- Email: [email protected]
Outcomes after use of two standard ablative regimens in patients with refractory acute myeloid leukaemia: a retrospective, multicentre, registry analysis
Articles
Outcomes after use of two standard ablative regimens in patients with refractory acute myeloid leukaemia: a retrospective, multicentre, registry analysis Arnon Nagler*, Bipin N Savani*, Myriam Labopin, Emmanuelle Polge, Jakob Passweg, Jürgen Finke, Slawomira Kyrcz-Krzemien, Liisa Volin, Achilles Anagnostopoulos, Mahmoud Aljurf, Dietrich W Beelen, Stephane Vigouroux, Noel Milpied, Felipe Suarez, Mohamad Mohty
Summary Background Cyclophosphamide plus intravenous busulfan has not been compared with cyclophosphamide plus total body irradiation (TBI) in adults with advanced refractory acute myeloid leukaemia before allogeneic haemopoietic stem-cell transplantation (HCT). We aimed to assess whether survival of patients receiving ablative intravenous busulfan-based conditioning regimens before a related or volunteer-unrelated donor HCT for refractory acute myeloid leukaemia is not inferior to that of patients receiving an ablative TBI-based regimen. Methods In this retrospective, multicentre, registry-based study, we obtained data for patients (aged >18 years) with refractory acute myeloid leukaemia in active phase of disease, who had received HCT from an HLA-identical sibling or an unrelated donor after intravenous busulfan plus cyclophosphamide or cyclophosphamide plus TBI conditioning between 2000 and 2012. Data was obtained from the European Group for Blood and Marrow Transplantation registry. The primary endpoints of the study were overall survival and leukaemia-free survival. Findings We obtained data for 514 patients who had received intravenous busulfan plus cyclophosphamide and 338 patients who had received cyclophosphamide plus TBI. The median percentage of blasts before HCT did not differ significantly between groups (20% [range 5–100; IQR 10–32] in the intravenous busulfan plus cyclophosphamide group vs 16% [5–95; 9–33] in the cyclophosphamide plus TBI group; p=0·16). Overall survival at 2 years did not differ between the groups in the univariate analysis (31·2% [95% CI 26·8–35·5] with intravenous busulfan plus cyclophosphamide vs 33·4% [28·1–38·7] wth cyclophosphamide plus TBI; p=0·65). Leukaemia-free survival at 2 years also did not differ between groups (25·0% [95% CI 21·0–29·0] vs 28·4% [23·4–33·5]; p=0·47). In multivariable analysis adjusting for differences between both groups, no difference was noted between the two groups in terms of overall survival (hazard ratio [HR] 0·99 [95% CI 0·83–1·20]; p=0·95) or leukaemia-free survival (HR 0·97 [0·81–1·16]; p=0·71). Main causes of non-relapse mortality were graft-versus-host disease (49 [10%] in the intravenous busulfan plus cyclophosphamide group vs 25 [7%] in the cyclophosphamide plus TBI group) and infection (36 [7%] vs 18 [5%]). Interpretation From a practical standpoint, the use of intravenous busulfan plus cyclophosphamide is likely to be a valid and efficient alternative to cyclophosphamide plus TBI conditioning regimen for patients with refractory acute myeloid leukaemia, especially for those transplant centres without access to radiation facilities. Funding None.
Introduction In patients with poor-prognosis acute myeloid leukaemia, allogeneic haemopoietic stem-cell transplantation (HCT) can be done in those who have achieved complete remission to consolidate their response to chemotherapy and prevent future relapse. Many patients, however, will not receive HCT because they are unable to achieve complete remission after chemotherapy as a result of resistant or rapidly progressive disease. The fate of such patients is not well described, despite expansion in the range of subsequent therapies available to patients who do not respond to induction therapy, and the increasing availability of HCT.1–6 The traditional preparative ablative regimens for patients with acute myeloid leukaemia include cyclophosphamide combined with total body irradiation (TBI) or the combination of busulfan and cyclophosphamide.7,8 The www.thelancet.com/haematology Vol 2 September 2015
development of these two widely used ablative preparative regimens for HCT occurred largely in parallel.8,9 Several retrospective registry-based studies and randomised studies compared the two preparative regimens for HCT in patients with acute myeloid leukaemia and reported conflicting results regarding outcome and toxic effects.10–21 In the past decade, intravenous busulfan has increasingly replaced oral busulfan in conditioning regimens for HCT. Intravenous busulfan is associated with more predictable pharmacokinetics than is oral busulfan and, in some studies, has improved the tolerability of ablative busulfan plus cyclophosphamide.12,13,22,23 A large retrospective study12 in patients with acute myeloid leukaemia in first remission reported significantly lower incidences of non-relapse mortality and late relapse and better leukaemia-free
Lancet Haematol 2015; 2: e384–92 Published Online August 25, 2015 http://dx.doi.org/10.1016/ S2352-3026(15)00146-5 See Comment page e354 *Joint first authors Hematology Division, Chaim Sheba Medical Center, Tel Hashomer, Israel (Prof A Nagler MD); European Group for Blood and Marrow Transplantation (EBMT) Paris Study Office/CEREST-TC, Paris, France (Prof A Nagler, M Labopin PhD, E Polge MSc, Prof M Mohty MD); Vanderbilt University Medical Center, Nashville, TN, USA (Prof B N Savani MD); Department of Haematology, Saint Antoine Hospital, Paris, France (M Labopin, E Polge, Prof M Mohty); INSERM UMR 938, Paris, France (M Labopin, E Polge, Prof M Mohty); Université Pierre et Marie Curie, Paris, France (M Labopin, E Polge, Prof M Mohty); University Hospital, Basel, Switzerland (Prof J Passweg MD); Department of Medicine, Division of Hematology, Oncology and Stem Cell Transplantation, University of Freiburg, Freiburg, Germany (Prof J Finke MD); University Department of Haematology and BMT, Medical University of Silesia, Katowice, Poland (Prof S Kyrcz-Krzemien MD); Helsinki University Central Hospital, Comprehensive Cancer Center, Helsinki, Finland (Prof L Volin MD); Haematology Department/BMT Unit, George Papanicolaou General Hospital, Thessaloniki, Greece (Prof A Anagnostopoulos MD); Department of Oncology, Section of Adult Haematology/ BMT, King Faisal Specialist Hospital and Research Centre Riyadh, Saudi Arabia (M Aljurf MD); Department of Bone Marrow Transplantation,
e384
Articles
University Hospital, Essen, Germany (Prof D W Beelen MD); Department of Haematology and Cell Therapy, University Hospital, Bordeaux, France (S Vigouroux MD, Prof N Milpied MD); and Department of Adult Hematology, Necker University Hospital, INSERM U1163, Institut Imagine, Sorbonne Paris Cité and Université Paris Descartes, Paris, France (Prof F Suarez MD) Correspondence to: Dr Bipin N Savani, Vanderbilt University Medical Center, Nashville, TN 37232, USA [email protected]
Research in context Evidence before this study We searched PubMed with the search terms “acute myeloid leukemia”, “advanced”, “refractory, and ”transplantation” for articles published between Jan 1, 1975, and Oct 3, 2014 (the date of the last search). Cyclophosphamide plus total body irradiation (TBI) is regarded as the standard ablative preparative regimen for eligible patients with advanced or refractory acute myeloid leukaemia. However, no previous evidence has shown that cyclophosphamide plus TBI might yield better outcomes compared with busulfan plus cyclophosphamide in patients with refractory acute myeloid leukaemia. Added value of this study This study is the largest so far into outcomes in patients with refractory acute myeloid leukaemia after receiving two widely used myeloablative conditioning regimens. This large retrospective registry study shows patients with refractory
survival and overall survival with intravenous busulfan than with TBI, and a prospective cohort analysis in patients with myelodysplastic syndrome, acute myeloid leukaemia, and chronic myeloid leukaemia reported better overall survival after intravenous busulfan than after TBI.13 Similarly, European Group for Blood and Marrow Transplantation (EBMT) data showed that the outcome of acute myeloid leukaemia in patients in first or second complete remission was not statistically different after intravenous busulfan plus cyclophosphamide when compared with a cyclophosphamide plus TBI regimen.10 By default, cyclophosphamide plus TBI is regarded as the standard ablative preparative regimen for eligible patients with advanced or refractory acute myeloid leukaemia. No previous evidence has shown that cyclophosphamide plus TBI might yield better leukaemia-free survival or overall survival than busulfan plus cyclophosphamide in patients with refractory acute myeloid leukaemia. We aimed to test the hypothesis that survival of patients receiving ablative intravenous busulfan-based conditioning regimens before a related or volunteer-unrelated donor HCT for refractory acute myeloid leukaemia is not inferior to that of patients receiving an ablative TBI-based regimen.
Methods Study design and patients This was a retrospective, multicentre, registry-based analysis. Data were provided by the Acute Leukemia Working Party (ALWP) of the EBMT registry. The EBMT registry is a voluntary working group of more than 500 transplant centres, mostly located in Europe, that are required to report all consecutive stem-cell transplantations and follow-up data once a year. Audits are routinely undertaken to establish the accuracy of the data. Since 1990, patients have been able to provide e385
acute myeloid leukaemia have similar outcomes after receiving cyclophosphamide plus intravenous busulfan or cyclophosphamide plus TBI. About a third of patients with refractory acute myeloid leukaemia achieved long-term survival with either intravenous busulfan plus cyclophosphamide or cyclophosphamide plus TBI conditioning regimen. Implications of all the available evidence Intravenous busulfan is probably a valid and efficient alternative to high-dose TBI for patients with refractory acute myeloid leukaemia, especially those treated in transplant centres without access to radiation facilities. Moreover, when considering long-term side-effects after allogeneic haemopoietic stem cell transplantation (HCT), the use of ablative dose TBI is now well established to be associated with late complications after HCT.
informed consent that authorises the use of their personal information for research purposes. The ALWP of the EBMT granted ethical approval for this study. Eligibility criteria for this analysis included adult patients (aged >18 years) with refractory acute myeloid leukaemia in active phase of disease who had received HCT from an HLA-identical sibling or a unrelated donor (9/10 or 10/10) with bone marrow or granulocyte colonystimulating factor-mobilised peripheral blood stem cells after intravenous busulfan plus cyclophosphamide or cyclophosphamide plus TBI conditioning between 2000 and 2012. All unrelated donors were HLA-matched (10/10) or mismatched at one loci (9/10) (-A, -B, -C, DRB1, -DQB1). We excluded patients who had undergone haploidentical or umbilical cord blood HCT so that our analysis was restricted to homogeneous study populations.
Procedures We collected data for recipient and donor characteristics (age, sex, cytomegalovirus serostatus), disease features (including primary refractory, first relapse or second relapse refractory disease), transplant-related factors such as the conditioning regimen (intravenous busulfan plus cyclophosphamide vs cyclophosphamide plus TBI), immunosuppression (in-vivo T-cell depletion vs none), stem-cell source (bone marrow vs peripheral blood), graft-versus-host disease prophylaxis (ciclosporin plus methotrexate vs others), and outcome variables (acute and chronic graft-versus-host disease, relapse, nonrelapase mortality, leukaemia-free survival, overall survival, and causes of death).
Outcomes The primary endpoints of the study were overall survival and leukaemia-free survival. Secondary endpoints included complete remission after transplantation, disease relapse www.thelancet.com/haematology Vol 2 September 2015
Articles
incidence, non-relapse mortality, engraftment, incidences and severity of acute and chronic graft-versus-host disease, and risk of liver veno-occlusive disease.
