Paediatric leukemia Allogeneic bone marrow transplantation ... - Nature

0 downloads 0 Views 183KB Size Report
Feb 2, 2004 - USA; and 2Department of Pediatric Hematology/Oncology, Vanderbilt University, ... Keywords: allogeneic BMT; pediatric MDS; RA; RARS;.
Bone Marrow Transplantation (2004) 33, 805–814 & 2004 Nature Publishing Group All rights reserved 0268-3369/04 $25.00

www.nature.com/bmt

Paediatric leukemia Allogeneic bone marrow transplantation in children with myelodysplastic syndrome or juvenile myelomonocytic leukemia: the Seattle experience U Yusuf1, HA Frangoul2, TA Gooley1, AE Woolfrey1, PA Carpenter1, RG Andrews1, HJ Deeg1, FR Appelbaum1, C Anasetti1, R Storb1 and JE Sanders1 1 Fred Hutchinson Cancer Research Center and University of Washington Departments of Pediatrics and Medicine, Seattle, WA, USA; and 2Department of Pediatric Hematology/Oncology, Vanderbilt University, Nashville, TN, USA

Summary: The purpose of this study was to evaluate the role of allogeneic bone marrow transplantation (BMT) in children with myelodysplastic syndrome (MDS). In total, 94 consecutive pediatric patients with MDS received an allogeneic BMT from 1976 to 2001 for refractory anemia (RA) (n ¼ 25), RA with ringed sideroblasts (RARS) (n ¼ 2), RA with excess blasts (RAEB) (n ¼ 20), RAEB in transformation (RAEB-T) (n ¼ 14), juvenile myelomonocytic leukemia (JMML) (n ¼ 32) or chronic myelomonocytic leukemia (CMML) (n ¼ 1). The estimated 3-year probabilities of survival, event-free survival (EFS), nonrelapse mortality and relapse were 50, 41, 28 and 29%, respectively. Patients with RA/RARS had an estimated 3-year survival of 74% compared to 68% in those with RAEB and 33% in patients with JMML/CMML. In multivariable analysis, patients with RAEB-T or JMML were 3.9 and 3.7 times more likely to die compared to those with RA/RARS and RAEB (P ¼ 0.005 and 0.004, respectively). Patients with RAEB-T were 5.5 times more likely to relapse (P ¼ 0.01). The median follow-up among the 43 surviving patients is 10 years (range 1–25). We conclude that allogeneic BMT for children with MDS is well tolerated and can be curative. Bone Marrow Transplantation (2004) 33, 805–814. doi:10.1038/sj.bmt.1704438 Published online 2 February 2004 Keywords: allogeneic BMT; pediatric MDS; RA; RARS; RAEB; RAEB-T; JMML; CMML

Myelodysplastic syndrome (MDS) comprises a heterogeneous group of clonal hematologic disorders characterized by peripheral blood cytopenia and dysplastic changes in the bone marrow.1,2 The use of the French–American– British (FAB) classification of preleukemic conditions, used widely for classification of adult MDS, has been less well

Correspondence: Dr JE Sanders, Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue North, D5-280, Seattle, WA 98109-1204 USA; E-mail: [email protected] Received 10 April 2003; accepted 13 November 2003 Published online 2 February 2004

accepted in childhood MDS.3,4 Various attempts to classify childhood MDS have met with difficulty and lack of agreement due in part to the fact that childhood MDS often evolves from genetic predisposing conditions and has rarely been therapy related.3 The classification that includes some of the FAB subtypes, juvenile myelomonocytic leukemia (JMML), the monosomy 7 syndrome and others has not been universally accepted.5 The actual incidence of MDS in children is not known, but is estimated to be between 1 and 3% of childhood leukemias.6–8 In contrast to adult MDS patients, the preleukemic phase in children is usually short with relatively rapid progression to overt leukemia.6,9 The less aggressive subtypes, refractory anemia (RA) and refractory anemia with ring sideroblasts (RARS), are extremely rare in childhood and the majority of children also have associated chromosome abnormalities, the most frequent of which, monosomy 7, is found in 40–50%.10,11 In fact, children with MDS are frequently referred to as either having ‘adult type,’ monosomy 7 or JMML. Monosomy 7 disease in children has a particularly poor prognosis.9–14 Those with chronic myeloproliferative disease may die of marrow failure, develop myelofibrosis or progress to AML.12–14 Monosomy 7 associated AML is very difficult to treat as it is often resistant to chemotherapy. Two studies report outcome of children with monosomy 7 treated with conventional approaches. Of 16, 13 progressed to AML, one of 16 developed myelofibrosis, one developed progressive marrow failure and infection and one patient’s disease had not yet progressed.13,14 Among the 13 with AML, only three treated with intensive chemotherapy achieved remission for 9 months before they relapsed and died of progressive disease. Others have also observed that the aggressive chemotherapy used to treat monosomy 7 associated AML in children and young adults results in survival at best in the range of 15%.6,15,16 Allogeneic marrow transplantation has been the only potentially curative treatment for adult and pediatric patients with MDS. There are few data regarding the outcome of transplantation for children with various types of MDS.5,17–21 This report presents the results of allogeneic marrow transplantation for 94 consecutive children with MDS or JMML. A total of 14 of the JMML patients have been previously reported,18 and 29 of the patients with other subtypes of MDS have been previously included in analyses including adult MDS patients.22–24 Analysis of

Marrow transplant for childhood MDS U Yusuf et al

806

prognostic factors and long-term survival demonstrates that bone marrow transplantation (BMT) is well tolerated and may be curative.

