Scholarly article on topic 'Comparable Long-Term Survival after Bone Marrow versus Peripheral Blood Progenitor Cell Transplantation from Matched Unrelated Donors in Children with Hematologic Malignancies'

Comparable Long-Term Survival after Bone Marrow versus Peripheral Blood Progenitor Cell Transplantation from Matched Unrelated Donors in Children with Hematologic Malignancies Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Allogeneic hematopoietic stem cell transplantation" / Children / "Matched unrelated donor" / "Peripheral blood progenitor cells" / "Bone marrow"}

Abstract of research paper on Clinical medicine, author of scientific article — Roland Meisel, Hans-Jürgen Laws, Stefan Balzer, Benedikt Bernbeck, Christof Kramm, et al.

Abstract Despite the increasing use of peripheral blood progenitor cells (PBPC) instead of bone marrow (BM) for allogeneic hematopoietic stem cell transplantation (allo HSCT) from human leukocyte antigen (HLA)-matched unrelated donors in children, the relative benefits and risks of both stem cell sources in the pediatric setting remain largely unknown. Recently, the only larger study comparing the value of the 2 stem cell sources in a young patient group was confined to transplantation from HLA-identical sibling donors in older children and adolescents with acute leukemia. Based on the paucity of data in pediatric HLA-matched unrelated donor transplantation, we analyzed the outcome of 23 BM and 38 PBPC transplantations performed at our center. Neutrophil and platelet engraftment were achieved significantly faster in PBPC compared to BM recipients (18 versus 22 days and 26 versus 33 days; P < .001 and P = .03) whereas the risk for grade II-IV acute graft-versus-host disease (aGVHD) (62% versus 55%; P = .53) and chronic GVHD (cGVHD 65% versus 59%; P = .54) was comparable. As overall survival (OS; PBPC versus BM: 47.5% ± 8.6% versus 51.8% ± 10.5%; P = .88) and relapse-free survival (43.3% ± 8.3% versus 51.8% ± 10.5%; P = .60) are without detectable difference, PBPC and BM appear both as a valid stem cell source for HLA-matched unrelated donor transplantation in children with hematologic malignancies.

Similar topics of scientific paper in Clinical medicine , author of scholarly article — Roland Meisel, Hans-Jürgen Laws, Stefan Balzer, Benedikt Bernbeck, Christof Kramm, et al.

Academic research paper on topic "Comparable Long-Term Survival after Bone Marrow versus Peripheral Blood Progenitor Cell Transplantation from Matched Unrelated Donors in Children with Hematologic Malignancies"

Biology of Blood and Marrow Transplantation 13:1338-1345 (2007) © 2007 American Society for Blood and Marrow Transplantation 1083-8791/07/1311-0001$32.00/0 doi:10.1016/j.bbmt.2007.07.009

AS BMI

American Society for Blood arid Marrow Transplantation

Comparable Long-Term Survival after Bone Marrow versus Peripheral Blood Progenitor Cell Transplantation from Matched Unrelated Donors in Children with Hematologic Malignancies

Roland Meisel,1 Hans-Jürgen Laws,1 Stefan Balzer,1 Benedikt Bernbeck,1 Christof Kramm,1

Stefan Schönberger,1 Kumar Sinha,1 Anja Tröger,1 Monika Schmitz,1 Johannes Fischer,2 Ulrich Göbel,1

Jürgen Enczmann,2 Dagmar Dilloo1

1Clinic for Pediatric Oncology, Hematology and Clinical Immunology, and 2Institute for Transplantation Diagnostics and Cell Therapeutics, University Clinic of Düsseldorf, Heinrich-Heine-University, Düsseldorf, Germany

Correspondence and reprint requests: Roland Meisel, MD, Universitätsklinikum Düsseldorf, Zentrum für Kinder- und Jugendmedizin, Klinik für Kinder-Onkologie, Hämatologie und Klinische Immunologie, Moorenstrasse 5, D-40225 Düsseldorf, Germany (e-mail: meisel@med.uni-duesseldorf.de).

