, regardless of whether the marrow blast count is 20% or 30%. But we believe that the WHO proposal to eliminate RAEBT as a disease category and to classify such patients as AML with multilineage dysplasia does not diminish any fundamental differences between AML and MDS, but instead emphasizes the continuity of the MDS disease process with a unique subtype of AML. Perhaps just as importantly, however, is that elimination of RAEBT will also eliminate a common practice that is a disservice to some patients: that of classifying a patient with true AML de novo, ie, a patient with no myelodysplastic features who has good-risk cytogenetic findings, as RAEBT only because there were less than 30% blasts on a single determination. The IMDSSG does accept this latter group as AML when blasts are less than 30% in the marrow, and thus they evidently suggest that two different thresholds of blast percentages be used to establish the diagnosis of AML. The WHO committee proposes that there is a rationale for one threshold.In summary, the WHO committee believes that the blast count is a useful but inexact marker for distinguishing one disease process from another. In this case, we believe there is evidence that the current distinction between RAEBT and the AML with myelodysplastic features is indeed arbitrary.2. All MDS categories require dysplasia in at least two hematopoietic cell lineages, and the WHO proposals for refractory anemia (RA), refractory anemia with ringed sideroblasts (RARS), and refractory cytopenia with multilineage dysplasia (RCMD) "would problematically blur relevant clinical distinctions." We acknowledge, as asserted by the IMDSSG, that some investigators require dysplasia in at least two cell lineages to make a diagnosis of MDS, including RA or RARS. Although the FAB guidelines for RA and RARS do state that "morphological abnormalities in the granulocytic and megakaryocytic series identical to those present in the other subtypes of MDS may occasionally be found in varying degrees," they also indicate that the erythroid series is mainly affected, and that "the granulocytic and megakaryocytic series almost always appear normal."2 However narrowly or loosely one interprets these criteria, in practice RA and RARS include a heterogeneous population of patients, ranging from those with only erythroid dysplasia to those with significant dysplasia in the other lineages as well. We believe there are data that indicate patients with RA and RARS can be further assigned to clinically relevant groups based on whether the erythroid series is principally involved or whether there is prominent dysplasia (more than 10% dysplastic cells) in the granulocytic and megakaryocytic series as well. As noted by Greenberg et al, numerous investigators (see review in Germing et al11) have shown that patients with pure sideroblastic anemia (PSA) in whom there is unilineage dysplasia in only the erythroid series enjoy a longer survival and have a much lower incidence of transformation to AML than those who have RARS with multilineage dysplasia. The WHO proposal (admittedly not made clear in the initially published outline) does recognize these two types of sideroblastic anemia, ie, RARS with unilineage dysplasia (or PSA) and RARS with multilineage dysplasia. The WHO committee believes that RA as defined by the FAB guidelines is similarly heterogeneous and has proposed that patients who demonstrate mainly unilineage erythroid dysplasia be designated as RA, whereas those with prominent multilineage dysplasia be classified as RCMD. There are data from four recent studies to support this view.12-15 Patients with RCMD have been reported to have a higher incidence of pancytopenia and of cytogenetic abnormalities, more frequent progression to AML, and a worse survival than patients with RA in whom only the erythroid series is affected. Whether there are major clinical or biologic differences between RCMD and RARS with multilineage dysplasia is not yet clear, although preliminary data from Germing et al,15 in a study of 284 patients with RCMD, show no significant difference in survival or progression to AML between RCMD and RARS with multilineage dysplasia.We concur with Greenberg et al that minimum essential criteria should be established for defining MDS. This is particularly true for RA, because as defined in the FAB classification, there is considerable leeway for inclusion of cases, including some that may not be myelodysplastic in nature. In the WHO proposal, the