Volume 11, Issue 11 , Pages 881-889, November 2005
Effect of Conditioning Regimen on the Outcome of Bone Marrow Transplantation from an Unrelated Donor
Article Outline
Abstract
Little information is available regarding the effect of the conditioning regimen on the outcome of bone marrow transplantation (BMT) from an unrelated donor. Therefore, we retrospectively compared the outcome after a cyclophosphamide/total body irradiation (Cy-TBI) regimen, an intensified Cy-TBI regimen (Cy-TBI+), a busulfan and cyclophosphamide (Bu-Cy) regimen, and a Bu-Cy regimen with total lymphoid irradiation (Bu-Cy-TLI). Clinical data of 1875 adult patients who underwent unmanipulated unrelated BMT for leukemia or myelodysplastic syndrome by using 1 of the 4 regimens between 1993 and 2002 were extracted from the database of the Japan Marrow Donor Program. The effect of the conditioning regimen was adjusted for other independent significant factors by multivariate analyses. The Cy-TBI regimen was significantly better than the Bu-Cy regimen with regard to the incidence of engraftment failure (odds ratio, 2.49; P = .046) and overall survival (relative risk [RR], 1.31; P = .050). The Bu-Cy-TLI regimen decreased relapse (RR, 0.13; P = .039) but increased nonrelapse mortality (RR, 1.89; P = .0061). The Cy-TBI+ regimen resulted in increased nonrelapse mortality (RR, 1.48; P = .0003) and inferior survival (RR, 1.45; P < .0001). The results of this retrospective study suggested that the Cy-TBI regimen was superior to other regimens in unrelated BMT.
Key words: Conditioning regimen , Unrelated bone marrow transplantation , Cyclophosphamide , Busulfan , Total body irradiation
Introduction
The conditioning regimen before allogeneic hematopoietic stem cell transplantation (HSCT) is intended to eradicate tumor cells and to promote immunosuppression to prevent graft rejection. Successful bone marrow transplantation (BMT) with a combination of cyclophosphamide (Cy) and total body irradiation (TBI) was reported in the 1970s [1, 2]. TBI is effective against a variety of malignancies without sanctuary sites, such as the central nervous system and testicles. However, there has been concern regarding long-term sequelae, including cataracts, second malignancies, and development problems in children. Thus, non-TBI regimens have been investigated by substituting busulfan (Bu) for TBI [3, 4]. Accordingly, the Cy-TBI and Bu-Cy regimens have been regarded as the standard conditioning regimens since the 1980s.
Four randomized controlled trials have been performed to compare the Cy-TBI and Bu-Cy regimens in HSCT from an HLA-identical sibling donor, but they gave conflicting results with regard to both survival and toxicities [5, 6, 7, 8]. Therefore, Hartman et al. [9] conducted a meta-analysis of these randomized controlled trials and 1 other trial that compared the etoposide/TBI regimen and the Bu-Cy regimen. They showed a significantly lower incidence of hepatic veno-occlusive disease (VOD) after the TBI-based regimen than the Bu-Cy regimen and a trend toward better survival after the TBI-based regimen (P = .09). Recently, Socié et al. [10] updated the 4 randomized controlled trials that compared the Cy-TBI and Bu-Cy regimens, with a mean follow-up for surviving patients of >7 years. Although the Cy-TBI and Bu-Cy regimens were associated with similar survival in patients with chronic myelocytic leukemia (CML), a nonsignificant (P = .068) 10% lower survival rate was observed after the Bu-Cy regimen in patients with acute myeloblastic leukemia (AML).
The feasibility of the Bu-Cy regimen in unrelated HSCT has been shown in several studies, but the incidence of engraftment failure ranged up to 12% [11, 12]. The number of patients in these studies was small, and there has been no randomized controlled trial to compare conditioning regimens in unrelated BMT. Therefore, we retrospectively compared the Cy-TBI and Bu-Cy regimens in a large series of patients who underwent unrelated BMT in Japan. We also evaluated the efficacy of total lymphoid irradiation (TLI), which was added to the Bu-Cy regimen to prevent engraftment failure. Another object of this study was to evaluate the effect of an intensified conditioning regimen in which antineoplastic agents other than Cy, such as cytarabine, etoposide, and Bu, were added to the Cy-TBI regimen [13, 14, 15].
