Born to Friends

James Boswell achieved fame as the biographer of Dr. Samuel Johnson, but as an only child it was clear that he felt keenly the lack of a sibling relationship. His words "I, who have no sisters or brothers, look with some degree of innocent envy on those who may be said to be born to friends," have resonance to a contemporary debate in hematology: How do we select a donor for a patient who needs a stem cell transplant? The conventional approach is to start by testing any available brothers or sisters who we know each has a 25 percent chance of being matched at both human leukocyte antigen (HLA) haplotypes. The statistics tell us that the chance of finding a donor is 25 percent if there is one sibling, 44 percent for two siblings, 58 percent for three, and so on. However, more than 70 percent of patients do not have a suitable sibling donor. Several options have become available, and the article by Dr. Abraham Kanate and colleagues adds important information to this lively debate.

The three main options are an HLA-matched unrelated donor (URD), umbilical cord blood transplantation, or a haploidentical donor (HD). URDs are the strong preference of most transplant physicians, which has led to the development of very large and sophisticated volunteer donor panels such as the U.S. National Marrow Donor Program (NMDP). The clinical outcomes for patients matched with URDs have improved enormously in the past 20 years. Indeed, given the fact that the donor immune system exploits genetic differences to mediate a graft-versus-leukemia response, the increased histoincompatibility of a URD can help to reduce the rates of disease relapse compared with sibling donors. Cord blood transplantation has also achieved marked improvements, and a major step forward has been the use of two cord donations for each adult patient.

Haploidentical transplantation, where only one HLA haplotype is shared between donor and patient, is a relative newcomer to the debate. It was initially considered impossible due to the potential high risk of graft-versus-host disease (GVHD), but the development of high-dose stem cell infusions and intense T cell depletion facilitated its introduction in the pediatric setting. A further remarkable development was the introduction of a novel approach to reduce GVHD through the administration of cyclophosphamide at day 3 and day 4 after transplantation. This acts to kill rapidly dividing alloreactive T cells and, remarkably, can restrain GVHD even without prior T cell depletion of the graft.

This paper was a comparison of the clinical outcomes for adult patients with lymphoma who received a reduced intensity transplant from either URDs or HDs. A registry analysis of 732 URDs (n = 241 with, or n = 491 without antithymocyte globulin after reduced-intensity conditioning) and 185 HDs was available and showed that relapse rates, nonrelapse mortality, progression-free survival, and overall survival were similar in both groups. However, HD transplants were associated with a reduction in both grade III-IV acute GVHD and also chronic GVHD in a multivariate analysis.

This topic is the subject of intense debate. Haploidentical transplantation offers the advantages of wide availability of donors, from parents, siblings, or children, and it is also much less costly. However, URD registries continue to deliver excellent outcomes for many patients. It is important to remember that this was a registry analysis, and there is now an urgent need for a randomized trial to directly compare these two very different approaches.