Recent developments in HLA-haploidentical transplantations

Abstract

While allogeneic hematopoietic stem cell transplantations have a curative potential, several patients with hematologic malignancies cannot avail themselves of this therapeutic option due to lack of matched donor availability. Although HLA-haploidentical transplantations were previously associated with poor outcomes, recent evidence with use of post transplantation cyclophosphamide indicate improved safety and efficacy. The following paper discusses the most recent developments in this area.

Introduction

Allogeneic hematopoietic stem cell transplantation is a curative option for some hematologic malignancies including leukemia. However, many patients are unable to find an HLA-matched sibling or an unrelated donor and donor availability remains an important problem. To address this issue, other options such as HLA-haploidentical transplantation, partially mismatched related or unrelated donor transplantation, or cord blood transplantation have been explored. HLA-haploidentical, or partially HLA-mismatched related donor stem cell transplantations provide the benefit of more potential donors per patient, faster graft acquisition time and improved treatability of post-transplantation relapse, over HLA-matched transplantations. Despite these advantages, historical outcomes have been poor with HLA-haploidentical transplantations (almost 10% survival at 5 or more years) [1]. This has been attributed to bidirectional alloreactivity which causes severe graft versus host disease (GVHD), delayed engraftment and graft failure [1–3]. Although depletion of T cells from the donor graft has been associated with a reduction in graft-versus-host disease, this approach can result in poor immune reconstitution, and high mortality from infections [4]. Therefore there has been a need to address alloreactivity in a selective manner in HLA-haploidentical transplantation. The following discussion provides a brief summary of the most recent progress in improved HLA-haploidentical transplantation protocols.

Selective allo-depletion with high-dose PTCy

Cyclophosphamide (Cy) is an alkylating agent that targets actively proliferating cells and has been previously used to induce immunological tolerance. The high dose post-transplantation cyclophosphamide (PTCy) protocol involves administration of a high dose of Cy (50 mg/kg/day) on days 3 and 4 after transplantation, which results in selective elimination of proliferating alloreactive T cells from the donor and the host. On the other hand, quiescent non-alloreactive T cells, including T cells against pathogens such as cytomegalovirus and herpes simplex virus from the host and the donor are resistant to Cy. This ensures selective mitigation of alloreactivity without complete immunosuppression [5,6].

RIC haplo-BMT with PTCy

The protocol of reduced intensity conditioning and HLA-haploidentical bone marrow transplantation (RIC haplo-BMT) with high-dose PTCy has incorporated this principle and was used for the treatment of 374 patients with hematologic malignancies (120 patients with myeloid disease and 249 with lymphoid disease). The median age of the patients was 55 years (18–75 years), median HLA mismatch was 4/10 (0–5 on the HLA-A, -B, -C, -DRB1 and -DQB1) and the disease risk index (DRI) included low- (19%), intermediate- (65%), and high-risk (16%) patients. Outcomes data on these patients have shown that the incidence of grade II–IV GVHD was 32% and grade III–IV GVHD was 3.3% at 200 days post-transplantation. Similarly, the cumulative incidence of chronic GVHD was reported as 11% at 300 days post transplantation. Notably the incidences of relapse and non-relapse mortality 4000 days after transplantation were 51% and 17% respectively. The overall survival after 5 years was 40% and progression-free survival was 31% (Fig. 1) [7]. Analysis of disease-specific outcomes for RIC haplo-BMT with PT-Cy demonstrate that the overall 5-year survival for AML, ALL, B-cell NHL and Hodgkin lymphoma was 43%, 32%, 49%, and 52%, respectively (Fig. 2) [7]. More specifically, overall survival and progression-free survival 10 years after transplantation among AML patients were 42% and 35%, respectively (Fig. 2). Patients with minimal residual disease (MRD) detectable by flow cytometry had significantly worse outcomes than patients with no evidence of MRD, with long-term survival of >50% in those without MRD versus 10% in those with MRD (unpublished results). Interestingly, older patients, including patients aged 60–75 years, seem to tolerate RIC haplo-BMT with PTCy (Fig. 3) [8]. Although older patients (age 70–75 years) were associated with higher GVHD (~50%) as compared to patients 60–69 years (~30%) and 50–59 years (~20%), they were associated with a lower rate of relapse and lower non-relapse mortality. This is evident from the fact that overall survival and progression-free survival rates were comparable among all patient age groups (Fig. 3). Altogether, these data are encouraging, since the median age at diagnosis of AML is around 66 years and thus this transplantation platform represents a viable option for AML patients.

Fig. 1.

Major outcomes of non-myeloablative, haploBMT with PTCy [7]. Fuchs EJ. Bone Marrow Transplant 2015; 50 Suppl 2:S31-6.

Fig. 2.

