Paul Anthony Keown

Transplantation is the preferred treatment for end-stage renal disease, but premature graft loss is common and antibody-mediated rejection is a leading cause of transplant failure. Human leukocyte antigens (HLA) are the principal targets of these alloantibodies and matching for HLA can improve outcomes. However, matching is not routinely performed in Canada due to challenges such as vast geographic distances for organ-sharing and the heterogeneity of the HLA complex.This dissertation addresses an alternative method of HLA matching, called epitope or eplet matching. Eplets are amino acids on the HLA protein that are antibody targets. Eplet matching requires the amino acid sequence of patient and donor HLA proteins. To achieve this, the performance of Next-Generation Sequencing (NGS) to sequence the 11 HLA genes was assessed in 2234 samples. These results identify the ambiguities in this method, demonstrate improvement in precision in sequencing chemistry, and enabled the implementation of NGS as a reliable technology for Canadian HLA laboratories.Utilizing this platform, HLA allele and eplet frequencies were analyzed in 1846 renal patients and donors to estimate the probability of matching in Canada. All 150 eplets were identified across class I and II alleles, demonstrating sharing between exemplary HLA proteins. Eplets occurred in higher frequencies than their allelic counterparts and distribution was comparable between donors and recipients. Simulations demonstrated prospective matching enables donor-recipient compatibility at class II genes with a waiting-list size of >250 patients, suggesting that eplet matching is achievable within province, avoiding the complexities of national organ sharing except for difficult-to-match patients.I then examined the eplet targets of antibodies formed to mismatched donor HLA under real-world conditions to explore relative immunogenicity. Donor-specific antibodies occurred in approximately 5% of 1155 recipients, were associated with accelerated graft loss, and were predicted by class II eplet mismatches. They were formed most commonly against HLA-DQB1 epitopes and both eplet location and amino acid composition appeared important suggesting a cardinal role in immunogenicity which may incorporated into matching strategies. This collective research provides a foundation of evidence that will support a national program to explore prospective eplet matching in renal transplantation.

View record