Bone Marrow Transplantation (2003) 32, 969–978 & 2003 Nature Publishing Group All rights reserved 0268-3369/03 $25.00 www.nature.com/bmt

Mini Review Myeloablative conditioning regimens for AML allografts: 30 years later

V Gupta1, HM Lazarus2 and A Keating1

1Department of Medical Oncology and Hematology, Princess Margaret Hospital/Ontario Cancer Institute, Toronto, Ontario, Canada; and 2Case Western Reserve University, Cleveland, OH, USA

Summary: a conditioning regimen can influence the incidence and severity of GVHD.6 The morbidity and mortality asso- During the last three decades, several myeloablative ciated with alloBMT therefore, depends a great deal on the conditioning regimens have been used for AML allografts. choice of conditioning regimen.7,8 Despite the long-stand- In this review, we systematically examine the data from ing use of alloBMT in AML, the optimal myeloablative studies reporting on myeloablative conditioning regimens conditioning regimen remains unknown.Few data from for AML allografts. High-dose busulfan combined with comparative or randomized studies are available to address cyclophosphamide (BuCy) and cyclophosphamide in this issue. combination with total body irradiation (CyTBI) are the The outcome of transplantation in AML is dependent two most commonly used conditioning regimens for AML upon a complex interaction of many disease-, patient- and allografts. From the available data, there are no treatment-related factors.9 Over the past decade, pretreat- significant differences in survival with these two regimens. ment cytogenetics has emerged as one of the most robust A small benefit of decreased relapse rate with CyTBIis prognostic determinants of outcome for AML.10–12 Cyto- counterbalanced by a nonsignificant increase in treatment- genetics data, however, were either not reported or related mortality. The incidence of veno-occlusive disease unavailable in a large proportion of patients in studies is significantly higher in patients treated with BuCy. comparing the conditioning regimens in AML.Also, Therapeutic monitoring of busulfan was not reported in several studies used different definitions of ‘good-risk’ any of the studies comparing the regimens. Either of the and ‘high-risk’ AML.Owing to the multiple factors regimens can be used for AML allografts, and the choice affecting outcome of alloBMT for AML (Table 1), may ultimately depend on local availability and expertise. interpretation of these data can be difficult in the absence Further improvements may be possible from modifications of well-designed randomized studies.Studies addressing of the standard regimens. Data from these latter studies this question in other diseases such as CML13,14 may seem to be encouraging, but are not based on comparative not be applicable to AML because of differences in the randomized trials. biology of the disease and the intensity of pre-transplant Bone Marrow Transplantation (2003) 32, 969–978. treatment. doi:10.1038/sj.bmt.1704285 The objective of this review is to examine the studies Keywords: AML; conditioning regimen; busulfan; cyclo- reporting on myeloablative conditioning regimens for phosphamide; total body irradiation AML allografts and to determine whether one particular regimen is superior.

Allogeneic bone marrow transplantation (alloBMT) is the Search strategy and selection criteria most intensive postremission therapy used for AML for more than 30 years.1,2 The relapse-risk with this treatment Studies for this review were identified by searches of is considerably lower than with autologous bone marrow PubMed/Medline, EMBASE databases and appropriate transplant (ABMT) or chemotherapy alone, although references from the relevant articles.Only papers in English high treatment-related mortality (TRM) continues to be a were reviewed and only studies reporting on the allograft major limitation.3–5 TRM is related to the direct toxicities conditioning regimens in AML with at least 20 patients in a of the conditioning regimens and includes interstitial disease specific setting (eg CR1 or advanced disease) were pneumonitis, veno-occlusive disease (VOD) and graft- included. versus-host disease (GVHD).Furthermore, the toxicity of

Evolution of myeloablative conditioning regimens in acute leukemia Correspondence: Dr V Gupta, Department of Medical Oncology and Hematology, Rm.5-224, 610-University Avenue, Toronto, Ontario, Cyclophosphamide and total body irradiation (CyTBI) Canada M5G 2M9; E-mail: [email protected] has long been used as a myeloablative conditioning or Conditioning regimens for AML V Gupta et al 970 Table 1 Factors affecting outcome of alloBMT in AML choice of one particular regimen over the other by the treating physician or center and are not possible to Clinicobiological features Cytogenetics at diagnosis exclude from these studies.Other center-specific effects de novo AML vs secondary AML that may affect outcome32 are also difficult to eliminate Remission status at transplant from these studies.Problems in the interpretation of data from the literature on conditioning regimens in Host Age AML are highlighted in Table 2.Results from the five Performance status studies comparing BuCy with CyTBI are summarized in Table 3. Graft Source of stem cells – PB or BM Dose of stem cells Survival outcomes.The multicenter randomized French Type of donor – related or unrelated study is the only study in adults that showed the superiority Graft manipulation such as T-cell depletion of CyTBI (n ¼ 50) over BuCy (n ¼ 51) in disease-free survival (DFS) (72 vs 47%, Po0.01) and overall survival Supportive care (OS) (75 vs 51%, P 0.02). The superior survival was GVHD prophylaxis o attributed to a lower relapse rate (14 vs 34%, Po0.04) and low TRM (8 vs 27%, Po0.06) in favor of the CyTBI arm.27 In contrast, the randomized study on a cohort of patients preparatory regimen after the initial reports of encouraging with AML and other leukemias (AML CR1 patients, BuCy results of alloBMT in acute leukemia.1,2,15 Based on studies (n ¼ 25), CyTBI (n ¼ 26)) from the Nordic group did not in rats16–18 and initial allotransplant trials in patients with show a significant difference in the outcome of patients relapsed acute leukemia,19,20 a myeloablative dose of with ‘early-disease’ treated with these two regimens.28 The busulfan (16 mg/kg p.o. over 4 days) was established. The results for AML in CR1 are difficult to interpret in this total daily dose of busulfan is usually given in four divided study because several parameters were reported on a mixed doses, mainly for ease of administration of the large total cohort of patients with AML, CML and ALL with dose.The busulfan/cyclophosphamide (BuCy) regimen was different remission status.Subsequent comparative studies considered an effective alternative to radiation-based regi- from the EBMT29 and IBMTR30 groups did not show any mens for AML.21 While initial studies with BuCy used significant difference in survival with these two treatment Cy at a dose of 200 mg/kg (50 mg/kg daily  4 days, also regimens. known as BuCy 4),21,22 further attempts were made to refine this regimen by decreasing the dose of Cy to 120 mg/ Treatment-related complications.No difference in the kg (60 mg/kg daily  2 days, also known as BuCy 2), which incidence of GVHD, either acute or chronic, was observed appeared to be equally effective and less toxic.23–25 It is with these two regimens in the registry database studies.29,30 noteworthy that there has been no randomized comparison The randomized comparative studies showed more deaths of BuCy 4 with BuCy 2.The CyTBI and BuCy regimens from GVHD in the BuCy Group27 and a higher number of have become the two most commonly used standard cases with grades III–V aGVHD or cGVHD.28 It is also myeloablative regimens in AML (IBMTR database, noteworthy that the diagnosis of GVHD was mainly personal communication, Horowitz, MM). clinical in these studies.In the absence of a pathologic diagnosis, it may be difficult to differentiate between VOD (consistently higher in the BuCy group) and GVHD. Moreover, how VOD modifies the course and severity Comparison of BuCy vs CyTBIin AML of GVHD is not known. The regimens will be compared for survival outcomes, treatment-related complications and relapse rate, according Table 2 Limitations of studies on conditioning regimens in AML to the remission status, a powerful factor affecting treatment outcome.21,26 1.Small number of study patients in randomized studies 2.Mixed cohorts of patients with different remission status 3.Different definitions of end points AML CR1 4.Center-specific effects in retrospective comparisons a.Bias of selection of one regimen over other These two conditioning regimens in AML in CR1 were Availability 27,28 Personal preference compared in two randomized and two registry-based b.Different perceptions of role of BMT in various risk-groups 29,30 31 studies in adults and one registry-based study in c.Practice of TBI children.The main limitation of the randomized studies Lung shielding from the French27 and Nordic group28 and the registry- Source of TBI based pediatric study is the small number of subjects.In Different doses and fractionation schedules of TBI d.Monitoring of busulfan blood concentration contrast, the observational database studies from the e.Supportive care differences EBMT29 and IBMTR30 groups have large numbers of Antimicrobial prophylaxis patients but selection bias remains an issue despite Growth factor use careful matching for patient and disease characteristics. f.T-cell depletion g.GVHD prophylaxis The biases relate to unknown factors that influence the