Patients and methods Patients Between March 1976 and March 2001, 94 consecutive patients younger than 18 years of age received an allogeneic BMT at the FHCRC for treatment of MDS or JMML. These patients had RA, RARS, refractory anemia with excess blasts (RAEB), refractory anemia with excess blasts in transformation (RAEB-T), juvenile myelomonocytic leukemia (JMML) and chronic myelomonocytic leukemia (CMML) as defined by the French–American–British (FAB) classification system3,4 for MDS and the diagnostic criteria for JMML established by the international consensus group.25 Patients who had acute myeloid leukemia (bone marrow examination showing 430% blasts) evolving from MDS were excluded. Patient characteristics at transplant are summarized in Table 1. The median age at transplant was 5.4 years and disease duration ranged from 1 to 76 months with a median of 8 months. No specific predisposing etiology for the MDS was apparent in 73 patients, 12 had a history of severe aplastic anemia and one had Fanconi’s anemia. Three patients developed MDS following previous chemotherapy. One of these three patients with juvenile rheumatoid arthritis had prolonged exposure to chlorambucil, the second patient was treated with Imuran and steroids for SLE for 6 years and the third patient received Etoposide as part of therapy for T-cell acute lymphoblastic leukemia. Treatment prior to referral for transplant consisted of transfusions of red blood cells and platelets for 16 patients. A total of 60 patients received therapies other than transfusions (Table 1); 18 patients received no chemotherapy or transfusions prior to transplant. All patients were evaluated at the FHCRC immediately prior to transplant with complete bone marrow morphologic examinations and chromosomal analyses. Chromosome analyses were performed by G-banding after 1.5 and 24 h in culture without added mitogens (Table 1). The bone marrow morphology showed RA for 25, RARS for two, RAEB for 20, RAEB-T for 14, JMML for 32 and CMML in one. In total, 30 had monosomy 7 and four had trisomy 8, one of whom also had trisomy 21. A total of 20 patients had a combination of other cytogenetic abnormalities. Histocompatibility testing was performed by the FHCRC Clinical Immunogenetics Laboratory for all patients and donors according to previously described methods 26,27 (Table 2).

Clinical and laboratory characteristics at diagnosis These patients were a median of 5.4 (0.7–17.8) years at diagnosis. The majority of patients presented with symptoms relating to neutropenia, thrombocytopenia and anemia (n ¼ 88). Unusual presentations included viral myocarditis in one patient, retinal hemorrhage in two, Bone Marrow Transplantation

Table 1 (N ¼ 94)

Clinical and laboratory characteristics at transplantation

Characteristic

No. of patients

Median age in years (range) Gender, male/female Median disease duration months (range)

5.4 (0.7–17.8) 58/36 8 (1–76)

Disease etiology Idiopathic Post aplastic anemia Post chemotherapy Neurofibromatosis Fanconi Kostman neutropenia Schwachman Diamond Amegakaryocytic thrombocytopenia

73 12 3 2 1 1 1 1

(76.8) (12.6) (3.2) (2.1) (1.1) (1.1) (1.1) (1.1)

Prior therapy None Transfusions only AML-type therapy Aplastic-anemia type therapy Other chemotherapy Splenectomy pretransplant

18 16 20 18 22 17

(29.5) (6.3) (21.1) (18.9) (23.2) (18.0)

Disease morphology at the time of transplant RA RARS RAEB RAEB-T CMML JMML

25 2 20 14 1 32

(26.3) (2.1) (21.3) (14.7) (0.01) (33.7)

Cytogenetic evaluation at the time of transplant Normal Monosomy 7 Monosomy 7 plus other abnormality Trisomy 8 Other abnormalities

40 25 5 4 20

(42.1) (26.3) (5.3) (4.2) (21.1)

Median Median (range) Median Median (range) Median

white count at transplant mm3 (range) granulocyte count at transplant mm3

3000 (400–20 000) 600 (20–4500)

hemoglobin at transplant g/dl (range) platelet count at transplant mm3

9.5 (3.3–15) 26 (2–656)

hemoglobin F (%) (range)

26 (9–60)

Patients cytomegalovirus antibody titer Negative Positive

49 (51.6) 45 (47.4)

RA ¼ refractory anemia; RAEB ¼ refractory anemia with excess blasts; RAEBT ¼ refractory anemia with excess blasts in transformation; CMML ¼ chronic myelomonocytic leukemia; JMML ¼ juvenile myelomonocytic leukemia.

intracerebral hemorrhage in one, syncopal attacks in five, failure to thrive in six, deep-vein thrombosis in one and deficiency of growth and thyroid hormone in one patient. Children with JMML were generally younger (median age 2.5 years) than those with other forms of MDS. Hepatosplenomegaly and lymphadenopathy were universal presentations in all patients with JMML, and 17 of these patients had a splenectomy for hypersplenism prior to transplant. Vesicular skin rash was also one of the presenting features in 12 JMML patients. Patients with JMML had higher white blood counts at presentation

Marrow transplant for childhood MDS U Yusuf et al

807 Table 2

Transplant characteristics

Characteristic Marrow donor Relation to patient Sibling Parent URD

No. of patients (%)

36 (38.3) 21 (22.3) 37 (39.4)

Donor matching Matched related Matched unrelated Mismatched related Mismatched unrelated

30 30 27 7

(32) (31) (30) (7)

HLA matching Match 1 Antigen mismatch 2 Antigen mismatcha 3 Antigen mismatch

59 11 12 12

(62) (12) (13) (13)

Preparative regimenb CY and 12.5–15.75 Gy TBI CY and BU CY, BU and 12 Gy TBI BU and 12 Gy TBI FLU and 200cG TBI TBI, MEL, ATG

57 18 11 6 1 1

(60) (18.9) (11.6) (6.3) (1.1) (1.1)

GVHD prophylaxis Cyclosporine and methotrexate FK506 and methotrexate Cyclosporine Methotrexate

90 1 1 2

(95.7) (1.1) (1.1) (2.1)

CY ¼ cyclophosphamide; BU ¼ busulfan; TBI ¼ total body irradiation; FLU ¼ fludarabine; MEL ¼ melphalan. a One patient with 2 antigen mismatch related donor received anti-CD3 monoclonal antibody infusion. b Three patients with HLA-incompatible donors also received antithymocyte globulin.

with elevated fetal hemoglobin (HbF) noted in 26 (81%) of these patients.