Statistical analysis Our sample size was defined by the number of eligible patients within the EMBT registry who met all the inclusion criteria. The starting point for time-to-event analysis was the date of transplantation. We defined overall survival as the time to death from any cause. We censored surviving patients at time of last follow-up. We defined leukaemia-free survival as survival without relapse or progression. Patients surviving in continuous complete remission were censored at time of last followup. We defined relapse incidence as time to onset of leukaemia recurrence. Non-relapse mortality was the competing risk, and patients surviving in continuous complete remission were censored at last contact. We defined non-relapse mortality as death without relapse or progression (relapse was the competing risk). We used standard definitions to define refractory acute myeloid leukaemia: primary refractory (ie, active disease after induction chemotherapy regimens without previous remission), first relapse (ie, refractory disease after first relapse), and second relapse (ie, refractory disease after second or subsequent relapses). We compared the two groups with the χ² method for qualitative variables, whereas we applied the MannWhitney test for continuous variables. We did univariate comparisons using the log-rank test for overall survival and leukaemia-free survival, and the Gray’s test for relapse incidence, non-relapse mortality, and graftversus-host disease cumulative incidences. We did multivariable analyses with logistic regression to compare the proportion of patients who achieved complete remission, and Cox proportional hazards model for all other endpoints, since we were interested in the effect of conditioning on the instantaneous risk of each event for patients remaining at risk.24 All factors known as potentially related to the outcome were included in the final model. All tests were two-sided. The type I error rate was fixed at 0·05 for determination of factors associated with time to event outcomes. We did statistical analyses with SPSS (version 22.0) and R (version 3.1.1) software packages.
the study period was 190 (range 2–855; IQR 123–283); in the busulfan plus cyclophosphamide group it was 215 (range 6–596; IQR 118–306) and in the cyclophosphamide plus TBI group it was 189 (2–855; 128–269). 852 patients with refractory acute myeloid leukaemia were included in the study. 514 patients received intravenous busulfan plus cyclophosphamide and
Intravenous busulfan plus cyclophosphamide (n=514)
The funders had no role in study design, data analysis, data interpretation, or writing of the report. ML, AN, EP, and MM had full access to all the data in the study. AN, BNS, and MM had final responsibility for the decision to submit for publication.
Results We obtained data from 206 reporting centres, 56 of whom used both conditioning regimens (appendix p 4). The median number of HCTs done in each centre during www.thelancet.com/haematology Vol 2 September 2015
Cyclophosphamide plus TBI (n=338)
Follow up (IQR), months
24 (11–58)
51 (11–85)
Age (range), years
43 (18–68)
39 (18–62)
Sex 288 (56%)
167 (49%)
Female patients
226 (44%)
171 (51%)
KPS at HCT (range; IQR)
p value
0·002 0·001 0·054
Male patients
90% (70–100; 80–90)
90% (70–100; 80–90)
0·87
Primary refractory
147 (44–415; 98–223)
142 (46–385; 101–190)
0·42
First relapse
275 (89–2760; 190–467)
212 (91–1925; 166–421)
0·02
Second relapse
473 (150–2328; 222–649)
439 (130–5118; 299–669)
0·81
Interval from diagnosis to HCT (range; IQR), days
Blasts in bone marrow at HCT (range; IQR)
20% (5–100; 10–52)
16% (5–95; 9–33)
Disease status at HCT 249 (48%)
160 (47%)
First relapse
207 (40%)
150 (44%)
58 (11%)
28 (8%)
32 (6%)
28 (8%)
159 (31%)
112 (33%)
Second relapse Cytogenetic risk classification Good Intermediate Poor Normal or unknown
0·16 0·26
Primary refractory
0·10
51 (10%)
48 (14%)
272 (53%)
150 (44%)
Secondary AML
74 (14%)
38 (11%)
0·18
Donor female to male recipient
97 (19%)
65 (19%)
0·88
Patient CMV serology positive
343/470 (73%)
210/323 (65%)
Donor CMV serology positive
276/456 (61%)
158/319 (50%)
Donor type 359 (70%)
206 (61%)
Unrelated (10/10)
109 (21%)
97 (29%)
Unrelated (9/10)
46 (9%)
35 (10%)
423 (82%)
270 (80%)
91 (18%)
68 (20%)
Graft Peripheral blood
0·00076 424/496 (85%)
243/319 (76%)
72/496 (15%)
76/319 (24%)
No
379/494 (77%)
231/319 (72%)
Yes
115/494 (23%)
88/319 (28%)
Other
0·0024
0·38
GVHD prophylaxis Ciclosporin + methotrexate
0·016 0·021
HLA-identical sibling
Bone marrow
Role of the funding source
See Online for appendix
In-vivo T-cell depletion
0·17
Data are median (IQR), median (range), median (range; IQR), n (%), or n/N (%). Some percentages do not add up to 100% because of rounding. TBI=total body irradiation. KPS=Karnofsky performance score. HCT=allogeneic haemopoietic stemcell transplantation. AML=acute myeloid leukaemia. CMV=cytomegalovirus. GVHD=graft-versus-host disease.
Table 1: Patients and disease characteristics
e386
Articles
338 patients received cyclophosphamide plus TBI conditioning regimen before HCT. The median total busulfan dose was 12·8 mg/kg (range 9·7–14; IQR 12·6–12·8). Our analysis focused on the intravenous formulation of busulfan and we do not have details about busulfan concentrations or area under the curve in the registry data. The median TBI dose was 12 Gy (IQR 12–12); only 46 (14%) patients received a TBI dose of greater than 12 Gy. The intravenous busulfan plus cyclophosphamide group was significantly older than the cyclophosphamide plus Intravenous busulfan plus cyclophosphamide (n=514)
Cyclophosphamide plus p value TBI (n=338)
485 (94%)
330 (98%)
Grade 0–I
345/489 (71%)
212/323 (66%)
Grade II–IV
144/489 (29%)
111/323 (34%)
Engraftment Acute GVHD
0·018 0·14
Outcome at 2 years Overall survival
31·2% (26·8–35·5)
33·4% (28·1–38·7)
0·65
Leukaemia-free survival
25·0% (21·0–29·0)
28·4% (23·4–33·5)
0·47
Relapse
53·5% (48·9–57·9)
54·0% (48·3–59·3)
0·55
Non-relapse mortality
21·5% (17·9–25·2)
17·5% (14·2–21·0)
0·15
29·5% (24·2–34·9)
0·98
Chronic GVHD
30·4% (26·0–34·8)
Complete remission after HCT*
354 (69%)
211 (62%)
0·051
Data are n (%), n/N (%), or n (%; 95% CI), unless otherwise specified. TBI=total body irradiation. GVHD=graft-versushost disease. HCT=allogeneic haemopoietic stem-cell transplantation. *Within 100 days of transplant only; no information about complete remission after day 100 was available in the registry, since this information is asked at the first report (100 days after transplantation); however, we expect that the results—in terms of leukaemia-free and overall survival—take into account, at least in part, the proportion of late complete remission after transplantation.
Table 2: Transplantation outcomes
Intravenous busulfan plus Cyclophosphamide plus TBI cyclophosphamide (n=514) (n=338)
p value
Primary refractory Overall survival
35·3% (28·8–41·9)
39·7% (31·6–47·7)
0·36
Leukaemia-free survival
28·1% (22·0–34·2)
35·2% (27·5–43·0)
0·33
Relapse
49·0% (42·3–55·5)
48·2% (40·0–56·0)
0·67
Non-relapse mortality
22·9% (17·6–28·5)
16·5% (12·0–21·6)
0·15
Chronic GVHD
31·0% (24·5–37·6)
30·7% (22·9–38·9)
0·77
Overall survival
29·3% (22·7–35·9)
28·8% (21·1–36·4)
0·63
Leukaemia-free survival
24·1% (17·9–30·2)
23·5% (16·5–30·6)
0·80
Relapse
56·0% (48·6–62·7)
57·1% (48·5–64·9)
0·71
Non-relapse mortality
20·0% (14·7–25·8)
19·2% (14·0–24·9)
0·86
Chronic GVHD
30·7% (24·0–37·7)
28·8% (21·2–36·9)
0·67
Overall survival
21·3% (10·0–32·6)
25·0% (9·0–41·0)
0·88
Leukaemia-free survival
15·3% (5·6–25·1)
17·9% (3·7–32·0)
0·55
Relapse
63·4% (48·8–74·9)
67·9% (46·2–82·3)
0·92
Non-relapse mortality
21·3% (11·6–32·9)
14·3% (6·7–24·7)
0·37
Chronic GVHD
26·8% (14·3–40·9)
25·9% (10·7–44·2)
0·76
First relapse
Second relapse
Data are % (95% CI), unless otherwise specified. TBI=total body irradiation. GVHD=graft-versus-host disease.
Table 3: Outcomes at 2 years, by phase of disease at transplantation
e387
TBI group (table 1). The median year of transplant for patients in the cyclophosphamide plus TBI group was 2007 (range 2000–2012), whereas patients in the intravenous busulfan plus cyclophosphamide group underwent HCT more recently (median 2008, range 2000–2012; p<0·0001). The median follow-up of all patients was 32 months (IQR 11–70), but median follow-up in the intravenous busulfan plus cyclophosphamide group was significantly shorter than in the cyclophosphamide plus TBI group (table 1). This difference of 17 months might be explained by the median date of transplant in the two groups (July, 2008 [IQR July, 2004, to November, 2010] in the busulfan plus cyclophosphamide group vs February, 2007 [June, 2003, to October, 2009] in the cyclophosphamide plus TBI group). All patients were refractory in active phase of disease at time of HCT as per study inclusion criteria. Median proportion of blasts (in the bone marrow) before HCT did not differ significantly between groups (table 1). 468 (91%) patients received matched related or unrelated donor HCT in the intravenous busulfan plus cyclophosphamide group versus 303 (90%) in the cyclophosphamide plus TBI group. The intravenous busulfan plus cyclophosphamide group contained more cytomegalovirus-positive recipients and recipients with cytomegalovirus-positive donors than did the cyclophosphamide plus TBI group (table 1). The source of stem cells did not differ between the two groups. Ciclosporin plus methotrexate was used as graft-versushost disease prophylaxis more frequently in the intravenous busulfan plus cyclophosphamide group than in the cyclophosphamide plus TBI group and the proportion of patients who received in-vivo T-cell depletion did not significantly differ between the groups (table 1). Background demographic and disease characteristics for only patients treated at one of the 56 centres that provided both conditioning regimens are presented in the appendix p 4. A smaller proportion of patients in the intravenous busulfan plus cyclophosphamide group engrafted than did those in the cyclophosphamide plus TBI group (table 2). Absolute neutrophil count became greater than 500 cells per μL a median of 16 days (range 8–46) after transplantation in the intravenous busulfan plus cyclophosphamide group and after a median of 16 days (7–43) in the cyclophosphamide plus TBI group (p=0·24). The incidence of grade II–IV (table 2) and grade III–IV acute graft-versus-host disease (67 [14%] of 489 patients in the busulfan plus cyclophosphamide group vs 41 [13%] of 323 in the cyclophosphamide plus TBI group; p=0·45) did not significantly differ between the groups. Overall survival at 2 years did not differ between the conditioning regimens in univariate analysis (tables 2 and 3; figure 1A). These results were also confirmed by multivariable analysis (table 4). Multivariable analysis also showed lower overall survival for patients who received the transplant in first or second relapse compared with those who received the transplant in primary refractory disease (table 4). Leukaemia-free survival at 2 years also did not differ between the two www.thelancet.com/haematology Vol 2 September 2015
Articles
Discussion Results of this large retrospective analysis show that patients with refractory acute myeloid leukaemia who received one of two widely used conditioning regimens— intravenous busulfan plus cyclophosphamide or www.thelancet.com/haematology Vol 2 September 2015
A 100
Busulfan + cyclophosphamide Cyclophosphamide + TBI
90 80 Overall survival (%)
70 60 50 40 30 20 10 0
Number at risk Busulfan + cyclophosphamide Cyclophosphamide + TBI
0
1
2
3
4
5
514 338
187 127
110 89
80 79
62 64
45 50
0
1
2
3
4
5
49 55
36 45
B 100 90 80 Leukaemia-free survial (%)
conditioning regimens in univariate analysis (tables 2 and 3; figure 1B) or multivariable analysis (table 4). In multivariable analysis, leukaemia-free survival was lower for patients who received the transplant in first relapse or second relapse compared with patients who received the transplant in in primary refractory disease (table 4). The incidence of chronic graft-versus-host disease was similar between the groups (tables 2 and 3, figure 2A). In multivariable analysis, incidence of chronic graft-versushost disease was not significantly different between the intravenous busulfan plus cyclophosphamide and the cyclophosphamide plus TBI groups (table 4); factors associated with chronic graft-versus-host disease are summarised are in table 4. No difference in relapse incidence was noted between the conditioning regimens in univariate analysis (tables 2 and 3; figure 2B). In the multivariate analysis, risk of relapse was significantly higher for patients who received the transplant in active disease after first or second relapse compared with those who received the transplant with primary refractory disease (table 4). When investigating the effect of chronic graft-versus-host disease on the incidence of relapse using Cox with timedependent variables (univariate analysis), we did not find a significant association between the two events (HR 0·72, 95% CI 0·48–1·08; p=0·11). We noted no difference in non-relapse mortality between the conditioning regimens in univariate analysis (tables 2 and 3; figure 2C). However, multivariable analysis showed that the cyclophosphamide plus TBI regimen was associated with a lower non-relapse mortality risk than was intravenous busulfan plus cyclophosphamide (table 4). Additionally, age at HCT was found to be an independent predictive factor for non-relapse mortality in the multivariable analysis (table 4). We noted no effect for year of transplant, disease status at transplant (primary refractory vs first or second relapse), donor types (related HLA-identical vs unrelated 10/10 or 9/10 HLA-matched), patient or donor cytomegalovirus status, graft-versus-host disease prophylaxis, or in-vivo T-cell depletion (table 4). Main causes of non-relapse mortality were graft-versushost disease and infection complications (table 5). The proportion of deaths from organ toxic effects was very low; however, venocclusive disease of the liver accounted for 16 (5%) of 340 deaths in the intravenous busulfan plus cyclophosphamide group and five (2%) of 225 deaths in the cyclophosphamide plus TBI group (p=0·12). Only four (1%) patients in the intravenous busulfan plus cyclophosphamide group and one (<1%) patient in the cyclophosphamide plus TBI group died from bleeding complications (table 5).
