Bone marrow transplantation for therapy-related ... - Nature

Bone Marrow Transplantation, (1997) 20, 737–743  1997 Stockton Press All rights reserved 0268–3369/97 $12.00

Bone marrow transplantation for therapy-related myelodysplasia: comparison with primary myelodysplasia KK Ballen1, DG Gilliland2, EC Guinan 2, C-C Hsieh1 , SK Parsons2 , IJ Rimm2, JLM Ferrara2, BE Bierer2 , HJ Weinstein2 and JH Antin2 2

Department of Medicine, Division of Hematology/Oncology, Brigham and Women’s Hospital and Harvard Medical School, Boston, and 1Cancer Center, University of Massachusetts Medical Center, Worcester, MA, USA

Summary: Therapy-related myelodysplasia (MDS) is a fatal marrow disorder distinct from primary MDS. We examined the efficacy of bone marrow transplantation (BMT) as a treatment for patients with therapy-related MDS. Eighteen patients with therapy-related MDS and twenty-five patients with primary MDS received an allogeneic, syngeneic, or unrelated donor BMT. Graft-versus-host disease prophylaxis included methotrexate, methotrexate plus cyclosporine, FK-506, or T cell depletion. Conditioning regimens consisted of cyclophosphamide/total body irradiation, with and without cytosine arabinoside, busulfan/cyclophosphamide, and cyclophosphamide/etoposide/carmustine. For patients with therapy-related MDS, the median age was 32 years and the actuarial disease- free survival was 24% (95% confidence interval 6, 42%) with a median follow-up of 3 years. For patients with primary MDS, the median age was 36 years and the actuarial diseasefree survival at 3 years was 43% (95% confidence interval 22, 64%). Four of the therapy-related patients and two of the primary patients have relapsed. Three patients experienced graft failure; all three had received T cell-depleted marrow and two had marrow fibrosis. Our results suggest that patients with therapy-related MDS can be successfully transplanted. Transplantation should be considered early in the disease, since longterm disease-free survival is achievable. Keywords: myelodysplasia; bone marrow transplantation

Myelodysplastic syndromes (MDS) are a group of clonal hematopoietic disorders characterized by hematopoietic failure and a risk of conversion to acute myelogenous leukemia (AML).1–3 MDS may arise de novo (primary MDS) or occur as a result of prior chemotherapy and/or radiation therapy (therapy-related). Because of an increasing number of long-term survivors of primary cancer and the extensive

Correspondence: Dr K Ballen, Division of Hematology/Oncology, University of Massachusetts Medical Center, 55 Lake Avenue North, Worcester, MA 01655, USA Presented in part at the 37th Meeting of the American Society of Hematology, Seattle, Washington, 2 December 1995 Received 21 October 1996; accepted 23 April 1997

use of intensive chemotherapy, the incidence of therapyrelated MDS is rising.4–7 There are also recent reports of therapy-related MDS occurring after autologous transplants for lymphoma and breast cancer.8–10 The prognosis for patients with therapy-related MDS is poor. The median survival is 4 months, with deaths due to infection, bleeding, or progression to acute leukemia.5,7,11 The prognosis for primary MDS is better, with a median survival of 15 months.2 However, the outcome is variable and the prognosis depends on blast percentage, cytogenetic abnormalities, platelet count and on stage of disease as described by French–American–British (FAB) classification.12,13 Treatment for therapy-related MDS has included intensive induction therapy similar to the regimens used for acute myelogenous leukemia. These regimens have induced a remission in 60% of patients but the remission duration is only a few months and the overall survival is poor.14–16 Other approaches have included retinoids, 5-azacytidine, cytosine arabinoside, growth factors, and hormonal agents, all with no improvement in survival.17–22 Bone marrow transplantation (BMT) is effective therapy for selected patients with primary MDS.23–26 The value of BMT in patients with therapy-related MDS has been difficult to assess because the number of reported cases is small. In addition, data for therapy-related MDS have been shown together with primary MDS, making the relative utility of transplantation for the two diseases difficult to determine.26–30 In this study, we examine the long-term outcome of allogeneic BMT in 18 patients with therapy-related MDS and compare these results to those obtained in 25 patients with primary MDS.