Patients and methods
Study Population and Transplantation Procedure
A total of 3543 patients who underwent allogeneic BMT from an unrelated donor between 1993 and 2002 for CML, AML, acute lymphoblastic leukemia (ALL), or myelodysplastic syndrome (MDS) were reported to the Japan Marrow Donor Program [16, 17, 18]. Those <16 years of age, those who received a manipulated graft, and those who received antithymocyte globulin or alemtuzumab as a part of their conditioning regimen were excluded from the study. The conditioning regimens before transplantation were classified into the following groups. The Cy-TBI regimen was defined as the combination of Cy and TBI only. The total dose of Cy was between 100 and 150 mg/kg. The total dose of TBI was between 10 and 15 Gy. Cy-TBI+ regimens were defined as those that included another antineoplastic agent added to the Cy-TBI regimen. The added agent was cytarabine in 61%, etoposide in 14%, and Bu in 24% of cases. The Bu-Cy regimen was defined as the combination of Bu and Cy. The total dose of Cy was between 100 and 150 mg/kg. The total dose of Bu was 16 mg/kg in most patients. The Bu-Cy-TLI regimen was the combination of Bu-Cy and TLI. TLI was typically performed at 5 to 8 Gy in 1 or 2 fractions. Patients who received a conditioning regimen that did not belong to these groups were excluded from the analysis. Finally, 1875 patients were included in the study.
The conditioning regimen was chosen at the discretion of each center. Bone marrow was exclusively used as a stem cell source. Prophylaxis for graft-versus-host disease (GVHD) mainly consisted of a combination of cyclosporin A and methotrexate (60%) or a combination of tacrolimus and methotrexate (32%).
Statistical Considerations
Data were collected by the Japan Marrow Donor Program by using a standardized report form. Follow-up reports were submitted at 100 days, 1 year, and annually after transplantation. Data for August 2003 were used in the following analyses. The primary end point was survival after transplantation. The incidences of engraftment failure and grade III/IV acute GVHD, which was graded according to the published criteria [19], were secondary end points. Engraftment was defined as a neutrophil count >500/μL for 3 consecutive days after transplantation. Engraftment failure was diagnosed when engraftment was not achieved at any time after transplantation. The incidences of secondary graft failure, defined as persistent neutropenia after engraftment and acute GVHD, were analyzed in 1744 patients who achieved initial engraftment.
The probability of survival and the cumulative incidence of acute GVHD were calculated with the Kaplan-Meier method. The cumulative incidence of relapse was calculated by Gray’s method by considering death without relapse as a competing risk [20]. Univariate comparison for dichotomous variables between groups was performed with the Fisher exact test or the χ2 test, and comparisons for time-to-event variables were performed with the log-rank test. Univariate analyses to compare the type of conditioning regimen were performed to test the null hypothesis that the effects of each conditioning regimen were the same. Multivariate analyses for dichotomous and time-to-event variables were performed by using logistic regression analysis and proportional hazards modeling, respectively. Potential confounding factors considered in the analysis included recipient age, recipient sex, donor age, donor sex, underlying disease, disease status, serologic/genotypic HLA mismatch, ABO mismatch, cytomegalovirus serostatus, conditioning regimen, cell dose in the graft, GVHD prophylaxis regimen, and the use of granulocyte colony-stimulating factor (G-CSF). Acute leukemia in first or second remission, CML in first or second chronic phase, and MDS without leukemic transformation were considered standard-risk diseases, whereas others were considered high-risk diseases [21]. An HLA mismatch in the graft-versus-host (GVH) direction was defined as when the recipient’s antigens or alleles were not shared by the donor, whereas mismatch in the host-versus-graft (HVG) direction was defined as when the donor’s antigens or alleles were not shared by the recipient. HLA-allele mismatch included the presence of HLA mismatch at both the antigen and allele levels. Factors other than the type of conditioning regimen that showed at least borderline significance (P < .10) in univariate analyses were included in the multivariate analyses and then deleted stepwise from the model, except that underlying disease was consistently kept in the model. The type of conditioning regimen was added in the final model.
Results
Characteristics of the Patients
The median age of the 1875 eligible patients was 33 years (range, 16-63 years). The number of patients who received the Cy-TBI, Cy-TBI+, Bu-Cy, and Bu-Cy-TLI regimens was 714, 861, 243, and 57, respectively (Table 1). A significant difference in the patients’ background characteristics was observed with regard to recipient age, diagnosis, disease risk category, GVHD prophylaxis regimen, and the use of G-CSF after transplantation. The Cy-TBI+ regimen tended to be used in younger patients and high-risk patients. The use of non-TBI regimens was less frequent in ALL. The lower incidence of FK506-based GVHD prophylaxis and posttransplant G-CSF in patients who received the Bu-Cy-TLI regimen was probably due to each center’s policy.
