Document 0qedz2J3Qgk4bYgODz5j1ym2b

Treatment of Myelodysplastic Syndromes With Hemopoit tic Growth Factors Peter L. Greenberg THE MYELODYSPLASTIC syndromes (MDS) provide a clinical model for evaluating the evolution of a relatively benign clonal myeloid hemopathy into the frankly malignant neoplasm, acute myeloid leukemia (AML). Because this disease is relatively indolent and predominates in the elderly, a therapeutic challenge has been to provide a treatment modality having adequate support for the patients' dominant cytopenias without causing excessive toxicity. Thus, with the biologic evidence of physiologic relevance and functional specificity of the hemopoietic growth factors and their availability for clinical use in recombinant form, these factors have been used to attempt to improve management of hematologic problems in MDS. Defectiveproliferation of hemopoietic precursors within MDS marrow has been suggested as being due to either decreased responsiveness to or decreased production of hemopoietic growth factors. However, as some leukemic cells have enhanced proliferative responses to the colonystimulating factors (CSFs) in vitro's3concern exists regarding the safety of using such agents in responsive neoplastic cells. Further, intrinsic abnormalities of responsiveness occur in the myeloid precursors of these patients. Therefore, in order to evaluate the proliferative versus differentiative responsivenessof hemopoietic precursors in MDS to hemopoietic growth factors and determine the possible clinical utility of CSF treatment, several laboratories, includingour own: have assessed in vitro proliferative,differentiative, and regenerative responses of marrow cells from these patients to recombinant human granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage From the Division of Hematology, Sranford University School of Medicine, and Veterans Administration Hospital, PaloAlto, CA. Part of this article has previously been published in Cancer Surveys 9(1):199, 1990 Address rrprint requests to Peter Greenberg, MD, Division of Hematology Rm S-161, Stanford University School of Medicine, 300 Pasteur Dr, Stanford, CA 94305. Copyright 0 I992 by W.B. Saunders Company 0093- 7 7 5 4 / 9 2 /1901-OO08$05.00i0 colony-stimulatingfactor (GM-CSF) (see accompanying article in this issue of the journal5). These in vitro findings have suggested thc possible efficacy of CSFs in this clinical setting and have led to therapeutic trials with these agents. CLINICAL FEATURES Numerous terms including preleukemia, hemopoietic dysplasia, refractory anemia with excess of blasts (RAEB), subacute or smoldering myeloid leukemia, oligoleukemia,and myelodys- plastic and dysmyelopoietic syndromes6'" have been used to describe patients with refractory cytopenias whose marrows showed dysplastic changes in at least two of the three hemopoietic cell lines and who have a propensity to undergo transformation into AML.The term "myelodys- plastic syndromes" describes the morphologic features of the marrows of these patients, and has been most commonly used to describe this clinical entity. Biologic data, particularly marrow cytogenetics and myeloid clonogenic culture and clonal analysis evaluating restriction fragment length polymorphisms, indicate that MDS patients already have their marrows involved with a myeloid malignant clone despite not having overt However, heterogeneity of these patients has been recognized regarding variations in their marrow morphology and their differing potential for transforming into AML. Thus, these patients were subclassifiedby the French, American and British (FAB) Morphologic Group, which formulated a set of criteria based on marrow morphology including the proportion of myeloblasts and degree of derangement of the hemopoietic cell lines.'3 This classification scheme subdivided patients into five subgroups: refractory anemia (RA), refractory anemia with ringed sideroblasts (RARS), refractory anemia with excess blasts (RAEB), refractory anemia with excess blasts in transformation (RAEB-T), and chronic myelornonocytic leukemia (CMML). The first four entities were characterized by abnormal marrow myeloid cell differentiation patterns (dysplasia) with RA or RARS patients having less than 5% blasts associated with 106 Seminars in Oncology, Vol 19, No 1 (February), 1992:pp 106-114 TREATMENT OF MYELODYSPLASTIC SYNDROMES 107 dysplasia of the three hemopoietic cell lines, RAEB having between 5% and 20% blasts and RAEB-T having 20% to 30% blasts. Acute myeloid leukemia was considered to be present if marrows had more than 30% blasts. This morphologic characterization was helpful, although due to evolution of these diseases into aggressive stages, an additional criterion necessary for categorizing these individuals is the pace of such progression. This characteristic is important to distinguish the relative indolence of MDS from the rapid evolution of patients with frank AML. Therefore, relative stability of the patients' peripheral blood counts and marrow morphology for 2 