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 971 Table 3 Comparison of BuCy and CyTBI in AML CR1

Study Design Conditioning n TRM (%) RR (%) DFS or OS (%) Comments (study population) (outcome at n years) regimen EFS (%)

Blaise et al27 Randomized BuCy2 51 27 34 47 51 CyTBI superior for (adults) (outcome at 2 years) survival, relapse rate and TRM CyTBI 50 8 14 72 75 Ringden et al28 Randomized BuCy2 25 NR* NR* 83 NR* Busulfan equally (adults) (outcome at 3 years) effective alternative to TBI in early disease, that is, CR1 (results reported for mixed cohort of patients) CyTBI 26 NR* NR* 58 NR* Ringden et al29 Comparative BuCy2 134 15 23 65 NR No difference in TRM, (adults) (outcome at 2 years) DFS or relapse rate. BuCy4 89 18 24 62 NR CyTBI 223 19 19 66 NR Litzow et al30 Comparative BuCy 381 27 19 54 55 CyTBI superior for (adults) (outcome at 5 years) lower relapse, no difference in survival CyTBI 200 30 12 58 60 Michel et al31 Comparative BuCy2 23 0 54 46 NR BuCy2 associated with (children) (outcome at 5 years) significantly higher-risk of relapse, no difference between BuCy4 and TBI containing regimen BuCy4 19 5 13 82 NR TBI** 32 10 10 80 NR

TRM, treatment-related mortality; RR, relapse rate; DFS, disease-free survival; EFS, event-free survival; OS, overall survival; NR, not reported; NR*, results reported for mixed cohort of patients but not separately for AML CR1 patients; TBI**, with TBI 24 patients received Cy, four Cy and additional drug and four melphalan; BuCy2, cyclophosphamide 120 mg/kg with busulfan; BuCy4, cyclophosphamide 200 mg/kg with busulfan.

It is noteworthy that TRM in the CyTBI arm (8%) was difference in earlier neutrophil recovery in the BuCy significantly lower in the French study, but this has not cohort.30 been matched in other studies (Table 3).The cause of this discrepancy is not clear.The fractionation of TBI and lung Relapse rate.The French randomized trial and the shielding reduce the incidence of interstitial pneumonitis IBMTR registry study, but not the EBMT registry study in TBI-based regimens.33,34 Lung shielding was used in all showed significantly lower relapse rates in the CyTBI arm. cases in the TBI group in the French study27 compared A different pattern of relapse, especially increased relapses with 60% of evaluable cases in the IBMTR registry study.30 at extramedullary (EM) sites, particularly the central The lungs receive 10–20% higher dose of radiation than nervous system (CNS), was observed in patients receiving the prescribed dose, because they are air-containing BuCy compared to CyTBI.30 A high incidence of EM organs.The practice of lung shielding to eliminate this relapses was also observed in patients with AML treated ‘overdose effect’ is not universal and varies at different with BuCy conditioning at a single center.35 These centers.It seems unlikely that this factor alone led to observations suggest that TBI-based regimens are more lower TRM in the TBI group in the French study and effective in treating occult CNS disease. may be a chance association due to the small number of It is noteworthy that in the French study, the dose of patients studied.While the EBMT study described a Cy was 120 mg/kg (BuCy2).It is unlikely however, that this significantly higher proportion of cases of interstitial led to the increased relapses in the BuCy arm as previous pneumonitis in the CyTBI arm,29 this was not observed in studies suggested that the efficacy of BuCy2 may be similar the IBMTR study.30 to BuCy4.23–25 Furthermore, both registry studies found Other complications appear to be more frequent in no difference in the outcome of patients treated with patients receiving BuCy.A consistent finding in all the lower dose of Cy.29,30 Nevertheless, the pediatric these studies was the significantly higher incidence of registry study, based on a small number of patients, VOD of the liver with BuCy compared to CyTBI.27–30 Also, demonstrated the superiority of BuCy4 to BuCy2 in a a significantly higher incidence of hemorrhagic cystitis was comparison of BuCy2, BuCy4 and TBI-based conditioning observed with the BuCy regimen.28,29 In addition, seizures regimens for patients with AML in CR1.31 This study occur in a significantly higher proportion of patients showed a significantly higher risk of relapse with BuCy2, receiving busulfan.28 but no difference between BuCy4 and CyTBI.Plasma busulfan levels, however, were not monitored in any of Engraftment.There were no differences in the these studies.Inter- and intrapatient variability in the engraftment rate between the two groups in the EBMT kinetics of busulfan may explain the different results among study,29 although the IBMTR study showed a significant the various studies.