Survival after nonmyeloablative, haploBMT with PTCy, by tumor histology. NHL = non-Hodgkin lymphoma [7]. Fuchs EJ. Bone Marrow Transplant 2015; 50 Suppl 2:S31-6.

Fig. 3.

Outcomes in haplo-BMT with PTCy based on patient age [8]. Reprinted with permission. ^©2015 American Society of Clinical Oncology. All rights reserved.

Comparison with HLA-matched BMT

The disease risk index (DRI) can be used to predict outcomes of BMT among heterogeneous patients based on diagnosis and disease status. DRI can be used as a prognostication tool irrespective of conditioning intensity or graft source [9,10]. Fig. 4 shows that DRI (low, intermediate or high risk) consistently prognosticates outcomes among patients undergoing RIC haplo-BMT with PT-Cy and the differences in the risk categories can be attributed to differences in the corresponding relapse rates (Fig. 5). The non-relapse mortality however, is not significantly different between the low-, intermediate-, and high-risk categories (Fig. 5). Given these results, outcomes from 614 patients that had received HLA-matched transplantation (matched sibling or matched unrelated donor) were compared to outcomes from patients who had received RIC haplo-BMT with PTCy based on DRI (Table 1). This comparison indicates that survival outcomes for RIC haplo-BMT with PTCy were similar to matched-BMT [11]. Notably, increased survival was observed among patients with high-risk disease who received RIC haplo-BMT with PTCy, which has been reported previously [12]. A recent study has shown that with respect to matched sibling and matched unrelated donors, HLA-haploidentical transplantation with PT-Cy was associated with comparable outcomes of overall survival, event-free survival, acute GVHD, and chronic GVHD [13]. Similarly, a comparison of HLA-haploidentical transplantation with PTCy versus matched (8/8) unrelated donor stem cell transplantation showed superimposable or non-significant differences in non-relapse mortality, relapse, disease-free survival, and overall survival. These retrospective outcomes were similar for both myeloablative as well as non-myeloablative RIC of matched unrelated donor stem cell transplantation [14].

Fig. 4.

Haplo-BMT with PTCy outcomes based on DRI [11]. Republished with permission of American Society of Hematology, from Risk-stratified outcomes of nonmyeloablative HLA-haploidentical BMT with high-dose post transplantation cyclophosphamide. McCurdy SR, Kanakry JA, Showel MM, Tsai HL, Bolanos-Meade J, Rosner GL et al. Blood 125 (19) ^©2015; permission conveyed through Copyright Clearance Center, Inc.

Fig. 5.

Relapse risk and non-relapse mortality in Haplo-BMT with PTCy based on DRI [11]. Republished with permission of American Society of Hematology, from Risk-stratified outcomes of nonmyeloablative HLA-haploidentical BMT with high-dose post-transplantation cyclophosphamide. McCurdy SR, Kanakry JA, Showel MM, Tsai HL, Bolanos-Meade J, Rosner GL et al. Blood125(19) ^© 2015; permission conveyed through Copyright Clearance Center, Inc.

Table 1.

Comparison of RIC haplo-BMT with PTCY and RIC-matched BMT [11]. Survival outcomes at 3 year post-BMT, by Disease Risk Index*.

Disease risk index (DRI)	Overall survival (%)		Progression-free survival (%)
	Matched (n = 614)	Haplo (n = 374)	Matched (n = 614)	Haplo (n = 374)
Low	70	73	66	65
Intermediate	47	49	31	39
High/very high	25	37	15	25

Summary

HLA-haploidentical transplantations with PTCy have demonstrated promising results in patients with AML, ALL, B-cell NHLs and Hodgkin lymphoma. The outcomes include low GVHD and non-relapse mortality. In addition, HLA-haploidentical transplantations with PTCy were well-tolerated by older patients. Furthermore, these most recent outcomes with HLA-haploidentical transplantations with PTCy were comparable to those with HLA-matched transplantations when adjusted for disease risk. Thus there is an equipoise to conduct randomized trials using different graft sources such as double unrelated donor umbilical cord blood (clinicaltrials.gov identifier NCT01597778). Finally, further studies on maintenance approaches (sorafenib, azacitidine) and cellular therapies (donor CAR T cells, natural killer cells, tumor specific T cells against WT-1, PRAME, survivin) to address the issue of relapse in HLA-haploidentical transplantations with PTCy are warranted.

Abbreviations

ALL

acute lymphoblastic leukemia

AML

acute myeloid leukemia

BMT

bone marrow transplantation

CAR T cell

chimeric antigen receptor T cell

Cy

cyclophosphamide

DRI

disease risk index

GVHD

graft versus host disease

haplo-BMT

HLA-haploidentical BMT

MRD

minimal residual disease

NHL

non-Hodgkin’s lymphoma

PRAME

preferentially expressed antigen in melanoma

PTCy

post transplantation cyclophosphamide

RIC

reduced intensity conditioning