Acute leukemia in adults: novel allogeneic transplant strategies

Abstract

Acute myeloid leukemia (AML) is the most common indication for hematopoietic cell transplantation (HCT). Significant improvements in histocompatibility testing have resulted in identifying better alternative donors to provide grafts; coupled with expanding use of reduced-intensity conditioning (RIC) regimens, patient outcomes have improved. Most AML patients in complete remission now can undergo allogeneic HCT procedures with an anticipated result that approaches that obtained with sibling donors and myeloablative conditioning (MAC). Treatment-related mortality (TRM), relapse, GVHD and, for umbilical cord blood (UCB) grafts, slow engraftment/engraftment failures, however, continue to plague HCT. Newer strategies include the use of double unit UCB HCT and ex vivo expansion of UCB units using: multipotent mesenchymal stromal cells (MSCs); polyamine copper chelator tetraethylenepentamine (TEPA); and Notch ligand. Haploidentical bone marrow and blood HCT are increasing in numbers given the improved outcome with post-HCT cyclophosphamide therapy. Novel chemotherapeutic preparative regimens incorporating agents such as clofarabine and treosulfan, and novel radiation preparative regimens utilizing selective radiation approaches (helical tomotherapy) and targeted myeloablative radioimmunotherapy appear to enhance the anti-AML effect. Immunological approaches with chimeric antigen receptor-redirected T lymphocytes have been added to the armamentarium. Other novel maneuvers include more sophisticated T-cell depletion of the donor graft and post-transplant immunotherapy with regulatory T cell (Treg) therapy and MSCs for GVHD prevention. These and other strategies are improving the outlook of AML HCT recipients.

Keywords:

• Acute leukemia
• Allogeneic hematopoietic cell transplantation
• Treatment-related mortality

Introduction

World-wide, hematopoietic cell transplantation (HCT) is regularly employed as a curative treatment option for patients with various disorders, including acute leukemia. Currently, acute myeloid leukemia (AML) is the most common indication for HCT. Despite the many advances in this life-saving art, HCT remains plagued by the risk of relapse or failure due to graft-versus-host disease (GVHD), infectious complications, and treatment-related mortality (TRM). Some of the emerging approaches that may improve overall patient outcome are reviewed herein.

Graft Sources

Expanding numbers of normal volunteers are included in world-wide registries and umbilical cord blood (UCB) banks are proliferating. Further, high-resolution DNA matching between donor and recipient for HLA-A,-B, -C, and -DRB1 alleles is associated with higher rates of survival. As a result, the past decade has seen the emergence of a significant increase in the use of alternative donors as shown in Table 1.1

Table 1. Graft sources for allogeneic HCT and percentage of all transplants

Donor type	Match probability	Percentage of all transplants
HLA-compatible sibling	30%	54
HLA-compatible unrelated donor	35%–80%	38
Umbilical cord blood	10–90%	5
Haploidentical donor	99%	<5

Blood versus marrow

The existence of multipotential hematopoietic cells in the blood has been recognized for many years. With the development of continuous-flow apheresis instruments and the widespread implementation of hematopoietic growth factor mobilization treatment, stimulated blood collections could contain numerous hematopoietic progenitors to be used in HCT. As a result, blood rapidly replaced bone marrow as the graft source of choice. Data from CIBMTR show that in 2008, only 12% of the related donor grafts and 14% of the unrelated donor grafts were derived from bone marrow as opposed to peripheral blood. Newer mobilization strategies include the use of plerixafor, a novel small-molecule CXCR4 chemokine antagonist that can enhance collection of hematopoietic progenitor cells from volunteer donors.

UCB

UCB continues to increase as an alternative source of hematopoietic progenitor cells as this product is relatively easy to procure and store and these grafts possess a low likelihood of transmitting infections. Further, for a given degree of HLA match, UCB grafts are associated with a reduced degree of GVHD when compared to a matched unrelated donor graft; further, UCB grafts retain the graft-versus-leukemia effect. Finally, no donor isohemagglutinins are produced in an ABO-incompatible UCB HCT. The relatively low cell dose content, however, contributes to the slower engraftment and elevated risk of engraftment failure. Strategies to overcome this limitation are discussed next.