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 972 Long-term complications.Concerns have been raised at was higher.Updated results of this study reported the the increased risk of secondary malignancies in patients same trend.42 treated with TBI-based regimens.36,37 There are few data on In another prospective study, the CyTBI regimen (n ¼ 36) the comparative risk of secondary malignancies with these was compared with melphalan and TBI (n ¼ 27) in AML two regimens.Long-term follow-up of patients treated with CR1.43 Neither the actuarial probability of remaining in these two regimens in the French study did not show remission (66 vs 94%, P40.1), nor the OS was significantly significant difference in the number of cases of second different between the two groups (53 vs 55%). malignancy.38 In the absence of clear evidence that one regimen causes more cases of cancer, it is important to keep Beyond CR1 patients treated with either regimen under long-term surveillance for secondary malignancies. A randomized study from the Southwest Oncology Group While long-term follow-up in the Nordic study found (SWOG) of 114 leukemia patients (including AML, ALL significantly more cases of obstructive bronchiolitis in and CML) not in first remission, compared fTBI (1320 cGy the BuCy group and cataracts in TBI cohort,39 no such of TBI in 11 fractions) and VP-16 (60 mg/kg) for one dose long-term differences were observed in AML patients vs BuCy2.44 The investigators stratified the patients to treated with these two conditioning regimens in the ‘good-risk’ (CR2 or accelerated phase) and ‘poor-risk’ French study.38 (CR3, induction failure, in relapse, blast phase) categories and showed equivalent survival outcomes with the fTBI Quality of life and costs.The quality of life (QOL) and and BuCy regimens (55 vs 34%, P ¼ 0.30). There were a health care utilization costs have not been studied total of 40 AML patients (good-risk (n ¼ 11), poor-risk prospectively in any of these studies.In a combined long- (n ¼ 29)) in this trial.Owing to the limited number of term analysis of the French and Nordic studies, equal patients, the results were not analyzed according to the type proportions of AML patients treated with CyTBI or BuCy of leukemia. returned to work/school (85 vs 88%).40 Modifications of standard myeloablative regimens in AML Beyond CR1 Very scanty data are available to determine whether In an effort to improve the efficacy of the regimens, several patients with advanced AML (second or later remissions, investigators have modified standard protocols by early relapse) benefit preferentially from TBI- or chemo- therapy-based regimens.In the EBMT comparative 1.Replacing cyclophosphamide with agents such as VP-16. study, no difference in TRM, relapse or LFS was observed 2.Intensifying the regimens with in patients receiving CyTBI (n ¼ 46) compared with (a) Additional chemotherapy. BuCy (n ¼ 46).29 In the Nordic study, the TBI-based (b) Radioimmunotherapy. regimen was superior for patients with advanced disease 3.Using newer strategies with busulfan. in a mixed cohort (beyond CR1 or CML patients 4.T-cell depletion. beyond first chronic phase), however, the number of AML patients with advanced disease in this study was too small to reach any meaningful conclusions (BuCy, Replacing cyclophosphamide with VP-16 28 n ¼ 12; CyTBI, n ¼ 6). The fTBI/VP-16 regimen was studied in 99 patients with acute leukemia in CR1 (AML, n ¼ 61).45 The regimen was Other comparative studies of myeloablative conditioning well tolerated and for AML patients, cumulative probabil- regimens in AML ities for DFS and relapse rates at 3 years were 61 and 12% (Table 4).This regimen has not been tested in AML CR1 AML CR1 patients in comparative studies.Another pilot study reported encouraging results (3 years DFS 56% and relapse The question of the optimum dose and schedule of TBI was rate 25%) by using VP-16 with busulfan for AML addressed by two randomized studies from the Seattle autografts.46 However, this regimen has not been tested group.In the first randomized study, AML patients in CR1 for AML allografts. were randomized between cyclophosphamide and single n vs exposure 10 Gy TBI ( ¼ 27) 12 Gy TBI in six fractions in Intensification of standard myeloablative regimens with combination with cyclophosphamide (n ¼ 26).41 This study showed a significant survival advantage for patients treated (a) Additional chemotherapy.Attempts were subsequently with the fractionated TBI regimen. made to improve the efficacy of standard regimens by the In an effort to increase the antileukemic efficacy of addition of other agents such as VP16 (etoposide),47–50 TBI, AML patients in CR1 were randomized to receive Thiotepa,51 and cytarabine.52 These regimens were well 12 Gy TBI in six fractions (n ¼ 34) vs 15.75 Gy TBI in tolerated and results from these studies are summarized in seven fractions (n ¼ 37).8 Indeed, the relapses in the Tables 4 and 5, according to remission status.The German cohort receiving higher dose of TBI was significantly group have reported impressive results using a combination lower, however, this did not translate into a survival of VP-16 with BuCy2 in AML CR1 in a mixed group of benefit because nonrelapse mortality from aGVHD patients (adults and children).48,49 Of particular note, none

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 973 Table 4 Studies using chemotherapy modifications of standard myeloablative regimens in AML CR1

Study Design Conditioning n TRM (%) RR (%) DFS or OS (%) Comments (study population) (outcome at n years regimen EFS (%) or months)