6-week periods aids in categorizing these cytopenic patients as MDS. The inclusion of CMML as belonging to MDS is problematic because this entity is predominantly a myeloproliferative rather than myelodysplastic disorder and appears to be more akin to chronic myeloid leukemia with associated monocytes than with myelodysplasia. The FAB classification has been useful in determining prognoses in MDS. There has been relative consistency of prognostic findings in seven major large studies evaluating this issue according to FAB morphologic criteria.l3-I9Patients with RAEB and RAEB-t had relatively poor prognoses with median survivals generally ranging from 5 to 12 months in contrast to RA or RARS patients who had median survivals of about 3 to 6 years. The proportion of these individuals who transformed to AML varied similarly, as in the high-risk RAEB and RAEB-t patients this incidencewas 40% to 50%, whereas in the low-risk group it was 5% to 15%. In a recent study evaluating time-to-disease evolution, 25% and 55% of patients with RAEB and RAEB-t, respectively, underwent transformation to AML at 1 year, and 35% and 65% at 2 years." In contrast, for patients with RA the incidence was 5% and 10% at 1 and 2 years, whereas none of the RARS patients underwent leukemic transformation within 2 years. Another parameter of negative prognostic significance was an increased proportion of marrow blasts," a finding parallelling data demonstrated for FAB subgroups. Mortality in these individuals is due to a variety of causes including evolution to AML, infection or bleeding complications related to the patients' dominant cytopenias or, because most of these patients are elderly, concomitant nonhematologic diseases associated with an older patient population. Three large s e r i e ~ ' ~ ~ ' ~ ~ ~ have analyzed these data. In RAEB and RAEB-t patients, AML was the cause of death in 20% to 55% of patients, whereas infection and hemorrhage due to marrow failure caused about 36% to 50% of the deaths, and nonhematologic causes caused about 10% to 20%. In RA and RARS, these figures were somewhat reversed, with AML causing death in 0 to 29%, infection and hemorrhage caused 15% to 44% and nonhematologiccauses 25% to 42%. A number of other methods have been used to analyze prognosis for MDS patients based either on biologic features, such as cytogenetic abnormalities and in vitro culture clonogenicity result^^^^ or abnormal clinical features (eg, pan- cytopenia as opposed to single cytopenia). Patients with abnormal cytogenetics and in vitro marrow myeloid clonal growth patterns, as well as more deranged clinical features have poorer prognoses. These abnormalities generally correlate with the more advanced FAB classifications. Thus, the FAB classification appears to be a useful first approximation of prognoses for these patients. Due to this variability of prognoses in subgroups of patients with MDS, it is necessary to analyze therapeutic outcomes with these parameters appropriately stratified. A variety of treatment approaches have been used in MDS, with general supportive care being the mainstay of therapy in the community. Thus, patients are treated as needed with antibiotics for infection, and red blood cell and platelet transfusions for anemia and thrombocytopenic bleeding. However, recently other therapeutic approaches are being more frequently used to treat MDS. Considering the direction in which treatment for this disease is now rapidly evolving, major focus of this review will be to evaluate the results of treatment of MDS with hemopoietic growth factors. HEMOPOIETIC GROWTH FACTORS The development of in vitro marrow clonogenic culture assays led to discovery of a family of interacting hematopoietic growth factorsthe CSFs-which have recently been demonstrated to have critical physiologic roles for 108 PETER L. GREENBERG controlling hemopoiesis in vivo.' Interleukin-3 eloid cells. Patients observed relatively few side (IL3) and GM-CSF have predominantly prolif- effects, but bone pain was dose-limitingwhen it erative effectson early hemopoieticcell compart- was associatedwith highwhite-cellcounts.These ments, whereas G-CSF and monocyte-CSF (M- results showed that GM-CSF was a potent in CSF) have major differentiative as well as vivo stimulator of hematopoiesis and produced proliferative effects on later more lineage- short-term hematologic improvement (8 to 32 restricted precursors cells (for granulocytes and weeks of observation) in patients with MDS. monocytes, respectively).Each hemopoietic cell In another GM-CSF was adminis- lineage appears to be regulated by both prolifer- tered as 1-hour or 4-hour intravenous infusions ative and differentiative stimuli. Marrow cul- daily for 7 days or as 12-hour intravenous ture studies in MDS have generally demon- infusionsfor 14days. Temporary improvements strated subnormal clonal growth and defective were seen in granulocyte counts, monocyte cellular maturation of myeloid and erythroid counts, and reticulocyte counts in five of seven precursors that become more abnormal as these patients with MDS, who also had increases in patients evolve toward AIv~L.