Pre-transplant donor-reactive IL-21 producing T cells as a tool to identify an increased risk for acute rejection

Pre-transplant screening focuses on the detection of anti-HLA alloantibodies. Previous studies have shown that IFN-γ and IL-21 producing T cells are associated with the development of acute rejection (AR). The aim of this study, was to assess whether pre-transplant donor-reactive T cells and/or B cells are associated with increased rejection risk. Samples from 114 kidney transplant recipients (transplanted between 2010 and 2013) were obtained pre-transplantation. The number of donor-reactive IFN-γ and IL-21 producing cells was analyzed by ELISPOT assay. The presence of donor specific antibodies (DSA) was also determined before transplantation. Numbers of donor-reactive IFN-γ producing cells were similar in patients with or without AR whereas those of IL-21 producing cells were higher in patients with AR (p = 0.03). Significantly more patients with AR [6/30(20%)] had detectable DSA compared to patients without AR [5/84(5.9%), p = 0.03]. Multivariate logistic regression showed that donor age (OR 1.06), pre-transplant DSA (OR 5.61) and positive IL-21 ELISPOT assay (OR 2.77) were independent predictors of an increased risk for the development of AR. Aside from an advanced donor-age and pre-transplant DSA, also pre-transplant donor-reactive IL-21 producing cells are associated with the development of AR after transplantation.

www.nature.com/scientificreports/ occur in so called ectopic lymphoid structures (functionally similar structures to germinal centers), that can form within transplanted renal and cardiac allografts 15,16 . Current pre-transplant screening mainly focuses on the detection of anti-HLA alloantibodies present in the serum of transplant candidates 17 . This strategy, however, does not account for the presence of donor-reactive memory T cells which may contribute to allograft rejection. A long-standing area of transplant research has focused on the development of in vitro methods that allow for the accurate detection of donor-specific T cell alloimmunity 18,19 . One such method was developed by Heeger et al., involving the use of an enzyme-linked immunosorbent spot (ELISPOT) assay in which donor-reactive memory IFN-γ producing T cells could be measured 20 . Since then, several research groups have found that the presence of pre-transplant donor-reactive IFN-γ producing cells is associated with the occurrence of early AR [21][22][23][24][25] . However, these findings could not be replicated in the Clinical Trials in Organ Transplantation-01 multicenter study, where no association between donor-specific pre-transplant IFN-γ ELISPOT and AR was found 26 . In 2019, Van Besouw et al. described an association between higher numbers of pre-transplant and post-transplant donor-specific interleukin (IL)-21 producing cells and AR in a case-control study 27 . IL-21 regulates the immune activity of different cells relevant in the setting of organ transplant rejection 28 . IL-21 enhances the cytotoxicity and production of the pro-inflammatory cytokine IFN-γ by natural killer (NK) cells and CD8+ T cells 29,30 , and it has also been shown to stimulate the expansion of Th17 cells 31,32 . Studies in heart transplant recipients have demonstrated that intragraft IL-21 mRNA levels were significantly increased in patients experiencing acute cellular rejection 33 . Additionally, IL-21 is crucial for T cell-dependent B cell differentiation into memory B cells and antibody-producing plasma cells. Due to the pleiotropic effects of IL-21, it is involved in both T cell-mediated rejection (TCMR) and antibodymediated rejection (ABMR) 16,[34][35][36] . The IL-21 ELISPOT assay is therefore, a promising assay for the detection and monitoring of donor-reactive T cells in transplant recipients. The aim of this study, was to assess whether donor-reactive IFN-γ and IL-21 producing T cells and B cell donor reactivity assessed pre-transplantation in kidney transplant recipients are associated with an increased rejection risk after transplantation.

