|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2004.06.102 on December 22 2003 © 2004 American Society of Clinical Oncology. Treatment With Granulocyte Colony-Stimulating Factor After Allogeneic Bone Marrow Transplantation for Acute Leukemia Increases the Risk of Graft-Versus-Host Disease and Death: A Study From the Acute Leukemia Working Party of the European Group for Blood and Marrow TransplantationFrom the Centre for Allogeneic Stem-Cell Transplantation and Division of Clinical Immunology, Huddinge University Hospital, Stockholm, Sweden; BA1638 Université Pierre et Marie Curie and Centre Claude Bernard, Centre International Greffes, Hôpital Saint-Antoine and European Group for Blood and Marrow Transplantation Data Center, Institut des Cordeliers; Hôpital Saint-Louis, Paris; Hôpital de Haut-Levèque, Pessac; Service des Maladie du Sang, Lille; Hôpital Henri Mondor, Creteil, France; University La Sapienza, Rome; Ospedale San Martino, Genova, Italy; and Bone Marrow Transplantation Centre, Leiden, the Netherlands. Address reprint requests to Olle Ringdén, MD, PhD, Karolinska Institute, Huddinge University Hospital, Centre for Allogeneic Stem Cell Transplantation, Division of Clinical Immunology, F79, SE-141 86 Stockholm, Sweden; e-mail: Olle.Ringden{at}labmed.ki.se
PURPOSE: Granulocyte colony-stimulating factor (G-CSF) is given after bone marrow transplantation (BMT) to shorten the neutropenic phase. Its effects have not been evaluated in a large patient population. PATIENTS AND METHODS: We studied 1,789 patients with acute leukemia receiving BMT and 434 patients receiving peripheral-blood stem cells (PBSCs) from HLA-identical siblings from 1992 to 2002 and reported the findings to the European Group for Blood and Marrow Transplantation. Among the BMT and PBSC patients, 501 (28%) and 175 (40%), respectively, were treated with G-CSF during the first 14 days after the transplantation. The outcome variables were entered into a Cox proportional hazards model. RESULTS: BMT and PBSC patients treated with G-CSF had a faster engraftment of absolute neutrophils greater than 0.5 x 109/L (P < .01), but platelet engraftment ( > 50 x 109/L) was slower (P < .001). In the BMT patients, acute graft-versus-host disease (GVHD) grades II to IV was 50% ± 5% (± 95% CI) in the G-CSF group versus 39% ± 3% in the controls (relative risk [RR], 1.33; P = .007, in the multivariate analysis). The incidence of chronic GVHD was also increased (RR, 1.29; P = .03). G-CSF was associated with an increase in transplantation-related mortality (TRM; RR, 1.73; P = .00016) and had no effect on relapse but reduced survival (RR, 0.59; P < .0001) and leukemia-free survival rates (LFS; RR, 0.64; P = .0003). No such effects of G-CSF were seen in patients receiving PBSC. CONCLUSION: After BMT, platelet engraftment was delayed, and GVHD and TRM were increased. Survival and LFS were reduced. This suggests that G-CSF should not be given shortly after BMT.
Bone marrow transplantation (BMT) is the treatment of choice for many patients with acute leukemia if an HLA-identical sibling donor is available [1-4]. The major problems after BMT include graft-versus-host disease (GVHD) and infections [1-3]. To reduce the risk of neutropenic infections, hematopoietic growth factors can be used to accelerate myeloid recovery and shorten the duration of posttransplant bone marrow aplasia [5,6]. After hematopoietic stem-cell transplantation, granulocyte colony-stimulating factor (G-CSF) has been shown to accelerate recovery of the absolute neutrophil count (ANC) significantly [7-12]. Current recommendations include the use of G-CSF [13]. Although most authors have found no statistically significant effect on acute and chronic GVHD, there seems to be some dispute about this [14]. A recent study showed a significant increase in the risk of acute GVHD when G-CSF was given after hematopoietic stem-cell transplantation [14]. The present analysis was performed to elucidate the clinical effects of G-CSF after BMT and peripheral-blood stem-cell (PBSC) transplantation in a larger group of patients with acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL).