~.T~h*e*'presence the numbers of eosinophils, immature myeloid of abnormal marrow cytogeneticsor defective in cells, and myeloblasts. There was no reduction vitro myeloid clonogenicity has been shown to in RBC transfusion requirements, and no effect have negative prognostic import in MDS."S*11*2' was observed on platelet counts. There was Erythropoietin (Epo) is a relatively late-acting minimal toxicity, consisting of transient low- factor that acts predominantly on CFU-E (col- back discomfort, anorexia, myalgias, arthral- ony-forming units-erythroid) and on a portion gias, and low-grade fever. of the earlier BFU-E (burst-forming unitserythroid), which generate CFU-E. Epo acts in In a study by Ganser et a1,25 the treatment schedule included dose escalation of GM-CSF synergywith G-CSFZoaand certain earlier acting administered by continuous intravenous infu- factors to enhance in vitro erythropoiesis. Defective in vitro erythropoiesis has been demon- sion for 7 to 14 days, repeated after a 2-week treatment-free interval. The blood leukocyte strated in MDS?5 and neutrophil counts increased markedly and Treatment with GM-CSF dose-dependently by 1.3- to 18-fold in 10 of 11 patients. A rise of monocytes and eosinophils To date, five therapeutic trials have been occurred in most patients. No sustained in- reported using recombinant human GM-CSF crease in reticulocytes or platelets was ob- short-term (generally treatment was adminis- served. Lymphocyte counts increased in all tered for 7 to 14days,with one to five courses of patients, affecting both T-helper and suppres- the rap^).^-*^ In the study of Vadhan Raj et al,23 sor cells; however, the lymphocytes were not GM-CSF treatment (continuous intravenous activated as analyzed by expression of IL2 infusion) was associated with marked increases receptors. In four of the patients, all with more in peripheral-blood leukocytes (five-to 70-fold), than 15% blast cells in the bone marrow, the including granulocytes (five- to 373-fold) in the percentage of bone marrow blast cells increased eight patients evaluated. The absolute number during treatment with GM-CSF and 5 of the 11 of monocytes, eosinophils, and lymphocytesalso treated patients developed AML during or increased in all patients. Three of eight patients within 1 month of treatment. Another patient also had two- to 10-fold increases in platelet developed a chronic myeloproliferative disor- counts and improvement in erythropoiesis, with der 3 months after treatment. Cytogenetic data the result that two of three patients who had indicated induction of both proliferation and required red blood cell (RBC) and platelet differentiation of the leukemic clones by GM- transfusions no longer needed them (at 20 to 27 CSF. Toxic side effects were minor with slight weeks of follow-up).Treatment was also associ- fever, phlebitis at the infusion site, and bone ated with increased marrow cellularity and a pain in the minority of patients. decreased percentage of marrow blasts in five In another study,'6 GM-CSF was adminis- patients, resulting in an increase in the ratio of tered by intravenous injections for 5 days to four differentiated myeloid cells to immature my- MDS patients. The treatment courses were TREATMENT OF MYELODYSPLASTIC SYNDROMES 109 interrupted by a 10-day rest period and repeated two to three times. Granulocyte-macrophage colony-stimulating factor was well tolerated, with only minor side effects seen, which included bone discomfort at the lower back, sternum, and ribs, and constitutional symptoms such as low-grade fever, nausea/vomiting, and mild myalgias. Whereas no increases in platelet and reticulocyte counts were recorded, elevations of absolute neutrophil counts above 100 cells/mm3occurred in all patients. However, a striking finding was the development of increases in the number of circulating and bone marrow blast counts that were observed particularly when GM-CSF doses of 500 Fg/m2of body surface area were administered. Thompson et al" demonstrated that daily subcutaneous GM-CSF treatment caused increased circulating neutrophil levels in a dosedependent manner in 11of 16 patients (two- to 194-fold increments). Although the neutrophils usually returned to pretreatment levels shortly after stopping GM-CSF, two patients continue to exhibit an elevation of neutrophils for 6 months. Dose-related increases in circulating monocytes and eosinophils were also noted. Transient increases in platelet and reticulocyte counts were observed in several patients. Five of the 16patients later received maintenance therapy with GM-CSF for 2 to 9 weeks. All patients had dramatic increases in neutrophils after 2 weeks. However, thereafter, despite continued therapy, the neutrophil count in 4 patients declinedmarkedly, 1patient developed antibodies to GM-CSF and 1patient's disease evolved into AML. This study indicated striking but usually temporary improvement of the neutropenia