definition of RA is sufficiently restrictive to exclude most nonclonal cases of refractory anemia, thus further reducing the heterogeneity of the RA group and allowing for more appropriate patient management.Overall, it is the aim of the WHO committee to refine the definition of the low-grade myelodysplastic disorders so that the subgroups are more homogeneous. This should permit clinicians and investigators to more accurately compare the results of future biologic and clinical studies.3. "Chronic myelomonocytic leukemia (CMML) is predominantly a myeloproliferative disorder." Although we respect the opinion of the IMDSSG that CMML is primarily a myeloproliferative disorder, there is considerable controversy among a number of other investigators regarding whether CMML is mainly myeloproliferative or myelodysplastic in nature, or both.16-18 Recent studies that divide CMML into MPD and MDS types according to the leukocyte count have concluded that this parameter alone is likely not sufficient to recognize subgroups of CMML with major biologic or prognostic differences,19 and to our knowledge, there are no specific cytogenetic or molecular differences between CMML with myelodysplastic features and CMML with myeloproliferative features. The WHO committee understands the rationale for including CMML with the chronic myeloproliferative disorders and discussed raising the threshold for the monocyte count required for defining CMML so that it could be considered in that group of diseases. Cases with lower monocyte counts might then be assigned to the appropriate myelodysplastic categories, as suggested by the IMDSSG. However, there are no studies that provide another value that has been proven to be more clinically or biologically relevant, so any change in the monocyte requirement would be arbitrary. Furthermore, Greenberg et al correctly acknowledge that some patients who start out as having nonproliferative CMML may eventually have quite proliferative CMML. It may be that, at present, there is no satisfactory definition for this disease, but it was the overwhelming consensus of the Airlie House Conference participants that arbitrarily subdividing CMML into two subtypes would not help to understand it any better. The WHO proposal does not make any significant changes in the criteria for the diagnosis of CMML. We have merely placed it in a category that acknowledges its multifaceted features. This category provides a less restrictive view of CMML than does the FAB classification, and permits clinicians, if they wish, to view the condition of those patients with low or normal leukocyte counts and monocytosis as being more like MDS, and those with higher counts as myeloproliferative in nature, and to manage them accordingly. We believe that atypical chronic myeloid leukemia also fits into this category, as it too has both myeloproliferative and myelodysplastic features, and may at times be difficult if not impossible to separate from CMML on the basis of morphologic and clinical features.16-18 If future studies provide more definitive evidence that CMML or atypical chronic myeloid leukemia are more accurately classified as myelodysplastic or myeloproliferative processes, then appropriate changes in their classification will be warranted. In our opinion, such evidence is not available at the present time.We believe that most of the other concerns raised in the letter of Greenberg et al are addressed in the details of the classification, which will be published shortly. In particular, refractory anemia with excess blasts (RAEB) is divided into two subgroups, RAEB-1 and RAEB-2, based mainly on the finding of less than or more than 10% marrow blasts, respectively. A similar subdivision is proposed for CMML. The 5q− syndrome is defined narrowly and includes only those patients with a de novo, isolated del(5q) abnormality, refractory anemia, hypolobated megakaryocytes, and less than 5% medullary blasts.20,21 We do believe, however, that the category of "MDS, unclassifiable" should remain. The frequency with which this category is used will vary with the experience of the morphologist and clinician, but all members of the WHO committee have encountered cases that defy accurate classification into one of the FAB categories, and we acknowledge that will also be the case with the WHO classification.Finally, we agree with Greenberg et al that the best prognostic classification scheme will incorporate a number of genetic, immunologic, and biologic markers. But