The study consisted of 1,789 patients with acute leukemia receiving bone marrow and 434 receiving PBSCs from HLA-identical siblings. Transplantations were performed between January 1, 1992, and December 31, 2001, and reported to the Acute Leukemia Working Party of the European Group for Blood and Marrow Transplantation. The patients were studied by 155 European BMT teams. The study included patients who had or had not been treated with G-CSF as prophylaxis during the immediate posttransplantation period, starting before day 15. This includes patients treated with G-CSF as prophylaxis in some centers during a certain time period. The comparison is with patients who did not receive G-CSF at the same centers at other time periods and with patients from centers that did not use G-CSF prophylaxis at all. The median interval from transplantation to G-CSF therapy was 4 days (range, 0 to 14 days) in BMT patients and 7 days (range, 0 to 14 days) in the PBSC patients. The 501 BMT patients who had been given G-CSF shortly after receiving the transplant were compared with 1,288 patients who had not been treated. Among the PBSC recipients, 175 did receive and 259 did not receive G-CSF. Patients who were given G-CSF after day 14 were excluded from this comparison. G-CSF was given for graft failure after day 14 to 199 BMT and 16 PBSC recipients. These patients were analyzed separately. The patient characteristics of BMT and PBSC recipients who did or did not receive G-CSF as prophylaxis are listed in Table 1. Among the BMT patients, more of those in the G-CSF group had ALL, fewer were in first complete remission (CR1), more had younger age, more had younger donors, and fewer had received cyclosporine (CsA) combined with methotrexate (MTX) as GVHD prophylaxis. Among the PBSC recipients, patients in the G-CSF group had received a lower nucleated cell dose.
Definition of Outcomes The primary outcomes were as follows: transplantation-related mortality (TRM), defined as all causes of death not due to leukemia; relapse incidence, defined on the basis of morphologic evidence of leukemia in the bone marrow or other extramedullary organs (to evaluate the probability of relapse, patients dying from direct toxicity of the procedure or from causes not related to leukemia were censored); leukemia-free survival (LFS), defined as the time from the transplantation to the first event (relapse or death in complete remission); and overall survival. Other outcomes were hematopoietic recovery (neutrophil and platelet recoveries were analyzed separately and defined by a neutrophil count of  0.5 x 109/L for 3 consecutive days and nontransfused platelet count of 50 x109/L for 7 consecutive days, respectively) and GVHD. Acute GVHD was diagnosed and graded from 0 to IV at each transplant center, according to the Seattle criteria [1,2,5]. Chronic GVHD was defined using standard criteria [5]. Patients surviving without relapse for more than 100 days after transplantation with sustained donor engraftment were considered as assessable for chronic GVHD.
Statistical Analysis Statistical analyses were done independently for each end point (ie, neutrophil, platelet recoveries, GVHD, relapse, TRM, LFS, and survival rates). Incidences of each event were nonparametrically estimated. Patients were censored at the time of relapse or at the last follow-up [15]. Probability of survival and LFS were estimated by the product-limit method [16]. The significance of differences between curves was estimated by the log-rank test (Mantel-Cox). Then, all variables were included in the Cox proportional hazards model [17]. Recovery of cell subsets and development of chronic GVHD were events that compete with relapse or death. Relapse and nonrelapse mortality were events that compete with themselves. Therefore, estimations of incidence of these events relied on the nonparametric estimator of cumulative incidence curves, whereas predictive analyses were based on the proportional hazards model for these subdistributions of competing risks [18]. Such analyses were conducted using the cmprsk package (developed by B. Gray, June 2001) on S-Plus 2000 software (Statistical Sciences, Seattle, WA) and SPSS software (SPSS Inc, Chicago, IL).
Engraftment Median time to ANC greater than 0.5 x 109/L was 16 days (range, 2 to 55 days) for BMT patients treated with G-CSF, as compared with 20 days (range, 4 to 195 days) for the BMT patients not treated with G-CSF (P = .008; Fig 1A). Median time to platelets greater than 50 x 109/L was 31 days (range, 2 to 395 days) for the BMT patients treated with G-CSF, as compared with 28 days (range, 6 to 941 days) for the controls (P = .0002; Fig 1B). In the BMT patients treated with G-CSF, median time to platelets greater than 50 x 109/L was 32 days (range, 10 to 315 days) in those with acute GVHD, as compared with 29 days (range, 2 to 395 days) in those without acute GVHD (P = .05). In the controls, the corresponding values were 29 days (range, 11 to 94 days) in patients with acute GVHD and 27 days (range, 6 to 391 days) in those without acute GVHD (P = .003). In the multivariate analysis in the BMT patients, G-CSF treatment was associated with a faster ANC engraftment and a slower platelet engraftment (Table 2).