Table 1. Characteristics of the Patients (N = 1875)
| Variable | Cy-TBI (n = 714) | Cy-TBI+ (n = 861) | Bu-Cy (n = 243) | Bu-Cy-TLI (n = 57) | P Value |
|---|---|---|---|---|---|
| Recipient sex male | 63% | 62% | 54% | 54% | .069 |
| Recipient age ≥40 y | 35% | 27% | 37% | 44% | .0005 |
| Donor sex male | 62% | 63% | 61% | 56% | .67 |
| Donor age ≥40 y | 26% | 28% | 30% | 25% | .56 |
| HLA-antigen mismatch in HVG direction | 4.6% | 4.3% | 4.8% | 2.0% | .85 |
| HLA-antigen mismatch in GVH direction | 3.0% | 3.7% | 2.7% | 2.0% | .82 |
| HLA-allele mismatch in HVG direction | 35% | 37% | 30% | 24% | .13 |
| HLA-allele mismatch in GVH direction | 35% | 37% | 31% | 29% | .44 |
| ABO major mismatch | 27% | 28% | 27% | 27% | .98 |
| ABO minor mismatch | 23% | 25% | 21% | 30% | .41 |
| Diagnosis | |||||
 ALL | 24% | 33% | 10% | 0% | <.0001 |
| AML | 30% | 31% | 32% | 37% | |
| CML | 32% | 26% | 44% | 53% | |
| MDS | 15% | 10% | 14% | 11% | |
| High-risk category | 17% | 34% | 20% | 16% | <.0001 |
| CMV serostatus positive | 80% | 80% | 81% | 76% | .81 |
| Cell dose in the graft ≥3.0 × 108 cells/kg | 46% | 53% | 49% | 46% | .047 |
| GVHD prophylaxis: tacrolimus + methotrexate | 35% | 37% | 32% | 9% | .0004 |
| G-CSF used | 85% | 86% | 86% | 51% | <.0001 |
Engraftment Failure
Engraftment failure was observed in 65 patients (3.6%). Univariate analysis identified 5 risk factors that affected the incidence of engraftment failure with a P value of <.10: higher recipient age, HLA-allele mismatch in the HVG direction, ABO major mismatch, high-risk disease, and low cell dose in the graft. By a multivariate analysis, all of these factors except for ABO major mismatch were identified as independent significant risk factors for engraftment failure (Table 2). When we added the type of conditioning regimen to this model, the use of a Bu-Cy regimen significantly increased the incidence of engraftment failure (odds ratio [OR], 2.49; 95% confidence interval [CI], 1.02-6.13; P = .046). There was no significant difference in the time to engraftment among the conditioning regimen groups (P = .26; Figure 1A).
Table 2. Multivariate Analyses for Engraftment Failure, Grade III/IV Acute GVHD, and Overall Survival before and after Adding the Type of Conditioning Regimen to the Model
| Factor | Before | After | ||
|---|---|---|---|---|
| OR (95% CI) | P Value | OR (95% CI) | P Value | |
| Engraftment failure | ||||
| Recipient age | ||||
| <40 y | 1.00 | 1.00 | ||
| ≥40 y | 2.00(1.05-3.80) | .035 | 2.02(1.06-3.88) | .032 |
| HLA-allele mismatch in HVG direction | ||||
| No | 1.00 | 1.00 | ||
| Yes | 3.36(1.74-6.47) | .0003 | 3.33(1.72-6.45) | .0003 |
| Risk category | ||||
| Standard | 1.00 | 1.00 | ||
| High | 3.10(1.50-6.41) | .0023 | 3.40(1.61-7.14) | .0014 |
| Cell dose | ||||
| <3.0 × 108 cells/kg | 1.00 | 1.00 | ||
| ≥3.0 × 108 cells/kg | 0.36(0.18-0.71) | .0036 | 0.36(0.18-0.73) | .0046 |
