Clinical and immunological characteristics for BK polyomavirus‐associated nephropathy after kidney transplantation

Abstract Introduction BK polyomavirus (BKPyV)‐associated nephropathy (BKPyVAN) can cause a significant risk of allograft impairment after kidney transplantation (KT). Intact BKPyV‐specific immunity is associated with viral containment. This study investigated BKPyV‐specific immunological factors among KT recipients. Methods This prospective study in a single transplant center from January 2019 to August 2019 assessed associations between clinical and immunological characteristics, with a focus on BKPyV‐cell‐specific immunity and BKPyVAN, among KT recipients aged ≥15 years. The numbers of interferon‐gamma (IFN‐γ)‐producing CD4+ T, CD8+ T, natural killer (NK), and natural killer T (NKT) cells were measured after stimulation with large T antigen and viral capsid protein 1 (VP1). Results In total, 100 KT recipients were included (mean age ± SD, 42 ± 11 years); 35% of the recipients were female patients, and 70% had received induction immunosuppressive therapy. The 1‐year cumulative incidence of high‐level BKPyV DNAuria (possible BKPyVAN) and (presumptive BKPyVAN) was 18%. Among 40 patients with immunological factor data, pre‐KT %NK cells (hazard ratio [HR], 1.258; 95% confidence interval [CI], 1.077–1.469; p = .004) and %VP1‐specific NK cells (HR, 1.209; 95% CI, 1.055–1.386; p = .006) were factors independently associated with possible and presumptive BKPyVAN. KT recipients with possible and presumptive BKPyVAN were more likely to exhibit significant mean coefficients of %NK, %VP1‐specific NK, and %NKT cells at 1 month after KT than before KT (all p < .05). Conclusion Individuals with nonspecific and VP1‐specific NK cells before KT and increasing numbers of these cells after KT may be at risk for high‐level BKPyV DNAuria and presumptive BKPyVAN. Further studies are needed to determine the utility of BKPyV‐specific innate immune surveillance in predicting the occurrence of BKPyVAN.


| INTRODUCTION
BK polyomavirus (BKPyV) infection can cause BKPyVassociated nephropathy (BKPyVAN) in kidney transplantation (KT) recipients, resulting in allograft dysfunction.Immunosuppressive drugs are important for efforts to prevent allograft rejection.3][4] Previous studies have focused on the restoration of BKPyV-specific immune responses in KT recipients. 3,5,6A lack of BKPyV-specific humoral immunity (e.g., antibodies) or BKPyV-specific cellmediated immunity (e.g., T-cell responses before transplantation) is associated with BKPyV reactivation after KT. 7 Adequate BKPyV-specific T-cell immunity, both before KT and during the early post-KT period, is reportedly associated with clearance of BKPyV. 7][10][11] Here, we assessed the incidences of BKPyV DNAuria, DNAemia, and BKPyVAN, as well as the associations of NK and NKT cell-specific immune responses with BKPyVAN, within 1 year after KT.We hypothesized that the incidences of BKPyV DNAuria (possible BKPyVAN) and proven BKPyVAN would be comparable to the incidences in previous studies. 12We also hypothesized that NK and natural killer T (NKT) cell-specific immune responses would be associated with BKPyV DNAuria, DNAemia, and BKPyVAN in our cohort.