Received May 29, 2007; accepted July 16, 2007

ABSTRACT

Despite the increasing use of peripheral blood progenitor cells (PBPC) instead of bone marrow (BM) for allogeneic hematopoietic stem cell transplantation (allo HSCT) from human leukocyte antigen (HLA)-matched unrelated donors in children, the relative benefits and risks of both stem cell sources in the pediatric setting remain largely unknown. Recently, the only larger study comparing the value of the 2 stem cell sources in a young patient group was confined to transplantation from HLA-identical sibling donors in older children and adolescents with acute leukemia. Based on the paucity of data in pediatric HLA-matched unrelated donor transplantation, we analyzed the outcome of 23 BM and 38 PBPC transplantations performed at our center. Neutrophil and platelet engraftment were achieved significantly faster in PBPC compared to BM recipients (18 versus 22 days and 26 versus 33 days; P < .001 and P = .03) whereas the risk for grade II-IV acute graft-versus-host disease (aGVHD) (62% versus 55%; P = .53) and chronic GVHD (cGVHD 65% versus 59%; P = .54) was comparable. As overall survival (OS; PBPC versus BM: 47.5% ± 8.6% versus 51.8% ± 10.5%; P = .88) and relapse-free survival (43.3% ± 8.3% versus 51.8% ± 10.5%; P = .60) are without detectable difference, PBPC and BM appear both as a valid stem cell source for HLA-matched unrelated donor transplantation in children with hematologic malignancies. © 2007 American Society for Blood and Marrow Transplantation

KEY WORDS

Allogeneic hematopoietic stem cell transplantation • Children • Matched unrelated donor • Peripheral blood progenitor cells • Bone marrow

INTRODUCTION

Peripheral blood progenitor cells (PBPC) are increasingly used instead of bone marrow (BM) for unrelated donor allogeneic hematopoietic stem cell transplantation (alloHSCT) in children with hemato-logic malignancies [1]. Initially, PBPCs have been introduced as an alternate stem cell source for alloHSCT hypothesizing that their increased hema-topoietic progenitor and immune cell content will lead

to prompter engraftment and more efficient graft-versus-leukemia responses with improved outcome from both reduced transplant-related mortality (TRM) and relapse rate [2,3]. Furthermore, the steadily increasing preference of volunteer donors for PBPC donation has also contributed to the observed shift in the use of stem cell sources. In contrast to substantial literature on the use of PBPC in adult recipients of matched related allografts, comparisons of BM versus PBPC in the unrelated donor setting are

sparse for adult patients and virtually nonexistent for children [4]. The few available studies in adults have yielded conflicting results with regard to relapse-free and overall survival after PBPC compared to BM transplantation from human leukocyte antigen (HLA)-matched unrelated donors [5-8]. However, relative benefits of PBPC versus BM transplantation may substantially differ in children because of a lesser propensity to graft-versus-host disease (GVHD) in younger transplant recipients and different disease kinetics in childhood leukemia.

Recently, the only other larger study comparing the value of the 2 stem cell sources in a young patient group was confined to transplantation from HLA-identical sibling donors in older children and adolescents with acute leukemia [9]. Based on the paucity of data in pediatric HLA-matched unrelated donor transplantation, we analyzed the long-term outcome of 61 transplants performed at our center between 1992 and 2004. In this first detailed comparison between PBPC and BM in unrelated donor transplantation in children with hematologic malignancies we demonstrate that there is no detectable difference in TRM, relapse rate and, most importantly, long-term relapse-free and OS in pediatric patients transplanted with either stem cell source.

PATIENTS, MATERIALS AND METHODS

Patient, Disease, and Transplant Characteristics

All pediatric patients (<18 years) with lymphohe-matologic malignancies who were transplanted at our center with unmanipulated PBPC (n = 38) or BM (n = 23) from >5 of 6 HLA-matched unrelated donors following myeloablative conditioning were included in this retrospective analysis. Table 1 provides details on the patient, disease, and transplant characteristics of the study cohort regarding the parameters of age, sex, and cytomegalovirus (CMV) serostatus of both recipient and donor as well as the patients' disease and risk status, the stem cell source (PBPC versus BM) and year of transplantation, transplanted total nucleated, and CD34+ cell dose (the latter only for PBPC recipients), degree of donor-recipient matching at the HLA-A, -B, and -DRßl locus, conditioning regimen, and GVHD prophylaxis.

