HLA similarity between donors (D) and recipients (R) of solid organ transplantation (SOT) plays a central role in determining the risk of rejection or graft-versus-host disease.Although prior studies have reported the value of HLA matching in SOT, these have largely been based on small clinical studies or retrospective analyses that use two-digit HLA matching at the antigenic level. To determine if D-R HLA mismatching negatively impacts SOT outcomes, we applied recent advances in HLA genotype imputation to infer accurate HLA alleles by using high-density genotyping platforms and known HLA-sequenced reference panels. We report on the refinement and application of a high-throughput, highly accurate analytical method for calling HLA alleles using genotyping data acquired from the DNA of SOT D-Rs. We first piloted this analysis and demonstrated that our method achieved over 93-98% accuracy in resolving HLA calls at the 4-digit level. We next extended this method to the analysis of genotype data derived from the genomic DNA of over 300 matched donor and recipient individuals enrolled in the CTOT-3 Trial. At present, we are evaluating the accuracy of this analysis in predicting HLA mismatches as well as assessing in a retrospective multivariate and cox-hazard proportional risk analysis to determine the of affect HLA D-R matching on the risk of EAD, chronic rejection, and transplant-related mortality in SOT recipients.

Finally, whole exome sequencing (WES) costs have declined recently, there has been significant interest in using existing WES results to accurately infer HLA genotypes. However, due to its extended sequence polymorphism and population frequency variability, accurate sequence alignment and the identification of non-variable HLA regions for exome capture platform design have been challenging. To tackle this problem, we propose the transformation of sequence-level results across the MHC from WES platforms to HLA genotype assignments. We have systematically extracted genomic markers that either are included directly or are informative tags of HLA genotypes and are currently using this data to build HLA imputation reference panels. We report results including HLA gene coverage and imputation panel reference markers using our pilot analysis, showing that this is a feasible method to apply existing WES data to the interpolation, analysis and imputation of HLA genotypes.

« Back to 2015 American Transplant Congress