Patients and methods Patient population Eighteen patients with therapy-related MDS were treated with allogeneic BMT at the Brigham and Women’s Hospital between January 1980 and June 1994. Nine of these patients have been reported previously.27 Patient characteristics are listed in Table 1. Diagnoses were classified according to FAB criteria. 31 Cytogenetics were available in 12 cases and were abnormal in nine cases. Cytogenetic abnormalities include four patients with monosomy 7, two

BMT for therapy-related myelodysplasia KK Ballen et al

738

Table 1

Patient characteristics

Median age (range) Sex (male/female) Chromosomes −5 −7 5q− trisomy 8 trisomy 11 20q− t(3;3) del 2 complex normal non-available FAB classification RA RARS RAEB RAEB-IT AML Prior illness Hodgkin’s disease non-Hodgkin’s lymphoma breast cancer Wilm’s tumor polycythemia vera Prior therapy radiation alonea chemotherapy alone chemo/radiotherapy MOPP containing CMF Pretransplantation therapy daunorubicin/Ara-C growth factors pyridoxine hormones none

Therapyrelated n = 18

Primary n = 25

32 (14–51) 8/10

36 (20–54) 9/16

0 4 1 2 0 0 0 0 2 3 6

0 3 4 3 1 2 1 1 1 8 1

4 0 5 3 6

9 3 4 6 3

14 1 1 1 1 2 1 14 1 1 0 0 0 17

5 1 3 3 13

a

One received oral 32P. RA = refractory anemia; RARS = refractory anemia with ringed sideroblasts; RAEB = refractory anemia with excess blasts; RAEB-IT = refractory anemia with excess blasts in transformation; AML = acute myeloid leukemia; CMF = cyclophosphamide, methotrexate, 5-FU; MOPP = nitrogen mustard, vincristine, procarbazine, prednisone.

patients with trisomy 8 and two patients with complex abnormalities, and one with 5q−. Bone marrow biopsies were stained with hematoxylin-eosin, Wright–Giemsa, and Gomori silver stains and evaluated for cellular morphology as well as reticulin deposition and myelofibrosis. Marrow reticulin was increased in five patients. The median age of patients with therapy-related MDS was 32 years (range 14– 51). Fourteen of the patients had prior therapy for Hodgkin’s disease, and one patient each had non-Hodgkin’s lymphoma, breast cancer, polycythemia vera, and Wilm’s tumor. Fifteen patients had received extensive chemoradiotherapy, one patient chemotherapy only, one patient radiation therapy alone, and one patient oral 32P alone. The median interval from primary diagnosis to secondary MDS was 66 months (range 24–360 months) and the median disease duration of the secondary MDS prior to transplant was 6 months (range 1–11 months). Marrow donors were histo-

compatibility locus antigen/mixed lymphocyte culture (HLA/MLC) compatible siblings in 13 patients, a single antigen (Ag)-matched family donor in one patient, an HLAmatched unrelated donor in one patient, a single Ag-mismatched unrelated donor in one patient, and syngeneic donors in two patients (Table 2). During the same time period, 25 patients with primary MDS were treated with allogeneic BMT. Their disease characteristics are outlined in Table 1. Seven of these patients have been reported previously.27 Cytogenetics were abnormal in 16 cases. Cytogenetic abnormalities included monosomy 7, trisomy 8, trisomy 11, 20q−, t (3;3), del 2, and complex abnormalities. Marrow fibrosis was increased in nine patients. The median age was 36 years (range 20– 54 years). The median time from the diagnosis to transplant was 8 months (range 2–132 months). Marrow donors were HLA/MLC compatible siblings in 20 patients, a single Ag mismatched family donor in three patients, an HLA matched unrelated donor in one patient, and a single Ag mismatched unrelated donor in one patient (Table 2). Disease status Three patients with primary MDS had converted to acute myelogenous leukemia (AML) and two achieved a remission with cytosine arabinoside and daunorubicin. One other patient with primary MDS (RAEB-IT) achieved a remission with induction therapy. Six patients with therapyrelated MDS converted to AML; one of them was treated unsuccessfully with cytosine arabinoside and daunorubicin; the transplant was performed with resistant disease. None of the other patients had attempts at remission induction. Two patients had active Hodgkin’s disease at the time of transplant.

Table 2

Bone marrow transplant

Donor syngeneic matched sibling mismatched relative matched unrelated mismatched unrelated Conditioning Cy/TBI Ara-C/Cy/TBI CBV Bu/Cy GVHD prophylaxis: none MTX alone MTX/CsA MTX/CsA/solumedrol FK506 T cell depletion

Therapyrelated n = 18

Primary n = 25

2 13 1 1 1

0 20 3 1 1

0 8 4 6

1 23 0 1

2 4 7 2 0 3

0 1 18 1 1 4

Cy = cyclophosphamide; TBI = total body irradiation; Ara-C = cytarabine; B = BCNU; V = etoposide; Bu = busulfan; MTX = methotrexate; CsA = cyclosporin A.