of MDS with GM-CSF. The combined data from these five indicated that 38 of 45 patients treated with GM-CSF had improvements in neutrophil counts, 9 with associated increases in marrow myeloid maturation documented, 14 had increased reticulocyte counts with three of these individuals having decreased RBC transfusion requirements, eight had transient increases in platelets (Table 1). In 12patients an increase in marrow and/or peripheral blood blasts were noted. Seven patients progressed to AML, particularly individuals with more than 15% marrow blasts (ie, those with RAEB or FL4EB-t). In Table 1. Effectsof Recombinant Human GM-CSF and G-CSF in Phase 1/11 Clinical Trials of Patients with Myelodysplastic Syndromes GM-CSF23.27 G-CSF3',32 Short-term treatment duration 7 to 14 days x 1-5 42 to 56 days coursesdaily, IV daily sub- or subcutaneously cutaneously 30-750 pglrn` Daily dose 30-750 pg/m2 0.1-3 pglkg Number of patients 45 18 FAB subtypes RA/RAEB- 14/26/5 211610 RAEB-t/CMML Responses Neutrophils 38 (84%) 16 (89%) Reticulocytes 14 5 Platelets 81 Marrow maturation 9 16 Increased blasts 12 4 Progressionto AML 7 5 Long-Term Treatment Duration 2-9 weeks" 6-28 months32 Persistent neutrophil 115" 10/1127 responses/ patients Modified with permission." the longer term study reported by Thompson et al,27only one of five individuals maintained the increased neutrophil counts. A preliminary report indicates results of a multi-institutional randomized trial of GM-CSF treatment versus observation in MDS, with crossover occurring in patients with infections.?s Only MDS patients with less than 15% marrow blasts (ie, those with RA, RARS, and some with RAEB) received GM-CSF. To date, 21 MDS patients have received GM-CSF, nearly all of whom demonstrated greater than twofold neutrophil responses, with 8 patients having persisting responses for 6 months. Severe toxicity requiring reduction or discontinuation of GMCSF dose (initially3 pg/kg per day) occurred in a substantial proportion of the patients. No improvements in the patients' platelet counts or hemoglobin levels were noted. Further patient accrual and follow-up is occurring to determine the impact of this therapy on the incidence of infectious episodes and transformation to AML. A recent trial using GM-CSF to treat nine patients with secondary MDS (ie, those in whom the MDS developed after prior cytotoxic chemotherapy for other malignancies) indicated neutrophil responses in seven of eight evaluable patients." One patient had a marked 110 PETER L. GREENBERG improvement in platelet counts and three had ~ resolution of active infections. Transformation to AML occurred in three patients after 5 to 11 months of therapy and three other patients died of progressive lymphoma (their underlying malignancy). A recent preliminary study has reported use of GM-CSF in combination with low doses of cytosine arabinoside to treat MDS.3' A low proportion of the patients (seven of 24 evaluable patients) had persisting improvements in neutrophil counts and marrow maturation patterns, whereas thrombocytopenia often became a major clinical problem. Further follow-up and evaluation of these patients is necessary to determine the value of this therapeutic approach. Treatment with G-CSF At Stanford, to date we have treated 18MDS patients (2 refractory anemia, 9 RAEB, 7 RAEB-t) with daily subcutaneous (Sa)injections of G-CSF, escalating dosage levels every 2 weeks from 0.1-3.0 pg/kg per day for a 6 to 8 week period?1JzSixteen patients had significant elevations in both WBC (two- to 10-fold) and absolute neutrophil counts (ANC, five- to 40fold). Nine of 11severely neutropenic patients (ANC < 500/mm3) had rises in ANC to 1,200 to 16,300 cells/mm3.All six moderately neutro- penic patients (ANC 500 to 1,800/mm3) had rises in ANC five- to 15-fold. In five patients a greater than twofold rise in reticulocyte counts occurred and 3 of 12 RBC transfusion-dependent patients had a decrease in RBC transfusions. Improved marrow myeloid maturation was noted in 16 of 18 patients. Cytogenetic abnormalities, initially present in 4 responding patients, persisted after treatment, suggesting that enhanced differentiation of the abnormal clone occurred. No significant changes in platelet, lymphocyte, monocyte, or eosinophil counts were found during treatment in 17 of 18 patients. After discontinuing G-CSF treatment peripheral WBCs returned to baseline levels over 2 to 4weeks. As a result of the clinical responsiveness and tolerance of these patients to short-term G-CSF treatment, the initial 2 month phase 1/11 trial was extended, using the same dose (generally to 1 to 3 pg/kg per day SQ) for maintenance therapy as that to which the patients initially r e s p ~ n d e dE. ~le~ven patients have received this more prolonged therapy and 10 of the 11 patients had continued improvements of their neutrophil counts, which have persisted for 6 to 28 months to date (Table 1). Marrow granulocytic maturation improved in seven of nine patients. Two of four RBC transfusion-dependent patients