Reprint of: Allogeneic Hematopoietic Cell Transplantation for Acute Myeloid Leukemia

Biol Blood Marrow Transplant 21 (2015) S3eS10

Biology of Blood and Marrow Transplantation journal homepage: www.bbmt.org

Educational Reviews

Reprint of: Allogeneic Hematopoietic Cell Transplantation for Acute Myeloid Leukemiaq Paresh Vyas 1, Frederick R. Appelbaum 2, Charles Craddock 3, * 1 2 3

Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, United Kingdom Fred Hutchinson Cancer Research Center, University of Washington, Seattle, Washington Queen Elizabeth Hospital, Birmingham, United Kingdom

Article history: Received 8 October 2014 Accepted 29 October 2014 Key Words: Acute myeloid leukemia Leukemic stem/progenitor cell Conditioning regimen Reduced-intensity conditioning Graft-versus-leukemia Disease relapse

a b s t r a c t Allogeneic stem cell transplantation is an increasingly important treatment option in the management of adult acute myeloid leukemia (AML). The major causes of treatment failure remain disease relapse and treatment toxicity. In this review, Dr Vyas presents an overview of important recent data defining molecular factors associated with treatment failure in AML. He also identifies the emerging importance of leukemia stem cell biology in determining both response to therapy and relapse risk in AML. Dr Appelbaum discusses advances in the design and delivery of both myeloablative and reduced-intensity conditioning regimens, highlighting novel strategies with the potential to improve outcome. Dr Craddock discusses the development of both novel conditioning regimens and post-transplantation strategies aimed at reducing the risk of disease relapse. Ó 2015 Published by Elsevier Inc. on behalf of American Society for Blood and Marrow Transplantation.

INTRODUCTION The advent of reduced-intensity conditioning (RIC) regimens, coupled with expansion in alternative donor stem cell sources, has dramatically increased the number of patients with acute myeloid leukemia (AML) in whom allogeneic transplantation can be considered a potential treatment option. Given its attendant toxicity, the choice of which patients have the potential to benefit from an allograft now assumes even greater importance. Advances in our understanding of the molecular determinants of disease relapse in patients treated with conventional chemotherapy, coupled with increased sophistication in predicting transplantationrelated mortality on the basis of both patient-donor HLA disparity and comorbidity scoring are now routinely used to inform such decision-making. At the same time, it remains the case that a substantial proportion of patients who proceed to transplantation are destined to die of either procedure-related toxicity or disease DOI of original article: http://dx.doi.org/10.1016/j.bbmt.2014.10.026. q This article is a reprint of a previously published article. For citation purposes, please use the original publication details; Biol Blood Marrow Transplant. 2015;21:8e15. Financial disclosure: See Acknowledgments on page S8. * Correspondence and reprint requests: Charles Craddock, FRCPath FRCP DPhil BMBCh, Queen Elizabeth Hospital, Birmingham, B15 2TH, United Kingdom. E-mail address: [email protected] (C. Craddock).

relapse. In this context, optimization of both patient selection and conditioning regimen design is central to improving outcome. The challenge of reducing the risk of disease relapse after allograft remains a particularly stubborn problem, but emerging and varied strategies focusing on either augmentation of the antitumor activity of the conditioning regimen, without increasing its toxicity, or enhancement of a graft-versus-leukemia (GVL) effect hold promise. Pivotal to improving transplantation outcome will be the incorporation of advances in our understanding of the biology of disease relapse into transplantation schedules, coupled with the prospective assessment of novel treatment strategies in early phase clinical trials and subsequently appropriately powered randomized studies.

MOLECULAR AND CELLULAR DETERMINANTS OF RELAPSE IN AML The most common cause of death in treated patients with AML is therapy-resistant disease. This review focuses on relapse in patients with nonacute promyelocytic AML. Approximately 50% of patients under the age of 60 and 90% of patients over the age of 60 with nonacute promyelocytic AML relapse from initial therapy [1,2]. Despite enormous strides in understanding the genetic basis of AML and the importance of the clinical relapse, it is remarkable how little we know about the cellular and molecular mechanisms mediating therapy

http://dx.doi.org/10.1016/j.bbmt.2014.12.032 1083-8791/Ó 2015 Published by Elsevier Inc. on behalf of American Society for Blood and Marrow Transplantation.

S4

P. Vyas et al. / Biol Blood Marrow Transplant 21 (2015) S3eS10

Table 1 European LeukemiaNET Genetic Prognostic Risk Groups Genetic Risk Group

Genetic Markers

Favorable

t(8;21)(q22;q22); RUNX1-RUNX1T1 inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB-MYH11 Mutated NPM1 without FLT3-ITD (CN-AML) Bi-allelic mutation of CEBPA (CN-AML) Mutated NPM1 and FLT3-ITD (CN-AML) Wild-type NPM1 and FLT3-ITD (CN-AML) Wild-type NPM1 without FLT3-ITD (CN-AML) t(9;11)(p22;q23); MLLT3-MLL Cytogenetic abnormalities not classified as favorable or adverse inv(3)(q21q26.2) or t(3;3)(q21;q26.2); RPN1-EVI1. t(6;9)(p23;q34); DEK-NUP214. t(v;11)(v;q23); MLL rearranged.
5 or del(5q);
7; abnormal (17p). Complex karyotype*

Intermediate-I

Intermediate-II

Adverse

* Complex karyotype is defined as 3 or more chromosome abnormalities in absence of World Health Organizationedesignated recurring translocations or inversions: t(15;17), t(8;21), inv(16) or t(16;16), t(9;11), t(v;11)(v;q23), t(6;9), inv(3), or t(3;3).