| Study design
This prospective cohort study was conducted at a single transplant center.All patients aged ≥15 years who underwent KT from January 2019 to August 2019 were enrolled.The primary endpoint was the 1-year cumulative incidence of BKPyV DNAuria (possible BKPyVAN), BKPyV DNAemia (presumptive BKPyVAN) and proven BKPyVAN.The secondary endpoints were DNAuria and BKPyVAN risk factors and outcomes.Finally, the exploratory outcomes were the associations of BKPyVspecific immunity with BKPyV DNAuria and presumptive BKPyVAN.Sample size analysis using a one-sample proportion test showed that, for 80% power in assessing the incidences of BKPyV DNAuria and presumptive BKPyVAN, 80 patients were required.Possible and proven BKPyVAN incidences of 20% (reported in previous studies [2][3][4]13 ) were used in this calculation. Conidering an expected dropout rate of 10%, we enrolled 100 patients.Allograft type, human leukocyte antigen match, panel reactive antibodies, and immunosuppressant type were recorded, along with clinical risk factors, immunological risk factors, and the outcomes of BKPyV DNAuria, BKPyV DNAemia, and proven BKPy-VAN.Preemptive measurements of urine BKPyV DNA loads were conducted before KT and at 1, 2, 3, 6, 9, and 12 months (±1 month) after KT (Supporting Information: Figure S1).Both blood and urine specimens were collected; for KT recipients with BKPyV DNAuria, assessments of archived plasma were conducted.Kidney biopsies were performed in accordance with nephrologist preferences.Because of laboratory constraints, we included all participants with evaluable immunological profiles in the initial analysis of BKPyV-specific immune cells.We investigated cluster of differentiation (CD)4 + T, CD8 + T, NK, and NKT cells that produced interferongamma (IFN-γ) after stimulation with BKPyV-specific antigens before KT and at 1-month post-KT, among subsets of cells from KT recipients; the results are reported as percentages.The study protocol was approved by the Institutional Review Board of the Faculty of Medicine at Ramathibodi Hospital, Mahidol University, Bangkok, Thailand (approval no.ID 10-61-11).

| Definition
BKPyVAN was defined in accordance with the American Society of Transplantation Infectious Diseases Community of Practice guidelines (Supporting Information: Table S1). 3,5Possible BKPyVAN was defined as BKPyV DNAuria >7 log10 copies/mL, presumptive BKPyVAN was defined as BKPyV DNAemia >4 log10 copies/mL, and proven BKPyVAN was defined via biopsy showing viral cytopathic changes, inflammatory infiltrates, tubulitis, or more than mild interstitial fibrosis/tubular atrophy.Immunohistochemistry was conducted to assess monoclonal antibody staining of polyomavirus simian virus 40 large T antigen (LT). 3,14Intracellular cytokine assays using VP1 and LT were performed to evaluate IFN-γ-producing CD4 + T cells, CD8 + T cells, NK cells, and NKT cells before and at 1 month ± 7 days after transplantation.

| Analysis of BK-specific T and NK cells
Peripheral blood mononuclear cell preparation for the quantification of virus-specific T and NK cells was performed as previously described. 15Briefly, heparinized blood was subjected to density gradient centrifugation using polysucrose-sodium diatrizoate (Lymphoprep™) purchased from Axis-Shield PoC AS.After separation, the cells were thoroughly washed with phosphatebuffered saline and resuspended in RPMI medium (Gibco) supplemented with 10% fetal calf serum (Gibco), then seeded in 96-well tissue culture plates (Corning Costar) at 1 × 10 5 cells/200 µL in each well.Subsequently, individual virus-specific antigens were added to each well.The BK antigens used were overlapping peptide pools (PepMix™ peptide pools) containing the LT and VP1 antigens, all purchased from JPT Peptide Technologies.In each overlapping peptide pool, the final concentration was 1 mg/mL.Wells containing only cells (no added antigen) served as negative controls.The plates were incubated at 37°C with 5% CO 2 in humidified air for 12 h before the addition of brefeldin A (eBioscience) to each well.The plates were incubated for an additional 6 h before cell fixation and staining, as described below.
Next, the cells were fixed with 1% formaldehyde (Sigma-Aldrich) for 15 min at room temperature.After fixation, the cells were washed, resuspended in 0.5% saponin (Sigma-Aldrich), and incubated at room temperature for 15 min; they were stained with fluorescencetagged antibodies suspended in 0.5% saponin.The antibody cocktail consisted of a 1:100 dilution of FITC-CD3, PE-CD56, APC-CD4, eFluor780-CD8, and PE-Cy7 IFN-γ.All antibodies were procured from eBioscience.Cells were incubated in the antibody cocktail for 30 min at 4°C, then subjected to flow cytometry analysis.