70 60 50 40 30 20 10 0
Number at risk Busulfan + cyclophosphamide Cyclophosphamide + TBI
Time from transplant (years) 514 338
143 101
89 76
66 65
Figure 1: Overall and leukaemia-free survival after HCT, according to conditioning regimen Overall (A) and leukaemia-free (B) survival after conditioning with intravenous busulfan-cyclophosphamide versus cyclophosphamide plus TBI. HCT=allogeneic haemopoietic stem-cell transplantation. TBI=total body irradiation.
cyclophosphamide plus TBI—before HCT had similar leukaemia-free survival and overall survival afterwards. This result supports findings from a retrospective study10 from the EBMT in acute myeloid leukaemia in patients in first and second complete remission, in which outcomes were similar in patients who had received cyclophosphamide plus TBI and those who had received intravenous busulfan plus cyclophosphamide. A major finding in this study is that about a third of patients with refractory acute myeloid leukaemia can achieve longterm survival after either an intravenous busulfan plus cyclophosphamide or a cyclophosphamide plus TBI conditioning regimen. Outcomes in patients with primary refractory acute myeloid leukaemia after HCT were better than in those with refractory disease after first or second relapse. This information is important since many centres in developed countries still do not e388
Articles
HR (95% CI)
p value
Overall survival Cyclophosphamide + TBI vs busulfan + cyclophosphamide
HR (95% CI)
p value
(Continued from previous column) 0·99 (0·83–1·20)
0·95
Donor CMV serology positive
1·15 (0·92–1·45)
0·23
0·84 (0·64–1·10)
0·21
1·31 (0·95–1·81)
0·095
0·69 (0·49–0·99)
0·044
Age at HCT (per 10 years)
1·01 (1·00–1·01)
0·21
GVHD prophylaxis (ciclosporin + methotrexate)
Year of HCT
0·98 (0·96–1·01)
0·27
In-vivo T-cell depletion Non-relapse mortality
Disease at HCT Primary refractory
1 (reference)
First relapse
1·19 (0·99–1·44)
0·070
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
Second relapse
1·50 (1·12–1·99)
0·0056
Age at HCT (per 10 years)
1·24 (1·07–1·45)
0·0052
Year of HCT
0·96 (0·92–1·01)
0·17
Donor
Disease at HCT
HLA-identical sibling
1 (reference)
Unrelated (10/10)
0·93 (0·71–1·23)
0·63
Primary refractory
1 (reference)
Unrelated (9/10)
1·12 (0·77–1·61)
0·56
First relapse
1·06 (0·75–1·50)
0·73
0·97 (0·70–1·35)
0·85
Second relapse
1·13 (0·64–1·98)
0·68
Secondary AML
Donor
Patient CMV serology positive
1·06 (0·85–1·31)
0·63
Donor CMV serology positive
1·11 (0·90–1·36)
0·32
HLA-identical sibling
1 (reference)
0·42
Unrelated (10/10)
1·02 (0·61–1·68)
0·95
Unrelated (9/10)
1·61 (0·85–3·04)
0·14
Secondary AML
1·09 (0·62–1·90)
0·77
Patient CMV serology positive
1·05 (0·71–1·55)
0·80
GVHD prophylaxis (ciclosporin + methotrexate)
0·91 (0·71–1·15)
In-vivo T-cell depletion
1·15 (0·86–1·53)
0·34
Leukaemia-free survival Cyclophosphamide + TBI vs busulfan + cyclophosphamide
0·97 (0·81–1·16)
0·71
Donor CMV serology positive
0·97 (0·68–1·40)
0·89 0·51
1·01 (0·94–1·09)
0·82
GVHD prophylaxis (ciclosporin + methotrexate)
1·15 (0·75–1·75)
Age at HCT (per 10 years) Year of HCT
0·99 (0·96–1·02)
0·43
In-vivo T-cell depletion
0·93 (0·55–1·56)
0·78
0·92 (0·68–1·24)
0·59
Chronic GVHD
Disease at HCT Primary refractory
1 (reference)
First relapse
1·20 (1·00–1·44)
0·053
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
Second relapse
1·54 (1·17–2·03)
0·0020
Age at HCT (per 10 years)
1·11 (0·98–1·27)
0·10
Year of HCT
0·99 (0·95–1·04)
0·80
Donor HLA-identical sibling
1 (reference)
Unrelated (10/10)
0·87 (0·67–1·13)
Disease at HCT 0·30
Primary refractory
1 (reference) 1·08 (0·80–1·46)
0·61
1·31 (0·81–1·12)
0·27
0·95 (0·67–1·35)
0·77
First relapse
Secondary AML
0·88 (0·63–1·22)
0·44
Second relapse
Patient CMV serology positive
1·03 (0·84–1·26)
0·79
Donor CMV serology positive
1·10 (0·91–1·34)
0·32
HLA-identical sibling
1 (reference)
0·46
Unrelated (10/10)
1·48 (0·97–2·24)
0·07
Unrelated (9/10)
1·41 (0·73–2·73)
0·31
Secondary AML
0·99 (0·60–1·64)
0·96
Patient CMV serology positive
0·90 (0·64–1·28)
0·56
Unrelated (9/10)
GVHD prophylaxis (ciclosporin + methotrexate)
0·92 (0·73–1·15)
In-vivo T-cell depletion
1·19 (0·91–1·57)
0·20
Relapse
Donor
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
1·08 (0·88–1·33)
0·46
Donor CMV serology positive
0·96 (0·69–1·35)
0·82 0·39
0·93 (0·85–1·02)
0·14
GVHD prophylaxis (ciclosporin + methotrexate)
1·18 (0·81–1·71)
Age at HCT (per 10 years) Year of HCT
1·00 (1·05–1·69)
0·86
In-vivo T-cell depletion
0·71 (0·44–1·15)
0·17
Complete remission within 100 days of HCT
Disease at HCT
1·39 (1·00–1·92)
0·048
Primary refractory
1 (reference)
First relapse
1·24 (1·00–1·54)
0·050
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
Second relapse
1·73 (1·26–2·38)
0·0007
Age at HCT (per 10 years)
1·02 (0·89–1·18)
0·72
Year of HCT
1·02 (0·97–1·07)
0·49
Donor HLA-identical sibling
1 (reference)
Unrelated (10/10)
0·83 (0·61–1·12)
Unrelated (9/10)
0·77 (0·50–1·18)
Disease at HCT 0·22
Primary refractory
1 (reference)
0·23
First relapse
1·17 (0·84–1·63)
0·36
Second relapse
1·10 (0·64–1·87)
0·73
Secondary AML
0·79 (0·52–1·19)
0·26
Patient CMV serology positive
1·02 (0·80–1·30)
0·87
(Table 4 continues in next column)
(Table 4 continues in next column)
e389
www.thelancet.com/haematology Vol 2 September 2015
Articles
HR (95% CI)
A
p value
(Continued from previous column)
100
HLA-identical sibling
1 (reference)
Unrelated (10/10)
1·04 (0·65–1·67)
0·88
Unrelated (9/10)
0·86 (0·45–1·64)
0·64
Secondary AML
1·36 (0·77–2·42)
0·29
Patient CMV serology positive
0·98 (0·67–1·42)
0·90
Donor CMV serology positive
1·27 (0·89–1·82)
0·19
GVHD prophylaxis (ciclosporin + methotrexate)
0·96 (0·64–1·43)
0·83
In-vivo T-cell depletion
1·33 (0·82–2·15)
0·25
Cumulative incidence of chronic GVHD (%)
Donor
90 80 70 60 50 40 30 20 10
HR=hazard ratio. TBI=total body irradiation. HCT=allogeneic haemopoietic stem-cell transplantation. AML=acute myeloid leukaemia. CMV=cytomegalovirus. GVHD=graft-versus-host disease. Number at risk Busulfan + cyclophosphamide Cyclophosphamide + TBI
0
1
2
3
4
5
441 294
61 54
28 30
20 25
16 22
13 16
1
2
3
4
5
143 101
89 76
66 65
49 55
36 45
1
2
3
4
5
49 55
36 45
B 100 90 Cumulative incidence of relapse (%)
undertake HCT in patients with acute myeloid leukaemia who are not in complete remission, and most patients with refractory acute myeloid leukaemia do not achieve complete remission. Transplants should preferentially be done as soon as necessary, rather than be postponed until a patient achieves complete remission (or reserved for only those who have achieved complete remission). Fewer than 20% of patients with refractory acute myeloid leukaemia will eventually be able to receive HCT in complete remission, because the patient will need to survive the disease and then be fit enough to undergo HCT after having achieved complete remission.25–27 The conditioning regimen has a key role in HCT and has important effects on transplant outcome in patients with refractory acute myeloid leukaemia.2–4,25 Busulfan plus cyclophosphamide and cyclophosphamide plus TBI are the two most widely used conditioning regimens for advanced acute myeloid leukaemia allografts. From the available data, survival does not significantly differ with these two regimens.2,11,28 Either of the regimens can be used for acute myeloid leukaemia allografts, and the choice might ultimately depend on local availability and expertise. The most recent randomised trial comparing busulfan plus cyclophosphamide with cyclophosphamide plus TBI was published more than 15 years ago.29 A gradual decrease in ablative TBI-based transplants and an increase in intravenous busulfan-based transplants have occurred without supportive prospective randomised data comparing the approaches. Our study does not include a comparison with the recently increasingly used intravenous fludarabine plus busulfan regimen, reported to have reduced toxic effects,
0
80 70 60 50 40 30 20 10 0 0
Number at risk Busulfan + cyclophosphamide 514 Cyclophosphamide + TBI 338
C 100 Cumulative incidence of non-relapse mortality (%)
Table 4: Multivariable analysis
Busulfan + cyclophosphamide Cyclophosphamide + TBI
90 80 70 60 50 40 30 20 10 0 0
Figure 2: Secondary outcomes after HCT, according to conditioning regimen Cumulative incidence of chronic GVHD (A), relapse (B), and non-relapse mortality (C) after conditioning with intravenous busulfan plus cyclophosphamide versus cyclophosphamide plus TBI. HCT=allogeneic haemopoietic stem-cell transplantation. TBI=total body irradiation.
www.thelancet.com/haematology Vol 2 September 2015
Number at risk Busulfan + cyclophosphamide 514 Cyclophosphamide + TBI 338
Time from transplant (years) 143 101
89 76
66 65
e390
Articles
Total deaths
Cyclophosphamide Intravenous busulfan plus cyclophosphamide plus TBI (n=338) (n=514)
All patients (n=852)
565 (66%)
340 (66%)
225 (67%)
Cardiac toxic effects
2 (<1%)
1 (<1%)
3 (<1%)
Haemorrhage
4 (1%)
1 (<1%)
5 (1%)
1 (<1%)
0
Graft failure or rejection
1 (<1%)
Venocclusive disease of the liver
16 (3%)
5 (1%)
21 (2%)
Infection
36 (7%)
18 (5%)
54 (6%)
Interstitial pneumonitis
12 (2%)
6 (2%)
18 (2%)
Graft-versus-host disease
49 (10%)
25 (7%)
74 (9%)
205 (40%)
159 (47%)
364 (43%)
Original disease Second malignancy
1 (<1%)
Other transplant-related causes
14 (3%)
0 10 (3%)
1 (<1%) 24 (3%)
Data are number of patients with at least one event (%).
