the role of cytotoxic therapy with hematopoietic stem cell ...stem cell transplantation in the...

Biology of Blood and Marrow Transplantation 14:137-180 (2008)Q 2008 American Society for Blood and Marrow Transplantation1083-8791/08/1402-0001$32.00/0doi:10.1016/j.bbmt.2007.11.002

The Role of Cytotoxic Therapy with HematopoieticStem Cell Transplantation in the Therapy of AcuteMyelogenous Leukemia in Adults: An Evidence-BasedReview

Denise M. Oliansky,1 Frederick Appelbaum,2 Peter A. Cassileth,3 Armand Keating,4 Jamie Kerr,5

Yago Nieto,6 Susan Stewart,7 Richard M. Stone,8 Martin S. Tallman,9 Philip L. McCarthy, Jr.,1

Theresa Hahn1

1 Roswell Park Cancer Institute, Buffalo, New York; 2 Fred Hutchinson Cancer Institute, Seattle, Washington;3 University of Miami Sylvester Cancer Center, Miami, Florida; 4 University of Toronto, Toronto, ON, Canada;5 Excellus Blue Cross/Blue Shield, Rochester, New York; 6 M.D. Anderson Cancer Center, Houston, Texas; 7 BMTInfonet, Chicago, Illinois; 8 Dana Farber Cancer Institute, Boston, Massachusetts; and 9 Northwestern UniversityFeinberg School of Medicine, Robert H. Lurie Comprehensive Cancer Center, Chicago, Illinois

Correspondence and reprint requests: Theresa Hahn, PhD, Roswell Park Cancer Institute, Elm and Carlton Streets,Buffalo, NY 14263. (e-mail: [email protected]).

Major funding for this study was provided by the National Marrow Donor Program.

Received November 8, 2007; accepted November 9, 2007

ABSTRACTClinical research examining the role of hematopoietic stem cell transplantation (HSCT) in the therapy of acutemyelogenous leukemia (AML) in adults is presented and critically evaluated in this systematic evidence-basedreview. Specific criteria were used for searching the published literature and for grading the quality and strengthof the evidence and the strength of the treatment recommendations. Treatment recommendations based on theevidence are presented in Table 3, entitled Summary of Treatment Recommendations Made by the Expert Panel for

Adult Acute Myelogenous Leukemia, and were reached unanimously by a panel of AML experts. The identifiedpriority areas of needed future research in adult AML include: (1) What is the role of HSCT in treating patientswith specific molecular markers (eg, FLT3, NPM1, CEBPA, BAALC, MLL, NRAS, etc.) especially in patients withnormal cytogenetics? (2) What is the benefit of using HSCT to treat different cytogenetic subgroups? (3) What isthe impact on survival outcomes of reduced intensity or nonmyeloablative versus conventional conditioning inolder (.60 years) and intermediate (40-60 years) aged adults? (4) What is the impact on survival outcomes ofunrelated donor HSCT vesus chemotherapy in younger (\40 years) adults with high risk disease?� 2008 American Society for Blood and Marrow Transplantation

KEY WORDS

Acute myeloid leukemia � Hematopoietic stem cell transplantation � Therapy � Adult

INTRODUCTION

The American Society for Blood and MarrowTransplantation (ASBMT), in 1999, began an initiativeto sponsor evidence-based reviews of the scientific andmedical literature for the use of blood and marrowtransplantation in the therapy of selected diseases.The steering committee that was convened to overseethe projects invited an independent panel of disease-specific experts to conduct each review. Five previous

reviews have been published in Biology of Blood andMarrow Transplantation: Diffuse Large B Cell Non-Hodgkin Lymphoma [1], Multiple Myeloma (MM) [2],Pediatric Acute Lymphocytic Leukemia (ALL) [3],Adult ALL [4], and Pediatric Acute Myeloid Leukemia(AML) [5].

The following is the sixth review to result from thisinitiative. Its goals are to assemble and critically evaluateall of the evidence regarding the role of hematopoietic

137

mailto:[email protected]

138 D. M. Oliansky et al.

Table 1. Grading the Quality of Design and Strength of Evidence

Levels of evidence

111 High-quality meta-analyses, systematic reviews of randomized controlled trials (RCTs), or RCTs with a very low risk of bias

11 Well-conducted meta-analyses, systematic reviews of RCTs, or RCTs with a low risk of bias

12 Meta-analyses, systematic reviews of RCTs, or RCTs with a high risk of bias

211 High-quality systematic reviews of case-controlled or cohort studies High-quality case-controlled or cohort studies with

a very low risk of confounding, bias, or chance and a high probability that the relationship is causal

21 Well-conducted case-controlled or cohort studies with a low risk of confounding, bias, or chance and a moderate

probability that the relationship is causal

22 Case-controlled or cohort studies with a high risk of confounding, bias, or chance and a significant risk that the relationship

is not causal

3 Nonanalytic studies, for example, case reports, case series

4 Expert opinion

Reprinted with permission from Harbour R, Miller J. A new system for grading recommendations in evidence-based guidelines. Br Med J.2001;323:334-336.

stem cell transplantation (HSCT) in the therapy ofadult ($15 years) patients with AML, make treatmentrecommendations based on the available evidence, andidentify areas of needed research.

LITERATURE SEARCH METHODOLOGY

PubMed and Medline, the Web sites developed bythe National Center of Biotechnology Information atthe National Library of Medicine of the National In-stitutes of Health, were searched August 8, 2006, usingthe search terms ‘‘acute myeloid leukemia’’ or ‘‘acutemyelogenous leukemia’’ and ‘‘transplant,’’ limited tohuman trials, English language, and a publicationdate of 1990 or later. An updated search was conductedin early May, 2007, limited to August 9, 2006, to April30, 2007. Manuscripts were excluded if they were pub-lished before 1990, included \50 AML patients, werenot peer reviewed, were editorials, letters to the editor,Phase I (dose-escalation or dose-finding) studies, re-views, consensus conference papers, practice guide-lines, laboratory studies with no clinical correlates,did not focus on an aspect of therapy with HSCT forthe treatment of adult AML, or if .50% of the studypopulation was \15 years and the results were notstratified by age. In addition, for a manuscript to be in-cluded, a minimum of 70% of study subjects had to beAML patients, or study results had to be stratified by

disease. Abstracts and presentations at national or in-ternational meetings were not included as evidencein this review for reasons previously described [3].

QUALITATIVE AND QUANTITATIVE GRADINGOF THE EVIDENCE

The hierarchy of evidence, including a grading sys-tem for the quality and strength of the evidence, andstrength of each treatment recommendation, was es-tablished and published as an editorial policy state-ment in Biology of Blood and Marrow Transplantation in2005 [6]. Tables 1 and 2 are reprinted from the policystatement and define criteria used to grade the studiesincluded in this review and to grade the treatment rec-ommendations. Study design, including sample size,patient selection criteria, duration of follow-up, andtreatment plan also were considered in evaluating thestudies. Several multicenter clinical trials were de-signed to biologically assign patients to a treatmentarm based on the availability of a donor (‘‘biologicallocation’’). These studies of allogeneic (allo) HSCTversus chemotherapy are therefore graded as level‘‘2’’ evidence, not level ‘‘1,’’ because they are not statis-tically randomized controlled trial designs. Autolo-gous (auto) HSCT versus chemotherapy studies weregraded as level ‘‘1’’ evidence if the study design in-cluded a statistically randomized controlled trial.

Table 2. Grading the Strength of the Treatment Recommendation

Grades of recommendation

A At least one meta-analysis, systematic review, or randomized controlled trial (RCT) rated as 111, and directly applicable

to the target population; or a systematic review of RCTs or a body of evidence consisting principally of studies rated as

11, directly applicable to the target population, and demonstrating overall consistency of results

B A body of evidence including studies rated as 211, directly applicable to the target population, and demonstrating overall

consistency of results; or extrapolated evidence from studies rated as 111 or 11

C A body of evidence including studies rated as 21, directly applicable to the target population and demonstrating overall

consistency of results; or extrapolated evidence from studies rated as 211

D Evidence level 3 or 4; or extrapolated evidence from studies rated as 21

Reprinted with permission from Harbour R, Miller J. A new system for grading recommendations in evidence-based guidelines. Br Med J.2001;323:334-336.

The Role of Cytotoxic Therapy with HSCT 139

Table 3. Summary of Treatment Recommendations Made by the Expert Panel for Adult Acute Myelogenous Leukemia

Indication for

HSCT

Treatment

Recommendation

Grade*

Highest

Level of

Evidence†

Reference

No.h Treatment Recommendation Comments

TRANSPLANTATION VERSUS CHEMOTHERAPY

Autologous (Auto) SCT versus Chemotherapy (Chemo) in CR1

Induction 1 intensive

consolidation chemo

(ICC), then ICC

versus auto

A 11 [9, 10] Based on the survival data presented, there is

no significant advantage of auto-SCT

over chemotherapy. Most of the data reflect

outmoded treatment strategies (i.e., BM versus

PBSC; maintenance therapy pre-SCT, etc.);

studies using modern technologies may effect

outcomes to an unknown extent.

Induction 1 ICC, then

no further Tx versus

auto

B 12 [13,14] Same as above

Induction, then ICC

versus auto

B 11 [16] Same as above

Allogeneic (Allo) SCT versus Chemo in CR1

Myeloablative Allo-

SCT versus chemo

A 11 [18] There is a survival advantage for allo-SCT versus

chemotherapy for patients \55 years with high

risk cytogenetics. There is insufficient evidence

to routinely recommend allo-SCT for patients

with intermediate risk cytogenetics, although

this is a reasonable strategy. There is no survival

advantage for allo-SCT in patients \55 years

with low risk cytogenetics. There are insufficient

data to make a recommendation on the use of

myeloablative regimens for patients .55 years.

Reduced-intensity

conditioning (RIC)

Allo versus

chemo

No

recommendation

22 [26] There are insufficient data to make a

recommendation.

SCT versus Chemo in CR2

Allo-SCT versus

chemo in CR2

D 4 [27] Based on expert opinion and clinical practice, it

is recommended that patients in CR2 receive an

allo-SCT if there is an available donor;

otherwise, an auto-SCT is

recommended.

TRANSPLANTATION TECHNIQUES

Allogeneic versus

Autologous SCT

B 211 [28,29] Based on data and expert opinion, an HLA-

matched related donor (MRD) allo-SCT is

recommended over auto-SCT, if an MRD is

available. There are insufficient data to make

a recommendation for matched unrelated

donor (MUD) allo-SCT versus auto-SCT.

These studies do not reflect modern techniques

in supportive care, stem cell source, or the use

of molecular HLA typing (for the MUD studies).

Autologous SCT

PBSCT versus BMT C 21 [37-39] Based on data and clinical practice, PBSCT is

recommended due to improvement in safety

and early mortality. However, long-term

outcomes have not been studied prospectively;

therefore, the impact of PBSCTon OS is

unknown.

Unpurged versus

purgedY

C 21 [40-42] There is no evidence of a survival advantage

with purged BMT. There are insufficient data

to make a recommendation for purging of

PBSCT.

Tandem versus

single

No

recommendation

2- [43] There are insufficient data to make a

recommendation.

Allogeneic SCT

Related versus

unrelated

C 21 [55,56] MRD allo-SCT is recommended if available. If

a MRD is not available, a MUD allo-SCT may

provide equivalent outcomes.


Table 3. (Continued)

Indication for

HSCT

Treatment

Recommendation

Grade*

Highest

Level of

Evidence†

Reference

No.h Treatment Recommendation Comments

T cell replete

versus

T cell depleted

B 11 [57] There is no evidence of a survival advantage

with T cell-depleted grafts.

PBSCT versus BMT C 211 [59] For high-risk disease, allo-PBSCT is recommended

over BMT. For low-risk disease, allo-PBSCT and

BMT have equivalent outcomes. There are

insufficient data to make a recommendation on

stem cell source for MUD allo-SCT.

Therapy Regimens

Auto-SCT—

Comparison of 2 or

more high-dose

therapy regimens

C 21 [77 - 78] There is no evidence of a survival advantage

with any one high-dose therapy regimen.

Allo-SCT—

Comparison of 2 or

more myeloablative

conditioning regimens

B 11 [79 - 80] There is no significant survival advantage with

any one myeloablative conditioning regimen.

Studies of late effects may change this

recommendation. If a TBI-containing

conditioning regimen is used, fractionated

rather than single-dose TBI is recommended.

Allo-SCT—

Use of RIC

regimens

No

recommendation

211 [86] There are insufficient data to make a

recommendation. The use of RIC is dependent

on patient characteristics such as age,

comorbidities, and cytogenetic risk.

HSCT indicates hematopoietic stem cell transplantation; CR, complete remission; ICC, intensive consolidation chemotherapy; PBSC, periph-eral blood stem cell; PBSCT, peripheral blood stem cell transplantation; BM, bone marrow; BMT, bone marrow transplantation; RIC, re-duced-intensity conditioning; HLA, human leukocyte antigen; MRD, matched-related donor; MUD, matched-unrelated donor; TBI, totalbody irradiation; OS, overall survival.

*Definitions: Grade of Recommendation (Table 2): (A) At least one meta-analysis, systematic review, or randomized controlled trial (RCT)rated as 111, and directly applicable to the target population; or a systematic review of RCTs or a body of evidence consisting principally ofstudies rated as 11, directly applicable to the target population, and demonstrating overall consistency of results. (B) A body of evidence includingstudies rated as 211, directly applicable to the target population, and demonstrating overall consistency of results; or extrapolated evidence fromstudies rated as 111 or 11. (C) A body of evidence including studies rated as 21, directly applicable to the target population and demonstratingoverall consistency of results; or extrapolated evidence from studies rated as 211; (D) Evidence level 3 or 4; or extrapolated evidence from studiesrated as 21†Definitions: Levels of Evidence (Table 1): 111 High-quality meta-analyses, systematic reviews of randomized controlled trials (RCTs), orRCTs with a very low risk of bias. 11 Well-conducted meta-analyses, systematic reviews of RCTs, or RCTs with a low risk of bias. 12 Meta-analyses, systematic reviews of RCTs, or RCTs with a high risk of bias. 211 High-quality systematic reviews of case-control or cohort studies; orHigh-quality case-control or cohort studies with a very low risk of confounding, bias, or chance and a high probability that the relationship iscausal. 21 Well-conducted case control or cohort studies with a low risk of confounding, bias, or chance and a moderate probability that therelationship is causal; 22 Case-control or cohort studies with a high risk of confounding, bias, or chance and a significant risk that the relationshipis not causal. 3 Nonanalytic studies, for example, case reports, case series. 4 Expert opinion.hThe references listed represent the highest level of evidence used to make the treatment recommendation and are not inclusive of all evidencedescribed in the review.YPurging: Techniques to remove tumor, either by negative or positive selection.

Clinical studies are described with enough detail togive a concise summary of study design, sample size,and eligibility criteria.

All data in the text and tables were abstracted fromthe original manuscripts by the first author (D.O.),then double checked for accuracy and clarity by 2 otherauthors (T.H. and P.L.M.) and 1 additional reviewer(see Acknowledgments). In some articles there werediscrepancies within the data reported and, in thesecases, the data most consistent with the text of thearticle were presented in this review. The last author(T.H.) takes responsibility if errors remain. AppendixA lists the common abbreviations used in this review.

TREATMENT RECOMMENDATIONS

The strength of this review is the detail conveyedin the text about the study designs and the presentationof the outcomes in the summary tables included ineach major section. Table 3 contains the summary oftreatment recommendations made by the adult AMLexpert panel.

FORMAT OF THE REVIEW

Evidence is taken from self-described studies ofadult populations that included patients $15 years ofage. In each section of this review, the highest quality


studies are presented first; studies of equal quality arepresented in descending order by sample size. Thedesign of each study is described in the text and, unlessotherwise noted, the accompanying table in eachsection presents the outcomes for each study.

HSCT VERSUS CHEMOTHERAPY IN ADULT AML

Table 4 presents a summary of the outcomes foreach of the studies in this section. Of the 10 studies in-vestigating auto-HSCT versus chemotherapy, 5 founda significant difference in disease-free survival (DFS),relapse-free survival (RFS), or leukemia-free survival(LFS), and 2 studies found a significant difference inoverall survival (OS). Of the 14 studies that examinedallo-HSCT versus chemotherapy 6 auto-HSCT, 7reported a statistically significant difference in DFS,LFS, or event-free survival (EFS), and 4 studies founda significant difference in OS. Table 5 presents a sum-mary of the treatment schema used in the auto-HSCTversus chemotherapy studies.

Autologous SCT versus Chemotherapy in FirstComplete Remission (CR1)

Levi et al. [7] and Nathan et al. [8] each presentedthe results of meta-analyses of the same 6 randomizedcontrolled studies comparing auto bone marrow trans-plantation (BMT) to chemotherapy or no further treat-ment. The methods and results of these 2 studies werecomparable; hence, only 1 study [7] is summarizedhere. Study requirements for potential inclusion inthe Levi et al. meta-analysis were: randomization toauto-BMT in 1 arm and chemotherapy or no furthertreatment in the second arm, study sample aged 15-56 years, only AML patients, and data analysis basedon intention-to-treat (ITT). Of 4410 recruited patientsacross the 6 studies, 3220 (73%) achieved CR1, and1044 (23% of recruited patients) were randomized toreceive an auto-BMT or chemotherapy only (or nofurther therapy). Of these, 835 (80%) received theintended treatment, 360 in the auto-BMT arm and475 in the control arm. The reasons for nontransplantincluded recurrence of disease, toxicity of prior treat-ment, and patient refusal to have a transplant. Risk ofdying (death rate) and risk of relapse or death (eventrate) were determined for the auto-BMT and otherarm. Relative risks (RR) and 95% confidence intervals(CI) of death and events rates were calculated for eachstudy, from which a combined estimate of each rate wascalculated across the 6 studies. Cytogenetic data fromthe studies were not considered in the meta-analysis.Figure 1 compares the RR of relapse or death betweenthe auto-BMT and control groups.

Treatment schema I: induction 1 intensive consolidationchemo (ICC), then auto-SCT versus additional ICC. Zittounet al. [9] presented the results of the European Organi-zation for Research and Treatment of Cancer

(EORTC) and Gruppo Italiano Malattie EmatologicheMaligne dell’Adulto (GIMEMA) Leukemia Coopera-tive Groups prospective study of 990 adult (11-59 years;.90% were $15 years) patients with previously un-treated AML, comparing the outcomes of postremis-sion allo-BMT, auto-BMT, and chemotherapy. Somecenters from the GIMEMA group excluded patientswith acute promyelocytic (M3) leukemia (APL). Ofthe 990 enrolled patients, 36 (4%) were ineligible (inad-equate diagnosis [n 5 19]; other exclusion criteria [n 5

17]) and 13 (1%) had missing data, leaving a total of 941(95%) evaluable patients. A total of 623 (66%) patientsachieved CR1. A human leukocyte antigen (HLA)-matched sibling donor was available for 230 patients,and 168 (27% of CR1 patients) were assigned to allo-BMT. Of the remaining patients, 254 (41% of CR1 pa-tients) were randomized to either auto-BMT (n 5 128)or a second course of chemotherapy (n 5 126). Therewere no significant differences among the 3 treatmentgroups by age, white blood cell (WBC) count at diagno-sis, French-American-British (FAB) AML classifica-tion, number of induction cycles needed to reach

Figure 1. Relative risk (RR) of relapse or death rate comparing autol-

ogous BMT with the control arm. (Reprinted with permission; [7].)


CR1, or cytogenetic group. The median times fromCR1 to the initiation of postremission therapy were10, 14, and 15 weeks for the chemotherapy, auto-BMT, and allo-BMT groups, respectively (P \ .001).Actual receipt of intended therapy was as follows: allo-BMT, n 5 144 (86%), auto-BMT, n 5 95 (74%), andchemotherapy, n 5 104 (83%). Outcome analyseswere on an ITT basis.

Reiffers et al. [10] presented the outcomes of 204adults (15-55 years) with de novo AML enrolled in theprospective, multicenter, French BGMT 87 trialcomparing allo-BMT, auto-SCT, and ICC. Of 204patients, 162 (79.4%) achieved CR1. Between CR1and consolidation, 26 patients were excluded for the fol-lowing reasons: fungal infection (n 5 8), early leukemicrelapse (n 5 5), cardiac contraindications for anthracy-cline administration or intensive treatment (n 5 4), en-cephalitis (n 5 3), persistent cytopenia (n 5 2), refusal offurther treatment (n 5 2), and hepatitis (n 5 2). One pa-tient who underwent syngeneic BMT was also excludedfrom the analyses. Of the remaining 135 patients, 36were #45 years of age with an HLA-identical siblingdonor and were assigned to allo-BMT. Of these, 33(92%) actually received the assigned treatment. The99 patients who did not meet the inclusion criteria forallo-BMT (39 were .45 years; 60 were #45 years,but with no HLA-identical sibling donor) received in-tensive chemotherapy, then, while still in CR (n 5 77),were randomly assigned to either auto-SCT (n 5 39) ormaintenance chemotherapy (n 5 38). ITT analysescompared the outcomes between the donor (n 5 39)and no donor (n 5 60) groups and the randomizedauto-SCT and chemotherapy groups, in addition toa non-ITT analysis by actual treatment (allo-BMT,n 5 33; auto-SCT, n 5 33; and chemotherapy, n 5 38).Median time from CR1 to allo-BMT was 56 days(range: 42-142 days). The patients in the donor/nodonor groups were similar with regard to sex, WBCcount at diagnosis, and FAB classification. The rate ofhigh-risk cytogenetic abnormalities was higher in thedonor group than the no donor group, but the differ-ence was not statistically significant (P 5 .17). Of the39 patients assigned to auto-SCT, 33 underwent eithertransplant with unpurged bone marrow (n 5 16) orperipheral blood stem cells (PBSC). There were nosignificant differences in patient characteristics be-tween the auto-SCT and maintenance chemotherapygroups.

Harousseau et al. [11] reported the results of 535adult (15-50 years) patients enrolled in the GroupeOuest Est Leuc�emies Aigues My�eloblastiques (GOE-LAM) study comparing the outcomes of allo-BMT,auto-BMT, and ICC. Of the 535 patients, 18 wereconsidered ineligible because of inadequate diagnosis(n 5 9), age (n 5 7), or other reasons (n 5 2). Inaddition, 13 patients were not evaluable because ofdeath (n 5 4), major protocol violation (n 5 5), or

wrong randomization (n 5 4). Of the 504 evaluable pa-tients, 367 (73%) achieved CR1. Patients #40 years ofage with an HLA-identical sibling were assigned toallo-BMT (n 5 88). All other patients received a firstcourse of ICC. The allo-BMT patients were comparedto a subset of 134 patients #40 years of age, without anHLA-identical sibling, who had received a first courseof ICC. There were no significant differences inpatient characteristics between these 2 groups exceptin cytogenetic risk group: intermediate and unfavor-able cytogenetics were less common in the allo-BMTgroup. The median interval between CR1 and allo-BMT was 68 days. All patients who received a firstcourse of ICC were randomly assigned to either a sec-ond course of ICC (n 5 78) or an unpurged auto-BMT(n 5 86). Analyses were on an ITT basis. Of 367patients who achieved CR1, 219 (59.5%) receivedthe planned postremission treatment (73 allo-BMT,75 auto-BMT, and 71 ICC). Multivariate analysesconsidered the impact of FAB type, WBC count atdiagnosis, and cytogenetic risk group on outcomes.

Miggiano et al. [12] reported the results of 89adults (15-59 years) with AML in late CR1 treated ata single center in Italy. Three patients had antecedentmyelodysplastic syndrome (MDS) and 1 patient hadtherapy-related AML. Of the 89 patients, 51 (57%)received an auto-BMT. Median time from CR1 toauto-BMT was 8 months. The remaining 38 patients,who served as a nonrandomized control group, werenot transplanted because of poor prognostic factors(n 5 23), severe toxicity after induction therapy (n 5 8),patient refusal (n 5 4), or logistic difficulties (n 5 3).The 2 groups differed significantly in age and FABsubtype. There were no significant differences bet-ween the therapy groups by sex, type of AML (denovo versus secondary), or WBC count at diagnosis.