had decreases in their transfusion requirements, Platelet counts were generally not altered by this therapy. Neutrophil function (in vitro chemotaxis and phagocytosis), which was maintained or improved after 2 months of treatment, was further augmented in five patients after an additional 6 months of G-CSF therapy. During the G-CSF treatment study period four patients had eight episodes of clinically significant bacterial infections, seven of which occurred at an ANC I1,500/mm3either before a patient responded or between treatment cyc l e ~P.ri~or~to treatment, during an observation period of 2 to 15 months, five patients had developed infections, all of which occurred at an ANC less than 1,500/mm3. Including this pretreatment period there was a significant reduction in infection risk in responding patients who achieved an ANC greater than 1,500/ mm3after G-CSF treatment. Toxicity to G-CSFwas minimal, with only two patients stopping G-CSF injections because of side effects related to preexisting clinical conditions (psoriasis, nausea and an~rexia).~M~i.l~d' fluid overload occurred in one patient. In several patients other medical problems occurred, likely related to their preexisting cardiac or pulmonary disorders. Five of the 18 patients, 4 initiallywith RAEB-t, converted to AML after 6 to 16 months of the study. In one individual, pretreatment in vitro culture had revealed a marked increase in proliferative sensitivity of his hemopoietic precursors to G-CSF and GMCSF associatedwith increased in vitro CFU-GM regeneration and poor myeloid differentiation. These data indicated that G-CSF injected subcutaneously on a chronic basis was well tolerated and effective for eliciting persistent improvement in neutrophil counts and in vitro function, marrow myeloid maturation, and possibly decreasing bacterial infections and RBC transfusion requirements in MDS patients. TREATMENT OF MYELODYSPIASTICSYNDROMES 111 Results of a very short-term trial of 40 MDS patients (20 R4,20 RAEB/RAEB-t) treated with G-CSF has recently been reported." At doses of 2 to 5 pg/kg per day intravenously, 20 of 22 patients (7-day treatment) and 16 of 18 patients (14-daytreatment) had substantialneutrophil responses associated with decrements in marrow blasts in 8of 13evaluated patients. The treatment was well tolerated. Evidence of Clonal Responses Cytogenetic evaluations were performed to determine whether selective responses to CSFs of normal versus abnormal clones occurred in MDS. In the two short-term studies using GMCSF,U*2w5hich perform these analyses, the cytogenetic abnormalities persisted after treatment in the 18patients having these abnormalities. In the studiesof MDS patients treated with G-CSF all six individualswith abnormal karyotypes had persistent cytogenetic abnormalities,three after longer-term (3 to 13 months) therapy?1fzThe generally persisting cytogenetic abnormalities after treatment with these CSFs suggested enhanced differentiation of the abnormal clone in MDS. Analyzing restriction fragment length polymorphisms of X-linked genes, one patient with abnormal marrow cytogenetics responding to G-CSF showed monoclonality of her neutrophils after treatment.32However, another similarly analyzed patient responding to GM-CSF demonstrated polyclonal hematopoiesis after therapy?' Direct evidence of this point will require further analysis of X-linked genetic polymorphisms or of interphase cytogenetics of neutrophils." Treatment with Interleukin-3 Two studies have reported the effects of IL3 therapy in MDS patients (Table 2). In the first trial, nine patients were treated with one to three 15-day courses of relatively high doses of IL3 (250 to 500 pg/m2,subcutaneously).%Six of these patients had RA and three had RAEB. All of the patients had leukocyte responses and in seven, neutrophils were elevated. In only three of the patients did an increase occur of their ANC from less than 1,OOO to more than l,000/mm3. Three patients had an increase of reticulocytes, but in only one of nine RBC transfusion-requiring patients was there a de- Table 2. Effectsof Recombinant Human IL-3Treatment in Myelodysplastic Syndromes Reference 34 Reference 35 Treatment Duration Daily Dose Number of Patients FAB Subtypes RAIRAEBRAEB-t1CMML Responses Leukocytes Neutrophils Reticulocytes Platelets t Blasts Progressionto AML 15 d, X1-3 courses 250 to 500 pg/m' SC 9 61310 9 7:3* 3; -1 transfusions 119 3t; 1 transfusions 214 2 PB;1 BM 1 20 d 30 to 1,000 Fglm` IV 13 5/5/3 a 6;3* 1 2 1 BM 0 Abbreviations: BM, bone marrow; PB, peripheral blood. `Increased ANC from less than to more than l.OOO/mm'. tllnsustained. crease in such transfusions. Similarly, three patients had responses of their platelets, and in two of four, there was a transient decrease in platelet transfusion requirements. One of these individuals progressed to AML. In the other study, 13 patients were treated with 28-day courses of varying doses of IL3 intraven~usly.