Snyder et al45 Single arm FTBI+VP-16 61 28 12 61 62 (i) Mixed population of (adults and (outcome at 3 years) adults and children, median children) age 26.3 years (1.9–47.4) (ii) Age a significant prognostic variable (iii) Not tested with a standard myeloablative regimen Zander et al48 Single arm BuCy2+VP16 31 19.5 0 80.5 80.5 (i) Well-tolerated regimen, (adults and (outcome at 30.5 months) (VP16 30-60 (ii) Adult and pediatric children) mg/kg) patients combined, median age 30 years (4–51) (iii) Not compared with a standard myeloablative regimen Kroger et al49 Comparison of two BuCy2+VP-16 13 30.7 0 66 NR (i) Highly efficacious, no (adults and doses of VP16 added 45 mg/kg relapses children) to BuCy2 (outcome at vs 3 years) BuCy+VP-16 25 20 0 80 NR (ii) Lower dose of etoposide 30 mg/kg (30 mg/kg) better tolerated (iii) Adult and pediatric patients combined (iv) Not compared with a standard myeloablative regimen

TRM, treatment-related mortality; RR, relapse rate; DFS, disease-free survival; EFS, event-free survival; OS, overall survival; NR, not reported. of the patients relapsed in these two studies.The data intestinal absorption, dosing uncertainties due suggest that VP-16 30 mg/kg for one dose with BuCy2 is to emesis, erratic bioavailability and a narrow thera- well tolerated.Another study reported severe pulmonary peutic window.A suboptimal dose of busulfan expressed toxicity of this regimen in patients with a prior history of as a low area under the curve (AUC) resulted in a higher chest irradiation undergoing ABMT and alloBMT.53 The frequency of graft rejection and relapses60 and a higher dose of VP-16 in this study was 40 mg/kg for one dose AUC correlated with increased toxicity, mainly VOD along with BuCy2. and seizures.61–63 The pharmacokinetics of busulfan are also age-dependent and children tend to clear busulfan (b) Radioimmunotherapy.Targeted radioimmunother- faster.61 Some of these issues may be addressed by apy, either in combination with radiation-based or chemo- monitoring blood busulfan levels using limited sampling therapy-based regimens is another means of increasing the models followed by dose adjustment to achieve a ‘targeted’ radiation dose without increasing toxicity.54 Clinical trials steady-state concentration.64,65 Cyclophosphamide levels using 131I-labeled anti-CD33 antibody M195,55 131I-labeled however, are not routinely monitored after conditioning anti-CD45 antibody56,57 and 188Rhenium-labeled anti- therapy.It was recently shown in patients treated CD66 antibody58 in combination with the CyTBI and with CyTBI protocols that increased exposure to toxic BuCy regimens demonstrate the feasibility of this ap- metabolites of cyclophosphamide resulted in increased proach.These approaches mainly use b-emitting isotopes liver toxicity and nonrelapse mortality.66 Whether a conjugated to monoclonal antibodies.More recently, an a- ‘targeting strategy’ with cyclophosphamide similar to the particle emitter, 213Bi in conjugation with Hum195, was approach used with busulfan will result in better survival is shown to be active in a phase I study in advanced myeloid currently not known. leukemias.59 Unlike b-particle emitting isotopes, a emitters More recently, intravenous (i.v.) preparations of busul- can selectively kill individual cancer cells.The use of a- fan have become available.Four different formulations particle-based targeted radiation with conditioning regi- of busulfan have been tested in phase I/II studies.67–70 The mens has not yet been studied.These strategies appear safety profile of i.v. preparations of busulfan is favorable promising and may further improve the efficacy of the with fewer acute side effects and better engraftment. conditioning regimens, but are restricted to larger centers A comparative retrospective analysis of i.v. busulfan vs because of the requirement for special facilities. p.o. busulfan in the BuCy2 regimen showed a significantly lower VOD (8 vs 33%, P=0.002) and significantly superior 71 Newer strategies with busulfan 100-day survival in favor of i.v. preparation. It is important to note that busulfan levels were not reported A major limitation of p.o. busulfan is wide interpatient in this study. The superiority of i.v. busulfan over p.o. variability of pharmacokinetics due to unpredictable busulfan with ‘targeted levels’ is not known at present

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 974 Table 5 Studies using chemotherapy modifications of standard myeloablative regimens in AML beyond CR1

Study Design Conditioning N TRM (%) RR (%) DFS or OS (%) Comments (study population) (outcome at n regimen EFS (%) years or months)

Brown et al50 Single arm Etoposide+ 40 47 44 29 NR Patients in untreated (adults) (outcome at 44 months) CyTBI first relapse Bibawi et al51 Single arm Thiotepa+ 50 19a 43a NR 26a (i) Active regimen in (adults) (outcome at 3 years) BuCy2 advanced AML (ii) Comparable TRM to standard myeloablative regimens (iii) Dose of busulfan reduced to 12 mg/kg (iv) Results reported for combined AML and advanced MDS patients

TRM, treatment-related mortality; RR, relapse rate; DFS, disease-free survival; EFS, event-free survival; OS, overall survival; NR, not reported. aResults reported for mixture of patients with AML and advanced MDS.