resistance and, thus, relapse. The review will summarize mature data on the AML mutational profile prognostic of relapse and discuss cellular heterogeneity in AML that may also be important in therapy leading to relapse. Genetic Markers Associated With Relapse Our current understanding is that survival of AML patients is dependent on patient-specific parameters (namely, age, comorbidities, and performance status) and the profile of the pathogenetic driver genetic variants in the AML. Investigators and trial groups have devised a number of different prognostic risk groups from clinical and/or diseasespecific variables at diagnosis. One of the most widely used is that devised by the European LeukemiaNET panel of experts that divides AML into 4 prognostic risk groups (Table 1) [2]. The highest chance of achieving a complete remission (CR) with intensive induction chemotherapy is in the favorable-risk group (80% to 90% in patients under 60 years of age and 70% to 80% in patients over the age of 60 years) and the lowest chance of CR occurs in the adverse-risk group (25% to 35%). Once a patient is in CR, the chance of relapse varies depending on disease-specific cytogenetic and genetic factors and the type of postinduction consolidation therapy the patient receives. In general, European LeukemiaNET favorable risk groups have the lowest risk of relapse (10% to 35%) and usually receive combination chemotherapy with or without gemtuzumab ozogamicin monoclonal anti-CD33 antibody conjugated with calicheamicin [3]. Relapse is highest in the adverse-risk group, where it varies from 70% to 90% with chemotherapy to 30% to 50% with myeloablative (MA) transplantation. Furthermore, within a risk category, coacquisition of additional specific mutations influences prognosis and relapse risk. Good examples of this are the increased relapse risk with coacquisition of mutations in the KIT gene in core binding factor (CBF) leukemia [4,5] and RUNX1 mutations in normal karyotype AML [6,7]. However, the data are not so clear in all situations. There are conflicting reports on the impact of TET2 mutations in AML patients with normal cytogenetics and mutations in either NPM1 but wild-type for FLT3 or in patients with bi-allelic mutations in CEPBA [8,9]. Thus, though we are beginning to understand how to use

genetic information together and we have an improved understanding of toxicity of therapies to develop risk-adapted approaches to offer optimal postinduction consolidation and decide who to offer transplantation [10], our understanding of markers of relapse are still rudimentary. Recent whole genome (WGS) and exome sequencing studies in diagnostic samples from de novo AML patients have begun to illustrate the extent of molecular heterogeneity in AML. Analysis of 200 AML samples (50 by WGS and 150 exomes) shows that 23 genes demonstrate a higher than expected mutation prevalence in AML; though, like most cancers, there is a much larger number of genes (at least 1893) where mutations in coding regions exist at low frequency and are predicted to cause significant change in protein function (tier 1 mutations) [11]. Further large-scale WGS studies (with deeper coverage) and targeted resequencing studies are in progress or are planned to validate mutation frequency and more accurately define the frequency of rare mutations. As our understanding of the functional significance of recurrent acquired synonymous [12] and noncoding mutations in RNA and regulatory elements [13] increases, we will gain a deeper and more complete picture of the genetic complexity of AML. Several of these studies are planned on samples from large clinical trials with the potential to refine our current prognostic risk classifications. Eventually, these studies will have the potential to provide information on genetic variants that predict response to specific therapies. Any 1 AML sample has an average of 13 (range, 2 to 15) tier 1 mutations, organized in at least 1 to 5 (average, 2) different clones [11]. Although this raises the possibility of an almost infinite number of different clonal structures across AML as a whole, as AML driver mutations are often coselected, there are likely to be a more restricted number of common clonal genotypes across the majority of patients [11,14]. Given that the biologic behavior of AML cells is the sum of all the acquired genetic changes, it is critical to define clonal structures. This can be imputed by studies of variant allele frequency (ie, clonal structures are inferred by studying how common different mutations are) but is best determined by extensive single-cell genotyping that experimentally proves clonal composition. Such studies are now just beginning in AML (Quek and Vyas, unpublished data [15]). Few studies have studied paired diagnostic and relapse samples at a global genetic level to define how the genetic composition of AML cells changes at relapse. The first WGS study of 8 paired samples showed that clonal evolution is common at relapse and occurs through acquisition of new mutations, either in the dominant founder clone or in a subclone of the founder clone [16]. This study also highlighted the mutagenic effect of DNA-damaging chemotherapy. This raises the specter of therapy itself possibly contributing to acquisition of therapy-resistant mutations. Clonal evolution at relapse has also been documented by single nucleotide polymorphism profiling and targeted resequencing [17]. More WGS studies of paired samples in similar or greater detail will be needed to understand if there are specific patterns of clonal evolution for each initial clonotype and if they vary with therapies used. Taking all the genetic variant data together, 2 remarkable features emerge. First, all of the current data suggest that the known genetic variants associated with higher rates of relapse most likely function on the cell fate of the AML cell

P. Vyas et al. / Biol Blood Marrow Transplant 21 (2015) S3eS10

itself (ie, on its quiescence, self-renewal, differentiation, and proliferation properties) rather than how the cell deals with therapy itself. If this is indeed true, it supports the notion that most therapy does not tackle the underlying aberrant biology of the AML cell itself. Second, the detailed mechanisms of why any of the genetic markers associated with high relapse rates (with the possible exception of TP53 mutation) cause therapy failure remain obscure. Thus, a considerable amount of mechanistic work remains to be done to understand therapy failure. Functional Cellular Heterogeneity in AML Seminal studies in AML (reviewed in [18]) have led to the cancer stem cell hypothesis, which postulates that cancers are organized in cellular hierarchies, as are normal tissues. At the hierarchy apex are multipotent, largely quiescent, long-lived cancer stem cells with marked self-renewal capacity that sustain disease and differentiate into “bulk cancer cells.” A key experimental attribute of AML leukemic stem cells (LSC) is that they are the only cell population within AML that can serially propagate disease when transplanted into immunodeficient mice. This observation raised the hypothesis that therapy may not need to eradicate all AML cells, but only the leukemia-propagating AML LSC. Furthermore, it has also been suggested that some experimentally defined AML LSC may be more chemoresistant, as they are more quiescent (being localized to a protective bone niche environment) [19]. However, there are open questions about the biological and clinical importance of experimentally AML LSC. First, only w 50% to 60% of AML samples engraft in immunodeficient mice, suggesting that current murine models are not ideally suited to propagate primary human AML and, thus, detect leukemia-propagating populations. Second, even in those primary samples that engraft, not all the clones seen in a patient are able to engraft (Quek and Vyas, unpublished data and [20]). Despite these important caveats, it is also clear from global gene expression profiling that there are marked differences in RNA composition between LSC and non-LSC populations within a sample, suggesting that there could be functional differences. The clinical relevance of LSC still needs further investigation. Several studies have hinted that LSC populations may be more therapy resistant and, thus, may potentially provide important cellular markers of clinically relevant minimal residual disease (MRD) [21-24]. Furthermore, transcriptional profiling studies have suggested that LSC-associated expression signatures are associated with a poorer outcome [25,26]. As better models of interrogating primary human normal and leukemic blood stem cell function are developed, some of the scientific questions will be addressed. Furthermore, on going large-scale studies to define the value of LSC quantitation as an MRD marker and, thus, potentially important cell populations to therapeutically target, will go some way to help us understand the clinical importance of LSCs in relapse. Finally, over the last 2 years, it has become clear that the initiating mutations in both de novo [27-29] AML and AML secondary to myelodysplasia (MDS) [30] occur in normal hemopoietic stem cells and that this establishes a preleukemic state. It is still unclear how commonly relapse originates not only from a founder leukemic clone or sublcone, but also from preleukemic clones. Studies over the next few years will address this and, thus, establish whether therapy needs to target not only leukemic clones but also preleukemic clones.