~' Five of these individuals had RA, 5 had RAEB or RAEB-t, and 3 had CMML. Eight patients had leukocyte responses, 6 of whom had increases in neutrophils. However, in only 3 patients was there an increase of ANC from less than 1,0oOto more than l,000/mm3with therapy. Reticulocytes rose in one patient and platelets rose in two. Aside from headaches, bone pain and fever at higher doses, this treatment was relativelywell tolerated. These short-termstudies indicated modest improvements in neutrophils, which, however, were not as prominent as those demonstrated with G-CSF or GM-CSF. Further, only limited responses occurred in the other cell lines. Thus, it is likely that IL3 will need to be combined with other hemopoietic growth factors to achieve substantial improvement in the cytopenias in MDS patients. Treatmentwith Erythropoietin Studieshave indicated that serum erythropoietin (Epo) levels may be suboptimally elevated in MDS patients relative to their degree of anemia?*Recombinant human Epo therapy has been instituted to correct their hypoproductive I 112 PETER L. GREENBERG anemias and three published reports and sev- vide marked acute improvement in neutrophil eral abstracts have detailed the erythroid re- counts and marrowmorphology in a high propor- sponses to this form of treatment (Table 3).3w3 tion of these patients, and for G-CSF and for In one study, eight MDS patients were treated some studies with GM-CSF these responses with Epo at doses ranging from 200 to 400 have been quite durable. In a low fraction of units/kg IV three times a week.39 Erythroid such patients, improved hemoglobin levels and responses to Epo, which were associated with decreased RBC transfusion requirements oc- decreases in RBC transfusion requirements or curred, whereas platelet counts generally re- increases in hemoglobin, were found in two of mained unchanged. Epo treatment caused im- these eight patients. In another study, neither of proved erythroid responses in about 20% to two MDS patientstreated with 50 to 500units/kg 25% of MDS patients. Retrospective analysis three times a week responded.@In a relatively indicated decreased infectious episodes in pa- large study of patients with the low-risk forms of tients receiving G-CSF whose neutrophil counts MDS (ie, RA and RARS), 4 of 17 individuals normalized, suggesting possible clinical efficacy treated with doses ranging from 800 to 1,600 for the responding patients.32The growth fac- units/kg IV two times per week re~ponded.~' tors appear to be well tolerated with little Generally, the responses took several months to associated toxicity, and they can be adminis- occur but marrow cytogenetics and serum Epo tered subcutaneously, indicating their potential 1 levels did not reliably predict such erythroid for treating outpatients. These hemopoietic responses. These published studies have been growth factors thus provide encouraging experi- corroborated by two recent reports cited in mental alternatives to chemotherapy or bone Table 3,4243and thus to date, 11of 51 (22%) of marrow transplantation for the generally eld- the patients have responded to Epo. Generally, erly patients comprising this patient population. the patients required relatively high doses of In vitro marrow clonogenic and suspension Epo for their responses. This somewhat limited culture studies suggested the potential therapeu- in vivo responsiveness of MDS marrow cells to tic utility of CSFs for MDS patients. The in vivo Epo is not totally unexpected as the defective and in vitro differentiative responses to CSFs of erythroid clones have demonstrated suboptimal marrow precursors from these patients indicate responses to Epo in vitro, particularly for BFU-E that decreasedresponsiveness andlor decreased responses.45 Thus, as suggested by recent in intramedullary availability of effective levels of vitro it is likely that a combination of certain hemopoietic growth factors may under- hemopoietic growth factors, which would en- lie certain cytopenias in patients with MDS. hance BFU-E numbers or responsiveness to The use of differentiation-inducing agents such Epo, may provide more prominent erythroid as G-CSF for treating myeloid clonal hemopa- responses. Such clinical studies are currently in thies is based on an evolving body of in vitro progress. marrow culture and preclinical data indicating CONCLUSIONS AND FUTURE DIRECTIONS that such agents may diminish self-replication of abnormal cell clones concomitant with en- The hemopoietic growth factors G-CSF, GM- hancing their differentiati~n.