Calculated PRA and PIRCHE Algorithm in Kidney Transplant Recipients

Kidney transplantation is a life-saving treatment option for individuals with an end-stage kidney disease. However, the success of a transplant relies on several critical factors, including the compatibility between the donor and the recipient. One crucial aspect in determining this compatibility is the assessment of the recipient's calculated Panel Reactive Antibody (cPRA) level. The cPRA testing plays a pivotal role in evaluating the risk of antibody-mediated rejection (AMR) and ensuring an optimal organ matching, thus increasing the chances of graft survival post-transplantation. Traditionally, cross-matching tests were used to determine HLA compatibility between the donor and the recipient. However, cPRA offers a more comprehensive assessment by quantifying the percentage of potential donors that a recipient may be incompatible with due to pre-existing antibodies (1).

Several studies have emphasized the significance of cPRA testing in kidney transplantation, where cPRA levels above 80% were associated with an increased likelihood of AMR and graft loss (1, 2).

Calculated PRA testing in kidney transplantation has revolutionized the field by enabling a more accurate assessment of compatibility and risk prediction for AMR. The utilization of PRA testing has led to advancements in personalized patient care, expansion of the donor pool for sensitized patients, and improved overall graft survival rates (3).

To improve the accuracy of assessing immune compatibility, a groundbreaking algorithm called PIRCHE (Predicted Indirectly ReCognizable HLA Epitopes) has emerged as a promising tool in kidney transplantation.

The PIRCHE algorithm, developed by Ingo Thiele and colleagues in 2013, aims to identify the potentially immunogenic human leukocyte antigens (HLA) epitopes on the donor graft that are recognized by the recipient's HLA antibodies (4). Unlike traditional approaches that focus solely on HLA-match compatibility, the PIRCHE algorithm takes into account the subtle variations in HLA epitopes and their potential for immune recognition. By considering indirect recognition of HLA epitopes, the PIRCHE algorithm provides a more comprehensive assessment of alloimmunization risk in kidney transplantation.

Several studies have highlighted the significance of the PIRCHE algorithm in improving transplantation outcomes. PIRCHE scores significantly correlated with the development of de novo donor-specific antibodies (DSA) and the occurrence of AMR (5, 6).

The aim of this analysis was to compare the determined immunological risk according to the standard protocol of the Transplant-Nephrology Department of the University Hospital Martin and the risk that would be determined by cPRA and PIRCHE. Another aim was to compare the result of the protocol biopsy at month 3 after the transplantation (including the determination of the presence of donor-specific antibodies) in the context of cPRA and PIRCHE at the time of the transplantation.

Patients who underwent transplantation of kidney from a donor after brain death at the Transplant Centre of the University Hospital Martin in 2022 and 2023 were included in the retrospective analysis. All included patients underwent protocol graft biopsy and donor-specific antibody testing. Both protocol biopsy and DSA determination were performed at month 3 after the kidney transplantation, and histological findings were assessed according to the Banff classification 2019. The determination of donor-specific anti- bodies was performed before the kidney transplantation and in the 3rd month for all patients included in the study by means of the LUMINEX methodology (positivity was stipulated at ≥ 500 MFI). Before the transplantation, anti HLA antibody screening (class I, class II, and MICA) was performed every 3 months by Luminex method and in case of a positive screening, the specification was also completed. Subsequently, during the kidney allocation itself, we evaluated any DSA positivity.

We retrospectively evaluated immunological risk in all included patients according to a valid immunosuppressive induction protocol as shown in table 1.

The maintenance immunosuppression consisted of tacrolimus and mycophenolic acid (1080 mg daily dose until the 2^nd week after the transplantation, followed by a 720 mg daily dose). Methylprednisolone 500 mg i.v. was administered on day 0 and day 1, followed by the administration of Prednisone 20 mg until the 2^nd week after the transplantation, Prednisone 15 mg until the 4^th week after the transplantation, Prednisone 10 mg until the 12^th week after the transplantation, and Prednisone 7.5 mg until the 12^th month after the transplantation, with Prednisone 5 mg administered daily thereafter.

A donor with extended criteria (ECD) means a donor per the definition of ECD codified in 2002: donors over the age of 60 years without co-morbidities or donors over the age of 50 years with at least two co-morbidities that include: blood hypertension, death from cerebrovascular accident, or terminal serum creatinine levels >1.5 mg/dL.

Delayed graft function was defined as the need for dialysis during the first week after the transplantation.