S5

CHOOSING A PREPARATIVE REGIMEN FOR PATIENTS UNDERGOING ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANTATION FOR AML A perfect preparative regimen for patients with AML undergoing allogeneic hematopoietic cell transplantation would eliminate leukemia, provide adequate immunosuppression to guarantee engraftment, and have negligible toxicity. Lacking perfection, we are forced to select among a variety of imperfect choices. The following discussion considers various trade-offs between regimens and how these might be influenced by other factors, including graftversus-host disease (GVHD) and patient-specific variables. Approaches to improve current regimens will also be mentioned. Dose Intensity and Disease Eradication Either busulfan (Bu) or total body irradiation (TBI) is a major component of most currently used preparative regimens. With either agent, as doses increases, relapse rates lessen, even at MA doses. A prospective randomized trial performed almost 25 years ago found that increasing TBI dosing from 12 Gy to 15.75 Gy reduced relapse rates from 35% to 12% in patients who underwent transplantation for AML in first remission [31]. Given the supportive care measures available in the 1990s, mortality rates for patients over age 60 or with serious comorbidities treated with MA regimens were presumed to be unacceptably high, and so regimens of reduced intensity that assured engraftment were developed. There have been numerous retrospective single-institution studies, including reports from Seattle, Houston, and Boston, showing that these lower intensity regimens, although effective in allowing older patients and those with comorbidities to safely undergo transplantation, are generally associated with higher relapse rates than seen with more intensive regimens [32-34]. In 2013, the European Group for Blood and Marrow Transplantation reported results of a retrospective analysis involving 878 adults with AML or MDS and found that relapse rate increased from 23% with MA preparative regimens to 39% in RIC transplantations [35]. They further reported that relapse rates were higher in those receiving “minimal” RIC regimens (41%) than in those receiving “intermediate” RIC regimens (22%). Most recently, the Bone Marrow Transplant Clinical Trials Network closed protocol 0901, a prospective randomized comparison of MA versus RIC for patients with AML or MDS, after enrollment of 272 out of a planned 356 patients, because of an increased relapse rate in the reduced-intensity arm. The dose-response curves are not so steep that every study has found a difference, particularly when the intervals in dose intensity are small or when the numbers of patients are limited. For example, in a retrospective study, the Boston group reported little difference in relapse rates (54% versus 50%) when Bu doses were increased from 3.2 mg/kg to 6.4 mg/kg [36]. In a prospective study that was stopped early because of poor accrual, the Dresden group did not find a difference in relapse rates between TBI doses of 8 and 12 Gy [37]. But overall, the data seem convincing that when studied over broad ranges, dose does make a difference. Dose Intensity and Transplantation-related Toxicity In the setting of ablative transplantations, although higher dose TBI is associated with reduced relapse rates, it is also associated with increased nonrelapse mortality (NRM). Similarly, in the retrospective studies cited above, comparing MA to RIC regimens, NRM rates tended to be

S6

P. Vyas et al. / Biol Blood Marrow Transplant 21 (2015) S3eS10

higher with higher dose regimens, especially during the first 100 days after transplantation [31,32,34,35,38]. In general, higher dose regimens have been associated with increased transfusion requirements, increased incidence of idiopathic interstitial pneumonia and sinusoidal obstruction syndrome, and more bacterial infections [39-41]. However, the incidence of viral and fungal disease does not appear to be influenced to a major degree by the intensity of the preparative regimen [41]. Interaction of GVHD Prophylaxis and Preparative Regimens Preclinical studies have demonstrated that more intensive preparative regimens result in a “proinflammatory cytokine milieu” that favors the development of acute GVHD. In fact, the very earliest randomized studies comparing 2 high-dose TBI regimens reported higher rates of acute GVHD with the higher dose regimen [31]. When similar methods of GVHD prophylaxis are used, acute GVHD and NRM are increased with MA compared with RIC regimens [42]. Exactly what proportion of the increased NRM seen after higher dose regimens is due to direct organ damage versus how much is contributed by the effects of GVHD is uncertain. Nonetheless, the general principles seem clear: increasing intensity of the preparative regimen results in more acute GVHD and a higher rate of NRM. The obverse also seems likely; more effective GVHD prophylaxis should result in less NRM and may allow for an escalation in preparative regimen intensity without an increase in NRM. When minimally intensive preparative regimens, such as fludarabine (Flu) 30 mg/m2/ day for 3 days plus 2 Gy TBI (Flu-TBI), are used, neither antithymocyte globulin (ATG) nor T cell depletion are commonly added because of concern for graft rejection. When the intensity of the preparative regimen is increased, such as with the use of Flu 30 mg/m2/days for 5 days plus Bu 4 mg/kg/day for 2 days, ATG can be added without a marked increase in graft rejection (Flu-Bu-ATG). Blaise et al. compared Flu-TBI versus Flu-Bu-ATG and found that the FluBu-ATG regimen was associated with a lower relapse rate, a higher NRM, and equivalent overall survival [43]. That group has since gradually increased the ATG dose from 2.5 mg/kg to 5 mg/kg with a further reduction in acute GVHD and NRM and a seeming improvement in overall survival [44]. Increasing the ATG dose above 5 mg/kg in the setting of RIC appears to result in an increased incidence of disease recurrence, presumably due to loss of a GVL effect, which may negatively impact overall survival [45,46]. There are emerging data that more effective prevention of acute GVHD may allow intensification of the preparative regimen without an increase in NRM. As 1 example, a very intensive preparative regimen used in the setting of T cell depletion includes standard dose cyclophosphamide (Cy) (60 mg/kg  2) along with high-dose TBI (1375 cGY) and thiotepa (5 mg/kg  2). Although no randomized trial has yet been completed, a retrospective comparison found no increase in NRM when this regimen was compared with somewhat lower intensity regimens delivered with conventional GVHD prophylaxis [47]. As in the setting of RIC regimens, whether the increased intensity afforded by improved GVHD control balances a reduction in GVL is unknown. Patient-specific Variables For any given patient, the challenge is choosing the preparative regimen that has the greatest likelihood of eradicating malignancy without causing death or long-term