| Statistical analyses
The cumulative incidence of BKPyVAN was estimated using Kaplan-Meier analysis.A Cox proportional hazards model was used to analyze clinical and immunological risk factors for BKPyV DNAuria.Categorical variables are presented as absolute numbers and relative frequencies; continuous variables are presented as means with standard deviations.The percentages of IFN-γproducing CD4 + T, CD8 + T, NK, and NKT cells before and 1 month after KT were compared by mixed linear regression analysis.p Values < 0.05 were considered statistically significant.Statistical analyses were conducted, and linear plots of BKPyV-specific immune cells were constructed, using Stata v.16 statistical software (StataCorp).

| DISCUSSION
This study investigated immunological factors, including nonspecific and BKPyV-specific T-cell, NK-cell, and NKT-cell immunity, in KT recipients who experienced BKPyVAN within 1 year after KT.The results showed that individuals who had underlying diabetic nephropathy and BKPyV-specific innate immunity before transplantation, indicated by the presence of BKPyV-specific NK cells that secreted IFN-γ after stimulation with VP1 antigen, were more likely to develop BKPyV DNAuria within 1 year after KT.Additionally, the proportion of VP1-specific NK cells in KT recipients showed a tendency to increase after transplantation, as indicated by BKPyV DNAuria, despite post-transplant immunosuppression.The incidences of BKPyV DNAuria and BKPyV DNAemia in our KT recipients were comparable to the findings in previous studies. 12,16Two retrospective studies reported that the prevalence of BKPyVAN was 10%-12% among KT recipients with similar demographic characteristics.However, the rate of possible BKPyVAN in our study was relatively high (20%), which could be explained by proactive monitoring to detect early BKPyV DNAuria.Furthermore, none of our patients had proven BKPyVAN at the time of initial diagnosis, presumably because we conducted early optimization of immunosuppression; this finding considerably differs from the results of studies, in which 6.4%-8.6% of patients had proven BKPyVAN. 12,16Furthermore, we reaffirmed the importance of underlying diabetes as a predictive factor for BKPyV DNAuria. 3 Because no effective anti-BKPyV agent is available, immunosuppressive therapy modification is a key management approach for the restoration of immune function.Therefore, BKPyV-specific immune surveillance may be a useful tool for posttreatment prevention and disease monitoring. 1,7,17Thus far, virus-specific Tcell immunity has played an important role in controlling viral replication in both solid organ and hematopoietic stem cell transplant recipients. 189][20] Although most studies have investigated the role of cytomegalovirus-specific T-cell immunity and their findings have been implemented in clinical practice, there is limited data regarding BKPyV-specific immunity. 21,22daptive immune responses, particularly T-cell responses, constitute the main mechanism for controlling BKPyV infection.Increases in CD4+ and CD8 + T cells were observed in patients with DNAemia who achieved clearance of BKPyV. 7LT-specific CD8 + T cells and VP1-specific CD4 + BKPyV-specific T cells may be particularly important. 8Furthermore, a recent study showed that KT recipients with BKPyVAN tended to exhibit recovery of LT-specific CD4 + T-cell responses after their immunosuppression treatment had been optimized. 19e speculate that increased pre-transplant BKPyVspecific innate immunity leads to greater percentages of NK and VP1-specific NK cells after KT, which may be associated with BKPyVAN.The innate immune system has key roles in suppressing viral replication and activating adaptive immunity to eradicate viruses. 23nflammatory NK cell antiviral responses consist of interactions between immunoglobulin-like receptors on NK cells and major histocompatibility complex class I molecules on virus-infected cells, which influence their sensitivity to lysis by NK cells. 24NK cells eradicate BKPyV by inducing apoptosis in infected cells through antibody-dependent cellular cytotoxicity. 25Ischemic injury to an allograft may reactivate BKPyV in that allograft, thereby triggering the innate immune system.NK cells recognize virus-infected cells through the downregulation of major histocompatibility complex class I receptors and inhibitory receptors, as well as the  upregulation of activator molecules. 23There is evidence that the proportion of activating NK-cell immunoglobulin-like receptors is significantly smaller in patients with BKPyVAN than in healthy controls. 8,23,26NKT cells have recently gained attention as a bridge between the innate and adaptive immune systems.In a herpes virus infection model, NKT cells functioned as T-helper cells and as cytotoxic T cells. 27KT cells can promote early antibody-based immunity after viral infection by enhancing B-cell antibody responses. 28Previous reviews have suggested that innate immunity plays a crucial role in BKPyVAN; however, our results contradict this notion.Thus, there is a need to explore the role and mechanism of innate immunity in BKPyVAN, particularly concerning NK cells, dendritic cells, neutrophils, eosinophils, Toll-like receptors, and chemokines.
To our knowledge, only a few studies have identified immunological factors associated with BKPyV infection.A notable strength of the present study is that it explored an early stage of BKPyV DNAuria (possible BKPyVAN) and early immune responses, such as the innate immunity represented by BKPyV-specific NK cell responses.We also emphasized the importance of both innate and adaptive BKPyV-specific immunity in the development of BKPyV DNAuria.We believe that our findings will address knowledge gaps in this field. 3Our findings highlight the need for a post-KT BKPyV screening surveillance system for KT recipients, considering the limited conventional treatment options.Additionally, BKPyV-specific T-cell and antibody responses may serve as adjunct markers of viral clearance, offering guidance for clinicians who encounter this infection.However, this study also had some limitations.First, although urine BKPyV DNA load can be used to screen for BKPyVAN, 3,29 it has a less obvious relationship with proven BKPyVAN and possibly lower cost-effectiveness, compared with plasma BKPyV DNA load. 30Second, we did not adjust for BKPyV-specific humoral immunity, which is reportedly associated with BKPyVAN.Third, our study was relatively small, which may have hindered our ability to identify other relevant variables.Fourth, we included patients aged ≥15 years because this is the age of adulthood in our country.However, the immune systems of adolescents and adults may differ; further assessments of BKPyVAN in adolescents are needed.We note that there is limited information regarding comparisons of BKPyV DNAuria, BKPyV DNAemia, and immunological factors between adults and children who have undergone KT.Thus, a larger prospective study involving both adults and children is warranted.Fifth, the lack of an IFN-γ response may be related to either the absence of specific T or NK cells or the impairment of IFN-γ production.This aspect could be addressed by tests involving tetramers.Finally, the lack of commercial assays could restrict the application of our findings to clinical practice, particularly concerning the use of high donor BKPyV-specific antibody titers as a marker of recent viral exposure and potentially higher BKPyV viral load in allografts. 31,32n summary, we found that initial levels and post-KT increases in NK, NKT, and VP1-specific NK cells were greater in KT recipients who exhibited BKPyV DNAuria within 1 year after KT.BKPyV-specific NKand NKT-cell immune surveillance could be used to identify patients at risk for post-transplant viral reactivation.Further studies regarding the underlying mechanisms and clinical impact are needed to confirm this association.