All studies were performed under the guidelines of the Human Research Committee of the Brigham and Women’s Hospital and informed consent was obtained from each patient prior to transplantation. Patients presented here represented all patients with myelodysplasia transplanted during this time period. All patients had to demonstrate an ejection fraction .50% and pulmonary function tests .60% predicted to proceed to transplantation. Conditioning regimens and graft-versus-host disease (GVHD) prophylaxis used are summarized in Table 2. The patients were treated according to protocols that were active at the time of their illness. Eight of the therapy-related and 23 of the primary MDS patients received cytosine arabinoside (500 mg/m2/day via continuous infusion for 7 days or 3 g/m2 twice a day for 3 days or 500 mg/m2 twice a day for 3 days) followed by cyclophosphamide (1800 mg/m2/day for 2 days), and fractionated total body irradiation (TBI) 1200–1400 cGy administered in six to eight doses at 5–10 cGy/min from opposing anteroposterior/posteroanterior 4 MeV linear accelerators. One patient received cyclophosphamide 1800 mg/m2/day for 2 days and TBI 1200 cGy. Those patients who had received mediastinal or abdominal radiation therapy in the past were conditioned with chemotherapy alone. Six therapy-related patients and one primary patient received busulfan 4 mg/kg/day for 4 days followed by cyclophosphamide 1600 mg/m2/day for 4 days. Four patients, all with therapy-related MDS, received carmustine (BCNU) 112.5 mg/m2/day for 4 days, cyclophosphamide 750 mg/m2 twice a day for 4 days, and etoposide 200 mg/m2 twice a day for 4 days (CBV).32,33 Prophylaxis for GVHD was administered according to active protocols at the time of treatment (Table 2). Methotrexate (10 or 15 mg/m2/day on days 1, and 10 mg/m2/day on days 3, 6 and 11 or 10 mg/m2/day on days 1, 8, 15, 22 and 29) plus cyclosporine were administered to 25 patients. Three of the four recipients of unrelated donor transplants also received solumedrol 3 mg/kg/day as part of their GVHD prophylaxis. Five patients received methotrexate alone 10 mg/m2 on days 1, 3, 6 and 11. One patient received FK506 alone. Seven patients had marrows purged of T cells with anti-CD5 monoclonal antibodies – either anti-Leu-1 (Becton Dickinson, Mountain View, CA, USA) plus baby rabbit complement or ST 1 immunotoxin (ST1 immunotoxin, Sanofi Research, Montpelier, France) conjugated to ricin A chain.34 These patients did not receive any pharmacologic GVHD prophylaxis. No GVHD prophylaxis was given to the syngeneic transplants. All patients were nursed in laminar air flow rooms and received low bacteria diets and gut decontamination. Patients were evaluated for GVHD by an attending physician and GVHD was graded according to standard criteria.35 Statistics The outcomes measured in this study were overall survival, relapse rate, and treatment-related toxicity. Disease-free survival measured from the time of transplantation was analyzed according to the method of Kaplan and Meier.36 Confidence limits were calculated according to the formula of

Peto et al.37 Individual prognostic factors including age, cytogenetics, FAB subclass, and type of donor, were evaluated first with the log-rank test.37 For multivariate analysis, these prognostic characteristics were evaluated simultaneously in a Cox proportional hazards model. Statistical analysis was performed using Systat software (Evanston, IL, USA). All P values are two-sided. Results Allogeneic BMT for secondary MDS Five of the eighteen patients are alive with a median followup of 3 years (range 1–14). All five are disease-free for an actuarial disease-free survival of 24% (95% confidence interval 6, 42%), as shown in Figure 1. The median time to engraftment (.500 granulocytes/ml) was 20 days (range 14–30 days). Graft failure occurred in one recipient of a T cell-depleted bone marrow who received two subsequent transplants and eventually died of adult respiratory distress syndrome (ARDS). Four patients relapsed and died of recurrent disease (Table 3). The 13 patients who received a transplant from an HLA-matched sibling had an actuarial disease-free survival of 27% at 3 years. Acute GVHD was considered assessable if the patient had engrafted by 30 days, and chronic GVHD was assessed if there was sustained engraftment at .100 days. No patient died of GVHD before engraftment. Acute GVHD, grades 2–4, occurred in seven patients and was fatal in two patients. Mild chronic GVHD developed in two patients, both of whom are living. Additional causes of death included veno-occlusive disease (VOD) of the liver in two patients and ARDS, pneumocystis pneumonia, aspergillus, and cytoxan cardiomyopathy in one patient each. Overall, regimen-related toxicity resulted in six deaths in the 18 patients (33%). Including the two deaths from GVHD and the one death from graft failure, the transplant-related mortality was 50% (nine of 18 patients).

