Hematopoietic stem cell transplantation for advanced ... - Nature

Leukemia (2011) 25, 455–462 & 2011 Macmillan Publishers Limited All rights reserved 0887-6924/11 www.nature.com/leu

ORIGINAL ARTICLE Hematopoietic stem cell transplantation for advanced myelodysplastic syndrome in children: results of the EWOG-MDS 98 study B Strahm1, P No¨llke1, M Zecca2, ET Korthof3, M Bierings4, I Furlan1, P Sedlacek5, A Chybicka6, M Schmugge7, V Bordon8, C Peters9, A O’Marcaigh10, CD de Heredia11, E Bergstraesser7, BD Moerloose8, MM van den Heuvel-Eibrink12,13, J Stary´5, M Trebo9, D Wojcik6, CM Niemeyer1 and F Locatelli14 for the EWOG-MDS study group 1 Department of Pediatrics and Adolescent Medicine, Division of Pediatric Hematology and Oncology, University of Freiburg, Freiburg, Germany; 2Department of Pediatrics, Fondazione, IRCCS Policlinico, San Matteo, Italy; 3Department of Pediatrics, Division of Stem Cell Transplantation, Leiden University Medical Center, Leiden, The Netherlands; 4Department of Hematology, University Hospital for Children, Utrecht, The Netherlands; 5Department of Pediatric Hematology and Oncology, University Hospital Motol, Prague, Czech Republic; 6Department of Pediatric Bone Marrow Transplantation, Oncology and Hematology, Medical University, Wroclaw, Poland; 7Department of Hematology and Oncology, University Children’s Hospital, Zurich, Switzerland; 8Department of Pediatric Hematology and Oncology, Ghent University Hospital, Ghent, Belgium; 9Department of Pediatrics, St Anna Children’s Hospital, Vienna, Austria; 10Department of Oncology and Hematology, Our Lady’s Hospital for Sick Children, Dublin, Ireland; 11Department of Pediatric Oncology and Hematology, Hospital Vall d’Hebron, Barcelona, Spain; 12 Erasmus Medical Center, Rotterdam, The Netherlands; 13Dutch Childhood Oncology Group, The Hague, The Netherlands and 14 Department of Pediatric Hematology and Oncology, IRCCS Ospedale Bambino Gesu`, University of Pavia, Rome, Italy

We report on the outcome of children with advanced primary myelodysplastic syndrome (MDS) transplanted from an HLAmatched sibling (MSD) or an unrelated donor (UD) following a preparative regimen with busulfan, cyclophosphamide and melphalan. Ninety-seven patients with refractory anemia with excess blasts (RAEB, n ¼ 53), RAEB in transformation (RAEB-T, n ¼ 29) and myelodysplasia-related acute myeloid leukemia (MDR-AML, n ¼ 15) enrolled in the European Working Group of MDS in Childhood (EWOG-MDS) 98 study and given hematopoietic stem cell transplantation (HSCT) were analyzed. Median age at HSCT was 11.1 years (range 1.4–19.0). Thirty-nine children were transplanted from an MSD, whereas 58 were given the allograft from a UD (n ¼ 57) or alternative family donor (n ¼ 1). Stem cell source was bone marrow (n ¼ 69) or peripheral blood (n ¼ 28). With a median follow-up of 3.9 years (range 0.1–10.9), the 5-year probability of overall survival is 63%, while the 5-year cumulative incidence of transplantation-related mortality (TRM) and relapse is 21% each. Age at HSCT greater than 12 years, interval between diagnosis and HSCT longer than 4 months, and occurrence of acute or extensive chronic graft-versus-host disease were associated with increased TRM. The risk of relapse increased with more advanced disease. This study indicates that HSCT following a myeloablative preparative regimen offers a high probability of survival for children with advanced MDS. Leukemia (2011) 25, 455–462; doi:10.1038/leu.2010.297; published online 7 January 2011 Keywords: myelodysplastic syndrome; stem cell transplantation; children

Introduction Accounting for 4% of all pediatric hematopoietic neoplasia, myelodysplastic syndromes (MDSs) are far less common in children than in adults.1–3 There is a great heterogeneity in Correspondence: Dr B Strahm, Department of Pediatrics and Adolescent Medicine, Division of Pediatric Hematology and Oncology, University of Freiburg, Mathildenstrasse 1, Freiburg 79106, Germany. E-mail: [email protected] Received 28 June 2010; revised 13 October 2010; accepted 15 November 2010; published online 7 January 2011