HLA-A and -B typing was performed either by se-rologic (until 1998) or low-resolution molecular typing, whereas HLA-DRßl typing was done by high-resolution molecular typing throughout the whole study period [10]. Pretransplant conditioning with a total-body irradiation (TBI-) versus busulfan-based regimen varied according to patients' age, disease, and risk status based on the guidelines of the respective study protocols of the "Gesellschaft für Pädiatrische Onkologie und Häma-tologie" (GPOH) and the "Pädiatrische Arbeitsgemein-

schaft fur Knochenmark- und Blutstammzelltransplantation" (PAD-AG-KBT) and was myeloablative in all cases. With regard to transplant indications patients with acute lymphoblastic leukemia (ALL) and acute myelogenous leukemia (AML) in first and second complete remission, with chronic myelogenous leukemia (CML) in first chronic phase, non-Hodgkin lymphoma (NHL) in remission and refractory cytopenia (RC) were considered standard risk, whereas patients with ALL/AML in equal to or greater than third complete remission or nonremission, CML equal to or greater than accelerated phase, NHL in nonremission, myelodysplastic syndrome (MDS) with excess of blasts and juvenile my-elomonocytic leukemia (JMML) were classified as high risk. Immunosuppressive therapy for prevention of graft rejection and GVHD consisted of cyclosporin A (CSA) + methotrexate (MTX) combined with an-tibody-/serotherapy in most cases (n = 54) with a minority of patients receiving MTX + CSA alone (n = 3) or CSA + antibody-/serotherapy (n = 4).

Study Endpoints

As the endpoints of this retrospective analysis, hematologic recovery, acute and chronic GVHD (aGVHD, cGVHD), relapse rate, TRM, death of disease (DOD), relapse-free survival (RFS) and OS were compared following PBPC versus BM transplantation. Neutrophil engraftment was defined as achievement of an absolute neutrophil count (ANC) >500/^L and/or a white blood cell count >1000/^L for 3 consecutive days, platelet engraftment as achievement of a platelet count >50,000/^L for 3 consecutive days without platelet support in the previous 7 days. For analysis of aGVHD and cGVHD only patients with evidence of donor engraftment and survival beyond day +100 (for cGVHD) were considered evaluable. GVHD assessment was based on clinical signs and symptoms, laboratory tests, and biopsy, if applicable. aGVHD and cGVHD were graded according to previously described criteria [11,12]. TRM was defined as death in remission, DOD as death following relapse of the underlying hematologic malignancy. RFS was defined as survival in complete remission; death and relapse were considered events and surviving patients were censored at last follow-up. For analysis of OS, death was the event and surviving patients were censored at last follow-up.

Statistical Analysis

For statistical comparisons between the PBPC and BM groups the Fisher's exact test was used for categoric parameters and the Mann-Whitney test for continuous variables. Kaplan-Meier estimates and the log-rank test were employed for comparison of time-dependent outcome parameters following PBPC versus BM transplantation [13]. Multivariate analysis us-

Table 1. Characteristics of the Study Cohorts PBPC BM

Variable No. (%) No. (%) P

Total No. 38 23

Patient age, years ns

Median 7.8 5.6

Range 0.6-18.0 0.8-18

Patient sex, male 28 (74%) II (48%) .06

Patient CMV serostatus ns

neg. 28 (74%) 17 (74%)

pos. 10 (26%) 5 (22%)

Donor age, years ns

Median 38 4I

Range 24-57 21-54

Donor sex, male 27 (71%) 13 (57%) ns

Donor CMV serostatus ns

neg. 23 (61%) 14 (61%)

pos. 15 (39%) 9 (39%)

Disease characteristics ns

ALL 17 (45%) 10 (44%)

CRI/CR2 I3 (34%) 7 (30%)

CR>3/NR 4 (11%) 3 (13%)

AML 10 (26%) 6 (26%)

CRI/CR2 6 (18%) 6 (26%)

CR>3/NR 4 (11%) 0 (0%)

MDS/JMML 7 (18%) 3 (13%)

RC 3 (8%) 0 (0%)

RAEB/RAEB-T 2 (5%) 2 (13%)

JMML 2 (5%) 1 (4%)

CML 2 (5%) 3 (13%)

CPI 2 (5%) 2 (9%)

