HLA-A Disparities Illustrate Challenges for Ranking the Impact of HLA Mismatches on Bone Marrow Transplant Outcomes in the United States

HLA disparity between hematopoietic stem cell donors and recipients is one of the most important factors influencing transplant outcomes, but there are no well-accepted guidelines to aid in selecting the optimal donor among several HLA mismatched donors. In this report, HLA-A is used as a model to illustrate factors that are barriers to delineating the relationship between specific HLA mismatches and transplant outcomes in the United States. Patients in this investigation received transplants for hematologic malignancies that were facilitated by the National Marrow Donor Program (NMDP) between 1990 and 2002 (n = 4226). High-resolution HLA typing was performed for HLA-A, -B, -C, -DRB1, -DQA1, -DQB1, -DPA1, and -DPB1. HLA-A mismatches were observed in 745 donor-recipient pairs and 62% of these pairs also had disparities at HLA-B, -C, and/or -DRB1. The HLA-A mismatches involved 190 different combinations of HLA-A alleles and 51% of these were observed in only 1 pair. Addition of a single HLA-A disparity when HLA-B, -C, and -DRB1 were matched (n = 282) was associated with increased mortality (odds ratio [OR] = 1.32, confidence interval [CI] 1.07-1.63). When HLA-B, -C, and -DRB1 were matched, the most frequent HLA-A mismatches were HLA-A∗0201:0205 (n = 28), HLA-A ∗0301:0302 (n = 15), HLA-A ∗0201:0206 (n = 15), HLA-A ∗0201:6801 (n = 12), HLA-A∗0101:1101 (n = 11), and HLA-A∗0101:0201 (n = 10). There were no statistically significant relationships between any of these disparities and transplant outcomes (engraftment, acute and chronic graft-versus-host disease [aGVHD, cGVHD] relapse, treatment-related mortality [TRM], or overall survival [OS]) when adjustments for multiple comparisons were considered. Achieving 80% power to detect an effect of any 1 of these 6 HLA-A disparities on survival is estimated to require a total transplant population of 11,000 to more than 1 million U.S. donor-recipient pairs depending upon the HLA disparity. Thus, alternative approaches are required to develop a clinically relevant ranking system for specific HLA disparities in the United States.