Abstract
Matching of HLA-A, -B and -DRB1 is the gold standard in histocompatibility testing for kidney transplantation. A good HLA match reduces the risk of alloimmunisation, thereby decreasing the likelihood of graft failure and potentially enabling reduced immunosuppression therapy. However, the role of the non-classical HLA and MIC genes is not well established. We hypothesised that polymorphisms in these genes contribute additional risk of acute rejection after kidney transplantation, beyond what is explained by classical HLA mismatching. We developed and validated an oligogenic risk score based on ten polymorphisms in the HLA-E, -F, -G, MICA and MICB genes associated with time to acute rejection using Cox proportional hazards analysis in a single centre retrospective cohort of 1025 Finnish adult kidney transplant recipient-donor pairs. The score, particularly at the extreme ends, that is, the 1st and 10th deciles, was associated with rejection-free survival in both the discovery cohort (Cox multivariable model HR 1.20, 95% CI 1.08-1.33, p-value 4.350 × 10(-4)) and the replication cohort (HR 1.33, 1.10-1.62, p-value 0.004). The score added value beyond the standard HLA-A, -B, -DRB1 matching or HLA class I and II eplet matching, differentiating high and low-risk patient groups. The association remained when adding HLA-C and HLA-DQB1 matching to the analysis. The results demonstrate the value of considering the non-classical HLA and MIC genes in risk assessment and even graft allocation in kidney transplantation, and the potential of using oligogenic and polygenic summary scores to predict transplant outcomes.