Treatment schema II: induction 1 ICC, then auto-SCT ver-sus no further treatment. Burnett et al. [13] presented theresults of 381 adult (74% were 15-55 years) patients en-rolled in the UK Medical Research Council (MRC)AML-10 trial who were randomized to receive eitherauto-BMT (n 5 190) or no further treatment (n 5 191)after ICC. The trial was open to patients with de novo orsecondary AML or refractory anemia. A total of 1966patients were enrolled in the trial, of which 1509(77%) achieved CR1. Of these, 508 were not eligiblefor randomization: 378 had a matched sibling donor,60 relapsed, and 70 died in remission. Of the 1001patients eligible for randomization, 620 (62%) werenot randomized because of patient or physician prefer-ence for an auto-BMT (n 5 79) or no further treatment(n 5 481), or for unknown reasons (n 5 60). The auto-BMT and no further treatment groups were similar inpatient characteristics, including age, sex, WBC count,FAB type, number of cycles needed to achieve CR1,type of AML, and cytogenetic risk group. Of the 190patients randomized to auto-BMT, 126 (66%) actually


underwent the procedure, and 5 (3%) of the 191patients randomized to no further treatment actuallyreceived an auto-BMT. Reasons for noncompliancewith the allocated treatment included death (n 5 7)or relapse (n5 11), patient refusal (n 5 20), clinicaldecision (n 5 16), and other (n 5 7) or unknownreasons (n 5 3). Outcome comparisons were by ITT,with adjustment for cytogenetic risk group.

Breems et al. [14] presented the results of 646 adult(\60 years) patients with AML enrolled in the Dutch-Belgian Haemato-Oncology Cooperative Group(HOVON) and Swiss Group for Clinical Cancer Re-search (SAKK) trial comparing auto-BMT versus nofurther treatment. Of the 646 enrolled, 425 (66%) pa-tients achieved CR1. Eighty-one patients underwentan allo-SCT and were not included in the analysis.An additional 214 patients were deemed ineligible forrandomization because of patient refusal (n 5 81), earlyrelapse (n 5 54), excessive toxicity (n 5 27), or notmeeting inclusion criteria (n 5 26), leaving 130 patientsto be randomized between auto-BMT (n 5 66) and nofurther treatment (n 5 64). Of these, 36 (55%) and 38(59%) received the assigned treatment, respectively.Analyses were on an ITT basis. The median intervalbetween randomization and auto-BMT was 64 days.There were no significant differences in patient charac-teristics between the 2 groups.

Rohatiner et al. [15] presented the results of a 2sequential cohorts study comparing the outcomes of144 adult (15-49 years) patients with AML enrolled inthe BXIII protocol (Italy, England) assigned to receivean unpurged auto-BMT versus a comparison group ofpatients who received traditional chemotherapy. Pa-tients with APL were excluded. Eleven patients withpreceding MDS, which progressed to AML, wereincluded in the auto-BMT group, as were 43 other pa-tients with varying degrees of MDS concurrent withthe diagnosis of AML. Of the 144 patients in theauto-BMT group, 106 achieved CR1 and, of these,61 proceeded to the assigned transplantation. Forty-five patients did not receive the auto-BMT for the fol-lowing reasons: early recurrence (n 5 17), medicallyunfit (n 5 12), patient refusal (n 5 8), elective allo-BMT (n 5 7), and insufficient cells (n 5 1). Patientsin the comparison group were from the same 3 insti-tutions as those in the auto-BMT group and receivedthe identical remission induction and consolidationtherapy, but without the high-dose therapy andauto-BMT. Analyses were on an ITT basis. Withthe exception of the presence of MDS in the auto-BMT group, patient characteristics were similar inthe 2 sequential cohorts.

Treatment schema III: induction, then ICC versus auto-SCT.Cassileth et al. [16] reported the results of an inter-group (Eastern Cooperative Oncology Group [ECOG],Southwest Oncology Group [SWOG], Cancer andLeukemia Group B [CALGB]) prospective study com-

paring the outcomes of chemotherapy, auto-BMT, andallo-BMT in 808 adult (16-55 years) patients with un-treated AML. Of the 808 patients enrolled in the study,36 were ineligible and 32 could not be evaluated (CRnot documented [n 5 19]; missing follow-up data [n 5

7]; withdrew before completing therapy [n 5 5]; CNSleukemia detected [n 5 1]), leaving 740 (92%) patientseligible for induction therapy. Of these, 518 (70%)achieved CR1; however, 172 patients were removedfrom the study prior to randomization or assignmentto postremission therapy, leaving 346 patients in theITT analysis. The primary reasons for removal in-cluded patient refusal to continue, persistent medicalproblems after induction, and relapse before randomi-zation. Patients with HLA-matched or single-antigen-mismatched family donors (n 5 113) were biologicallyassigned to receive allo-BMT. The remaining 233 pa-tients were randomized between auto-BMT (n 5 116)and high-dose cytarabine-based chemotherapy (n 5

117). Randomization to the 2 arms was stratifiedaccording to age (#45 versus .45 years), FAB type(M1, M2, M3, M4 versus M0, M5, M6, M7), thenumber of induction courses to achieve CR (1 versus2), and cytogenetics (favorable 5 t[8;21], t[15;17], orinv[16]; normal or having a single abnormality;unfavorable55q2, 25, 7q2, abnormal chromosome 9or 11, or 3 or more clonal abnormalities). There wereno statistically significant differences among the treat-ment groups on any of these patient characteristics.Median times from CR1 to postremission therapywere 14.1, 14.6, and 12.4 weeks for allo-BMT,auto-BMT, and chemotherapy, respectively. Actual re-ceipt of intended therapy was as follows: allo-BMT,n 5 90 (80%); auto-BMT, n 5 63 (54%); chemother-apy, n 5 106 (91%).

Bassan et al. [17] presented the results of 153 adult(15-60 years) patients with AML enrolled in the BXIIIProtocol, a postremission, nonrandomized, collabora-tive study between the Bergamo/Vicenza Hospitals inItaly and St. Bartholomew Hospital in England.Patients with APL were excluded. There were 147patients with de novo AML and 6 with secondaryAML. Of the 153 patients, 108 (70%) achieved CR1;however, 1 patient refused further treatment, leaving107 in CR1. Seventy-four patients \50 years startedfirst consolidation, of whom 41 had an auto-BMT(55%). Seven patients with HLA-matched siblingdonors underwent allo-BMT and were excluded fromthe analysis. Additional reasons patients did not receivean auto-BMT included relapse (n 5 8), infection(n 5 15), thrombocytopenia (n 5 1), patient refusal(n 5 1), or inadequate bone marrow harvest (n 5 1).Of the 33 patients .50 years, 2 died of infection, 2relapsed, and 10 had complications or very poor perfor-mance status, leaving 19 patients (plus 5 patientsfrom the\50 age group) to receive the planned chemo-therapy. Both groups were similar with regard to sex,


FAB, blast cell count, cytogenetics, and incidence ofhepatosplenomegaly.

Allogeneic SCT versus Chemotherapy ±Autologous SCT in CR1

Four of the studies already described in the previ-ous section included allo-BMT versus chemotherapy(6 auto-BMT) comparisons [9-11,16]. Because allo-BMT involves biologic assignment, as opposed to ran-domization, comparisons of allo-BMT versus auto/chemo are graded as cohort studies and rated as level‘‘2’’ evidence. Therefore, outcome results that relateto allo-BMT from these studies are presented in Table4 under the section ‘‘Allo-BMT versus Chemotherapy6 Auto-BMT in CR1.’’ Following are the descriptionsof several additional cohort studies of allo-BMT versuschemotherapy (6 auto-BMT) in CR1, the outcomes ofwhich are presented in the same section of Table 4.

Myeloablative conditioning for allo-SCT. Yanada et al.[18] presented the results of a meta-analyses of 5 stud-ies comparing the efficacy of allo-SCT versus chemo-therapy and/or auto-SCT in adult (10-55 years)patients with AML in CR1. Study requirements forpotential inclusion in the meta-analysis included: pub-lished in English between 1995 and 2003, only patientswith AML, offered allo-HSCT to all patients in CR1with an HLA-matched sibling and chemotherapy orauto-SCT to all others, used ITT analyses to comparepatients on the basis of donor availability, and assessedoutcomes in terms of OS. Studies were excluded if theydealt exclusively with children, analyzed outcomes byactual treatment given, did not offer allo-SCT to allpatients with a donor, if other articles from the sametrials were eligible, or the hazard ratio (HR) and95% CI for OS could not be assessed. A total of3100 patients received allo-SCT (n 5 1151) or alterna-tive treatments (n 5 1949) in the 5 studies. HRs wereused to assess the survival advantage of allo-SCT ver-sus alternative treatments. HRs and 95% CIs were ab-stracted from each study, and a general variance-basedmethod was used to estimate the summary HR and the95% CI for OS. Outcomes were analyzed by cytoge-netic risk. Figure 2 presents a Forrest plot of theHRs and 95% CIs for OS from the 5 studies includedin the meta-analysis.

Cornelissen et al. [19] reported the results ofa retrospective analysis of the Dutch-BelgiumHemato-Oncology Cooperative Group and the SwissGroup for Clinical Cancer Research (HOVON-SAKK) collaborative study of 925 adult (15-55 years)patients with AML in 3 consecutive studies (AML4,AML29, and AML42) comparing donor (n 5 326) ver-sus no donor outcomes (n 5 599). Patients with APLwere excluded. Patients were assigned to the no donorgroup as a result of absence of siblings, HLA-incom-patibility, or ineligibility of a potential donor. The 2groups were comparable with respect to age, FAB

type, WBC count at diagnosis, number of cycles toCR1, and cytogenetics. Of the 326 patients with anHLA-matched sibling donor, 268 (82%) underwentan allo-HSCT, 17% received a third cycle of chemo-therapy, and 1% an autograft. In the no donor group,65% underwent chemotherapy consolidation, 28% re-ceived an auto-SCT, and 8% received a mismatchedrelated or unrelated allo-SCT. Analyses were on anITT basis. Figure 3 compares the actuarial rates ofDFS between the donor and no donor groups.

Burnett et al. [20] presented a donor versus nodonor analysis of 1063 HLA-typed adult (\55 years)

Figure 2. Forrest plot of the HRs and 95% CIs for OS. Study iden-

tifications are provided below. FEM and REM denote summary HRs

by the fixed-effect and random-effect models. The varying sizes of

the filled diamonds represent the weight for the fixed-effect model

in the meta-analysis. A hazard ratio greater than unity means that al-

logeneic transplantation is superior to nonallogeneic transplanta-

tion. Study IDs: 1-1: Reiffers et al. [10]; 2-1: Keating et al., Br J

Haematol. 1998;102:1344-1353; 3-1 to 3-6: Slovak et al. [116]; 4-1

to 4-5: Burnett et al. [20]; 5-1 to 5-4: Suciu et al. [30]. (Reprinted

with permission; [18].)

Figure 3. Actuarial rates of DFS of patients with AML in first

complete remission according to donor availability. (Reprinted

with permission; [19].)

tion

(Interval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif) Outcome Defined

(Interval)

% OS

(Signif)

(Overall EFS)

RR§ 5 0.85

(CI 0.75-.97)

Auto versus Chemo

(P signif, but not

stated)

Not stated (Not stated)

RR 5.94

(CI .81-1.09)

(P not signif)

(4-yr DFS)

48 ± 5

30 ± 4

(P 5 .05)

Time from CR1

to first relapse or

death in CR

(4-yr OS)

56 ± 5

46 ± 5

(P 5 .43)

(3-yr DFS)

51 ± 17

58 ± 16

(P not signif)

Time from CR1 to

relapse, death, or

last visit

(3-yr OS)

56 ± 16

55 ± 16

(P not signif)

(4-yr DFS)

44 ± 5.5

40 ± 5.5

(P 5 .41)

Time from CR1 to

relapse or death

(4-yr OS)

50 ± 6

54 ± 6

(P 5 .72)

(5-yr LFS)

69.6

41.9

(P 5 .008)

Not stated Not stated

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T145

Table 4. Comparison of Patient Characteristics and Outcomes from Studies Included in the Transplantation versus Chemotherapy Sec

Reference [No.]

Quality

and

Strength

of

Evidence* Patient Population†

Number of Patients by

Study Group

Upper

Limit

(median)

Age at

Diagnosis

(Interval)

% Tx-

Related

Mortality

Median

Follow-up

AUTOLOGOUS SCT VERSUS CHEMOTHERAPY IN CR1‡

Levi et al., 2004

[7]

111 Meta-analysis

(6 RCT studies)

1984-1996

Recruited n 5 4410

Achieved CR1 n 5 3220

Randomized n 5 1044

(ITT)

Auto 524

Chemo/no further TX

520

\60 yrs

(Not

stated)

Not

stated

Not stated

Treatment Schema I: Induction 1 ICC, then Auto-SCT versus Additional ICC

Zittoun et al., 1995

[9]

(included in Levi et

al., meta-analysis)

11 EORTC and GIMEMA

1986-1993

Multicenter (59)

Recruited n 5 990

Evaluable n 5 941


Randomized n 5 254

(ITT)

Auto 128

Chemo 126

#59 yrs

(33 yrs)

Not

stated

3.3 yrs

Reiffers et al., 1996

[10]


al. and Yanada et al.,

meta-analyses)

11 BGMT 87 trial

1987-1990

Multicenter (4)

Enrolled n 5 204


Randomized n 5 77

(ITT)

Auto 39

Chemo 38

#55 yrs

(39.8 yrs)

(Not

stated)

6

0

Minimum

Follow-up

40 months

Harousseau et al.,

1997

[11]


al., meta-analysis)

12 GOELAM

1987-1994

Multicenter (16)

Recruited n 5 535

Evaluable n 5 504


Randomized n 5 164

(ITT)

Auto 86

Chemo 78

#50 yrs

(36 yrs)

(Not

stated)

6.5

3

62 mos

Miggiano et al., 1996

[12]

21 Retrospective

Nonrandomized

1987-1994

Single center

Included n 5 89

Total 89

(Actual TX)

Auto 51

Chemo 38

\60 yrs

(mean age)

(36)

(44)

(Not

stated)

0

(Not

stated)

Mean

Follow-up

35 mos

Outcome Defined

(Interval)

% OS

(Signif)

Time from

randomization to

relapse or death

in CR1

(7-yr OS)

57

45

(P 5 .20)

Not stated (5-yr OS)

39 ± 6

56 ± 6

(P 5 .08)


K-M curves only

(Auto versus Chemo,

P 5 .009)

Time from CR1

to 1st relapse or

death from any

cause

(4-yr OS)

43 ± 9

52 ± 9

(P 5 .05)

Time from CR1 to

relapse or death

in CR

(5-yr OS)

60

65

(P not signif)

146

D.M

.Olian

sky

etal.


Reference [No.]

Quality

and

Strength

of



Study Group

Upper

Limit

(median)

Age at

Diagnosis

(Interval)

% Tx-

Related

Mortality

Median

Follow-up

(Interval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif)

Treatment Schema II: Induction 1 ICC, then Auto-SCT versus No Further Treatment

Burnett et al., 1998

[13]


al. meta-analysis)

12 UK MRC AML 10

1988-1995

Multicenter (163)

Recruited n 5 1966

Evaluable n 5 1857


Eligible for

Randomization

n 5 1001

Randomized n 5 381

(ITT)

Auto 190

Chemo only 191

\55 yrs

(Not

stated)

Not stated 4.8 yrs (7-yr DFS)

53

40

(P 5 .04)

Breems et al., 2004

[14]

12 HOVON & SAKK

1987-1995

Recruited n 5 646


Randomizable n 5 344

Randomized n 5 130

Randomized n 5 130

(ITT)

Auto 66

No further TX 64

(Chemo only)

\60 yrs

(Mean age

52 yrs)

Not stated 154 mos (5-yr DFS)

35 ± 6

44 ± 6

(P 5 .68)

Rohatiner et al., 2000

[15]

21 BXIII Protocol

1988-1994

Multicenter (3)

Recruited n 5 144


Pre-1988 Comparison

Group n 5 133

Total n 5 239 (ITT)

Auto 106

Chemo 133

\50 yrs

(38 yrs)

Not stated 5.5 yrs (5-yr RFS)

K-M curves only

(Auto versus Chemo,

P 5 .002)

Treatment Schema III: Induction, then ICC versus Auto-SCT

Cassileth et al., 1998

[16]


al., meta-analysis)

11 CALGB, ECOG,

SWOG

1990-1995

Intergroup,

multicenter study

Recruited n 5 808

Evaluable n 5 740

Achieved CR1 n5518

Randomized n5233

Randomized n 5 233

(ITT)

Auto 116

Chemo 117

# 55 yrs

(Not

stated)

(\100 d)

14

3

4 yrs (4-yr DFS)

35 ± 9

35 ± 9

(P 5 .77)

Bassan et al.,

1998

[17]

22 BXIII Protocol

Non-randomized

1987-1993

2 centers

Enrolled n 5 153

Achieved CR1 n 5108

Total n 5 65

Auto BMT 41

Chemo 24

#60 yrs

(37)

(55)

(Not stated)

7

8

Minimum

3.3 yrs

(5-yr RFS)

53

54

(P not signif)

Outcome Defined

(Interval)

% OS

(Signif)

Not stated (Not stated)

HR¶ 51.17

(CI 1.06 - 1.30)

(P 5 .003, Allo versus

alternative)

Time from

consolidation to

relapse or death

(4-yr OS)

54 ± 3

46 ± 2

(P 5 .09)

Time from CR1 to

relapse or death

(4-yr OS)

53 ± 5.5

53 ± 4.5

(P 5 .74)

Time from CR1 to

relapse, death, or

last visit

(3-yr OS)

65 ± 16

50 ± 13

(P not signif)

Time from CR1 to

relapse or death

in CR

(7-yr OS)

56

50

(P 5 .10)


48 (CI 43-53)

42 (CI 33-51)

(P 5 .28)

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T147


Reference [No.]

Quality

and

Strength

of



Study Group

Upper

Limit

(median)

Age at

Diagnosis

(Interval)

% Tx-

Related

Mortality

Median

Follow-up

(Interval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif)

ALLOGENEIC SCT VERSUS CHEMOTHERAPY ± AUTOLOGOUS SCT IN CR1

Myeloablative Conditioning for Allogeneic SCT

Yanada et al.

2005

[18]

11 Meta-analysis (5 allo-

SCT studies)

Published 1995-2003

Accrual 1986-1999

Total n 5 3100

Allo-SCT 1151

Alternative 1949

(Auto or chemo)

#55 yrs

(Not

stated)

Not stated Not stated Not stated

Cornelissen et al. 2007

[19]

211 HOVON-SAKK

AML4, AML29, and

AML42 trials

1987-2004

Recruited n 5 2287

Evaluable n 5 925

Total n 5 925 (ITT)

Donor 326

No Donor 599

(Auto or chemo)

#55 yrs

(39 yrs)

(4-yr)

21

4

63 mos (4-yr DFS)

48 ± 3

37 ± 2

(P \.001)

Harousseau

et al., 1997

[11]

211 GOELAM

1987-1994

Multicenter (16)

Recruited n 5 535

Evaluable n 5 504


Total n 5 222 (ITT)

Donor 88

No Donor 134

(Chemo)

#40 yrs

(Not

stated)

(Not stated)

22

3

62 mos (4-yr DFS)

44 ± 5.5

38 ± 4

(P 5 .62)

Reiffers et al., 1996

[10]

211 BGMT 87

1987-1990

Multicenter (4)

Eligible n 5 204


Total n 5 96 (ITT)

Donor 36

No Donor 60

(Auto or chemo)

#45 yrs

(32 yrs)

(34 yrs)

(Not stated)

12

3

Minimum

Follow-up

40 months

(3-yr DFS)

66 ± 16

56 ± 13

(P \.05)

Burnett et al., 2002

[20]

(Included in Yanada

et al. meta-

analysis)

21 UK MRC AML 10

1988-1995

Multicenter (163)

Recruited n 5 1966

Evaluable n5 1857

Achieved CR1 n51509

Total n 5 1287 (ITT)

Donor 419

No Donor 868

(including 200 auto)

\55 yrs

(Not

stated)

Not stated 80 mos (7-yr DFS)

50

42

(P 5 .01)

Gale et al., 1996

[21]

21 Retrospective

IBMTR (transplant);

GAMLCG (chemo)

1985-1993

Multicenter (401)

Included n 5 1097

Total n 5 1097

(Actual TX)

Allo 901

Chemo 196

#50 yrs

(30)

(36)

(5-yr)

43 (37-49)

7 (3-11)

Not stated (5-yr LFS)

46 (CI 42-50)

35 (CI 28-41)

(P 5 .01)

Outcome Defined

(Interval)

% OS

(Signif)

)

Time from CR1

to 1st relapse or

death in CR

(4-yr OS)

59 ± 4

46 ± 5

(P not signif)

Time from CR1

to 1st relapse or

death from any

cause

(4-yr OS)

46 ± 10

52 ± 9

(Chemo versus Allo,

P 5 .04)

o

Time from CR1

relapse or death

(4-yr OS)

42 ± 14

43 ± 18

19 ± 17

(Allo versus Maint.

Chemo

P 5 .047)


45 ± 24

53 ± 16

(P not signif)

Time from CR1 to

relapse or death

Not stated

Time from CR1 to

relapse, death, or

last contact

(7-yr OS)

41 ± 22

46 ± 19

(P 5 .10)

148

D.M

.Olian

sky

etal.


Reference [No.]

Quality

and

Strength

of



Study Group

Upper

Limit

(median)

Age at

Diagnosis

(Interval)

% Tx-

Related

Mortality

Median

Follow-up

(Interval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif)

Zittoun et al.,

1995

[9]

21 EORTC & GIMEMA

1986-1993

Multicenter (59)

Recruited n 5 990


Total n 5 294 (ITT)

Allo 168

Chemo 126

#59 yrs

(33 yrs)

Not stated 3.3 yrs (4-yr DFS)

55 ± 4

30 ± 4

(Allo versus chemo, P\.05


[16]

21 CALGB, ECOG,

SWOG

1990-1995

Intergroup,

multicenter study

Eligible n 5 808


Total n 5 230 (ITT)

Allo 113

Chemo 117

#55 yrs

(Not

stated)

(\100 d)

21

3

4 yrs (4-yr DFS)

43 ± 10

35 ± 9

(P not signif)

Cassileth et al.,

1992

[22]

21 ECOG

1984-1988

Recruited n 5 534

Evaluable n 5 449


Allo, n 5 54, compared

with subset of chemo

pts.\41 yr, n 5 50)

Total n 5 104 (ITT)

Allo 54

Consolidation

chemo 29

Maintenance

chemo 21

\ 41 yrs

(26)

(31)

(34)

(Not stated)

36

21

0

4 yrs (4-yr EFS)

42 ± 13

30 ± 17

14 ± 15

(Allo versus Maint. Chem

P 5 .01)

Schiller et al.,

1992

[23]

21 ALP 3 and 4

1982-1990

2 centers

Recruited n 5 103

Achieved CR1 n 5 82

Total n 5 82

(Actual TX)

Allo 28

Chemo 54

#45 yrs

(33 yrs)

(4-yr)

32

6

4 yrs (5-yr DFS)

48 ± 21

38 ± 14

(P not signif)

Willemze et al.,

1991

[24]

21 Retrospective

1983-1989

Single center

Included n 5 107

Total n 5 78

(Actual TX)

Allo 44

Chemo 34

#62 yrs

(41 yrs)

(8-yr)

17

6

18 mos (3-yr DFS)

30 (CI 19-45)

25 (CI not stated)

(P 5 .45)

Archimbaud et al,

1994

[25]

21 LYLAM-85

1985-1990

Single center

Recruited n 5 172

Subset \40 yr of age

n 5 78

Achieved CR1 n 5 58

Total n 5 58

(ITT)

Donor 27

No Donor 31

(Chemo)

\40 yrs

(31 yrs)

(Not

stated)

22

13

63 mos (7-yr LFS)

41 ± 22

27 ± 16

(P not signif)

erval)

Tx-

lated

rtality

Median

Follow-up

(Interval)

% DFS/

% EFS/

% RFS/

% LFS


(Interval)

% OS

(Signif)

Not

tated)

12

stated

31 mos (4-yr LFS)

54

30

(P 5 .01)

Time from DX to

relapse

(4-yr OS)

Data Not stated

(P 5 .04)

-yr)

56

7

60 mos

30 mos

(3-yr LFS)

26 (CI 20-32)

17 (12-23)

(P not signif)


chemotherapy; BMT, bone marrow transplantation; SCT, stem cell transplantation; CR, completeanization of Research and Treatment of Cancer; GIMEMA, Gruppo Italiano Malattie Ematologichegy Group; SWOG, Southwest Oncology Group; GOELAM, Groupe Ouest Est Leuc�emies Aiguesative Group; ALP, acute leukemia protocol; HOVON, Dutch Hemato-Oncology Working Party;

-treat; DFS, disease-free survival; EFS, event-free survival; RFS, relapse-free survival; Signif., signif-galy; PS, performance status; RR, rate ratio; CI, 95% confidence interval.

motherapy.