1.00

0.75

Probability

Treatment

0.50

Primary MDS

0.25 Secondary MDS 0.00 0.0

1.0

2.0

3.0

4.0

5.0

12-15.0

Years from transplantation Figure 1 Allogeneic BMT for myelodysplasia: all patients. Actuarial disease-free survival of patients with primary MDS compared with secondary MDS.

739


740

Table 3

Outcome Therapyrelated n = 18 (%)

Acute GVHD grade 0 I II III IV not evaluable

(33) (11) (22) (17) 0 3 (17)

Causes of death GVHD relapse graft failure cardiac failure respiratory failure veno-occlusive disease infection

2 4 1 1 1 2 2

6 2 4 3

(11) (22) (6) (6) (6) (11) (11)

Primary n = 25 (%)

4 10 6 2 1 2

(16) (40) (24) (8) (4) (8)

5 (20) 4 (4) 2 (8) 0 4 (16) 1 (4) 3 (12)

Primary MDS Ten of the 25 patients are alive with a median follow-up of 4 years (range 1.0–12 years) Nine of these 10 patients are disease-free for an actuarial disease-free survival at 3 years of 43% (95% confidence interval 22, 64%) (Figure 1). The median time to engraftment (.500 granulocytes/ml) was 20 days (range 15–28 days). Graft failure occurred in two patients who received second transplants and eventually succumbed to infection. Both patients had increased fibrosis in their marrow and received T cell-depleted transplants. Two patients have relapsed, one of whom has died. Those patients (21 patients) who received a transplant from an HLA-matched sibling had an actuarial disease-free survival of 47% at 3 years. Acute GVHD grades 2–4 occurred in nine patients and was the cause of death in three patients. Chronic GVHD occurred in three patients. Death was attributed to chronic GVHD in two patients. Other causes of death included two cases each of cytomegalovirus (CMV) pneumonia, ARDS, and invasive fungal infections, and one case each of VOD and bowel necrosis. Thus, regimen-related toxicity resulted in eight deaths in 25 patients (31%). Including the five patients who died of GVHD and the two patients who died of graft failure, the transplant-related mortality was 60% (15 of 25 patients). We compared the outcomes between patients with therapy-related MDS and primary MDS. The difference in survival rates between the two groups did not reach statistical significance (P = 0.42). The relapse rate was 22% (four of 18 patients) for patients with therapy-related MDS and 8% (two of 25 patients) for patients with primary MDS. This difference also did not reach statistical significance (P = 0.38). There was no difference in regimen-related toxicity between the two group of patients (31 vs 33%, P = 0.77). However, the small number of patients treated made it difficult to detect small differences. We combined data from both therapy-related and the primary MDS patients to study the effect of various prognostic factors on outcome. In univariate analyses, the differences

in outcome by age, cytogenetics, FAB subclass, and type of donor did not reach statistical significance. Patients evaluated for cytogenetic anomalies were further grouped following the suggestions by Greenberg et al.38 The group considered to have a poor prognosis included patients with complex (ie more than two anomalies) and chromosome 7 anomalies. Patients who had chromosome 7 anomalies and other abnormal cytogenetics were included with the chromosome 7 group. The 3-year Kaplan–Meier survival rate for this group was 20% as compared with 49% for patients with other cytogenetic subgroups as listed in Table 1. The difference in survival rates, however, did not reach statistical significance (P = 0.13). The differences in survival rates were also not statistically significant after stratifying the analysis according to therapy or primary myelodysplasia. In a multivariate analysis that included the 36 patients with concurrent information on age, FAB subtypes, donor match and cytogenetics, patients with a fully matched sibling donor, had a better survival rate than those without a fully matched sibling donor (relative risk 11.8, P = 0.003). Patients younger than 30 had a better survival rate than older patients (relative risk 4.6, P = 0.02). Relative risk estimates for FAB subtypes and cytogenetics did not reach statistical significance in the four-factor multivariate analysis. A good risk group of patients less than age 30 and with FAB subtypes refractory anemia, refractory anemia with ringed sideroblasts, and a fully matched sibling donor had a better survival rate. However, only six patients fell into this good risk group. We have analyzed our data separating the MDS patients from those patients who had already transformed to AML. The survival experience was not different (log-rank test P = 0.98) among the three groups that included primary MDS, therapy-related MDS and AML patients. Two patients developed AML post-primary MDS and received successful induction chemotherapy. One of these patients died of relapse disease and the other is alive with chronic GVHD. We have previously reported that patients with marrow fibrosis and T cell depletion had a high incidence of graft failure.27 Fourteen patients had evidence of marrow fibrosis; nine were in the primary MDS group. Three patients – two with primary MDS and one with therapyrelated MDS – experienced graft failure. All three patients received T cell-depleted transplants and two of the three also had marrow fibrosis (Table 4). Although the type of GVHD prophylaxis did not affect overall survival (P = 0.54), T cell depletion was highly predictive for graft failure (P = 0.006). Two of 14 patients with marrow fibrosis had graft failure; one of 29 patients without marrow fibrosis had graft failure. Table 4

Effect of T cell depletion on graft failure

GVHD prophylaxis

T cell depletion Pharmacologic prophylaxis

P value by Fisher’s exact test.

