Control of BKPyV-DNAemia by a Tailored Viro-Immunologic Approach Does Not Lead to BKPyV-Nephropathy Progression and Development of Donor-Specific Antibodies in Pediatric Kidney Transplantation

Abstract

Polyomavirus BK (BKPyV)-associated nephropathy (BKPyV-nephropathy) remains a significant cause of premature kidney allograft failure. In the absence of effective antiviral treatments, current therapeutic approaches rely on immunosuppression (IS) reduction, possibly at the risk of inducing alloimmunity. Therefore, we sought to explore the long-term effects of a tailored viro-immunologic surveillance and treatment program for BKPyV on the development of alloimmunity and kidney graft outcome. Forty-five pediatric kidney transplant recipients were longitudinally monitored for BKPyV replication, virus-specific immunity, and donor-specific HLA antibodies (DSAs). DNAemia developed in 15 patients who were treated with stepwise IS reduction. Among the other 30 patients, 17 developed DNAuria without DNAemia and 13 always resulted as BKPyV-negative. All patients with DNAemia cleared BKPyV after having mounted a virus-specific cellular immune response, and no biopsy-proven BKPyV-nephropathy was observed. The presence of cytotoxic populations directed to the BKPyV Large-T (LT) antigen early after transplantation protected kidney recipients from developing BKPyV replication, and the appearance of LT-specific T cells in viruric patients prevented the development of BKPyV-DNAemia. In our cohort, no significant correlation was observed between BKPyV-DNAemia and the development of DSA and antibody-mediated rejection. However, patients who experienced and cleared BKPyV-DNAemia had a worse allograft survival at a median follow-up of 18.9 years (p = 0.048). These data need to be confirmed in larger cohorts.

Keywords: cellular immunity; donor-specific antibodies; humoral immunity; pediatric kidney transplantation; polyomavirus BK.

Figure 1

Flow diagram of study cohort and main outcome results for BKPyV infection, DSA, and graft survival. Left-bottom diagram: allograft survival in kidney graft recipients stratified by absence or presence of BKPyV replication. Right-bottom diagram: allograft survival in patients with or without BKPyV-DNAemia. Right-upper diagram: DSA development in patients with or without BKPyV-DNAemia. The statistical difference between Kaplan–Meier survival curves was evaluated with the log-rank test; differences with p values < 0.05 were considered statistically significant. BKPyV: polyomavirus BK; IS: immunosuppression; CNI: calcineurin inhibitors; DSA: donor-specific antibodies.

Figure 2

Kinetics of BKPyV-specific cellular and humoral immune responses in the KTx cohort. Data on the frequency of IFNγ-secreting lymphocytes (A,B) and virus-specific cytotoxicity (C,D), measured in patients’ cultured PBMCs, after a 10-day stimulation with VP1 (IFNγ-secreting lymphocytes: (A); cytotoxicity: (C)) and LT (IFNγ-secreting lymphocytes: (B); cytotoxicity: (D)) peptides, are reported. In (E), data on patients’ anti-BKV IgG are shown. Responses of KTx recipients who did not develop BKPyV-DNaemia and/or DNAuria (group 1, Neg), of DNAuric only (group 2, U+), and BKPyV-DNaemia-positive patients (U+P+, group 3) are reported. Blue dots represent pre-Tx data while white dots are for post-Tx measures; red rectangle represent median values. For groups 2 and 3, data obtained at the earliest time point before development of DNAuria (group 2) or DNAemia (group 3), at DNAuria or DNAemia increase, and at decrease and after viral clearance, are reported. IFNγ-secreting cells are represented as number of spots/10⁵ cells (median spots of triplicate experiments ± SD). Cytotoxicity is represented as % specific lysis at an effector to target ratio of 10:1 (mean of triplicate experiments ± SD). Differences among results obtained at the different time points were analyzed by the Kruskal–Wallis test (*: p < 0.05; **: p < 0.005; ***: p < 0.001).

Figure 3

Immune response to polyclonal T cell activator PHA, and its relationship with BKPyV-specific immune response, in the KTx cohort. In (A), data on the frequency of IFNγ-secreting lymphocytes in response to PHA in KTx recipients who did not develop BKPyV-DNaemia and/or -DNAuria (group 1, Neg), of DNAuric only (group 2, U+), and of BKPyV-DNaemia-positive patients (U+P+, group 3) are reported. For group 1, assessment was at month + 1 post-KTx, while for groups 2 and 3, data were obtained at the earliest time point before development of viruria (group 2) or viremia (group 3). IFNγ-secreting cells are represented as number of spots/10⁵ cells (single patients as white dots, median spots of triplicate experiments ± SD as boxes). In (B,C), the correlation between response to PHA, as depicted in panel (A), and BKPyV VP1 and LT antigens at the same points of time, is reported.

Figure 4

BKPyV LT-specific cytotoxicity in patients early after KTx. BKPyV LT-specific cytotoxicity, measured in samples obtained at 1 month after KTx in patients’ cultured PBMCs after 10-day stimulation with LT peptides, is reported. Responses of KTx recipients who did not develop BKPyV-DNAemia and/or DNAuria (group 1, Neg, blue dots), and of BKPyV-DNAuric only and DNAemic patients (U+/U+P+, groups 2 and 3, white dots) are reported. Cytotoxicity is represented as % specific lysis at an effector to target ratio of 10:1 (mean of triplicate experiments ± SD). Differences among results were analyzed by the Wilcoxon test. Red rectangle represent median values.

Figure 5

Kinetics of cellular immune responses in KTx recipients with BKPyV-DNAuria only versus BKPyV-DNaemia. Data on the frequency of IFNγ-secreting lymphocytes in response to LT peptides (A) or PHA (B) are reported. Responses of KTx recipients belonging to group 2 before (U_pre) and at peak (U_peak) DNAuria, and of group 3 DNAemic patients before replication (P_pre), at peak DNAuria (P_peak U) and at DNAemia clearance (P_post) are reported. IFNγ-secreting cells are reported as number of spots/10⁵ cells (median spots of triplicate experiments ± SD). Differences among results obtained at the different time points were analyzed by the Wilcoxon test; ns: not significant.