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T149


Reference [No.]

Quality

and

Strength

of



Study Group

Upper

Limit

(median)

Age at

Diagnosis

(Int

%

Re

Mo

REDUCED INTENSITY CONDITIONING (RIC) FOR ALLOGENEIC SCT

Mohty et al., 2005

[26]

22 Retrospective

1999-2003

Single center

Included n 5 95

Total n 5 95 (ITT)

Donor 35

No Donor 60

#65 yrs

(52 yrs)

(

s

Not

ALLOGENEIC SCT VERSUS CHEMOTHERAPY IN CR2

Gale et al.,

1996

[27]

21 Retrospective

MRC, ECOG, IBMTR

1980-1989

Multicenter (801)

Total n 5 501

Allo 257

Chemo 244

#50 yrs

(26 yrs)

(35 yrs)

(3

*Quality and strength of evidence definitions are listed in Table 1.†AML indicates acute myelogenous (myeloid) leukemia; Allo, allogeneic; Auto, autologous; Chemo,remission; DX, diagnosis; TX, treatment; MRC, Medical Research Council; EORTC, European OrgMaligne dell’Adulto; CALGB, Cancer and Leukemia Group B; ECOG, Eastern Cooperative OncoloMy�eloblastiques; IBMTR, International Bone Marrow Registry; GAMLCG, German AML CooperSTT, short-term therapy; FAB, French, American, British morphology classification; ITT, intent-toicance (probability); OS, overall survival; WBC ct, white blood cell count; Hepato., hepatosplenome‡Auto versus allo-SCT outcomes are presented in Table 5.§Relative Risk (RR) less than 1 indicates significant survival benefit for auto-SCT as compared to che¶Hazard Ratio (HR) greater than 1 indicates allo-SCT is superior to nonallogeneic SCT.

Further Treatment

rses:

mg/

to

If donor available, Allo-SCT (34% of pts) using

Cy (60 mg/kg/day � 2) 1 TBI (12 Gy)

Or

Bu (4 mg/kg/day � 4) 1 Cy (60 mg/kg/day X 2)

If no donor, randomized to HiDAC (Ara-C 4

mg/m2/day � 4 1 DNR 45 mg/m2/day � 3)

Or

Auto-SCT (55% of pts) using Bu (4 mg/kg/day

X 4) 1 Cy (60 mg/kg/day � 2)

es:

/m2/

If donor available and #45 years, Allo-SCT

using Bu (4 mg/kg/day � 4) 1 Mel (140/m2/

day � 1)

If no donor or . 45 years, randomized to 2

years maintenance using Ara-C (100 mg/m2/

day � 5) 1 DNR (1 mg/kg i.v. � 1)

administered at 1, 3, 6, 9, and 12 mos

Or

Auto-SCT using Bu (4 mg/kg/day � 4) 1 Mel

(140/m2/day � 1)

If donor available and #40 years, Allo-SCT

using myeloablative regimen:

Bu (4 mg/kg/day � 4) 1 Cy (50 mg/kg/day � 4 or

60 mg/kg/day � 2)

Or

Cy (60 mg/kg/day � 2) 1 TBI (dose not stated)

If no donor or .40 years, randomized to:

second ICC using AMSA (150 mg/m2/day � 5) 1

VP-16 (100 mg/m2/day � 5)

Or

Auto-SCT using Bu (4 mg/kg/day � 4) 1 Cy

(50 mg/kg/day � 4)

Auto-SCT using Bu (4mg/kg/day X 4) 1 Cy

(50 mg/kg � 4 days or 60 mg/kg/day � 2 days)

Nonrandomized control group received

monthly maintenance chemotherapy (drug

and dosage not stated)

150

D.M

.Olian

sky

etal.

Table 5. Treatment Schema Used in Autologous SCT versus Chemotherapy Studies

SCT versus Chemo Study

[Strength of Evidence*] [Ref No.] Induction Intensive Consolidation

Induction 1 ICC, then Autologous SCT versus Additional ICC

Zittoun et al., 1995 [11] [9] DNR†(45 mg/m2/day � 3) 1 Ara-C (200 mg/m2/

day � 7); second course if partial or no CR

For patients who achieved CR after 1-2 cou

AMSA (120 mg/m2/day � 3) 1 Ara-C (2000

m2/day � 6 for first 75 pts, then decreased

1000 mg/m2/day � 6 for subsequent pts)

Reiffers et al., 1996 [11] [10] Ara-C (100 mg/m2/day � 10) 1 DNR (60 mg/m2/

day � 3); second course if partial or no CR

For all pts who achieved CR after 1-2 cours

Ara-C (100 mg/m2/day � 7) 1 DNR (60 mg

day � 2)

Harousseau et al., 1997 [11] [11] Ara-C (200 mg/m2/day � 7) 1 IDR (8 mg/m2/day

� 5) or RBZ (200 mg/m2/day � 4)

For pts in CR waiting for Allo-BMT:

AMSA (150 mg/m2/day � 1) 1 Ara-C

(100 mg/m2/day � 5)

For all other pts in CR:

HiDAC (6 g/m2/day � 4) 1 IDR

(10 mg/m2/day � 2)

Or

HiDAC (6000 mg/m2/day � 4) 1 RBZ

(200 mg/m2/day � 2)

Miggiano et al., 1996 [21] [12] (Exact dosage not provided)

DNR 1 Ara-C ± VP-16

Or

ICE (IDR 10 mg/m2 1 Ara-C 1 VP-16)

Or

IDR (10 mg/m2) ± Ara-C

For all pts who achieved CR after

1-2 courses: (Exact dosage not provided)

Ara-C 1 AMSA

Or

Ara-C (conventional dose) 1 MITO

Further Treatment

If donor available, pts received 1 course

MidAC: MITO (10 mg/m2/day � 5) 1 Ara-C

(2000 mg/m2/day � 3); followed by Allo-SCT

using Cy (120 mg/kg/day � 2) 1 TBI (1200 or

1440 Gy)

No donor available, pts received 1 dose MidAC,

then randomized to:

No further treatment

Or

Auto-SCT using Cy (120 mg/kg/day � 2) 1 TBI

(1200 or 1440 Gy)

If donor available, pts received Allo-SCT

(regimen not stated)

No donor available, pts randomized to:

No further treatment

Or


(1550 mg/m2/day � 4)

Auto-SCT using Ara-C (2000 mg/m2/day � 6) 1

TBI (400 cGy/day � 3)

Nonrandomized, historical control group

received no further treatment

If donor available, pts received Allo-SCT using

Bu (4 mg/kg/day � 4) 1 Cy (50 mg/kg/day � 4)

No donor available, pts randomized to:

HiDAC (6000 mg/m2/day � 6)

Or


(50 mg/kg/day � 4)

Pts . 50 years received 1 course HiDAC

(2000 mg/m2/day � 6) ± Dox at the

physician’s discretion (dose not stated)

Pts 15-50 years received HiDAC (2000 mg/m2/

day � 6) 1 TBI (400 cGy/day � 3) 1 Auto-SCT

mide; TBI, total body irradiation; Bu, Busulfan, IDR,

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T151


SCT versus Chemo Study

[Strength of Evidence*] [Ref No.] Induction Intensive Consolidation

Induction 1 ICC, then Autologous SCT versus No Further Treatment

Burnett et al., 1998 [12] [13] Pts randomized to:

DNR (50 mg/m2/day � 3) 1 Ara-C (200 mg/m2/

day � 10) 1 TG (200 mg/m2/day � 10);

followed by DNR (50 mg/m2/day � 3) 1

Ara-C (200 mg/m2/day � 8) 1 TG

(200 mg/m2/day � 8) whether in CR after

first course or not

Or

Ara-C (200 mg/m2/day � 10) 1 DNR

(50 mg/m2/day � 3) 1 VP-16 (100 mg/m2/

day � 5); followed by Ara-C (200 mg/m2/day

� 8) 1 DNR (50 mg/m2/day � 3) 1 VP-16 (100

mg/m2/day � 5) whether in CR after first

course or not

All pts in CR received MACE [AMSA (100 mg/

m2/day � 5) 1 Ara-C (200 mg/m2/day � 5) 1 VP-

16 (100 mg/m2/day � 5)]

Breems et al., 2004 [12] [14] DNR (45 mg/m2/day � 3) 1 Ara-C (200 mg/m2/

day � 7); followed by AMSA (120 mg/m2/day �3) 1 Ara-C (2000 mg/m2/day � 6) whether in

CR after first course or not

All pts who achieved CR received:

MITO (10 mg/m2/d � 5) 1 VP-16

(100 mg/m2/day � 5)

Rohatiner et al., 2000 [21] [15] Dox (25 mg/m2 i.v. � 3) 1 Ara-C (100 mg/m2 i.v.

� 7) 1 6-TG (100 mg/m2 po � 7); second

course if partial or no CR

All pts who achieved CR received 2 additional

cycles of induction regimen

Induction, then ICC versus Autologous SCT

Cassileth et al., 1998 [11] [16] IDR (12 mg/m2/day � 3) 1 Ara-C (25 mg/m2

followed by 100 mg/m2/day � 7)

All pts who achieved CR received another

course of induction therapy at the same daily

doses, but with only 2 days of IDR and 5 days

of Ara-C

Bassan et al., 1998 [22] [17] Dox (25 mg/m2/day � 3) 1 Ara-C (100 mg/m2/

day � 7) 1 6-TG (200 mg/m2/day � 7)

All pts in CR received 2 additional induction

therapy courses

ICC, intensive consolidation chemotherapy; DNR, daunorubicin; AMSA, Amsacrine; Ara-C, Cytarabine; HiDAC, high-dose Ara-C; Cy, cyclophosphaIdarubicin; Mel, Melphalan, TG, Thioguanine; VP-16, etoposide; MITO, Mitozantrone; RBZ, Rubidazone; Dox, Doxorubicin.

*Quality and strength of evidence definitions are listed in Table 1.


patients in the UK MRC AML-10 trial. Of these, 419had a matched sibling donor and 644 had no match.An additional 224 patients had no siblings and wereincluded in the no-donor group. Of the 419 patientswith a donor, 39% did not receive an allo-transplant.Of the 868 patients with no donor, 661 received che-motherapy and 207 received a transplant (auto-BMT,n 5 200). Analyses were on an ITT basis and wereadjusted for cytogenetic risk group.

Gale et al. [21] reported the outcomes of a retro-spective study of 1097 adults (16-50 years) withAML. Patients who received an allo-BMT (n 5 901)and were reported in the International Bone MarrowTransplant Registry (IBMTR) were compared to 196patients treated with chemotherapy in the GermanAML Cooperative Group (GAMLCG) AML86 trial.The cohorts differed significantly by sex, age, WBCcount at diagnosis, and FAB types, which wereadjusted for in the multivariate analysis of outcomes.Figure 4 compares the adjusted probability of LFSbetween the allo-BMT and chemotherapy groups.

Cassileth et al. [22] compared the outcomes of allo-BMT versus chemotherapy in 534 adult (15-65 years)patients with AML enrolled in an ECOG study. Four-teen patients were cancelled and 39 deemed ineligible(inadequate pathology submission, n 5 13; age .65years, n 5 2; other eligibility violations, n 5 6; anderroneous diagnosis, n 5 18). Another 32 patientswere excluded from analysis because of major protocolviolations (n 5 19) or no follow-up documentation(n 5 13). Of the 449 evaluable patients, 305 (68%)achieved CR1. A subset of 104 patients \41 yearswas assigned to allo-BMT (n 5 54) if an HLA-compat-ible sibling donor was available, or randomly assignedto consolidation chemotherapy (n 5 29) or mainte-nance chemotherapy (n 5 21). Forty-five (83%) ofthe patients assigned an allo-BMT underwent thetreatment. Analyses were on an ITT basis. Except for

Figure 4. Adjusted probability of LFS of persons 16-50 years old

with AML in first remission in the IBMTR and GAMLCG data-

bases. Numbers in parentheses indicate numbers of persons at risk

at different intervals. (Reprinted with permission; [21].)

age, there were no significant differences among thegroups in patient or leukemic characteristics (sex, per-formance status, degree of weight loss, marrow cellu-larity, splenic involvement, FAB types, cytogenetics,WBC count at presentation, number of circulatingblasts, percentage of marrow blasts, number of induc-tion cycles needed to achieve CR1, and time to CR1).

Schiller et al. [23] reported the outcomes of 103adult (16-45 years) patients with AML treated accord-ing to the Acute Leukemia Protocol (ALP) 3 and 4regimens comparing allo-BMT versus chemotherapy.Eighty-two (80%) of the patients achieved CR1.Of these, 28 patients were assigned to undergo allo-BMT and were compared with 54 age-matched patientsassigned to consolidation chemotherapy. The medianinterval from remission to allo-BMT was 44 days.Patient characteristics of the 2 groups were similar forage, sex, WBC count at diagnosis, FAB type, numberof cycles needed to achieve CR1, and cytogenetics.

Willemze et al. [24] presented the results of a singlecenter, retrospective study of 107 adult (15-65 years)AML patients, comparing allo-BMT, auto-BMT,and chemotherapy postremission therapies. One pa-tient died during the first consolidation course andwas excluded from the analysis. Thirty-four patientsreceived ICC, 28 were allocated to auto-BMT, and44 to allo-BMT. There were no significant differencesamong the 3 groups by age, sex, FAB type, or meaninterval (days) to reach CR1.

Archimbaud et al. [25] presented the results of 78adult (17-39 years) AML patients treated on theLYLAM-85 protocol comparing allo-BMT versuschemotherapy at a single center in France, of whom58 (74%) achieved a CR1. Twenty-seven of thesepatients had an HLA-identical sibling donor andwere assigned to allo-BMT (donor arm), and 20 (74%)actually underwent the procedure. The 31 patientswithout a matched sibling donor were assigned to thechemotherapy (no donor) arm, 24 (77%) of whomreceived the scheduled 3 courses of chemotherapy.Analyses were on an ITT basis. The 2 groups were sim-ilar with regard to age, sex, WBC count at diagnosis,FAB, percent blasts, platelet count, number of cyclesto CR1, days to CR1, and cytogenetics. There werealso no differences between the groups for the presenceof hemorrhage, organomegaly, or fever at diagnosis.

Reduced intensity conditioning (RIC) for allo-SCT. Mohtyet al. [26] presented the results of 95 adult (26-65 years)patients with AML enrolled at a single center in France.All 95 patients were considered candidates for RICallo-BMT because of their high-risk leukemic or clini-cal characteristics. Patients with an HLA-identicalsibling donor were assigned to the donor group (n 5

35, 37%), whereas the remaining 60 patients withouta donor were assigned to the no donor group andtreated according to ‘‘standard institutional proce-dures.’’ In the donor group, 25 (71%) patients actually


underwent allo-BMT. The reasons for not receiving anallo-BMT included patient refusal (n 5 60), earlyrelapse (n 5 2), and psychiatric disorders (n 5 2). Out-come analyses were on an ITT basis. There were nosignificant differences between the 2 groups by age,sex, type of AML, FAB, cytogenetic risk group, ornumber of induction cycles to achieve CR1.

Allogeneic SCT versus Chemotherapy in SecondComplete Remission (CR2)

Gale et al. [27] presented the outcomes of 501adult (\50 years) patients with AML in CR2 who un-derwent chemotherapy (n 5 244) in MRC, ECOG, orsingle center trials, or HLA-matched sibling donorallo-HSCT (n 5 257) at IBMTR centers. Allo-HSCTpatients were younger, had briefer first remissions,and more had FAB M2 subtype than chemotherapy pa-tients. The 2 groups were similar with respect to sex andWBC count at diagnosis. Outcomes were adjusted fortime-to-treatment, age, and duration of CR1.

AUTOLOGOUS VERSUS ALLOGENEIC HSCT

Three studies previously described in the HSCTversus Chemotherapy section [9,16,24] also includedcomparisons between allo-HSCT versus auto-HSCToutcomes, which are presented in Table 6. This sec-tion provides descriptions of additional studies focus-ing exclusively on the comparison of allo-HSCTversus auto-SCT, which are also presented in Table 6.The highest quality studies are presented first; studiesof equal quality are presented in descending order bysample size.

Ringd�en et al. [28] presented the results of a retro-spective study of 4536 patients (75% .17 years) withAML in CR1 reported to the European Group forBlood and Marrow Transplantation (EBMT) registry,comparing auto-BMT (n53796), HLA-identical sib-ling allo-BMT (n 5 696), and twins allo-BMT (n 5 44).Auto-BMT patients tended to be older, the year oftheir transplant was later, and the interval from diagno-sis to BMT was longer than for the other 2 groups.

Lazarus et al. [29] presented the retrospective anal-ysis of 1144 AML patients (#58 years; 81% .10 years)transplanted in CR1 or CR2 and reported to the Cen-ter for International Blood and Marrow TransplantResearch (CIBMTR), comparing the outcomes ofauto-BMT (n 5 668 patients) to unrelated donor(URD) allo-BMT (n 5 476 patients). Analysis of pa-tient characteristics between the 2 groups indicatedthat a higher proportion of auto-BMT patients were\ 10 years of age, and allo-BMT patients were morelikely to be male, have a lower performance score, un-favorable cytogenetics, and require more time toachieve CR1. Median times from remission to trans-plantation were 102 and 104 days for auto-BMT andallo-BMT patients, respectively. Multivariate analyses

considered the impact of disease status, FAB, cytoge-netic risk group, and cytomegalovirus (CMV) statuson outcomes.

Suciu et al. [30] compared the outcomes of allo-BMT versus auto-HSCT in 1198 adult (15-46 yrs) pa-tients with AML enrolled in the EORTC/GIMEMAAML-10 protocol. Patients with APL were excluded.Of the 1198 patients, 62 could not be evaluated becauseof a lack of clinical documentation. Of the remaining1136 patients, 822 (72.4%) achieved CR1. Fifty pa-tients were excluded from further protocol treatmentbecause of toxicity (n 5 27), poor documentation(n 5 15), or other reasons (n 5 8). Of the 772 patientswho received consolidation, 38 were not HLA-typedbecause of patient refusal (n 5 9), early death/toxicity(n 5 20), logistic reasons (n 5 5), or other reasons(n 5 4). Of the remaining patients, 293 had an HLA-matched sibling donor (donor group) and were as-signed to receive an allo-HSCT. The remaining 441patients either had no siblings or no HLA-matched sib-ling and were assigned to receive an unpurged auto-PBSCT or BMT (no-donor group). Allo-HSCT wasactually performed in 202 of 293 (68.9%) donor grouppatients, and 246 of 441 (55.8%) patients in the no-donor group received the assigned auto-HSCT. Eleven(2.5%) no-donor group patients received an allo-BMTwith stem cells from a matched unrelated or matchedrelated donor, while 7 (2.5%) patients with a sibling do-nor underwent auto-HSCT. Outcome analyses wereon an ITT basis. Patient characteristics, includingage, WBC count at diagnosis, FAB subtype, cytogenet-ics, and the CR rate after the first induction course weresimilar in the 2 groups. Figure 5 compares the DFSbetween the donor and no donor groups.

Figure 5. DFS from CR according to donor availability. The esti-

mates of the 4-year DFS rates (6SE) for the donor group (dotted

line) and the no donor group (solid line) are given. The 4-year cumu-

lative incidence of relapse and of death in CR are given in italics at

the right of the graph. N indicates the number of patients; O, ob-

served number of events (relapse or death in first CR). P is deter-

mined by the log-rank test. (Reprinted with permission; [30].)

rval)

FS/

S/

FS/

FS

nif)

Definition of DFS/

EFS/

RFS/

LFS

(Interval)

% OS

(Signif)

LFS)

± 2

± 8

± 1

Allo

sus

uto

001)

Time from BMT

to relapse

or death

Not stated

LFS)

29-38)

42-50)

01)

Onset not stated;

events were

relapse or death

(5-yr OS)

36

(CI 31-40)

51

(CI 47-55)

(P \.001)

DFS)

.2

.2

.04)

Time from CR1 to

first relapse or

death in CR

(4-yr OS)

58.3

50.8

(P 5 .18)

DFS)

± 4

± 5

tated)

Time from CR1 to

first relapse or

death in CR

(4-yr OS)

59 ± 4

56 ± 5

(Not stated)

DFS)

10

± 9

signif)

Time from CR1 to

first relapse

or death

(4-yr OS)

46 ± 10

43 ± 9

(P not signif)

tated Not stated (8-yr OS)

52 (CI 40-61)

49 (CI 33-62)

(P not signif)

154

D.M

.Olian

sky

etal.

Table 6. Comparison of Patient Characteristics and Outcomes from Studies Included in the Autologous SCT versus Allogeneic SCT Section

Reference No.