one must first accurately classify the patient into homogeneous groups in order to accurately interpret the results of studies using these techniques, and we believe that the WHO classification system does provide a useful framework for achieving this latter goal. Recently, Germing et al15 applied the WHO proposals to 1,600 MDS patients for whom long-term follow-up was available. In their study, multivariate analysis of the data demonstrated that the WHO proposals provided more homogeneous categories, fewer unclassifiable cases, and more precise prognostic information than did the FAB classification.We hope that the members of the IMDSSG will consider evaluating the WHO modifications in their studies of patients with MDS. Thank you for giving us the opportunity to respond.Reply 2:The letter from the International MDS Study Group (IMDSSG) and the World Health Organization (WHO) committee response raise thoughtful issues. An analysis of differences persisting between the two groups follows:1. Separation of myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Both groups agree that a marrow blast percentage threshold is irrelevant in one subset of AML, true de novo AML with recurring cytogenetic translocations. This AML subset should be eliminated from discussion. The other major subset, myelodysplasia-related or MDR-AML (WHO classification: AML with multilineage dysplasia), is clearly related to MDS. Both the WHO and IMDSSG groups agree that any marrow blast percentage is imprecise for separating MDS and MDR-AML; when possible, serial marrows and clinical parameters should be used. They further agree that the blast threshold is a surrogate for poorly understood changes that result in transformation of stem cells in MDS and progression to AML. The remaining difference between the groups is which arbitrary blast percentage threshold should be used in making this distinction in the absence of serial marrows or more definitive criteria. A low blast percentage threshold will misclassify as AML more cases that will continue to behave as MDS, whereas a high threshold risks the opposite result. Neither is satisfactory, and we must strive to find a better means to recognize transformation to AML in this setting.2. Definitions and minimal diagnostic criteria for low-grade MDS. Disagreement here is philosophical. Should we still consider MDS a syndrome, as it was historically and as still implied by its name, or rather should we now consider it a specific set of diseases sharing common biologic and clinical parameters, genetics, and pathogenesis and requiring common treatments? The WHO group has retained the syndrome approach, with more inclusive but less specific diagnostic criteria (unilineage dysplasia) for low-grade MDS. With this approach low-grade MDS will be heterogeneous, including diseases with no acquired genetic abnormality or risk for progression to high-grade MDS or MDR-AML. The IMDSSG has adopted the disease approach, attempting to restrict MDS to a specific set of diseases united at least by involvement of a multipotential hematopoietic progenitor, hence the requirement for bilineage dysplasia. Neither approach is inherently incorrect. However, for clinical protocols and for studies of the biology and pathogenesis of the specific disease group MDS, the disease approach seems preferable, with its more stringent diagnostic criteria (bilineage dysplasia).3. Definition and placement of chronic myelomonocytic leukemia (CMML). CMML comprises a heterogeneous spectrum of disease, on one end predominantly hypoproductive with minor monocytosis, on the other predominantly hyperproductive and sharing features with myeloproliferative diseases, with a continuous spectrum between. The entire spectrum shares cytogenetic and dysplastic features with MDS. Current data do not indicate how to subdivide CMML, although intuitively the hypoproductive end of the spectrum belongs in MDS and the hyperproductive end in a separate group with a major proliferative component. Placing the entire group in MDS (French-American-British classification) seems to err in placement of the hyperproductive high end, whereas removing all CMML from MDS (WHO classification) may err in removing the hypoproductive end from MDS. Although available data are not helpful, a compromise is obviously possible for use in clinical trials to allow study of hypoproductive CMML cases in common protocols with other cases of MDS, pending clarification of this issue.1. Harris N, Jaffe