presentation and clinical course. Refractory anemia with excess blasts (RAEB) in children has features similar to those observed in adults. Children with RAEB generally have relatively stable counts for weeks or months. A similar biological behavior without clinical features of leukemia is noted in some children with myelodysplasia and blast percentage between 20 and 29%, a condition classified as RAEB in transformation (RAEB-T) in the French–American–British (FAB) Cooperative Group classification. Because children with RAEB-T may not benefit from intensive chemotherapy before hematopoietic stem cell transplantation (HSCT), the European Working Group of MDS in Childhood (EWOG-MDS) has maintained their inclusion in pediatric MDS.4,5 The treatment of patients with RAEB, RAEB-T and acute myeloid leukemia (AML) evolving from MDS (MDR-AML) remains a major challenge. Therapy of these disorders has been associated with intensive treatment-related toxicities and a high risk of relapse. Conventional AML-type chemotherapy without HSCT resulted in survival rates below 30%.6,7 In contrast, it has been shown that a large proportion of children with advanced MDS can be successfully treated with HSCT.7–11 However, all previously published studies include patients with a wide spectrum of diagnoses, ranging from refractory cytopenia to more advanced MDS and juvenile myelomonocytic leukemia, transplanted following variable conditioning regimens. The use of intensive chemotherapy before HSCT, aimed at decreasing the risk of relapse, is controversial, and none of the previously published studies has been able to show a robust benefit. Here, we report on the largest series of children with advanced primary MDS transplanted following a uniform preparative regimen including busulfan, cyclophosphamide and melphalan.

Patients and methods Ninety-seven consecutive patients who were enrolled in the observational study EWOG-MDS 98 (Clinical Trials.gov Identifier: NCT00047268), had a diagnosis of advanced primary MDS before January 2007 and were transplanted according to the recommendations of the EWOG-MDS Study group between January 1998 and May 2007 were included in the analysis.

HSCT for myelodysplastic syndrome in children B Strahm et al

456 Peripheral blood smears, bone marrow aspirates, bone marrow biopsies (if available) and metaphase cytogenetics were assessed in local and national reference laboratories according to published criteria.12 Patients with moderate to severe myelofibrosis were excluded. HSCT was performed in 30 European centers (Supplementary Data). Data were reported on standardized data collection forms to the Coordinating Study Centre of the EWOG-MDS Study Group. Informed consent was obtained from patients’ parents or legal guardians according to the Declaration of Helsinki. The study was approved by the Institutional Review Board of each participating institution.

Patient characteristics Median age at diagnosis of the 97 children (65 males and 32 females) was 10.7 years (range 1.0–18.2). According to the most advanced morphological subtype before HSCT, patients were classified as RAEB (n ¼ 53), RAEB-T (n ¼ 29) or MDR-AML (n ¼ 15). Cytogenetic analysis at diagnosis was available in 94 patients; 32 of them had a normal karyotype and 62 had chromosomal aberrations including monosomy 7 (defined as monosomy 7 þ / additional aberrations with the exception of structurally complex karyotypes, n ¼ 29), a structurally complex karyotype (defined as more than three chromosomal aberrations in the presence of at least one structural aberration, n ¼ 5) or other abnormalities (n ¼ 28).

Transplantation procedure Donor and HLA typing. Thirty-nine patients (40%) were grafted from an HLA-matched sibling donor (MSD), whereas the remaining were transplanted from a UD (n ¼ 57, 59%) or an alternative matched family donor (n ¼ 1, 1%). This latter patient was analyzed within the UD group. On the basis of intermediate resolution typing for HLA-A, -B and -C and high-resolution typing for HLA DRB1, unrelated donor–recipient pairs were either identical (UD-8/8; n ¼ 31) or mismatched in one antigen (UD-7/8; n ¼ 10). In 17 donor–recipient pairs, HLA typing did not include HLA-C or was based on serological methods only (UD-other).

Pre-transplant therapy, preparative regimen and prophylaxis of graft-versus-host disease (GVHD). Twentyfour patients (25%) received AML-like chemotherapy before HSCT, whereas the majority of patients (n ¼ 73, 75%) were transplanted without having received intensive chemotherapy. The preparative regimen was based on the use of busulfan (16 mg/kg given orally over four consecutive days), cyclophosphamide (60 mg/kg/day for two consecutive days) and melphalan (140 mg/m2 in a single dose). More recently, 24 patients received busulfan intravenously given according to the recommendations of the pharmaceutical company. To prevent occurrence of GVHD, most patients (27/39, 69%) transplanted from an MSD received cyclosporine-A alone. The combination of cyclosporine-A, short-term methotrexate and either anti-thymocyte globulin (ATG) or, in four cases, the monoclonal antibody Campath1-H was employed in the majority of patients (48/58, 83%) transplanted from a UD.