| Diagnosis | ||||
| ALL | 1.00 | |||
| AML | 0.42(0.16-1.06) | .066 | 0.38(0.15-0.98) | .045 |
| CML | 0.58(0.24-1.36) | .21 | 0.50(0.21-1.21) | .12 |
| MDS | 2.51(0.95-6.62) | .063 | 2.33(0.87-6.25) | .094 |
| Regimen | ||||
| Cy-TBI | 1.00 | |||
| Cy-TBI+ | 0.87(0.41-1.83) | .71 | ||
| Bu-Cy | 2.49(1.02-6.13) | .046 | ||
| Bu-Cy-TLI | 0.00 | .98 | ||
| RR (95% CI) | P Value | RR (95% CI) | P Value | |
| Grade III/IV acute GVHD | ||||
| HLA allele mismatch in GVH direction | ||||
| No | 1.00 | 1.00 | ||
| Yes | 1.95(1.47-2.57) | <.0001 | 1.96(1.48-2.59) | <.0001 |
| ABO-minor mismatch | ||||
| No | 1.00 | 1.00 | ||
| Yes | 1.36(1.01-1.82) | .045 | 1.36(1.01-1.83) | .043 |
| GVHD prophylaxis | ||||
| Cyclosporine + methotrexate | 1.00 | 1.00 | ||
| Tacrolimus + methotrexate | 0.53(0.38-0.74) | <.0002 | 0.53(0.38-0.74) | .0002 |
| Diagnosis | ||||
| ALL | 1.00 | 1.00 | ||
| AML | 0.86(0.55-1.32) | .48 | 0.86(0.55-1.33) | .49 |
| CML | 1.86(1.29-2.67) | .0009 | 1.85(1.27-2.68) | .0014 |
| MDS | 1.41(0.84-2.38) | .19 | 1.44(0.85-2.44) | .17 |
| Regimen | ||||
| Cy-TBI | 1.00 | |||
| Cy-TBI+ | 1.19(0.87-1.63) | .29 | ||
| Bu-Cy | 1.28(0.83-1.98) | .27 | ||
| Bu-Cy-TLI | 1.18(0.58-2.38) | .65 | ||
| RR (95% CI) | P Value | RR (95% CI) | P Value | |
| Overall survival | ||||
| Recipient age | ||||
| <40 y | 1.00 | 1.00 | ||
| ≥40 y | 1.50(1.27-1.77) | <.0001 | 1.54(1.31-1.83) | <.0001 |
| Donor age | ||||
| <40 y | 1.00 | 1.00 | ||
| ≥40 y | 1.20(1.01-1.42) | .036 | 1.17(0.99-1.38) | .074 |
| HLA-allele mismatch in GVH direction | ||||
| No | 1.00 | 1.00 | ||
| Yes | 1.55(1.32-1.82) | <.0001 | 1.56(1.33-1.83) | <.0001 |
| G-CSF | ||||
| No | 1.00 | |||
| Yes | 1.30(1.04-1.64) | .024 | 1.32(1.05-1.67) | .020 |
| Risk category | ||||
| Standard | 1.00 | 1.00 | ||
| High | 2.48(2.09-2.93) | <.0001 | 2.31(1.95-2.76) | <.0001 |
| GVHD prophylaxis | ||||
| Cyclosporine + methotrexate | 1.00 | 1.00 | ||
| Tacrolimus + methotrexate | 0.73(0.62-0.87) | .0005 | 0.72(0.60-0.86) | .0003 |
| Diagnosis | ||||
| ALL | 1.00 | 1.00 | ||
| AML | 0.86(0.69-1.07) | .18 | 0.88(0.70-1.09) | .24 |
| CML | 0.82(0.66-1.01) | .063 | 0.84(0.68-1.04) | .11 |
| MDS | 1.24(0.93-1.66) | .13 | 1.30(0.98-1.74) | .070 |
| Regimen | ||||
| Cy-TBI | 1.00 | |||
| Cy-TBI+ | 1.45(1.20-1.74) | <.0001 | ||
| Bu-Cy | 1.31(1.00-1.73) | .050 | ||
| Bu-Cy-TLI | 1.43(0.91-2.26) | .12 | ||
Figure 1.
Days to engraftment (A) and days to grade III/IV acute GVHD (B) grouped according to the type of conditioning regimen.
Secondary graft failure was observed in 1.6% of patients who achieved initial engraftment. Logistic regression analysis revealed that only higher donor age was an independent significant risk factor for secondary graft failure (OR, 2.38; 95% CI, 1.07-5.29; P = .034). The type of conditioning did not significantly affect the incidence of secondary graft failure.