E, Diebold J, et al: World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: Report of the Clinical Advisory Committee meeting—Airlie House, Virginia, November 1997. J Clin Oncol 17:: 3835,1999-3849, Link, Google Scholar2. Bennett JM, Catovsky D, Daniel MT, et al: Proposals for the classification of the myelodysplastic syndrome. Br J Haematol 51:: 189,1982-199, Crossref, Medline, Google Scholar3. Bennett JM, Catovsky D, Daniel MT, et al: Proposed revised criteria for the classification of acute myeloid leukemia: A report of the French-American-British Cooperative Group. Ann Intern Med 103:: 620,1985-625, Crossref, Medline, Google Scholar4. Head DR: Revised classification of acute myeloid leukemia. Leukemia 10:: 1826,1996-1831, Medline, Google Scholar5. Sonneveld P, van Dongen JJM, Hagemeijer A, et al: High expression of the multidrug resistance P-glycoprotein in high-risk myelodysplasia is associated with immature phenotype. Leukemia 7:: 963,1993-969, Medline, Google Scholar6. Leith CP, Kopecky KJ, Godwin J, et al: Acute myeloid leukemia in the elderly: Assessment of multidrug resistance (MDR1) and cytogenetics distinguishes biologic subgroups with remarkable distinct responses to standard chemotherapy—A Soutwest Oncology Group study. Blood 89:: 3323,1997-3329, Medline, Google Scholar7. Estey E, Pierce S, Kantarjian H, et al: Treatment of myelodysplastic syndromes with AML-type chemotherapy. Leuk Lymphoma 11:: 59,1993-63, (suppl 2) Crossref, Medline, Google Scholar8. Estey E, Thall P, Beran M, et al: Effects of diagnosis (refractory anemia with excess blasts, refractory anemia with excess blasts in transformation or acute myeloid leukemia [AML]) on outcome of AML-type chemotherapy. Blood 90:: 2969,1997-2977, Crossref, Medline, Google Scholar9. Berstein SH, Brunetto VL, Davey FR: Acute myeloid leukemia-type chemotherapy for newly diagnosed patients without antecedent cytopenias having myelodysplastic syndromes as defined by French-American-British criteria: A Cancer and Leukemia Group B study. J Clin Oncol 14:: 2486,1996-2494, Link, Google Scholar10. Greenberg P, Cox C, Le Beau MM, et al: International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 89:: 2079,1997-2088, Crossref, Medline, Google Scholar11. Germing U, Gatterman N, Aivado M, et al: Two types of acquired sideroblastic anaemia (AISA): A time-tested distinction. Br J Haematol 108:: 724,2000-728, Crossref, Medline, Google Scholar12. Michels S, Chan W, Jakubowski D, et al: Unclassifiable myelodysplastic syndrome: A study of sixteen cases with a proposal for a new subtype. Lab Invest 62:: 67,,1990 (abstr) Google Scholar13. Rosati S, Mick R, Xu F, et al: Refractory cytopenia with multilineage dysplasia: Further characterization of an "unclassifiable" myelodysplastic syndrome. Leukemia 10:: 20,1996-26, Medline, Google Scholar14. Balduini CL, Guarnone R, Pecci A, et al: Multilineage dysplasia without increased blasts identifies a poor prognosis subset of myelodysplastic syndrome. Leukemia 12:: 1655,1998-1656, (letter) Crossref, Medline, Google Scholar15. Germing U, Gatterman N, Strupp C, et al: Validation of the WHO proposals for a new classification of primary myelodysplastic syndromes: A retrospective analysis of 1600 patients. Leukemia Res 24:: 983,2000-992, Crossref, Medline, Google Scholar16. Michaux JL, Martiat P: Chronic myelomonocytic leuaemia (CMML): A myelodysplastic or myeloproliferative syndrome? Leuk Lymphoma 9:: 35,1993-41, Crossref, Medline, Google Scholar17. Groupe Francais de Cytogenetique Hamatologique: Chronic myelomonocytic leukemia: Single entity or heterogeneous disorder? A prospective multicenter study of 100 patients. Cancer Genet Cytogenet 55:: 57,1991-65, Crossref, Medline, Google Scholar18. Bennett JM, Catovsky D, Daniel MT, et al: The chronic myeloid leukaemias: Guidelines for distinguishing the chronic granulocytic, atypical chronic myeloid and chronic myelomonocytic leukaemicas. Br J Haematol 87:: 746,1994-754, Crossref, Medline, Google Scholar19. Germing U, Gatterman N, Minning H, et al: Problems in the classification of CMML: Dysplastic versus proliferative type. Leukemia Res 22:: 871,1998-878, Crossref, Medline, Google Scholar20. Van Den Berghe H, Cassiman JJ, David G, et al: Distinct haematological disorder with deletion of long arm of No. 5 chromosome. Nature 251:: 437,1974-438, Crossref, Medline, Google Scholar21. Boultwood J, Lewis S, Wainscoat JS: The 5q− syndrome. Blood 84:: 3253,1994-3260, Crossref, Medline, Google Scholar
Referência(s)