Identification of a 10/10 matched donor for patients with an uncommon haplotype is unlikely

• Published in: HLA Vol. 89; no. 2; pp. 77 - 81
• Main Authors: Olson, J. A., Gibbens, Y., Tram, K., Kempenich, J., Novakovich, J., Buck, K., Dehn, J.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01.02.2017
• Subjects: Algorithms; Alleles; Bone Marrow Transplantation - methods; Continental Population Groups; Cord Blood Stem Cell Transplantation - ethnology; Cord Blood Stem Cell Transplantation - methods; Gene Expression; Gene Frequency; haplotype; Haplotypes; Hematopoietic Stem Cell Transplantation - ethnology; Hematopoietic Stem Cell Transplantation - methods; Histocompatibility Testing; HLA Antigens - classification; HLA Antigens - genetics; HLA Antigens - immunology; HLA typing; Humans; Probability; Registries; uncommon HLA; United States; unrelated donor; Unrelated Donors - statistics & numerical data; Unrelated Donors - supply & distribution; haplotype; unrelated donor; HLA typing; uncommon HLA
• DOI: 10.1111/tan.12946
• ISSN: 2059-2302

Background Despite over 6 million subjects contributing to the National Marrow Donor Program human leukocyte antigen (HLA) haplotype frequency reference data (HFD), haplotypes cannot be predicted from the HLA assignments of some patients searching for an unrelated donor (URD) in the Be The Match Registry®. We aimed to determine the incidence of these patient searches and whether haplotypes lacking from the HFD can be found among the low‐resolution typed URD pool. Materials and Methods New NMDP searches with uncommon patient haplotypes (UPH), defined as a lack of haplotype pairs in any single ethnic group in the HFD based upon HLA‐A˜C˜B˜DRB1˜DQB1, were identified. Each search had up to 20 potential 10/10 or 8/8 URDs typed to determine the likelihood of an allele match. Results The incidence of patient searches without haplotype pairs in a single ethnic group in the HFD was 1.2% (N=144 out of 12,172) and a majority of these patients (117; 81%) had one uncommon haplotype previously uncharacterized in the HFD. Non‐White patients had the highest incidence of UPH. Importantly, no patients with UPH had a 10/10 URD identified. The transplant rate among UPH patients was 15%, and a majority of these patients utilized cord blood units as their transplant stem cell source. Conclusion Therefore, the HLA HFD that informs the HapLogic matching algorithm is thorough as UPH patient searches were infrequent. Since such patients are highly unlikely to have a fully 10/10 matched URD identified, this study supports the identification of alternative stem cell sources including cord blood or a mismatched URD early in the search process.