reported criteria.15 Treatment of GVHD was administered according to protocols in use at each single institution. Myeloid engraftment was defined as the first of three consecutive days with a neutrophil count 40.5 109/l, and platelet engraftment as the first of three consecutive days with an unsupported platelet count 420 109/l. For statistical analysis, the reference date was set to 1 April 2009. Overall survival was defined as the probability of survival irrespective of disease state at any point in time, event-free survival (EFS) as the probability of being alive and free of disease at any time point, considering death, relapse, rejection and graft failure as events. The Kaplan–Meier method was used to estimate survival rates, whereas the two-sided log-rank test was employed to evaluate the equality of the survivorship functions in different subgroups.16 Relapse incidence (RI) was defined as the probability of having a relapse; death without experiencing a relapse was considered a competing event. On the contrary, transplantationrelated mortality (TRM) was defined as the probability of dying without previous occurrence of a relapse, which was the competing event. Both these probabilities were estimated as cumulative incidence curves.17–19 Also, the probabilities of acute and chronic GVHD were estimated as cumulative incidence. For both aGVHD and chronic GVHD, relapse, death, rejection and graft failure were treated as competing events. All results were expressed as 5-year probability or 5-year cumulative incidence (%) and 95% confidence interval (95% CI). Univariate analyses of EFS, RI and TRM were performed considering potential risk factors and patient-related variables, such as age at diagnosis, gender and year of diagnosis. For multivariate analysis, a Cox proportional hazard regression model was designed for each outcome (that is, EFS, RI and TRM). Generally, demographic data (gender, age) and variables with a P-value o0.10 in univariate analysis for at least one of the outcomes were included in each Cox model.20 However, the variable karyotype was not included as the structural complex karyotype occurred in five patients only. Furthermore, the blast count at HSCT was excluded because there was a high correlation with the morphological MDS subtype, and the information was not available for all patients. w2-test was used to examine the statistical significance of an association between categorized variables. In the case of a 2 2 contingency table, Fisher’s exact test was calculated. Because normal distribution cannot be assumed, median values and ranges were reported and nonparametric statistics were used to test for differences in continuous variables for different subgroups (Kruskal–Wallis test with adjacent post-hoc Mann–Whitney U-Test).21,22 All P-values were two-sided and values lower than 0.05 were considered statistically significant. Non-significant P-values greater than 0.15 were reported as NS, while those between 0.05 and 0.15 were reported in detail. SPSS for Windows 15.0.1 (SPSS Inc., Chicago, IL, USA) and NCSS 2004 (Number Cruncher Statistical Systems, Kaysville, UT, USA) were used for the statistical analysis of data.

Results

Definitions and statistics. Acute GVHD (aGVHD) was diagnosed and graded by investigators at each transplant center according to previously reported criteria.13,14 Patients surviving more than 10 days after HSCT were considered at risk for aGVHD. Children alive at day þ 100 post-transplantation with sustained donor engraftment were considered to be evaluable for chronic GVHD, which was classified according to previously Leukemia

Engraftment Details of myeloid recovery were available for all patients. Sustained neutrophil engraftment was achieved in all but two patients who died of TRM (day 11 and 15) before engraftment. The median time to neutrophil engraftment was 16 days (range 10–43). For patients achieving platelet engraftment (87/97), the


457 median time to obtain platelet recovery was 23 days (range 7–148). Nine patients died before platelet engraftment at a median time of 58 days (range 14–118) after HSCT.

The majority of patients who experienced TRM in the absence of GVHD had developed liver veno-occlusive disease (n ¼ 4) or pulmonary toxicity (n ¼ 3).

GVHD

Relapse incidence

Grade II–IV aGVHD developed in 44 patients. The cumulative incidence of grade II–IV aGVHD at day 100 was 46% (37–57), while that of grade III–IV aGVHD was 24% (17–34). Patients given HSCT from an MSD had a significantly higher incidence of grade II–IV and grade III–IV aGVHD than those transplanted from a UD (57 and 33% vs 39 and 18%, respectively, P ¼ 0.02 and P ¼ 0.04) (Figure 1). Within the group of patients grafted from an MSD, the intensity of GVHD prophylaxis (cyclosporineA vs cyclosporine-A/methotrexate) significantly affected the risk of developing grade II–IV aGVHD (74 vs 17%, Po0.01) and grade III–IV (44 vs 8%, P ¼ 0.04). Thirty patients out of the 85 at risk (35%) developed chronic GVHD, which was limited in 18 cases and extensive in the remaining 12 patients. The overall cumulative incidence of chronic GVHD was 37% (28–49). Patients given HSCT from an MSD had a cumulative incidence of chronic GVHD similar to that of patients transplanted from a UD (43 vs 33%, respectively, P ¼ NS). Grade II–IV aGVHD and age at HSCT X12 years were the only factors significantly associated with an increased risk of developing chronic GVHD.

Eighteen patients relapsed at a median time of 10 months (0.5–54) after HSCT, resulting in a 5-year cumulative incidence of relapse of 21% (14–32) (Figure 2). Thirteen of these patients died because of progression of their disease; five patients were rescued with a second allograft and are alive and free of disease with a median follow-up of 3.0 years (0.9–5.0). In univariate analysis, the most advanced morphological subtype of MDS before HSCT and a structurally complex karyotype were the only significant risk factors for disease recurrence (Table 1). The use of intensive chemotherapy before transplantation and the presence of a lower blast count at HSCT were not associated with a significantly decreased risk of relapse. In multivariate analysis, the most advanced morphological subtype type before HSCT remained a significant indicator for an increased risk of relapse; also, transplantation from an insufficiently matched UD (UD–other) was identified to be a risk factor for disease recurrence (see also Table 2 for details). Furthermore, a longer time interval between diagnosis and HSCT was associated with a lower relapse incidence, possibly reflecting the selection of patients with less aggressive disease in the later transplantation group.