Table 5: Causes of death
because of the small numbers of patients with long-term follow-up. However, we believe that the intensity of a conditioning regimen might have a substantial effect on post-transplantation relapse or disease progression in patients with refractory or active leukaemia. The results of a randomised trial30 suggested that the fludarabine plus busulfan regimen is not a suitable replacement for the busulfan plus cyclophosphamide regimen in adults younger than 60 years who are eligible for ablative conditioning therapy for HCT. Decreasing the intensity of the conditioning regimen is likely to be associated with increased relapse risk or progression, especially in patients with active leukaemia before HCT.2–6,30 Also, in terms of disease control, we did not note a significant difference in long-term leukaemia-free survival between the two conditioning regimens, suggesting that the use of intravenous busulfan is probably associated with a similar drug-versus-leukaemia effect to TBI in patients with refractory acute myeloid leukaemia. These results are of major importance because they support the findings of previous studies,10 but in refractory patients rather than just in patients in remission. Leukaemia-free survival in patients in second complete remission was also similar between the intravenous busulfan plus cyclophosphamide and cyclophosphamide plus TBI groups,10 and now in refractory disease, suggesting a potent anti-leukaemic effect after intravenous busulfan. Thus, from a practical standpoint, intravenous busulfan is probably a valid and efficient alternative to high-dose TBI for patients with acute myeloid leukaemia in remission (first or second) or active disease, especially those treated in transplant centres without access to radiation facilities. Finally, when considering long-term side-effects after HCT, the use of ablative TBI is now well known to play a major role in the onset of diverse late complications after HCT.31–34 Several studies have shown an effect of TBI on cataract formation, azoospermia, diabetes, e391
hypertension, late cardiovascular events, and secondary malignancies.31,32,35–39 The limitations of this study are important to recognise. First, the study is retrospective. Second, patient characteristics vary among the groups for several factors including age, proportion of unrelated donors, and the year in which transplantation was done. Also, we do not have information about why patients were allocated to a specific regimen in the registry and distinguishing the role of conditioning from the role of a potential centre effect is difficult. The aim of our analysis was to compare the two conditioning regimens using EBMT registry data, which allows us to study a large population of patients with refractory acute myeloid leukaemia. Both the design of the study and inclusion criteria were intended to answer this clinical question and therefore are not adapted to develop a prognostic score based on information that is not routinely collected in the registry. However, only through the conduct of well designed clinical trials can we understand and appreciate the complexities of conditioning regimen selection and the associated outcomes after HCT for refractory acute myeloid leukaemia. Unfortunately, no ongoing trials are comparing outcomes after intravenous busulfan plus cyclophosphamide with that after cyclophosphamide plus TBI conditioning regimen for refractory acute myeloid leukaemia. Therefore, in the absence of any prospect of such comparative studies, our data suggest that both regimens are equally effective in patients with refractory acute myeloid leukaemia. Contributors AN, BNS, ML, and MM designed the study or analysed the data (or both). ML, EP, JP, JF, SK-K, LV, AA, MA, DWB, SV, NM, FS, and MM provided important clinical data. All authors contributed to the writing of the report and approved the final version of the Article. Declaration of interests JF has received speaker honoraria from Fresenius, Neovii, and Riemser. All other authors declare no competing interests. Acknowledgments We thank all European Group for Blood and Marrow Transplantation (EBMT) centres and national registries for contributing patients to the study, and data managers for their excellent contribution. References 1 Kanakry JA, Kasamon YL, Gocke CD, et al. Outcomes of related donor HLA-identical or HLA-haploidentical allogeneic blood or marrow transplantation for peripheral T cell lymphoma. Biol Blood Marrow Transplant 2013; 19: 602–06. 2 Feldman EJ, Gergis U. Management of refractory acute myeloid leukemia: re-induction therapy or straight to transplantation? Curr Hematol Malig Rep 2012; 7: 74–77. 3 Hamadani M, Mohty M, Kharfan-Dabaja MA. Reduced-intensity conditioning allogeneic hematopoietic cell transplantation in adults with acute myeloid leukemia. Cancer Control 2011; 18: 237–45. 4 Chemnitz JM, von Lilienfeld-Toal M, Holtick U, et al. Intermediate intensity conditioning regimen containing FLAMSA, treosulfan, cyclophosphamide, and ATG for allogeneic stem cell transplantation in elderly patients with relapsed or high-risk acute myeloid leukemia. Ann Hematol 2012; 91: 47–55. 5 Duval M, Klein JP, He W, et al. Hematopoietic stem-cell transplantation for acute leukemia in relapse or primary induction failure. J Clin Oncol 2010; 28: 3730–38.
www.thelancet.com/haematology Vol 2 September 2015
Articles
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Craddock C, Labopin M, Pillai S, et al. Factors predicting outcome after unrelated donor stem cell transplantation in primary refractory acute myeloid leukaemia. Leukemia 2011; 25: 808–13. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: a randomized trial of two irradiation regimens. Blood 1990; 76: 1867–71. Santos GW, Tutschka PJ, Brookmeyer R, et al. Marrow transplantation for acute nonlymphocytic leukemia after treatment with busulfan and cyclophosphamide. N Engl J Med 1983; 309: 1347–53. Thomas ED, Buckner CD, Banaji M, et al. One hundred patients with acute leukemia treated by chemotherapy, total body irradiation, and allogeneic marrow transplantation. Blood 1977; 49: 511–33. Nagler A, Rocha V, Labopin M, et al. Allogeneic hematopoietic stem-cell transplantation for acute myeloid leukemia in remission: comparison of intravenous busulfan plus cyclophosphamide versus total-body irradiation plus cyclophosphamide as conditioning regimen—a report from the Acute Leukemia Working Party of the European Group for Blood and Marrow Transplantation. J Clin Oncol 2013; 31: 3549–56. Uberti JP, Agovi MA, Tarima S, et al. Comparative analysis of BU and CY versus CY and TBI in full intensity unrelated marrow donor transplantation for AML, CML and myelodysplasia. Bone Marrow Transplant 2011; 46: 34–43. Copelan EA, Hamilton BK, Avalos B, et al. Better leukemia-free and overall survival in AML in first remission following cyclophosphamide in combination with busulfan compared with TBI. Blood 2013; 122: 3863–70. Bredeson C, LeRademacher J, Kato K, et al. Prospective cohort study comparing intravenous busulfan to total body irradiation in hematopoietic cell transplantation. Blood 2013; 122: 3871–78. Copelan EA, Biggs JC, Thompson JM, et al. Treatment for acute myelocytic leukemia with allogeneic bone marrow transplantation following preparation with BuCy2. Blood 1991; 78: 838–43. Blaise D, Maraninchi D, Archimbaud E, et al. 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: 2578–82. Ringden O, Remberger M, Ruutu T, et al, for the Nordic Bone Marrow Transplantation Group. Increased risk of chronic graftversus-host disease, obstructive bronchiolitis, and alopecia with busulfan versus total body irradiation: long-term results of a randomized trial in allogeneic marrow recipients with leukemia. Blood 1999; 93: 2196–201. Ringden O, Ruutu T, Remberger M, et al. 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: 2723–30. Clift RA, Buckner CD, Thomas ED, et al. Marrow transplantation for chronic myeloid leukemia: a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide. Blood 1994; 84: 2036–43. Michel G, Gluckman E, Esperou-Bourdeau H, et al. Allogeneic bone marrow transplantation for children with acute myeloblastic leukemia in first complete remission: impact of conditioning regimen without total-body irradiation—a report from the Societe Francaise de Greffe de Moelle. J Clin Oncol 1994; 12: 1217–22. Hartman AR, Williams SF, Dillon JJ. 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: 439–43. Devergie A, Blaise D, Attal M, et al. 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: 2263–68.
www.thelancet.com/haematology Vol 2 September 2015
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36 37
38
39
Rezvani AR, McCune JS, Storer BE, et al. Cyclophosphamide followed by intravenous targeted busulfan for allogeneic hematopoietic cell transplantation: pharmacokinetics and clinical outcomes. Biol Blood Marrow Transplant 2013; 19: 1033–39. Hassan M, Andersson BS. Role of pharmacogenetics in busulfan/ cyclophosphamide conditioning therapy prior to hematopoietic stem cell transplantation. Pharmacogenomics 2013; 14: 75–87. Latouche A, Allignol A, Beyersmann J, Labopin M, Fine JP. A competing risks analysis should report results on all cause-specific hazards and cumulative incidence functions. J Clin Epidemiol 2013; 66: 648–53. Chen GL, Liu H, Zhang Y, et al. Early versus late preemptive allogeneic hematopoietic cell transplantation for relapsed or refractory acute myeloid leukemia. Biol Blood Marrow Transplant 2014; 20: 1369–74. Jabbour E, Daver N, Champlin R, et al. Allogeneic stem cell transplantation as initial salvage for patients with acute myeloid leukemia refractory to high-dose cytarabine-based induction chemotherapy. Am J Hematol 2014; 89: 395–98. Ravandi F, Cortes J, Faderl S, et al. Characteristics and outcome of patients with acute myeloid leukemia refractory to 1 cycle of high-dose cytarabine-based induction chemotherapy. Blood 2010; 116: 5818–23. Gupta V, Lazarus HM, Keating A. Myeloablative conditioning regimens for AML allografts: 30 years later. Bone Marrow Transplant 2003; 32: 969–78. Socie G, Clift RA, Blaise D, et al. 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: 3569–74. Lee JH, Joo YD, Kim H, et al. Randomized trial of myeloablative conditioning regimens: busulfan plus cyclophosphamide versus busulfan plus fludarabine. J Clin Oncol 2013; 31: 701–09. Savani BN. Blood and marrow transplantation long term management: prevention and complications, 1st edn. Oxford, UK: John Wiley & Sons Ltd, 2014. Mohty M, Malard F, Savani BN. High-dose total body irradiation and myeloablative conditioning before allogeneic stem cell transplantation: time to rethink? Biol Blood Marrow Transplant 2014; 21: 620–24. Savani BN, Griffith ML, Jagasia S, Lee SJ. How I treat late effects in adults after allogeneic stem cell transplantation. Blood 2011; 117: 3002–09. Savani BN. How can we improve life expectancy and quality of life in long-term survivors after allogeneic stem cell transplantation? Semin Hematol 2012; 49: 1–3. Khera N, Storer B, Flowers ME, et al. Nonmalignant late effects and compromised functional status in survivors of hematopoietic cell transplantation. J Clin Oncol 2012; 30: 71–77. Rizzo JD, Curtis RE, Socie G, et al. Solid cancers after allogeneic hematopoietic cell transplantation. Blood 2009; 113: 1175–83. Wei C, Thyagiarajan M, Hunt L, et al. Reduced beta-cell reserve and pancreatic volume in survivors of childhood acute lymphoblastic leukaemia treated with bone marrow transplantation and total body irradiation. Clin Endocrinol (Oxf) 2014; 82: 59–67. Sanders JE, Hoffmeister PA, Woolfrey AE, et al. Thyroid function following hematopoietic cell transplantation in children: 30 years’ experience. Blood 2009; 113: 306–08. Hoffmeister PA, Hingorani SR, Storer BE, Baker KS, Sanders JE. Hypertension in long-term survivors of pediatric hematopoietic cell transplantation. Biol Blood Marrow Transplant 2010; 16: 515–24.
