Human leukocyte antigen eplet mismatching is associated with increased risk of graft loss and rejection after pediatric heart transplant

Background While mismatching between donor and recipient human leukocyte antigen (HLA) alleles has been associated with increased graft loss in pediatric heart recipients, it is actually the surface amino acid structures, termed eplets, which determine the antigenicity of each HLA molecule. We hypothesized that HLA eplet mismatch analysis is a better predictor of adverse outcomes after pediatric heart transplant than conventional allele mismatch comparison. Methods A retrospective review of the Pediatric Heart Transplant Society database identified pediatric heart recipients (<18 years at listing) with complete donor and recipient HLA typing (A, B, and DR). Imputed high-resolution HLA genotypes were entered into HLAMatchmaker software which then calculated the number of eplet mismatches between each donor-recipient pair. Multivariable Cox regression analysis was used to examine associations between allele or eplet mismatching and adverse outcomes. Results Compared to those with <20 HLA class I eplet mismatches, recipients with 20 or more HLA class I eplet mismatches had an increased risk of graft loss (HR 1.46 [1.01-2.12], p = .049). HLA class I eplet mismatching was also associated with rejection (>20 mismatches: HR 1.30 [1.03-1.65], p = .030), while HLA class II eplet mismatching was associated with specified antibody-mediated rejection (10-20 mismatches: HR 1.57 [1.06-2.34], p = .025; >20 mismatches: HR 3.14 [1.72-5.71], p < .001). Neither HLA class I nor class II allele mismatching was significantly associated with graft loss or rejection. Conclusion Eplet mismatch analysis was more predictive of adverse post-transplant outcomes (including graft loss and rejection) than allele mismatch comparison. Further study, including prospective high-resolution HLA typing, is warranted.