Transplantation-related mortality

CUMULATIVE INCIDENCE

1.0 Grade II-IV, MSD 0.57 [0.41-0.73] UD 0.39 [0.26-0.52] Grade III-IV, MSD 0.33 [0.18-0.48] UD 0.18 [0.08-0.28]

0.8

Survival and event-free survival Overall, 64 patients are alive after HSCT, resulting in a Kaplan– Meier estimate of survival of 63% (53–73) at 5 years (Figure 2). With a median observation time of 5.2 years (1.0–10.9), 59 patients are alive in first complete remission after HSCT. The 5-year probability of EFS after the first allograft is 59% (49–69), with no significant difference for patients grafted from either an MSD or a UD (67 vs 53%, P ¼ NS). However, for patients transplanted from a UD matched only by incomplete or serological HLA typing, the EFS was significantly lower (35%, in comparison: 67% for MSD, 62% for UD-8/8 and 56% for UD-7/8, P ¼ 0.03). In univariate analysis, the presence of a

1.0

0.8 PROBABILITY (95%)

Twenty patients died at a median time of 113 days (range 28–796) after HSCT because of transplantation-related causes, resulting in a 5-year cumulative incidence of 21% (14–31) (Figure 2). The 5-year cumulative incidence of TRM for patients transplanted from either an HLA-identical sibling or an unrelated volunteer was comparable (18 vs 23%, P ¼ NS), whereas there was a trend towards a higher TRM for patients transplanted from a UD, matched only by either incomplete or serological HLA typing (35%, in comparison: 18% for MSD, 16% for UD-8/8 and 20% for UD-7/8, P ¼ NS). In univariate analysis, age at transplantation X12 years, time interval between diagnosis and HSCT X4 months, a structural complex karyotype, and presence of both acute and extensive chronic GVHD were significantly associated with a higher probability of TRM (Table 1). The significant influence of the occurrence of grade III–IV aGVHD and age at HSCT X12 years on the risk of dying because of transplantation-related causes was confirmed in multivariate analysis using a Cox regression model (Table 2).

0.6

0.6 Overall-Survival P=0.63 [0.53-0.73] EFS P=0.59 [0.49-0.69]

0.4

0.2

P=0.21 [0.14-0.32] P=0.21 [0.14-0.32]

Relapse TRM

p=0.02

0.4

0.0

p=0.04

0.2 0.0 0

20 40 60 80 DAYS AFTER TRANSPLANTATION

100

Figure 1 Cumulative incidence of grade II–IV and grade III–IV acute graft-versus-host disease (GVHD) according to type of donor. MSD, matched sibling donor; UD, unrelated donor.

0

2

4

6

8

10

YEARS AFTER TRANSPLANTATION No. of patients at risk 97 45 11 1 28 63 97 40 10 1 25 60 10 1 97 40 25 60 97 40 10 1 25 60

12

Overall Survival EFS Relapse TRM

Figure 2 Kaplan–Meier estimate of overall survival and event-free survival (EFS) and cumulative incidence of relapse and transplantationrelated mortality (TRM) for 97 children with advanced MDS. Leukemia


458 Table 1

Univariate analysis of the probability of 5-year EFS, RI and TRM EFS

TRM

No. of patients

Probability

(95% CI)

Cumulative incidence

(95% CI)


(95% CI)

97

59%

(49–69)

21%

(14–32)

21%

(14–31)

65 32

60% 55% NS

(48–72) (36–74)

20% 23% NS

(12–33) (11–47)

20% 22% NS

(12–33) (11–43)

Age at diagnosis o12 years X12 years P-value

54 43

63% 53% NS

(49–77) (37–69)

22% 19% NS

(13–38) (9–37)

15% 28% 0.14

(8–28) (17–45)

Age at HSCT o12 years X12 years P-value

51 46

67% 50% 0.11

(53–81) (35–65)

24% 17% NS

(13–40) (9–34)

10% 33% o0.01

(4–23) (22–50)

Year of HSCT o2001 2001–2003 X2004 P-value

34 36 27

56% 60% 65% NS

(39–73) (44–76) (46–84)

21% 23% 12% 0.13

(11–40) (13–43) (4–34)

24% 17% 24% 0.13

(13–43) (8–35) (12–49)

Interval diagnosisFHSCT o4 months X4 months P-value

49 48

63% 52% NS

(47–79) (38–65)

27% 15% NS

(16–46) (8–30)

10% 31% 0.01

(4–23) (21–48)

Donor Matched sibling donor Matched unrelated donor P-value

39 58

67% 53% 0.12

(51–83) (40–66)

15% 24% 0.13

(7–35) (15–39)

18% 23% NS

(9–36) (14–36)

39

67%

(51–83)

15%

(7–35)

18%

(9–36)

31 10 17

62% 56% 35% 0.03

(44–80) (22–90) (12–58)

22% 24% 29% NS

(11–45) (7–82) (14–61)

16% 20% 35% NS

(7–36) (6–69) (19–67)

63% 64% 32% 0.07

(49–77) (46–82) (8–56)

15% 22% 41% 0.02

(7–31) (11–45) (22–77)

23% 14% 27% NS

(14–37) (6–34) (12–62)