disability. Age is useful as a starting point but is an imperfect surrogate for a patient’s ability to tolerate intensive therapy. A pretransplantation assessment of mortality score has been developed that provides a more accurate estimate of NRM than age alone [48]. A second tool, the hematopoietic cell transplantationespecific comorbidity index also aids in predicting NRM [49]. Exercises have been conducted in which age and comorbidities have been used to predict NRM after nonmyeloablative, RIC, and MA transplantation, and have found that the combination of age and comorbidities does a better job than either alone, but the predictions are not perfect, suggesting that other factors are important [50]. We and others have found that individual variability in the pharmacokinetics of Bu and Cy are associated with differences in transplantation-related toxicities [51,52]. These differences are partly the result of polymorphisms in genes responsible for drug metabolism and deposition. Likewise, polymorphisms in selected histocompatibility genes have been shown to associate with differences in GVHD risk [53]. Almost certainly, genetically determined differences in other inflammatory and damage response pathways will be found that predict tolerance to currently used preparative regimens. Although current models do a reasonable job of predicting NRM after transplantation, one can easily imagine the creation of better models that include genetic data as they become better established. Nonetheless, the biggest risk most patients undergoing transplantation for AML face is the possibility of disease recurrence. Thus, in choosing among preparative regimens, physicians must weigh the risk of NRM carefully against the burden and biologic behavior of the primary disease [54]. Alternative Agents A number of alternatives to traditional preparative regimens have been developed in the hopes of lessening toxicity without sacrificing antitumor effects. Intravenous Bu reduces some of the pharmacologic variation seen with the oral drug [55]. Flu has been used as a substitute for Cy in a number of variations of the standard Bu-Cy regimens. A retrospective matched-pair analysis found that NRM was somewhat lower with Flu-Bu compared with standard BuCy, but relapse rates were higher, resulting in no overall benefit [56], and the only prospective randomized trial comparing Flu-Bu with Bu-Cy found that overall survival was superior with Bu-Cy [57]. Treosulfan is a prodrug of a bifunctional alkylating agent. Several phase 2 studies have shown that combinations of treosulfan with either Flu alone or Flu plus low-dose TBI are well tolerated and result in excellent overall outcomes [58]. Currently, we and others continue to explore the use of antibody targeted radio-immunotherapy as a way to deliver higher dose radiation therapy to marrow, lymph nodes, and other sites of disease without increasing toxicity to normal organs. The results of phase 2 studies using an I131-labeled antibody reactive with CD45 combined with Flu and TBI have been very encouraging and a phase 3 trial is being initiated [59]. Strategies to Prevent and Treat Relapse After Allograft Disease relapse occurs in 30% to 80% of patients allografted for AML and represents the most common cause of treatment failure [60]. At the same time, the outcome for patients with recurrent disease after allograft remains extremely poor [61,62]. Approaches with the potential to reduce the risk of disease relapse include optimizing the

P. Vyas et al. / Biol Blood Marrow Transplant 21 (2015) S3eS10

antitumor activity of the conditioning regimen without an unacceptable increase in toxicity and the design of strategies that optimize a GVL effect without increasing the risk of GVHD. Identification of Patients at Risk of Relapse Central to the design and rational deployment of strategies with the potential to reduce relapse is a more accurate characterization of patients at risk of disease recurrence. Disease-specific factors associated with an increased risk of relapse include underlying disease biology, at both cytogenetic and molecular levels, and evidence of prior chemoresistance, as demonstrated by failure to achieve a CR after the first or second course of induction chemotherapy [63,64]. More recently, immunophenotypic evidence of residual disease in patients who underwent transplantation in CR has also been shown to be an important factor determining relapse risk [65]. AML is a hierarchical disease originating from a clonally transformed LSC cell and it has been hypothesized that chemoresistant LSC serve as a reservoir of disease in patients destined to relapse after both myelosuppressive chemotherapy and epigenetic therapies [18,23,24]. There is, however, a paucity of clinical studies addressing the impact of stem cell transplantation on the LSC compartment. Such studies will be important to assess whether quantitation of this cellular population assists in risk stratification of transplantation-eligible patients as well as serves as a biomarker of disease-free survival after allogeneic transplantation. Transplantation-specific factors, including both the use of RIC regimens and augmented GVHD prophylaxis strategies such as the use of in vivo T cell depletion, are also associated with an increased risk of disease relapse. An integrated analysis of the combined impact of leukemia-associated somatic mutations and transplantation characteristics on relapse risk in patients allografted for AML will be important, as has recently been performed in patients allografted for MDS [66]. The majority of patients destined to relapse after an allogeneic transplantation for AML will do so within the first year after transplantation but a more precise characterization of the factors determining the kinetics of disease relapse is required if post-transplantation interventions are to be optimally deployed. Optimizing the Conditioning Regimen in High-risk AML and MDS A centrally important approach towards reducing relapse risk is the design of both MA and RIC regimens that possess augmented antitumor activity without a concomitant increase in transplantation toxicity. The ameliorated toxicity of an intravenous formulation of Bu has permitted the refinement of an MA Bu/Cy regimen [67]. At the same time, its combination with the highly immunosuppressive purine analogue, Flu, has resulted in the development of the promising Flu/Bu regimen, both of which are currently under evaluation in prospective trials. The extent to which the antitumor activity of current RIC regimens differ was revealed by a recent randomized trial in which patients with a range of hematologic malignancies who underwent transplantation using a Flu/Bu/ATG regimen demonstrated a lower relapse rate than patients who received a Flu/low-dose TBI regimen [43]. In light of the high relapse rates reported in patients with high-risk AML using current RIC regimens, a number of regimens with augmented antitumor activity