Quality

and

Strength

of



Study Group

Upper

Limit

(Median)

Age at

Diagnosis

(Interval)

% TX-

Related

Mortality

Median

Follow-

up

(Inte

%D

%EF

%R

%L

(Sig

Ringd�en et al.,

2000

[28]

211 Retrospective

EBMT

1987-1999

Included n 5 6306

AML n 5 4536 (72%)

Total AML n 5 4536

HLA-ident. Sib allo 696

Twins allo 44

Auto 3796

#77 yrs

(30 yrs)

(2-yr)

12 ± 1

5 ± 3

11 ± 1

32 mos (2-yr

64

59

51

(Sib

ver

A

P \.0

Lazarus et al.,

2006

[29]

211 Retrospective

CIBMTR

1989-1996

Multicenter (12)

Included n 5 1144

Total n 5 1144

(Actual TX)

MUD Allo 476

Auto 668

#58 yrs

(25 yrs)

(27 yrs)

(100 day)

31

6

95 mos

81 mos

(5-yr

34 (CI

46 (CI

(P\.0

Suciu et al., 2003

[30]

(Included in

Yanada et al.

meta-analysis)

21 Prospective

EORTC and GIMEMA

AML-10 Trial 1993-1999

Multicenter (80)

Recruited n 5 1198

Evaluable n 5 1136


Total n5734 (ITT)

Donor 293

No Donor 441

(Auto only)

#45 yrs

(35 yrs)

(33 yrs)

(Not stated)

20

5

4 yrs (4-yr

52

42

(P 5

Zittoun et al., 1995

[9]

(Described in SCT

versus Chemo

section)

21 Prospective EORTC and

GIMEMA 1986- 1993

Multicenter (59)

Recruited n 5 990


Total n 5 296

(ITT)

Allo 168

Auto 128

#59 yrs

(33 yrs)

Not stated 3.3 yrs (4-yr

55

48

(Not s


[16]

(Described in SCT

versus Chemo

section)

21 Prospective CALGB,

ECOG, SWOG

1990-1995

Multicenter

Eligible n 5 808


Total n 5 229 (ITT)

Allo 113

Auto 116

#55 yrs

(Not

stated)

(\100 d)

21

14

4 yrs (4-yr

43 ±

35

(P not

Carella et al.,

1992

[31]

21 Retrospective

1983-1991

Single center

Included n 5 159

Total n 5 159

(Actual TX)

Allo 104

Auto 55

\46 Allo

\62 Auto

(26 yrs)

(37 yrs)

(Not

stated)

24

12

45 mos

29 mos

Not s

rval)

FS/

FS/

FS/

FS

gnif)

Definition of DFS/

EFS/

RFS/

LFS

(Interval)

% OS

(Signif)

DFS)

2

7

.02)


13

29

(P 5 .004)

EFS)

37-75)

31-67)

.54)


DFS)

19-45)

19-35)

signif)

Time from CR1

to relapse or

death

Not stated

DFS)

± 22

± 16

stated)


RFS)

1

5

.12)


66

37

(P 5 .05)

LFS)

± 7

± 6

.08)


32 ± 7

50 ± 6

(P 5 .07)

cell transplantation; CR, complete remission; TX,od stem cell transplant; EORTC, European Orga-a Group B; ECOG, Eastern Cooperative Oncologyransplantation; CIBMTR, Center for InternationalFS, event-free survival; RFS, relapse-free survival;

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T155


Reference No.

Quality

and

Strength

of



Study Group

Upper

Limit

(Median)

Age at

Diagnosis

(Interval)

% TX-

Related

Mortality

Median

Follow-

up

(Inte

%D

%E

%R

%L

(Si

Singhal et al.,

2003

[32]

21 Retrospective

1984-1999

2 centers

Included n 5 130 AML

Total n 5 130

(Actual TX)

PMRDT Allo 80

Auto 50

#73 yrs

(23 yrs)

(18 yrs)

(5-yr)

62

20

5 yrs (5-yr

1

2

(P 5

Mitus et al., 1995

[33]

21 Prospective

1987-1993

Multicenter (3)

Recruited n 5 94

Achieved CR1 n 5 84

Total n 5 84 (ITT)

Allo 31

Auto 53

#63 yrs

(33 yrs)

(42 yrs)

(Not stated)

32

9

Not

stated

(5-yr

56 (CI

45 (CI

(P 5

Willemze et al.,

1991

[24]

(Described in SCT

versus Chemo

section)

21 Retrospective

1983-1989

Single center

Included n 5 107

Total n 5 72

(Actual TX)

Allo 44

Auto 28

#62 yrs

(41 yrs)

(8-yr)

17

7

18 mos (3-yr

30 (CI

40 (CI

(P not

Cassileth et al.,

1993

[34]

21 Prospective

ECOG P-C 486

No dates provided

Multicenter

Recruited n 5 123

Evaluable n 5 111

Achieved CR1 n 5 83

Total n 5 58 (ITT)

Allo 19

Auto 39

#55 yrs

(32 yrs)

(36 yrs)

Not stated 31 mos (3-yr

42

54

(P not

Lowenberg et al.,

1990

[35]

21 Retrospective HOVON

1984 -1987

Recruited n 5 117

Achieved CR1 n 5 90

Total n 5 53

(Actual Tx)

Allo 21

Auto 32

#60 yrs

(43 yrs)

(3-yr)

19

9

30 mos (3-yr

5

3

(P 5

Sierra et al., 1996

[36]

22 Prospective 1988-1993

Multicenter (10)

Recruited n 5 159


Total n 5 115 (ITT)

Allo 47

Auto 68

#50 yrs

(30 yrs)

(Not stated)

23

0

45 mos

37 mos

(4-yr

31

50

(P 5

*Quality and strength of evidence definitions are listed in Table 1.†AML indicates acute myelogenous leukemia; Allo, allogeneic; Auto, autologous; Chemo, chemotherapy; BMT, bone marrow transplantation; SCT, stemtreatment; URD, unrelated donor; MUD, matched-rnrelated donor; PMRDT, partially mismatched related donor transplant; PBSCT, peripheral blonization of Research and Treatment of Cancer; GIMEMA, Gruppo Italiano Malattie Ematologiche Maligne dell’Adulto; CALGB, Cancer and LeukemiGroup; SWOG, Southwest Oncology Group; HOVON, Dutch Hemato-Oncology Working Party; EBMT, European Group for Blood and Marrow TBlood and Marrow Transplant Research; FAB, French, American, British morphology classification; ITT, intent-to-treat; DFS, disease-free survival; ESignif., significance (probability); OS, overall survival; CI, 95% confidence interval.


Carella et al. [31] retrospectively analyzed 159 adult(#62 years) patients with AML in CR1 who received anallo-BMT (n 5 104) or unpurged auto-BMT (n 5 55)at a single institution. Median times from CR1 to trans-plant were 150 and 144 days for allo-BMT and auto-BMT patients, respectively. Patient characteristics forthe 2 groups, including sex, age, FAB, and WBC countat diagnosis, were not compared.

Singhal et al. [32] presented a retrospective analysisof adult (62% .15 years) patients with AML (n 5 130,44% of sample) beyond first CR, who received eithera partially mismatched related donor (PMRD) T cell-depleted allo-BMT (n 5 80, 61.5%) or an unpurgedauto-BMT (n 5 50, 38.5%). Patients in CR1, those un-dergoing a second auto-BMT, and those with knownsecondary leukemia were excluded. Age and sex distri-bution were comparable between the 2 treatmentgroups. PMRD allo-BMT patients were less likelythan auto-BMT patients to be in CR2 and more likelyto have active disease at the time of transplant.

Mitus et al. [33] reported the outcomes of 94 adult(17-63 years) patients with AML enrolled in a prospec-tive, multicenter study of induction therapy followedby allo-BMT versus auto-BMT. Patients with ante-cedent MDS, anemia, or other cytopenia were ex-cluded. Of the 94 patients, 84 (89%) achieved CR1.Patients \55 years with an HLA-compatible relativewere assigned to allo-BMT, and those without a donoror who were age 55-65 years were assigned to auto-BMT. Median times from CR1 to transplantationwere 236 and 190 days for auto-BMT and allo-BMT, respectively. One patient in each group under-went the opposite procedure. Outcome analyses wereon an ITT basis. There were no significant differencesbetween the 2 treatment groups by age, sex, FAB type,WBC count at diagnosis, or cytogenetics.

Cassileth et al. [34] reported the results of 123adult (15-55 years) patients with AML enrolled inthe ECOG P-C 486 trial. Twelve (10%) patientswere deemed ineligible for various reasons. Of the111 eligible patients, 83 (75%) achieved CR1.Twenty-five (30%) patients were off-study becauseof infections or other disease (n 5 14), patient refusal(n 5 9), or other reasons (n 5 2). The remainingpatients were assigned to allo-BMT (n 5 19) orauto-BMT (n 5 39). Patient characteristics were notsignificantly different between the 2 groups, exceptthat allo-BMT patients were significantly youngerthan auto-BMT patients. The median time fromCR1 to auto-BMT was 8 weeks, and 35 of 39 patientshad an auto-BMT. The median time from CR1 to allo-BMT was 9 weeks, and 17 of 19 patients underwent theprocedure. Analyses were on an ITT basis.

Lowenberg et al. [35] presented the outcomes of117 adult (15-60 years) patients with AML enrolledin a Dutch Hemato-Oncology Working Party(HOVON) study comparing allo-BMT, auto-BMT,

and no further treatment. Ninety (77%) patientsachieved CR1 and, of these, 37 (41%) patients whowere eligible for auto-BMT received no further treat-ment because of early relapse or death (n 5 19), refusal(n 5 6), insufficient graft (n 5 3), chronic infection(n 5 3), or other reasons (n 5 6). Of the remaining53 patients, 32 (36%) underwent an unpurgedauto-BMT and 21 (23%) underwent an HLA-matchedallo-BMT. Median times from CR1 to transplantationwere 3.8 and 3.0 months for auto-BMT and allo-BMT, respectively. Auto-BMT patients were, onaverage, 10 years older than allo-BMT patients. Therewere no significant differences between the allo-BMTand auto-BMT groups by sex, FAB type, or time toCR1. Analyses were by actual treatment received.

Sierra et al. [36] presented the outcomes of adult(79% $15 years) patients with AML enrolled in a pro-spective, multicenter study comparing allo-BMT ver-sus auto-BMT. Patients .51 years or those havinga history of MDS, previous treatment with cytotoxicdrugs or radiation, or a severe concomitant diseasewere excluded. Of 159 patients, 120 (75%) achievedCR1, and 115 were HLA-typed. Forty-seven (41%)patients had an HLA-identical sibling donor and wereassigned to allo-BMT; 26 (55%) underwent the proce-dure. Reasons for not receiving the assigned allo-BMTincluded medical contraindications (n 5 13) and earlyrelapse (n 5 8). Sixty-eight patients (59%) were as-signed to auto-BMT and, of these, 47 (69%) underwentthe procedure. Reasons for not receiving the assignedauto-BMT included medical contraindications (n 5 9),early relapse (n 5 8), and patient refusal (n 5 3). Out-come analyses were on an ITT basis. Median intervalsfrom CR1 to transplantation were 166 and 147 daysfor auto-BMT and allo-BMT, respectively. There wereno significant differences between the 2 groups on age,sex, WBC count at diagnosis, number of inductioncourses to CR1, or time to CR1.

AUTOLOGOUS HSCT

This section provides descriptions of auto-HSCTstudies, the outcomes of which are presented inTable 7. The highest quality studies are presented first;studies of equal quality are presented in descendingorder by sample size.

PBSCT Versus BMT

Reiffers et al. [37] presented the results of a retro-spective, multicenter study comparing the use ofPBSC versus BM for auto-HSCT in 1393 patients(#64 years) with AML and registered with theEBMT. Patients received an auto-PBSCT (n 5 100),purged auto-BMT (n 5 252), or unpurged auto-BMT (n 5 1041). The significantly different patientcharacteristics among the groups were age (greater inPBSCT patients than purged auto-BMT patients)

l)

y

Median

Follow-up

(Interval)

% DFS/

% EFS/

% RFS/

% LFS


(Interval)

% OS

(Signif)


44 ± 6

49 ± 2

57 ± 3

(PB vs Purged

BM, P 5 .01;

PB vs Unpurged

BM, P 5 .22)

Onset not stated;

events were

relapse or death

(2-yr OS)

53 ± 6

56 ± 2

66 ± 3

(PB vs Purged

BM, P 5 .10;

PB vs Unpurged

BM, P 5 .85)

Not stated Not stated Not stated (4-yr OS)

77

63

(P 5 .15)

d) 67 mos (6-yr DFS)

58

33

(P 5 .009)

Time from SCT to

relapse

or death

(6-yr OS)

54

35

(P 5 .02)

57 mos (3-yr LFS)

56 (CI 47-64)

31 (CI 18-45)

(P not stated)

Onset not stated;

events were

relapse or death

(3-yr OS)

63 (CI 52-70)

40 (CI 25-55)

(P not stated)

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T157

Table 7. Outcomes from Autologous SCT Studies

Reference No.

Quality

and

Strength

of


Number of Patients

by Study Group

Upper

Limit

(Median)

Age at

Diagnosis

(Interva

% Tx -

Related

Mortalit

PBSCT versus BMT

Reiffers et al.,

2000

[37]

21 Retrospective

EBMT Registry

1986-1994

Included n 5 1393

Total n 5 1393

PB Auto 100

Unpurged BM Auto 1041

Purged BM Auto 252

#64 yrs

(43 yrs)

(52 yrs)

(52 yrs)

(2-yr)

12 ± 5

11 ± 1

9 ± 2

Sirohi et al.,

2004

[38]

21 Retrospective,

matched

1996-2000

EBMT Registry

(n 5 114)

and Single center

(n 5 57)

Total n 5 171

PB Auto 114

Unpurged BM Auto 57

#66 yrs

(39 yrs)

(43 yrs)

(4-yr)

1

13

Vey et al.,

2004

[39]

21 Retrospective

1984-1998

Single center

Included n 5 101

Total n 5 101

PB Auto 43

BM Auto 58

#60 yrs

(46 yrs)

(Not state

0

8

Purged versus Unpurged

Miller et al.,

2001

[40]

21 Retrospective

ABMTR

1989-1993

Multicenter (41)

Included n 5 294

Subgroup CR1 w/in

6 mos n 5 209

Total CR1 w/in 6 mos

n 5 209

Purged BM auto 148

Unpurged BM auto 61

#60 yrs

(33 yrs)

(18 yrs)

(3-yr)

12

8

(Interval)

% DFS/

% EFS/

% RFS/

% LFS


(Interval)

% OS

(Signif)

(4-yr LFS)

63 ± 8

34 ± 7

(P 5 .05)

Onset not stated;

events were

relapse or death

Not stated

Not stated)

57

32

(P 5 .36)


(3-yr LFS)

67

40

(P 5 .052)


73

47

(Not stated)

(5-yr DFS)

8 (CI 63-73)

Time from CR1 to

relapse or

death in CR

(5-yr OS)

71

(8-yr EFS)

5 (CI 44-64)

Time from SCT to

relapse, 2ndary

AML or MDS, or

death

(8-yr OS)

62 (CI 50-72)

(3-yr LFS)

36 ± 5

Onset not stated;

events were

relapse or death

(3-yr OS)

47 ± 5

158

D.M

.Olian

sky

etal.


Reference No.

Quality

and

Strength

of


Number of Patients

by Study Group

Upper

Limit

(Median)

Age at

Diagnosis

(Interval)

% Tx -

Related

Mortality

Median

Follow-up

Gorin et al.,

1990

[41]

21 Retrospective EBMT

Registry 1982-1987

Multicenter (34)

Included n 5 263

Subgroup SR, CR1,

TBI n 5 107

Total CR1, SR, TBI

n 5 107

Purged BM auto 30

Unpurged BM auto 77

#47 yrs

(30 yrs)

Not stated 28 mos

Chao et al.,

1993

[42]

21 Retrospective

1987-1992

Single Center

Included n 5 50

Total n 5 50

Purged BM auto 30

Unpurged BM auto 20

#60 yrs

(40 yrs)

(Not

stated)

14

overall

31 mos (

Tandem versus Single

McMillan et al.,

1990

[43]

2- No dates stated

Multicenter

Rec’d first auto BMT

n 5 82

Eligible for second auto

n 5 54

Received second auto

n 5 26

Total n 5 51

Double auto 26

Single auto 25

#57 yrs

(40 yrs)

(Not stated)

6

overall

31 mos

Single Arm Cohort Studies—Purging status not stated

Kim et al.,

2004

[44]

211 Retrospective

1993—not stated

Single center

Included n 5 174

Total CR1 n 5 174 \65 yrs

(34 yrs)

(Not

stated)

4

51 mos

6

Molee et al.,

2004

[45]

211 Retrospective

1986-2001

Single center

Transplanted n 5 145

Transplanted in CR1

n 5 117

Total CR1 n 5 117 #65 yrs

(43 yrs)

(1-yr)

6

53 mos

5

Gorin et al.,

2000

[46]

21 Retrospective

1984-1998

EBMT Registry

Included n 5 193

CR1 only n 5 147

Total CR1

n 5 147

#69 yrs

(63 yrs)

(3-yr)

15 ± 4

14 mos

rval)

FS/

FS/

FS/

FS

nif) Outcome Defined

(Interval)

% OS

(Signif)

LFS)

± 5

Onset not stated;

events were

relapse

or death

(5-yr OS)

56 ± 5

tated,

S)

12

Time from BMT to

relapse

or TRM

(Not stated, OS)

72 ± 12

DFS)

± 7

Time from CR1 to

relapse

or death

(5-yr OS)

61 ± 6

LFS)

27-51)

36-61)

.12)

Time from CR1 to

relapse

or death

(5-yr OS)

41 (CI 29-54)

53 (CI 39-66)

(P 5 .14)

DFS)

2

Time from BMT to

relapse

or death

(10-yr OS)

45

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T159


Reference No.

Quality

and

Strength

of


Number of Patients

by Study Group

Upper

Limit

(Median)

Age at

Diagnosis

(Interval)

% Tx -

Related

Mortality

Median

Follow-up

(Inte

% D

% E

% R

% L

(Sig

Single Arm Cohort Studies—Purged BMT

Gorin et al.,

1999

[47]

21 Retrospective

1983-1997

Single center

Mafosfamide Purge

AML pts. n 5 165

CR1 only n 5 123


(37 yrs)

(5-yr)

18 ± 4

Not stated (5-yr

53

Linker et al.,

1998

[48]

21 Retrospective

1986-1993

2 Centers

4-HC Purge

Included n 5 165

CR1 only n 5 50

Total CR1 n 5 50 \ 60 yrs

(37 yrs)

(Not

stated)

4

6.8 yrs (Not s

DF

70 ±

Isnard et al.,

2001 [49]

2- Retrospective

1982-1994

Single center

Mafosfamide Purge

Included n 5 108

Achieved CR n 5 92

Rec’d transplant n 5 61

(89% purged auto

BMT; 11% allo)


(39 yrs)


55

Single Arm Cohort Studies—Unpurged BMT

Jourdan et al.,

2005

[50]

21 Prospective BGMT 95

Randomized

No dates stated

Multicenter

Enrolled n 5 437


Randomized to

HiDAC regimens

n 5 128

Total n 5 128 (ITT)

HiDAC 1 dose 65

HiDAC 2 dose 63

#60 yrs

(49 yrs)

(48 yrs)

(Not stated)

8

8

73 mos (5-yr

39 (CI

48 (CI

(P 5

Meloni et al.,

1996

[51]

21 Retrospective

1984-1994

Single center

CR2 patients

Total n 5 60

(78% adults)

Total CR2

n 5 60

#54 yrs

(28 yrs)


4

rval)

x -

ted

ality

Median

Follow-up

(Interval)

% DFS/

% EFS/

% RFS/

% LFS


(Interval)

% OS

(Signif)

tated 37 mos (2-yr DFS)

49 (CI 37-62)

Time from

consolidation

chemo to

relapse or death

Not stated

ot

ed)

2

30 mos (5-yr DFS)

55 (CI 45-65)

Time from

start of

consolidation

chemo to

relapse

or death

Not stated

ot

ed)

1

18 mos (Not stated,

DFS)

61

Not stated (Not stated, OS)

62

BMT, bone marrow transplantation; SCT, stem cell transplantation; CR, completew Transplant Registry; BGMT, Bordeaux-Grenoble-Marseille-Toulouse; ITT, in-robability); OS, overall survival; CI, 95% confidence interval.

160

D.M

.Olian

sky

etal.


Reference No.

Quality

and

Strength

of


Number of Patients

by Study Group

Upper

Limit

(Median)

Age at

Diagnosis

(Inte

% T

Rela

Mort

Stein et al.,

1996

[52]

21 Prospective

1989-1993

Single center

In CR1 n 5 60

Rec’d Auto BMT

n 5 44

Total CR1 n 5 60 (ITT) #55 yrs

(39 yrs)

Not s

Single Arm Cohort Studies – Unpurged PBSCT

Linker et al.,

2000

[53]

21 Prospective

1993-1998

Multicenter (6)

Enrolled in CR1

n 5 128

Rec’d auto PBSCT

n 5 117

Total CR1 n 5 128 (ITT) \70 yrs

(39 yrs)

(N

stat

Pavlovsky et al.,

1998

[54]

21 Prospective 1991-1996

Single center

Eligible in CR1 n 5 56

(80% adults)

Rec’d auto-PBSCT

n 5 54

Total CR1 n 5 56 (ITT) #64 yrs

(32 yrs)

(N

stat

1

*Quality and strength of evidence definitions are listed in Table 1.†AML indicates acute myelogenous (myeloid) leukemia; Allo, allogeneic; Auto, autologous; Chemo, chemotherapy;remission; TX, treatment; PBSCT, peripheral blood stem cell transplant; ABMTR, Autologous Blood and Marrotent-to-treat; DFS, Disease-free survival; EFS, event-free survival; RFS, relapse-free survival; Signif., significance (p


and FAB subtype (fewer patients with APL in thePBSCT group than in either the purged or unpurgedBM groups). The interval between CR1 and HSCTwas longer for purged or unpurged BMT than forPBSCT, and the proportion of patients receiving totalbody irradiation (TBI) was higher in the BMT groupsthan the PBSCT group.

Sirohi et al. [38] reported the results of a retrospec-tive matched case-control study comparing auto-BMTversus PBSCT in adult (16-66 years) patients withAML. Fifty-seven patients from a single institutionwho underwent an unpurged auto-BMT were matchedon cytogenetics, FAB subtype, CR1 to HSCT interval,and age with 114 patients from the EBMT registry whounderwent a PBSCT. Patients in the auto-BMT groupwere significantly older than the PBSCT patients, andWBC count at diagnosis was higher in the PBSCT pa-tients than the auto-BMT group. Median times fromCR1 to HSCT were comparable at 121 days and 119days in the BM and PBSC groups, respectively.

Vey et al. [39] retrospectively studied the outcomesof 101 adult (16-60 years) patients with AML who un-derwent an auto-SCT at a single center. Patients withAPL (n 5 4) were included. The stem cell source forpatients transplanted pre-1994 was BM (n 5 58); thosetreated after 1994 received PBSC (n 5 43). HSCT wasperformed at a median of 3.5 months from CR1. Therewere no significant clinical or biologic differencesbetween the 2 treatment groups.

Purged versus Unpurged

Miller et al. [40] presented the results of 209 patients(#60 years) reported to the Autologous Blood andMarrow Transplant Registry (ABMTR) who receivedeither a purged (n 5 148) or unpurged (n 5 61)auto-BMT for AML in CR1. BM was purged with4-hydroperoxycyclophosphamide (4-HC). Patientsreceiving purged BM were significantly younger, hadlower performance scores, were less likely to receiveconsolidation chemotherapy, and were transplantedsooner after CR1 than patients receiving unpurged BM.

Gorin et al. [41] presented the results of 263 pa-tients (86% adult) with AML from the EBMT registrywho underwent a purged or unpurged auto-BMT. Ofthe 263 patients, 231 were classified as standard risk(SR) and 32 were high risk (HR). A comparison ofpurged (n 5 30) versus unpurged (n 5 77) auto-BMT included SR patients autografted in CR1 afterTBI-containing regimen(s). Marrow was purged withmafosfamide. There were no significant differencesin patient characteristics between the 2 groups.

Chao et al. [42] retrospectively studied the out-comes of 50 adult (96% .15 years) patients treatedfor AML (n 5 48) or biphenotypic acute leukemia(n 5 2) at a single U.S. center with purged (n 5 30)or unpurged (n 5 20) auto-BMT. BM was purged

with 4-HC, and for the 2 biphenotypic acute leukemiapatients BM purging also included etoposide. Thirty-four patients were in CR1, 12 in CR2, and 4 in firstrelapse at time of transplantation. Median time fromCR1 to transplantation was 4 months. No significantdifferences in patient characteristics between the 2groups were reported.

Tandem versus Single

McMillan et al. [43] presented the results of a retro-spective study of 82 adult (16-57 years) patients withAML who underwent an unpurged auto-BMT. Therewere 4 (5%) procedure-related deaths and 9 (11%) re-lapses within 90 days of the first auto-BMT. Of the 69remaining patients, 54 (78%) were eligible to proceedto a second auto-BMT. Of these, 26 (48%) underwentthe second procedure. Reasons for not receiving thesecond BMT included patient refusal (n 5 22), toxicity(n 5 4), infection (n 5 1), failed harvest (n 5 1), andsevere psychologic problems (n 5 1). The outcomesof 25 patients who remained in remission at 90 days fol-lowing the single auto-BMT were compared to thosepatients who underwent the tandem auto-BMT. Thecomparison single-graft group differed from the tan-dem BMT group in that the interval between remissionand auto-BMT was significantly longer (P 5 .003).