62% 65% 45%

(41–83) (50–80) (23–67)

19% 16% 30%

(8–47) (7–33) (15–59)

18% 20% 25%

(8–45) (11–35) (12–53)

Overall probability or incidence Patient gender Male Female P-value

Donor Matched sibling donor Matched unrelated donor UD-8/8a UD-7/8b UD-otherc P-value

Most advanced morphological subtype RAEB 53 RAEB-T 29 MDR-AML 15 P-value Bone marrow blast percentage at HSCT o5% 22 5–o20% 46 X20% 20 Missing 9 P-value

Leukemia

RI

0.11

0.09

NS

Stem cell source Bone marrow Peripheral blood P-value

69 28

58% 60% NS

(46–70) (42–78)

20% 22% NS

(12–34) (11–45)

22% 18% NS

(14–34) (8–40)

Preparative regimen Busulfan p.o. Busulfan i.v. P-value

73 24

59% 56% NS

(47–71) (33–79)

18% 31% NS

(11–30) (16–63)

23% 13% NS

(15–36) (4–36)

Karyotype Normal Monosomy 7 Structurally complex Other aberration Missing P-value

32 29 5 28 3

67% 52% 0% 61%

(50–84) (32–72)

11% 24% 40% 28%

(4–32) (12–51) (14–100) (15–53)

22% 24% 60% 11%

(12–43) (13–46) (29–100) (4–31)

o0.01

(41–81)

o0.01

o0.01


459 Table 1 (Continued ) EFS

Donor/recipient gender Donor female/recipient male Other Missing P-value Pre-HSCT cytotoxic treatment All MDS subtypes None or low-dose AML-like P-value MDR AML only None or low-dose AML-like P-value

RI

TRM

No. of patients

Probability

(95% CI)


(95% CI)


(95% CI)

17 78 2

63% 58%

(38–88) (46–70)

24% 21%

(10–55) (13–33)

14% 22%

(4–51) (14–33)

NS

NS

NS

73 24

58% 62% NS

(46–70) (42–82)

20% 21% NS

(12–34) (10–46)

22% 17% NS

(14–34) (7–41)

7 8

14% 47% 0.06

(0–40) (9–85)

71% 16% 0.01

(45–100) (3–92)

14% 38% NS

(2–87) (15–92)

27 12

69% 59% NS

(51–87) (25–93)

8% 32% 0.09

(2–32) (12–51)

22% 8% NS

(11–45) (1–54)

Acute GVHD Grade 0–I Grade II–IV P-value

53 44

62% 55% NS

(48–76) (40–70)

29% 10% 0.12

(18–45) (4–27)

9% 35% o0.01

(4–22) (22–52)

Acute GVHD Grade 0–II Grade III–IV P-value

74 23

63% 42% 0.04

(51–75) (21–63)

24% 10% NS

(16–38) (3–36)

12% 48% o0.01

(7–23) (31–74)

Chronic GVHD Absent Present Not evaluable P-value

55 30 12

66% 68%

(52–80) (50–86)

24% 12%

(15–40) (4–34)

9% 21%

(4–21) (10–42)

Chronic GVHD Absent Limited Extensive Not evaluable P-value

55 18 12 12

GVHD prophylaxis Sibling donor transplants CSA CSA+MTX P-value

NS 66% 80% 50%

NS (52–88) (59–99) (22–78)

0.13

24% 14% 8% NS

NS (15–40) (4–51) (1–54)

9% 6% 42%

(4–21) (1–42) (21–81)

o0.01

Abbreviations: AML, acute myeloid leukemia; CI, confidence interval; CSA, cyclosporine-A; EFS, event-free survival; GVHD, graft-versus-host disease; HSCT, hematopoietic stem cell transplantation; i.v., intravenous; MDR, myelodysplasia-related; MTX, methotrexate; NS, not significant; p.o., per os; RAEB, refractory anemia with excess blasts; RI, cumulative incidence of relapse; TRM, transplant-related mortality. a Identical. b Mismatched in one antigen based on intermediate resolution typing for HLA-A, -B, and -C and high-resolution typing for HLA DRB1. c Identical based on intermediate resolution typing for HLA-A and -B and high-resolution typing for HLA DRB1 or serological typing only. Bold entries represent significance levels below 5%.

structurally complex karyotype and grade III–IV aGvHD significantly decreased the probability of EFS (Table 1). Patients who were classified as MDR-AML had a lower probability of EFS compared with patients with RAEB and RAEB-T (32 vs 63 and 64%, respectively); however, the difference was not statistically significant (P ¼ 0.07) (Figure 3). No other disease- or treatmentrelated variable including the blast percentage at HSCT and the use of intensive chemotherapy before HSCT had a significant influence on the probability of EFS (Table 1). In multivariate analysis, higher disease stage (MDR-AML vs RAEB), the occurrence of grade III–IV aGVHD and transplantation from an insufficiently matched UD (UD–other) were associated with a lower probability of EFS (Table 2).