and this remains a key study question for future clinical main aim is to exploit the beneficial graft-versus-leukemia trials in this area. (GVL) effect and reduce regimen-related complications Initial studies with i.v. busulfan used the four times daily such as GVHD.These approaches offer patients who dosing schedule as with p.o. busulfan. A recent study are not candidates for myeloablative transplants due to reported once daily i.v. busulfan (3.2 mg/kg i.v. Â 4 days) age or concomitant medical problems, the opportunity with fludarabine as a myeloablative conditioning regimen to benefit from a GVL effect.79,80 Experience with AML for hematologic malignancies.72 Pharmacokinetic studies however is limited.81–84 A high relapse rate continues on a subset of patients showed linear kinetics with this to be a major issue with nonmyeloablative protocols.84,85 approach with cumulative AUC comparable to that Although the comparison of nonmyeloablative and established for p.o. busulfan. The once daily regimen may myeloablative protocols for the incidence of GVHD is help to further simplify the treatment regimen. difficult because of different patient eligibility criteria, a recent study showed that the cumulative incidence of T-cell depletion grades II–IV aGVHD was significantly in favor of nonmyeloablative transplants (64 vs 85%, P ¼ 0.001).86 As GVHD is one of the main factors that influences TRM, Nonetheless, there was no difference in cGVHD requiring approaches to decrease the incidence and severity of treatment (73 vs 71%).More disease-specific and long-term GVHD by decreasing the number of T cells in the donor data are needed to fully evaluate the efficacy of these marrow have caught investigative attention.Several meth- approaches. ods of T-cell depletion such as negative selection by At present, myeloablative conditioning regimens, clini- physical separation or antibody-based purging have been cally tested for over 30 years, remain the gold standard tried.73 The results of clinical trials using this approach in of treatment and allotransplants with novel strategies for AML patients in CR1 are summarized in Table 6.74–76 No patients at high risk of TRM should only be offered as a additional GVHD prophylaxis to CR1 patients was given part of a clinical trial. in these trials.The main problem with this approach remains the higher risk of relapse, graft failure and immune reconstitution.77 Graft rejection with T-cell depletion Conclusion allografts may be overcome by increasing pretransplant immunosuppression and myeloablation with ATG and Based on an assessment of the literature, survival with the thiotepa.76,78 All AML patients in these two studies two most commonly used myeloablative conditioning achieved primary and sustained engraftment with full regimens, BuCy and CyTBI, are similar.Local availability donor chimerism.Moreover, GVHD was either signifi- and expertise may determine the preference of one over the other.Neither regimen is suitable for all the situations and cantly reduced (grade I aGVHD o5%, no cases of grades II–IV aGVHD, cGVHD 3%)76 or not seen at all.78 a particular regimen should be avoided in selected clinical Despite the increased scientific knowledge gained over situations; for example, BuCy should be avoided in a the last two decades in this area, the exact role of T-cell patient with abnormal liver function and CyTBI, in depletion in allotransplants for AML remains unclear. patients with a prior history of radiation to the lung. Modification of these regimens by intensification with additional chemotherapy or radioimmunotherapy, newer Novel strategies for conditioning regimens preparations of busulfan and T-cell depletion may enhance efficacy and tolerability.Whether these modifications will More recently, several groups have focused on nonmyelo- result in better survival in AML patients remain to be ablative or reduced intensity conditioning regimens.The established in randomized studies.

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 975 Table 6 Studies evaluating myeloablative conditioning regimens with T-cell depletion in AML CR1

Study (study Design (outcome at Conditioning n TRM (%) RR (%) DFS or OS (%) Comments population) n years or months) regimen EFS (%)

Young et al74 Single arm (out- CyTBI 31 —a 13 45 45 (i) 12 patients died (adults) come at 3 years of treatment related T-cell depletion of causes BM grafts (physical (ii) Graft rejection method) major cause of treatment failure Soiffer et al75 Single arm (out- CyTBI mainly 28 16b 31 63 71 Several results (adults) come at 4 years BuCy 1 patient reported for CD6 depleted combined AML and donor grafts by ALL patients T12 monoclonal antibody) Papadopoulos Single arm (out- ATG & Thiotepa+ 31 19.4 3.2 77.4 NR (i) Low relapse rate et al76 (adults) come at 4 years CyTBI (ii) aGVHD grade I T-cell depletion o5%, no cases of of BM grafts) II–IV aGVHD (physical method) (iii) cGVHD 3%

TRM, treatment-related mortality; RR, relapse rate; DFS, disease-free survival; EFS, event-free survival; OS, overall survival; aGVHD, acute GVHD; cGVHD, chronic GVHD. a12/31 patients died of causes related to transplant and one indeterminate cause. bReported for mixture of AML and ALL patients.

Acknowledgements leukemia in first remission: a randomized trial of two irradiation regimens. Blood 1990; 76: 1867–1871. We thank Dr JM Rowe for a careful review of the manuscript 9 Appelbaum FR.Is there a best transplant conditioning regimen and helpful comments.AK holds the Gloria and Seymour for acute myeloid leukemia? Leukemia 2000; 14: 497–501. Epstein Chair in Cell Therapy and Transplantation at University 10 Grimwade D, Walker H, Oliver F et al. The importance of Health Network and University of Toronto. diagnostic cytogenetics on outcome in AML: analysis of 1,612 patients entered into the MRC AML 10 trial.The Medical Research Council Adult and Children’s Leukaemia Working References Parties. Blood 1998; 92: 2322–2333. 11 Slovak ML, Kopecky KJ, Cassileth PA et al. Karyotypic 1 Thomas ED, Buckner CD, Banaji M et al. One hundred analysis predicts outcome of preremission and postremission patients with acute leukemia treated by chemotherapy, total therapy in adult acute myeloid leukemia: a Southwest body irradiation, and allogeneic marrow transplantation. Oncology Group/Eastern Cooperative Oncology Group Blood 1977; 49: 511–533. Study. Blood 2000; 96: 4075–4083. 2 Thomas ED, Buckner CD, Clift RA et al. Marrow transplan- 12 Byrd JC, Mrozek K, Dodge RK et al. Pretreatment cytogenetic tation for acute nonlymphoblastic leukemia in first remission. abnormalities are predictive of induction success, cumulative N Engl J Med 1979; 301: 597–599. incidence of relapse, and overall survival in adult patients 3 Zittoun RA, Mandelli F, Willemze R et al. Autologous or with de novo acute myeloid leukemia: results from Cancer allogeneic bone marrow transplantation compared with and Leukemia Group B (CALGB 8461). Blood 2002; 100: intensive chemotherapy in acute myelogenous leukemia. 4325–4336. European Organization for Research and Treatment of Cancer 13 Clift RA, Buckner CD, Thomas ED et al. Marrow transplan- (EORTC) and the Gruppo Italiano Malattie Ematologiche tation for chronic myeloid leukemia: a randomized study Maligne dell’Adulto (GIMEMA) Leukemia Cooperative comparing cyclophosphamide and total body irradiation with Groups. N Engl J Med 1995; 332: 217–223. busulfan and cyclophosphamide. Blood 1994; 84: 2036–2043. 4 Cassileth PA, Harrington DP, Appelbaum FR et al. Che- 14 Devergie A, Blaise D, Attal M et al. Allogeneic bone marrow motherapy compared with autologous or allogeneic bone transplantation for chronic myeloid leukemia in first chronic marrow transplantation in the management of acute myeloid phase: a randomized trial of busulfan-cytoxan vs cytoxan-total leukemia in first remission. N Engl J Med 1998; 339: body irradiation as preparative regimen: a report from the 1649–1656. French Society of Bone Marrow Graft (SFGM). Blood 1995; 5 Burnett AK, Wheatley K, Goldstone AH et al. The value of 85: 2263–2268. allogeneic bone marrow transplant in patients with acute 15 Beutler E, Blume KG, Bross KJ et al. Bone marrow myeloid leukaemia at differing risk of relapse: results of the transplantation as the treatment of choice for ‘good risk’ adult UK MRC AML 10 trial. Br J Haematol 2002; 118: 385–400. patients with acute leukemia. Trans Assoc Am Phys 1979; 92: 6 Ferrera J, Antin J. Pathophysiology of Graft-vs-Host Disease. 189–195. Hematopoietic Stem Cell Transplantation.Blackwell Scientific 16 Santos GW, Tutschka PJ.Effect of busulfan on antibody Publications: Boston, MA, 1999, pp 305–315. production and skin allograft survival in the rat. J Natl Cancer 7 Gluckman E.Influence of conditioning on the outcome of Inst 1974; 53: 1775–1780. allogeneic bone marrow transplantation. Int J Radiat Oncol 17 Tutschka PJ, Santos GW.Bone marrow transplantation in the Biol Phys 1990; 19: 1325–1327. busulfan-treated rat.I.Effect of cyclophosphamide and rabbit 8 Clift RA, Buckner CD, Appelbaum FR et al. Allogeneic antirat thymocyte serum as immunosuppression. Transplanta- marrow transplantation in patients with acute myeloid tion 1975; 20: 101–106.