Double unit UCB HCT

Engraftment delays and failure rates remain higher compared to bone marrow or mobilized blood sources even with infusion of two UCB units. Furthermore, investigators still cannot explain why only one UCB unit ultimately will predominate. On the other hand, relapse rates appear lower with two compared to one unit UCB HCT; the non-engrafting unit may enhance immune activation.2

Ex vivo UCB unit expansion: multipotent mesenchymal stromal cells (MSCs)

de Lima and co-workers at the MD Anderson Cancer Center3 ex vivo co-cultured one of two UCB units infused in the course of a double UCB HCT using either third-party haploidentical family member, marrow-derived MSCs, or commercially available mesenchymal progenitor cells (Angioblast Systems, Inc., New York, USA). The culture increased total nucleated cells 14-fold and CD34⁺ cells 40-fold. Thirty-two patients given myeloablative conditioning (MAC) had median neutrophil and platelet engraftment at 15 and 40 days, respectively. The expanded unit contributed only to early blood cell recovery; long-term engraftment was provided by the unexpanded unit.

Ex vivo UCB unit expansion: polyamine copper chelator tetraethylenepentamine (TEPA)

In a pre-clinical model, TEPA was shown to augment long-term ex vivo expansion of UCB-derived CD34⁺ cells and increased their engraftment potential. The MD Anderson Cancer Center investigators4 treated 10 high-risk hematologic malignancy patients with cytotoxic therapy followed by infusion of a low cell dose UCB unit that had been cryopreserved in two fractions. One fraction was administered unexpanded, while the other component was CD133-selected and expanded in vitro in media containing TEPA. The total nucleated cell dose was expanded 219-fold and the CD34⁺ cell content increased six-fold. Despite the infusion of low cell numbers in the starting materials, 9 of 10 patients engrafted at median times comparable to those associated with infusion of unmanipulated single UCB units, e.g. neutrophils at 30 days and platelets at 48 days.

Ex vivo UCB unit expansion: Notch ligand

Investigators at the Fred Hutchinson Cancer Research Center treated 10 high-risk acute leukemia patients using a MAC regimen followed by one non-manipulated UCB and one Notch ligand Delta1 supplemented media ex vivo expanded UCB graft.5 CD34⁺ cells expanded 164-fold and median time to neutrophils >500/μl was 16 days, numbers comparable to a concurrently treated cohort of 20 patients undergoing double UCB transplantation.

Haploidentical bone marrow

This approach has been hampered historically by a high rate of graft rejection, slow immune recovery, and substantial TRM, often due to opportunistic infection. The group at Johns Hopkins has extensive experience infusing unmodified haploidentical marrow followed by post-transplantation treatment with cyclophosphamide.6 They speculated that post-HCT chemotherapy eliminates infused donor alloreactive T cells during their early proliferative phase. Severe acute and chronic GVHD and TRM rates were lowered, but engraftment failure rates remained high.

The Blood and Marrow Transplant Clinical Trials Network (BMT CTN) reported parallel, multi-center, phase II reduced-intensity conditioning (RIC) trials for leukemia and lymphoma patients who lacked suitable related donors7 using either double UCB, or HLA-haploidentical related donor grafts. At 1 year, overall and progression-free survivals were 54% and 46%, respectively, after double UCB transplantation (n = 50) and 62% and 48%, respectively, after haploidentical marrow transplantation (n = 50). Neutrophil recoveries, incidence of grade II–IV acute GVHD, non-relapse mortality, and relapse incidence at 1 year after double UCB transplantation were similar. These studies demonstrate that almost every patient can count on an adequately matched UCB or haploidentical donor, but further work is needed to directly compare these approaches and optimize their outcome.

Paradigms in Conditioning

Preparative regimen intensity

With greater understanding of patient tolerance to therapy, three categories of preparative regimen intensity have been identified: MAC, RIC, and non-myeloablative conditioning regimens. Retrospective data show no superiority of one regimen intensity and most centers rely upon the conditioning with which they have generated the most experiences. Survival in patients older than 55 years is comparable to younger patients after RIC HCT with no significant differences in non-relapse mortality and overall survival. As expected, the data show significantly lower 1-year infection-related mortality after RIC compared to MAC. The field awaits the completion of prospective trials comparing regimens of differing intensity.