Acute and Chronic GVHD
The incidence of grade II to IV and grade III/IV acute GVHD was 43.9% and 16.7%, respectively. Male sex, higher donor age, HLA mismatch in the GVH direction, ABO minor mismatch, underlying disease, high-risk disease, and the GVHD prophylaxis regimen affected the incidence of grade III/IV acute GVHD with at least borderline significance (P < .10). Among these, HLA-allele mismatch in the GVH direction, ABO minor mismatch, underlying disease, and GVHD prophylaxis were identified as independent risk factors by a multivariate analysis (Table 2). There was no difference in the incidence of acute GVHD among the 4 types of conditioning regimens after adjustment for these risk factors (Table 2 and Figure 1B).
Chronic GVHD was observed in 49.7% of patients who achieved engraftment and survived disease free for at least 100 days after transplantation. Only the presence of an HLA-allele mismatch in the GVH direction significantly affected the incidence of chronic GVHD by multivariate analysis. The type of conditioning did not significantly affect the incidence of chronic GVHD.
Survival after Transplantation
Overall survival and disease-free survival at 5 years after transplantation for all of the patients was 46.2% and 42.5%, respectively. Overall survival stratified by disease status, grouped according to the conditioning regimen, is shown in Figure 2. A significant difference in survival was observed in standard-risk patients. Risk factors for shorter survival with a P value of <.10 identified by the log-rank test included male sex, higher recipient age, higher donor age, HLA mismatch in both the GVH and HVG directions, ABO major mismatch, high-risk disease, cytomegalovirus seropositivity, use of G-CSF after transplantation, and GVHD prophylaxis consisting of cyclosporin A and methotrexate. Proportional hazard modeling identified 6 independent significant risk factors: higher patient age, higher donor age, HLA-allele mismatch in the GVH direction, use of G-CSF, high-risk disease, and the use of the combination of cyclosporin A and methotrexate (Table 2). When we added the type of conditioning regimen to the proportional hazard model, the Cy-TBI+ and Bu-Cy regimens were significantly inferior to the Cy-TBI regimen (relative risk [RR], 1.45; 95% CI, 1.20-1.74; P < .0001 and RR, 1.31; 95% CI, 1.00-1.73; P = .050, respectively).
Figure 2.
Overall survival grouped according to the type of conditioning regimen in standard-risk (A) and high-risk (B) patients.
Analyses Based on Detailed HLA Matching
We added analyses based on detailed HLA matching because it has been reported that the outcome of unrelated BMT is affected not only by the presence of HLA-allele mismatch, but also by whether the HLA-allele mismatch belongs to class I or class II [16, 18]. In this study, none of the HLA-A/-B antigen, HLA-C antigen, HLA-DR antigen, HLA-A/-B allele, HLA-C allele, or HLA-DRB1 allele mismatches in the HVG direction significantly affected the incidence of engraftment failure, probably because of the small number of patients in each group. However, mismatches in the GVH direction at the HLA-A/-B antigen, HLA-C antigen, HLA-A/-B allele, HLA-C allele, and HLA-DRB1 allele significantly affected the incidence of grade III/IV acute GVHD in univariate analyses. These factors were included in the multivariate analysis, and HLA-A/-B allele, HLA-C allele, and HLA-DRB1 allele mismatches were shown to be independently significant. However, the effect of the conditioning regimen on the incidence of grade III/IV acute GVHD was not significant after adjustment for the independent significant factors. As for survival after transplantation, mismatches in both the HVG and GVH directions at the HLA-A/-B antigen, HLA-C antigen, HLA-A/-B allele, HLA-C allele, and HLA-DRB1 allele significantly affected overall survival in univariate analyses. Among these, the presence of an HLA-A/-B antigen mismatch in the HVG direction and an HLA-A/-B allele mismatch in the GVH direction were identified as independent significant risk factors for overall survival. After adjustment for these factors, as well as other independent significant risk factors, the adverse effects of the Cy-TBI+ and Bu-Cy regimens remained significant (RR, 1.42; 95% CI, 1.18-1.70; P = .0002 and RR, 1.31; 95% CI, 1.00-1.72; P = .052, respectively).
Other Statistical Analyses to Ensure the Results
We added statistical analyses to ensure the findings of this study. First, we repeated the analyses by using only patients who received the Cy-TBI or Bu-Cy regimen, to confirm the difference between the 2 regimens. The findings were almost the same, and the use of Bu-Cy adversely affected the incidence of engraftment failure and overall survival (OR, 2.53; 95% CI, 1.00-6.39; P = .049 and RR, 1.32; 95% CI, 1.00-1.75; P = .053, respectively).