Graft failure

No graft failure

3 0

4 36 P = 0.006


Discussion The prognosis for therapy-related MDS is poor, with survival of only a few months. Consequently, allogeneic BMT has been offered to patients with therapy-related as well as primary MDS. 15,26–30,39 The appropriate timing, conditioning regimen, and GVHD prophylaxis for transplantation of these patients is uncertain, and there has not been an organized effort to determine the optimal therapy. Most survival data include mixtures of therapy-related and primary MDS.26,40 Our study represents the largest single-institution collection of therapy-related patients that are compared directly to patients with primary MDS. We have shown here that bone marrow transplantation is a curative procedure for patients with therapy-related MDS. There was some difference in the outcome of the therapy-related as compared with the primary MDS patients. However, this observed difference did not attain statistical significance, perhaps due to the small number of subjects in this study. Four patients with therapy-related MDS and two patients with primary MDS relapsed; the four therapy-related patients and one of the two primary MDS patients have died of relapsed disease. This difference also did not reach statistical significance. Cytogenetics had no predictive significance in this study, perhaps due to the small number of patients. In our earlier reports, we found that patients with therapy-related MDS had a higher regimen-related toxicity than patients with primary MDS.27,41 We postulated that this difference might be related to the toxicity of prior therapy. In our current study, with more than twice the number of patients, this difference was not significant (33% for therapy-related vs 31% for primary). The transplant-related mortality, which also includes deaths from GVHD and graft failure, was high for both groups, 50% for the therapyrelated MDS patients and 60% for the primary MDS patients. The two groups did not receive the same preparative regimens; more patients in the therapy-related MDS group were conditioned without TBI, as they had received prior radiation treatment for their initial disease. Because this is a retrospective study over many years of a relatively uncommon disease, patients were treated with different conditioning and GVHD prophylactic regimens based on the year of transplant and current protocols active at that time. Because of the size of the study it was difficult to determine if the year of transplant had any impact on survival. Six therapy-related patients were transplanted after 1990 and three survive; 17 primary MDS patients were transplanted after 1990 and seven survive. The transplant-related mortality was 50% in our patients. This may represent the intensive conditioning regimen which included Ara-C, in addition to cytoxan/TIB. This transplant-related mortality is similar to other studies in the literature. O’Donnell et al23 report 50% transplant-related mortality in their 20 patients, of whom one had therapyrelated MDS. The Seattle group has also shown transplantrelated mortalities ranging from 36 to 68%.42 The high transplant-related mortality in MDS is confirmed in a recent report in the literature.43 Only six of our patients received cytotoxic induction chemotherapy prior to transplant; remission was achieved