ACKNOWLEDGMENTS
This study was supported by the Infectious Disease Association of Thailand-Institut Merieux (research grant 2019) and research funding from the Faculty of Medicine Ramathibodi Hospital, Mahidol University, Bangkok, Thailand (RF_62047 and CF_65001).

F
I G U R E 1 Flowchart of patient selection for this study.BKPyV, BK polyomavirus; BKPyVAN, BKPyV-associated nephropathy.overtime from pre-KT to 1-month post-KT.Linear plots of those BKPyV-specific immune cells in BKPyVAN-free KT recipients, before and 1 month after transplantation, are presented in Figure4.Analyses of laboratory-developed intracellular cytokine assays measuring the percentages of IFN-γ-producing CD4 + T cells, CD8 + T cells, NK cells, and NKT cells after incubation with LT and VP1 are shown in Supporting Information: FigureS2; PE-CD56 staining results and findings in control cells are shown in Supporting Information: FigureS3.

F I G U R E 2
Cumulative incidences of possible and presumptive BKPyV-associated nephropathy.(A, B) Kaplan-Meier plots of possible (A) and presumptive (B) BKPyV-associated nephropathy within 1 year after KT.BKPyV, BK polyomavirus; KT, kidney transplantation.T A B L E 2 Factors associated with possible and presumptive BKPyVAN among kidney transplant recipients.