CP>2 0 (0%) 1 (4%)

NHL 2 (5%) 1 (4%)

CRI 1 (3%) 0 (0%)

NR 1 (3%) 1 (4%)

Risk status* ns

Standard 25 (66%) 16 (70%)

High 13 (34%) 7 (30%)

Nonremission (ALL + AML) 6 (16%) 1 (4%) .2

Conditioning .01

TBI-based 22 (58%) 21 (91%)

Busulfan-based 16 (42%) 2 (9%)

GVHD prophytaxis ns

MTX + CSA 3 (8%) 0

MTX + CSA + antibody^ 3I (82%) 23 (100%)

CSA + antibody^ + other 4 (10%) 0

Year of transplantation <.001

Median 2001 I995

Range I997-2004 I992-2003

HLA match (HLA-A,-B,-DR) ns

6/6 36 (95%) 19 (83%)

5/6 2 (5%) 4 (17%)

Nucleated cell dose, X I08/kg <.001

Median 15.5 7.0

Range 6.2-42.0 2.8-14.6

CD34 cell dose Xl06/kg

Median 10.0 nd

Range 2.1-21.0 nd

PBPC indicates peripheral blood progenitor cells; BM, bone marrow; No, number; ns, not significant; CMV, cytomegalovirus; ALL, acute lymphoblastic leukemia; CR, complete remission; NR, nonremission; AML, acute myeloblastic leukemia; MDS, myelodysplastic syndrome; JMML, juvenile myelomonocytic leukemia; RC, refractory cytopenia; RAEB, refractory anemia with excess blasts; RAEB-T, refractory anemia with excess blasts in transformation; CML, chronic myelogenous leukemia; CP, chronic phase; NHL, non-Hodgkin lymphoma; TBI, total-body irradiation; GVHD, graft-versus-host disease; CSA, cyclosporine A; MTX, methotrexate; HLA, human leukocyte antigen; nd, not determined. *Standard risk: ALL/AML in first and second complete remission, CML in first chronic phase, MDS-refractory cytopenia, NHL in complete remission; high risk: ALL/AML in equal to or greater than third complete remission or nonremission, CML equal to or greater than accelerated phase, MDS-refractory anemia with excess of blasts, JMML, NHL in nonremission. tAntibody was ATG/ALG during conditioning and/or anti-interleukin2-receptor antibody BT563/leukotac™ posttransplantation.

ing Cox regression was carried out to determine the impact of potential risk factors on OS as the clinically most relevant outcome parameter [14]. Those patient, disease, and transplant variables with a P-value <.2 in univariate analysis were entered into the final model with the exception of the parameter stem cell source, which was held in the model regardless of P-value. All P-values are 2 sided, with P < .05 considered statistically significant. Statistical analyses were performed using the SPSS software package (Version 12.0.1).

RESULTS Study Cohort

Sixty-one patients were included into the study: 38 in the PBPC, and 23 in the BM transplant group. Patient and donor characteristics were comparable between the 2 groups with the exception of a statistically nonsignificant trend toward a higher proportion of male patients (74% versus 48%; P = .06) in the PBPC compared to the BM group (Table 1). Likewise, disease categories were similarly distributed in both groups. However, 16% versus 4% of acute leukemia patients in the PBPC versus BM group did not achieve remission prior to transplantation (P = .2), and therefore constitute a very high-risk group. With regard to transplant characteristics, about half of the patients in the PBPC group and the majority of patients in the BM group received TBI-based conditioning therapy (58% versus 91%, P = .01). Because of the later establishment of PBPC as a stem cell source for HSCT, the time period during which transplantation with PBPC was performed is shorter than for BM (1997-2004 versus 1992-2003), and consequently, median year of transplant was more recent in the PBPC group (P < .001). Also as expected, PBPC allografts contained an approximately 2.2 times higher total nucleated cell dose than those from BM donors (P < .001).