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 976 18 Tutschka PJ, Santos GW.Bone marrow transplantation in the 33 Pino y Torres JL, Bross DS, Lam WC et al. Risk factors in busulfan-treated rat.II.Effect of cyclophosphamide and interstitial pneumonitis following allogenic bone marrow antithymic serum on the presensitized state. Transplantation transplantation. Int J Radiat Oncol Biol Phys 1982; 8: 1975; 20: 116–122. 1301–1307. 19 Tutschka PJ, Santos GW, Elfenbein GJ.Marrow transplanta- 34 Labar B, Bogdanic V, Nemet D et al. Total body tion in acute leukemia following busulfan and cyclopho- irradiation with or without lung shielding for allogeneic bone sphamide. Haematol Blood Transfus 1980; 25: 375–380. marrow transplantation. Bone Marrow Transplant 1992; 9: 20 Santos GW, Kaizer H.Bone marrow transplantation in acute 343–347. leukemia. Semin Hematol 1982; 19: 227–239. 35 Simpson DR, Nevill TJ, Shepherd JD et al. High incidence of 21 Santos GW, Tutschka PJ, Brookmeyer R et al. Marrow extramedullary relapse of AML after busulfan/cyclophospha- transplantation for acute nonlymphocytic leukemia after mide conditioning and allogeneic stem cell transplantation. treatment with busulfan and cyclophosphamide. N Engl J Bone Marrow Transplant 1998; 22: 259–264. Med 1983; 309: 1347–1353. 36 Witherspoon RP, Fisher LD, Schoch G et al. Secondary 22 Geller RB, Saral R, Piantadosi S et al. Allogeneic bone cancers after bone marrow transplantation for leukemia or marrow transplantation after high-dose busulfan and cyclo- aplastic anemia. N Engl J Med 1989; 321: 784–789. phosphamide in patients with acute nonlymphocytic leukemia. 37 Curtis RE, Rowlings PA, Deeg HJ et al. Solid cancers Blood 1989; 73: 2209–2218. after bone marrow transplantation. N Engl J Med 1997; 336: 23 Tutschka PJ, Copelan EA, Klein JP.Bone marrow transplan- 897–904. tation for leukemia following a new busulfan and cyclopho- 38 Blaise D, Maraninchi D, Michallet M et al. Long-term follow- sphamide regimen. Blood 1987; 70: 1382–1388. up of a randomized trial comparing the combination of 24 Tutschka PJ, Copelan EA, Kapoor N.Replacing total body cyclophosphamide with total body irradiation or busulfan as irradiation with busulfan as conditioning of patients with conditioning regimen for patients receiving HLA-identical leukemia for allogeneic marrow transplantation. Transplant marrow grafts for acute myeloblastic leukemia in first complete Proc 1989; 21: 2952–2954. remission. Blood 2001; 97: 3669–3671. 25 Copelan EA, Biggs JC, Thompson JM et al. Treatment for 39 Ringden O, Remberger M, Ruutu T et al. Increased risk of acute myelocytic leukemia with allogeneic bone marrow chronic graft-vs-host disease, obstructive bronchiolitis, and transplantation following preparation with BuCy2. Blood alopecia with busulfan vs total body irradiation: long-term 1991; 78: 838–843. results of a randomized trial in allogeneic marrow recipients 26 Zapatero A, Martin de Vidales C, Pinar B et al. Prognostic with leukemia.Nordic Bone Marrow Transplantation Group. factors affecting leukemia relapse after allogeneic BMT Blood 1999; 93: 2196–2201. conditioned with cyclophosphamide and fractionated TBI. 40 Socie G, Clift RA, Blaise D et al. Busulfan plus cyclopho- Bone Marrow Transplant 1996; 18: 591–596. sphamide compared with total-body irradiation plus cyclopho- 27 Blaise D, Maraninchi D, Archimbaud E et al. Allogeneic bone sphamide before marrow transplantation for myeloid marrow transplantation for acute myeloid leukemia in first leukemia: long-term follow-up of 4 randomized studies. Blood remission: a randomized trial of a busulfan-cytoxan 2001; 98: 3569–3574. vs cytoxan-total body irradiation as preparative regimen: a 41 Thomas ED, Clift RA, Hersman J et al. Marrow transplanta- report from the Group d’Etudes de la Greffe de Moelle tion for acute nonlymphoblastic leukemic in first remission Osseuse. Blood 1992; 79: 2578–2582. using fractionated or single-dose irradiation. Int J Radiat 28 Ringden O, Ruutu T, Remberger M et al. A randomized trial Oncol Biol Phys 1982; 8: 817–821. comparing busulfan with total body irradiation as condition- 42 Clift RA, Buckner CD, Appelbaum FR et al. Long-term ing in allogeneic marrow transplant recipients with leukemia: a follow-up of a randomized trial of two irradiation regimens report from the Nordic Bone Marrow Transplantation Group. for patients receiving allogeneic marrow transplants during Blood 1994; 83: 2723–2730. first remission of acute myeloid leukemia. Blood 1998; 92: 29 Ringden O, Labopin M, Tura S et al. A comparison of 1455–1456. busulphan vs total body irradiation combined with cyclopho- 43 Helenglass G, Powles RL, McElwain TJ et al. Melphalan and sphamide as conditioning for autograft or allograft bone total body irradiation (TBI) versus cyclophosphamide and TBI marrow transplantation in patients with acute leukaemia. as conditioning for allogeneic matched sibling bone marrow Acute Leukaemia Working Party of the European Group for transplants for acute myeloblastic leukaemia in first remission. Blood and Marrow Transplantation (EBMT). Br J Haematol Bone Marrow Transplant 1988; 3: 21–29. 1996; 93: 637–645. 44 Blume KG, Kopecky KJ, Henslee-Downey JP et al. A 30 Litzow MR, Perez WS, Klein JP et al. Comparison of outcome prospective randomized comparison of total body irradia- following allogeneic bone marrow transplantation with cyclo- tion-etoposide vs busulfan-cyclophosphamide as preparatory phosphamide-total body irradiation vs busulphan-cyclopho- regimens for bone marrow transplantation in patients with sphamide conditioning regimens for acute myelogenous leukemia who were not in first remission: a Southwest leukaemia in first remission. Br J Haematol 2002; 119: Oncology Group study. Blood 1993; 81: 2187–2193. 1115–1124. 45 Snyder DS, Chao NJ, Amylon MD et al. Fractionated total 31 Michel G, Gluckman E, Esperou-Bourdeau H et al. Allogeneic body irradiation and high-dose etoposide as a preparatory bone marrow transplantation for children with acute myelo- regimen for bone marrow transplantation for 99 patients with blastic leukemia in first complete remission: impact of acute leukemia in first complete remission. Blood 1993; 82: conditioning regimen without total-body irradiation – a report 2920–2928. from the Societe Francaise de Greffe de Moelle. J Clin Oncol 46 Linker CA, Ries CA, Damon LE et al. Autologous bone 1994; 12: 1217–1222. marrow transplantation for acute myeloid leukemia using 32 Frassoni F, Labopin M, Powles R et al. Effect of centre on busulfan plus etoposide as a preparative regimen. Blood 1993; outcome of bone-marrow transplantation for acute myeloid 81: 311–318. leukaemia.Acute Leukaemia Working Party of the European 47 Spitzer TR, Cottler-Fox M, Torrisi J et al. Escalating doses Group for Blood and Marrow Transplantation. Lancet 2000; of etoposide with cyclophosphamide and fractionated 355: 1393–1398. total body irradiation or busulfan as conditioning for bone