Preparative regimen and supportive care Transplant characteristics are summarized in Table 2. Preparative regimens depended on protocols in use at the time of transplant. The most commonly used regimen consisted of cyclophosphamide (CY) 120 mg/kg i.v. over 2 days and total body irradiation (TBI) 12–15.75 Gy. The TBI was delivered at 6–8 cGy/min using opposing Cobalt 60 sources in single daily fractions of 200 or 225 cGy for 6 or 7 consecutive days or hyperfractionated schedules administered as 1.2 Gy three times per day for 4 consecutive days as previously described.28,29 In total, 18 patients received busulfan (BU) 16 mg/kg orally over 4 days and CY 120 mg/kg IV over 2 days; 11 patients received BU, CY 50 mg/kg over 2 days and 12 Gy fractionated TBI. Six patients received BU and 12 Gy fractionated TBI, and in these six patients the BU dose was adjusted to achieve steady-state targeted levels of 600–900 ng/dI.30 Of the remaining two patients, one received Fludarabine 30 mg/ m2 for 3 days and low dose TBI (200 cGy) as the preparative regimen for a nonmyeloablative transplant,

while the other received Melphalan, TBI and ATG prior to receiving cord blood. Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporine and methotrexate in the majority of patients (Table 2). Unmanipulated bone marrow was used in all patients except one who received marrow and G-CSF-mobilized peripheral blood stem cells. All patients received antibacterial prophylaxis, and some also received antifungal and antiviral prophylaxis according to the standard of practice at the time of transplant. Regimen-related toxicity was scored by the method of Bearman.31 Transplant protocols and consent forms were approved by the Institutional Review Board (IRB) at FHCRC, and informed consent was obtained from parents or responsible guardians according to IRB policies.

Engraftment, relapse and GVHD Engraftment was defined as the first of three consecutive days with an absolute neutrophil count (ANC of 4500/ mm3) and presence of donor markers. Donor engraftment was determined by in situ DNA hybridization with Y or Xbody-specific-probes32 for sex-mismatched transplants, or by restriction fragment length polymorphism (RFLP) analysis,33 or PCR assay of genomic DNA for variable number of tandem repeats (VNTR).34 Platelet recovery was defined as a platelet count of 420 000/mm3 for 7 days without any transfusions. Relapse was identified by recurrence of a prior cytogenetic abnormality or by return of morphologic evidence of disease. Acute and chronic GVHD were diagnosed according to conventional criteria and treated as previously described.28,35,36 Graft failure was defined as lack of donor cell engraftment by 28 days post BMT, or marrow hypoplasia following engraftment and loss of donor cells with or without return of host hematopoiesis.

Follow-up and quality of life Patients were followed yearly by visits to the FHCRC longterm follow-up clinic and/or by detailed questionnaires completed by the patient’s family and primary referring physician. Quality of life was evaluated using the Lansky Play Performance37 or Karnofsky scale.38 Annual evaluation included measurement of growth hormone (GH), thyroxine (T4) and thyroid-stimulating hormone (TSH) and measurements of engraftment.39

Statistics Survival and event-free survival (EFS) estimates were obtained by the method of Kaplan and Meier,40 where patients failing to reach the appropriate end point were censored at date of last contact. Death, relapse and graft rejection were considered events for EFS. Nonrelapse mortality (NRM), relapse and clinical extensive chronic GVHD were summarized using cumulative incidence estimates,41 where relapse was considered a competing risk for NRM, death without relapse a competing risk for relapse and death without chronic GVHD a competing risk for chronic GVHD. Proportional hazards regression models were fit to assess the association of various Bone Marrow Transplantation

Marrow transplant for childhood MDS U Yusuf et al

808

explanatory variables with the hazard of failure appropriate for the end points of survival, EFS, NRM, relapse and chronic GVHD. Logistic regression was used to assess the association of the explanatory variables with the probability of grades II–IV acute GVHD. The variables examined included presence of monosomy 7, diagnosis at transplant, patient/donor gender, donor type (HLA-mismatched related vs matched related vs unrelated), age at transplant, patient/donor CMV serostatus, disease duration prior to transplant, preparative regimen and prior therapy. For purposes of regression modeling for the timeto-event end points, patients failing to reach the appropriate end point were censored at time of last contact or failure from a competing risk, whichever occurred first. The Wald test was used to obtain two-sided P-values from the regression models, and no adjustments were made for multiple comparisons. P-values between 0.01 and 0.05 are therefore considered to be suggestive of a true difference rather than as definitive evidence. Data presented were as of 31 March 2001.

Results

Graft-versus-host disease

Engraftment Of the 94 (97.8%) patients, 92 achieved a sustained ANC of 4 500/mm3 at a median of 21 days (range 11–33). Among these 92, 87 (94%) achieved such a level on or before day 28. In all, 68 of the 94 (72%) patients had a documented sustained platelet count of 420 000/mm3 at a median of 21.5 days (range of 11–119). Of these 68 patients, 90% achieved this count on or before day 45. All 11 patients who failed to reach a platelet count 420 000/mm3 died from 32 to 120 days post transplant.

Second bone marrow transplant Seven patients developed graft failure or graft rejection at a median of 41 (26–95) days after the initial transplant.

Table 3a

Diagnosis

Regimen

Donor

2601 2601 3609 3609 6163 6163 8068 8068

RAEB

CY/12 Gy CY/7.5 Gy CY-ATG-15.75 Gy CY+ATG BU-CY(2) BUTCY(4) CY+14.40 Gy Anti CD3+MP

2 Ag mm sibling 2 Ag mm mother 2 Ag mm mother 2 Ag mm mother 1 Ag mm URD 1 Ag mm URD 1 Ag mm URD 1 Ag mm URD

CY+ATG+14.40 Anti CD3+MP BU-CY(2) CY+12.0 Gy FLU+200TBI BUTCY(2)

2 Ag mm URD 0.13 3 Ag mm sibling Unknown 0 Ag mm mother 1.38 0 Ag mm mother 2.7 Match URD 3.7 Match URD 3.8

9099 #1 9099 #2 9717 #1 9717 #2 15390 #1 15390 #2

The two patients who died without achieving engraftment were not evaluable for development of GVHD. Among the 92 patients evaluable for acute GVHD, 12 (13%) developed grades 0–I, 59 (64%) had grades II–III and 7 (7.7%) had grade IV. Of the 30 (40%) recipients of HLA-matched related marrow developed grade II–IV acute GVHD compared to 22 (76%) of the 29 recipients of matched unrelated marrow, seven (100%) of HLA-mismatched unrelated marrow and 25 (89%) of the 28 of HLAmismatched related marrow. In a multivariable logistic regression model, the odds of developing grades II–IV acute GVHD were higher among unrelated and mismatched related recipients