Materials and methods
Study population. A cohort of 114 renal transplant recipients transplanted between 2010 and 2013 was sampled cross-sectionally within 24 h prior to transplantation. In order to be able to study more patients with a rejection, the study cohort was enriched with patients who experienced one or more rejection events after transplantation. All patients provided written informed consent and the study was approved by the Medical Ethical Committee of the Erasmus Medical Center in Rotterdam, the Netherlands (biobank protocol MEC-2010-022, MEC-2016-718). All transplantations were performed adhering to the Declaration of Istanbul and all experiments were performed in accordance with the relevant guidelines and regulations of our institution and in accordance with the ethical standards of the Declaration of Helsinki. No transplants from inmates were used. Kidney function was assessed by estimated glomerular filtration rate (eGFR, mL/min per 1.73 m2, calculated by the CKD-EPI equation), serum creatinine (umol/L) and urine protein-to-creatinine ratio (mg/mmol) until graft failure or until a follow-up period of 7 years after transplantation. At time of transplantation all patients had a negative complement-dependent cytotoxicity cross-match. Transplant recipients received induction therapy with basiliximab [Simulect; Novartis, Basel, Switzerland; 20 mg intravenously on days 0 and 4]. The post-operative immunosuppressive regimen after transplantation consisted of tacrolimus (Prograf; Astellas Pharma, Tokyo, Japan; aiming for pre-dose concentrations of 10-15 ng/mL in weeks 1-2, 8-12 ng/mL in weeks 3-4 and 5-10 ng/ mL thereafter), MMF (Cellcept); Roche, Basel, Switzerland; starting dose of 1 g twice a day, aiming for pre-dose concentrations of 1.5-3.0 mg/L) and prednisolone. Prednisolone was tapered to 5 mg at month 3 and withdrawn at months 4-5. Only 'for cause' biopsies were performed in this patient cohort. Rejection was defined as biopsyproven acute rejection (BPAR) within the first 6 postoperative months by a renal pathologist using 2 μm paraffin sections stained for HE, PAS, Jones and immunohistochemistry for C4d on 4 μm sections. After the completion of the study, all biopsies were reviewed again by a clinical pathologist (M.C.C.) in a blinded fashion and scored according to the Banff '15 classification 37 .
Anti-HLA antibodies. Pre-transplant serum samples from recipients were screened for the presence of anti-HLA antibodies using the Lifecodes Lifescreen Deluxe (LMX) kit, according to the manufacturer's manual (Immucor Transplant Diagnostics Inc. Stamford, CT, USA). Thereafter, anti-HLA class I (HLA-A, HLA-B, and/ or HLA-C) or HLA class II (HLA-DR and/or HLA-DQ) antibodies were analyzed with a Luminex Single Antigen assay using LABscreen HLA class I and class II antigen beads (One Lambda, Canoga Park, GA, USA), as described in our previous study 13 . A cut-off mean fluorescence intensity value of 5000 was used to determine the presence of anti-HLA antibodies. The presence of donor-specific antibodies (DSA) was determined by comparing the measured HLA specificities with donor HLA typing.
IFN-γ and IL-21 ELISPOT assay. As described in our previous study 27  Statistical analysis. Statistical analyses were performed using SPSS 21.0 (SPSS Inc, Chicago, IL, US) and figures were made using GraphPad Prism version 6.01 (GraphPad, Inc., La Jolla, CA). The Mann-Whitney U-test was used to analyze the number of IFN-γ and IL-12 producing cells between patients with and without rejection. Data are presented as median and interquartile range (IQR). Pearson's chi-squared test was used to analyze the frequency of AR in patients with and without DSA. Receiver operating characteristic (ROC) curve analysis was used to calculate the cut-off value of number of donor-reactive IL-21 producing cells. Finally, multivariable binary logistic regression was performed to assess the odds ratio (OR) and 95% confidence interval (CI) for developing rejection. The regression was done using a stepwise backward selection method. A two-sided p-value ≤ 0.05 was considered statistically significant.