Novel chemotherapeutic preparative regimens

Clofarabine-based

Clofarabine, a new potent purine nucleoside antimetabolite that inhibits DNA polymerase and ribonucleotide reductase, is synergistic with cytarabine and has significant activity in acute leukemia. Agura and co-workers8 described a single-institution phase II clofarabine and busulfan RIC regimen in 20 hematologic malignancy (n = 16 AML) patients. Eleven subjects with active malignancy attained complete response (CR) by day 30; seven experienced relapse and overall 12 died but six remain in remission at a median follow-up of 31 months. Kirshbaum and colleagues9 performed a phase I dose escalation study of clofarabine and melphalan in 16 AML patients (median age: 63 years); 11 remain in remission at a median of 23 months after HCT.

Treosulphan-based

Treosulfan is pharmacologically similar to busulfan but more potent in vitro; unlike busulfan, it does not require dose adjustment and has been incorporated into HCT several regimens.10 A RIC trial from France11 combined treosulfan 12 mg/m²/day for 3 days with fludarabine and anti-thymocyte globulin in 56 hematological malignancy patients. Overall survival was not reached at a median follow-up of 13 months with a 52% 3-year probability of survival. Nemecek and colleagues12 treated 60 patients (n = 44 AML) at high-risk for relapse or TRM using treosulfan 12–14 mg/m²/day for 3 days and fludarabine followed by HCT from HLA-identical siblings (n = 30) or unrelated donors (n = 30). Two-year relapse-free survival for all patients was 58% and was 88% for patients without high-risk cytogenetics.

Novel radiation preparative regimens

Selective radiation approaches

While total body irradiation (TBI) is used to prevent immunologic graft rejection and can eradicate residual malignant cells, toxic effects may be significant. Helical tomotherapy is a modality that integrates CT image-guided radiotherapy and intensity-modulated radiation therapy that can deliver conformal targeted radiation to the marrow and lymphatic systems. Rosenthal et al.13 recently described the results of a phase I–II investigation using 1200 cGy total marrow and lymph node irradiation to augment RIC transplantation for 33 advanced, poor-risk hematologic malignancy patients. At 1 year after HCT, overall survival, event-free survival, and non-relapse-related mortality rates were 75%, 65%, and 19%, respectively. The addition of total marrow and lymph node irradiation to RIC is feasible and safe. Those centers equipped with this device can offer it to patients with advanced hematological malignancies who might not otherwise be candidates for RIC.

Targeted myeloablative radioimmunotherapy

One alternative strategy for acute leukemia patients undergoing HCT is myeloablative radioimmunotherapy, a technique that targets radiotherapy to the bone marrow using radio-labeled monoclonal antibodies. The cross-firing beta-particles create a field radiation effect, potentially killing antigen-negative resident marrow tumor cells. Candidate isotopes usually are the beta-emitter radio-nuclides (¹³¹I, ⁹⁰Y, ¹⁸⁸Re) which are conjugated to monoclonal antibodies including anti-CD33, anti-CD45, and anti-CD66. Investigators now are considering use of alpha-particle emitters (helium nuclei) such as ²¹³Bi, ²²⁵Ac, and ²¹¹At; this approach could result in less non-specific toxicity to normal bystander cells and provide more efficient single-cell killing than beta-emitting constructs.

Pre-targeted myeloablative radioimmunotherapy (PRIT)

PRIT allows for maximal antibody targeting to take place before delivery of the therapeutic radionuclide while maintaining targeting specificity and substantially reduces whole-body radiation. Synthetic chasers (‘clearing agents’) have been introduced as an additional refinement to PRIT. Presently, this technique has not yet been implemented in clinical practice.

GVHD Prevention Strategies

T-cell depletion of the graft

For a long time, T-cell depletion of the graft has been recognized to reduced GVHD, but at the expense of increased relapse and engraftment failure rates. Recently, Devine and colleagues14 for the BMT CTN performed a phase II single-arm multi-center study in 44 adult AML CR1 (n = 37) and CR2 (n = 7) with MAC/fractionated TBI (1375 cGy) and immune-magnetically selected CD34-enriched, T cell-depleted allografts from HLA-identical siblings (without pharmacologic GVHD prophylaxis). The incidence of grade II–IV acute GVHD was 23%; extensive chronic GVHD at 24 months was 7%. Disease-free survival for CR1 patients was 73% at 12 months and 58% at 36 months.