As a next step, we retrospectively performed cPRA and PIRCHE calculations in all patients included in the analysis at the time of the kidney transplantation. According to these values, we then compared the immunological risk that was determined at the real time of the kidney transplantation according to the induction protocol and the hypothetical risk that would have been attributable to the patients according to cPRA or PIRCHE.

The cPRA was determined at the time of the kidney transplantation using a virtual calculator to determine the profile of the panel of reactive antibodies (cPRA), (https://www.etrl.org/InformationVPRA.aspx), found in the patient's serum. The virtual PRA calculator is a web-based application determining the non-acceptable HLA antigens: HLA-A, -B, -C, DRB1, -DRB3/4/5, -DQB1, -DQA1, DPB1, and -DPA1. The unacceptable HLA antigens were entered into the virtual PRA calculator based on anti-HLA antibodies detected in the patient's serum using the Luminex method, single-antigen bead class I (SAB1), and single-antigen bead class II (SAB2) (One Lambda/Thermofisher). Anti-HLA antibodies in each patient's serum were determined individually on the basis of MFI values and cut-off boundaries, and on the basis of the common and specific eplets (https://www.epregistry.com.br/) to which anti-HLA antibodies in the Luminex assay bound to each HLA allele. After the determination of unacceptable HLA antigens, we then calculated cPRA (https://www.etrl.org/vPRA.aspx) % values for individual patients.

PIRCHE score was determined at the time of the kidney transplantation using the PIRCHE-II algorithm and netMHCIIpan-3.0_IMGT-3.47 database (https://www.pirche.com/). Incompatibilities in HLA alleles between the donor and the recipient were determined at the low-resolution level using LinkSeq Real-Time PCR HLA typing (One Lambda/Thermofisher). HLA typing of the donor and the patient at the low-resolution level: HLA-A, -B, -DRB1, and -DQB1 were imported into the PIRCHE-II algorithm with unique identifiers that paired HLA typing, the patient, and the donor. For all incompatible HLA alleles between the donor and the recipient, we determined the number of peptides derivatized from incompatible HLA alleles of the donor and presented by HLA-DRB1 alleles of the recipient to CD4+ T-lymphocytes of the recipient (PIRCHE-II). Based on PIRCHE-II scores and derived peptides, we determined the immunological risk of predicting the initial de novo formation of de novo DSA and anti-HLA antibodies against individual peptides of the HLA-A, -B, -DRB1, and DQB1 donor alleles.

Twenty patients were included in the cohort, with only one patient undergoing a secondary kidney transplantation. In a first step, the cohort was divided according to the standard immunological risk resulting from the induction immunosuppression protocol (as shown in table 1), resulting in 3 groups – low immunological risk (green), intermediate (orange), and high (red) – figure 1.

Fig. 1

According to the induction protocol, 8 patients were low risk, 2 patients were high risk, and the remaining patients were classified as intermediate immunologic risk. We proceeded with the hypothesis of using only immunological factors to determine risk (thus only PRA value, Luminex, positive cross match in history, and secondary or higher KT were taken into account). In this case, the risk distribution would change as follows (figure 2): the patients at high risk remained unchanged, but there was a reduction in risk from intermediate to low in 4 patients.

Fig. 2

Analogously, we created a hypothesis of risk determination with only non-immunological criteria, where neither patient was at a high immunological risk (figure 3); on the contrary, both patients who were at a very high risk according to immunological factors were in the green and thus at low risk in this case.

Fig. 3

We further advanced our hypotheses by including cPRA in the immunological risk assessment (figure 4 – Immunological risk determined by cPRA). According to cPRA, the two patients we assessed as intermediate risk in real time would be classified as low risk. Both of these patients did not have DSA identified at month 3 and did not have rejection changes in the protocol biopsy.

Fig. 4

Finally, we evaluated the hypothetical immunological risk based on PIRCHE, figure 5. In this case, two patients would have their immunological risk changed from moderate to low, one patient even from very high to low. In none of these patients did we subsequently observe DSA formation at month 3 post-transplantation and also all three protocol biopsies were free of rejection changes in these patients. Interestingly, a patient who was assessed as high risk according to the standard induction protocol (based on secondary kidney transplantation) would have had a low immunological risk according to PIRCHE, and, as mentioned above, even in this patient there was no evidence of DSAs at month 3 post-transplantation and the protocol biopsies were free of rejection findings. Conversely, up to 5 patients would have been classified from low or intermediate risk to high risk in the PIRCHE risk assessment. We did not observe any rejection changes in the protocol biopsy in these patients; also, there were no confirmed DSAs in these patients at month 3.