e392
Outcomes after use of two standard ablative regimens in patients with refractory acute myeloid leukaemia: a retrospective, multicentre, registry analysis Arnon Nagler*, Bipin N Savani*, Myriam Labopin, Emmanuelle Polge, Jakob Passweg, Jürgen Finke, Slawomira Kyrcz-Krzemien, Liisa Volin, Achilles Anagnostopoulos, Mahmoud Aljurf, Dietrich W Beelen, Stephane Vigouroux, Noel Milpied, Felipe Suarez, Mohamad Mohty
Summary Background Cyclophosphamide plus intravenous busulfan has not been compared with cyclophosphamide plus total body irradiation (TBI) in adults with advanced refractory acute myeloid leukaemia before allogeneic haemopoietic stem-cell transplantation (HCT). We aimed to assess whether survival of patients receiving ablative intravenous busulfan-based conditioning regimens before a related or volunteer-unrelated donor HCT for refractory acute myeloid leukaemia is not inferior to that of patients receiving an ablative TBI-based regimen. Methods In this retrospective, multicentre, registry-based study, we obtained data for patients (aged >18 years) with refractory acute myeloid leukaemia in active phase of disease, who had received HCT from an HLA-identical sibling or an unrelated donor after intravenous busulfan plus cyclophosphamide or cyclophosphamide plus TBI conditioning between 2000 and 2012. Data was obtained from the European Group for Blood and Marrow Transplantation registry. The primary endpoints of the study were overall survival and leukaemia-free survival. Findings We obtained data for 514 patients who had received intravenous busulfan plus cyclophosphamide and 338 patients who had received cyclophosphamide plus TBI. The median percentage of blasts before HCT did not differ significantly between groups (20% [range 5–100; IQR 10–32] in the intravenous busulfan plus cyclophosphamide group vs 16% [5–95; 9–33] in the cyclophosphamide plus TBI group; p=0·16). Overall survival at 2 years did not differ between the groups in the univariate analysis (31·2% [95% CI 26·8–35·5] with intravenous busulfan plus cyclophosphamide vs 33·4% [28·1–38·7] wth cyclophosphamide plus TBI; p=0·65). Leukaemia-free survival at 2 years also did not differ between groups (25·0% [95% CI 21·0–29·0] vs 28·4% [23·4–33·5]; p=0·47). In multivariable analysis adjusting for differences between both groups, no difference was noted between the two groups in terms of overall survival (hazard ratio [HR] 0·99 [95% CI 0·83–1·20]; p=0·95) or leukaemia-free survival (HR 0·97 [0·81–1·16]; p=0·71). Main causes of non-relapse mortality were graft-versus-host disease (49 [10%] in the intravenous busulfan plus cyclophosphamide group vs 25 [7%] in the cyclophosphamide plus TBI group) and infection (36 [7%] vs 18 [5%]). Interpretation From a practical standpoint, the use of intravenous busulfan plus cyclophosphamide is likely to be a valid and efficient alternative to cyclophosphamide plus TBI conditioning regimen for patients with refractory acute myeloid leukaemia, especially for those transplant centres without access to radiation facilities. Funding None.
Introduction In patients with poor-prognosis acute myeloid leukaemia, allogeneic haemopoietic stem-cell transplantation (HCT) can be done in those who have achieved complete remission to consolidate their response to chemotherapy and prevent future relapse. Many patients, however, will not receive HCT because they are unable to achieve complete remission after chemotherapy as a result of resistant or rapidly progressive disease. The fate of such patients is not well described, despite expansion in the range of subsequent therapies available to patients who do not respond to induction therapy, and the increasing availability of HCT.1–6 The traditional preparative ablative regimens for patients with acute myeloid leukaemia include cyclophosphamide combined with total body irradiation (TBI) or the combination of busulfan and cyclophosphamide.7,8 The www.thelancet.com/haematology Vol 2 September 2015
development of these two widely used ablative preparative regimens for HCT occurred largely in parallel.8,9 Several retrospective registry-based studies and randomised studies compared the two preparative regimens for HCT in patients with acute myeloid leukaemia and reported conflicting results regarding outcome and toxic effects.10–21 In the past decade, intravenous busulfan has increasingly replaced oral busulfan in conditioning regimens for HCT. Intravenous busulfan is associated with more predictable pharmacokinetics than is oral busulfan and, in some studies, has improved the tolerability of ablative busulfan plus cyclophosphamide.12,13,22,23 A large retrospective study12 in patients with acute myeloid leukaemia in first remission reported significantly lower incidences of non-relapse mortality and late relapse and better leukaemia-free
Lancet Haematol 2015; 2: e384–92 Published Online August 25, 2015 http://dx.doi.org/10.1016/ S2352-3026(15)00146-5 See Comment page e354 *Joint first authors Hematology Division, Chaim Sheba Medical Center, Tel Hashomer, Israel (Prof A Nagler MD); European Group for Blood and Marrow Transplantation (EBMT) Paris Study Office/CEREST-TC, Paris, France (Prof A Nagler, M Labopin PhD, E Polge MSc, Prof M Mohty MD); Vanderbilt University Medical Center, Nashville, TN, USA (Prof B N Savani MD); Department of Haematology, Saint Antoine Hospital, Paris, France (M Labopin, E Polge, Prof M Mohty); INSERM UMR 938, Paris, France (M Labopin, E Polge, Prof M Mohty); Université Pierre et Marie Curie, Paris, France (M Labopin, E Polge, Prof M Mohty); University Hospital, Basel, Switzerland (Prof J Passweg MD); Department of Medicine, Division of Hematology, Oncology and Stem Cell Transplantation, University of Freiburg, Freiburg, Germany (Prof J Finke MD); University Department of Haematology and BMT, Medical University of Silesia, Katowice, Poland (Prof S Kyrcz-Krzemien MD); Helsinki University Central Hospital, Comprehensive Cancer Center, Helsinki, Finland (Prof L Volin MD); Haematology Department/BMT Unit, George Papanicolaou General Hospital, Thessaloniki, Greece (Prof A Anagnostopoulos MD); Department of Oncology, Section of Adult Haematology/ BMT, King Faisal Specialist Hospital and Research Centre Riyadh, Saudi Arabia (M Aljurf MD); Department of Bone Marrow Transplantation,
e384
Articles
University Hospital, Essen, Germany (Prof D W Beelen MD); Department of Haematology and Cell Therapy, University Hospital, Bordeaux, France (S Vigouroux MD, Prof N Milpied MD); and Department of Adult Hematology, Necker University Hospital, INSERM U1163, Institut Imagine, Sorbonne Paris Cité and Université Paris Descartes, Paris, France (Prof F Suarez MD) Correspondence to: Dr Bipin N Savani, Vanderbilt University Medical Center, Nashville, TN 37232, USA [email protected]
Research in context Evidence before this study We searched PubMed with the search terms “acute myeloid leukemia”, “advanced”, “refractory, and ”transplantation” for articles published between Jan 1, 1975, and Oct 3, 2014 (the date of the last search). Cyclophosphamide plus total body irradiation (TBI) is regarded as the standard ablative preparative regimen for eligible patients with advanced or refractory acute myeloid leukaemia. However, no previous evidence has shown that cyclophosphamide plus TBI might yield better outcomes compared with busulfan plus cyclophosphamide in patients with refractory acute myeloid leukaemia. Added value of this study This study is the largest so far into outcomes in patients with refractory acute myeloid leukaemia after receiving two widely used myeloablative conditioning regimens. This large retrospective registry study shows patients with refractory
survival and overall survival with intravenous busulfan than with TBI, and a prospective cohort analysis in patients with myelodysplastic syndrome, acute myeloid leukaemia, and chronic myeloid leukaemia reported better overall survival after intravenous busulfan than after TBI.13 Similarly, European Group for Blood and Marrow Transplantation (EBMT) data showed that the outcome of acute myeloid leukaemia in patients in first or second complete remission was not statistically different after intravenous busulfan plus cyclophosphamide when compared with a cyclophosphamide plus TBI regimen.10 By default, cyclophosphamide plus TBI is regarded as the standard ablative preparative regimen for eligible patients with advanced or refractory acute myeloid leukaemia. No previous evidence has shown that cyclophosphamide plus TBI might yield better leukaemia-free survival or overall survival than busulfan plus cyclophosphamide in patients with refractory acute myeloid leukaemia. We aimed to test the hypothesis that survival of patients receiving ablative intravenous busulfan-based conditioning regimens before a related or volunteer-unrelated donor HCT for refractory acute myeloid leukaemia is not inferior to that of patients receiving an ablative TBI-based regimen.
Methods Study design and patients This was a retrospective, multicentre, registry-based analysis. Data were provided by the Acute Leukemia Working Party (ALWP) of the EBMT registry. The EBMT registry is a voluntary working group of more than 500 transplant centres, mostly located in Europe, that are required to report all consecutive stem-cell transplantations and follow-up data once a year. Audits are routinely undertaken to establish the accuracy of the data. Since 1990, patients have been able to provide e385
acute myeloid leukaemia have similar outcomes after receiving cyclophosphamide plus intravenous busulfan or cyclophosphamide plus TBI. About a third of patients with refractory acute myeloid leukaemia achieved long-term survival with either intravenous busulfan plus cyclophosphamide or cyclophosphamide plus TBI conditioning regimen. Implications of all the available evidence Intravenous busulfan is probably a valid and efficient alternative to high-dose TBI for patients with refractory acute myeloid leukaemia, especially those treated in transplant centres without access to radiation facilities. Moreover, when considering long-term side-effects after allogeneic haemopoietic stem cell transplantation (HCT), the use of ablative dose TBI is now well established to be associated with late complications after HCT.
informed consent that authorises the use of their personal information for research purposes. The ALWP of the EBMT granted ethical approval for this study. Eligibility criteria for this analysis included adult patients (aged >18 years) with refractory acute myeloid leukaemia in active phase of disease who had received HCT from an HLA-identical sibling or a unrelated donor (9/10 or 10/10) with bone marrow or granulocyte colonystimulating factor-mobilised peripheral blood stem cells after intravenous busulfan plus cyclophosphamide or cyclophosphamide plus TBI conditioning between 2000 and 2012. All unrelated donors were HLA-matched (10/10) or mismatched at one loci (9/10) (-A, -B, -C, DRB1, -DQB1). We excluded patients who had undergone haploidentical or umbilical cord blood HCT so that our analysis was restricted to homogeneous study populations.
Procedures We collected data for recipient and donor characteristics (age, sex, cytomegalovirus serostatus), disease features (including primary refractory, first relapse or second relapse refractory disease), transplant-related factors such as the conditioning regimen (intravenous busulfan plus cyclophosphamide vs cyclophosphamide plus TBI), immunosuppression (in-vivo T-cell depletion vs none), stem-cell source (bone marrow vs peripheral blood), graft-versus-host disease prophylaxis (ciclosporin plus methotrexate vs others), and outcome variables (acute and chronic graft-versus-host disease, relapse, nonrelapase mortality, leukaemia-free survival, overall survival, and causes of death).
Outcomes The primary endpoints of the study were overall survival and leukaemia-free survival. Secondary endpoints included complete remission after transplantation, disease relapse www.thelancet.com/haematology Vol 2 September 2015
Articles
incidence, non-relapse mortality, engraftment, incidences and severity of acute and chronic graft-versus-host disease, and risk of liver veno-occlusive disease.