S7

have been designed. Foremost among these is the FLAMSA regimen, developed by the Munich group, that incorporates both intensive pretransplantation cytoreduction and early administration of donor lymphocyte infusions (DLI) [68]. Phase II studies in patients with poor-risk disease, specifically AML in first CR with adverse risk cytogenetics and primary refractory disease, report a reduction in relapse risk in this challenging population [69]. Recent studies suggest that substitution of Bu for low-dose TBI, within the FLAMSA regimen, may further improve outcome and reduce transplantation toxicity, particularly in patients over the age of 60 years. In separate studies, clofarabine, which possesses additional antileukemic activity, has been combined with Bu or melphalan with promising initial results. Encouraging outcomes have also been reported using a combination of Flu, melphalan, and BCNU [70,71]. The long-heralded potential of adjunctive radioimmunotherapy to increase antitumor effect without increasing toxicity is supported by the observation that relapse risk is reduced in patients with very higherisk AML whose conditioning regimen includes radiolabeled 131I-labeled anti-CD45 antibodies [72]. Despite this proliferation of novel conditioning regimens, there are, to date, no reports of the results of randomized trials in highrisk AML patients and these must now be prioritized by the transplantation community. Optimizing a GVL Effect in Patients Undergoing Transplantation for High-risk AML A potent GVL effect is evident in patients undergoing transplantation for AML using either an MA or RIC regimen, as evidenced by the inverse correlation between relapse risk and both the occurrence of chronic GVHD and the intensity of post-transplantation immunosuppression [73-75]. The magnitude of this GVL effect presents a number of opportunities to reduce relapse risk, although the kinetics of disease relapse mandate early intervention. The duration and intensity of post-transplantation immunosuppression represents an important and potentially manipulable determinant of relapse risk. At present, however, the optimal duration and dose of cyclosporin or tacrolimus administration in patients alllografted for high-risk AML remain unknown and there remains considerable heterogeneity in clinical practice. Administration of DLI as a strategy to reduce relapse in patients allografted for high-risk AML Preemptive administration of DLI, guided by conventional MRD assessments or LSC quantitation, represents an attractive strategy by which a GVL effect can be augmented in patients deemed to be at a high risk of disease relapse. Currently, a number of challenges are associated with administration of DLI in high-risk patients undergoing transplantation for AML. Because of the rapid kinetics of disease relapse in AML, DLI must be delivered early, at a time when it is associated with significant GVHD-related morbidity and mortality. This not only compromises its clinical benefit but also limits DLI dosing. In this context, the emergence of biologically targeted drugs, such as tyrosine kinase inhibitors and epigenetic therapies, which exert a potent antitumor effect with limited hematopoietic toxicity, presents an opportunity to manipulate the kinetics of disease relapse permitting the postponement of DLI to a time at which it can be administered with less risk of toxicity. This concept, by which GVHD and GVL might be dissociated, was first developed in the context of chronic myeloid leukemia,

S8

P. Vyas et al. / Biol Blood Marrow Transplant 21 (2015) S3eS10

where the administration of imatinib after transplantation was shown to manipulate the kinetics of disease relapse allowing the postponement of DLI administration [76] to a time when it was associated with less toxicity. Building on this model, the adjunctive administration of FLT3 inhibitors and azacitidine in is of potential interest in patients with AML deemed to be at a high risk of disease relapse after transplantation [77]. Pharmacological manipulation of the alloreactive response after transplantation Pharmacological strategies aimed at augmenting a GVL effect after transplantation have focused either on upregulating the expression of putative target antigens on target cells or manipulation of cellular effector function. DNA methyltransferase (DNMT) inhibitors, such as azacitidine and decitabine, upregulate the expression of both minor histocompatibility antigens and putative tumor antigens, including members of the cancer testis antigen family. In addition, prior exposure to these drugs augments killing of tumor targets by CD8þ T cells [78]. At the same time, azacitidine, possibly through demethylation of the Foxp3 promoter, has the capacity to increase the number of circulating regulatory T cells in murine transplantation models [79,80]. The possibility that epigenetic therapies such as azacitidine might improve transplantation outcome by epigenetically manipulating the alloreactive response has recently been explored in patients undergoing a RIC allograft for high-risk AML/MDS. Azacitidine is both well tolerated after transplantation and appears to both accelerate regulatory T cell reconstitution as well as induce a CD8þ T cell response to candidate tumor antigens, presenting a novel strategy by which both the risk of GVHD and disease relapse can be reduced [81,82]. A number of other biological agents, including histone deacetylase inhibitors and bortezomib, modulate alloreactivity in animal models and present an alternative strategy for improving outcome in patients at high risk of disease relapse after transplantation. Lenalidomide, because of its capacity to activate CD8þ T cells, represents an alternative method of pharmacologically augmenting a GVL response after transplantation. Early phase studies incorporating post-transplantation lenalidomide have been complicated by a high-risk GVHD, confirming its ability to augment an alloreactive response after transplantation [83]. However, given the excessive GVHD-related toxicity associated with early lenalidomide administrationemodified treatment regimens, perhaps utilizing T cell depletion or concurrent azacitidine administration are required. Management of Patients With Disease Relapse Treatment options in patients who relapse after an allogeneic stem cell transplantation for AML remain limited. In patients with no evidence of active GVHD who relapse more than 12 to 18 months after transplantation, a second allograft can be produce long-term disease-free survival in a proportion of patients, although there is no evidence that use of a different stem cell donor is beneficial [84]. DLI also has the capacity to salvage a proportion of patients who have achieved CR after salvage therapy [85]. The optimal salvage regimen remains to be established. Conventional myelosuppressive chemotherapy is associated with significant toxicity and there are emerging data that azacitidine, with or without adjunctive DLI, may be an effective salvage therapy [86].