Fig. 5

The patient who developed DSA at month 3 and also had a positive finding in terms of rejection changes in the protocol biopsy was classified as highest risk according to PIRCHE.

Finally, we compared the PIRCHE value in patients with negative versus positive Luminex while on the waiting list (WL) and also 3 months after the transplantation (table 2). We observed only numerically higher PIRCHE values in patients with positive Luminex examination versus negative patients, without statistical significance.

Table 2

In our analysis, we confirmed the importance of both immunological and non-immunological parameters for determining the risk at the time of the transplantation. Comparing our determined immunological risk at the time of the transplantation (according to the valid induction immunological risk) and immunological risk according to cPRA and PIRCHE, respectively, we identified patients in whom the determined risk did not match. Nevertheless, although their "new" immunological risk according to cPRA or PIRCHE was assessed to be lower compared to the actual established risk, we did not find any evidence of rejection changes or DSA in these patients at month 3 post-transplantation. This can be mainly explained by the fact that there was only one patient with a high cPRA value in the whole cohort. It is in patients with a cPRA value of more than 85 and 95%, respectively, that such an immunological risk assessment is warranted. Understanding a recipient's PRA level allows transplant professionals to identify potential mismatched donors and develop strategies to mitigate the risk of AMR. Advanced techniques such as virtual cross-matching and desensitization protocols have emerged as effective tools for improving transplantation outcomes in patients with higher PRA levels. Virtual cross-matching uses a combination of PRA results and donor-specific antibody (DSA) testing to predict the risk of AMR, assisting in donor selection, and promoting better graft survival rates. Moreover, desensitization procedures involving therapeutic interventions such as plasmapheresis, intravenous immunoglobulin administration, and rituximab have shown promising results in reducing PRA levels, thereby expanding the pool of compatible donors for highly sensitized patients (3).

When we evaluated PIRCHE in the context of positive and negative results of Luminex examination in the studied patients, we did not confirm a significant difference (numerical only). However, we know that the high immunological risk evaluated by the PIRCHE algorithm is mainly indicative of DSA formation in the long term. A study by Spitznagel et al. confirmed in 123 patients that a high PIRCHE score (median 101.50 vs. 74.00) predicted TCMR changes (6).

Specific HLA-antigen disparities can trigger considerable clinical alloreactivity and are classified as notably immunogenic HLA mismatches. Conversely, certain HLA-antigen mismatches typically do not result in clinical alloreactivity. This underscores the importance of pinpointing molecular HLA-epitope mismatch rather than solely relying on the number of HLA-antigen mismatches. The PIRCHE algorithm emerges as a more suitable tool for stratifying the risk of immune-mediated injury. Geneugelijk et al. documented in 2018 a robust association between higher PIRCHE-II scores and an elevated risk of kidney allograft failure (7, 8). Furthermore, based on their multivariate analysis, the prognostic value of PIRCHE-II scores exceeded that of HLA-antigen mismatches. Studies have demonstrated that the predictive prowess of PIRCHE-II scores in predicting de novo DSA development is most pronounced for HLA-DQ and HLA-DR, followed by HLA-A and HLA-B (9, 10, 11, 12).

The implementation of the PIRCHE algorithm in clinical practice holds several advantages. Firstly, it provides a more precise assessment of immune compatibility, allowing physicians to make informed decisions regarding donor selection and risk stratification. Moreover, the algorithm can guide personalized immunosuppressive strategies to prevent and manage alloimmune responses, ultimately improving graft survival rates. PIRCHE algorithm represents a significant advancement in assessing alloimmunity and improving kidney transplantation outcomes. By evaluating the compatibility of HLA epitopes and their potential immunogenicity, this algorithm provides a better understanding of the risks associated with graft rejection. As the field of kidney transplantation continues to evolve, the integration of the PIRCHE algorithm into clinical practice has the potential to enhance personalized patient care and optimize long-term transplant success (13, 14).

In conclusion, calculated PRA testing in kidney transplantation has revolutionized the field by enabling a more accurate assessment of compatibility and risk prediction for ABMR. The utilization of PRA testing has led to advancements in personalized patient care, expansion of the donor pool for sensitized patients, and improved overall graft survival rates (15).

The utilization of PIRCHE for organ allocation has been documented in small studies involving uniform populations. Larger studies are required to confirm this strategy. Challenges in employing this method of epitope matching for donor selection involve the need for clearer differentiation between eplet-mismatch load and immunogenic eplets, as well as the availability of analysis software (8, 16).