Statistical analysis Our sample size was defined by the number of eligible patients within the EMBT registry who met all the inclusion criteria. The starting point for time-to-event analysis was the date of transplantation. We defined overall survival as the time to death from any cause. We censored surviving patients at time of last follow-up. We defined leukaemia-free survival as survival without relapse or progression. Patients surviving in continuous complete remission were censored at time of last followup. We defined relapse incidence as time to onset of leukaemia recurrence. Non-relapse mortality was the competing risk, and patients surviving in continuous complete remission were censored at last contact. We defined non-relapse mortality as death without relapse or progression (relapse was the competing risk). We used standard definitions to define refractory acute myeloid leukaemia: primary refractory (ie, active disease after induction chemotherapy regimens without previous remission), first relapse (ie, refractory disease after first relapse), and second relapse (ie, refractory disease after second or subsequent relapses). We compared the two groups with the χ² method for qualitative variables, whereas we applied the MannWhitney test for continuous variables. We did univariate comparisons using the log-rank test for overall survival and leukaemia-free survival, and the Gray’s test for relapse incidence, non-relapse mortality, and graftversus-host disease cumulative incidences. We did multivariable analyses with logistic regression to compare the proportion of patients who achieved complete remission, and Cox proportional hazards model for all other endpoints, since we were interested in the effect of conditioning on the instantaneous risk of each event for patients remaining at risk.24 All factors known as potentially related to the outcome were included in the final model. All tests were two-sided. The type I error rate was fixed at 0·05 for determination of factors associated with time to event outcomes. We did statistical analyses with SPSS (version 22.0) and R (version 3.1.1) software packages.
the study period was 190 (range 2–855; IQR 123–283); in the busulfan plus cyclophosphamide group it was 215 (range 6–596; IQR 118–306) and in the cyclophosphamide plus TBI group it was 189 (2–855; 128–269). 852 patients with refractory acute myeloid leukaemia were included in the study. 514 patients received intravenous busulfan plus cyclophosphamide and
Intravenous busulfan plus cyclophosphamide (n=514)
The funders had no role in study design, data analysis, data interpretation, or writing of the report. ML, AN, EP, and MM had full access to all the data in the study. AN, BNS, and MM had final responsibility for the decision to submit for publication.
Results We obtained data from 206 reporting centres, 56 of whom used both conditioning regimens (appendix p 4). The median number of HCTs done in each centre during www.thelancet.com/haematology Vol 2 September 2015
Cyclophosphamide plus TBI (n=338)
Follow up (IQR), months
24 (11–58)
51 (11–85)
Age (range), years
43 (18–68)
39 (18–62)
Sex 288 (56%)
167 (49%)
Female patients
226 (44%)
171 (51%)
KPS at HCT (range; IQR)
p value
0·002 0·001 0·054
Male patients
90% (70–100; 80–90)
90% (70–100; 80–90)
0·87
Primary refractory
147 (44–415; 98–223)
142 (46–385; 101–190)
0·42
First relapse
275 (89–2760; 190–467)
212 (91–1925; 166–421)
0·02
Second relapse
473 (150–2328; 222–649)
439 (130–5118; 299–669)
0·81
Interval from diagnosis to HCT (range; IQR), days
Blasts in bone marrow at HCT (range; IQR)
20% (5–100; 10–52)
16% (5–95; 9–33)
Disease status at HCT 249 (48%)
160 (47%)
First relapse
207 (40%)
150 (44%)
58 (11%)
28 (8%)
32 (6%)
28 (8%)
159 (31%)
112 (33%)
Second relapse Cytogenetic risk classification Good Intermediate Poor Normal or unknown
0·16 0·26
Primary refractory
0·10
51 (10%)
48 (14%)
272 (53%)
150 (44%)
Secondary AML
74 (14%)
38 (11%)
0·18
Donor female to male recipient
97 (19%)
65 (19%)
0·88
Patient CMV serology positive
343/470 (73%)
210/323 (65%)
Donor CMV serology positive
276/456 (61%)
158/319 (50%)
Donor type 359 (70%)
206 (61%)
Unrelated (10/10)
109 (21%)
97 (29%)
Unrelated (9/10)
46 (9%)
35 (10%)
423 (82%)
270 (80%)
91 (18%)
68 (20%)
Graft Peripheral blood
0·00076 424/496 (85%)
243/319 (76%)
72/496 (15%)
76/319 (24%)
No
379/494 (77%)
231/319 (72%)
Yes
115/494 (23%)
88/319 (28%)
Other
0·0024
0·38
GVHD prophylaxis Ciclosporin + methotrexate
0·016 0·021
HLA-identical sibling
Bone marrow
Role of the funding source
See Online for appendix
In-vivo T-cell depletion
0·17
Data are median (IQR), median (range), median (range; IQR), n (%), or n/N (%). Some percentages do not add up to 100% because of rounding. TBI=total body irradiation. KPS=Karnofsky performance score. HCT=allogeneic haemopoietic stemcell transplantation. AML=acute myeloid leukaemia. CMV=cytomegalovirus. GVHD=graft-versus-host disease.
Table 1: Patients and disease characteristics
e386
Articles
338 patients received cyclophosphamide plus TBI conditioning regimen before HCT. The median total busulfan dose was 12·8 mg/kg (range 9·7–14; IQR 12·6–12·8). Our analysis focused on the intravenous formulation of busulfan and we do not have details about busulfan concentrations or area under the curve in the registry data. The median TBI dose was 12 Gy (IQR 12–12); only 46 (14%) patients received a TBI dose of greater than 12 Gy. The intravenous busulfan plus cyclophosphamide group was significantly older than the cyclophosphamide plus Intravenous busulfan plus cyclophosphamide (n=514)
Cyclophosphamide plus p value TBI (n=338)
485 (94%)
330 (98%)
Grade 0–I
345/489 (71%)
212/323 (66%)
Grade II–IV
144/489 (29%)
111/323 (34%)
Engraftment Acute GVHD
0·018 0·14
Outcome at 2 years Overall survival
31·2% (26·8–35·5)
33·4% (28·1–38·7)
0·65
Leukaemia-free survival
25·0% (21·0–29·0)
28·4% (23·4–33·5)
0·47
Relapse
53·5% (48·9–57·9)
54·0% (48·3–59·3)
0·55
Non-relapse mortality
21·5% (17·9–25·2)
17·5% (14·2–21·0)
0·15
29·5% (24·2–34·9)
0·98
Chronic GVHD
30·4% (26·0–34·8)
Complete remission after HCT*
354 (69%)
211 (62%)
0·051
Data are n (%), n/N (%), or n (%; 95% CI), unless otherwise specified. TBI=total body irradiation. GVHD=graft-versushost disease. HCT=allogeneic haemopoietic stem-cell transplantation. *Within 100 days of transplant only; no information about complete remission after day 100 was available in the registry, since this information is asked at the first report (100 days after transplantation); however, we expect that the results—in terms of leukaemia-free and overall survival—take into account, at least in part, the proportion of late complete remission after transplantation.
Table 2: Transplantation outcomes
Intravenous busulfan plus Cyclophosphamide plus TBI cyclophosphamide (n=514) (n=338)
p value
Primary refractory Overall survival
35·3% (28·8–41·9)
39·7% (31·6–47·7)
0·36
Leukaemia-free survival
28·1% (22·0–34·2)
35·2% (27·5–43·0)
0·33
Relapse
49·0% (42·3–55·5)
48·2% (40·0–56·0)
0·67
Non-relapse mortality
22·9% (17·6–28·5)
16·5% (12·0–21·6)
0·15
Chronic GVHD
31·0% (24·5–37·6)
30·7% (22·9–38·9)
0·77
Overall survival
29·3% (22·7–35·9)
28·8% (21·1–36·4)
0·63
Leukaemia-free survival
24·1% (17·9–30·2)
23·5% (16·5–30·6)
0·80
Relapse
56·0% (48·6–62·7)
57·1% (48·5–64·9)
0·71
Non-relapse mortality
20·0% (14·7–25·8)
19·2% (14·0–24·9)
0·86
Chronic GVHD
30·7% (24·0–37·7)
28·8% (21·2–36·9)
0·67
Overall survival
21·3% (10·0–32·6)
25·0% (9·0–41·0)
0·88
Leukaemia-free survival
15·3% (5·6–25·1)
17·9% (3·7–32·0)
0·55
Relapse
63·4% (48·8–74·9)
67·9% (46·2–82·3)
0·92
Non-relapse mortality
21·3% (11·6–32·9)
14·3% (6·7–24·7)
0·37
Chronic GVHD
26·8% (14·3–40·9)
25·9% (10·7–44·2)
0·76
First relapse
Second relapse
Data are % (95% CI), unless otherwise specified. TBI=total body irradiation. GVHD=graft-versus-host disease.
Table 3: Outcomes at 2 years, by phase of disease at transplantation
e387
TBI group (table 1). The median year of transplant for patients in the cyclophosphamide plus TBI group was 2007 (range 2000–2012), whereas patients in the intravenous busulfan plus cyclophosphamide group underwent HCT more recently (median 2008, range 2000–2012; p<0·0001). The median follow-up of all patients was 32 months (IQR 11–70), but median follow-up in the intravenous busulfan plus cyclophosphamide group was significantly shorter than in the cyclophosphamide plus TBI group (table 1). This difference of 17 months might be explained by the median date of transplant in the two groups (July, 2008 [IQR July, 2004, to November, 2010] in the busulfan plus cyclophosphamide group vs February, 2007 [June, 2003, to October, 2009] in the cyclophosphamide plus TBI group). All patients were refractory in active phase of disease at time of HCT as per study inclusion criteria. Median proportion of blasts (in the bone marrow) before HCT did not differ significantly between groups (table 1). 468 (91%) patients received matched related or unrelated donor HCT in the intravenous busulfan plus cyclophosphamide group versus 303 (90%) in the cyclophosphamide plus TBI group. The intravenous busulfan plus cyclophosphamide group contained more cytomegalovirus-positive recipients and recipients with cytomegalovirus-positive donors than did the cyclophosphamide plus TBI group (table 1). The source of stem cells did not differ between the two groups. Ciclosporin plus methotrexate was used as graft-versushost disease prophylaxis more frequently in the intravenous busulfan plus cyclophosphamide group than in the cyclophosphamide plus TBI group and the proportion of patients who received in-vivo T-cell depletion did not significantly differ between the groups (table 1). Background demographic and disease characteristics for only patients treated at one of the 56 centres that provided both conditioning regimens are presented in the appendix p 4. A smaller proportion of patients in the intravenous busulfan plus cyclophosphamide group engrafted than did those in the cyclophosphamide plus TBI group (table 2). Absolute neutrophil count became greater than 500 cells per μL a median of 16 days (range 8–46) after transplantation in the intravenous busulfan plus cyclophosphamide group and after a median of 16 days (7–43) in the cyclophosphamide plus TBI group (p=0·24). The incidence of grade II–IV (table 2) and grade III–IV acute graft-versus-host disease (67 [14%] of 489 patients in the busulfan plus cyclophosphamide group vs 41 [13%] of 323 in the cyclophosphamide plus TBI group; p=0·45) did not significantly differ between the groups. Overall survival at 2 years did not differ between the conditioning regimens in univariate analysis (tables 2 and 3; figure 1A). These results were also confirmed by multivariable analysis (table 4). Multivariable analysis also showed lower overall survival for patients who received the transplant in first or second relapse compared with those who received the transplant in primary refractory disease (table 4). Leukaemia-free survival at 2 years also did not differ between the two www.thelancet.com/haematology Vol 2 September 2015
Articles
Discussion Results of this large retrospective analysis show that patients with refractory acute myeloid leukaemia who received one of two widely used conditioning regimens— intravenous busulfan plus cyclophosphamide or www.thelancet.com/haematology Vol 2 September 2015
A 100
Busulfan + cyclophosphamide Cyclophosphamide + TBI
90 80 Overall survival (%)
70 60 50 40 30 20 10 0
Number at risk Busulfan + cyclophosphamide Cyclophosphamide + TBI
0
1
2
3
4
5
514 338
187 127
110 89
80 79
62 64
45 50
0
1
2
3
4
5
49 55
36 45
B 100 90 80 Leukaemia-free survial (%)
conditioning regimens in univariate analysis (tables 2 and 3; figure 1B) or multivariable analysis (table 4). In multivariable analysis, leukaemia-free survival was lower for patients who received the transplant in first relapse or second relapse compared with patients who received the transplant in in primary refractory disease (table 4). The incidence of chronic graft-versus-host disease was similar between the groups (tables 2 and 3, figure 2A). In multivariable analysis, incidence of chronic graft-versushost disease was not significantly different between the intravenous busulfan plus cyclophosphamide and the cyclophosphamide plus TBI groups (table 4); factors associated with chronic graft-versus-host disease are summarised are in table 4. No difference in relapse incidence was noted between the conditioning regimens in univariate analysis (tables 2 and 3; figure 2B). In the multivariate analysis, risk of relapse was significantly higher for patients who received the transplant in active disease after first or second relapse compared with those who received the transplant with primary refractory disease (table 4). When investigating the effect of chronic graft-versus-host disease on the incidence of relapse using Cox with timedependent variables (univariate analysis), we did not find a significant association between the two events (HR 0·72, 95% CI 0·48–1·08; p=0·11). We noted no difference in non-relapse mortality between the conditioning regimens in univariate analysis (tables 2 and 3; figure 2C). However, multivariable analysis showed that the cyclophosphamide plus TBI regimen was associated with a lower non-relapse mortality risk than was intravenous busulfan plus cyclophosphamide (table 4). Additionally, age at HCT was found to be an independent predictive factor for non-relapse mortality in the multivariable analysis (table 4). We noted no effect for year of transplant, disease status at transplant (primary refractory vs first or second relapse), donor types (related HLA-identical vs unrelated 10/10 or 9/10 HLA-matched), patient or donor cytomegalovirus status, graft-versus-host disease prophylaxis, or in-vivo T-cell depletion (table 4). Main causes of non-relapse mortality were graft-versushost disease and infection complications (table 5). The proportion of deaths from organ toxic effects was very low; however, venocclusive disease of the liver accounted for 16 (5%) of 340 deaths in the intravenous busulfan plus cyclophosphamide group and five (2%) of 225 deaths in the cyclophosphamide plus TBI group (p=0·12). Only four (1%) patients in the intravenous busulfan plus cyclophosphamide group and one (<1%) patient in the cyclophosphamide plus TBI group died from bleeding complications (table 5).