Among patients receiving PBSCs, time to ANC greater than 0.5 x 109/L was significantly faster (median, 12 v 15 days) using G-CSF (relative risk [RR], 1.49; P = .0035). However, time to platelets greater than 50 x 109/L was slower (18 v 15 days) in the G-CSF group (RR, 1.58; P = .0005).
Acute and Chronic GVHD
In the PBSC patients, the G-CSF group had a probability of acute GVHD grades II to IV of 33% ± 8% (± 95% CI) versus 35% ± 7% in the controls (P = .43). The cumulative incidence of chronic GVHD was 54% ± 8% and 53% ± 7% in the two groups, respectively (P = .58).
TRM, Relapse, Survival, and LFS in the BMT Patients
In BMT patients, G-CSF treatment had a similar effect on outcome in AML and ALL patients. For example, overall survival was significantly reduced in AML patients (P = .0008) and ALL patients (P = .0002) treated with G-CSF.
Outcome in Children Versus Adults
Causes of Death
TRM, Relapse, Survival, and LFS of Patients Receiving PBSCs Patients treated with G-CSF and given PBSC grafts did not differ significantly from those not given G-CSF in regard to TRM (19% ± 6% v 18% ± 6%; P = .71). At the 5-year follow-up, relapse was 33% ± 9% in PBSC patients treated with G-CSF versus 37% ± 9% in the controls (P = .43). The corresponding figures for survival were 52% ± 5% versus 47% ± 5% (P = .68). LFS was 48% ± 5% in the G-CSF group and 45% ± 5% in the controls (P = .62).
Outcome in Patients Treated With G-CSF for Graft Failure
Several findings were significant in this retrospective analysis of the role of G-CSF prophylaxis in patients with AML and ALL who received a BMT from HLA-identical sibling donors. Treatment with G-CSF had a significant effect on engraftment of ANC, platelets, acute and chronic GVHD, TRM, survival, and LFS rates. In the present study, the median initiation of G-CSF occurred on day 4 after transplantation. However, when G-CSF is used as prophylaxis to enhance engraftment of ANC, we found no difference in time to ANC greater than 0.5 x 109/L after BMT if G-CSF was started on day 0, day 5, or day 10 in prospective randomized studies [6,19]. It is well known that G-CSF enhances engraftment of ANC [5-12] (Fig 1, Table 2), which was seen after BMT and PBSC transplantation. However, this is offset by a delay in the time to reach platelets greater than 50 x 109/L. These findings confirm a recent study based on data in the International Bone Marrow Transplant Registry (IBMTR), in which G-CSF treatment was significantly associated with a delay in platelet engraftment [20]. Randomized studies of autologous PBSC transplantation in children have also shown that G-CSF had a detrimental effect on platelet recovery [21]. An improvement in ANC recovery and lower platelet counts has also been reported in a double-blind placebo-controlled trial in BMT patients, using granulocyte-macrophage colony-stimulating factor [22]. The delayed platelet reconstitution in patients treated with G-CSF may reflect an accelerated intravascular consumption of platelets. G-CSF increases platelet aggregation and promotes a hypercoagulable state in healthy PBSC donors [23]. During PBSC mobilization by G-CSF, the donors also develop thrombocytopenia [5,6]. Moreover, the increase in the incidence of GVHD may adversely affect platelet recovery. In this study, we also saw a delayed platelet recovery in patients with GVHD, regardless of whether they were treated with G-CSF. A striking and important finding was that G-CSF treatment increased the risk of acute GVHD in recipients of bone marrow but not of PBSC grafts (Fig 2; Table 3). This is not a center effect, because center was included in the multivariate analysis. Furthermore, a single-center study supported the finding of a significant increase in the risk of acute GVHD associated with G-CSF treatment [14]. In addition, a recent study by the IBMTR in children and adolescents undergoing bone marrow or PBSC transplantation for leukemia found that G-CSF posttransplantation was associated with TRM, poor survival, and poor LFS in multivariate analysis (Eapen et al, manuscript submitted for publication). However, in a randomized pediatric study, G-CSF did not increase the risk of GVHD after BMT [12]. In the present study, G-CSF increased the risk of GVHD and death in children and adults when analyzed separately. The