CONCLUSION Scientific advances in our understanding of the biology of disease relapse after allogeneic stem cell transplantation coupled with insights into how a GVL response be selectively harnessed are pivotal to improving transplantation outcomes in patients allografted for AML. At the same time, the advent of novel conditioning regimens and biological agents with the potential to manipulate the alloreactive response after transplantation present real opportunities to reduce both transplantation toxicity and disease relapse. Integration of biomarkers of response into well-designed early and late phase prospective clinical trials remain the key to improving patient outcome and mandate appropriate investment and support by the wider transplantation community. ACKNOWLEDGMENTS Financial disclosure: P.V. and C.C. have served on advisory panels and received research support from Celgene. Conflict of interest statement: There are no conflicts of interest to report. REFERENCES 1. Burnett A, Wetzler M, Lowenberg B. Therapeutic advances in acute myeloid leukemia. J Clin Oncol. 2011;29:487-494. 2. Dohner H, Estey EH, Amadori S, et al. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European LeukemiaNet. Blood. 2010;115:453-574. 3. Burnett AK, Russell NH, Hills RK, et al. Addition of gemtuzumab ozogamicin to induction chemotherapy improves survival in older patients with acute myeloid leukemia. J Clin Oncol. 2012;30:3924-3931. 4. Paschka P, Marcucci G, Ruppert AS, et al. Adverse prognostic significance of KIT mutations in adult acute myeloid leukemia with inv(16) and t(8;21): a Cancer and Leukemia Group B Study. J Clin Oncol. 2006; 24:3904-3911. 5. Cairoli R, Beghini A, Grillo G, et al. Prognostic impact of c-KIT mutations in core binding factor leukemias: an Italian retrospective study. Blood. 2006;107:3463-3468. 6. Gaidzik VI, Bullinger L, Schlenk RF, et al. RUNX1 mutations in acute myeloid leukemia: results from a comprehensive genetic and clinical analysis from the AML study group. J Clin Oncol. 2011;29:1364-1372. 7. Mendler JH, Maharry K, Radmacher MD, et al. RUNX1 mutations are associated with poor outcome in younger and older patients with cytogenetically normal acute myeloid leukemia and with distinct gene and MicroRNA expression signatures. J Clin Oncol. 2012;30:3109-3118. 8. Metzeler KH, Maharry K, Radmacher MD, et al. TET2 mutations improve the new European LeukemiaNet risk classification of acute myeloid leukemia: a Cancer and Leukemia Group B study. J Clin Oncol. 2011;29:1373-1381. 9. Gaidzik VI, Paschka P, Spath D, et al. TET2 mutations in acute myeloid leukemia (AML): results from a comprehensive genetic and clinical analysis of the AML study group. J Clin Oncol. 2012;30:1350-1357. 10. Cornelissen JJ, Gratwohl A, Schlenk RF, et al. The European LeukemiaNet AML Working Party consensus statement on allogeneic HSCT for patients with AML in remission: an integrated-risk adapted approach. Nat Rev Clin Oncol. 2012;9:579-590. 11. Cancer Genome Atlas Research N. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med. 2013;368: 2059-2074. 12. Supek F, Minana B, Valcarcel J, et al. Synonymous mutations frequently act as driver mutations in human cancers. Cell. 2014;156:1324-1335. 13. Groschel S, Sanders MA, Hoogenboezem R, et al. A single oncogenic enhancer rearrangement causes concomitant EVI1 and GATA2 deregulation in leukemia. Cell. 2014;157:369-381. 14. Patel JP, Gonen M, Figueroa ME, et al. Prognostic relevance of integrated genetic profiling in acute myeloid leukemia. N Engl J Med. 2012; 366:1079-1089. 15. Hughes AE, Magrini V, Demeter R, et al. Clonal architecture of secondary acute myeloid leukemia defined by single-cell sequencing. PLoS Genet. 2014;10:e1004462. 16. Ding L, Ley TJ, Larson DE, et al. Clonal evolution in relapsed acute myeloid leukaemia revealed by whole-genome sequencing. Nature. 2012;481:506-510. 17. Kronke J, Bullinger L, Teleanu V, et al. Clonal evolution in relapsed NPM1-mutated acute myeloid leukemia. Blood. 2013;122:100-108. 18. Goardon N, Marchi E, Atzberger A, et al. Coexistence of LMPP-like and GMP-like leukemia stem cells in acute myeloid leukemia. Cancer Cell. 2011;19:138-152.

P. Vyas et al. / Biol Blood Marrow Transplant 21 (2015) S3eS10

19. Ishikawa F, Yoshida S, Saito Y, et al. Chemotherapy-resistant human AML stem cells home to and engraft within the bone-marrow endosteal region. Nat Biotechnol. 2007;25:1315-1321. 20. Klco JM, Spencer DH, Miller CA, et al. Functional heterogeneity of genetically defined subclones in acute myeloid leukemia. Cancer Cell. 2014;25:379-392. 21. van Rhenen A, Feller N, Kelder A, et al. High stem cell frequency in acute myeloid leukemia at diagnosis predicts high minimal residual disease and poor survival. Clin Cancer Res. 2005;11:6520-6527. 22. van Rhenen A, Moshaver B, Kelder A, et al. Aberrant marker expression patterns on the CD34þCD38- stem cell compartment in acute myeloid leukemia allows to distinguish the malignant from the normal stem cell compartment both at diagnosis and in remission. Leukemia. 2007; 21:1700-1707. 23. Gerber JM, Smith BD, Ngwang B, et al. A clinically relevant population of leukemic CD34(þ)CD38(
) cells in acute myeloid leukemia. Blood. 2012;119:3571-3577. 24. Craddock C, Quek L, Goardon N, et al. Azacitidine fails to eradicate leukemic stem/progenitor cell populations in patients with acute myeloid leukemia and myelodysplasia. Leukemia. 2013;27:1028-1036. 25. Gentles AJ, Plevritis SK, Majeti R, Alizadeh AA. Association of a leukemic stem cell gene expression signature with clinical outcomes in acute myeloid leukemia. JAMA. 2010;304:2706-2715. 26. Eppert K, Takenaka K, Lechman ER, et al. Stem cell gene expression programs influence clinical outcome in human leukemia. Nat Med. 2011;17:1086-1093. 27. Jan M, Snyder TM, Corces-Zimmerman MR, et al. Clonal evolution of preleukemic hematopoietic stem cells precedes human acute myeloid leukemia. Sci Transl Med. 2012;4:149ra18. 28. Corces-Zimmerman MR, Hong WJ, Weissman IL, et al. Preleukemic mutations in human acute myeloid leukemia affect epigenetic regulators and persist in remission. Proc Natl Acad Sci U S A. 2014;111:2548-2553. 29. Shlush LI, Zandi S, Mitchell A, et al. Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia. Nature. 2014;506: 328-333. 30. Woll PS, Kjallquist U, Chowdhury O, et al. Myelodysplastic syndromes are propagated by rare and distinct human cancer stem cells in vivo. Cancer Cell. 2014;25:794-808. 31. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: a randomized trial of two irradiation regimens. Blood. 1990;76: 1867-1871. 32. Appelbaum FR. Dose intensity of preparative regimens for acute myeloid leukemia - one-size-fits-all or tailor-made? Best Pract Res Clin Hematol. 2010;23:509-517. 33. Alyea EP, Kim HT, Ho V, et al. Comparative outcome of nonmyeloablative and myeloablative allogeneic hematopoietic cell transplantation for patients older than 50 years of age. Blood. 2005;105: 1810-1814. 34. de Lima M, Anagnostopoulos A, Munsell M, et al. Nonablative versus reduced-intensity conditioning regimens in the treatment 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. 35. Martino R, de Wreede L, Fiocco M, et al. Comparison of conditioning regimens of various intensities for allogeneic hematopoietic SCT using HLA-identical sibling donors in AML and MDS with <10% BM blasts: a report from EBMT. Bone Marrow Transplant. 2013;48:761-770. 36. Chen YB, Coughlin E, Kennedy KF, et al. Busulfan dose intensity and outcomes in reduced-intensity allogeneic peripheral blood stem cell transplantation for myelodysplastic syndrome or acute myeloid leukemia. Biol Blood Marrow Transplant. 2013;19:981-987. 37. Bornhauser M, Kienast J, Trenschel R, et al. Reduced-intensity conditioning versus standard conditioning before allogeneic haemopoietic cell transplantation in patients with acute myeloid leukaemia in first complete remission: a prospective, open-label randomised phase 3 trial. Lancet Oncol. 2012;13:1035-1044. 38. Alyea EP, Kim HT, Ho V, et al. Impact of conditioning regimen intensity on outcome of allogeneic hematopoietic cell transplantation for advanced acute myelogenous leukemia and myelodysplastic syndrome. Biol Blood Marrow Transplant. 2006;12:1047-1055. 39. Weissinger F, Sandmaier BM, Maloney DG, et al. Decreased transfusion requirements for patients receiving nonmyeloablative compared with conventional peripheral blood stem cell transplants from HLA-identical siblings. Blood. 2001;98:3584-3588. 40. Fukuda T, Hackman RC, Guthrie KA, et al. Risks and outcomes of idiopathic pneumonia syndrome after nonmyeloab lative and conventional conditioning regimens for allogeneic hematopoietic stem cell transplantation. Blood. 2003;102:2777-2785. 41. Junghanss C, Marr KA, Carter RA, et al. Incidence and outcome of bacterial and fungal infections following nonmyeloablative compared with myeloablative allogeneic hematopoietic stem cell transplantation: a matched control study. Biol Blood Marrow Transplant. 2002;8:512-520.