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 977 marrow transplantation. Bone Marrow Transplant 1989; 4: 63 Ljungman P, Hassan M, Bekassy AN et al. High busulfan 559–565. concentrations are associated with increased transplant-related 48 Zander AR, Berger C, Kroger N et al. High dose chemother- mortality in allogeneic bone marrow transplant patients. Bone apy with busulfan, cyclophosphamide, and etoposide as Marrow Transplant 1997; 20: 909–913. conditioning regimen for allogeneic bone marrow transplanta- 64 Chattergoon DS, Saunders EF, Klein J et al. An improved tion for patients with acute myeloid leukemia in first complete limited sampling method for individualised busulphan dosing remission. Clin Cancer Res 1997; 3: 2671–2675. in bone marrow transplantation in children. Bone Marrow 49 Kroger N, Zabelina T, Sonnenberg S et al. Dose-dependent Transplant 1997; 20: 347–354. effect of etoposide in combination with busulfan plus 65 Schuler U, Schroer S, Kuhnle A et al. Busulfan pharmacoki- cyclophosphamide as conditioning for stem cell transplanta- netics in bone marrow transplant patients: is drug monitoring tion in patients with acute myeloid leukemia. Bone Marrow warranted? Bone Marrow Transplant 1994; 14: 759–765. Transplant 2000; 26: 711–716. 66 McDonald GB, Slattery JT, Bouvier ME et al. Cyclopho- 50 Brown RA, Wolff SN, Fay JW et al. High-dose etoposide, sphamide metabolism, liver toxicity, and mortality following cyclophosphamide, and total body irradiation with allogeneic hematopoietic stem cell transplantation. Blood 2003; 101: bone marrow transplantation for patients with acute myeloid 2043–2048. leukemia in untreated first relapse: a study by the North 67 Andersson BS, Madden T, Tran HT et al. Acute safety and American Marrow Transplant Group. Blood 1995; 85: pharmacokinetics of intravenous busulfan when used with oral 1391–1395. busulfan and cyclophosphamide as pretransplantation con- 51 Bibawi S, Abi-Said D, Fayad L et al. Thiotepa, busulfan, and ditioning therapy: a phase I study. Biol Blood Marrow cyclophosphamide as a preparative regimen for allogeneic Transplant 2000; 6: 548–554. transplantation for advanced myelodysplastic syndrome and 68 Olavarria E, Hassan M, Eades A et al. A phase I/II study of acute myelogenous leukemia. Am J Hematol 2001; 67: multiple-dose intravenous busulfan as myeloablation prior to 227–233. stem cell transplantation. Leukemia 2000; 14: 1954–1959. 52 Ratanatharathorn V, Karanes C, Lum LG et al. Allogeneic 69 Schuler US, Renner UD, Kroschinsky F et al. Intravenous bone marrow transplantation in high-risk myeloid disorders busulphan for conditioning before autologous or allogeneic using busulfan, cytosine arabinoside and cyclophosphamide human blood stem cell transplantation. Br J Haematol 2001; (BAC). Bone Marrow Transplant 1992; 9: 49–55. 114: 944–950. 53 Crilley P, Topolsky D, Styler MJ et al. Extramedullary toxicity 70 Hassan M, Nilsson C, Hassan Z et al. A phase II trial of a conditioning regimen containing busulphan, cyclopho- of liposomal busulphan as an intravenous myeloablative agent sphamide and etoposide in 84 patients undergoing autologous prior to stem cell transplantation: 500 mg/m(2) as a optimal and allogenic bone marrow transplantation. Bone Marrow total dose for conditioning. Bone Marrow Transplant 2002; 30: Transplant 1995; 15: 361–365. 833–841. 54 Reske SN, Bunjes D, Buchmann I et al. Targeted bone 71 Kashyap A, Wingard J, Cagnoni P et al. Intravenous vs oral marrow irradiation in the conditioning of high-risk leukaemia busulfan as part of a busulfan/cyclophosphamide preparative prior to stem cell transplantation. Eur J Nucl Med 2001; 28: regimen for allogeneic hematopoietic stem cell transplantation: 807–815. decreased incidence of hepatic venoocclusive disease (HVOD), 55 Jurcic JG, Caron PC, Nikula TK et al. Radiolabeled anti- HVOD-related mortality, and overall 100-day mortality. Biol CD33 monoclonal antibody M195 for myeloid leukemias. Blood Marrow Transplant 2002; 8: 493–500. Cancer Res 1995; 55: 5908s–5910s. 72 Russell JA, Tran HT, Quinlan D et al. Once-daily intravenous 56 Matthews DC, Appelbaum FR, Eary JF et al. Development of busulfan given with fludarabine as conditioning for allogeneic a marrow transplant regimen for acute leukemia using targeted stem cell transplantation: study of pharmacokinetics and early hematopoietic irradiation delivered by 131I-labeled anti-CD45 clinical outcomes. Biol Blood Marrow Transplant 2002; 8: 468–476. antibody, combined with cyclophosphamide and total body 73 Ho VT, Soiffer RJ.The history and future of T-cell depletion irradiation. Blood 1995; 85: 1122–1131. as graft-versus-host disease prophylaxis for allogeneic hema- 57 Matthews DC, Appelbaum FR, Eary JF et al. Phase I study of topoietic stem cell transplantation. Blood 2001; 98: 3192–3204. (131)I-anti-CD45 antibody plus cyclophosphamide and total 74 Young JW, Papadopoulos EB, Cunningham I et al. T-cell- body irradiation for advanced acute leukemia and myelodys- depleted allogeneic bone marrow transplantation in adults plastic syndrome. Blood 1999; 94: 1237–1247. with acute nonlymphocytic leukemia in first remission. Blood 58 Bunjes D, Buchmann I, Duncker C et al. Rhenium 188-labeled 1992; 79: 3380–3387. anti-CD66 (a, b, c, e) monoclonal antibody to intensify the 75 Soiffer RJ, Fairclough D, Robertson M et al. CD6-depleted conditioning regimen prior to stem cell transplantation for allogeneic bone marrow transplantation for acute leukemia in patients with high-risk acute myeloid leukemia or myelodys- first complete remission. Blood 1997; 89: 3039–3047. plastic syndrome: results of a phase I–II study. Blood 2001; 98: 76 Papadopoulos EB, Carabasi MH, Castro-Malaspina H et al. 565–572. T-cell-depleted allogeneic bone marrow transplantation as 59 Jurcic JG, Larson SM, Sgouros G et al. Targeted alpha particle postremission therapy for acute myelogenous leukemia: free- immunotherapy for myeloid leukemia. Blood 2002; 100: dom from relapse in the absence of graft-versus-host disease. 1233–1239. Blood 1998; 91: 1083–1090. 60 Slattery JT, Sanders JE, Buckner CD et al. Graft-rejection and 77 Marmont AM, Horowitz MM, Gale RP et al. T-cell depletion of toxicity following bone marrow transplantation in relation to HLA-identical transplants in leukemia. Blood 1991; 78: 2120–2130. busulfan pharmacokinetics. Bone Marrow Transplant 1995; 16: 78 Aversa F, Terenzi A, Carotti A et al. Improved outcome 31–42. with T-cell-depleted bone marrow transplantation for acute 61 Vassal G, Deroussent A, Hartmann O et al. Dose-dependent leukemia. J Clin Oncol 1999; 17: 1545–1550. neurotoxicity of high-dose busulfan in children: a clinical and 79 Maris M, Woolfrey A, McSweeney PA et al. Nonmyeloabla- pharmacological study. Cancer Res 1990; 50: 6203–6207. tive hematopoietic stem cell transplantation: transplantation 62 Dix SP, Wingard JR, Mullins RE et al. Association of busulfan for the 21st century. Front Biosci 2001; 6: G13–G16. area under the curve with veno-occlusive disease following 80 McSweeney PA, Niederwieser D, Shizuru JA et al. Hemato- BMT. Bone Marrow Transplant 1996; 17: 225–230. poietic cell transplantation in older patients with hematologic