in three of these patients. Two of the six patients who had received prior cytotoxic therapy survived the transplant. Although the effectiveness of cytotoxic chemotherapy prior to transplant on the outcome of transplant cannot be judged adequately from this study, pre-transplant chemotherapy may not be necessary to achieve a remission and may only delay the transplant. Patients who received a T cell-depleted transplant had a higher risk of graft failure. We and others have shown that T cell-depleted transplants demonstrate a higher risk of graft failure in patients with MDS compared to patients with leukemia.34,41,44 Thus, our data would indicate that pharmacologic GVHD prophylaxis may be safer for this group of patients. No patients who received pharmacologic GVHD prophylaxis experienced graft failure, including 12 patients with marrow fibrosis. The Seattle group has recently shown that fibrosis had no effect on engraftment rates.45 The optimal conditioning regimen for these patients remains unknown.46 In this study, 32 patients were conditioned with a TBI containing regimen. Those patients who had received radiation therapy in the past were conditioned with a chemotherapy-only conditioning regimen, either busulfan/cyclophosphamide or cyclophosphamide/ BCNU/etoposide (CBV). Although the size of our study and the differences between the TBI and chemotherapy only groups did not permit a direct comparison of conditioning regimens, three of the 10 therapy-related patients who were conditioned with chemotherapy alone are alive. The efficacy of the busulfan/ cyclophosphamide regimen for patients with myelodysplasia was recently shown.47,48 Cyclophosphamide alone is ineffective; thus either busulfan or BCNU and etoposide must be added to the conditioning regimen.49 Allogeneic transplants have traditionally been limited to young patients with an HLA-matched sibling donor. However, only 30% of patients would be estimated to have a matched sibling to serve as donor. Therefore, the use of unrelated donors has been increasingly important. Two primary MDS patients and two therapy-related MDS patients received unrelated donor transplants; two of these four transplants were from a single antigen-mismatched unrelated donor, a group that has been shown to have a high mortality.50 One of these four patients, who had therapyrelated MDS and received a fully matched unrelated donor transplant, survived. Owing to the small number of patients, the survival rate for patients who received an unrelated donor transplant (n = 4) was not statistically different from the survival rate for those who received a family donor transplant (n = 39). The National Marrow Donor Program reported 18% disease-free survival in 32 patients with myelodysplasia who received transplants from unrelated donors; these data included both therapy-related and primary MDS patients.50 As the number of patients with unrelated donor transplants increases, we may better be able to analyze prognostic factors in this particular group. Other investigators have reported similar survival rates for patients with therapy-related MDS undergoing BMT.15,23,25,28–30,39 These studies all confirmed survival rates ranging from 18 to 50%. This wide range of results may reflect the small numbers as well as the different

741


742

patient populations in these studies. Anderson et al26 analyzed the Seattle experience with bone marrow transplants in 93 patients with MDS, of whom eight patients were therapy-related. Disease-free survival was 41% at 4 years. The therapy-related MDS patients were not analyzed separately. Multivariate analysis revealed that younger age and shorter disease duration were associated with improved diseasefree survival, a finding similar to larger studies of patients with chronic myelogenous leukemia (CML).51,52 A more recent study from the same group confirms that younger age and shorter disease duration are associated with improved survival in patients with refractory anemia.47 The European Bone Marrow Transplant Group compared results of transplants in 28 patients with therapy-related MDS/AML with those of 53 patients with primary MDS or AML evolved from MDS. Disease-free survival was similar in the primary and the therapy-related MDS groups, but the relapse rate was higher in the therapy-related group. 14,53 Marrow grafting offers a chance for cure for an otherwise rapidly fatal disease. Patients with therapy-related MDS have a more aggressive natural history than patients with primary MDS. Therefore, bone marrow transplantation may be safer if performed early in the course of the disease before increasing progression of the disease, infection, and toxicity from blood products make the procedure less effective.

9

10 11 12

13

14 15 16

17

Acknowledgements We are indebted to Carol McGarigle, the transplant nurses, the Hematology/Oncology fellows, and medical house staff of the Brigham and Women’s Hospital for their excellent care of these patients. We thank Dr Robert Handin for his encouragement and critical review of the manuscript. This work was supported in part by National Institutes of Health Grants 5K12-AG00294, CA58661 and CA39542.

18

19

20

References 1 Buzaid AC, Garewal HS, Greenberg BR. Management of myelodysplastic syndromes. Am J Med 1986; 80: 1149–1157. 2 Cheson BD. The myelodysplastic syndromes: current approaches to therapy. Ann Intern Med 1990; 112: 932–941. 3 Fialkow PJ, Singer JW, Raskind WH et al. Clonal development, stem-cell differentiation, and clinical remissions in acute nonlymphocytic leukemia. New Engl J Med 1987; 317: 468–473. 4 Kantarjian HM, Keating MJ. Therapy-related leukemia and myelodysplastic syndrome. Semin Oncol 1987; 14: 435–443. 5 Levine EG, Bloomfield CD. Leukemias and myelodysplastic syndromes secondary to drug, radiation, and environmental exposure. Semin Oncol 1992; 19: 47–84. 6 Levine EG, Bloomfield CD. Secondary myelodysplastic syndromes and leukemias. Clin Hematol 1986; 15: 1037–1078. 7 Michels SD, McKenna RW, Arthur DC et al. Therapy-related acute myeloid leukemia and myelodysplastic syndrome: a clinical and morphologic study of 65 cases. Blood 1986; 65: 1364–1372. 8 Roman-Unfer S, Bitran JD, Hanauer S et al. Acute myeloid leukemia and myelodysplasia following intensive chemo-