differences between this and the IBMTR study on one hand, and the pediatric study on the other hand, may be due to chance and the few patients in the latter. In a study by Berger et al, in recipients of unrelated BMT, 11 (50%) of 22 patients given G-CSF developed grades II to IV acute GVHD, as compared with seven (28%) of 25 patients who were not given G-CSF (P = .14) [10]. The corresponding rates of chronic GVHD were 71% and 50% in the two groups, respectively. The figures were not statistically significant, which may be due to the small number of patients included. Kobayashi et al showed that the levels of soluble interleukin-2 receptor alfa were higher in patients who developed acute GVHD during administration of G-CSF than in those who developed it after the treatment had been stopped [24]. A higher level of soluble interleukin-2 receptor alfa in BMT patients treated with G-CSF has also been confirmed (M. Remberger, May 2003, personal communication). This would suggest that the administration of G-CSF may aggravate acute GVHD. We would like to highlight that the adverse effect of G-CSF posttransplantation was not seen in recipients of PBSC. The lack of an effect in the PBSC group does not seem to be primarily due to lower numbers and lack of power, because there does not seem to be any adverse trend. Several studies in animals and in humans suggest that G-CSF treatment of the donor decrease the risk of GVHD. G-CSF was shown to lower the production of tumor necrosis factor alpha, interleukin-2, and interferon g, which may be immunoregulatory rather than stimulatory [6]. Stem cells from donor mice treated with G-CSF showed a marked reduction in their ability to induce acute lethal GVHD [25]. Pan et al also reported that in mice, pretreatment of donors with G-CSF polarized donor T cells toward production of anti-inflammatory type-2 cytokine, which could reduce the severity of acute GVHD after transplantation [26]. In humans, G-CSF–stimulated PBSCs contain predominantly T-helper cell type 2 dendritic cells [27]. These data agree with the finding that G-CSF did not increase GVHD in recipients of PBSC grafts mobilized by G-CSF in the donor. A randomized study compared outcome after stem-cell transplantation in patients receiving PBSC or BMT from donors treated with G-CSF [28]. Those receiving BMT from G-CSF–treated donors had a significantly reduced severity of acute and chronic GVHD. This clinical study further demonstrates the decreased risk of GVHD if the donor is treated with G-CSF. There is a close correlation between acute and chronic GVHD [3,29]. Because acute GVHD occurred more frequently in BMT patients treated with G-CSF, chronic GVHD should also be more common (Fig 2; Table 3). In the present study, PBSC recipients treated with G-CSF did not have a greater risk of chronic GVHD. This may be due to the high incidence seen with PBSCs. Recipients of PBSCs have a higher incidence of chronic GVHD than those of bone marrow grafts, which may be due to the high T-cell dose in combination with the content of the PBSC grafts [6,30]. Because more recipients of bone marrow treated with G-CSF developed acute GVHD, they also had a higher TRM (Fig 3; Table 4). GVHD is one of the main reasons why patients die after BMT [1,2,5]. Schriber et al also observed a higher than expected early mortality rate in patients receiving G-CSF after BMT [7]. Treatment with G-CSF had no effect on relapse, which may seem surprising. Because of the graft-versus-leukemia effect of acute and chronic GVHD, such treatment should reduce the probability of relapse [31]. Because G-CSF adversely affected TRM in the BMT patients, it also had significant effects on the survival and LFS rates. Apart from the adverse effect of G-CSF in the BMT patients, the outcomes were worse in terms of TRM, survival, and LFS rates in adults than in children, patients with more advanced leukemia, in those not treated with CsA + MTX, and in various centers. These risk factors are well known from previous studies [3,30-32]. The strong center effect may be due to greater experience in some centers, a higher volume of patients treated, and better protocols and patient selection [32]. We included only patients in whom G-CSF was started within the first 14 days after transplantation and given as prophylaxis. Patients