Discussion This study reports on the results of HSCT in a large cohort of children with advanced primary MDS diagnosed according to defined criteria and transplanted following a homogeneous preparative regimen including three alkylating agents, namely busulfan, cyclophosphamide and melphalan.8,12 The study demonstrates that allogeneic HSCT offers a probability of survival of 63%. Relapse and TRM contributed equally to treatment failure. Data on outcome of children transplanted for childhood MDS are scarce and previously published reports include a limited number of patients with various variants of MDS transplanted following heterogeneous regimens.7,9–11 The largest pediatric Leukemia


460 Table 2 Multivariate analysis of variables influencing the probability of event-free survival, relapse incidence and transplantation-related mortality

Event-free survival Most advanced morphological subtype RAEB-T vs RAEB MDR-AML vs RAEB Patient gender Male vs female Patient age at HSCT X12 years vs o12 years Interval from diagnosis of MDS to HSCT X4 months vs o4 months Donor UD-8/8a vs MSD UD-7/8b vs MSD UD otherc vs MSD Acute GVHD Grade III/IV vs grade 0–II Relapse incidence Most advanced morphological subtype RAEB-T vs RAEB MDR-AML vs RAEB Patient gender Male vs female Patient age at HSCT X12 years vs o12 years Interval from diagnosis of MDS to HSCT X4 months vs o4 months Donor UD-8/8a vs MSD UD-7/8b vs MSD UD otherc vs MSD Acute GVHD grade III/IV vs grade 0–II Transplantation-related mortality Most advanced morphological subtype RAEB-T vs RAEB MDR-AML vs RAEB Patient gender Male vs female Patient age at HSCT X12 years vs o 12 years Interval from diagnosis of MDS to HSCT X4 months vs o4 months Donor UD-8/8a vs MSD UD-7/8b vs MSD) UD otherc vs MSD Acute GVHD Grade III/IV vs grade 0–II

1.08 2.71

(0.47–2.47) (1.16–6.39)

P

NS 0.02

0.6

0.4 P=0.63 [0.49-0.77] RAEB P=0.64 [0.46-0.82] RAEB-T MDR-AML P=0.32 [0.08-0.56]

0.2

1.26

(0.61–2.60)

NS

1.28

(0.63–2.58)

NS

0.90

(0.41–2.02)

NS

1.50 1.25 3.59

(0.57–3.95) (0.36–4.36) (1.36–9.48)

NS NS 0.01

2.39

(1.15–4.96)

0.02

1.25 10.73

(0.37–4.19) NS (2.54–45.38) o0.01

1.34

(0.45–4.00)

NS

0.54

(0.19–1.57)

NS

0.25

(0.07–0.88)

0.03

2.86 1.46 5.60

(0.71–11.59) (0.23–9.43) (1.38–22.66)

NS NS 0.02

0.75

(0.15–3.76)

NS

1.05 1.20

(0.30–3.74) (0.36–4.00)

NS NS

1.17

(0.42–3.25)

NS

2.95

(1.01–8.65)

0.04

3.16

(0.90–11.15)

0.07

0.58 0.49 1.36

(0.16–2.09) (0.08–2.96) (0.37–4.97)

NS NS NS

4.13

(1.63–10.48) o0.01

a

Identical. Mismatched in one antigen based on intermediate resolution typing for HLA-A, -B, and -C and high resolution typing for HLA DRB1. c Identical based on intermediate resolution typing for HLA-A,and -B and high resolution typing for HLA DRB1 or serological typing only. b

study reported the results of allogeneic HSCT in 94 children with MDS, including patients with refractory cytopenia, advanced MDS, as well as juvenile myelomonocytic leukemia and secondary MDS, demonstrating a probability of EFS of 41%.11 Leukemia

0.8 PROBABILITY (95%)

Relative risk (95% CI)

1.0

Log Rank: p=0.07

0.0 0

2

4

6

8

10

12

YEARS AFTER TRANSPLANTATION No. of patients at risk 53

34

24

13

4

0

RAEB

29

19

12

9

5

1

RAEB-T

15

7

4

3

1

0

AML

Figure 3 Kaplan–Meier estimate of event-free survival (EFS) according to highest MDS-subtype before HSCT. MDR-AML: myelodysplasiarelated acute myeloid leukemia; RAEB, refractory anemia with excess blasts; RAEB-T: RAEB in transformation.