Second transplant for graft rejection

Graft rejection #1 #2 #1 #2 #1 #2 #1 #2

Details of their first and second transplant are in Table 3a. These patients were transplanted from matched or mismatched unrelated donors (n ¼ 4) or from matched or mismatched family member donors (n ¼ 3). All patients received bone marrow containing a median of 1.7 (0.13– 4.9)  108 nucleated cells/kg recipient weight. Graft rejection occurred at a median of 43 (26–95) days after the first transplant, and the second transplant occurred at a median of 94 (41–224) days after the first transplant. Two patients survive more than 8 and 11 years after second transplant and five patients died of infections (n ¼ 4) or GVHD (n ¼ 1). A total of 12 patients relapsed at a median of 277 (40–1670) days after initial transplant. All patients received TBI-containing preparative regimens and 10 received matched family member or unrelated donor marrow (Table 3b). The marrow cell dose was a median of 4.3 (1.3–8.6)  108/kg recipient weight. Two patients survive more than 15 years after second transplant, two died of venoocclusive disease, one died of fungal infection, one died of chronic GVHD and six died of relapse.

JMML RA JMML

RA RA RA

Cell dose  108/kg

Day of rejection

Day of BMT#2

0.4 4.81 4.9 4.8 2.8 7.1 1.7 1.7

26

94

2.7

42

95

224

43

75

Survival from COD BMT#2 (years) 0.1

CNS Bleed

0.04

Aspergillus

411

27

41

41

125

62

145

0.01

Candida parapsilosis

0.3

GVHD

48 0.2

CMV pneumonia

COD ¼ cause of death; RAEB ¼ refractory anemia with excess blasts; MM ¼ mismatch; JMML ¼ juvenile myelomonocytic leukemia; RA ¼ refractory anemia; URD ¼ unrelated donor; BUT ¼ targeted busulfan to 800–900 ng/dl; Flu ¼ fludarabine; UPN ¼ unique patient number; Ag ¼ antigen; #1 ¼ first transplant; #2 ¼ second transplant.

Bone Marrow Transplantation

Marrow transplant for childhood MDS U Yusuf et al

809 Table 3b UPN

1429 1429 4157 4157 4430 4430 4502 4502 4510 4510 4613 4613 4647 4647 5303 5303 5447 5447 7005 7005 7183 7183 7534 7534

#1 #2 #1 #2 #1 #2 #1 #2 #1 #2 #1 #2 #1 #2 #1 #2 #1 #2 #1 #2 #1 #2 #1 #2

Second transplant for relapsed disease DX

Regimen

Donor

JMML

CY+10.0 Gy BU+CY(4) CY+15.75 Gy BU+CY(4) CY+14.40 TBI BU CY CY-200x6 BU+CY CY+1500 Gy BU+CY (4) BU+CY+12 Gy BU+CY (4) CY+ 14.4 Gy BU+CY (4) CY+14.4 Gy BU+CY (4) CY+12 Gy BU+CY (4) CY+13.2 Gy CY+13.2 Gy CY-ATG-14.40 GY BU+CY (4) CY+14.4 Gy BU+CY(4)

¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ Sib ¼ URD ¼ URD 2 Ag mm father 2 Ag mm father ¼ Sister ¼ Sister 1 Ag mm brother 1 Ag mm brother ¼ father ¼ father ¼ URD ¼ URD

JMML RAEB-T RAEB RA JMML RAEB-T JMML RAEB RAEB-T JMML JMML

 108/kg cell dose

After #1 day relapse

4.4 3.0 8.6 14.9 2.0 1.2 3.48 2.4 Unknown 2.9 3.3 3.0 7.1 4.1 4.6 3.1 1.46 3.86 2.9 Unknown 1.3 3.7 4.3 7.5

1670

After #1 day of BMT #2

COD

Survival post #2 (days) 48

145

221

40 77 277

494 311

310

425

147

158

399

435

108

178

961

1009

Relapse

415 years 182

Relapse

928

Relapse

410 yrs

605 147

720 232

71

395

956

Relapse

164

Relapse

237

C GVHD

75 73

Relapse Infection

33

VOD

33

VOD

¼ refers to HLA identical; UPN ¼ unique patient number; COD ¼ cause of death; BU ¼ busulfan; CY ¼ cytoxan; JMML ¼ juvenile myelomonocytic leukemia; CGVHD ¼ chronic graft versus host disease; VOD ¼ venoocclusive disease; Ag ¼ antigen.

Table 4

Multivariate cox regression models Event-free survival HR

MDS type

Donor type

RA/RARS

1.0

RAEB RAEB-T JMML/ CMML

1.0 2.4 2.3

Matched related Unrelated Mismatched related

1.0

Patient/ MF, FF, donor FM sex MM

1.9 2.1

95% Cl

Non-relapse mortality

P-value HR

95% Cl

P-value

Relapse HR

Acute GVHD

95% Cl

P-value

0.2–5.7 1.4–21.5 0.5–8.8

0.85 0.01 0.33

OR

95% Cl

P-value

Chronic GVHD OR

95% Cl

P-value

1.2–24.2 1.3–29.6

0.03 0.02

1.0 0.4–2.5 1.0–5.4 1.1–5.0

0.99 0.04 0.03

1.2 5.5 2.1

1.0 1.0–3.9 1.0–4.2

0.06 0.05

2.0 0.7–5.9 3.0 1.0–8.6

1.0 0.22 0.04

9.1 2.6–31.5 0.0005 13.5 3.2–56.7 0.0004

1.0 5.3 6.3

1.0

2.6 1.2–5.6

0.01

HR ¼ hazard Ratio; Cl ¼ confidence limit; OR ¼ odds ratio.