who were given G-CSF for graft failure were analyzed separately. Such patients were expected to have a poor prognosis and might introduce a bias in the study. One reason for graft failure may be weak donor alloreactivity. This could explain the low probability of grades II to IV acute GVHD in only 15% of these patients. This also suggests that G-CSF can be given for graft-failure without a high risk for GVHD. TRM, survival, and LFS rates were also encouraging in these patients with threatening graft failure. Because this is a retrospective analysis, the data should be interpreted with some caution. It compares centers who use G-CSF as prophylaxis with centers who do not use G-CSF. There may be several other differences between these centers that may affect outcome. However, center was included in the multivariate analysis and was found to be significant for time to ANC greater than 0.5 x 109/L, survival, and LFS (Tables 2 and 4). In contrast, the center effect was not significant for time to platelets greater than 50 x 109/L, GVHD, TRM, and relapse (Tables 2, 3, and 4). For all these outcomes, except for relapse, other risk factors, among them G-CSF treatment, were more important than the center effect in this analysis. Despite the limitations of the study, there is little reason to treat allogeneic BMT patients with G-CSF as prophylaxis after transplantation. Although engraftment of ANC is faster, this is offset by a delay in the engraftment of platelets. Moreover, the risk of GVHD, along with an increase in TRM and reduction in the survival and LFS rates, call into question the use of G-CSF as prophylaxis in BMT patients. Even in patients who receive PBSCs, the use of G-CSF after transplantation delays platelet engraftment. To conclude, G-CSF given as prophylaxis within 14 days after BMT to HLA-identical siblings with acute leukemia reduced the time to ANC greater than 0.5 x 109/L, delayed the time to platelets greater than 50 x 109/L, increased the risk of acute and chronic GVHD and TRM, and reduced survival and LFS rates.
The following teams reported data to this study: Hopital St Louis, Paris, France; University "La Sapienza," Rome, Italy; Hopital Haut-Leveque, Pessac, France; Leiden University Hospital, Leiden, the Netherlands; Hopital Claude Huriez, Lille, France; Hopital Henri Mondor, Creteil, France; Hotel Dieu, Paris, France; Hopital Necker, Paris, France; Huddinge University Hospital, Stockholm, Sweden; Hospital Universitario "Marques de Valdecilla," Santander, Spain; Hopital Jean Minjoz, Besancon; Hopital A. Michallon, Grenoble, France; Policlinico San Matteo, Pavia, Italy; Institut Paoli Calmettes, Marseille, France; Kantonspital, Basel, Switzerland; Hospital Santa Creu I Sant Pau, Barcelona, Spain; Medizinische Klinik und Poliklinik, Ulm, Germany; Hopital La Miletrie, Poitiers; Fédération de Greffe de Moelle de Thérapie Cellulaire d'Auvergne, Clermont-Ferrand, France; Children's Hospital, University of Helsinki, Helsinki; Turku University Central Hospital, Finland, Turku; Clinica di Oncoematologia Pediatrica, Padova, Italy; Pitie-Salpetriere, Paris, France; Cliniques Universitaires St Luc, Brussels, Belgium; The Oxford Radcliffe Hospital, Oxford, United Kingdom; Ospedale Maggiore di Milano, Istituto di Ricovero e Cura a Carattere Scientifico, Milano, Italy; University Hospital Gasthuisberg, Leuven, Belgium; Hospital Clinico Universitario, Valencia, Spain; Hopital Saint Antoine, Paris, France; Institut Jules Bordet, Brussels, Belgium; Royal Free Hospital and School of Medicine, London, United Kingdom; Belfast City Hospital, Belfast, United Kingdom; University Medical Centre, Utrecht, the Netherlands; Glasgow Royal Infirmary, Glasgow, United Kingdom; The National Hospital, Oslo, Norway; Centro Trapianti Midollo Osseo, Parma, Italy; University Medical Center St Radboud, Nijmegen, the Netherlands; Birmingham Heartlands Hospital, Birmingham, United Kingdom; Imperial College School of Medicine af the Hammersmith Hospital, London, United Kingdom; St James' University Hospital, Leeds, United Kingdom; Hopital Cantonal Universitaire, Geneva, Switzerland; Royal Victoria Infirmary, Newcastle on Tyne, United Kingdom; Patras University Medical School, Patras; University Hospital, Lund, Sweden; and Ospedale S. Camillo-Forlanini, Rome, Italy.