Single-Arm Cohort Studies

The remaining 11 studies, presented only in Table 7,are single-arm, noncomparative studies examiningauto-BMT or auto-PBSCT: purging status not stated(3 studies); purged auto-BMT (3 studies); unpurgedauto-BMT (3 studies); and unpurged auto-PBSCT(2 studies).

ALLOGENEIC HSCT

This section provides descriptions of 22 allo-HSCT studies, the outcomes of which are presentedin Table 8. Seven comparative studies examined theimpact on patient survival of related versus unrelatedallo-HSCT (n 5 2), T cell-depleted (TCD) versus Tcell-replete HSCT (n 5 3), and the use of BM versusPBSC (n 5 2). The remaining 15 studies represent sin-gle arm, noncomparative studies of related and unre-lated HSCT (n 5 6) and related allo-HSCT only(n 5 9). Within each subsection, the highest qualitystudies are presented first; studies of equal quality arepresented in descending order by sample size.

Related versus Unrelated

Hegenbart et al. [55] presented the results of a pro-spective, multicenter study comparing related versusunrelated donor allo-HSCT following reduced inten-sity, TBI-based conditioning in 122 adult (17-74 years)patients with AML. Patients were eligible for unre-lated HSCT if they were .50 years old and for related

l)

/

/

/

)

Definition of

DFS/

EFS/

RFS/

LFS

(Interval)

% OS

(Signif)

S)

)


48

48

(P 5 .68)

S)

8)


32 overall

S)

)

Onset not stated;

events were

relapse or

death

Not stated

S)

71)

47)

57)

5 .002;

5 .24)


62 (CI 50-74)

35 (CI 22-49)

58 (CI 53-62)

(P 1 versus 2 5 .001;

P 1 versus 3 5 .27)

S)

)


S)

)

Time from SCT

to relapse

or death

(2-yr OS)

42 ± 6

38 ± 9

(P 5 .86)

162

D.M

.Olian

sky

etal.

Table 8. Outcomes from Allogeneic SCT Studies

Reference No.

Quality

and

Strength

of


Number of Patients

by Study Group

Upper Limit

(Median)

Age at

Diagnosis

(Interval)

% TX -

Related

Mortality

Median

Follow-

up

(Interva

% DFS

% EFS

% RFS

% LFS

(Signif

Related versus Unrelated—Reduced-Intensity Conditioning (RIC)

Hegenbart et al.,

2006

[55]

21 Prospective

1998-2002

Multicenter (13)

Total n 5 122

Related donor 58

Unrelated donor 64

#74 yrs

(57.5 yrs)

(2-yr)

10

22

44 mos (2-yr DF

43

45

(P 5 .93

Sayer et al.,

2003

[56]

21 Retrospective

1998-2000

Multicenter (10)

Total n 5 113

Related donor 63

Unrelated donor 50

#67 yrs

(51 yrs)

(2-yr)

53 overall

1 yr (2-yr EF

39.9

16.7

(P 5 .02

T cell Depleted versus T Cell Replete

Wagner et al.,

2005

[57]

11 Prospective

Randomized

1995-2000

Multicenter (15)

Total n 5 405

AML n 5 101 (25%)

Total AML n 5 101

(ITT)

T cell depleted 48

T cell replete 53

Not stated

(31.2 yrs)

Not stated 4.2 yrs (3-yr DF

29

33

(P 5 .79

Hale et al.,

1998

[58]

211 Retrospective

Nonrandomized

Case-controlled,

1984-1995

Multicenter (3)

Total n 5 579

(1) in vitro and in vivo

T cell depleted 70

(2) in vitro T cell

depleted control 50

(3) IBMTR T cell

replete control 459

#56 yrs

(36 yrs)

(30 yrs)

(32 yrs)

(5-yr)

15

58

26

44 mos

116 mos

58 mos

(5-yr LF

60 (CI 47-

33 (CI 21-

52 (CI 47-

(P 1 versus 2

P 1 versus 3

Marmont et al.,

1991

[59]

21 Retrospective

IBMTR

Nonrandomized

1982-1987

Multicenter (137)

Total n 5 3211

AML n 5 1154 (62%)

AML CR1 n 5 717

Total AML CR1 n 5 717

T cell depleted 159

T cell replete 560

#56 yrs

(30 yrs)

(25 yrs)

Not stated Not

stated

(2-yr LF

45

57

(P \.02

PBSCT versus BMT

Garderet et al.,

2003

[60]


Registry

Nonrandomized Case-

controlled 1994-1999

Total n 5 213

AML n 5 111 (52%)

Total AML n 5 111

BM 73

PBSC 38

Not stated

92% adult

.16 yrs

(2-yr)

40

38

18 mos

25 mos

(2-yr LF

42 ± 6

33 ± 8

(P 5 .49

Definition of

DFS/

EFS/

RFS/

LFS

(Interval)

% OS

(Signif)


65 ± 2

65 ± 3

(P 5 .94)

Time from BMT

to relapse

or death

(5-yr OS)

22 ± 4

)

Time from SCT

to relapse

or death

(2-yr OS)

81 (CI 67-95)

21 (CI 7-36)

(P \.001)


41

)



39

FS) Not stated Not stated


59 ± 2

)

)

)

)

Onset not stated;

events were

relapse or

death

(5-yr OS)

56 (CI 49-63)

60 (CI 51-67)

60 (CI 46-71)

(P not signif.)

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T163


Reference No.

Quality

and

Strength

of


Number of Patients

by Study Group

Upper Limit

(Median)

Age at

Diagnosis

(Interval)

% TX -

Related

Mortality

Median

Follow-

up

(Interval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif)

Ringd�en et al.,

2002

[61]

21 Retrospective

EBMT Registry

Nonrandomized

Multicenter (224)

Total n 5 3465

AML n 5 2294 (44%)

AML CR1 n 5 1532

Total AML CR1

n 5 1532

BM 1072

PBSC 460

#65 yrs

(Mean)

(36 yrs)

(38 yrs)

(2-yr)

23

24

(Mean)

29 mos

15 mos

(2-yr LFS)

61 ± 2

57 ± 3

(P 5 .46)

Single Arm Cohort Studies—Related and Unrelated Allo-SCT

Michallet et al.,

2000

[62]

211 Retrospective SFGM

Registry 1980-1993

Advanced AML

Total n 5 397 #77 yrs

(28 yrs)

86% adult

.15 yrs

(5-yr)

45 ± 6

7.5 yrs (5-yr DFS)

20 ± 4

Stelljes et al.,

2005

[63]

211 Prospective

1999-2004

Multicenter (4)

RIC

Total n 5 71

In CR 36

Non-CR 35

#66 yrs

(51 yrs)

(2-yr)

8

37

25.9 mos (2-yr RFS)

78 (CI 63-93

16 (CI 3-29)

(P \.001)

Greinix et al.,

2002

[64]

21 Retrospective

1982-2000

Multicenter (4)


(35 yrs)

(Not stated) 5.6 yrs (5-yr LFS)

40

Sierra et al.,

2000

[65]

21 Retrospective

1985-1998

Single center


(30 yrs)

84% adult

$18 yrs

(5-yr)

43

2.9 yrs (5-yr LFS)

14 ± 2 (adults

Cook et al.,

2006

[66]

21 Retrospective

1989-2003

Multicenter (15)

CR1 after refractory

to first induction

Total n 5 68 \60 yrs

(36 yrs)

(Not stated)

38

. 4 yrs (4-yr DFS)

34

Bunjes et al.,

2002

[67]

21 Retrospective

Dates not stated

Single center

High-risk AML


(45 ys)

(Not stated)

25

26 mos (Not stated, D

47

Single Arm Cohort Studies—Related Allogeneic SCT

Rocha et al.,

2002

[68]


Registry 1992-1999

Multicenter (4501)


(35 yrs)

(5-yr)

24 ± 4


57 ± 2

Tallman et al.,

2000 [69]

21 Retrospective IBMTR

Nonrandomized

1989-1995

Multicenter (108)

Total n 5 431

Standard-dose Cy 222

High-dose Cy 147

No remission Tx 62

#56 yrs

(29 yrs)

(28 yrs)

(32 yrs)

(5-yr)

22

24

30

61 mos (5-yr LFS)

56 (CI 49-63

59 (CI 50-66

50 (CI 36-63

(P not signif.

nterval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif)

Definition of

DFS/

EFS/

RFS/

LFS

(Interval)

% OS

(Signif)

-yr RFS)

23

Time from BMT

to relapse

or death

Not stated

-yr EFS)

51 ± 5


53 ± 5

-yr DFS)

(CI 21-43)

Time from BMT

to relapse

or death

Not stated

0-yr LFS)

60

10

ot stated)


ot stated Not stated (5-yr OS)

71


26 (CI 14-37)

ot stated Not stated (Not stated, OS)

50

40

(P not stated)

T, European Group for Blood and Marrow Transplan-lapse-free survival; Signif., significance (probability); OS,CI, 95% confidence interval.

164

D.M

.Olian

sky

etal.


Reference No.

Quality

and

Strength

of


Number of Patients

by Study Group

Upper Limit

(Median)

Age at

Diagnosis

(Interval)

% TX -

Related

Mortality

Median

Follow-

up

(I

Clift et al.,

1992

[70]

21 Retrospective

1974-1990

Multicenter (3)

Untreated first relapse


(.30 yrs)

75% adult

.18 yrs

(5-yr)

44

Not stated (5

Mehta et al.,

2002

[71]

21 Retrospective 1978-

1986

Single center


(32 yrs)

(5-yr)

31 ± 4

9 yrs (5

Brown et al.,

1996 [72]

21 Retrospective

NAMTG

1984-1992

Multicenter

Resistant AML (40 in

untreated first relapse)


(33 yrs)

(3-yr)

37 (CI 23-51)

Not stated (5

32

Greinix et al.,

1996

[73]

21 Retrospective

1982-1995

Single center

Total n 5 62

CR1 35

. CR1 27

#51 yrs

(32 yrs)

(10 yrs)

11

39

50 mos (1

(P n

Zikos et al.,

1998 [74]

21 Prospective

1990-1994

Single center


(29 yrs)

(5-yr)

14

4.4 yrs N

Bibawi et al.,

2001

[75]

21 Retrospective

1991-1998

Single center

Total n 5 62

AML n 5 50 (81%)

Advanced AML


(37 yrs)

(100 d)

19

3 yrs N

Frassoni et al.,

1991

[76]

22 Retrospective

1981-1989

Single center

Total n 5 168

AML n 5 118 (63%)

Total n 5 118

CR1 88

Advanced Disease 30

Not stated Not stated 767 days N

*Quality and strength of evidence definitions are listed in Table 1.†AML indicates acute myelogenous leukemia; BMT, bone marrow transplantation; SCT, stem cell transplantation; CR, complete remission; EBMtation; IBMTR, International Bone Marrow Transplant Registry; ITT, intent-to-treat; DFS, disease-free survival; EFS, event-free survival; RFS, reoverall survival; ctrl grp, control group; NAMTG, North American Marrow Transplant Group; SFGM, Soci�et�e Francaise de Greffe de Moelle;


HSCT if they were .55 years old. Younger patientswere included if they had comorbid conditions thatexcluded them from standard conditioning for allo-HSCT. Eighteen patients had secondary AML, 51were in CR1, 39 were in CR2, and 32 had moreadvanced AML. Patients received either a related(n 5 58, 47.5%) or unrelated (n 5 64, 52.5%) donorallo-HSCT. Median time from last chemotherapy toSCT was 76 days, with no difference between the 2groups. Patient characteristics were comparable be-tween the 2 groups except more related than unrelateddonor allo-recipients were in CR1 (55% versus 30%),and more unrelated than related recipients had under-gone a previous auto-HSCT (23% versus 3%).

Sayer et al. [56] reported the results of a multicen-ter, retrospective analysis of 113 adult (16-67 years) pa-tients with AML who received either a matched relateddonor (n 5 63) or matched unrelated donor (n 5 50)allo-SCT following RIC with fludarabine and busulfan(Flu 1 Bu, n 5 93) or TBI (n 5 20). Stem cell sourcewas PBSC (n 5 102), BM (n 5 10), or both (n 5 1).Differences in patient characteristics between the 2groups were not reported.

T Cell Depleted (TCD) versus T Cell Replete

Wagner et al. [57] presented the outcomes of 103adult (77% .18 years) patients with AML randomizedto undergo an unrelated allo-BMT with either TCDmarrow and cyclosporine (n 5 48) or a T cell-repleteBMT with methotrexate (MJX) and cyclosporine(n 5 53). Two methods of TCD were used: counter-flow centrifugal elutriation, a physical method ofseparating T cells; and T10B9, an antibody. Althoughprognostic factors were similar between the 2 random-ized groups, there were differences by age, race, andtotal nucleated cell dose among centers differentiatedby the 2 methods of TCD.

Hale et al. [58] presented the results of a retrospec-tive, case-controlled study of adult (14-56 years) pa-tients with AML in CR1, comparing TCD versus Tcell-replete marrow in HLA-matched sibling allo-BMT. CAMPATH-1M (CAM-IgM) was used for invitro depletion of the graft and CAMPATH-1G(CAM-IgG) for in vivo depletion of the recipient priorto graft infusion (n 5 70). The results were comparedwith 2 control groups: (1) a historic group of 50 pa-tients who received CAM-IgM depleted BM, but noin vivo CAM-IgG; and (2) a matched group of concur-rently treated patients (n 5 459) who received T cell-replete allo-BMT and were reported to the IBMTR.No differences in patient characteristics among thegroups were reported.

Marmont et al. [59] analyzed the effects of TCDon the outcome of HLA-matched sibling donor BMTin a multicenter, retrospective study of 1154 adult(#56 years) patients with AML from 137 centers that

reported to the IBMTR. A total of 245 AML patientsreceived a TCD allo-BMT. A variety of TCD tech-niques were used by the centers, including physical,broad antibody, anti-T and NK, and anti-T. Duringthe same time period, 909 control group patients re-ceived a non-TCD allo-BMT. No differences in patientcharacteristics between the 2 groups were reported.

PBSCT versus BMT

Garderet et al. [60] presented the results of a retro-spective, case-controlled study of 111 adult (92% .16years) patients with AML reported to the EBMT,comparing PBSC versus BM as the stem cell source.The data from 38 AML patients who received G-CSF mobilized PBSC from a matched unrelated donorwere compared with a historic control group of 73matched unrelated allo-BMT patients computer-matched on disease status at transplant, age, year oftransplant, and whether or not the graft was TCD.

Ringd�en et al. [61] reported the results of a retro-spective comparison of BM (n 5 1537) versus PBSC(n 5 757) as the stem cell source in 2294 adult(16-65 years) patients with AML who underwent anHLA-matched sibling HSCT and were reported tothe EBMT Registry. Patient characteristics betweenthe 2 groups were comparable, except patients receiv-ing PBSC were older, had older donors, received moreTCD grafts, and received methotrexate (MTX) andTBI less frequently than patients receiving BM.

Single-Arm Cohort Studies

The remaining 15 studies, presented only in Table8, examined the outcomes of single arm, noncompara-tive studies of related and unrelated allo-HSCT (n 5 6)and related allo-HSCT only (n 5 9). No single-armunrelated allo-HSCT studies met the inclusion criteriafor this review.

COMPARISON OF TREATMENT REGIMENS

Although the term ‘‘conditioning regimen’’ is oftenapplied to the treatment immediately prior to allo- orauto-HSCT, in this review it is used only in relationto allo-HSCT, as conditioning regimen implies treat-ment that includes facilitation of engraftment of alloge-neic cells. In this review, the term ‘‘high-dose therapyregimen’’ is used to refer to the treatment immediatelyprior to auto-HSCT. This section provides descrip-tions of 14 studies comparing treatment regimens andtheir impact on HSCT outcomes, which are presentedin Table 9. These studies included high-dose auto-HSCT (n 5 2), myeloablative (n 5 8), or RIC (n 5 4)allo-SCT (n 5 12), and 1 study that did not differenti-ate between auto-and allo-SCT (presented in textonly). Within each subsection, the highest quality stud-ies are presented first; studies of equal quality are pre-sented in descending order by sample size.

nterval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif)

Definition of

DFS/

EFS/

RFS/

LFS

(Interval)

% OS

(Signif)

-yr LFS)

52 ± 3

51 ± 3

ot signif.)


-yr DFS)

42

37

20

ot stated)


37

34

20

(P not stated)

-yr DFS)

49

73

P 5 .01)


51

75

(P 5 .02)

-yr EFS)

59

56

P 5 .70)


for AML only


27.3

(CI 18.9-35.6)

30.1

(CI 20.2-39.1)

(P 5 .48)

-yr DFS)

(CI 44-67)

(CI 31-56)

P 5 .07)

Time from BMT

to relapse

or death

(5-yr OS)

64 (CI 53-72)

45 (CI 33-58)

(P 5 .04)

-yr LFS)

(CI 48-59)

(CI 50-65)

P 5 .44)

Onset not stated;

events were

relapse or

death

(5-yr OS)

55 (CI 49-60)

60 (CI 52-66)

(P 5 .58)

166

D.M

.Olian

sky

etal.

Table 9. Studies of SCT Outcome by Treatment Regimen

Reference No.

Quality and

Strength of



Study Group

Upper Limit

(median)

Age at Diagnosis

(Interval)

% TX -Related

Mortality

Median

Follow-up

(I

Autologous SCT—Comparison of 2 or More High-Dose Therapy Regimens

Ringd�en et al.,

1996

[77]

21 Retrospective

EBMT Registry

Nonrandomized

1987-1994

Multicenter

Total n 5 660

Bu 1 Cy 330

Cy 1 TBI 330

.40 yrs

(Not stated)

(2-yr)

12 ± 2

15 ± 2

Not stated (2

(P n

Ball et al.,

2000

[78]

21 Retrospective

Nonrandomized

1984-1997

Multicenter (13)

Total n 5 138

Bu 1 Cy 93

Cy 1 TBI 35

Bu 1 VP-16 10

\65 yrs

(40 yrs)

Not stated Not stated (2

(P n

Allogeneic SCT—Comparison of 2 or More Myeloablative Conditioning Regimens

Blaise et al.,

1992

[79]

11 Prospective

GEGMO

Randomized

1987-1990

Multicenter (15)

Total n 5 101

Bu 1 Cy 51

Cy 1 TBI 50

Not stated

(31 yrs)

(32 yrs)

(2-yr)

27

8

(mean)

23 mos

(2

(

Ringd�en et al.,

1999

[80]

11 Retrospective

Randomized

1988-1992

Multicenter (6)

AML n 5 69 (41%)

Total n 5 69

Bu 1 Cy 37

Cy 1 TBI 32

#68 yrs

(35 yrs)

(34 yrs)

(7-yr)

(Overall)

34

14

Not stated (2

(

Farag et al.,

2005

[81]

211 Retrospective

Nonrandomized

1984-2001

2 centers

Total n 5 237

Bu 1 Cy 1 VP-16 127

Bu 1 Cy 110

#62 yrs

(40 yrs)

(35 yrs)

(Day 1100)

26.8

21.8

4.3 yrs

10.9 yrs

N

Resbeut et al.,

1995

[82]

211 Prospective

SFGM

Nonrandomized

1983-1990

Multicenter (17)

Total n 5 171 (ITT)

Cy 1 Fractionated

TBI 106

Cy 1 Single dose

TBI 65

(Not stated)

(29 yrs)

(25.7 yrs)

24

38

63 mos (5

56

43

(

Litzow et al.,

2002

[83]

21 Retrospective

Nonrandomized

IBMTR

1988-1996

Multicenter (350)

Total n 5 581

Bu 1 Cy 381

Cy 1 TBI 200

#57 yrs

(35 yrs)

(34 yrs)

(5-yr)

27

30

55 mos

45 mos

(5

54

58

(

(Interval)

% DFS/

% EFS/

% RFS/

% LFS

(Signif)

Definition of

DFS/

EFS/

RFS/

LFS

(Interval)

% OS

(Signif)

2-yr LFS)

64 ± 3

66 ± 3

not signif.)



73

41

(P \.0009)

5-yr LFS)

(CI 10-58)

(CI 34-68)

(P 5 .053)

Onset not stated;

events were

relapse

or death

(5-yr OS)

38 (CI 15-61)

51 (CI 36-66)

(P 5 .149)

2-yr DFS)

(CI 29-60)

(CI 24-62)

(CI 25-72)

not signif)

Time from BMT

to relapse

or death

(2-yr OS)

50 (CI 34-66)

47 (CI 30-65)

49 (CI 24-74)

(P not signif)

2-yr LFS)

44 ± 3

40 ± 3

(P 5 .80)


46 ± 3

47 ± 3

(P 5 .43)

3-yr PFS)

32

19

not stated)


35

30

(P not signif.)

n; EBMT, European Group for Blood and Marrowvival; RFS, relapse-free survival; Signif., significance% confidence interval.

The

Role

ofCyto

toxic

Therap

yw

ithHSC

T167


Reference No.

Quality and

Strength of



Study Group

Upper Limit

(median)

Age at Diagnosis

(Interval)

% TX -Related

Mortality

Median

Follow-up

Ringd�en et al.,

1996

[77]

21 Retrospective

EBMT Registry

Nonrandomized

1987-1994

Multicenter

Total n 5 446

Bu 1 Cy 223

Cy 1 TBI 223

.40 yrs

(Not

stated)

(2-yr)

16 ± 3

19 ± 3

Not stated (

(P

Bacigalupo et al.,

2000

[84]

21 Retrospective

Non-randomized

1980-1998

Single center

Total n 5 150

Cy 1 TBI $ 9.9 Gy 116

Cy 1 TBI \ 9.9 Gy 34

#47 yrs

(28 yrs)

(5-yr)

18

26

2467 days N

Schapp et al.,

1997

[85]

21 Retrospective

Nonrandomized

1981-1995

Single center

Total n 5 181

AML n 5 70 (39%)

Total AML n 5 70

Cy 1 TBI 29

Cy 1 TBI 1 DNR or Ida

41

#59 yrs

(Overall)

(32 yrs)

(36 yrs)

(Not

stated)

26

22

63 mos (

34

51

Allogeneic SCT—Reduced-Intensity Conditioning (RIC) Regimens

Shimoni et al.,

2006

[86]

211 Prospective

Nonrandomized

No dates given

(5 years)

Single center

AML/MDS (85%

AML)

Total n 5 112 (ITT)

Cy 1 Bu 45

RIC Flu 1 Bu2 41

Modified ablative

Flu 1 Bu4 26

#70 yrs

(50 yrs)

(57 yrs)

(51 yrs)

(2-yr)

10

3

2

22 mos (

45

43

49

(P

Aoudjhane et al.,

2005

[87]

21 Retrospective

EBMT Registry

Nonrandomized

1997-2003

Multicenter (182)

Total n 5 722

Myeloablative 407

RIC 315

$50 yrs

(54 yrs)

(57 yrs)

(2-yr)

32

18

13 mos

14 mos

(

de Lima

et al., 2004

[88]

21 Retrospective

Nonrandomized

1995-2000

Single center

AML/MDS (72%

AML)

Total n 5 94

Flu 1 Mel 62

Flu 1 Cy 32

#75 yrs

(54 yrs)

(61 yrs)

(3-yr)

39

16

40 mos (

(P

*Quality and strength of evidence definitions are listed in Table 1.†AML indicates acute myelogenous (myeloid) leukemia; BMT, bone marrow transplantation; SCT, stem cell transplantation; CR, complete remissioTransplantation; IBMTR, International Blood and Marrow Transplant Research; ITT, intent-to-treat; DFS, disease-free survival; EFS, event-free sur(probability); OS, overall survival; SFGM, Soci�et�e Francaise de Greffe de Moelle; GEGMO, Groupe d’Etude des Greffes de Moelle Osseuse; CI, 95


Autologous SCT—High-Dose Therapy Regimens

Ringd�en et al. [77] reported the outcomes of660 adult (79% .20 years) patients with AML inCR1 who underwent auto-SCT after treatment withBu 1 Cy (n 5 330) or Cy 1 TBI (n 5 330) and werereported to the EBMT Registry. Significantly morepatients treated with Bu 1 Cy were in first relapse,and patients in the Cy 1 TBI group were more likelyto be male and in CR2. Figure 6 compares the actuarialLFS between patients receiving Bu 1 Cy versus Cy 1

TBI prior to auto-SCT.Ball et al. [78] reported the results of a retrospective,

nonrandomized, multicenter study of 138 adult (#67years) patients with AML comparing 3 high dose ther-apy regimens followed by auto-BMT. The treatmentregimen prior to 1988 was Cy 1 fractionated TBI(n 5 35), post-1988 it was changed to Bu 1 Cy (n 5 93),and 10 patients at 1 center (post-1988) were treated withBu 1 VP-16. No differences in patient characteristicsamong the 3 groups were reported. In all, 110 patientswere in CR (CR1, n 5 23; CR2/3, n 5 87) and 28were in first relapse at the time of transplantation.

