Abstract
BACKGROUND: Given the lack of specificity of current blood and urine testing and the resistance/inability to perform pancreas allograft biopsies, additional noninvasive investigational tools to assess the risk for rejection are needed. This study examines the clinical impact of molecular HLA matching in a large single-center simultaneous pancreas-kidney (SPK) transplant program. METHODS: The study cohort comprised 238 SPK recipients between 2012 and 2021. The number of HLA mismatches, eplet, Snow (that counts the number of protein-specific surface-accessible donor HLA amino acid mismatches), and predicted indirectly recognizable T-cell epitope (PIRCHE, version 4.2; 100%) loads were calculated on the basis of 2-field HLA-A, -B, -C, -DRB1, and -DQB1 typing of recipients and donors. Univariable and multivariable Cox proportional hazard, as well as Kaplan-Meier analyses, were performed considering either first rejection events of a graft or a composite endpoint of de novo donor-specific antibodies, first rejection, and uncensored graft failure of either organ. RESULTS: Kaplan-Meier analyses considered class II PIRCHE groups separated by a threshold of 7. From the considered histocompatibility metrics, multivariable regression analysis revealed only PIRCHE-II derived from donor HLA class II as statistically significantly correlated with clinical events and rejection after SPK, mostly driven by pancreas rejections. Furthermore, longer dialysis time and the induction agent had significant negative impacts on the defined composite endpoint. CONCLUSIONS: Our data support the clinical benefit of incorporating PIRCHE scores for the interpretation of class II HLA mismatches among patients undergoing SPK transplantation.