70 60 50 40 30 20 10 0
Number at risk Busulfan + cyclophosphamide Cyclophosphamide + TBI
Time from transplant (years) 514 338
143 101
89 76
66 65
Figure 1: Overall and leukaemia-free survival after HCT, according to conditioning regimen Overall (A) and leukaemia-free (B) survival after conditioning with intravenous busulfan-cyclophosphamide versus cyclophosphamide plus TBI. HCT=allogeneic haemopoietic stem-cell transplantation. TBI=total body irradiation.
cyclophosphamide plus TBI—before HCT had similar leukaemia-free survival and overall survival afterwards. This result supports findings from a retrospective study10 from the EBMT in acute myeloid leukaemia in patients in first and second complete remission, in which outcomes were similar in patients who had received cyclophosphamide plus TBI and those who had received intravenous busulfan plus cyclophosphamide. A major finding in this study is that about a third of patients with refractory acute myeloid leukaemia can achieve longterm survival after either an intravenous busulfan plus cyclophosphamide or a cyclophosphamide plus TBI conditioning regimen. Outcomes in patients with primary refractory acute myeloid leukaemia after HCT were better than in those with refractory disease after first or second relapse. This information is important since many centres in developed countries still do not e388
Articles
HR (95% CI)
p value
Overall survival Cyclophosphamide + TBI vs busulfan + cyclophosphamide
HR (95% CI)
p value
(Continued from previous column) 0·99 (0·83–1·20)
0·95
Donor CMV serology positive
1·15 (0·92–1·45)
0·23
0·84 (0·64–1·10)
0·21
1·31 (0·95–1·81)
0·095
0·69 (0·49–0·99)
0·044
Age at HCT (per 10 years)
1·01 (1·00–1·01)
0·21
GVHD prophylaxis (ciclosporin + methotrexate)
Year of HCT
0·98 (0·96–1·01)
0·27
In-vivo T-cell depletion Non-relapse mortality
Disease at HCT Primary refractory
1 (reference)
First relapse
1·19 (0·99–1·44)
0·070
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
Second relapse
1·50 (1·12–1·99)
0·0056
Age at HCT (per 10 years)
1·24 (1·07–1·45)
0·0052
Year of HCT
0·96 (0·92–1·01)
0·17
Donor
Disease at HCT
HLA-identical sibling
1 (reference)
Unrelated (10/10)
0·93 (0·71–1·23)
0·63
Primary refractory
1 (reference)
Unrelated (9/10)
1·12 (0·77–1·61)
0·56
First relapse
1·06 (0·75–1·50)
0·73
0·97 (0·70–1·35)
0·85
Second relapse
1·13 (0·64–1·98)
0·68
Secondary AML
Donor
Patient CMV serology positive
1·06 (0·85–1·31)
0·63
Donor CMV serology positive
1·11 (0·90–1·36)
0·32
HLA-identical sibling
1 (reference)
0·42
Unrelated (10/10)
1·02 (0·61–1·68)
0·95
Unrelated (9/10)
1·61 (0·85–3·04)
0·14
Secondary AML
1·09 (0·62–1·90)
0·77
Patient CMV serology positive
1·05 (0·71–1·55)
0·80
GVHD prophylaxis (ciclosporin + methotrexate)
0·91 (0·71–1·15)
In-vivo T-cell depletion
1·15 (0·86–1·53)
0·34
Leukaemia-free survival Cyclophosphamide + TBI vs busulfan + cyclophosphamide
0·97 (0·81–1·16)
0·71
Donor CMV serology positive
0·97 (0·68–1·40)
0·89 0·51
1·01 (0·94–1·09)
0·82
GVHD prophylaxis (ciclosporin + methotrexate)
1·15 (0·75–1·75)
Age at HCT (per 10 years) Year of HCT
0·99 (0·96–1·02)
0·43
In-vivo T-cell depletion
0·93 (0·55–1·56)
0·78
0·92 (0·68–1·24)
0·59
Chronic GVHD
Disease at HCT Primary refractory
1 (reference)
First relapse
1·20 (1·00–1·44)
0·053
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
Second relapse
1·54 (1·17–2·03)
0·0020
Age at HCT (per 10 years)
1·11 (0·98–1·27)
0·10
Year of HCT
0·99 (0·95–1·04)
0·80
Donor HLA-identical sibling
1 (reference)
Unrelated (10/10)
0·87 (0·67–1·13)
Disease at HCT 0·30
Primary refractory
1 (reference) 1·08 (0·80–1·46)
0·61
1·31 (0·81–1·12)
0·27
0·95 (0·67–1·35)
0·77
First relapse
Secondary AML
0·88 (0·63–1·22)
0·44
Second relapse
Patient CMV serology positive
1·03 (0·84–1·26)
0·79
Donor CMV serology positive
1·10 (0·91–1·34)
0·32
HLA-identical sibling
1 (reference)
0·46
Unrelated (10/10)
1·48 (0·97–2·24)
0·07
Unrelated (9/10)
1·41 (0·73–2·73)
0·31
Secondary AML
0·99 (0·60–1·64)
0·96
Patient CMV serology positive
0·90 (0·64–1·28)
0·56
Unrelated (9/10)
GVHD prophylaxis (ciclosporin + methotrexate)
0·92 (0·73–1·15)
In-vivo T-cell depletion
1·19 (0·91–1·57)
0·20
Relapse
Donor
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
1·08 (0·88–1·33)
0·46
Donor CMV serology positive
0·96 (0·69–1·35)
0·82 0·39
0·93 (0·85–1·02)
0·14
GVHD prophylaxis (ciclosporin + methotrexate)
1·18 (0·81–1·71)
Age at HCT (per 10 years) Year of HCT
1·00 (1·05–1·69)
0·86
In-vivo T-cell depletion
0·71 (0·44–1·15)
0·17
Complete remission within 100 days of HCT
Disease at HCT
1·39 (1·00–1·92)
0·048
Primary refractory
1 (reference)
First relapse
1·24 (1·00–1·54)
0·050
Cyclophosphamide + TBI vs busulfan + cyclophosphamide
Second relapse
1·73 (1·26–2·38)
0·0007
Age at HCT (per 10 years)
1·02 (0·89–1·18)
0·72
Year of HCT
1·02 (0·97–1·07)
0·49
Donor HLA-identical sibling
1 (reference)
Unrelated (10/10)
0·83 (0·61–1·12)
Unrelated (9/10)
0·77 (0·50–1·18)
Disease at HCT 0·22
Primary refractory
1 (reference)
0·23
First relapse
1·17 (0·84–1·63)
0·36
Second relapse
1·10 (0·64–1·87)
0·73
Secondary AML
0·79 (0·52–1·19)
0·26
Patient CMV serology positive
1·02 (0·80–1·30)
0·87
(Table 4 continues in next column)
(Table 4 continues in next column)
e389
www.thelancet.com/haematology Vol 2 September 2015
Articles
HR (95% CI)
A
p value
(Continued from previous column)
100
HLA-identical sibling
1 (reference)
Unrelated (10/10)
1·04 (0·65–1·67)
0·88
Unrelated (9/10)
0·86 (0·45–1·64)
0·64
Secondary AML
1·36 (0·77–2·42)
0·29
Patient CMV serology positive
0·98 (0·67–1·42)
0·90
Donor CMV serology positive
1·27 (0·89–1·82)
0·19
GVHD prophylaxis (ciclosporin + methotrexate)
0·96 (0·64–1·43)
0·83
In-vivo T-cell depletion
1·33 (0·82–2·15)
0·25
Cumulative incidence of chronic GVHD (%)
Donor
90 80 70 60 50 40 30 20 10
HR=hazard ratio. TBI=total body irradiation. HCT=allogeneic haemopoietic stem-cell transplantation. AML=acute myeloid leukaemia. CMV=cytomegalovirus. GVHD=graft-versus-host disease. Number at risk Busulfan + cyclophosphamide Cyclophosphamide + TBI
0
1
2
3
4
5
441 294
61 54
28 30
20 25
16 22
13 16
1
2
3
4
5
143 101
89 76
66 65
49 55
36 45
1
2
3
4
5
49 55
36 45
B 100 90 Cumulative incidence of relapse (%)
undertake HCT in patients with acute myeloid leukaemia who are not in complete remission, and most patients with refractory acute myeloid leukaemia do not achieve complete remission. Transplants should preferentially be done as soon as necessary, rather than be postponed until a patient achieves complete remission (or reserved for only those who have achieved complete remission). Fewer than 20% of patients with refractory acute myeloid leukaemia will eventually be able to receive HCT in complete remission, because the patient will need to survive the disease and then be fit enough to undergo HCT after having achieved complete remission.25–27 The conditioning regimen has a key role in HCT and has important effects on transplant outcome in patients with refractory acute myeloid leukaemia.2–4,25 Busulfan plus cyclophosphamide and cyclophosphamide plus TBI are the two most widely used conditioning regimens for advanced acute myeloid leukaemia allografts. From the available data, survival does not significantly differ with these two regimens.2,11,28 Either of the regimens can be used for acute myeloid leukaemia allografts, and the choice might ultimately depend on local availability and expertise. The most recent randomised trial comparing busulfan plus cyclophosphamide with cyclophosphamide plus TBI was published more than 15 years ago.29 A gradual decrease in ablative TBI-based transplants and an increase in intravenous busulfan-based transplants have occurred without supportive prospective randomised data comparing the approaches. Our study does not include a comparison with the recently increasingly used intravenous fludarabine plus busulfan regimen, reported to have reduced toxic effects,
0
80 70 60 50 40 30 20 10 0 0
Number at risk Busulfan + cyclophosphamide 514 Cyclophosphamide + TBI 338
C 100 Cumulative incidence of non-relapse mortality (%)
Table 4: Multivariable analysis
Busulfan + cyclophosphamide Cyclophosphamide + TBI
90 80 70 60 50 40 30 20 10 0 0
Figure 2: Secondary outcomes after HCT, according to conditioning regimen Cumulative incidence of chronic GVHD (A), relapse (B), and non-relapse mortality (C) after conditioning with intravenous busulfan plus cyclophosphamide versus cyclophosphamide plus TBI. HCT=allogeneic haemopoietic stem-cell transplantation. TBI=total body irradiation.
www.thelancet.com/haematology Vol 2 September 2015
Number at risk Busulfan + cyclophosphamide 514 Cyclophosphamide + TBI 338
Time from transplant (years) 143 101
89 76
66 65
e390
Articles
Total deaths
Cyclophosphamide Intravenous busulfan plus cyclophosphamide plus TBI (n=338) (n=514)
All patients (n=852)
565 (66%)
340 (66%)
225 (67%)
Cardiac toxic effects
2 (<1%)
1 (<1%)
3 (<1%)
Haemorrhage
4 (1%)
1 (<1%)
5 (1%)
1 (<1%)
0
Graft failure or rejection
1 (<1%)
Venocclusive disease of the liver
16 (3%)
5 (1%)
21 (2%)
Infection
36 (7%)
18 (5%)
54 (6%)
Interstitial pneumonitis
12 (2%)
6 (2%)
18 (2%)
Graft-versus-host disease
49 (10%)
25 (7%)
74 (9%)
205 (40%)
159 (47%)
364 (43%)
Original disease Second malignancy
1 (<1%)
Other transplant-related causes
14 (3%)
0 10 (3%)
1 (<1%) 24 (3%)
Data are number of patients with at least one event (%).