compared to recipients of matched related marrow after adjusting for use of TBI-containing regimens. In addition, patients who received TBI as part of their conditioning regimen appeared to have 3.8 times the risk of developing acute GVHD compared to patients who did not receive TBI. Two of 30 (6.7%) patients who received marrow from a matched related donor

developed clinical extensive chronic GVHD compared to 8 of 28 (28.6%) recipients of mismatched related marrow and 11 of 36 (30.6%) recipients of unrelated marrow. Type of donor was the only variable that was statistically significantly associated with the hazard of developing clinical extensive chronic GVHD, as summarized in Table 4. Bone Marrow Transplantation

Marrow transplant for childhood MDS U Yusuf et al

810

Relapse Relapse was detected in 28 patients between 27 and 1670 (median 139) days after first transplant, resulting in a cumulative incidence estimate of 30% at 5 years (Figure 1). There were three relapses among 27 (11%) patients with RA/RARS compared to five of the 20 (25%) patients with RAEB, seven of the 14 (50%) patients with RAEB-T and 13 of 33 (39%) patients with JMML/CMML. In a multivariable regression model, the hazard of relapse among patients with RAEB-T was 5.5 times higher compared to patients with RA (95% CI, 1.4 to 21.5; P ¼ 0.01). In addition, increasing patient age was associated with a decreased hazard of relapse (with age modeled as a continuous variable, P ¼ 0.02). Among the 69 patients who were relapse-free 80 days post transplant and at risk for developing chronic GVHD, 15/41 (37%) with chronic GVHD and 13/28 (46%) without chronic GVHD relapsed. One patient who had a cytogenetic relapse had a rapid taper of immunosuppressive therapy resulting in subsequent cytogenetic remission with more than 7 years of follow-up. Another patient with RAEB and monosomy 7 was noted to have one metaphase showing monosomy 7 among 50 analyzed at day 80 after BMT. Cyclosporine was then discontinued and a cytogenetic remission was achieved until 2.6 years after transplant when frank relapse occurred. At the time of relapse, monosomy 7 and other cytogenetic abnormalities in host cells were present.

of death in these patients are summarized in Table 5. In a multivariable regression model, patients transplanted from mismatched related donors had a hazard of failure 3.0 times greater than patients who received a matched related transplant (95% CI, 1.0 to 8.6; P ¼ 0.04). In addition, male patients transplanted from male donors were 2.6 times as likely to have NRM compared to all other patient/donor gender combinations (HR ¼ 2.6; 95% CI, 1.2 to 5.6; P ¼ 0.01). With these variables in the regression model for NRM, no other variables examined significantly improved the model.

EFS and overall survival The estimated EFS among all patients at 3 years was 41%. The estimated EFS for patients with RA/RARS at 3 years was 59% compared to 58, 18 and 27% for patients with RAEB, RAEB-T and JMML/CMML, respectively (Figure 2). The hazard of failure in a multivariable regression model was higher among patients with RAEBT and JMML/CMML compared to patients with RA/RARS (HR ¼ 2.4; 95% CI, 1.0–5.4; P ¼ 0.04 and HR ¼ 2.3; 95% CI 1.1–5.0; P ¼ 0.03, respectively). In addition, there was a suggestion that the hazard of failure was higher among recipients of unrelated or mismatched related marrow compared to patients transplanted from a matched related donor (HR ¼ 1.9; 95% CI, 1.0–3.9; P ¼ 0.06 and HR ¼ 2.1 Primary causes of death post transplantation

Table 5

NRM

Cause

In total, 27 patients died without a prior relapse between 6 and 4082 days (median 80 days) post transplant. The 5-year cumulative incidence estimate of NRM is 28% (Figure 1). Infections were the most common cause of NRM with seven patients dying of fungal infections, four bacterial infections and two from CMV pneumonitis. Four patients died of acute GVHD complications, while four died of complications related to chronic GVHD. Regimen-related toxicity was the cause of death in four patients and two patients died secondary to graft failure. The primary causes

Relapse Graft failure Regimen-related toxicity Acute GVHD Chronic GVHD

18 2 7 4 4

Infection Bacterial Fungal Viral Intracranial hemorrhage

5 7 3 1

1.0

1.0

Overall Survival (n=94) Event free Survival Non relapse Mortality Relapse Censored

0.6 0.4

JMML/CMML (n=33) RA/RARS (n=27) RAEB (n=19) RAEBT (n=15) Censored

0.8 Probability

0.8 Probability

No. of patients

0.6

0.4

0.2

0.2

0.0

0.0 0 Figure 1

3

6 9 12 Years after Transplant

15

Probability of survival, EFS, NRM and relapse after transplant in 94 children with myelodysplasia.

Bone Marrow Transplantation

0

3

6 9 Years after Transplant

12

15

Figure 2 Kaplan–Meier estimates of EFS of patients with RA/RARS, RAEB, RAEB-T and JMML/CMML.

Marrow transplant for childhood MDS U Yusuf et al

811

95% CI, 1.0–4.2; P ¼ 0.05, respectively (Table 4, Figure 3)). Use of a TBI-based conditioning regimen did not significantly improve this regression model (P ¼ 0.37). The estimated overall survival among all patients was 50% at 3 years and 48% at 5 years (Figure 1). Estimated survival by diagnosis at 3 years was 74, 68, 18 and 33% for patients with RA/RARS, RAEB, RAEB-T and JMML/ CMML, respectively. Compared to patients with RA/ RARS, the hazard of death was 3.9 times higher among patients with RAEB-T (95% CI, 1.5–9.9; P ¼ 0.005) and 3.7 times higher among patients with JMML/CMML (95% CI, 1.5–8.9; P ¼ 0.004) in a multivariable regression model (Table 4). In addition, patients transplanted from an unrelated donor or mismatched related donor had a higher hazard of death compared to recipients of matched related marrow. No other variables statistically significantly improved the regression model summarized in Table 4.

