Predictive ability of HLA antigen incompatibility prediction algorithms for donor heart rejection risk stratification

Background. Standard HLA typing and simple allele mismatch counts do not always provide an objective assessment of rejection risk. Platforms PIRCHE (Predicted Indirectly ReCognizable HLA Epitopes) and HLAEMMA (HLA Epitope MisMatch Algorithm) estimate donor–recipient HLA incompatibility at the epitope level, based on low-resolution HLA typing. Objective: To evaluate the predictive ability of PIRCHE and HLA-EMMA for stratifying the risk of heart allograft rejection within the first year after orthotopic heart transplantation. Materials and Methods. This retrospective study included 160 heart recipients. During the first-year post-transplantation 77 recipients (control group) showed no rejection (ACR 0R–1R and/or pAMR 0–1). Eighty-three recipients experienced an adverse event: a rejection episode (ACR 2R–3R and/or pAMR 2–3) or death from any cause without preceding rejection. HLA incompatibility was calculated using the PIRCHE-T2, PIRCHE-B, and six HLA-EMMA algorithms. The prognostic significance was assessed using the Kaplan-Meier method. Results. Statistically significant differences between groups were observed for five algorithms: Allele-MM-L2, AA-MM-L1, AA-MM-L2, SA-L2, and PIRCHE-T2. PIRCHE-T2 demonstrated the strongest predictive value with a threshold >80: adverse events occurred in 73.7 % (28/38) of recipients with PIRCHE-T2 >80 versus 45 % (55/122) with PIRCHE-T2 ≤80. Conclusion. The obtained results indicate the potential utility of the PIRCHE-T2 with a threshold value of >80 for risk stratification during the first post-transplant year.