21 22

23 24 25

26 27

therapy for breast cancer. Bone Marrow Transplant 1995; 16: 163–168. Miller JS, Arthur DC, Litz CE et al. Myelodysplastic syndrome after autologous bone marrow transplantation: an additional late complication of curative cancer therapy. Blood 1994; 83: 3780–3786. Stone RM. Myelodysplastic syndrome after autologous transplantation for lymphoma: the price of progress? Blood 1994; 83: 3437–3440. Kantarjian HM, Estey EH, Keating MJ. Treatment of therapyrelated leukemia and myelodysplastic syndrome. Hematol/ Oncol Clin N Am 1993; 7: 81–106. Le Beau MM, Albain KS, Larson RA et al. Clinical and cytogenetic correlations in 63 patients with therapy-related myelodysplastic syndromes and acute nonlymphocytic leukemia: further evidence for characteristic abnormalities of chromosomes no. 5 and 7. J Clin Oncol 1986; 4: 325–345. Pedersen-Bjergaard J, Philip P, Larsen SO et al. Therapyrelated myelodysplasia and acute myeloid leukemia. Cytogenetic characteristics of 115 consecutive cases and risk in seven cohorts of patients treated intensively for malignant disease in the Copenhagen series. Leukemia 1993; 7: 1975–1986. De Witte T. New treatment approaches for myelodysplastic syndromes and secondary leukemias. Ann Oncol 1994; 5: 401–408. Cheson BD. Chemotherapy and bone marrow transplantation for myelodysplastic syndromes. Semin Oncol 1992; 19: 85–94. De Witte T, Blacklock HA, Prentice HG et al. Allogeneic bone marrow transplantation in a patient with acute myeloid leukemia secondary to Hodgkin’s disease. Cancer 1984; 53: 1507–1508. Cheson BD, Jasperse DM, Simon R et al. A critical appraisal of low-dose cytosine arabinoside in patients with acute nonlymphocytic leukemia and myelodysplastic syndromes. J Clin Oncol 1986; 4: 1857–1864. Chitambar CR, Libnoch JA, Matthaeus WG. Evaluation of continuous infusion low-dose 5-azacytidine in the treatment of myelodysplastic syndromes. Am J Hematol 1991; 37: 100–104. Clark RE, Ismail AD, Jacobs A et al. A randomized trial of 13-cis retinoic acid with or without cytosine arabinoside in patients with the myelodysplastic syndrome. Br J Haematol 1987; 66: 77–83. Ganser A, Lindermann A, Seipelt G et al. Effects of recombinant human interleukin-3 in patients with normal hematopoiesis and in patients with bone marrow failure. Blood 1990; 75: 666–676. Priesler HD, Raza A, Barcos M et al. High-dose cytosine arabinoside as the initial treatment of poor-risk patients with acute nonlymphocytic leukemia. J Clin Oncol 1987; 5: 75–82. Gradishar WJ, Le Beau MM, O’Laughlin R et al. Clinical and cytogenetic responses to granulocyte–macrophage colonystimulating factor in therapy-related myelodysplasia. Blood 1992; 80: 2463–2470. O’Donnell JR, Nademanee AP, Snyder DS et al. Bone marrow transplantation for myelodysplastic and myeloproliferative syndromes. J Clin Oncol 1987; 5: 1822–1826. Guinan EC, Tarbell NJ, Tantravahi R et al. Bone marrow transplantation for children with myelodysplastic syndromes. Blood 1989; 73: 619–622. Belanger R, Gyger M, Perreault C et al. Bone marrow transplantation for myelodysplastic syndromes. Br J Haematol 1988; 69: 29–33. Anderson JE, Appelbaum FR, Fisher LD et al. Allogeneic bone marrow transplantation for 93 patients with myelodysplastic syndrome. Blood 1993; 82: 677–681. Longmore G, Guinan EC, Weinstein HJ et al. Bone marrow