Engraftment

Primary neutrophil engraftment was attained in 37 of 38 (97%) PBPC and 22 of 23 (96%) BM transplant recipients despite the lack of prophylactic administration of hematopoietic growth factors. Two patients died without achieving neutrophil engraft-ment on day +6 and day +31 after transplantation, respectively, because of refractory progressive AML and sepsis with multiorgan failure. Time to myeloid engraftment was significantly shorter after PBPC transplantation (median 18 days, range: 9-28 days) compared to BM transplantation (median 22 days, range 14-43 days, P < .001; Figure 1A). Similarly, platelet recovery in patients transplanted with PBPC (median 26 days, range: 15-104 days) was achieved 7

Figure 1. Time to neutrophil and platelet engraftment. The cumulative probability of reaching neutrophil (A) and platelet (B) engraft-ment is shown for the PBPC group (dotted line) and BM group (solid line). N, number of patients evaluable in each arm of the study cohort; Ev, number of events (neutrophil/platelet engraftment) observed in each arm.

days earlier than in patients who received BM (median 33 days, range: 20-128 days, P = .03; Figure 1B).

aGVHD and cGVHD

The cumulative incidence of clinically relevant aGVHD grade II-IV (PBPC versus BM: 62% ± 8% versus 55% ± 11%, P = .53; Figure 2A) and severe aGVHD grade III and IV (PBPC versus BM: 30% ± 8% versus 23% ± 9%, P = .52; Figure 2B) was comparable between both groups. Also, the incidence of cGvHD did not differ significantly between PBPC

Figure 2. Incidence of aGVHD. The cumulative incidences of grade II-IV (A) and grade III-IV (B) aGVHD is shown for the PBPC group (dotted line) and BM group (solid line). N, number of evaluable patients in each arm of the study cohort; Ev, number of events (grade II-IV/III-IV aGVHD) observed in each arm.

and BM allograft recipients (65% ± 10% versus 59% ± 13%, P = .54; Figure 3A). However, there was a statistically nonsignificant trend towards a higher risk for clinically extended cGVHD with 50% ± 10% versus 26% ± 11% of patients in the PBPC versus BM groups, respectively (P = .14; Figure 3B). Further analyses of the clinical impact of extended cGVHD on long-term outcome revealed that in 9 of 12 (75%) surviving patients extended cGVHD resolved without disabling sequelae, whereas 1 patient was lost to follow-up. Consequently, only 2 patients still exhibited GVHD-related functional impairment involving ei-

ther the skin or lungs at last follow-up. With regard to GVHD-related mortality we could not detect any difference between the 2 stem cell sources: 7 patients, 2 of 23 (9%) from the BM and 5 of 38 (13%) from the PBPC group died from GVHD-related complications (P = .70).

TRM, Relapse, and Survival

The trend toward a higher risk for extended cGVHD in the PBPC group (Figure 3B). did not translate into any detectable difference in TRM: A

§ 0-6 -

PBPC: 65% ± 10% ■■♦•«•■■■■ft.................tt—"

BM: 59% ± 13%

BM: N=16, Ev= 9(56%) PBPC: N = 27, Ev= 11 (63%) p = 6.54

1 2 Years after transplantation

1 08 -

о Ё О

u 0.6-

2 0.2 -о

PBPC: 56% ±16%

BM: 26% ± 11%

BM: N=16, Ev= 4(25%) PBPC: N = 27, Ev= 13 (48%) p = 6.14

0 12 3

Years after transplantation

Figure 3. Incidence of cGvHD. The cumulative incidences of any grade (A) and clinically extensive (B) cGVHD is shown for the PBPC group (dotted line) and BM group (solid line). N, number of evaluable patients in each arm of the study cohort; Ev, number of events (any grade/clinically extensive cGvHD) observed in each arm.

Table 2. Multivariate Analysis of Risk Factors for Mortality

Variable* Relative Risk 95% CI P

Stem cell source, PBPC 1.17 0.54-2.53 .69

Risk status, high 2.40 1.12-5.15 .02

Donor sex, female 2.21 1.04-4.70 .04

HLA-match, 5/6 antigens 1.25 0.41-3.80 .69

CI indicates confidence interval; PBPC, peripheral blood progenitor cells; HLA, human leukocyte antigen.