Next, we changed the method of the multivariate analyses to include all factors with at least borderline significance (P < .10) in univariate analyses, as well as the underlying disease and the type of conditioning regimen, followed by a stepwise deletion of nonsignificant factors. This change in the statistical method did not change the major findings of this study. The Bu-Cy regimen was inferior to the Cy-TBI regimen in the incidence of engraftment failure and overall survival (OR, 2.49; 95% CI, 1.02-6.12; P = .045 and RR, 1.33; 95% CI, 1.02-1.75; P = .046, respectively). The Cy-TBI+ regimen was inferior to the Cy-TBI regimen in overall survival (RR, 1.46; 95% CI, 1.21-1.75; P < .0001).
Relapse and Nonrelapse Mortality
To evaluate the cause of the difference in survival among the different types of conditioning regimens, we further analyzed the incidences of relapse and nonrelapse mortality. Multivariate analyses revealed that the incidence of relapse after the Bu-Cy-TLI regimen was significantly lower than that after the Cy-TBI regimen (RR, 0.13; 95% CI, 0.02-0.90; P = .039, adjusted for ABO major mismatch, underlying disease, disease status, and GVHD prophylaxis), although this benefit was offset by a significant increase in the incidence of nonrelapse mortality (RR, 1.89; 95% CI, 1.20-3.00; P = .0061, adjusted for recipient age, donor age, underlying disease, disease status, HLA-allele mismatch in the HVG direction, G-CSF, and GVHD prophylaxis); this resulted in similar survival. The incidence of nonrelapse mortality after the Cy-TBI+ regimen was significantly higher than that after the Cy-TBI regimen (RR, 1.48; 95% CI, 1.20-1.84; P = .0003, adjusted as described previously), whereas there was no difference in the incidence of relapse (RR, 0.84; 95% CI, 0.64-1.11; P = .22). There was no significant difference in the incidence of relapse and nonrelapse mortality between the Cy-TBI and Bu-Cy regimens (RR, 0.89; 95% CI, 0.57-1.38; P = .59 and RR, 1.21; 95% CI, 0.89-1.65; P = .23, respectively).
Other Complications after Transplantation
The incidence of interstitial pneumonitis was significantly different among the 4 conditioning regimens (P = .019; Figure 3). The incidence of interstitial pneumonitis after the Cy-TBI+ regimen was significantly higher than that after the Cy-TBI regimen (OR, 1.59; 95% CI, 1.13-2.23; P = .0076, adjusted for underlying disease, HLA-allele mismatch in the HVG direction, and GVHD prophylaxis). A statistically significant difference was not observed between the Cy-TBI and Bu-Cy regimens (P = .66). The incidence of VOD was also significantly different among the 4 conditioning groups (P = .0049). It was significantly higher after the Cy-TBI+, Bu-Cy, and Bu-Cy-TLI regimens than after the Cy-TBI regimen (OR, 1.64; 95% CI, 1.00-2.71; P = .052; OR, 3.00; 95% CI, 1.62-5.45; P = .0005; and OR, 3.20; 95% CI, 1.11-8.24; P = .032, respectively, adjusted for underlying disease, HLA-allele mismatch in the HVG direction, ABO major mismatch, ABO minor mismatch, and G-CSF). The incidence of hemorrhagic cystitis was significantly affected by the type of conditioning regimen (P = .0003). It was also significantly higher after the Cy-TBI+, Bu-Cy, and Bu-Cy-TLI regimens than after the Cy-TBI regimen (OR, 1.37; 95% CI, 1.09-1.72; P = .0075; OR, 1.85; 95% CI, 1.34-2.56; P = .0002; and OR, 2.11; 95% CI, 1.16-3.85; P = .015, respectively, adjusted for underlying disease and donor sex).
Figure 3.
Incidence of interstitial pneumonitis, hepatic veno-occlusive disease, and secondary malignancies, excluding posttransplantation lymphoproliferative disorders.
Secondary malignancies excluding posttransplantation lymphoproliferative disorders developed in 8 patients a median of 35 months (range, 15-84 months) after transplantation, including MDS in 2 and AML, thyroid cancer, uterine body cancer, esophageal cancer, breast cancer, and squamous cell cancer in 1 each. The incidence of secondary malignancies was not significantly different among the 4 conditioning groups.