Effect of the presence of monosomy 7 on outcome Monosomy 7 was present in 30 (32%) of the 94 patients and its presence was correlated with diagnosis. This cytogenetic abnormality was seen in 11/27 (41%) of patients with RA/RARS, 12/19 (63%) with RAEB, 6/15 (40%) with RAEB-T and 1/33 (3%) with JMML/CMML. Relapse occurred in eight (27%) of the 30 patients with monosomy 7 compared to 20 (31%) of 64 with no monosomy 7 (Table 6). In a multivariate regression model, there was no statistically significant difference in overall

1.0

Matched Related (n=30) Mismatched Related (n=28) Unrelated (n=36) Censored

Probability

0.8 0.6 0.4 0.2 0.0 0

3

6

9

12

15

Years after Transplant Figure 3 Kaplan–Meier estimates of survival in patients receiving a transplant from a matched related, mismatched related or unrelated donors.

Table 6

survival (P ¼ 0.91), EFS (P ¼ 0.54) and relapse (P ¼ 0.62) between patients with or without monosomy 7.

Quality of life A total of 43 patients are alive with a median follow-up of 10 years, and 41 of these had a performance score available. Among these, 35 had a performance score of 100%, three a score of 90%, one a score of 70% and two a score of 60%. The three patients with performance scores of less than 90% have ongoing complications related to chronic GVHD. Information regarding growth was available for 37 surviving patients. Of the 37 patients, 11 developed growth hormone deficiency 0.9–4 years post BMT and have received treatment with growth hormone replacement. All 11 patients received TBI as part of their preparative regimen, for the first (n ¼ 10) or second transplant (n ¼ 1). Nine patients developed thyroid hormone deficiency from 0.9 to 5.2 years post BMT and all are receiving therapy. Six of these nine patients received a TBI-containing regimen, two received BU/CY and, one had two consecutive transplants with BU/CY.

Discussion Allogeneic BMT remains the only curative treatment for MDS. The estimated overall survival at 3 years of 50% and EFS of 41% observed in the current series of 94 children with MDS are comparable to those in previously reported smaller series of children transplanted for MDS.5,17,18,20,21,29,42–44 Smith et al17 recently reported 42% survival and 24% DFS in 46 children with JMML who underwent unrelated donor BMT with a follow up of 2 years. Guinan et al 20 in 1989 reported long-term survival for four of eight children with MDS receiving marrow transplants from matched-sibling donors. Locatelli et al 21 reported 58% EFS in 33 children with MDS receiving allogeneic transplant, but this report included 10 patients with AML secondary to MDS. Davies et al 42 and Casper et al 43 reported long-term survival in 11 out of 19 children with MDS transplanted from an URD.42,43 In this analysis, patients with RA/RARS had an estimated 3-year survival of 74 vs 18% in the RAEB-T group. Of interest, children with the RAEB subtype had comparable survival and EFS to that of patients with RA/RARS. The poor outcome for patients with RAEB-T and JMML/CMML suggests that different approaches are needed for these patients, which may include a more intensive preparative regimen or post BMT immune-modulation. Recent results in adult patients

Effect of monosomy 7 on relapse and death

Diagnosis

Relapse Monosomy 7

RA/RARS (n ¼ 27) RAEB (n ¼ 19) RAEB-T (n ¼ 15) JMML/CMML (n ¼ 33)

2/11 (18%) 3/12 (25%) 3/6 (50%) 0/1

Death or relapse No monosomy 7 1/16 2/7 4/9 13/32

(6%) (29) (44%) (41%)

Monosomy 7 5/11 6/12 4/6 1/1

(45%) (50%) (67%) (100%)

No monosomy 7 4/16 3/7 8/9 24/32

(25%) (43%) (89%) (75%)

Bone Marrow Transplantation

Marrow transplant for childhood MDS U Yusuf et al

812

with advanced leukemia utilizing targeted radiolabeledantibody therapy in addition to standard conditioning is showing encouraging results and might be considered for children with advanced MDS.45 Previous analyses including adult and pediatric patients have shown that age of less than 20 years and shorter disease duration had a favorable impact on outcome.19,46 In the current analysis, neither disease duration nor age had a demonstrable influence on EFS or survival. The presence of monosomy 7 in childhood AML is associated with poor prognosis.47 The effect of monosomy 7 on outcome following BMT in children with MDS has not been previously described. The current analysis did not show a statistically significant association between the presence of monosomy 7 and any outcome. Whether this lack of statistical significance is real or due to lack of power needs to be further studied before a definitive conclusion can be made. Prior reports of outcome following allogeneic transplantation for adult MDS have demonstrated increased relapse rates for patients with RAEB, RAEB-T and CMML.19,21,44 Others have shown fewer relapses among patients with longer disease duration before transplant.24 In the present series, patients with RAEB-T had a statistically significant higher relapse rate compared to patients with RA/RARS. Relapse was the major contributor to overall mortality in this report. Development of methods to detect minimal residual disease to allow identification of those patients who are at high risk of relapse and early initiation of immune-modulation to facilitate a graft-versus-leukemia reaction may be useful.48 Donor type is often suggested as a major factor influencing outcome following marrow transplantation. Previous reports have demonstrated that transplantation using URD for MDS did not influence overall survival.24 Davies et al42 and Casper et al43 reported on the use of unrelated donor transplantation for children with MDS. In the current analysis, the use of URD was associated with a higher incidence of acute and chronic GVHD and, in general, poorer overall outcome compared to transplant from a matched related donor. Outcome for URD recipients, however, appeared to be similar to that observed among mismatched related transplants. Another factor associated with an increased risk of acute GVHD, but not chronic GVHD was the use of TBI. Although the reasons for this observation are not clear, the basic biology of the pathogenesis of GVHD must be considered. A multistep model of antigen expression, cytokine production, T-cell activation and tissue injury has been proposed as the interactive events leading to GVHD. Tissue injury from infection, chemotherapy and/or TBI results in the release of interleukins, tumor necrosis factor and/or interferon that results in the clinical manifestations of GVHD.49,50 The use of TBI may result in greater tissue injury than chemotherapy with the subsequent observation of an increased incidence of acute GVHD. The high rate of acute and chronic GVHD noted in the recipients of URD transplant in this analysis might be due in part to the fact that only serological typing for HLA class I and II was performed prior to 1990. Recent reports Bone Marrow Transplantation