S9

42. Mielcarek M, Martin PJ, Leisenring W, et al. Graft-versus-host disease after nonmyeloablative versus conventional hematopoietic stem cell transplantation. Blood. 2003;102:756-762. 43. Blaise D, Tabrizi R, Boher JM, et al. Randomized study of 2 reducedintensity conditioning strategies for human leukocyte antigenmatched, related allogeneic peripheral blood stem cell transplantation: prospective clinical and socioeconomic evaluation. Cancer. 2013;119: 602-611. 44. Blaise D, Castagna L. Do different conditioning regimens really make a difference? Hematology Am Soc Hematol Educ Program. 2012;2012: 237-245. 45. Appelbaum FR, Bacigalupo A, Soiffer R. Anti-T cell antibodies as part of the preparative regimen in hematopoietic cell transplantationea debate. Biol Blood Marrow Transplant. 2012;18:S111-115. 46. Soiffer RJ, Lerademacher J, Ho V, et al. Impact of immune modulation with anti-T-cell antibodies on the outcome of reduced-intensity allogeneic hematopoietic stem cell transplantation for hematologic malignancies. Blood. 2011;117:6963-6970. 47. Pasquini MC, Devine S, Mendizabal A, et al. Comparative outcomes of donor graft CD34þ selection and immune suppressive therapy as graftversus-host disease prophylaxis for patients with acute myeloid leukemia in complete remission undergoing HLA-matched sibling allogeneic hematopoietic cell transplantation. J Clin Oncol. 2012;30:3194-3201. 48. Parimon T, Au DH, Martin PJ, Chien JW. A risk score for mortality after allogeneic hematopoietic cell transplantation. Ann Intern Med. 2006; 144:407-414. 49. Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood. 2005;106:2912-2919. 50. Sorror ML, Storb RF, Sandmaier BM, et al. Comorbidity-age index: a clinical measure of biologic age before allogeneic hematopoietic cell transplantation. J Clin Oncol. 2014;32:3249-3256. 51. McCune JS, Batchelder A, Guthrie KA, et al. Personalized dosing of cyclophosphamide in the total body irradiation-cyclophosphamide conditioning regimen: a phase II trial in patients with hematologic malignancy. Clin Pharmacol Ther. 2009;85:615-622. 52. Slattery JT, Clift RA, Buckner CD, et al. Marrow transplantation for chronic myeloid leukemia: the influence of plasma busulfan levels on the outcome of transplantation. Blood. 1997;89:3055-3060. 53. Petersdorf EW, Malkki M, Gooley TA, et al. MHC-resident variation affects risks after unrelated donor hematopoietic cell transplantation. Sci Transl Med. 2012;4:144ra01. 54. Sorror ML, Appelbaum FR. Risk assessment before allogeneic hematopoietic cell transplantation for older adults with acute myeloid leukemia. Expert Rev Hematol. 2013;6:547-562. 55. Ciurea SO, Andersson BS. Busulfan in hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2009;15:523-536. 56. Bredeson CN, Zhang MJ, Agovi MA, et al. Outcomes following HSCT using fludarabine, busulfan, and thymoglobulin: a matched comparison to allogeneic transplants conditioned with busulfan and cyclophosphamide. Biol Blood Marrow Transplant. 2008;14:993-1003. 57. Lee JH, Joo YD, Kim H, et al. Randomized trial of myeloablative conditioning regimens: busulfan plus cyclophosphamide versus busulfan plus fludarabine. J Clin Oncol. 2013;31:701-709. 58. Gyurkocza B, Gutman J, Nemecek ER, et al. Treosulfan, fludarabine, and 2-Gy total body irradiation followed by allogeneic hematopoietic cell transplantation in patients with myelodysplastic syndrome and acute myeloid leukemia. Biol Blood Marrow Transplant. 2014;20:549-555. 59. Pagel JM, Gooley TA, Rajendran J, et al. Allogeneic hematopoietic cell transplantation after conditioning with 131I-anti-CD45 antibody plus fludarabine and low-dose total body irradiation for elderly patients with advanced acute myeloid leukemia or high-risk myelodysplastic syndrome. Blood. 2009;114:5444-5453. 60. Appelbaum FR. Optimising the conditioning regimen for acute myeloid leukaemia. Best Pract Res Clin Hematol. 2009;22:543-550. 61. Schmid C, Labopin M, Nagler A, et al. Treatment, risk factors, and outcome of adults with relapsed AML after reduced intensity conditioning for allogeneic stem cell transplantation. Blood. 2012;119: 1599-1606. 62. Bejanyan N, Oran B, Shanley R, et al. Clinical outcomes of AML patients relapsing after matched-related donor and umbilical cord blood transplantation. Bone Marrow Transplant. 2014;49:1029-1035. 63. Craddock C, Labopin M, Pillai S, et al. Factors predicting outcome after unrelated donor stem cell transplantation in primary refractory acute myeloid leukaemia. Leukemia. 2011;25:808-813. 64. Walter RB, Sandmaier BM, Storer BE, et al. Number of courses of induction therapy independently predicts outcome after allogeneic transplantation for AML in first morphological remission. Biol Blood Marrow Transplant. 2014. http://dx.doi.org/10.1016/ j.bbmt.2014.09.022 [Epub ahead of print]. 65. Walter RB, Gooley TA, Wood BL, et al. Impact of pretransplantation minimal residual disease, as detected by multiparametric flow cytometry, on outcome of myeloablative hematopoietic cell transplantation for acute myeloid leukemia. J Clin Oncol. 2011;29:1190-1197.