Discussion
In this study, we retrospectively evaluated the effect of the conditioning regimen on the outcome of unrelated BMT. The Cy-TBI regimen was superior to the Bu-Cy regimen, not only with regard to the incidence of engraftment failure, but also for overall survival after transplantation. The addition of TLI to the Bu-Cy regimen decreased the incidences of engraftment failure and relapse but increased nonrelapse mortality. Intensified conditioning regimens in which another antineoplastic agent was added to the Cy-TBI regimen resulted in increased nonrelapse mortality and inferior survival.
On the basis of the results of randomized controlled trials and their meta-analysis, the Cy-TBI regimen is generally preferred to the Bu-Cy regimen except for patients with CML in chronic phase in HSCT from an HLA-identical sibling donor [5, 6, 7, 8, 9, 10]. This study showed that Cy-TBI may be the first-choice regimen in most patients who undergo unrelated BMT unless the patient has a condition that precludes the use of TBI, such as previous high-dose irradiation to a major organ. The weakness of the Bu-Cy regimen was apparent in the increased incidences of engraftment failure and VOD. As a current general practice in Japan, Bu is administered orally without monitoring the plasma concentration. Therefore, the use of intravenous Bu or oral Bu targeted to a predetermined plasma level may improve the outcome after the Bu-Cy regimen [22]. However, further trials are required to evaluate the efficacy of intravenous Bu and targeted oral Bu.
Higher nonrelapse mortality after the intensified Cy-TBI+ regimen might reflect the possibility that the regimen was preferentially used in patients with advanced diseases. However, the incidence of nonrelapse mortality was significantly higher after adjustment for disease status and also when the comparison was limited to patients with standard-risk disease (RR, 1.47; 95% CI, 1.14-1.90; P = .0031). Conversely, a decrease in the relapse incidence was not observed either in standard-risk or in high-risk patients (RR, 0.81; 95% CI, 0.57-1.15; P = .24 and RR, 0.89; 95% CI, 0.57-1.39; P = .60). Therefore, these results did not show any benefit for the intensified regimens.
This was a retrospective study, and it was impossible to completely eradicate biases. First, non-TBI regimens were preferentially used in older patients. Second, the use of Bu-based regimens was less frequent in ALL compared with myeloid malignancies. Third, the intensified Cy-TBI+ regimen was most frequently used in young patients with high-risk diseases. Therefore, we adjusted the effect of the conditioning regimen for these variables in multivariate analyses. We should also consider the “center” effect as a possible bias. However, a study from the Japan Society for HSCT did not show a significant center effect in unrelated BMT in Japan [23]. The inclusion of patients who underwent transplantation from 1993 and 2002 might have resulted in the significant variations in transplantation procedures. We could not obtain detailed information of supportive care, and this is one of the limitations of this type of registry data study.
The use of G-CSF after transplantation significantly adversely affected survival. A similar result was observed in a retrospective study by the European Group for Blood and Marrow Transplantation [24]. However, such an adverse effect has not been shown in prospective randomized controlled trials that evaluated the use of G-CSF after transplantation [25]. Patients with preexisting infections or other comorbidities might have tended to receive G-CSF. These data were not included in the analyses and thus might have biased the results.
Although a definite conclusion cannot be made without a randomized controlled trial, >1000 patients will be required to detect the meaningful difference (RR, 1.31) in survival between the Cy-TBI and Bu-Cy groups that was seen in this study at a statistically significant level with α and β errors of 5% and 20%, respectively. Thus, realistically, this retrospective study that considered possible biases in multivariate analyses may be the best evidence. More than 30 years have passed since the introduction of the Cy-TBI regimen. Nevertheless, the Cy-TBI regimen still seems to be the most suitable regimen not only in HSCT from an HLA-identical sibling donor, but also in unrelated BMT.