~'-T~h' us, it is of CSF, and IL3 have been demonstrated to pro- interest that G-CSF has substantial, although Table 3. Erythroid Responsesto RecombinantHuman Erythropoietinin MyelodysplasticSyndromes References MDS Subtype RAIRARS RAEBIRAEB-t Epo Dose Wkg) Responses/ Patients Responder Serum Epo Levels(mU/mll 39 4 40 2 41 17 42 5 43 10 4 200 to 400 IV, 3x/wk 2/8 (25%) 694,919 0 50 to 500 IV, 3x/wk 012 - 0 800 to 1,600 IV, 2x/wk 4/17 (24%) 16,515,589,1030 5 200 to 1,000 IV. 3x/wk 4/10 (40%) 183 (mean) 4 80 to 640 SO, 3x/wk 1/14 (7%) 1750 Totals 38 13 11/51 (22%) TREATMENT OF MYELODYSPIASTIC SYNDROMES 113 subnormal, myeloid differentiating activity for MDS marrow cells.s.slThis myeloid differentiation-inducing effect of G-CSF is greater than that of GM-CSF, particularly for W E B / RAEB-t patients and those with normal cytogenetkxa Methods are being sought to further enhance differentiation of cells from patients with myeloid clonal hemopathies. Despite the demonstration of improvements in neutrophil levels and marrow morphology in a substantial proportion of MDS patients with these hemopoietic growth factors, randomized controlled studies are needed to determine whether the natural history of these disorders (survival, evolution to AML, or infectiouscom.plications) wiIl be altered by treatment with the CSFs. Still required are growth factors necessary to augment platelet counts in these patients. Investigation is also warranted using erythropoietin in combination with the myeloid CSFs, to determine whether more consistent improvement in hemoglobin levels is achievable in MDS. REFERENCES 1. Metcalf D: The molecular biology and functionsof the granulocyte-macrophagecolony-stimulating factors. Blood 67257-267,1986 2. Miyauchi J, Kelleher CA, Yang YC, et al: The effects of three recombinant growth factors, IL3, GM-CSF and G-CSF, on the blast cells of acute myeloblastic leukemia maintained in short-term suspension culture. Blood 76657663,1987 3. Vellenga E, Young DC, Wagner K, et al: The effectsof GM-CSF and G-CSF in promoting growth of clonogenic cells in acute myeloblastic leukemia. Blood 69:1771-1776, 1987 4. Greenberg P L In vitro culture techniques defining biologic abnormalities in the myelodysplastic syndromes and myeloproliiferativedisorders. Clinicsin Haematol15:973993,1986 5. Greenberg P In vitro marrow culture studies in myelodysplasticsyndromes. Semin Oncol 19:34-46,1992 6. Block M, Jacobson LO, Bethard WF: Preleukemic acute leukemia. JAMA 152:1018-1021,1953 7. Saarni M, Linman J: Preleukemia. The hematologic syndrome preceding acute leukemia. Am J Med 55:38-48, 1973 8. Linman JW,Bagby GC: The preleukemic syndrome (hemopoietic dysplasia). Cancer 42:854-864,1978 9. Rheingold JJ, Kaufman R, Adelson E, et al: Smoldering acute leukemia. N Engl J Med 268:812-815,1963 10. Dreyfus B: Preleukemicstates. 11. Refractory anemia with excess of myeloblasts. Blood Cells 2:33-45, 1976 11. LeBeau MM, Albain KS, Larson R, et al: Clinical and cytogenetic correlations in 63 patients with therapy-related myelodysplasticsyndromesand acute nonlymphocyticleukemia: Further evidence for characteristic abnormalities of chromosomes No. 5 and 7. J Clin Oncol4325-46,1986 12. Janssen JWG, Buschle M, Layton M, et al: Clonal analysisof myelodysplastic syndromes: Evidenceof multipotent stem cell origin. Blood 73:248-254,1989 13. Bennett JM, Catovsky D, Daniel MT, et al: FAB Cooperative Group: Proposal for the classification of the myelodysplasticsyndromes.Br J Haematol5 1:189-199,1982 14. Foucar K, Langdon RM, ArmitageJO: Myelodysplastic syndromes. A clinical and pathologic analysis of 109 cases. Cancer 56553-561,1985 15. Mufti GJ, StevensJR, Oscier DG, et al: Myelodysplastic syndromes: A scoring system with prognostic significance. Br J Haematol59:425-433,1985 16. Kerhofs H, Hermans J, Haak HL, et al: Utility of the FAB classificationfor myelodysplastic syndromes:Investigation of prognostic factors in 237 cases. Br J Haematol 65:73-81,1987 17. Tricot G, Vlietinck R, BoogaertsMA, et al: Prognostic factors in the myelodysplasticsyndromes:Importance of initial data on peripheral blood counts, bone marrow cytology, trephine biopsy and chromosomal analysis. Br J Haematol W19-32,1985 18. Vallespi T, Torrabadella M, Julia A, et al: A study of 101cases according to the FAB classification.Br J Haemato1 61:83-, 1985 19. Sanz GF, Sanz MA, Vallespi T, et al: Two regression models and a scoring system for predicting survival and planning treatment in myelodysplasticsyndromes:A multivariate analysis of prognostic factors in 370 patients. Blood 74395-408,1989 20. Coiffier B, Adeleine P, Viala JJ, et al: Dysmyelopoietic syndromes. A search for prognostic factors in 193 patients. Cancer 52:83-90,1983 20a. Souza L,Boone T, Gabrilove J, et ai: Recombinant human granulocyte colony-stimulating factor: Effects on normal and leukemicmyeloid cells. Science 232:61-65, 1986 21. Greenberg P L The smoldering myeloid leukemic states. Clinical and biologic features. Blood 61:1035-1045, 1983 22. Jacobs RH, Cornbleet MA, Vardiman JW, et al: Prognostic implications of morphology and karyotype in primary myelodysplastic syndromes. Blood 67:1765-1772, 1986 23. Vadhan-Raj S, Keating M, LeMaistreA, et al: Effects of recombinant human granulocyte-macrophage colonystimulating factor in patients with myelodysplastic syndromes. N Engl J Med 317:1545-1552,1987 24. Antin JH, Smith BR, Holmes W,et al: Phase 1/11 study of recombinant granulocyte-macrophage colonystimulating factor in aplastic anemia and myelodysplastic syndrome. Blood 72705-713,1988 25. Ganser A, Volkers B, Greher J, et al: Recombinant human