"Variables analyzed for prognostic impact on mortality: age, sex, and cytomegalovirus serostatus of both recipient and donor, patients' disease and risk status, stem cell source (PBPC versus BM), year of transplantation, transplanted total nucleated cell dose, degree of donor-recipient HLA matching, conditioning regimen, and GVHD prophylaxis. Only variables with a P-value <.2 in univariate analysis were entered into the final multivar-iate model with the exception of the parameter stem cell source, which was held in the model irrespective of P-value.

total of 17 patients, 11 (29%) from the PBPC and 6 (26%) from the BM cohort, died of transplant-related complications between days +22 and +967 (median 124 days) after transplantation (P = .90). Likewise, relapse of the underlying malignant disease occurred in comparable frequency in both groups with 10 of 38 (26%) patients from the PBPC group and 6 of 23 (26%) patients from the BM group at a median of 88 days after transplantation (P = 1.0). With a median follow-up of 3.4 years (PBPC) and 10.0 years (BM) RFS and OS was without detectable difference between both groups (Figure 4). When nonremission patients, whereas constitute an exceptional high risk group and were not evenly distributed between the PBPC and the BM group, were excluded from analysis, the survival curves became superimposable with RFS rates of 51.5% ± 9.2% versus 54.2% ± 10.7% (P = .97) and OS rates of 56.4% ± 9.5% versus 54.2% ± 10.7% (P = .73) for PBSC versus BM, respectively. These clinically most relevant results were confirmed in a multivariate analysis showing that advanced disease status at transplant (relative risk [RR] 2.4, 95% confidence interval [CI] 1.1-5.2, P = .02) and female donor sex (RR 2.2, 95% CI 1.04-4.70, P = .04) were significant, independent risk factors for mortality, whereas the stem cell source (PBPC versus BM) had no effect (RR 1.2, 95% CI 0.5-2.5, P = .69; Table 2). Moreover, those factors that were unevenly distributed between the PBPC and BM groups, that is, conditioning regimen, year of transplantation, and transplanted nucleated cell dose (Table 1), had no significant impact on relapse-free and OS in univariate and multivariate analysis (data not shown). Further analyses of the potential mechanism by which donor sex affects OS revealed that children transplanted from a female donor carry a significantly higher risk for aGVHD grade III-IV (50% versus 15% in female versus male donor transplantion; P = .01), and this

translates into a higher rate of TRM (48% versus 18%; P = .03).

DISCUSSION

After more than a decade of experience in G-CSF-mobilized T cell replete PBPC transplantation, a number of clinical studies comparing PBPC with BM for alloHSCT have consistently documented more rapid engraftment but have yielded conflicting results with regard to other critical outcome parameters [4-8]. Particularly in pediatric transplantation, a detailed comparison between the 2 stem cell sources in

Figure 4. Probability of relapse-free survival (RFS) and overall survival (OS). RFS (A) and OS (B) is depicted for the PBPC group (dotted line) and BM group (solid line). N, number of patients in each arm of the study cohort; Ev, number of events observed in each arm.

alloHSCT from unrelated donors is missing. Here we present the first single center analysis of the outcome following PBPC versus BM transplantation for hematologic malignancies from HLA-matched unrelated donors in an entirely pediatric cohort.

In keeping with most adult studies, we observed a significantly shorter time to neutrophil and platelet engraftment after PBPC compared to BM transplantation with a difference in median time to engraftment of 4 and 7 days, respectively. As the use of MTX has been associated with a delay in neutrophil engraftment following alloHSCT [15,16], it is of particular note that in our study >90% of patients in both the PBPC and the BM group received an MTX-containing GVHD prophylaxis regimen such that a potential bias from differential MTX usage can be excluded. Thus, the swifter engraftment in the PBPC group is most likely attributable to the higher progenitor and stem cell content in PBPC compared to BM grafts. This observation is in accordance with the results of virtually all studies in the HLA-identical sibling setting and the few available studies in adult recipients of HLA-matched unrelated donor alloHSCT [4-9,17]. However, as in the majority of these studies, in our analysis faster engraftment in the PBPC group did not translate into any detectable reduction in TRM. This may be a consequence of the routine implementation of early broad spectrum antimicrobial therapy as the standard of care for neutropenic fever after alloHSCT.