Allogeneic HSCT

Myeloablative conditioning regimens. Blaise et al. [79]presented the results of a prospective, randomizedstudy by the Groupe d’Etudes de la Greffe de MoelleOsseuse (GEGMO) of 101 adult (.14 years) patientswith AML in CR1 conditioned prior to an HLA-matched sibling donor allo-HSCT with either Bu 1

Cy (n 5 51) or Cy 1 TBI (n 5 50). Sixty three(62%) patients (Cy 1 TBI, n 5 31; Bu 1 Cy, n 5 32)were transplanted \120 days after diagnosis. Therewere no statistical differences in patient characteristicsbetween the 2 groups.

Figure 6. Actuarial LFS in autograft recipients with AML in CR1

treated with Bu/Cy (broken line) (n 5 330) or Cy/TBI (solid line)

(n 5 330). The survival curves are not significantly different (P 5

.64). (Reprinted with permission; [77].)

Ringd�enetal. [80]reportedtheresultsofa retrospec-tive analysis of 69 adult (#55 years) patients with AMLin CR1 (n 5 51) or in $CR2 or early relapse (n 5 18) ina randomized trial comparing Bu 1 Cy (n 5 37) versusCy 1 TBI (n 5 32) followed by an HLA-matchedsibling donor allo-BMT. There were no differences inpatient characteristics between the 2 groups.

Farag et al. [81] presented the outcomes of 237 adult(17-62 years) patients with AML in a retrospective,nonrandomized study comparing Bu 1 Cy 1 VP-16(n 5 127) versus Bu 1 Cy (n 5 110) followed byallo-HSCT from HLA-matched related or unrelateddonors. Patients who received Bu 1 Cy 1 VP-16were significantly more likely to be older, have high-risk disease, and to receive a matched-unrelated donorallo-HSCT than the patients in the Bu 1 Cy group.

Resbeut et al. [82] reported the results of a prospec-tive study by the Soci�et�e Francaise de Greffe de Moelle(SFGM) of 171 adult (median age 5 27 years) patientswith AML in CR1 who were nonrandomly assigned toCy 1 single dose TBI (10 Gy) (n 5 65) or Cy 1 frac-tionated TBI (range 10-15 Gy; 76% received 12 Gy in6 fractions over 3 days) (n 5 106) prior to undergoingan HLA-matched sibling allo-BMT. Patient charac-teristics were comparable between the 2 groups exceptfor a shorter diagnosis-to-BMT interval in the frac-tionated TBI group.

Litzow et al. [83] presented the outcomes of 581adult (20-57 years) patients with AML in CR1 in a ret-rospective, nonrandomized, IBMTR study comparingBu 1 Cy versus Cy 1 fractionated TBI (dose .11.25Gy) conditioning regimens followed by HLA-matchedsibling allo-HSCT. Significantly more patients in theBu 1 Cy group were .40 years, and a higher percent-age of Cy 1 TBI patients required more than 1 cycle ofchemotherapy to achieve CR1.

Ringd�en et al. [77] reported the outcomes of 446adult (79% .20 years) patients with AML in CR1who underwent allo-HSCT following conditioningwith Bu 1 Cy (n 5 223) or Cy 1 TBI (n 5 223) andwere reported to the EBMT Registry. Significantlymore patients treated with Bu 1 Cy were in first re-lapse, and patients in the Cy 1 TBI group weremore likely to be male and in CR2. Figure 7 comparesthe actuarial LFS between patients receiving Bu 1 Cyversus Cy 1 TBI prior to allo-HSCT.

Bacigalupo et al. [84] presented the results of a ret-rospective, nonrandomized study of 150 adult (#47years) patients with AML in CR1, comparing condi-tioning with Cy 1 low dose TBI (\9.9 Gy, n 5 34)versus Cy 1 high dose TBI ($9.9 Gy, n 5 116) priorto HLA-matched related donor allo-BMT. Therewere no significant differences in patient characteris-tics between the 2 groups.

Schapp et al. [85] reported the results of a retro-spective, nonrandomized study of 70 adult (13-59years) patients with AML in CR1 who underwent


conditioning with Cy 1 TBI (n 5 29) or Cy 1 TBI 1

DNR or Ida (n 5 41) followed by an HLA-matchedsibling donor allo-HSCT. No significant differencesin patient characteristics between the 2 groups were re-ported.

Reduced-intensity conditioning regimens. Shimoni et al.[86] presented the outcomes of 112 adult (17-70 years)patients with AML (n 5 95) or MDS (n 5 17) in CR1enrolled in a prospective, nonrandomized study com-paring standard myeloablative conditioning with Cy1 Bu (12.8 mg/kg) (n 5 45) versus RIC with Flu 1

Bu (6.4 mg/kg, FluBu2, n 5 41) or a modified myeloa-blative conditioning regimen with Flu 1 Bu (12.8mg/kg, FluBu4, n 5 26) followed by matched ormismatched, related or unrelated allo-SCT. Older pa-tients (.65 years) were more likely to receive FluBu2conditioning, patients with active disease were morelikely to receive FluBu4 conditioning.

Aoudjhane et al. [87] reported the results of a retro-spective, nonrandomized study of 722 adult (.50years) patients with AML reported to the EBMTRegistry who underwent an HLA-matched siblingallo-SCT, comparing RIC (n 5 315) with Flu 1

low-dose TBI (\3 Gy), Bu (#8 mg/kg), or other non-myeloablative drugs versus myeloablative condition-ing (n 5 407) with TBI (.10 Gy) or Bu (.8 mg/kg)1 other myeloablative drugs. RIC patients were olderand more likely to be male than patients who receiveda myeloablative (MA) conditioning regimen. Figure 8compares the unadjusted cumulative incidence ofLFS between the RIC and MA cohorts.

de Lima et al. [88] reported the outcomes of94 adult (22-75 years) patients with AML (n 5 68)or high-risk MDS (n 5 26) in a retrospective,nonrandomized study comparing a nonmyeloablativeregimen of Flu 1 Cy 1 Ida (FAI, n 5 32) versus a mye-losuppressive, RIC regimen of Flu 1 Mel (FM, n 5 62)prior to HLA-matched related or unrelated donor

Figure 7. ActuarialLFSin allograft recipientswithAMLinCR1treated

with Bu/Cy (n 5 223) or Cy/TBI (n 5 223). Survival curves are not

significantly different (P 5 .63). (Reprinted with permission; [77].)

allo-SCT. Compared to patients who received FM,those treated with FAI were significantly older andmore likely to be in CR1 at transplant.

Kroger et al. [89] reported the results of a prospec-tive, nonrandomized study of 90 adult (#64 years)patients with AML comparing Bu 1 Cy 1 low doseVP-16 (30 mg/kg) (n 5 60) versus Bu 1 Cy 1 highdose VP-16 (45 mg/kg) (n 5 30) followed by auto-orallo-HSCT (HSCT type not differentiated; outcomesare presented only in the text, not in Table 9). Stemcell sources were allo-related BM (n 5 53), allo-unre-lated BM (n 5 5), allo-unrelated PBSC (n 5 2), synge-neic BM (n 5 2), auto-purged BM (n 5 9),auto-unpurged BM (n 5 9), or auto-PBSC (n 5 10).Differences in patient characteristics between the 2conditioning regimens were not reported. Medianfollow-up was 16 months. Two-year treatment-relatedmortality (TRM) was 16% versus 26%, 3-year DFSwas 62% versus 40% (P 5 .03), and 3-year OS was63% versus 41% (P 5 .06) for patients receivinglow-versus high-dose VP-16, respectively.

TREATMENT FOR RELAPSE AFTER FIRST ALLO-HSCT

This section presents 2 studies of treatment forrelapse after an initial allo-HSCT, 1 using donorlymphocyte infusion (DLI) and the other a secondallo-HSCT. Although these studies may not meet allcriteria for inclusion, they were included in this reviewto illustrate the typical outcome of these treatments foradult AML patients. Study designs and outcomes arepresented in the text.

DLI

Collins et al. [90] reported the results of a retro-spective study of 46 adult (median age 33.5 years) pa-tients with AML who received a DLI to treat relapseafter a related (n 5 44) or unrelated (n 5 2)

Figure 8. Unadjusted cumulative incidence of LFS for patients over

50 years of age with AML receiving myeloablative (MA) versus re-

duced intensity conditioning (RIC) prior to HLA-identical sibling

HSCT. (Reprinted with permission; [87].)


allo-BMT. Six patients were not assessable for re-sponse to DLI because of prior chemotherapy. Ofthe remaining 39 patients, 6 (15.4%) achieved CR.Median follow-up of survivors after DLI was 506days. An OS value was not provided; however, a Ka-plan-Meier curve indicated a 3-year OS of approxi-mately 15%.

Second Allogeneic SCT

Eapen et al. [91] presented the outcomes for 125adult (overall, 66% .21 years) AML patients fromthe IBMTR who received a second allo-SCT forrecurrent or persistent leukemia after initial HLA-matched sibling SCT. Overall, 85% of the patientsreceived allografts from the same donor for both trans-plants. Median follow-up of survivors was 93 months.The 3-year LFS and OS were both 27% (95% CI,19%-35%).

AUTO-SCT AS TREATMENT OF THERAPY-RELATEDOR SECONDARY AML

The following 2 studies examined auto-SCT astreatment for secondary AML after MDS or therapy-related AML.

De Witte et al. [92] reported the outcomes of 60adult patients (median age 39 years), retrieved fromthe registries of the Chronic and Acute LeukemiaWorking Parties of the European Group for Bloodand Marrow Transplantation, who underwent an auto-BMT in CR1 for secondary (n 5 39) or therapy-relatedMDS/AML (n 5 21). The median interval betweendiagnosis and auto-BMT was 7 months. Median fol-low-up of survivors was 10 months. For secondaryAML and therapy-related MDS/AML patients, 2-yearOS from BMT was 34% and 41%, DFS was 30% and36%, and TRM was 5% and 10%, respectively.

Kroger et al. [93] reported the results of 65 adult(median age 39 years) patients with therapy-relatedMDS/AML who underwent an auto-SCT and were re-ported to the EBMT. Stem cell source was BM (n 5 31),PBSC (n 5 30), or a combination of both (n 5 4). Themedian time between diagnosis and transplant was 5months. Median follow-up was not reported. Two-year TRM was 12% (95% CI, 6%-38%) and 3-yearDFS and OS from BMT were 32% (95% CI,18%-45%) and 35% (95% CI, 21%-49%), respectively.

LATE EFFECTS AND QUALITY OF LIFE

Late Effects

The following studies investigated late effects inadult patients who underwent auto- or allo-SCT fortreatment of AML, providing additional insight intothe potential outcomes related to specific treatments.

Frassoni et al. [94] investigated the occurrence oflate events in patients with AML who underwentallo-HSCT (n 5 1059) or auto-BMT (n 5 656).

They found that the incidence of late relapse continu-ously decreased with time. Patients with no recurrenceat 2 years had an 82% chance of remaining in CR at 9years following transplantation. The latest relapseswere observed following allo-HSCT at 6.6 years forpatients transplanted in CR1, as opposed to 3.7 yearsfor patients transplanted in CR2, and followingauto-HSCT at 6 years and 5.1 years, respectively.Compared to auto-SCT, patients who underwentallo-HSCT experienced a lower frequency of laterelapse.

Abdallah et al. [95] analyzed the long-term out-comes and toxicities of 98 high-risk AML patients inCR1 or CR2 treated with TBI and high-dose Cy priorto auto-SCT purged with mafosfamide. Long-termcomplications (median observation period of 11.67years) included cataracts (44.4%), HCV infections(5%), cardiac complications (4%), MDS (4%), andrenal insufficiency (2%).

Wadleigh et al. [96] conducted a retrospectivestudy of 62 patients who underwent allo-HSCT forthe treatment of AML or AML arising from MDS todetermine whether gemtuzumab ozogamicin (GO),a monoclonal antibody (mAb) used in the treatmentof AML, increased the risk of veno-occlusive disease(VOD). Nine of 14 (64%) of patients who receivedGO prior to allo-SCT developed VOD comparedto 4 (8%) of 48 patients without prior GO exposure(P \ .0001).

Toor et al. [97] investigated life-threatening bleed-ing associated with platelet transfusion refractorinessamong 39 patients who underwent high-dose chemo-radiotherapy followed by sibling donor allo-BMT(n 5 12) or auto-BMT (n 5 27). Increased plateletrequirements in HLA alloimmunized auto-BMT pa-tients were observed at a median of 211 platelet trans-fusions versus 0 transfusions in nonalloimmunizedauto-BMT patients (P \ .01) and 17 in allo-BMTpatients. Five of 6 HLA alloimmunized auto-BMT pa-tients experienced delayed bleeding, which contrib-uted to their death while still in CR.

Butt and Clark [98] studied the iron status of 32adult AML patients enrolled in the MRC AML 10and 12 trials and found that patients who underwentauto-SCT had a higher median first serum ferritinlevel (3245 mg/L) than patients who received chemo-therapy alone (1148 mg/L) or allo-SCT (1334 mg/L)because of increased use of transfused blood. Nine ofthe 10 auto-SCT recipients underwent venisection,but no patient suffered end organ damage.

Quality of Life (QOL)

The following studies investigated the QOL re-ported by or observed in patients who underwentallo- or auto-SCT for the treatment of AML.


Watson et al. [99] surveyed 481 patients 1 yearfrom the end of treatment in the MRC AML 10 trialto compare QOL following allo-BMT (n 5 97),auto-BMT (n 5 74), or chemotherapy (n 5 310). Onthe EORTC Quality of Life-Core 30 Questionnaire(QLQ-C30) with 5 functional scales (physical, role,cognitive, emotional, and social) and a leukemia-spe-cific measure (QLQ-LEU) designed to assess late ef-fects of BMT, allo-BMT was found to havea significantly adverse impact on most QOL dimen-sions compared to auto-BMT or chemotherapy.

Zittoun et al. [100] studied the self-reported QOLof 98 patients in continued CR for 1-7.4 years afterundergoing allo-BMT, auto-BMT, or ICC in theEORTC-GIMEMA AML 84 trial. On all parameters,including somatic symptoms (mouth sores, cough, hairloss, headache), repeated acute medical problems,physical functioning, role functioning, leisure activi-ties, sexual functioning, overall physical condition,and overall QOL, there were significant differencesamong the 3 groups with a consistent ranking ofallo-BMT lower than auto-BMT and auto-BMTlower than chemotherapy.

Hsu et al. [101] used the EORTC QLQ-C30, aswell as the World Health Organization QOL ques-tionnaire (WHOQOL-BREF), to examine differencesin QOL between 41 AML patients who underwentallo-BMT as consolidation or salvage therapy versus63 patients who received traditional chemotherapyalone. The mean scores on the 2 QOL questionnairesdid not differ significantly between the 2 groups; how-ever, when comparing survival-weighted psychometricscores (SWPS), patients who underwent allo-BMThad significantly (P \ .01) higher SWPS on all func-tioning domains and symptom items of the QLQ-C30 and all 4 domains (physical, psychologic, social,and environmental) of the WHOQOL-BREF thanthose who received chemotherapy only.

Wellisch et al. [102] found no significant differ-ences in QOL between AML patients who underwentBMT (n 5 11) or chemotherapy only (n 5 27), with re-gard to occurrence of depressive symptoms, multifocalpsychiatric symptomatology, or on any subscale of theCARES Questionnaire, a 139-item inventory designedto evaluate the problems and rehabilitation needs ofpatients with cancer.

PROGNOSTIC FACTORS

Table 10 presents a summary of patient and diseaseprognostic factors and their reported effect on survivaloutcomes as determined by multivariate analysis inadult AML studies. Atlhough many of the referencedstudies in Table 10 were described in the text andtables of this review, others were specifically prognos-tic factor studies [references in brackets] that are notpresented in this evidence-based review because they

did not meet the inclusion criteria. Studies were notincluded in this table if the study did not conducta multivariate analysis and/or if transplantation wasnot an included treatment modality. The data in thistable are provided for the reader’s information andwere not used to make treatment recommendations.

FUTURE DIRECTIONS

Areas of Needed Research

After reviewing the evidence, the panel recommen-ded that large clinical trials be conducted to investigatethe following important areas of needed research inadult onset AML: (1) What is the role of HSCT intreating patients with specific molecular markers (eg,FLT3, NPM1, CEBPA, BAALC, MLL, NRAS, etc.)especially in patients with normal cytogenetics? (2)What is the benefit of using SCT to treat differentcytogenetic subgroups? (3) What is the impact on sur-vival outcomes of reduced intensity or nonmyeloabla-tive versus conventional conditioning in older (.60years) and intermediate (40-60 years) aged adults? (4)What is the impact on survival outcomes of unrelateddonor SCT vesus chemotherapy in younger (\40years) adults with high risk disease?

Ongoing Studies

Several studies are summarized below that addressareas of needed research or other critical issues thatmay affect the treatment recommendations made inTable 3. These studies are currently accruing patientsand/or have been published in abstract form.

Effect of specific molecular markers on outcome in AML pa-tients with normal cytogenetics. Schlenk et al. [142] evalu-ated the prognostic value of NPM1, FLT3, CEBPA,MLL, and NRAS gene mutations on RFS and OS fol-lowing allo-SCT in adult (16-60 years) AML patientswith normal cytogenetics. Between 1993 and 2004,872 patients with normal cytogenetics were enteredon 4 AMLSG treatment trials [AML-2/95, AML-1/99, AML HD93, AML HD98A]. In all 4 trials therewas a biologic allocation to an MRD allo-SCT inCR1. The median age of patients was 48 years; medianfollow-up time was 49 months. Of 666 patients achiev-ing a CR after induction therapy, 171 had a MRD and143 (84%) of these received an allo-HSCT in CR1.Subgroup analyses were performed on an ITT basisby mutation status (NPM11/FLT3-internal tandemduplication [ITD] versus all other combinations),which revealed a significant improvement in RFS(HR 5 0.56, 95% CI 0.39-0.81) and OS (HR 5 0.69(95% CI 0.48-0.98) in the subgroup of patients with-out the NPM11/FLT3-ITD marker constellationwho received an allo-SCT in CR1.

Dose intensity prior to auto-SCT. The EORTC andGIMEMA Leukemia Cooperative Groups havesponsored a phase III randomized, international,

Studies

References Reporting No Significant Impact of

Prognostic Factor on Survival Outcome

,15,24,35,39,41,46,49-51,63,64,74,75,[117-119],[129]

,19,24,32,40,41,45,49,58,[106],[113],[116],[117],[125],[129]

,55,63

,40,45,46,[117],[119],[125],[129]

,15,24,26,35,40,41,49,50,[106],[116]

,44,75

,19,26,39,[117]

,40

17]

,41

,40,41,46,51,[106]

,49,56,74,75,[124]

17]

, 58, 59,[106]

ality.

172

D.M

.Olian

sky

etal.

Table 10. Significance of Patient and Disease-Based Prognostic Factors on Survival* Outcomes as Determined by Multivariate Analysis in Individual

Prognostic Factor†

References Reporting Significant Impact of

Prognostic Factor on Survival Outcome

Patient Characteristics

Age 19,20,27,28,32,37,40,45,47,55,57,58,62,68,

69,71,[103],‡[104-116],[125],[127],[128],[131-135],[137],[139]

10-12

Sex 15, 69,[104],[118] 11,12

Disease Characteristics at Diagnosis

AML type (de novo or secondary) [114],[126],[134],[135] 15,26

WBC count 10,11,19,20,30,49,50,[104],[112],[114],[116],[118],[120],

[125],[127],[130],[133],[134],[137],[139]

12,39

FAB subtype 28-30,37,61,68,[105],[113],[114] 11,12

Performance status 56,57,69,[109],[116],[118],[121] 11,40

Cytogenetic risk group/Karyotype 11,15,18,19,26,29,30,39,45,49,50,55,57,66, 68,[104],

[105],[107],[110],[112-114],[116],

[117],[122-124],[129],[131],[132],[134],[135],

[137],[138],[141]

10,40

Molecular markers (eg, FLT3,§ BAALC, CEBPA,

NPM1, ERG, etc.)

[107],[125-136]

Flow cytometric cell surface expression (eg, CD2,

CD34, CD36, CD56, etc.)

[131],[137-141]

CMV status 29,57 63

Hepatomegaly/Splenomegaly 15 [116]

Extramedullary disease 40

Disease Characteristics at SCT

No. cycles to achieve CR1 10,45,50,66 11,12

Time to CR1 24,35

Duration of CR1 27,40,51,65

Time from diagnosis to CR1 37,[106] 41,46

Time from diagnosis to SCT 28,66,[108] 74,[1

Time from CR1 to stem cell harvest 12,40

Time from CR1 to SCT 37,[105],[115] 12,19

Disease status at SCT 29,55,60-65,72,[109],[114] 32,41

% BM blasts at SCT 56,[104],[109],[121] 15,[1

Year of Transplant 46,49,64,69,[113] 40, 41

*Survival 5 any one or more of the following: OS, DFS, RFS, LFS, EFS, PFS (see other tables for explanation of abbreviations).†Factors were not included if the study did not conduct a multivariate analysis and/or if transplantation was not an included treatment mod‡Bracketed [ ] references indicate prognostic factor studies that are not described in the text of the evidence-based review.§Only FLT3-ITD has been shown to have prognostic significance, other FLT3 mutations have unknown significance.


multicenter, prospective study (Protocol ID:EORTC-1992.00) of adult (15-60 years) patientswith AML, comparing high-dose versus standarddose induction therapy (Ara-C 1 DNR 1 VP-16), fol-lowed by consolidation and auto-PBSCT or BMTwith or without interleukin-2.

Dose intensity and in vivo purged vs unpurged auto-PBSCT.The ECOG has sponsored a phase III randomized,multicenter, prospective trial (Protocol ID: ECOG-1900) in adult (16-60 years) patients with AML, com-paring survival outcomes after high or standard doseDNR 1 Ara-C followed by in vivo purged (with gem-tuzumab ozogamicin) versus unpurged auto-PBSCT.Patients with appropriate donors are biologicallyassigned to allo-SCT.

Comparison of allo-SCT conditioning regimens. TheFred Hutchinson Cancer Research Center (FHCRC)has sponsored a phase III, randomized, multicenter, pro-spective study (Protocol ID: FHCRC-1992.00) of adult(#65 years) patients with AML or MDS, comparingsurvival outcomes after myeloablative (Flu 1 Bu orCy 1 Bu) versus nonmyeloablative (Flu 1 TBI) condi-tioning prior to allo-PBSCT. Patients are stratified bytransplant center (FHCRC versus other), diagnosis(AML versus MDS), donor (MRD versus URD), andetiology (primary versus treatment related).