indicate that the use of DNA typing for HLA class I, class II and H antigens can decrease the incidence of acute GVHD and improve outcome.51 Infections were a major cause of NRM in our patients, most likely due to immunosuppression from GVHD and its therapy, and also a fairly large number of patients had received a second transplant. This analysis demonstrated a 30% DFS in patients who received a second bone marrow transplant. This is comparable to the report by Kishi et al52 who identified early relapse (o6 months), second BMT less than 12 months from the first and not being in remission at second BMT as poor prognostic factors in patients receiving second BMT for post transplant leukemia relapse. Early relapse was a feature in 50% of the patients who relapsed and 86% of the deaths occurred in those patients who had received the second transplant within a year of the first BMT. Overall, quality of life for most patients is excellent with all but three of 41 surviving patients having a performance score of greater than or equal to 90%. These data demonstrate that children with MDS may achieve encouraging overall survival and EFS following allogeneic BMT. However, methods to further improve outcome in patients with MDS or JMML undergoing allogeneic transplantation are needed. In particular, innovative measures to decrease the relapse rate among patients with advanced MDS and JMML would be likely to improve EFS. In addition, measures to prevent and treat GVHD more effectively would probably improve quality of life in recipients of mismatched or unrelated donor marrow. We recommend that children with MDS or JMML undergo allogeneic BMT as soon as a donor is identified, prior to progression of their disease.

Acknowledgements We are indebted to Mr Chris Olson and Ms Meg Bender for their valuable assistance in the preparation of this manuscript. This study was supported in part by National Institutes of Health Grants No. HL 36444, CA 15704, CA 18029, CA 18221 and CA 47748.

References 1 Kobrinsky NIL, Nesbit Jr MIE, Ramsay NK et al. Hematopoietic dysplasia and marrow hypocellularity in children: a preleukemic condition. J Pediatr 1982; 10: 907–913. 2 Gadner H, Haas OA. Experience in pediatric myelodysplastic syndromes. Hematol Oncol Clin North Am 1992; 6: 655–672. 3 Passmore SJ, Hann IM, Stiller CA et al. Pediatric myelodysplasia: a study of 68 children and a new prognostic scoring system. Blood 1995; 85: 1742–1750. 4 Bennett JM, Catovsky D, Daniel MT et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982; 51: 189–199. 5 Woods WG, Barnard DR, Alonzo TA et al. Prospective study of 90 children requiring treatment for juvenile myelomonocytic leukemia or myelodysplastic syndrome: a report from the children’s cancer group. J Clin Oncol 2002; 20: 434–440.

Marrow transplant for childhood MDS U Yusuf et al

813 6 Tuncer MA, Pagliuca A, Hicsonmez G et al. Primary myelodysplastic syndrome in children: the clinical experience in 33 cases. Br J Haematol 1992; 82: 347–353. 7 Creutzig U, Cantu-Rajnoldi A, Ritter J et al. Myelodysplastic syndromes in childhood. Report of 21 patients from Italy and West Germany. Am J Pediatr Hematol Oncol 1987; 9: 324–330. 8 Hasle H. Myelodysplastic syndromes in childhood – classification, epidemiology, and treatment. Leuk Lymphoma 1994; 13: 11–26. 9 Wegelius R. Preleukaemic states in children. Scan J Haematol Supple 1986; 45: 133–139. 10 Martinez-Climent JA. Molecular cytogenetics of childhood hematological malignancies. Leukemia 1997; 11: 1999–2021. 11 Nowell PC. Cytogenetics of preleukemia. Cancer Genet Cytogenet 1982; 5: 265–278. 12 Luna-Fineman S, Shannon KM, Lange BJ. Childhood monosomy 7: epidemiology, biology, and mechanistic implications. Blood 1995; 85: 1985–1999. 13 Evans JP, Czepulkowski B, Gibbons B et al. Childhood monosomy 7 revisited. Br J Haematol 1988; 69: 41–45. 14 Sieff CA, Chessells JM, Harvey BA et al. Monosomy 7 in childhood: a myeloproliferative disorder. Br J Haematol 1981; 49: 235–249. 15 Festa RS, Shende A, Lanzkowsky P. Juvenile chronic myelocytic leukemia: experience with intensive combination chemotherapy. Med Pediatr Oncol 1990; 18: 311–316. 16 Hasle H, Kerndrup G, Yssing M et al. Intensive chemotherapy in childhood myelodysplastic syndrome. A comparison with results in acute myeloid leukemia. Leukemia 1996; 10: 1269– 1273. 17 Smith FO, King R, Nelson G et al. Unrelated donor bone marrow transplantation for children with juvenile myelomonocytic leukemia. Br J Haematology 2002; 116: 716–724. 18 Sanders JE, Buckner CD, Thomas ED et al. Allogeneic marrow transplantation for children with juvenile chronic myelogenous leukemia. Blood 1988; 71: 1144–1146. 19 Anderson JE, Appelbaum FR, Fisher LD et al. Allogeneic bone marrow transplantation for 93 patients with myelodysplastic syndrome. Blood 1993; 82: 677–681. 20 Guinan EC, Tarbell NJ, Tantravahi R, Weinstein HJ. Bone marrow transplantation for children with myelodysplastic syndromes. Blood 1989; 73: 619–622. 21 Locatelli F, Zecca M, Niemeyer C et al. Role of allogeneic bone marrow transplantation for the treatment of myelodysplastic syndromes in childhood. The European Working Group on Childhood Myelodysplastic Syndrome (EWOGMDS) and the Austria-Germany-Italy (AGI) Bone Marrow Transplantation Registry. Bone Marrow Transplant 1996; 18 (Suppl 2): 63–68. 22 Anderson JE, Appelbaum FR, Storb R. An update on allogeneic marrow transplantation for myelodysplastic syndrome. Leuk Lymphoma 1995; 17: 95–99. 23 Appelbaum FR, Barrall J, Storb R et al. Bone marrow transplantation for patients with myelodysplasia. Pretreatment variables and outcome. Ann Intern Med 1990; 112: 590–597. 24 Anderson JE, Anasetti C, Appelbaum FR et al. Unrelated donor marrow transplantation for myelodysplasia (MDS) and MDS-related acute myeloid leukaemia. Br J Haematol 1996; 93: 59–67. 25 Emmanuel PD. Myelodysplastic syndrome and myeloproliferative disease in childhood: an update. Br J Haematol 1999; 105: 852–863. 26 Dupont B, Yang S. Histocompatibility. In: Anonymous (ed.) Blackwell Scientific Publications: Boston, MA, 1994. 27 Petersdorf EW, Smith AG, Mickelson EM et al. Ten HLADR4 alleles defined by sequence polymorphisms within the DRB I first domain. Immunogenetics 1991; 33: 267–275.