S10

P. Vyas et al. / Biol Blood Marrow Transplant 21 (2015) S3eS10

66. Bejar R, Stevenson KE, Caughey B, et al. Somatic mutations predict poor outcome in patients with myelodysplastic syndrome after hematopoietic stem-cell transplantation. J Clin Oncol. 2014;32: 2691-2698. 67. Sobecks RM, Rybicki L, Yurch M, et al. Intravenous compared with oral busulfan as preparation for allogeneic hematopoietic progenitor cell transplantation for AML and MDS. Bone Marrow Transplant. 2012;47: 633-638. 68. Schmid C, Schleuning M, Ledderose G, et al. Sequential regimen of chemotherapy, reduced-intensity conditioning for allogeneic stem-cell transplantation, and prophylactic donor lymphocyte transfusion in high-risk acute myeloid leukemia and myelodysplastic syndrome. J Clin Oncol. 2005;23:5675-5687. 69. Schmid C, Schleuning M, Schwerdtfeger R, et al. Long-term survival in refractory acute myeloid leukemia after sequential treatment with chemotherapy and reduced-intensity conditioning for allogeneic stem cell transplantation. Blood. 2006;108:1092-1099. 70. Bertz H, Potthoff K, Finke J. Allogeneic stem-cell transplantation from related and unrelated donors in older patients with myeloid leukemia. J Clin Oncol. 2003;21:1480-1484. 71. Chevallier P, Labopin M, Socie G, et al. Results from a clofarabinebusulfan-containing, reduced-toxicity conditioning regimen prior to allogeneic stem cell transplantation: the phase 2 prospective CLORIC trial. Haematologica. 2014;99:1486-1491. 72. Mawad R, Gooley TA, Rajendran JG, et al. Radiolabeled anti-CD45 antibody with reduced-intensity conditioning and allogeneic transplantation for younger patients with advanced acute myeloid leukemia or myelodysplastic syndrome. Biol Blood Marrow Transplant. 2014;20: 1363-1368. 73. 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 undergoing allogeneic bone marrow transplantation. Br J Haematol. 2000;108:99-104. 74. Baron F, Labopin M, Niederwieser D, et al. Impact of graft-versus-host disease after reduced-intensity conditioning allogeneic stem cell transplantation for acute myeloid leukemia: a report from the Acute Leukemia Working Party of the European group for blood and marrow transplantation. Leukemia. 2012;26:2462-2468. 75. Craddock C, Nagra S, Peniket A, et al. Factors predicting long-term survival after T-cell depleted reduced intensity allogeneic stem cell transplantation for acute myeloid leukemia. Haematologica. 2010;95: 989-995.

76. Olavarria E, Siddique S, Griffiths MJ, et al. Posttransplantation imatinib as a strategy to postpone the requirement for immunotherapy in patients undergoing reduced-intensity allografts for chronic myeloid leukemia. Blood. 2007;110:4614-4617. 77. Platzbecker U, Wermke M, Radke J, et al. Azacitidine for treatment of imminent relapse in MDS or AML patients after allogeneic HSCT: results of the RELAZA trial. Leukemia. 2012;26:381-389. 78. Goodyear O, Agathanggelou A, Novitzky-Basso I, et al. Induction of a CD8þ T-cell response to the MAGE cancer testis antigen by combined treatment with azacitidine and sodium valproate in patients with acute myeloid leukemia and myelodysplasia. Blood. 2010;116: 1908-1918. 79. Choi J, Ritchey J, Prior JL, et al. In vivo administration of hypomethylating agents mitigate graft-versus-host disease without sacrificing graftversus-leukemia. Blood. 2010;116:129-139. 80. Schroeder T, Frobel J, Cadeddu RP, et al. Salvage therapy with azacitidine increases regulatory T cells in peripheral blood of patients with AML or MDS and early relapse after allogeneic blood stem cell transplantation. Leukemia. 2013;27:1910-1913. 81. Goodyear OC, Dennis M, Jilani NY, et al. Azacitidine augments expansion of regulatory T cells after allogeneic stem cell transplantation in patients with acute myeloid leukemia (AML). Blood. 2012;119:3361-3369. 82. Mohty M, Chevallier P. Azacitidine after allo-SCT: the good without the bad? Blood. 2012;119:3199-3200. 83. Sockel K, Bornhaeuser M, Mischak-Weissinger E, et al. Lenalidomide maintenance after allogeneic HSCT seems to trigger acute graft-versushost disease in patients with high-risk myelodysplastic syndromes or acute myeloid leukemia and del(5q): results of the LENAMAINT trial. Haematologica. 2012;97:e34-e35. 84. Christopeit M, Kuss O, Finke J, et al. Second allograft for hematologic relapse of acute leukemia after first allogeneic stem-cell transplantation from related and unrelated donors: the role of donor change. J Clin Oncol. 2013;31:3259-3271. 85. Schmid C, Labopin M, Nagler A, et al. Donor lymphocyte infusion in the treatment of first hematological relapse after allogeneic stem-cell transplantation in adults with acute myeloid leukemia: a retrospective risk factors analysis and comparison with other strategies by the EBMT Acute Leukemia Working Party. J Clin Oncol. 2007;25:4938-4945. 86. Schroeder T, Czibere A, Platzbecker U, et al. Azacitidine and donor lymphocyte infusions as first salvage therapy for relapse of AML or MDS after allogeneic stem cell transplantation. Leukemia. 2013;27: 1229-1235.