One of the major concerns when using PBPC instead of BM is the increased risk of clinically relevant aGVHD and cGVHD related to the approximately 10-fold higher T cell content in PBPC grafts. We found no difference in the incidences of aGVHD grade II-IV (62% versus 55%, P = .53) and severe aGVHD grade III-IV (30% versus 23%, P = .52) between the PBPC and the BM group. Again, our data on comparable aGVHD in pediatric unrelated alloHSCT using PBPC and BM reflect the results of the IBMTR study in pediatric HLA-identical sibling transplantation [9] and a larger retrospective comparison of PBPC versus BM transplantation in HLA-matched unrelated donor transplantation in adults [6]. With regard to overall incidence of cGVHD, we could also not detect any significant difference (PBPC versus BM: 65% versus 59%, P = .54). However, a strong but statistically nonsignificant trend toward a higher risk for clinically extended cGVHD (50% versus 26%; P = .14) was observed in the PBPC group. A higher propensity to cGVHD in recipients of PBPC grafts has been observed in the majority of randomized studies in related alloHSCT comparing PBPC with BM for adults and the above-mentioned retrospective IBMTR study of HLA-identical sibling transplantation in older children and adolescents [4,9]. Likewise, a recent follow-up report to the above mentioned retrospective study on the long-term outcome

following PBPC versus BM transplantation in the adult matched unrelated donor setting found a higher incidence of extensive cGVHD in the PBPC group [18]. To assess the potential clinical impact of a higher rate of extensive cGVHD in pediatric unrelated donor transplantation we compared the proportion of GVHD-related deaths in the PBPC and the BM group in our cohort and found the overall incidence to be low and without difference between the PBPC and BM group (13% versus 9%, P = .7). Moreover, we found that overall 9 of 12 patients suffering from clinically extensive cGVHD subsequently recovered without disabling sequelae, whereas only 2 patients still exhibited GVHD-related functional impairment at last follow-up. Thus, the trend toward an increased risk for extended cGVHD following PBPC from matched-unrelated donors in children does not result in a higher rate of TRM. Although quality of life may be affected in long-term survivors with cGVHD, this issue still needs to be adequately addressed in future studies of larger patient cohorts.

As the clinically most relevant outcome parameters for children with hematologic malignancies we analyzed RFS and OS in recipients of PBPC and BM transplantation. As a consequence of comparable rates of relapses and TRM in both groups, we found no significant difference in RFS (43.3% versus 51.8%; P = .60) and OS (47.5% versus 51.8%; P = .88) between PBPC and BM recipients. This result was verified in a multivariate analysis of OS including those risk factors as covariates that were unevenly distributed between the PBPC and the BM groups, thus confirming that the stem cell source has no significant impact on overall outcome in unrelated pedi-atric alloHSCT. This result is in contrast to the IBMTR study on HLA-identical sibling transplantation in older children with acute leukemia, which revealed a 10% higher mortality rate in the PBPC group [9]. Thus, our observation underscores the notion that the relative benefits of PBPC versus BM as a stem cell source may well depend on the disease entities treated, the donor type (ie, related versus unrelated) as well as the patient age. However, as our study was retrospective and involved a limited number of patients, it may currently not be sufficiently powered to detect a smaller difference in outcome between recipients of BM and PBPC transplants. Therefore, to further define the role of PBPC versus BM for alloHSCT from HLA-matched unrelated donors in children with he-matologic malignancies, prospective studies comprising a higher number of patients are clearly warranted. In this regard, a prospective randomized trial currently performed by the Blood and Marrow Transplant Clinical Trials Network (BMT CTN) in the United States adresses this issue by comparing gran-ulocyte-colony stimulating factor (G-CSF) mobilized PBPC with marrow transplantation from HLA com-

patible unrelated donors. However, the results of this trial, which also includes pediatric patients, will only be available in a couple of years, as recruitment is still ongoing. In the meantime, however, it remains important to note that there are currently no pediatric data justifying the preferential use of either PBPC or BM as stem cell source for unrelated alloHSCT. We believe that this fact may also be of relevance when counseling unrelated volunteer donors.

ACKNOWLEDGMENTS

We thank Arndt Borkhardt (Clinic for Pediatric Oncology, Hematology and Clinical Immunology, University Clinic of Düsseldorf, Heinrich-Heine-University, Düsseldorf, Germany) for critical reading of the manuscript and the nursing staff of the pediatric BMT unit of University Clinic Düsseldorf for excellent patient care. The stem cell transplantation programme of the University Clinic Düsseldorf is generously supported by the Elterninitiative Kinderkrebsklink Düsseldorf e.V. All authors declare no conflicts.