granulocyte-macrophage colony-stimulating factor 114 in patients with myelodysplastic syndromes-A Phase 1/11 trial. Blood 73:31-37,1989 26. Henmann F, Lindemann A, Klein H, et al: Effect of recombinant granulocyte-macrophage colony-stimulating factor in patients with myelodysplastic syndrome with excess blasts. Leukemia 3:335-338,1989 27. Thompson JA, Lee DJ, Kidd P, et al: Subcutaneous granulocyte-macrophage colony-stimulating factor in patients with myelodysplasticsyndrome:Toxicity, pharmacokinetics, and hematological effects. J Clin Oncol 7629-637, 1989 28. Schuster Mw, Thompson JA, Larson R, et al: Randomized trial of subcutaneous granulocyte-macrophage colony-stimulating factor versus observation in patients with myelodysplastic syndrome or aplastic anemia. Proceedings of the Annual Meeting of the American Society of Clinical Oncology 1990,p A793 29. Gradishar W, O'Laughlin R, Le Beau MM, et al: Colony stimulating factor in therapy-related myelodysplastic syndrome. Proceedings of the Annual Meeting of the American Societyof Clinical Oncology 1990, p A1160 30. Hoelzer D, Ganser A, Seipelt G, et al: Simultaneous treatment with recombinant human granulocyte-macrophage colony-stimulatingfactor and low-dose cytosinearabinoside in patients with myelodysplastic syndromes. Blood 74118,1989 (abstr) 31. Negrin RS, Haeuber DH, Nagler A, et al: Treatment of myelodysplastic syndromes with recombinant human granulocyte colony stimulating factor. Ann Intern Med 110976-984,1989 32. Negrin RS, Nagler A, Kobayashi Y, et al: Maintenance treatment of patients with myelodysplasticsyndromes using recombinant human granulocyte colony stimulating factor. Blood 78:36-43,1990 33. Greenberg PL, Negrin R, Nagler A The use of haemopoietic growth factors in the treatment of myelodysplastic syndromes. Cancer Surv 9(i):199-212,1990 34. Yoshida Y,Hirashima K, Asano S, et al: A Phase I1 trial of recombinant human granulocyte colony-stimulating factor in the myelodysplastic syndromes. Br J Haematol 78:378-384,1991 35. Vadhan-Raj S, Broxrneyer HE, Spitzer G, et al: Stimulation of nonclonal hematopoiesisand suppressionof the neoplastic clone after treatment with recombinant human granulocyte-macrophage colony-stimulating factor in a patient with therapy-related myelodysplasticsyndrome. Blood 741491-1498,1989 36. Ganser A, Seipelt G, Lindemann A, et al: Effects of recombinant human interleukin-3in patients with myelodysplastic syndromes. Blood 76455462,1990 37. Kurzrock R, Talpaz M, Estrov Z, et al: Phase I study of recombinant human interleukin-3 in patients with bone marrow failure. J Clin Oncol9:1241-1250, 1991 PETER L. GREENBERG 38. Jacobs A, Janowska-Wienorek A, Car0 J, et al: Circulating erythropoietin in patients with myelodysplastic syndromes. Br J Haematol73:36-39,1989 39. Bessho M, Jinnai I, Matsuda A Improvement of anemia by recombinant erythropoietin in patients with myelodysplastic syndromes and aplastic anemia. Int J Cell Cloning 8 4 5 - 4 5 8 , 1 9 9 0 40. Stebler C, Tichelli A, Dazzi H, et al: High-dose recombinant human erythropoietinfor treatment of anemia in myelodysplastic syndromes and paroxysmal nocturnal hemoglobinuria: A pilot study. Exp Hematol181204-1208, 1990 41. Stein R, Abels R, Krantz S: Pharmacologic doses of recombinant human erythropoietin in the treatment of myelodysplasticsyndromes. Blood 78:1658-1665,1991 42. Hellstrom E, Birgegard G, Lockner D, et ak Treatment of myelodysplastic syndromes with recombinant human erythropoietin. Blood 76279a, 1990(abstr) 43. Schouten HC, Vellenga E, van Rhenen D, et al: Recombinant human erythropoietinfor patientswith myelodysplastic syndromes. Leukemia 5:432-436,1991 44. Greenberg PL, Negrin R, Ginzton N GCSF syner- gizes with erythropoietin for enhancing erythroid colony formation in myeiodysplastic syndromes. Proc Am SOC Hematol, Dec 1991.Blood 1991(in press) 45. Sachs L: Annotation: The differentiationof myeloid leukemia cells-New possibilitiesfor therapy. Br J Haemato1 40509-517,1978 46. Hozumi M: Fundamentals of chemotherapy of myeloid leukemia by inductionof leukemiacell differentiation. Adv Cancer Research 38:121-169,1983 47. Nicola N A Granulocyte colony-stimulating factor and differentiation-inductionin myeloid leukemic cells. Int J Cell Cloning 5:l-15,1987 48. Metcalf D: Regulator-induced suppression of myelomonocytic leukemic c e k Clonal analysis of early cellular events. Int J Cancer 38:121,1982 49. Lotem J, Sachs L In vitro inhibition of the development of myeloid leukemia by injection of macrophage and granulocyte-inducingprotein. Int J Cancer 28:375-386,1981 50. Lotem J, Sachs L Controlof in vivo differentiationof myeloid leukemic cells. IV. Inhibition of leukemia development by myeloid differentiation-inducing protein. Int J Cancer 33:147-154,1984 51. Nagler A, Mackichan ML, Negrin RS, et al: Impact of marrow cytogenetics and morphology on in vitro hemopoiesis in the myelodysplasticsyndromes: Comparison between recombinant human granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor. Blood 7611299-1307,1990