28 Thomas ED, Storb R, Clift RA et al. Bone-marrow transplantation (second of two parts). N Engl J Med 1975; 292: 895–902. 29 Johnson FL, Sanders JE, Ruggiero M et al. Bone marrow transplantation for the treatment of acute nonlymphoblastic leukemia in children aged less than 2 years. Blood 1988; 71: 1277–1280. 30 Slattery J, Sanders J, Buckner CD, Schaffer RL. Graftrejection and toxicity following bone marrow transplantation in relation to busulfan pharmacokinetics. Bone Marrow Transplant 1995; 16: 31–42. 31 Bearman SI, Appelbaum FR, Buckner CD et al. Regimenrelated toxicity in patients undergoing bone marrow transplantation. J Clin Oncol 1988; 6: 1562–1568. 32 Durnarn DM, Anders KR, Fisher L et al. Analysis of the origin of marrow cells in bone marrow transplant recipients using a Y-chromosome-specific in situ hybridization assay. Blood 1989; 74: 2220–2226. 33 Blazar BR, Orr ITF, Arthur DC et al. Restriction fragment length polymorphisms as markers of engraftment in allogeneic marrow transplantation. Blood 1985; 66: 1436–1444. 34 Ugozzoli L, Yam P, Petz LD et al. Amplification by the polymerase chain reaction of hypervariable regions of the human genome for evaluation of chimerism after bone marrow transplantation. Blood 1991; 77: 1607–1615. 35 Doney KC, Weiden PL, Storb R, Thomas ED. Treatment of graft-versus-host disease in human allogeneic marrow graft recipients: a randomized trial comparing antithymocyte globulin and corticosteroids. Am J Hematol 1981; 11: 1–8. 36 Sullivan KM, Shulman HM, Storb R et al. Chronic graftversus-host disease in 52 patients: adverse natural course and successful treatment with combination immunosuppression. Blood 1981; 57: 267–276. 37 Lansky SB, List MA, Lansky LL et al. The measurement of performance in childhood cancer patients. Cancer 1987; 60: 1651–1656. 38 Karnofsky D, Burchenal J. The clinical evaluation of chemotherapeutic agents in cancer. In: Anonymous (ed.) The Clinical Evaluation of Chemotherapeutic Agents in Cancer. Columbia University Press: New York, 1949. 39 Sanders JE. Growth and development after hematopoietic cell transplantation. In: Thomas ED, Blume KG, Forman SJ (eds). Hematopoietic cell transplantation, Chapter 65. pp 764–775. 40 Kaplan E, Meier P. Non parametric estimation from incomplete observations. Am Stat Assoc 1957; 53: 457–481. 41 Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999; 18: 695–706. 42 Davies SM, Wagner JE, Defor T et al. Unrelated donor bone marrow transplantation for children and adolescents with aplastic anemia or myelodysplasia. Br J Haematol 1997; 96: 749–756. 43 Casper J, Camitta B, Truitt R et al. Unrelated bone marrow donor transplants for children with leukemia or myelodysplasia. Blood 1995; 85: 2354–2363. 44 de Witte T, Zwaan F, Hermans J et al. Allogeneic bone marrow transplantation for secondary leukaemia and myelodysplastic syndrome: a survey by the Leukaemia Working Party of the European Bone Marrow Transplantation Group (EBMTG). Br J Haematol 1990; 74: 151–155. 45 Matthews DC, Appelbaum FR, Eary JF et al. Development of a marrow transplant regimen for acute leukemia using targeted hematopoietic irradiation delivered by 13 1 I-labeled antiCD45 antibody, combined with cyclophosphamide and total body irradiation. Blood 1995; 85: 1122–1131. 46 Runde V, de Witte T, Arnold R et al. Bone marrow transplantation from HLA-identical siblings as first-line Bone Marrow Transplantation

Marrow transplant for childhood MDS U Yusuf et al

814 treatment in patients with myelodysplastic syndromes: early transplantation is associated with improved outcome. Chronic Leukemia Working Party of the European Group for Blood and Marrow Transplantation. Bone Marrow Transplant 1998; 21: 255–261. 47 Woods WG, Nesbitt ME, Buckley JD et al. Correlation of chromosomal abnormalities with characteristics, histologic subtype and induction success in children with acute nonlymphoblastic leukemia (ANLL). J Clin Oncol. 1985; 3: 3–11. 48 Sievers EL, Lange BJ, Buckley JD et al. Prediction of relapse of pediatric acute myeloid leukemia by use of multidimensional flow cytometry. J Natl Cancer Inst 1996; 88: 1483–1488.

Bone Marrow Transplantation

49 Antin JH, Ferrara JLN. Cytokine dysregulation and acute graft-versus-host disease. Blood 1992; 80: 2964–2968. 50 Vogelsang GB, Hess AD. Graft-versus-host disease: new directions for a persistent problem. Blood 1994; 84: 2061–2067. 51 Petersdorf EW, Gooley TA, Anasetti C et al. Optimizing outcome after unrelated marrow transplantation by comprehensive matching of HLA class I and 11 alleles in the donor and recipient. Blood 1998; 92: 3515–3520. 52 Kishi K, Takahashi S, Gondo H et al. Second allogeneic bone marrow transplantation for post transplant leukemia relapse: results of a survey of 66 cases in 24 Japanese institutes. Bone Marrow Transplant 1997; 19: 461–466.