REFERENCES

1. Miano M, Labopin M, Hartmann O, et al. Haematopoietic stem cell transplantation trends in children over the last three decades: a survey by the pediatric diseases working party of the European Group for Blood and Marrow Transplantation. Bone Marrow Transplant. 2007;39:89-99.

2. Bensinger WI, Clift R, Martin P, et al. Allogeneic peripheral blood progenitor cell transplantation in patients with advanced hematologic malignancies: a retrospective comparison with marrow transplantation. Blood. 1996;88:2794-2800.

3. Ottinger HD, Beelen DW, Scheulen B, Schaefer UW and Grosse-Wilde H. Improved immune reconstitution after allotransplantation of peripheral blood progenitor cells instead of bone marrow. Blood. 1996;88:2775-2779.

4. Allogeneic peripheral blood stem-cell compared with bone marrow transplantation in the management of hematologic malignancies: an individual patient data meta-analysis of nine randomized trials. J Clin Oncol. 2005;23:5074-5087.

5. Ringden O, Remberger M, Runde V, et al. Peripheral blood progenitor cell transplantation from unrelated donors: a comparison with marrow transplantation. Blood. 1999;94:455-464.

6. Remberger M, Ringden O, Blau IW, et al. No difference in graft-versus-host disease, relapse, and survival comparing peripheral stem cells to bone marrow using unrelated donors. Blood. 2001;98:1739-1745.

7. Elmaagacli AH, Basoglu S, Peceny R, et al. Improved disease-free-survival after transplantation of peripheral blood progenitor cells as compared with bone marrow from HLA-identical unrelated donors in patients with first chronic phase chronic myeloid leukemia. Blood. 2002;99:1130-1135.

8. Garderet L, Labopin M, Gorin NC, et al. Patients with acute lymphoblastic leukemia allografted with a matched unrelated donor may have a lower survival with a peripheral blood progenitor cell graft compared to bone marrow. Bone Marrow Transplant. 2003;31:23-29.

9. Eapen M, Horowitz MM, Klein JP, et al. Higher mortality after allogeneic peripheral-blood transplantation compared with bone marrow in children and adolescents: the Histocompatibil-ity and Alternate Stem Cell Source Working Committee of the International Bone Marrow Transplant Registry. J Clin Oncol. 2004;22:4872-4880.

10. Knipper AJ, Hinney A, Schuch B, Enczmann J, Uhrberg M, Wernet P. Selection of unrelated bone marrow donors by PCR-SSP typing and subsequent nonradioactive sequence-based typing for HLA DRB1/3/4/5, DQB1, and DPB1 alleles. Tissue Antigens. 1994;44:275-284.

11. Glucksberg H, Storb R, Fefer A, et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HL-A-matched sibling donors. Transplantation. 1974;18: 295-304.

12. Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graft-versus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. Am J Med. 1980;69:204-217.

13. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457-481.

14. Cox DR. Regression models and life tables. J R Stat Soc. 1972; 34:187-220.

15. Deeg HJ, Storb R, Thomas ED, et al. Cyclosporine as prophylaxis for graft-versus-host disease: a randomized study in patients undergoing marrow transplantation for acute nonlym-phoblastic leukemia. Blood. 1985;65:1325-1334.

16. Locatelli F, Bruno B, Zecca M, et al. Cyclosporin A and short-term methotrexate versus cyclosporin A as graft versus host disease prophylaxis in patients with severe aplastic anemia given allogeneic bone marrow transplantation from an HLA-identical sibling: results of a GITMO/EBMT randomized trial. Blood. 2000;96:1690-1697.

17. Meisel R, Enczmann J, Balzer S, et al. Similar survival following HLA-identical sibling transplantation for standard indication in children with haematologic malignancies: a single center comparison of mobilized peripheral blood progenitor cell with bone marrow transplantation. Klin Padiatr. 2005;217:135-141.

18. Remberger M, Beelen DW, Fauser A, Basara N, Basu O, Ringden O. Increased risk of extensive chronic graft-versus-host disease after allogeneic peripheral blood progenitor cell transplantation using unrelated donors. Blood. 2005;105:548-551.