Within the Children’s Cancer Group, patients with MDS and juvenile myelomonocytic leukemia received intensive chemotherapy that, depending on the remission status post induction therapy and the availability of a sibling donor, was followed by allogeneic HSCT.7 This strategy resulted in a probability of survival of 29–50% depending on MDS subtype. Other investigators focused on patients with MDS or AML associated with monosomy 723 or therapy-related MDS.24 On analysing the factors influencing the patients’ outcome, we found that the cumulative incidence of grade III–IV aGVHD was higher in patients transplanted from an MSD (33%) compared with patients transplanted from a UD (18%), demonstrating that more intensive GVHD prophylaxis (that is, the use of methotrexate and the addition of anti-thymocyte globulin in the majority of patients) applied in unrelated HSCT was effective in preventing GVHD. Support to this interpretation is given by the observation that the addition of methotrexate to GVHD prophylaxis in one-third of the patients transplanted from an HLA-matched sibling was able to reduce the incidence of acute GVHD (data not shown). Although the more intensive GVHD prophylaxis reduced the risk of aGVHD and consecutively the risk of dying from transplantation-related causes, this did not translate into a better EFS, as the positive effect in terms of TRM was counterbalanced by a higher risk of relapse (Table 1). Although these findings support the hypothesis of an effective graft-versus-malignancy effect in pediatric MDS, the increased risk of TRM in the presence of GVHD and the efficacy of more intensive GVHD prophylaxis in preventing GVHD asks for a risk-adapted strategy of GVHD prophylaxis. In the cohort reported here, the probability of dying from transplantation-related causes was 21%, age X12 years at HSCT and occurrence of grade III–IV aGVHD being the main risk factors. Similar or even higher rates of TRM have been reported in patients transplanted for advanced MDS.9–11 Interestingly, patients transplanted for de novo AML following the same preparative regimen have been reported to experience a lower risk of TRM.25,26 These observations may support the hypothesis that the high rates of TRM are not exclusively related to the preparative regimen, but that children with MDS suffer from unrecognized underlying disorders that are associated with a


461 higher sensitivity towards chemotherapeutic drugs or alloreactive effect. Our results confirm the relevant risk of relapse reported in most published series6,9–11 for children receiving allogeneic HSCT for advanced MDS and identify more advanced disease as a risk factor for relapse.11,24 Interestingly, the difference in RI and EFS was only observed on comparing MDR-AML with RAEB, whereas RAEB and RAEB-T seem to behave in a similar way. This observation provides support and justifies the pediatric approach to the classification of MDS, where, in contrast to the WHO classification that omitted RAEBT as a separate entity and included patients with more than 19% bone marrow blasts in the AML group, RAEB-T was maintained as a separate MDS subtype.4 One of the most controversial issues in the treatment of children with advanced MDS is the impact of intensive chemotherapy before HSCT. Children with MDS treated on AML protocols have been reported to experience a high rate of induction failure and relapse, resulting in a probability of OS of approximately 30%.6,7,27–29 Likewise, patients receiving allogeneic HSCT as primary treatment have a considerable risk of relapse, as well as transplantation-related toxicity.6,9,11 Therefore, the role of intensive chemotherapy before HSCT for patients with advanced MDS has remained a matter of debate. In this series, the use of intensive chemotherapy before the allograft did not improve survival or EFS. There was also no difference in both RI and TRM for patients who did or did not receive intensive chemotherapy before HSCT. When the analysis was restricted to children with MDR-AML, there was a significantly decreased risk of relapse in the intensive chemotherapy group, resulting in a non-significant advantage in terms of EFS (Table 1). The application of intensive chemotherapy was not tested in a systematic way and therefore the results must be interpreted with caution. However, we suggest that intensive chemotherapy cannot generally be recommended for children with RAEB and RAEB-T, but may be considered for children with MDR-AML. In conclusion, this study indicates that allogeneic HSCT from either an MSD or UD following a myeloablative conditioning regimen including three alkylating agents is feasible and offers a high probability of survival for children with advanced MDS. Outcomes are comparable for children with RAEB and RAEB-T, whereas patients with MDR-AML have an increased risk of relapse. For these patients, new strategies including the systematic application of novel antileukemic therapies, that is epigenetically active substances before HSCT, will have to be evaluated. Likewise, strategies able to reduce the risk of TRM in adolescents, possibly including better GVHD prevention, are warranted.

Conflict of interest The authors declare no conflict of interest.

References 1 Niemeyer CM, Kratz CP. Paediatric myelodysplastic syndromes and juvenile myelomonocytic leukaemia: molecular classification and treatment options. Br J Haematol 2008; 140: 610–624. 2 Hasle H, Wadsworth LD, Massing BG, McBride M, Schultz KR. A population-based study of childhood myelodysplastic syndrome in British Columbia, Canada. Br J Haematol 1999; 106: 1027–1032. 3 Passmore SJ, Chessells JM, Kempski H, Hann IM, Brownbill PA, Stiller CA. Pediatric myelodysplastic syndromes and juvenile myelomonocytic leukaemia in the UK: a population-based study of incidence and survival. Br J Haematol 2003; 121: 758–767.

