Related or unrelated donor

In this issue of Blood, Kanda et al have analyzed data from the Japanese Hematopoietic Cell Transplant Registry to address an important question about the risks associated with hematopoietic cell grafts from HLA 1-antigen mismatched related donors contrasted with 8 of 8 HLA allele matched unrelated donors.¹ 

The study reported by Kanda and colleagues addressed the question of whether a serologically defined 1-antigen HLA-A, B, or DRB mismatched related donor should be preferred to an “8/8” HLA-A, B, C, DR allele MUD. The study population consisted of 327 1-HLA antigen mismatched related donor transplant recipients and 452 8/8-HLA allele matched unrelated donor transplant recipients accessed through the Japanese Transplant Registry Unified Management Program (TRUMP), which includes data from the Japan Society for Hematopoietic Cell Transplantation (JSHCT) and the Japan Marrow Donor Program (JMDP). Patients underwent transplantation for the treatment of acute myeloid leukemia, acute lymphoblastic leukemia, chronic myeloid leukemia, or myelodysplasia between January 2001 and December 2008. Patients received T cell–replete marrow or blood stem cells and 90% to 97% received a conventional cyclosporine- or Tacrolimus-based regimen for GVHD prophylaxis. The related and unrelated cases differed for certain pretransplantation risk factors. One-antigen mismatched related donor recipients were more likely to have high-risk disease, to have received peripheral blood rather than marrow stem cells, and be a male receiving a transplant from a female donor than MUD recipients. Results of a multivariate analysis showed that 1-HLA antigen mismatched related cases overall had a lower survival rate compared with 8/8-MUD cases (HR = 1.49; CI = 1.19-1.86). The related donor mismatching effect was most prominent in patients receiving a transplant for standard-risk disease (OR = 1.72; CI = 1.24-2.39). Based on these results, Kanda et al conclude that an 8/8-HLA allele matched unrelated donor should be preferentially selected over a related donor mismatched for one serologically defined HLA-A, B, or DR antigen.¹ 

Kanda et al also examined risk of mortality for 1-antigen mismatched related donor transplantations according to mismatched locus, and found that the greatest risk of reduced survival was associated with mismatching for HLA-B¹ . HLA-C locus typing, which was available for a subset of the related patient-donor pairs, revealed that most of the cases mismatched for HLA-B were also mismatched for HLA-C, implying that the remainder of the presumed 1-antigen HLA-B mismatches were probably two antigen mismatches.

The HLA matching data available to Kanda et al for this retrospective analysis were generated with different HLA typing methods.¹  At the time these transplants were performed, the standard clinical practice for HLA typing for the selection of related donor searches was based on serology and matching for the HLA-A, B, and DR antigens, while HLA typing for the selection of an unrelated donor was based on DNA-based genomic typing and matching for HLA-A, B, C, and DRB1 alleles.

Related donor transplant studies from the 1980s found that risk of severe GVHD and mortality was greater with increasing disparity for serologically defined HLA antigens.²  During the pre-unrelated donor era, mismatched related donor transplantations limited to disparity for 1-HLA antigen were generally deemed acceptable for patients failing conventional therapy when an HLA identical sibling was not available.^2-4 

The utility of DNA-based genomic approaches to HLA genotyping and the importance of allele level HLA matching was initially demonstrated in studies of unrelated donor transplantations,^5,6  and subsequently reinforced by HLA studies facilitated by large transplant registry cohorts.^7-9  Improved statistical power has also made it possible to estimate more accurately the risks of mismatching for antigens and alleles of individual HLA loci, and demonstrated that mismatching for ≥ 2 alleles is associated with progressively increasing risk of severe GVHD and mortality.^6-9 

The analysis presented by Kanda et al leaves open the possibility that a 1-allele HLA-A, B, C, or DR mismatched related donor might be preferred or at least equal to an 8/8-HLA matched unrelated donor. The probability of finding this type of a related alternative donor, however, is very low. Fortunately, recently introduced novel immune suppression strategies have been successful in overcoming the HLA barrier and achieving sustained engraftment without increased risk of severe GVHD, and this has been accomplished with HLA fully mismatched haploidentical donors.^10,11  Future studies are need to confirm these preliminary results, define in more detail the disease-specific indications, and determine the relative advantages of the different alternative donor sources including HLA haploidentical related, matched unrelated, and also umbilical cord blood.