Allogeneic hematopoietic stem-cell transplantation (HSCT) remains the key curative treatment modality for various hematologic malignancies, bone marrow failure syndromes, and some hemoglobinopathies. An HLA-matched donor has traditionally been the donor of choice in HSCT due to superior outcomes associated with HLA-matching. Unfortunately, it is sometimes challenging to identify HLA-matched donors, particularly amongst ethnic minorities who are typically poorly represented in donor registries. This has led to the expansion of the donor pool to include "alternative" donors such as haploidentical or 50% matched related donors (i.e., parent or offspring) and mismatched unrelated donors.
While transplantation across HLA barriers has historically been very difficult due to high rates of severe graft-versus-host disease (GVHD) and poor engraftment, the landscape has changed with the advent of post-transplant cyclophosphamide (PTCy), a novel GVHD prophylaxis strategy. Haplo-transplants using PTCy-based prophylaxis result in such encouraging GVHD outcomes, in some studies superior to HLA-matched transplantation with calcineurin inhibitor-based prophylaxis, that PTCy is now being evaluated in the context of HLA-matched transplantation, as well. While PTCy may well become the new standard in GVHD prophylaxis, the optimal donor choice — haploidentical related or HLA-matched unrelated donor — remains an important question, which we sought to answer.
In our analysis of more than 2,000 transplants for acute leukemia reported to the Center for International Blood and Marrow Transplant Research (CIBMTR), we compared transplant outcomes between the two most common donor types, HLA-matched unrelated donors (MUD) and haploidentical related (haplo) donors, when PTCy-based prophylaxis was used with both donor types. Among patients receiving reduced intensity conditioning regimens, MUD transplantation resulted in lower non-relapse mortality (NRM), lower grade III-IV acute GVHD, poorer engraftment, and higher overall survival (OS) when compared to haplo transplants. The risk of grade III-IV acute and chronic GVHD were also higher after haploidentical transplantation in the myeloablative setting, suggesting an advantage with fully matched donors here, as well.
The current study was driven by an earlier publication from the CIBMTR (Ciurea S et al., Blood 2015; 126(8): 1033-1040) that compared survival between HLA-matched unrelated and haploidentical donor transplants. As the GVHD platforms in the earlier study differed, the pattern of treatment failure differed without a survival advantage for either donor type. Our current analyses included transplantations with a uniform GVHD prophylaxis platform with both donor types for the first time. Our results extend and confirm the importance of donor-recipient HLA matching for allogeneic transplantation even in the PTCy era. While a MUD appears to be the preferred donor, especially for transplantations with reduced-intensity conditioning regimens, it also results in better GVHD outcomes in the myeloablative setting. We believe this is clinically informative and practice changing regarding donor selection for adults with acute leukemia.