The authors indicated no potential conflicts of interest.
We thank the data managers from all centers for their help in collecting data, Virginie Chesnel and Patricia Palut from the European Group for Blood and Marrow Transplantation Central Data Office for their assistance with data management, Inger Hammarberg for excellent typing of the manuscript, and Drs Zoe and Francis P. Walsh for checking the language.
Supported by European Group for Blood and Marrow Transplantation funds and by convention 6113 from the Association pour la Recherche contre le Cancer, Villejuif, France, and grant BMH1-CT-94-0300 from the European Community. O.R. was supported by grants from the Swedish Cancer Society (0070-B99-13XAC), the Children's Cancer Foundation (2000/067), the Swedish Medical Research Council (K2000-06X-05971-20A), the Cancer Society in Stockholm, the Tobias Foundation, and Karolinska Institute. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Thomas ED, Storb R, Clift RA, et al: Bone-marrow transplantation: Part I. N Engl J Med 292:832-843, 1975[Medline] 2. Thomas ED, Storb R, Clift RA, et al: Bone-marrow transplantation: Part II. N Engl J Med 292:895-902, 1975[Medline] 3. Ringdén O: Allogeneic bone marrow transplantation for hematological malignancies: Controversies and recent advances. Acta Oncol 36:549-564, 1997[Medline] 4. Goldman JM, Schmitz N, Niethammer D, et al: Allogeneic and autologous transplantation for hematological diseases, solid tumours and immune disorders: Current practice in Europe in 1998—Accreditation Sub-Committee of the European Group for Blood and Marrow Transplantation. Bone Marrow Transplant 21:1-7, 1998[CrossRef][Medline] 5. Appelbaum FR: Allogeneic bone marrow transplantation and the use of hematopoietic growth factors. Stem Cells 13:344-350, 1995[Abstract] 6. Rutella S, Rumi C, Sica S, et al: Recombinant human granulocyte colony-stimulating factor (fHuG-CSF): Effects on lymphocyte phenotype and function. J Interferon Cytokine Res 19:989-994, 1999[CrossRef][Medline]
7. Schriber JR, Chao NJ, Long GD, et al: Granulocyte colony-stimulating factor after allogeneic bone marrow transplantation. Blood 84:1680-1684, 1994 8. Gisselbrecht C, Prentice HG, Bacigalupo A, et al: Placebo-controlled phase III trial of lenograstim in bone-marrow transplantation. Lancet 343:696-700, 1994[CrossRef][Medline] 9. Hirokawa M, Kitabayashi A, Niitsu H, et al: Modulation of allogeneic immune responses by filgrastim (recombinant human granulocyte colony-stimulating factor) in bone marrow transplantation. Int J Hematol 62:235-241, 1995[CrossRef][Medline] 10. Berger C, Bertz H, Schmoor C, et al: Influence of recombinant human granulocyte colony-stimulating factor (filgrastim) on hematopoietic recovery and outcome following allogeneic bone marrow transplantation (BMT) from volunteer unrelated donors. Bone Marrow Transplant 23:983-990, 1999[CrossRef][Medline]
11. Bishop MR, Tarantolo SR, Geller RB, et al: A randomized double-blind trial of filgrastim (granulocyte colony-stimulating factor) versus placebo following allogeneic blood stem cell transplantation. Blood 96:80-85, 2000
12. Dallorso S, Rondelli R, Messina C, et al: Clinical benefits of granulocyte colony-stimulating factor therapy after hematopoietic stem cell transplant in children: Results of a prospective randomized trial. Haematologica 87:1274-1280, 2002