The Cooperative German Transplant StudyGroup has sponsored 2 single center, phase III, rando-mized, prospective studies comparing nonmyeloabla-tive (Flu 1 TBI, 8 Gy) versus myeloablative (Cy 1

TBI, 12 Gy) conditioning prior to allo-SCT in adult(18-60 years) AML patients in CR1 (Protocol ID:9005-2005) or in CR2 (Protocol ID: AML_CR2_Allo_HSCT).

The Hopital Saint Antoine in France has spon-sored a multicenter, phase III, randomized, prospec-tive study (Protocol ID: P040420) of adult (35-55years) patients with AML, comparing myeloablative(Cy 1 TBI) versus nonmyeloablative (Flu 1 TBI,2Gy) conditioning prior to allo-SCT.

The Technische Universitaet Dresden, Germany,has sponsored a single center, phase III, randomized,prospective study (AML2003, Protocol ID: MK1-95)investigating standard versus intensified therapy foradult (16-60 years) AML patients prior to allo-HSCT. In the first 8 months of the trial, 107 AML pa-tients with a median age of 48 years (17-60 years) wererecruited, 57 of whom were randomized to the inten-sive treatment arm [143].

RIC allo-HSCT versus chemotherapy. The CooperativeGerman Transplant Study Group (Protocol ID:TRALG1/02) has sponsored a multicenter, prospec-tive study of older adult (51-70 years) AML patientsin a donor versus no donor design comparing allo-HSCT with RIC versus traditional chemotherapy.

G-CSF-mobilized PBSC versus BM allo-HSCT. TheNational Marrow Donor Program (NMDP) and the

National Institutes of Health (NIH) have sponsoreda phase III randomized, multicenter, prospective trial(Protocol ID: BMTCTN-0201) of adult (#66 years)patients with hematologic malignancies (AML, ALL,CML, MDS, myeloproliferative diseases, and therapy-related AML or MDS), comparing survival outcomesafter granulocyte colony stimulating factor (G-CSF)-mobilized PBSCT versus BMT from HLA-compatibleunrelated donors.

Nonmyeloablative allo-HSCT in older patients. TheCALGB has sponsored a phase II trial (Protocol ID:CALGB-100103) studying the efficacy of a RIC regi-men of Flu 1 Bu followed by allo-HSCT to treat olderadult (60-74 years) patients with AML in CR1.

STRENGTHS/LIMITATIONS AND DISCUSSION

The strengths of this systematic evidence-based re-view are the details conveyed in the text about eachstudy’s design, the presentation of outcomes in sum-mary tables for each major section, and the treatmentrecommendations made by the adult AML expertpanel. A limitation of this systematic evidence-basedreview is the inclusion of only published data, specifi-cally peer-reviewed articles published since 1990. Un-published data can represent ‘‘negative’’ findings thatcould lead to publication bias; however, the inclusionof high-quality, peer-reviewed publicly available datawas of paramount importance. Also excluded weredata published in abstract form because abstracts donot adequately convey the full details of the study de-sign or patient characteristics to meet evidence-basedcriteria for inclusion in systematic reviews, nor formaking a true assessment of the widespread applicabil-ity or impact of the treatment outside the scope of thetrial.

A limitation of the studies included in this review isthe inability to provide level ‘‘1’’ evidence for allo-HSCT trials because of the low rate of patientsallocated to the allo-HSCT arm who would actuallyreceive the assigned treatment (approximately 35%of patients have a matched-related donor [144]).Therefore, trials that biologically allocate patients toallo-HSCT based on donor availability have level‘‘2’’ as their highest evidence grade.

Other study-specific limitations that affect thequality of this systematic evidence-based review in-clude the variability in reporting patient characteristicspre-HSCT, changing treatment modalities over time,and the paucity of RCT data on sufficiently largepatient populations. The success of most therapies isaffected by cytogenetic risk, which is either not re-ported, not collected, or missing on too many patients.Chemotherapy regimens, HLA typing techniques,pre-HSCT treatment regimens, stem cell sources,and post-HSCT supportive care have changed consid-erably over the 17 years of trials included in this review.


The clinical research process is lengthy, making thedata from many of these studies outmoded at thetime of publication. RCT data were lacking in manyareas of this review, leading to several treatment rec-ommendations based on small prospective studiesand/or large retrospective registry reports.

To address some of these limitations, the authorsrecommend methodology standardization, includinguse of consistent study designs, endpoint definitions,and reporting of study results. Multicenter random-ized phase III comparative trials with large enroll-ments and high statistical power are required toadvance the field more constructively than single insti-tution phase II trials with 1 treatment arm, or retro-spective multicenter or registry studies. Much oftoday’s therapies for cancer result from the random-ized clinical trial process. It is currently estimatedthat \5% of adult cancer patients who are eligible toparticipate in clinical trials actually enroll in a trial.The authors acknowledge the importance of removingbarriers to participation in clinical trials, which mayinclude patients’ reluctance to be randomized, lackof patient access to clinical trials (eg, geographic,transportation, cultural), financial restraints (no or

incomplete insurance coverage for trial expenses),stringent trial eligibility criteria, and reluctance ofcommunity physicians to refer patients for clinical trialparticipation.

ACKNOWLEDGEMENTSThe American Society for Blood and Marrow

Transplantation and Drs. Hahn and McCarthy are in-debted to the members of the AML Review ExpertPanel and the ASBMT Evidence-Based Review Steer-ing Committee, who voluntarily and enthusiasticallyparticipated in this endeavor. The authors of this re-view and the American Society for Blood and MarrowTransplantation thank the National Marrow DonorProgram, and especially Jeffrey Chell, MD andMichael Boo, for their support of this project. Theauthors acknowledge Dr. C. Fred LeMaistre for pio-neering and supporting this effort, Alan Leahigh andDianne O’Rourke for their invaluable administrativeassistance, and Amy Gillen for serving as the literatureabstractor and additional reviewer of this manuscript.The authors are grateful to all the patients who partic-ipated in the clinical trials which led to the evidenceupon which this review is based.

APPENDIX A: GLOSSARY OF TERMS

4-HC 4-hydroperoxcyclophosphamideABMTR Autologous Blood and Marrow Trans-

plant RegistryAllo AllogeneicALP Acute leukemia protocolAML acute myelogenous leukemiaAMSA AmsacrineAPL acute promyelocytic leukemiaAra-C CytarabineAuto autologousBM bone marrowBMT bone marrow transplantationBu BusulfanCALGB Cancer and Leukemia Group BCI (95%) confidence intervalCIBMTR Center for International Blood and

Marrow ResearchCMV cytomegalovirusCNS central nervous systemCR complete remissionCR1 first complete remissionCR2 second complete remissionCy cyclophosphamideDFS disease-free survivalDLI donor lymphocyte infusionDS Down syndromeDNR DaunorubicinDOX DoxorubicinEBMT European Group for Blood and Marrow

Transplant

ECOG Eastern Cooperative Oncology GroupEFS event-free survivalEMD extramedullary diseaseEORTC European Organization of Research and

Treatment of CancerFAB French-American-Britishmorphology

classificationGAMLCG German AML Cooperative GroupGEGMO Groupe d’Etudes de la Greffe de Moelle

OuestGIMEMA Gruppo Italiano Malattie Ematologiche

Maligne dell’AdultoGOELAM Groupe Ouest Est Leuc�emies Aigues

My�eloblastiquesHiDAC high-dose Ara-CHLA human leukocyte antigenHOVON (Dutch-Belgian) Haemato-Oncology Co-

operative GroupHR high-risk and hazard ratioHSCT hematopoietic stem cell transplantationIBMTR International Bone Marrow Transplant

RegistryICC intensive consolidation chemotherapyIDR IdarubicinITT intention to treatLFS leukemia-free survivalMDS myelodysplastic syndromeMel MelphalanMITO MitozantroneMRC Medical Research Council


MRD matched-related donorMUD matched-unrelated donorOS overall survivalPBSC peripheral blood stem cellsPBSCT peripheral blood stem cell transplanta-

tionPMRD partially matched related donorQOL quality of lifeRCT randomized controlled trialRFS relapse-free survivalRR relative riskRBZ Rubidazone

SAKK Swiss Group for Clinical Cancer ResearchSCT stem cell transplantationSFGM Soci�et�e Francaise de Greffe de MoelleSR standard riskSWOG Southwest Oncology GroupSWPS survival-weighted psychometric scoresTBI total body irradiationTG ThioguanineTRM treatment-related mortalityURD unrelated donorVP-16 EtoposideWBC white blood cell

APPENDIX B: Outline of Article

AbstractIntroductionLiterature Search MethodologyQualitative and Quantitative Grading of the

EvidenceTreatment RecommendationsFormat of the ReviewSCT versus Chemotherapy in Adult AML

� Autologous SCT versus Chemotherapy in CR1� Allogeneic SCT versus Chemotherapy 6 Autolo-

gous SCT in CR1� Allogeneic SCT versus Chemotherapy in CR2

Autologous versus Allogeneic SCTAutologous SCT

� PBSCT versus BMT� Purged versus Unpurged� Tandem versus Single� Single Arm Cohort Studies

Allogeneic SCT

� Related versus Unrelated Donor� T Cell Depleted versus T Cell Replete

� PBSCT versus BMT� Single Arm Cohort Studies

Comparison of Treatment Regimens

� Autologous SCT� Allogeneic SCT (myeloablative; myeloablative

versus RIC; nonmyeloablative versus RIC)

Treatment for Relapse after First Allo-SCT

� DLI� Second Transplant

Autologous SCT as Treatment for Therapy-RelatedAML or Secondary AML

Late Effects and Quality of LifePrognostic FactorsFuture Directions

� Areas of Needed Research� Ongoing Studies

Strengths/Limitations and DiscussionAcknowledgementsAppendix A: Glossary of AbbreviationsReferences

REFERENCES

1. Hahn T, Wolff SN, Czuczman M, et al. The role of cytotoxic

therapy with hematopoietic stem cell transplantation in the

therapy of diffuse large cell B-cell non-Hodgkin’s lymphoma:

an evidence-based review. Biol Blood Marrow Transplant. 2001;

7:308-331.

2. Hahn T, Wingard JR, Anderson KC, et al. The role of cyto-

toxic therapy with hematopoietic stem cell transplantation in

the therapy of multiple myeloma: an evidence-based review.

Biol Blood Marrow Transplant. 2003;9:4-37.

3. Hahn T, Wall D, Camitta B, et al. The role of cytotoxic

therapy with hematopoietic stem cell transplantation in the

therapy of acute lymphoblastic leukemia in children: an

evidence-based review. Biol Blood Marrow Transplant. 2005;

11:823-861.

4. Hahn T, Wall D, Camitta B, et al. The role of cytotoxic therapy

with hematopoietic stem cell transplantation in the therapy of

acute lymphoblastic leukemia in adults: an evidence-based

review. Biol Blood Marrow Transplant. 2006;12:1-30.

5. Oliansky DM, Rizzo JD, Aplan PD, et al. The role of cyto-

toxic therapy with hematopoietic stem cell transplantation in

the therapy of acute myelogenous leukemia in children: an

evidence-based review. Biol Blood Marrow Transplant. 2007;13:

1-25.

6. Jones RJ, Nieto Y, Rizzo JD, et al. The evolution of the

evidence-based review: evaluating the science enhances the

art of medicine-statement of the Steering Committee for

Evidence-Based Reviews of the American Society for Blood

and Marrow Transplantation. Biol Blood Marrow Transplant.

2005;11:819-822.

7. Levi I, Grotto I, Yerushalmi R, et al. Meta-analysis of autolo-

gous bone marrow transplantation versus chemotherapy in

adult patients with acute myeloid leukemia in first remission.

Leuk Res. 2004;28:605-612.


8. Nathan PC, Sung L, Crump M, et al. Consolidation therapy

with autologous bone marrow transplantation in adults with

acute myeloid leukemia: a meta-analysis. J Natl Cancer Inst.

2004;96:38-45.

9. Zittoun RA, Mandelli F, Willemze R, et al. Autologous or

allogeneic bone marrow transplantation compared with

intensive chemotherapy in acute myelogenous leukemia.

European Organization for Research and Treatment of

Cancer (EORTC) and the Gruppo Italiano Malattie Emato-

logiche Maligne dell’Adulto (GIMEMA) Leukemia Coopera-

tive Groups. N Engl J Med. 1995;332:217-223.

10. Reiffers J, Stoppa AM, Attal M, et al. Allogeneic vs autologous

stem cell transplantation vs chemotherapy in patients with

acute myeloid leukemia in first remission: the BGMT 87 study.

Leukemia. 1996;10:1874-1882.

11. Harousseau JL, Cahn JY, Pignon B, et al. Comparison of autol-

ogous bone marrow transplantation and intensive chemother-

apy as postremission therapy in adult acute myeloid leukemia.

The Groupe Ouest Est Leucemies Aigues Myeloblastiques

(GOELAM). Blood. 1997;90:2978-2986.

12. Miggiano MC, Gherlinzoni F, Rosti G, et al. Autologous bone

marrow transplantation in late first complete remission im-

proves outcome in acute myelogenous leukemia. Leukemia.

1996;10:402-409.

13. Burnett AK, Goldstone AH, Stevens RM, et al. Randomised

comparison of addition of autologous bone-marrow transplan-

tation to intensive chemotherapy for acute myeloid leukaemia

in first remission: results of MRC AML 10 trial. UK Medical

Research Council Adult and Children’s Leukaemia Working

Parties. Lancet. 1998;351:700-708.

14. Breems DA, Boogaerts MA, Dekker AW, et al. Autologous

bone marrow transplantation as consolidation therapy in the

treatment of adult patients under 60 years with acute myeloid

leukaemia in first complete remission: a prospective random-

ized Dutch-Belgian Haemato-Oncology Co-operative Group

(HOVON) and Swiss Group for Clinical Cancer Research

(SAKK) trial. Br J Haematol. 2005;128:59-65.

15. Rohatiner AZ, Bassan R, Raimondi R, et al. High-dose treat-

ment with autologous bone marrow support as consolidation

of first remission in younger patients with acute myelogenous

leukaemia. Ann Oncol. 2000;11:1007-1015.

16. Cassileth PA, Harrington DP, Appelbaum FR, et al. Chemo-

therapy compared with autologous or allogeneic bone marrow

transplantation in the management of acute myeloid leukemia

in first remission. N Engl J Med. 1998;339:1649-1656.

17. Bassan R, Raimondi R, Lerede T, et al. Outcome assessment of

age group-specific (1/2 50 years) post-remission consolida-

tion with high-dose cytarabine or bone marrow autograft for

adult acute myelogenous leukemia. Haematologica. 1998;83:

627-635.

18. Yanada M, Matsuo K, Emi N, et al. Efficacy of allogeneic he-

matopoietic stem cell transplantation depends on cytogenetic

risk for acute myeloid leukemia in first disease remission.

Cancer. 2005;103:1652-1658.

19. Cornelissen JJ, van Putten WLJ, Verdonck LF, et al. Results of

a HOVON/SAKK donor versus no-donor analysis of myeloa-

blative HLA-identical sibling stem cell transplantation in first

remission acute myeloid leukemia in young and middle-aged

adults: benefits for whom? Blood. 2007;109:3658-3666.

20. Burnett AK, Wheatley K, Goldstone AH, et al. The value of

allogeneic bone marrow transplant in patients with acute mye-

loid leukaemia at differing risk of relapse: results of the UK

MRC AML 10 trial. Br J Haematol. 2002;118:385-400.

21. Gale RP, Buchner T, Zhang MJ, et al. HLA-identical sib-

ling bone marrow transplants vs chemotherapy for acute

myelogenous leukemia in first remission. Leukemia. 1996;10:

1687-1691.

22. Cassileth PA, Andersen JW, Bennett JM, et al. Escalating the

intensity of post-remission therapy improves the outcome in

acute myeloid leukemia: the ECOG experience. The Eastern

Cooperative Oncology Group. Leukemia. 1992;6(Suppl 2):

116-119.

23. Schiller GJ, Nimer SD, Territo MC, et al. Bone marrow trans-

plantation versus high-dose cytarabine-based consolidation

chemotherapy for acute myelogenous leukemia in first remis-

sion. J Clin Oncol. 1992;10:41-46.

24. Willemze R, Fibbe WE, Kluin-Nelemans JC, et al. Bone mar-

row transplantation or chemotherapy as post-remission treat-

ment of adult acute myelogenous leukemia. Ann Hematol.

1991;62:59-63.

25. Archimbaud E, Thomas X, Michallet M, et al. Prospective ge-

netically randomized comparison between intensive postinduc-

tion chemotherapy and bone marrow transplantation in adults

with newly diagnosed acute myeloid leukemia. J Clin Oncol.

1994;12:262-267.

26. Mohty M, de Lavallade H, Ladaique P, et al. The role of

reduced intensity conditioning allogeneic stem cell transplan-

tation in patients with acute myeloid leukemia: a donor vs no

donor comparison. Leukemia. 2005;19:916-920.

27. Gale RP, Horowitz MM, Rees JKH, et al. Chemotherapy

versus transplants for acute myelogenous leukemia in second

remission. Leukemia. 1996;10:13-19.

28. Ringden O, Labopin M, Gorin NC, et al. Is there a graft-ver-

sus-leukaemia effect in the absence of graft-versus-host disease

in patients undergoing bone marrow transplantation for acute

leukaemia? Br J Haematol. 2000;111:1130-1137.

29. Lazarus HM, Perez WS, Klein JP, et al. Autotransplantation

versus HLA-matched unrelated donor transplantation for

acute myeloid leukaemia: a retrospective analysis from the

Center for International Blood and Marrow Transplant

Research. Br J Haematol. 2006;132:755-759.

30. Suciu S, Mandelli F, de Witte T, et al. Allogeneic compared

with autologous stem cell transplantation in the treatment of

patients younger than 46 years with acute myeloid leukemia

(AML) in first complete remission (CR1): an intention-to-treat

analysis of the EORTC/GIMEMA AML-10 trial. Blood. 2003;

102:1232-1240.

31. Carella AM, Frassoni F, van Lint MT, et al. Autologous and

allogeneic bone marrow transplantation in acute myeloid

leukemia in first complete remission: an update of the Genoa

experience with 159 patients. Ann Hematol. 1992;64:128-131.

32. Singhal S, Henslee-Downey PJ, Powles R, et al. Haploidentical

vs autologous hematopoietic stem cell transplantation in pa-

tients with acute leukemia beyond first remission. Bone Marrow

Transplant. 2003;31:889-895.

33. Mitus AJ, Miller KB, Schenkein DP, et al. Improved survival

for patients with acute myelogenous leukemia. J Clin Oncol.

1995;13:560-569.

34. Cassileth PA, Andersen J, Lazarus HM, et al. Autologous bone

marrow transplant in acute myeloid leukemia in first remission.

J Clin Oncol. 1993;11:314-319.

35. Lowenberg B, Verdonck LJ, Dekker AW, et al. Autologous

bone marrow transplantation in acute myeloid leukemia in first


remission: results of a Dutch prospective study. J Clin Oncol.

1990;8:287-294.

36. Sierra J, Brunet S, Granena A, et al. Feasibility and results of

bone marrow transplantation after remission induction and in-

tensification chemotherapy in de novo acute myeloid leukemia.

Catalan Group for Bone Marrow Transplantation. J Clin Oncol.

1996;14:1353-1363.

37. Reiffers J, Labopin M, Sanz M, et al. Autologous blood cell vs

marrow transplantation for acute myeloid leukemia in com-

plete remission: an EBMT retrospective analysis. Bone Marrow

Transplant. 2000;25:1115-1119.

38. Sirohi B, Powles R, Kulkarni S, et al. Reassessing auto-

transplantation for acute myeloid leukaemia in first remis-

sion—a matched pair analysis of autologous marrow vs

peripheral blood stem cells. Bone Marrow Transplant. 2004;33:

1209-1214.

39. Vey N, Bouabdallah R, Stoppa A, et al. Autologous stem cell

transplantation for acute myelogenous leukemia in first com-

plete remission: a 6-year follow-up study of 101 patients from

a single institution. Bone Marrow Transplant. 2004;33:177-182.

40. Miller CB, Rowlings PA, Zhang MJ, et al. The effect of graft

purging with 4-hydroperoxycyclophosphamide in autologous

bone marrow transplantation for acute myelogenous leukemia.

Exp Hematol. 2001;29:1336-1346.

41. Gorin N, Aegerter P, Auvert B, et al. Autologous bone marrow

transplantation for acute myelocytic leukemia in first remis-

sion: a European survey of the role of marrow purging. Blood.

1990;75:1606-1614.

42. Chao NJ, Stein A, Long GD, et al. Busulfan/Etoposide-initial

experience with a new preparatory regimen for autologous

bone marrow transplantation in patients with acute nonlym-

phoblastic leukemia. Blood. 1993;81:319-323.

43. McMillan AK, Goldstone AH, Linch DC, et al. High-dose

chemotherapy and autologous bone marrow transplantation

in acute myeloid leukemia. Blood. 1990;76:480-488.

44. Kim HJ, Min WS, Eom KS, et al. Autologous stem cell

transplantation using modified TAM or combination of triple-

alkylating agents conditioning regimens as one of the post-

remission treatments in patients with adult acute myeloid

leukemia in first complete remission. Bone Marrow Transplant.

2004;34:215-220.

45. Mollee P, Gupta V, Song K, et al. Long-term outcome after in-

tensive therapy with etoposide, melphalan, total body irradia-

tion and autotransplant for acute myeloid leukemia. Bone

Marrow Transplant. 2004;33:1201-2108.

46. Gorin N, Labopin M, Pichard P, et al. Feasibility and recent

improvement of autologous stem cell transplantation for acute

myelocytic leukaemia in patients over 60 years of age: impor-

tance of the source of stem cells. Br J Haematol. 2000;110:

887-893.

47. Gorin N, Labopin M, Laporte JP, et al. Importance of marrow

dose on posttransplant outcome in acute leukemia: models de-

rived from patients autografted with mafosfamide-purged mar-

row at a single institution. Experimental Hematology. 1999;27:

1822-1830.

48. Linker CA, Ries CA, Damon LE, et al. Autologous bone mar-

row transplantation for acute myeloid leukemia using 4-hydro-

peroxycyclophosphamide-purged bone marrow and the

busulfan/etoposide preparative regimen: a follow-up report.

Bone Marrow Transplant. 1998;22:865-872.

49. Isnard F, Guiguet M, Laporte JP, et al. Improved efficiency of

remission induction facilitates autologous BMT harvesting and

improves overall survival in adults with AML: 108 patients

treated at a single institution. Bone Marrow Transplant. 2001;

27:1045-1052.

50. Jourdan E, Rigal-Huguet F, Marit G, et al. One versus 2 high-

dose cytarabine-based consolidation before autologous stem

cell transplantation for young acute myeloblastic leukaemia pa-

tients in first complete remission. Br J Haematol. 2005;129:

403-410.

51. Meloni G, Vignetti M, Avvisati G, et al. BAVC regimen and au-

tograft for acute myelogenous leukemia in second complete re-

mission. Bone Marrow Transplant. 1996;18:693-698.

52. Stein AS, O’Donnell MR, Chai A, et al. In vivo purging with

high-dose cytarabine followed by high-dose chemoradiother-

apy and reinfusion of unpurged bone marrow for adult acute

myelogenous leukemia in first complete remission. J Clin Oncol.

1996;14:2206-2216.

53. Linker CA, Ries CA, Damon LE, et al. Autologous stem cell

transplantation for acute myeloid leukemia in first remission.