References
- Bone-marrow transplantation for hematologic neoplasia in 16 patients with identical twins . N Engl J Med . 1974;290:1389–1393
- Bone-marrow transplantation (second of two parts) . N Engl J Med . 1975;292:895–902
- . Bone marrow transplantation for leukemia following a new busulfan and cyclophosphamide regimen . Blood . 1987;70:1382–1388
- Marrow transplantation for acute nonlymphocytic leukemia after treatment with busulfan and cyclophosphamide . N Engl J Med . 1983;309:1347–1353
- Allogeneic bone marrow transplantation for acute myeloid leukemia in first remission: a randomized trial of a busulfan-Cytoxan versus Cytoxan-total body irradiation as preparative regimen: a report from the Group d’Etudes de la Greffe de Moelle Osseuse . Blood . 1992;79:2578–2582
- Marrow transplantation for chronic myeloid leukemia (a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide) . Blood . 1994;84:2036–2043
- Allogeneic bone marrow transplantation for chronic myeloid leukemia in first chronic phase: a randomized trial of busulfan-cytoxan versus cytoxan-total body irradiation as preparative regimen: a report from the French Society of Bone Marrow Graft (SFGM) . Blood . 1995;85:2263–2268
- A randomized trial comparing busulfan with total body irradiation as conditioning in allogeneic marrow transplant recipients with leukemia (a report from the Nordic Bone Marrow Transplantation Group) . Blood . 1994;83:2723–2730
- . Survival, disease-free survival and adverse effects of conditioning for allogeneic bone marrow transplantation with busulfan/cyclophosphamide vs total body irradiation (a meta-analysis) . Bone Marrow Transplant . 1998;22:439–443
- Busulfan plus cyclophosphamide compared with total-body irradiation plus cyclophosphamide before marrow transplantation for myeloid leukemia (long-term follow-up of 4 randomized studies) . Blood . 2001;98:3569–3574
- Unrelated allogeneic bone marrow transplantation using high-dose busulfan and cyclophosphamide (BU-CY) for the preparative regimen . Bone Marrow Transplant . 1996;17:685–689
- . Busulfan/cyclophosphamide in volunteer unrelated donor (VUD) BMT (excellent feasibility and low incidence of treatment-related toxicity) . Bone Marrow Transplant . 1997;19:1169–1173
- The Stanford experience with high-dose etoposide cytoreductive regimens and autologous bone marrow transplantation in Hodgkin’s disease and non-Hodgkin’s lymphoma (preliminary data) . Ann Oncol . 1991;2(suppl 1):47–50
- High-dose cytarabine and total body irradiation with or without cyclophosphamide as a preparative regimen for marrow transplantation for acute leukemia . J Clin Oncol . 1988;6:576–582
- High doses of cyclophosphamide, etoposide and total body irradiation followed by autologous stem cell transplantation in the management of patients with chronic myelogenous leukemia . Bone Marrow Transplant . 1994;14:57–61
- The clinical significance of human leukocyte antigen (HLA) allele compatibility in patients receiving a marrow transplant from serologically HLA-A, HLA-B, and HLA-DR matched unrelated donors . Blood . 2002;99:4200–4206
- Analysis of 500 bone marrow transplants from unrelated donors (UR-BMT) facilitated by the Japan Marrow Donor Program (confirmation of UR-BMT as a standard therapy for patients with leukemia and aplastic anemia) . Bone Marrow Transplant . 1999;24:995–1003
- Effect of matching of class I HLA alleles on clinical outcome after transplantation of hematopoietic stem cells from an unrelated donor. Japan Marrow Donor Program . N Engl J Med . 1998;339:1177–1185
- Clinical manifestations of graft-versus-host disease in human recipients of marrow from HL-A-matched sibling donors . Transplantation . 1974;18:295–304
- . Estimation of failure probabilities in the presence of competing risks (new representations of old estimators) . Stat Med . 1999;18:695–706
- Allogeneic hematopoietic stem cell transplantation from family members other than HLA-identical siblings over the last decade (1991-2000) . Blood . 2003;102:1541–1547
- Conditioning with targeted busulfan and cyclophosphamide for hemopoietic stem cell transplantation from related and unrelated donors in patients with myelodysplastic syndrome . Blood . 2002;100:1201–1207
- . Hospital capacity and post-transplant survival after allogeneic bone marrow transplantation (analysis of data from the Japan Society for Hematopoietic Cell Transplantation) . Bone Marrow Transplant . 2000;26:1061–1067
- 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 Transplantation) . J Clin Oncol . 2004;22:416–423
- . Use of granulocyte colony-stimulating factor following hematopoietic cell transplantation (does haste make waste?) . J Clin Oncol . 2004;22:390–391
PII: S1083-8791(05)00429-5
doi:10.1016/j.bbmt.2005.07.005
© 2005 American Society for Blood and Marrow Transplantation. Published by Elsevier Inc. All rights reserved.
Volume 11, Issue 11 , Pages 881-889, November 2005