4 Hasle H, Niemeyer CM, Chessells JM, Baumann I, Bennett JM, Kerndrup G et al. A pediatric approach to the WHO classification of myelodysplastic and myeloproliferative diseases. Leukemia 2003; 17: 277–282. 5 Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC Press: Lyon, 2008. 6 Sasaki H, Manabe A, Kojima S, Tsuchida M, Hayashi Y, Ikuta K et al. Myelodysplastic syndrome in childhood: a retrospective study of 189 patients in Japan. Leukemia 2001; 15: 1713–1720. 7 Woods WG, Barnard DR, Alonzo TA, Buckley JD, Kobrinsky N, Arthur DC 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. 8 Locatelli F, Pession A, Bonetti F, Maserati E, Prete L, Pedrazzoli P et al. Busulfan, cyclophosphamide and melphalan as conditioning regimen for bone marrow transplantation in children with myelodysplastic syndromes. Leukemia 1994; 8: 844–849. 9 Munoz A, de Heredia CD, Badell I, Bureo E, Gomez P, Martinez A et al. Allogeneic stem cell transplantation for myelodysplastic syndromes in children: a report from the Spanish Working Party for Blood and Marrow Transplantation in Children (GETMON). Pediatr Hematol Oncol 2009; 26: 345–355. 10 Parikh SH, Mendizabal A, Martin PL, Prasad VK, Szabolcs P, Driscoll TA et al. Unrelated donor umbilical cord blood transplantation in pediatric myelodysplastic syndrome: a single-center experience. Biol Blood Marrow Transplant 2009; 15: 948–955. 11 Yusuf U, Frangoul HA, Gooley TA, Woolfrey AE, Carpenter PA, Andrews RG et al. Allogeneic bone marrow transplantation in children with myelodysplastic syndrome or juvenile myelomonocytic leukemia: the Seattle experience. Bone Marrow Transplant 2004; 33: 805–814. 12 Cantu RA, Fenu S, Kerndrup G, van Wering ER, Niemeyer CM, Baumann I. Evaluation of dysplastic features in myelodysplastic syndromes: experience from the morphology group of the European Working Group of MDS in Childhood (EWOG-MDS). Ann Hematol 2005; 84: 429–433. 13 Glucksberg H, Storb R, Fefer A, Buckner CD, Neiman PE, Clift RA et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HLA-matched sibling donors. Transplantation 1974; 18: 295–304. 14 Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 Consensus Conference on acute GVHD grading. Bone Marrow Transplant 1995; 15: 825–828. 15 Storb R, Prentice RL, Sullivan KM, Shulman HM, Deeg HJ, Doney KC et al. Predictive factors in chronic graft-versus-host disease in patients with aplastic anemia treated by marrow transplantation from HLA-identical siblings. Ann Intern Med 1983; 98: 461–466. 16 Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 456–481. 17 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. 18 Klein JP, Rizzo JD, Zhang MJ, Keiding N. Statistical methods for the analysis and presentation of the results of bone marrow transplants. Part 2: Regression modeling. Bone Marrow Transplant 2001; 28: 1001–1011. 19 Pepe MS, Longton G, Pettinger M, Mori M, Fisher LD, Storb R. Summarizing data on survival, relapse, and chronic graft-versushost disease after bone marrow transplantation: motivation for and description of new methods. Br J Haematol 1993; 83: 602–607. 20 Cox D. Regression models and life tables (with discussion). J Royal Stat Soc B 1972; 34: 187. 21 Kruskal WH, Wallis WA. Use of ranks in one-criterion variance analysis. J Am Stat Assoc 1952; 47: 583–621. 22 Mann HB, Whitney DR. On a test of whether one or two random variables is stochastically larger than the other. Ann MathStat 1947; 18: 50–60. 23 Trobaugh-Lotrario AD, Kletzel M, Quinones RR, McGavran L, Proytcheva MA, Hunger SP et al. Monosomy 7 associated with pediatric acute myeloid leukemia (AML) and myelodysplastic Leukemia


462 syndrome (MDS): successful management by allogeneic hematopoietic stem cell transplant (HSCT). Bone Marrow Transplant 2005; 35: 143–149. 24 Woodard P, Barfield R, Hale G, Horwitz E, Leung W, Ribeiro R et al. Outcome of hematopoietic stem cell transplantation for pediatric patients with therapy-related acute myeloid leukemia or myelodysplastic syndrome. Pediatr Blood Cancer 2006; 47: 931–935. 25 Pession A, Rizzari C, Putti MC, Masetti R, Casale F, Fagioli F et al. Results of the AIEOP AML 2002/01 study for treatment of children with acute myeloid leukemia. Blood (ASH Annu Meet Abstr) 2009; 114: 17. 26 Zecca M, Masetti R, Balduzzi A, Fagioli F, Dini G, Messina C et al. Long-term outcome of children with acute myeloid leukemia in first remission given allogeneic HSCT from a matched family donor after a conditioning regimen comprising busulfan, cyclopho-

sphamide and melphalan. Blood (ASH Annu Meet Abstr) 2009; 114: 2288. 27 Chan GC, Wang WC, Raimondi SC, Behm FG, Krance RA, Chen G et al. Myelodysplastic syndrome in children: differentiation from acute myeloid leukemia with a low blast count. Leukemia 1997; 11: 206–211. 28 Hasle H, Kerndrup G, Yssing M, Clausen N, Ostergaard E, Jacobsen N et al. Intensive chemotherapy in childhood myelodysplastic syndrome. A comparison with results in acute myeloid leukemia. Leukemia 1996; 10: 1269–1273. 29 Webb DK, Harrison G, Stevens RF, Gibson BG, Hann IM, Wheatley K. Relationships between age at diagnosis, clinical features, and outcome of therapy in children treated in the Medical Research Council AML 10 and 12 trials for acute myeloid leukemia. Blood 2001; 98: 1714–1720.

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Leukemia