54. Pavlovsky S, Fernandez I, Milone G, et al. Autologous periph-

eral blood progenitor cell transplantation mobilized with high-

dose cytarabine in acute myeloid leukemia in first complete

remission. Ann Oncol. 1998;9:151-157.

55. Hegenbart U, Niederwieser D, Sandmaier BM, et al. Treat-

ment for acute myelogenous leukemia by low-dose, total-

body, irradiation-based conditioning and hematopoietic cell

transplantation from related and unrelated donors. J Clin Oncol.

2006;24:444-453.

56. Sayer HG, Kroger M, Beyer J, et al. Reduced intensity condi-

tioning for allogeneic hematopoietic stem cell transplantation

in patients with acute myeloid leukemia: disease status by mar-

row blasts is the strongest prognostic factor. Bone Marrow

Transplant. 2003;31:1089-1095.

57. Wagner JE, Thompson JS, Carter SL, et al. Effect of graft-

versus-host disease prophylaxis on 3-year disease-free survival

in recipients of unrelated donor bone marrow (T-cell Deple-

tion Trial): a multi-centre, randomised phase II-III trial.

Lancet. 2005;366:733-741.

58. Hale G, Zhang MJ, Bunjes D, et al. Improving the outcome of

bone marrow transplantation by using CD52 monoclonal anti-

bodies to prevent graft-versus-host disease and graft rejection.

Blood. 1998;92:4581-4590.

59. Marmont AM, Horowitz MM, Gale RP, et al. T-cell depletion

of HLA-identical transplants in leukemia. Blood. 1991;78:

2120-2130.

60. Garderet L, Labopin M, Gorin NC, et al. Patients with acute

lymphoblastic leukaemia allografted with a matched unrelated

donor may have a lower survival with a peripheral blood stem

cell graft compared to bone marrow. Bone Marrow Transplant.

2003;31:23-29.

61. Ringden O, Labopin M, Bacigalupo A, et al. Transplantation of

peripheral blood stem cells as compared with bone marrow

from HLA-identical siblings in adult patients with acute mye-

loid leukemia and acute lymphoblastic leukemia. J Clin Oncol.

2002;20:4655-4664.

62. Michallet M, Thomas X, Vernant JP, et al. Long-term outcome

after allogeneic hematopoietic stem cell transplantation for ad-

vanced stage acute myeloblastic leukemia: a retrospective study

of 379 patients reported to the Societe Francaise de Greffe de

Moelle (SFGM). Bone Marrow Transplant. 2000;26:1157-1163.

63. Stelljes M, Bornhauser M, Kroger M, et al. Conditioning with

8-Gy total body irradiation and fludarabine for allogeneic


hematopoietic stem cell transplantation in acute myeloid leuke-

mia. Blood. 2005;106:3314-3321.

64. Greinix HT, Nachbaur D, Krieger O, et al. Factors affecting

long-term outcome after allogeneic haematopoietic stem cell

transplantation for acute myelogenous leukaemia: a retrospec-

tive study of 172 adult patients reported to the Austrian Stem

Cell Transplantation Registry. Br J Haematol. 2002;117:

914-923.

65. Sierra J, Storer B, Hansen JA, et al. Unrelated donor marrow

transplantation for acute myeloid leukemia: an update of the

Seattle experience. Bone Marrow Transplant. 2000;26:397-404.

66. Cook G, Clark RE, Crawley C, et al. The outcome of sibling

and unrelated donor allogeneic stem cell transplantation in

adult patients with acute myeloid leukemia in first remission

who were initially refractory to first induction chemotherapy.


67. Bunjes D. Re-labeled anti-CD66 monoclonal antibody in stem

cell transplantation for patients with high-risk acute myeloid

leukemia. Leuk Lymphoma. 2002;43:2125-2131.

68. Rocha V, Labopin M, Gluckman E, et al. Relevance of bone

marrow cell dose on allogeneic transplantation outcomes for

patients with acute myeloid leukemia in first complete remis-

sion: results of a European survey. J Clin Oncol. 2002;20:

4324-4330.

69. Tallman MS, Rowlings PA, Milone G, et al. Effect of postre-

mission chemotherapy before human leukocyte antigen-identi-

cal sibling transplantation for acute myelogenous leukemia in

first complete remission. Blood. 2000;96:1254-1258.

70. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic

marrow transplantation during untreated first relapse of acute

myeloid leukemia. J Clin Oncol. 1992;10:1723-1729.

71. Mehta J, Powles R, Sirohi B, et al. Does donor-recipient ABO

incompatibility protect against relapse after allogeneic bone

marrow transplantation in first remission acute myeloid leuke-

mia? Bone Marrow Transplant. 2002;29:853-859.

72. Brown RA, Wolff SN, Fay JW, et al. High-dose etoposide, cy-

clophosphamide and total body irradiation with allogeneic

bone marrow transplantation for resistant acute myeloid leuke-

mia: a study by the North American Marrow Transplant

Group. Leuk Lymphoma. 1996;22:271-277.

73. Greinix HT, Keil F, Brugger SA, et al. Long-term leukemia-

free survival after allogeneic marrow transplantation in patients

with acute myelogenous leukemia. Ann Hematol. 1996;72:

53-59.

74. Zikos P, Van Lint MT, Frassoni F, et al. Low transplant mor-

tality in allogeneic bone marrow transplantation for acute my-

eloid leukemia: a randomized study of low-dose cyclosporin

versus low-dose cyclosporin and low-dose methotrexate. Blood.

1998;91:3503-3508.

75. Bibawi S, Abi-Said D, Fayad L, et al. Thiotepa, busulfan, and

cyclophosphamide as a preparative regimen for allogeneic

transplantation for advanced myelodysplastic syndrome and

acute myelogenous leukemia. Am J Hematol. 2001;67:

227-233.

76. Frassoni F. Eradication of leukaemic marrow and prevention of

leukaemia relapse with total body irradiation and bone marrow

transplantation. Med Oncol Tumor Pharmacother. 1991;8:

189-201.

77. Ringden O, Labopin M, Tura S, et al. A comparison of busul-

phan versus total body irradiation combined with cyclophos-

phamide as conditioning for autograft or allograft bone

marrow transplantation in patients with acute leukaemia. Acute

Leukaemia Working Party of the European Group for Blood

and Marrow Transplantation (EBMT). Br J Haematol. 1996;

93:637-645.

78. Ball ED, Wilson J, Phelps V, et al. Autologous bone marrow

transplantation for acute myeloid leukemia in remission or first

relapse using monoclonal antibody-purged marrow: results of

phase II studies with long-term follow-up. Bone Marrow Trans-

plant. 2000;25:823-829.

79. 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 Cy-

toxan-total body irradiation as preparative regimen: a report

from the Group d’Etudes de la Greffe de Moelle Osseuse.

Blood. 1992;79:2578-2582.

80. Ringden O, Remberger M, Ruutu T, et al. Increased risk of

chronic graft-versus-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. Nordic Bone Marrow Transplanta-

tion Group. Blood. 1999;93:2196-2201.

81. Farag SS, Bolwell BJ, Elder PJ, et al. High-dose busulfan,

cyclophosphamide, and etoposide does not improve outcome

of allogeneic stem cell transplantation compared to BuCy2 in

patients with acute myeloid leukemia. Bone Marrow Transplant.

2005;35:653-661.

82. Resbeut M, Cowen D, Blaise D, et al. Fractionated or single-

dose total body irradiation in 171 acute myeloblastic leukemias

in first complete remission: is there a best choice? SFGM.

Societe Francaise de Greffe de Moelle. Int J Radiat Oncol Biol

Phys. 1995;31:509-517.

83. Litzow MR, Perez WS, Klein JP, et al. Comparison of outcome

following allogeneic bone marrow transplantation with cyclo-

phosphamide-total body irradiation versus busulphan-cyclo-

phosphamide conditioning regimens for acute myelogenous

leukaemia in first remission. Br J Haematol. 2002;119:

1115-1124.

84. Bacigalupo A, Vitale V, Corvo R, et al. The combined effect of

total body irradiation (TBI) and cyclosporin A (CyA) on the

risk of relapse in patients with acute myeloid leukaemia under-

going allogeneic bone marrow transplantation. Br J Haematol.

2000;108:99-104.

85. Schaap N, Schattenberg A, Bar B, et al. Outcome of transplan-

tation for standard-risk leukaemia with grafts depleted of lym-

phocytes after conditioning with an intensified regimen. Br J

Haematol. 1997;98:750-759.

86. Shimoni A, Hardan I, Shem-Tov N, et al. Allogeneic hemato-

poietic stem-cell transplantation in AML and MDS using mye-

loablabtive versus reduced-intensity conditioning: the role of

dose intensity. Leukemia. 2006;20:322-328.

87. Aoudjhane M, Labopin M, Gorin NC, et al. Comparative out-

come of reduced intensity and myeloablative conditioning reg-

imen in HLA identical sibling allogeneic haematopoietic stem

cell transplantation for patients older than 50 years of age with

acute myeloblastic leukaemia: a retrospective survey from the

Acute Leukemia Working Party (ALWP) of the European

group for Blood and Marrow Transplantation (EBMT). Leuke-

mia. 2005;19:2304-2312.

88. de Lima M, Anagnostopoulos A, Munsell M, et al. Nonablative

versus reduced-intensity conditioning regimens in the treat-

ment of acute myeloid leukemia and high-risk myelodysplastic

syndrome: dose is relevant for long-term disease control after


allogeneic hematopoietic stem cell transplantation. Blood. 2004;

104:865-872.

89. Kroger N, Zabelina T, Sonnenberg S, et al. Dose-dependent

effect of etoposide in combination with busulfan plus cyclo-

phosphamide as conditioning for stem cell transplantation in

patients with acute myeloid leukemia. Bone Marrow Transplant.

2000;26:711-716.

90. Collins RH Jr., Shpilberg O, Drobyski WR, et al. Donor leu-

kocyte infusions in 140 patients with relapsed malignancy after

allogeneic bone marrow transplantation. J Clin Oncol. 1997;15:

433-444.

91. Eapen M, Giralt SA, Horowitz MM, et al. Second transplant for

acute and chronic leukemia relapsing after first HLA-identical

sibling transplant. Bone Marrow Transplant. 2004;34:721-727.

92. De Witte T, Van Biezen A, Hermans J, et al. Autologous bone

marrow transplantation for patients with myelodysplastic syn-

drome (MDS) or acute myeloid leukemia following MDS.

Chronic and Acute Leukemia Working Parties of the European

Group for Blood and Marrow Transplantation. Blood. 1997;90:

3853-3857.

93. Kroger N, Brand R, van Biezen A, et al. Autologous stem cell

transplantation for therapy-related acute myeloid leukemia

and myelodysplastic syndrome. Bone Marrow Transplant. 2006;

37:183-189.

94. Frassoni F, Labopin M, Gluckman E, et al. Are patients with

acute leukaemia, alive and well 2 years post bone marrow

transplantation cured? A European survey. Acute Leukaemia

Working Party of the European Group for Bone Marrow

Transplantation (EBMT). Leukemia. 1994;8:924-948.

95. Abdallah A, Egerer G, Weber-Nordt RM, et al. Long-term

outcome in acute myelogenous leukemia autografted with ma-

fosfamide-purged marrow in a single institution: adverse events

and incidence of secondary myelodysplasia. Bone Marrow

Transplant. 2002;30:15-22.

96. Wadleigh M, Richardson PG, Zahrieh D, et al. Prior gemtu-

zumab ozogamicin exposure significantly increases the risk

of veno-occlusive disease in patients who undergo myeloabla-

tive allogeneic stem cell transplantation. Blood. 2003;102:

1578-1582.

97. Toor AA, Choo SY, Little JA. Bleeding risk and platelet trans-

fusion refractoriness in patients with acute myelogenous leuke-

mia who undergo autologous stem cell transplantation. Bone


98. Butt NM, Clark RE. Autografting as a risk factor for persisting

iron overload in long-term survivors of acute myeloid leukae-

mia. Bone Marrow Transplant. 2003;32:909-913.

99. Watson M, Buck G, Wheatley K, et al. Adverse impact of bone

marrow transplantation on quality of life in acute myeloid leu-

kaemia patients; analysis of the UK Medical Research Council

AML 10 Trial. Eur J Cancer. 2004;40:971-978.

100. Zittoun R, Suciu S, Watson M, et al. Quality of life in patients

with acute myelogenous leukemia in prolonged first complete

remission after bone marrow transplantation (allogeneic or au-

tologous) or chemotherapy: a cross-sectional study of the

EORTC-GIMEMA AML 8A trial. Bone Marrow Transplant.

1997;20:307-315.

101. Hsu C, Wang JD, Hwang JS, et al. Survival-weighted health

profile for long-term survivors of acute myelogenous leukemia.

Qual Life Res. 2003;12:503-517.

102. Wellisch DK, Centeno J, Guzman J, et al. Bone marrow

transplantation vs. high-dose cytorabine-based consolidation

chemotherapy for acute myelogenous leukemia. A long-term

follow-up study of quality-of-life measures of survivors.

Psychosomatics. 1996;37:144-154.

103. Delaunay J, Vey N, Leblanc T, et al. Prognosis of in(16)/

t(16;16) acute myeloid leukemia (AML): a survey of 110 cases

from the French AML Intergroup. Blood. 2003;102:462-469.

104. Farag SS, Archer KJ, Mrozek K, et al. Isolated trisomy of chro-

mosomes 8, 11, 13 and 21 is an adverse prognostic factor

in adults with de novo acute myeloid leukemia: results from

Cancer and Leukemia Group B 8461. Int J Oncol. 2002;21:

1041-1051.

105. Ferrant A, Labopin F, Frassoni F, et al. Karoytype in acute my-

eloblastic leukemia: prognostic significance for bone marrow

transplantation in first remission: a European Group for Blood

and Marrow Transplantation Study. Blood. 1997;90:2931-2938.

106. Frassoni F, Labopin M, Powles R, et al. Effect of centre on out-

come of bone-marrow transplantation for acute myeloid leu-

kaemia. Acute Leukaemia Working Party of the European

Group for Blood and Marrow Transplantation. Lancet. 2000;

355:1393-1398.

107. Gale RE, Hills R, Kottaridis PD, et al. No evidence that FLT3

status should be considered as an indicator for transplantation

in acute myeloid leukemia (AML): an analysis of 1135 patients,

excluding acute promyelocytic leukemia, from the UK MRC

AML10 and 12 trials. Blood. 2005;106:3658-3665.

108. Jourdan E, Maraninchi D, Reiffers J, et al. Early allogeneic

transplantation favorably influences the outcome of adult pa-

tients suffering from acute myeloid leukemia. Societe Francaise

de Greffe de Moelle (SFGM). Bone Marrow Transplant. 1997;

19:875-881.

109. Kebriaei P, Kline J, Stock W, et al. Impact of disease burden at

time of allogeneic stem cell transplantation in adults with acute

myeloid leukemia and myelodysplastic syndromes. Bone


110. Marcucci G, Mrozek K, Ruppert AS, et al. Prognostic factors

and outcome of core binding factor acute myeloid leukemia pa-

tients with t(8;21) differ from those of patients with inv(16): A

Cancer and Leukemia Group B Study. J Clin Oncol. 2005;23:

5705-5717.

111. Ogawa H, Ikegame K, Kawakami M, et al. Impact of cytoge-

netics on outcome of stem cell transplantation for acute mye-

loid leukemia in first remission: a large-scale retrospective

analysis of data from the Japan Society for Hematopoietic

Cell Transplantation. Int J Hematol. 2004;79:495-500.

112. Schoch C, Schnittger S, Klaus M, et al. AML with 11q23/MLL

abnormalities as defined by the WHO classification: incidence,

partner chromosomes, FAB subtype, age distribution, and

prognostic impact in an unselected series of 1897 cytogeneti-

cally analyzed AML cases. Blood. 2003;102:2395-2402.

113. Wahlin A, Markevarn B, Golovleva I, et al. Improved outcome

in adult acute myeloid leukemia is almost entirely restricted to

young patients and associated with stem cell transplantation.

Eur J Haematol. 2002;68:54-63.

114. Wheatley K, Burnett A, Goldstone A, et al. A simple, robust,

validated and highly predictive index for the determination of

risk-directed therapy in acute myeloid leukaemia derived

from the MRC AML 10 trial. Br J Haematol. 1999;107:69-79.

115. Cahn JY, Labopin M, Mandelli F, et al. Autologous bone

marrow transplantation for first remission acute myeloblastic

leukemia in patients older than 50 years: a retrospective analysis

of the European Bone Marrow Transplant Group. Blood. 1995;

85:575-579.


116. Slovak ML, Kopecky KJ, Cassileth PA, et al. Karyotypic anal-

ysis predicts outcome of preremission and postremission ther-

apy in adult acute myeloid leukemia: a Southwest Oncology

Group/Eastern Cooperative Oncology Group Study. Blood.

2000;96:4075-4083.

117. Fung HC, Stein A, Slovak M, et al. A long-term follow-up re-

port on allogeneic stem cell transplantation for patients with

primary refractory acute myelogenous leukemia: impact of cy-

togenetic characteristics on transplantation outcome. Biol Bone


118. Lee KW, Choi IS, Roh EY, et al. Adult patients with t(8;21)

acute myeloid leukemia had no superior treatment outcome

to those without t(8;21): a single institution’s experience.

Ann Hematol. 2004;83:218-224.

119. Schlenk RF, Benner A, Hartmann F, et al. Risk-adapted post-

remission therapy in acute myeloid leukemia: results of the

German multicenter AML HD93 treatment trial. Leukemia.

2003;17:1521-1528.

120. Nguyen S, Leblanc T, Fenaux P, et al. A white blood cell index

as the main prognostic factor in t(8;21) acute myeloid leukemia

(AML): a survey of 161 cases from the French AML Inter-

group. Blood. 2002;99:3517-3523.

121. Wong R, Shahjahan M, Wang X, et al. Prognostic factors for

outcomes of patients with refractory or relapsed acute myelog-

enous leukemia or myelodysplastic syndromes undergoing

allogeneic progenitor cell transplantation. Biol Blood Marrow

Transplant. 2005;11:108-114.

122. Mehta J, Powles R, Sirohi B, et al. Impact of cytogenetics on

the outcome of autotransplantation for acute myeloid leukemia

in first remission: is the benefit of intensive pretransplant ther-

apy limited to patients with good karyotypes? Bone Marrow

Transplant. 2003;32:157-164.

123. Slovak ML, Kopecky KJ, Wolman SR, et al. Cytogenetic cor-

relation with disease status and treatment outcome in advanced

stage leukemia post bone marrow transplantation: a Southwest

Oncology Group study (SWOG-8612). Leukemia Res. 1995;19:

381-388.

124. Gale RP, Horowitz MM, Weiner RS, et al. Impact of cytoge-

netic abnormalities on outcome of bone marrow transplants

in acute myelogenous leukemia in first remission. Bone Marrow

Transplant. 1995;16:203-208.

125. Baldus CD, Tanner SM, Ruppert AS, et al. BAALC expression

predicts clinical outcome of de novo acute myeloid leukemia pa-

tients with normal cytogenetics: a Cancer and Leukemia Group

B Study. Blood. 2003;102:1613-1618.

126. Dohner K, Schlenk RF, Habdank M, et al. Mutant nucleophos-

min (NPM1) predicts favorable prognosis in younger adults

with acute myeloid leukemia and normal cytogenetics: interac-

tion with other gene mutations. Blood. 2005;106:3740-3746.

127. Frohling S, Schlenk RF, Breitruck J, et al. Prognostic signifi-

cance of activating FLT3 mutations in younger adults (16 to

60 years) with acute myeloid leukemia and normal cytogenet-

ics: a study of the AML Study Group Ulm. Blood. 2002;100:

4372-4380.

128. Frohling S, Schlenk RF, Stolze I, et al. CEBPA mutations in

younger adults with acute myeloid leukemia and normal cyto-

genetics: Prognostic relevance and analysis of cooperating mu-

tations. J Clin Oncol. 2004;22:624-633.

129. Kottaridis PD, Gale RE, Frew ME, et al. The presence of

a FLT3 internal tandem duplication in patients with acute

myeloid leukemia (AML) adds important prognostic informa-

tion to cytogenetic risk group and response to the first cycle

of chemotherapy: analysis of 854 patients from the United

Kingdom Medical Research Council AML 10 and 12 trials.

Blood. 2001;98:1752-1759.

130. Marcucci G, Baldus CD, Ruppert AS, et al. Overexpression of

the ETS-related gene, ERG, predicts a worse outcome in acute

myeloid leukemia with normal karyotype: a Cancer and Leuke-

mia Group B Study. J Clin Oncol. 2005;23:9234-9242.

131. Rombouts EJC, Pavic B, Lowenberg B, et al. Relation between

CXCR-4 expression, Flt3 mutations, and unfavorable progno-

sis of adult acute myeloid leukemia. Blood. 2004;104:550-557.

132. Rombouts WJC, Lowenberg B, Van Putten WLJ, et al. Im-

proved prognostic significance of cytokine-induced prolifera-

tion in vitro in patients with de novo acute myeloid leukemia

of intermediate risk: Impact of internal tandem duplication in

the Flt3 gene. Leukemia. 2001;15:1046-1053.

133. Schnittger S, Schoch C, Kern W, et al. Nucleophosmin gene

mutations are predictors of favorable prognosis in acute mye-

logenous leukemia with a normal karyotype. Blood. 2005;106:

3733-3739.

134. Thiede C, Koch S, Creutzig E, et al. Prevalence and prognostic

impact of NPM1 mutations in 1485 adult patients with acute

myeloid leukemia (AML). Blood. 2006;107:4011-4020.

135. Thiede C, Steudel C, Mohr B, et al. Analysis of FLT3-activat-

ing mutations in 979 patients with acute myelogenous leuke-

mia: association with FAB subtypes and identification of

subgroups with poor prognosis. Blood. 2002;99:4326-4335.

136. Whitman SP, Archer KJ, Feng L, et al. Absence of the wild-

type allele predicts poor prognosis in adult de novo acute

myeloid leukemia with normal cytogenetics and the internal

tandem duplication of FLT3: A Cancer and Leukemia Group

B Study. Cancer Res. 2001;61:7233-7239.

137. Casanovas RO, Slimane FK, Garand R, et al. Immunological

classification of acute myeloblastic leukemias: relevance to

patient outcome. Leukemia. 2003;17:515-527.

138. Perea G, Domingo A, Villamor N, et al. Adverse prognostic

impact of CD36 and CD2 expression in adult de novo acute my-

eloid leukemia patients. Leukemia Res. 2005;29:1109-1116.

139. Raspadori D, Damiani D, Lenoci M, et al. CD56 antigenic ex-

pression in acute myeloid leukemia identifies patients with

poor clinical prognosis. Leukemia. 2001;15:1161-1164.

140. Repp R, Schaekel U, Helm G, et al. Immunophenotyping is an

independent factor for risk stratification in AML. Cytometry

Part B: Clin Cytometry. 2003;53B:11-19.

141. Spoo AC, Lubbert M, Wierda WG, et al. CXCR4 is a prognostic

marker inacute myelogenous leukemia. Blood. 2007;109:786-791.

142. Schlenk RF, Corbacioglu A, Krauter J, et al. Gene mutations as

predictive markers for postremission therapy in younger adults

with normal karyotype AML. ASH Ann Meet Abstr. 2006;108:4.

143. Schaich M, Fuessel M, Bornhauser M, et al. Rapid risk-profil-

ing including fast donor search within a multi-center trial

allows for early allogeneic stem cell transplantation during

aplasia in patients with high-risk acute myeloid leukemia

[abstract]. Blood. 2004;104:2767.

144. Davies SM RN, Weisdorf DJ. Feasibility and timing of unre-

lated donor identification for patients with ALL. Bone Marrow

Transplant. 1996;17:734-740.

the role of cytotoxic therapy with hematopoietic stem cell ...stem cell transplantation in the...

Documents