28 29

30

31

32

33 34

35 36 37 38 39 40

transplantation for myelodysplasia and secondary acute nonlymphocytic leukemia. J Clin Oncol 1990; 8: 1707–1714. Bunin NJ, Casper JT, Chitambar C et al. Partially matched bone marrow transplantation in patients with myelodysplastic syndromes. J Clin Oncol 1988; 5: 1851–1855. De Witte T, Zwaan F, Hermans J et al. Allogeneic bone marrow transplantation for secondary leukemia and myelodysplastic syndrome; a survey by the leukemia working party of the European Bone Marrow Transplantation Group. Br J Haematol 1990; 74: 151–155. Geller RB, Vogelsang GB, Wingard JR et al. Successful marrow transplantation for acute myelocytic leukemia following therapy for Hodgkin’s disease. J Clin Oncol 1988; 6: 1558– 1561. Bennett JM, Catovsky D, Daniel MT et al. Proposals for the classification of the acute leukemias: French–American–British (FAB) co-operative group. Br J Haematol 1976; 33: 451–458. Wheeler C, Antin JH, Churchill WH et al. Cyclophosphamide, carmustine, and etoposide with autologous bone marrow transplantation in refractory Hodgkin’s disease and non-Hodgkin’s lymphoma: a dose-finding study. J Clin Oncol 1990; 8: 648–656. Mitus AJ, Miller KB, Schenkein DP et al. Improved survival for patients with acute myelogenous leukemia. J Clin Oncol 1995; 3: 560–569. Antin JH, Bierer BE, Smith BR et al. Selective depletion of bone marrow T lymphocytes with anti-CD5 monoclonal antibodies: effective prophylaxis for graft-versus-host disease in patients with hematologic malignancies. Blood 1991; 78: 2139–2149. Glucksberg H, Storb R, Fefer A et al. Clinical manifestations of graft-versus-host disease in human recipients from HLAmatched sibling donors. Transplantation 1974; 18: 295–304. Kaplan G, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481. Peto R, Pike MC, Armitage P et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Br J Cancer 1977; 35: 1–39. Greenberg P, Fenaux P, Morel P et al. International workshop consensus risk analysis system for myelodysplastic syndromes (MDS). Blood 1995; 86 (Suppl. 1): 270a. Sargur M, Buckner CD, Appelbaum FR et al. Marrow transplantation for acute nonlymphocytic leukemia following therapy for Hodgkin’s disease. J Clin Oncol 1987; 5: 731–734. Sutton L, Leblond V, Le Maignan C. Bone marrow transplantation for myelodysplastic syndrome and secondary leukemia: outcome of 86 patients. Bone Marrow Transplant 1991; 7 (Suppl. 2): 39.

41 Ballen KK, Antin JH. Treatment of therapy-related acute myelogenous leukemia and myelodysplastic syndromes. Hematol/Oncol Clin N Am 1993; 7: 477–491. 42 Anderson JE, Appelbaum FR, Schoch G et al. Allogeneic marrow transplantation for myelodysplastic syndrome with advanced disease morphology: a phase II study of busulfan, cyclophosphamide, and total-body irradiation and analysis of prognostic factors. J Clin Oncol 1996; 14: 220–226. 43 Fung HC, Shepherd JD, Nantel SH et al. Allogeneic bone marrow transplantation for adults with primary myelodysplastic syndrome (MDS): evaluation of prognostic factors. Blood 1996: 88 (Suppl. 1): 480a (Abstr. 1909). 44 O’Reilly RJ, Collins NH, Kernan NA et al. Transplantation of marrow depleted of T cells by soybean lectin agglutination and E rosette depletion: major histocompatibility complexrelated graft resistance in leukemic transplant recipients. Transplant Proc 1985; 17: 455–459. 45 Soll E, Massumoto C, Clift RA et al. Relevance of marrow fibrosis in bone marrow transplantation: a retrospective analysis of engraftment. Blood 1995; 86: 4667–4673. 46 Schultz AB, Geller RB, Hillyer CD. The role of bone marrow transplantation in the treatment of myelodysplastic syndromes. J Hematother 1995; 4: 323–324. 47 Anderson JE, Appelbaum FR, Schoch G et al. Allogeneic marrow transplantation for refractory anemia: a comparison of two preparative regimens and analysis of prognostic factors. Blood 1996; 87: 51–58. 48 O’Donnell MR, Long GD, Parker PM et al. Busulfan/ cyclophosphamide as conditioning regimen for allogeneic bone marrow transplantation for myelodysplasia. J Clin Oncol 1995; 13: 2973–2979. 49 Nevill TJ, Shepherd JD, Reece DE et al. Treatment of myelodysplastic syndrome with busulfan-cyclophosphamide conditioning followed by allogeneic BMT. Bone Marrow Transplant 1992; 10: 445–450. 50 Kernan NA, Bartsch G, Ash RC et al. Analysis of 462 transplantations from unrelated donors facilitated by the national marrow donor program. New Engl J Med 1993; 328: 593–602. 51 Thomas ED, Clift RA. Indications for marrow transplantation in chronic myelogenous leukemia. Blood 1989; 73: 861–865. 52 Thomas ED, Clift RA, Fefer A et al. Marrow transplantation for the treatment of chronic myelogenous leukemia. Ann Intern Med 1986; 104: 155–161. 53 De Witte T, Hermans J, Van Viezen A et al. Prognostic variables in bone marrow transplantation for secondary leukemia and myelodysplastic syndromes: a survey of the Working Party of Leukemia. Bone Marrow Transplant 1991; 8 (Suppl. 1): 40.

743