13. American Society of Clinical Oncology: Update of recommendations for the use of hematopoietic colony-stimulating factors: Evidence-based clinical practice guidelines. J Clin Oncol 14:1957-1960, 1996 14. Remberger M, Naseh N, Aschan J, et al: G-CSF given after hematopoietic stem cell transplantation using HLA-identical sibling donors increases the risk of acute GVHD II-IV. Bone Marrow Transplant 32:217-223, 2003[CrossRef][Medline] 15. Clift R, Goldman JM, Gratwohl A, et al: Proposal for standardized reporting of bone marrow transplantation for leukemia. Bone Marrow Transplant 4:445-448, 1989[Medline] 16. Fine JP, Gray RJ: A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 94:496-509, 1999[CrossRef] 17. Kaplan EL, Meier P: Non parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958[CrossRef] 18. Cox DR: Regression models and life tables. J R Stat Soc B 34:187-202, 1972 19. Hägglund H, Ringdén O, Öhman S, et al: A prospective randomized trial of filgrastim (r-metHuG-CSF) given at different times after unrelated bone marrow transplantation. Bone Marrow Transplant 24:831-836, 1999[CrossRef][Medline] 20. Ringdén O, Barrett AJ, Zhang M-J, et al: Decreased treatment failure in recipients of HLA-identical bone marrow or peripheral blood stem cell transplants with high CD34 cell doses. Br J Haematol 121:874-885, 2003[CrossRef][Medline]
21. Kawano Y, Takaue Y, Mimaya J, et al: Marginal benefit/disadvantage of granulocyte colony-stimulating factor therapy after autologous blood stem cell transplantation in children: Results of a prospective randomized trial—The Japanese Cooperative Study Group of PBSCT. Blood 92:4040-4046, 1998 22. Powles R, Smith C, Milan S, et al: Human recombinant GM-CSF in allogeneic bone marrow transplantation for leukemia: Double-blind, placebo-controlled trial. Lancet 336:1417-1420, 1990[CrossRef][Medline] 23. Shimoda K, Okamura S, Inaba S, et al: Granulocyte colony-stimulating factor and platelet aggregation. Lancet 341:633, 1993[Medline] 24. Kobayashi S, Imamura M, Hashino S, et al: Possible role of granulocyte colony-stimulating factor in increased serum soluble interleukin-2-receptor-alpha levels after allogeneic bone marrow transplantation. Leuk Lymphoma 33:559-566, 1999[Medline]
25. Zeng D, Dejbakhsh-Jones S, Strober S: Granulocyte colony-stimulating factor reduces the capacity of blood mononuclear cells to induce graft-versus-host disease: Impact on blood progenitor cell transplantation. Blood 90:453-463, 1997
26. Pan L, Delmonte J Jr, Jalonen CK, et al: Pretreatment of donor mice with granulocyte colony-stimulating factor polarized donor T lymphocytes toward type-2 cytokine production and reduces severity of experimental graft-versus-host disease. Blood 86:4422-4449, 1995
27. Arpinatti M, Green CL, Heimfeld S, et al: Granulocyte-colony stimulating factor mobilizes T helper 2-inducing dendritic cells. Blood 95:2484-2490, 2000
28. Morton J, Hutchins C, Durrant S: Granulocyte-colony-stimulating factor (G-CSF)-primed allogeneic bone marrow: Significantly less graft-versus-host disease and comparable engraftment to G-CSF-mobilized peripheral blood stem cells. Blood 98:3186-3193, 2001 29. Storb R, Prentice RL, Sullivan KM, 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 98:461-466, 1983
30. Ringdén O, Labopin M, Bacigalupo A, et al: Transplantation of peripheral blood stem cells as compared with bone marrow from HLA-identical siblings in patients with acute myeloid leukemia and acute lymphoblastic leukemia. J Clin Oncol 20:4655-4664, 2002
31. Horowitz MM, Gale RP, Sondel PM, et al: Graft-versus-leukemia reactions following bone marrow transplantation in humans. Blood 75:555-562, 1990 32. Frassoni F, Labopin M, Powles R, et al: Effect of centre on outcome of bone-marrow transplantation for acute myeloid leukemia. Lancet 355:1393-1398, 2000[CrossRef][Medline] Submitted June 23, 2003; accepted September 25, 2003. Related Editorial
Related Correspondence
Related Reply
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
2 test for categoric variables and the nonparametric Mann-Whitney U test for continuous variables. 