Bone Marrow Transplantation Conditioning regimens for AML V Gupta et al 978 malignancies: replacing high-dose cytotoxic therapy with graft- cell transplantation with reduced-intensity conditioning in vs-tumor effects. Blood 2001; 97: 3390–3400. acute myelogenous leukemia and myelodysplastic syndromes. 81 Devine SM, Sanborn R, Jessop E et al. Fludarabine and Blood 2002; 100: 2243–2245. melphalan-based conditioning for patients with advanced 84 Feinstein LC, Sandmaier BM, Hegenbart U et al. Non- hematological malignancies relapsing after a previous hema- myeloablative allografting from human leucocyte antigen- topoietic stem cell transplant. Bone Marrow Transplant 2001; identical sibling donors for treatment of acute myeloid 28: 557–562. leukaemia in first complete remission. Br J Haematol 2003; 82 Pawson R, Potter MN, Theocharous P et al. Treatment of 120: 281–288. relapse after allogeneic bone marrow transplantation with 85 Djulbegovic B, Seidenfeld J, Bonnell C et al.Nonmyeloablative reduced intensity conditioning (FLAG +/À Ida) and second allogeneic stem-cell transplantation for hematologic malignan- allogeneic stem cell transplant. Br J Haematol 2001; 115: cies: a systematic review. Cancer Control 2003; 10: 17–41. 622–629. 86 Mielcarek M, Martin PJ, Leisenring W et al. Graft-versus-host 83 Martino R, Caballero MD, Simon JA et al. Evidence for a disease after nonmyeloablative vs conventional hematopoietic graft-vs-leukemia effect after allogeneic peripheral blood stem stem cell transplantation. Blood 2003; 102: 756–762.

Bone Marrow Transplantation