Eplet Mismatching as a Predictor of Acute Rejection in Simultaneous Pancreas Kidney Transplantation

Background: Eplet Mismatching is a high-resolution molecular-level analysis. Eplets are small, specific patches (3-5 amino acids) on the surface of the HLA molecule that are physically accessible to the recipient's immune system (antibodies and T-cell receptors). Pancreas transplantation is the definitive treatment for patients with brittle type 1 diabetes or severe type 2 diabetes, offering the potential for insulin independence and improved quality of life. However, long-term graft survival remains challenged by the high incidence of acute rejection, particularly antibody-mediated rejection (AMR), which is a leading cause of graft loss.
Objective: Given the limitations of traditional HLA matching and the clinical imperative to prevent rejection-driven graft loss, a more precise tool for pre-transplant risk stratification is needed. We hypothesize that a high number of HLA class II eplet mismatches, particularly at the HLA-DQ locus, is significantly correlated with an increased incidence of biopsy-proven acute rejection and the development of de novo DSA in pancreas transplant recipients.
Method: This will be a retrospective, Multiple-center cohort study. All patients who underwent simultaneous pancreas-kidney transplantation (SPK) at Shenzhen Third People's Hospital and Tianjin First Center Hospital， between January 2015 and November 2024.
Inclusion Criteria: 1. Recipient age ≥ 18 years. 2. Availability of high-resolution (4-digit) HLA typing for both donor and recipient .3. A minimum of one year of post-transplant follow-up data. Exclusion Criteria:1. Loss of graft due to technical failure (e.g., thrombosis, anastomotic leak) within the first 30 days. 2. Incomplete HLA typing data or insufficient clinical follow-up. 3. Multi-organ transplants beyond SPK. Patients were divided into two groups: Rejection Group and Non-Rejection Group. Existing donor and recipient DNA samples will be re-analyzed, or existing data will be used to confirm high-resolution typing for HLA-A, -B, -C, -DRB1, -DQB1, and -DPB1 loci.
Acute Rejection: Defined according to the latest Banff classification for pancreas allograft pathology. This was categorized as: T-Cell Mediated Rejection (TCMR), Antibody-Mediated Rejection (AMR), Mixed Rejection. Donor-Specific Antibody (DSA) Monitoring: Protocol and for-cause DSA testing was reviewed. The development of de novo DSA (dnDSA) post-transplant, defined as DSA not present pre-transplant, was recorded. The Mean Fluorescence Intensity (MFI) and specificity of the DSA was correlated with the underlying eplet mismatches. Eplet Mismatch Calculation: Eplet mismatches will be determined using the HLA MatchMaker algorithm (HLA-ABC Eplet Matching Version 4.0 and HLA- DRDQDP Eplet Matching Version 3.1) within a defined HLA epitope database. The following data was extracted from electronic medical records: Recipient and donor demographics (age, sex, BMI), HLA antigen mismatch level, Panel Reactive Antibody (PRA) level at transplant, immunosuppression regimen (induction and maintenance), timing and histology of all rejection episodes, graft survival and patient survival data.
Result: A total of 76 SPK recipients met the inclusion criteria. The cohort was divided into two groups: the Rejection Group (n=11, 14.5%) and the Non-Rejection Group (n=65, 85.5%). The groups were well-matched for baseline characteristics such as recipient age, sex, and induction therapy. As expected, the Rejection Group had a significantly higher proportion of patients who developed de novo DSA (54.55% vs. 10.77%, p<0.001). The total eplet mismatch Antibody-verified (AbVer) load was significantly higher in the Rejection Group compared to the Non Rejection Group (median 15 vs. 11 p=0.01). This difference was driven almost entirely by Class II eplet mismatches. The median Class II eplet mismatch was 6 in the Rejection Group vs. 3 in the Control Group (p<0.001). Class I eplet mismatch was not significantly different between groups (p=0.25). In a Cox proportional hazards model adjusting for recipient age, PRA level, and traditional HLA mismatch, a high HLA-DQ eplet mismatch load (≥6) was identified as the strongest independent risk factor for acute rejection. In 17.11%（13/76) of cases where de novo DSA was detected, the antibody specificity corresponded directly to one or more of the pre-existing high-risk eplet mismatches identified in the donor-recipient pair. The number of DQ eplet mismatches was strongly correlated with the peak MFI of the resulting DQA/DQB (R=0.73, p<0.001). Kaplan-Meier analysis revealed that patients in the "high-risk" DQ eplet mismatch group had significantly inferior 1-year pancreas graft survival (82.9%) compared to the "low-risk" group (100%) (Log-rank test, p<0.001).
Conclusion: These findings provide a strong rationale for prospective evaluation of eplet-based matching into clinical practice to improve outcomes for pancreas transplant recipients.