Table 5: Causes of death
because of the small numbers of patients with long-term follow-up. However, we believe that the intensity of a conditioning regimen might have a substantial effect on post-transplantation relapse or disease progression in patients with refractory or active leukaemia. The results of a randomised trial30 suggested that the fludarabine plus busulfan regimen is not a suitable replacement for the busulfan plus cyclophosphamide regimen in adults younger than 60 years who are eligible for ablative conditioning therapy for HCT. Decreasing the intensity of the conditioning regimen is likely to be associated with increased relapse risk or progression, especially in patients with active leukaemia before HCT.2–6,30 Also, in terms of disease control, we did not note a significant difference in long-term leukaemia-free survival between the two conditioning regimens, suggesting that the use of intravenous busulfan is probably associated with a similar drug-versus-leukaemia effect to TBI in patients with refractory acute myeloid leukaemia. These results are of major importance because they support the findings of previous studies,10 but in refractory patients rather than just in patients in remission. Leukaemia-free survival in patients in second complete remission was also similar between the intravenous busulfan plus cyclophosphamide and cyclophosphamide plus TBI groups,10 and now in refractory disease, suggesting a potent anti-leukaemic effect after intravenous busulfan. Thus, from a practical standpoint, intravenous busulfan is probably a valid and efficient alternative to high-dose TBI for patients with acute myeloid leukaemia in remission (first or second) or active disease, especially those treated in transplant centres without access to radiation facilities. Finally, when considering long-term side-effects after HCT, the use of ablative TBI is now well known to play a major role in the onset of diverse late complications after HCT.31–34 Several studies have shown an effect of TBI on cataract formation, azoospermia, diabetes, e391
hypertension, late cardiovascular events, and secondary malignancies.31,32,35–39 The limitations of this study are important to recognise. First, the study is retrospective. Second, patient characteristics vary among the groups for several factors including age, proportion of unrelated donors, and the year in which transplantation was done. Also, we do not have information about why patients were allocated to a specific regimen in the registry and distinguishing the role of conditioning from the role of a potential centre effect is difficult. The aim of our analysis was to compare the two conditioning regimens using EBMT registry data, which allows us to study a large population of patients with refractory acute myeloid leukaemia. Both the design of the study and inclusion criteria were intended to answer this clinical question and therefore are not adapted to develop a prognostic score based on information that is not routinely collected in the registry. However, only through the conduct of well designed clinical trials can we understand and appreciate the complexities of conditioning regimen selection and the associated outcomes after HCT for refractory acute myeloid leukaemia. Unfortunately, no ongoing trials are comparing outcomes after intravenous busulfan plus cyclophosphamide with that after cyclophosphamide plus TBI conditioning regimen for refractory acute myeloid leukaemia. Therefore, in the absence of any prospect of such comparative studies, our data suggest that both regimens are equally effective in patients with refractory acute myeloid leukaemia. Contributors AN, BNS, ML, and MM designed the study or analysed the data (or both). ML, EP, JP, JF, SK-K, LV, AA, MA, DWB, SV, NM, FS, and MM provided important clinical data. All authors contributed to the writing of the report and approved the final version of the Article. Declaration of interests JF has received speaker honoraria from Fresenius, Neovii, and Riemser. All other authors declare no competing interests. Acknowledgments We thank all European Group for Blood and Marrow Transplantation (EBMT) centres and national registries for contributing patients to the study, and data managers for their excellent contribution. References 1 Kanakry JA, Kasamon YL, Gocke CD, et al. Outcomes of related donor HLA-identical or HLA-haploidentical allogeneic blood or marrow transplantation for peripheral T cell lymphoma. Biol Blood Marrow Transplant 2013; 19: 602–06. 2 Feldman EJ, Gergis U. Management of refractory acute myeloid leukemia: re-induction therapy or straight to transplantation? Curr Hematol Malig Rep 2012; 7: 74–77. 3 Hamadani M, Mohty M, Kharfan-Dabaja MA. Reduced-intensity conditioning allogeneic hematopoietic cell transplantation in adults with acute myeloid leukemia. Cancer Control 2011; 18: 237–45. 4 Chemnitz JM, von Lilienfeld-Toal M, Holtick U, et al. Intermediate intensity conditioning regimen containing FLAMSA, treosulfan, cyclophosphamide, and ATG for allogeneic stem cell transplantation in elderly patients with relapsed or high-risk acute myeloid leukemia. Ann Hematol 2012; 91: 47–55. 5 Duval M, Klein JP, He W, et al. Hematopoietic stem-cell transplantation for acute leukemia in relapse or primary induction failure. J Clin Oncol 2010; 28: 3730–38.
www.thelancet.com/haematology Vol 2 September 2015
Articles
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Craddock C, Labopin M, Pillai S, et al. Factors predicting outcome after unrelated donor stem cell transplantation in primary refractory acute myeloid leukaemia. Leukemia 2011; 25: 808–13. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: a randomized trial of two irradiation regimens. Blood 1990; 76: 1867–71. Santos GW, Tutschka PJ, Brookmeyer R, et al. Marrow transplantation for acute nonlymphocytic leukemia after treatment with busulfan and cyclophosphamide. N Engl J Med 1983; 309: 1347–53. Thomas ED, Buckner CD, Banaji M, et al. One hundred patients with acute leukemia treated by chemotherapy, total body irradiation, and allogeneic marrow transplantation. Blood 1977; 49: 511–33. Nagler A, Rocha V, Labopin M, et al. Allogeneic hematopoietic stem-cell transplantation for acute myeloid leukemia in remission: comparison of intravenous busulfan plus cyclophosphamide versus total-body irradiation plus cyclophosphamide as conditioning regimen—a report from the Acute Leukemia Working Party of the European Group for Blood and Marrow Transplantation. J Clin Oncol 2013; 31: 3549–56. Uberti JP, Agovi MA, Tarima S, et al. Comparative analysis of BU and CY versus CY and TBI in full intensity unrelated marrow donor transplantation for AML, CML and myelodysplasia. Bone Marrow Transplant 2011; 46: 34–43. Copelan EA, Hamilton BK, Avalos B, et al. Better leukemia-free and overall survival in AML in first remission following cyclophosphamide in combination with busulfan compared with TBI. Blood 2013; 122: 3863–70. Bredeson C, LeRademacher J, Kato K, et al. Prospective cohort study comparing intravenous busulfan to total body irradiation in hematopoietic cell transplantation. Blood 2013; 122: 3871–78. Copelan EA, Biggs JC, Thompson JM, et al. Treatment for acute myelocytic leukemia with allogeneic bone marrow transplantation following preparation with BuCy2. Blood 1991; 78: 838–43. Blaise D, Maraninchi D, Archimbaud E, et al. 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: 2578–82. Ringden O, Remberger M, Ruutu T, et al, for the Nordic Bone Marrow Transplantation Group. Increased risk of chronic graftversus-host disease, obstructive bronchiolitis, and alopecia with busulfan versus total body irradiation: long-term results of a randomized trial in allogeneic marrow recipients with leukemia. Blood 1999; 93: 2196–201. Ringden O, Ruutu T, Remberger M, et al. 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: 2723–30. Clift RA, Buckner CD, Thomas ED, et al. Marrow transplantation for chronic myeloid leukemia: a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide. Blood 1994; 84: 2036–43. Michel G, Gluckman E, Esperou-Bourdeau H, et al. Allogeneic bone marrow transplantation for children with acute myeloblastic leukemia in first complete remission: impact of conditioning regimen without total-body irradiation—a report from the Societe Francaise de Greffe de Moelle. J Clin Oncol 1994; 12: 1217–22. Hartman AR, Williams SF, Dillon JJ. 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: 439–43. Devergie A, Blaise D, Attal M, et al. 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: 2263–68.
www.thelancet.com/haematology Vol 2 September 2015
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36 37
38
39
Rezvani AR, McCune JS, Storer BE, et al. Cyclophosphamide followed by intravenous targeted busulfan for allogeneic hematopoietic cell transplantation: pharmacokinetics and clinical outcomes. Biol Blood Marrow Transplant 2013; 19: 1033–39. Hassan M, Andersson BS. Role of pharmacogenetics in busulfan/ cyclophosphamide conditioning therapy prior to hematopoietic stem cell transplantation. Pharmacogenomics 2013; 14: 75–87. Latouche A, Allignol A, Beyersmann J, Labopin M, Fine JP. A competing risks analysis should report results on all cause-specific hazards and cumulative incidence functions. J Clin Epidemiol 2013; 66: 648–53. Chen GL, Liu H, Zhang Y, et al. Early versus late preemptive allogeneic hematopoietic cell transplantation for relapsed or refractory acute myeloid leukemia. Biol Blood Marrow Transplant 2014; 20: 1369–74. Jabbour E, Daver N, Champlin R, et al. Allogeneic stem cell transplantation as initial salvage for patients with acute myeloid leukemia refractory to high-dose cytarabine-based induction chemotherapy. Am J Hematol 2014; 89: 395–98. Ravandi F, Cortes J, Faderl S, et al. Characteristics and outcome of patients with acute myeloid leukemia refractory to 1 cycle of high-dose cytarabine-based induction chemotherapy. Blood 2010; 116: 5818–23. Gupta V, Lazarus HM, Keating A. Myeloablative conditioning regimens for AML allografts: 30 years later. Bone Marrow Transplant 2003; 32: 969–78. Socie G, Clift RA, Blaise D, et al. 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: 3569–74. Lee JH, Joo YD, Kim H, et al. Randomized trial of myeloablative conditioning regimens: busulfan plus cyclophosphamide versus busulfan plus fludarabine. J Clin Oncol 2013; 31: 701–09. Savani BN. Blood and marrow transplantation long term management: prevention and complications, 1st edn. Oxford, UK: John Wiley & Sons Ltd, 2014. Mohty M, Malard F, Savani BN. High-dose total body irradiation and myeloablative conditioning before allogeneic stem cell transplantation: time to rethink? Biol Blood Marrow Transplant 2014; 21: 620–24. Savani BN, Griffith ML, Jagasia S, Lee SJ. How I treat late effects in adults after allogeneic stem cell transplantation. Blood 2011; 117: 3002–09. Savani BN. How can we improve life expectancy and quality of life in long-term survivors after allogeneic stem cell transplantation? Semin Hematol 2012; 49: 1–3. Khera N, Storer B, Flowers ME, et al. Nonmalignant late effects and compromised functional status in survivors of hematopoietic cell transplantation. J Clin Oncol 2012; 30: 71–77. Rizzo JD, Curtis RE, Socie G, et al. Solid cancers after allogeneic hematopoietic cell transplantation. Blood 2009; 113: 1175–83. Wei C, Thyagiarajan M, Hunt L, et al. Reduced beta-cell reserve and pancreatic volume in survivors of childhood acute lymphoblastic leukaemia treated with bone marrow transplantation and total body irradiation. Clin Endocrinol (Oxf) 2014; 82: 59–67. Sanders JE, Hoffmeister PA, Woolfrey AE, et al. Thyroid function following hematopoietic cell transplantation in children: 30 years’ experience. Blood 2009; 113: 306–08. Hoffmeister PA, Hingorani SR, Storer BE, Baker KS, Sanders JE. Hypertension in long-term survivors of pediatric hematopoietic cell transplantation. Biol Blood Marrow Transplant 2010; 16: 515–24.
e392