HLA-DQ heterodimer mismatch as a predictor of post-transplant alloimmunity and allograft survival

HLA-DQ mismatch has been correlated with worse kidney allograft outcomes. However, there is linkage disequilibrium between HLA-DQ and HLA-DR, and the independent effect of HLA-DQ mismatches, particularly HLA-DQαβ heterodimer mismatches, on allograft outcomes after adjustment for HLA-DR mismatches has been understudied. It is also unknown whether the potential for HLA-DQ molecules to form cis- or trans-heterodimers impacts allograft outcomes. We analyzed 866 well-characterized kidney transplant recipients for the relevance of potential cis/trans-HLA-DQ⍺β heterodimers on T cell-mediated rejection, antibody-mediated rejection, death-censored allograft survival, and all-cause allograft survival. We found that including potential HLA-DQ⍺β cis/trans-heterodimer mismatches did not improve prediction models for all outcomes compared to HLA-DQ⍺β cis-heterodimers alone. We also studied the independent effect of HLA-DQ⍺β mismatch on these outcomes after adjusting for HLA-DR mismatch in multivariate models. Both HLA-DR antigen and HLA-DQ⍺β cis-heterodimer mismatch were univariate but not multivariate correlates of these outcomes when adjusted for each other. Collinearity between HLA-DRB1, HLA-DQB1, and HLA-DQA1 alleles, and the dilution of survival outcomes with non-alloimmune related events are potential reasons. This underlies the need for novel approaches to precision in alloimmune risk assessment, e.g., molecular mismatch, for use in prediction models and allocation algorithms.