T regulatory cell (Treg) therapy for GVHD prevention

Pre-clinical models suggested that donor Tregs reliably can suppress GVHD and possibly spare leukemia-specific immune responses. Brunstein and associates15 reported the ‘first-in-human’ clinical trial of ex vivo expansion of CD4⁺CD25⁺FoxP3⁺ T regulatory cell enrichment obtained from cryopreserved UCB. Twenty-three patients (median age: 52 years) were given chemotherapy, single fraction TBI (200 cGy), and double UCB HCT followed by the expanded cell population. When compared with 108 identically-treated historic controls who did not receive Treg infusions, the incidence of grade II–IV acute GVHD was reduced from 61% to 43% (P = 0·05). Di Ianni and the Perugia group16 infused Tregs early followed by conventional T cells in 28 high-risk hematologic malignancy patients (n = 22 AML) who underwent HLA-haploidentical HCT after chemotherapy and single-fraction TBI (800 cGy). This group showed for the first time in humans that adoptive transfer of Tregs prevented GVHD in the absence of post-transplantation immunosuppression. Twelve patients remain alive at a median 12 months after haploidentical HCT.

Post-transplant maneuvers

Donor lymphocyte infusion

Lutz and co-workers17 gave donor lymphocyte infusion in prophylactic fashion to 26 of 85 ALL allograft recipients. Twelve of 13 patients who had mixed chimerism converted to complete donor chimerism; 18 of 26 patients developed GVHD and are alive at a median follow-up of 42 months after HCT. The majority of patients, however, are ineligible for this strategy because of onset of GVHD or disease progression after HCT.

Chimeric antigen receptor-redirected T lymphocytes

To redirect specificity, investigators now are using chimeric antigen receptor T cells that combine antigen-specificity and T-cell activating properties into a single fusion molecule. Receptors encompassing both CD28 and CD3 now are rapidly expanding to a diverse array of receptors with different functional properties. A few highly committed and experienced centers are incorporating these approaches into the HCT post-transplant therapy in select patients at high-risk for relapse.

Ancillary Approaches

MSCs

MSCs are cells that secrete hematopoietic colony-stimulating factors and cytokines yet also exhibit immunosuppressive properties. The major advantages of using cellular therapies such as MSCs for GVHD prophylaxis or therapy are that their immunosuppressive properties appear to be selective and may not lead to an increase in opportunistic infections. Lazarus et al.18 co-transplanted marrow-derived, culture-expanded MSCs along with hematopoietic progenitor cells obtained from HLA-identical sibling donors after MAC therapy in a GVHD prophylaxis trial. The incidence of both acute and chronic GVHD appeared to be reduced in this 46 patient study. In a landmark case report, Le Blanc and colleagues19 described the successful use of therapy third-party MSCs obtained from the haploidentical mother of a 9-year-old boy who had treatment-resistant grade IV acute GVHD in the course of an HLA-matched unrelated donor transplant. Subsequent trials, however, have yielded mixed results and use of MSCs cannot be recommended as part of routine clinical care.

Enhancing homing and engraftment capabilities of hematopoietic cells

Based on pre-clinical models, several groups are exploring CD26/DPPIV inhibition to enhance ability of cells to home more efficiently to marrow. Clinical trials testing feasibility enhance engrafting capability in UCB transplants are about to begin.

Conclusions

The standard of care in HCT for hematologic malignancies previously has been established based on therapies undertaken on a relative uncommon population of young patients with minimal comorbidities who were able to withstand the challenges of using a MAC modality using a sibling-matched donor as the usual graft source. The majority of the AML population, however, is elderly, likely have multiple comorbidities and generally do not have an available sibling-matched donor. Significant improvements in histocompatibility testing have resulted in identifying better alternative donors to provide grafts and patient outcomes. Coupled with expanding use of RIC and non-myeloablative conditioning regimens, most AML patients in complete remission now can undergo allogeneic HCT procedures with an anticipated result that approaches that obtained with sibling donors and MAC. TRM, relapse, and GVHD, however, continue to plague HCT. Investigators are placing a greater emphasis on pursuing rational, targeted approaches. These many options appear feasible and include as ex vivo-expanded grafts, targeted and novel preparative regimens that with radioimmunotherapy, chimeric antigen receptors, Treg enhancement, and MSCs as well as post-HCT maneuvers. Finding the optimal individualized strategy for patients, however, depends on the development of well-designed, randomized, multi-center prospective trials. Enrollment of patients into clinical trials cannot be overemphasized.