Results
Patient characteristics. In the study population of 114 patients, 30 (26.3%) patients experienced one or more rejections within the first six months after transplantation. First rejections were scored as 24 TCMR, 3 ABMR and 3 mixed TCMR and ABMR. Table 1 depicts the characteristics of the patients with and without rejection. Univariate analysis showed that donor age, historical PRA, presence of anti-HLA antibodies and DSA were significantly higher in patients who experienced rejection within the first 6 months after transplantation compared to non-rejectors.  www.nature.com/scientificreports/ resulted in an AUC of 0.64 and was able to discriminate patients with an early rejection with a sensitivity of 63% and specificity of 60% (Fig. 3). This cut-off also resulted in a positive predictive value (PPV) of 21.5% and negative predictive value (NPV) of 90.1%. These PPV and NPV were based on a rejection incidence of 15%. Because no association was found between IFN-γ and AR, no ROC curve was analyzed for this cytokine.

Discussion
The aim of this study was to investigate whether donor-reactive IFN-γ producing cells, donor-reactive IL-21 producing cells and B cell alloreactivity were associated with an increased rejection risk after kidney transplantation. We found that a higher donor age, the presence of pre-transplant DSA and a positive IL-21 assay were independent risk factors for the development of AR within 6 months after transplantation. Donor-reactive IL-21 producing cells have previously been found to be present in higher numbers in patients who develop AR 27 . Our data are in line with our earlier findings and this study confirms that a high number of donor-reactive IL-21 producing cells is a risk factors for AR. Both advanced donor age and the presence of pre-transplant DSA  www.nature.com/scientificreports/ have been previously reported as being risk factors for a higher incidence of rejection and/or graft loss [38][39][40][41] . In particular, monitoring of pre-transplant as well as de novo DSA has become more widespread with an increasing number of studies focusing on the relationship between the presence of DSA and incidence of ABMR [42][43][44][45] .
Although ABMR is often accompanied by the presence of DSA, this is not always the case, nor is the presence of DSA associated with adverse outcomes in all kidney transplant recipients [46][47][48] . While much remains unknown about the individual contributions of these risk factors, it is clear that donor-reactive immune memory can be difficult to detect and difficult to inhibit. Most research into the assessment of donor-reactive memory T cells, has been performed using the IFN-γ ELISPOT assay. In this study we found a trend towards a higher number of pre-transplant donor-reactive IFN-γ producing cells and AR. One possible explanation for the disparate findings is, that there are more patients with a humoral rejection component (20%) in our cohort compared to other studies which have primarily analyzed the association between donor-reactive IFN-γ producing cells in relation to acute cellular rejection [21][22][23]26 . As with other rejection biomarkers, implementation of the IFN-γ ELISPOT in clinical practice has been difficult due to its inability to predict clinical risk at the individual patient level. Similarly, the IL-21 ELISPOT assay may be limited in its capacity to accurately predict adverse clinical outcomes at an individual level, however, the high NPV of the assay may be of direct use in clinical practice. This may help to identify patients who are at reduced risk for the development of AR and may be ideal candidates for studies investigating the efficacy and safety of reduced intensity immunosuppressive protocols (for example in elderly patients). Following the data from van Besouw et al. published in 2019, this is the second study to find that a positive pre-transplant IL-21 ELISPOT assay is associated with AR. Due to the prominent role of IL-21 in both TCMR and ABMR, this assay may be well suited to predict several types of rejection. The IL-21 ELISPOT assay however, is not without its challenges. The assay has a relatively long incubation time of 44 h and requires the use of donor cells, as the predictive power was lost with third-party cells. Similar to other assays which require the use of donor cells, it is unlikely that results are available before a donor kidney from a deceased donor is transplanted. In its current form the assay is most useful for living donor transplants where PBMCs of the donor and the results of the IL-21 ELISPOT assay can be made available before the actual transplant is performed.
In addition to pre-transplant risk assessment, it would also be of great interest to assess the frequency of donor-reactive IL-21 producing cells during an active or chronic rejection event. The addition of the IL-21 ELISPOT assay to current pre-transplant screening may serve as a valuable addition to assess memory T cell reactivity, thereby providing a more complete view of the allogeneic immune response in transplant recipients.

Data availability
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.