Polyomavirus Nephropathy in ABO Blood Group-Incompatible Kidney Transplantation: Torque Teno Virus and Immunosuppressive Burden as an Approximation to the Problem

Abstract

Introduction: Earlier reports suggest that patients after ABO-incompatible kidney transplantation (ABOi) are at enhanced risk of developing BK-virus (BKV, also known as BK polyomavirus [BKPyV]) nephropathy (BKPyVAN). It remains elusive whether this is a result of more intense immunosuppression or an ABOi-associated "intrinsic attribute." To address this question, we measured Torque Teno virus (TTV) loads as a quantitative proxy for immunosuppressive depth in ABOi recipients and compared them to human leukocyte antigen-incompatible (HLAi, i.e. pretransplant donor-specific antibody-positive) and standard-risk transplant recipients.

Methods: Our retrospective study screened 2256 consecutive kidney transplantations performed between 2007 and 2020 at the Medical University of Vienna. Out of 629 in-principle eligible transplantations, we were able to include 465 patients: 42 ABOi, 106 HLAi, and 317 control recipients. Longitudinal TTV- polymerase chain reaction (PCR) and BKV-PCR was carried out at predefined timepoints and ranged from pretransplant until month 24 posttransplantation. TTV loads and immunosuppression were evaluated in the context of BKV-associated complications.

Results: ABOi recipients had a higher TTV load compared to HLAi and controls both at month 3 (median 1.5 × 10⁹ vs. 2.4 × 10⁸ vs. 9.1 × 10⁷; P = 0.010) and at month 6 (3.1 × 10⁹ vs. 1.4 × 10⁷ vs. 6.4 × 10⁷; P = 0.014) posttransplantation. Tacrolimus exposure was significantly higher in ABOi patients compared to HLAi and control patients (ABOi vs. HLAi: P = 0.007; ABOi vs. controls: P < 0.0001). Biopsy-proven BKPyVAN was more frequent in ABOi recipients when compared to HLAi and control recipients (11.9% vs. 2.8% vs. 4.1%; P = 0.046).

Conclusion: Our data support the assumption that ABOi patients are indeed at higher risk to develop BKPyVAN. A higher TTV load and immunosuppressive burden suggest that intense immunosuppression, rather than an "intrinsic attribute" conferred by ABOi, may contribute to this finding.

Keywords: ABOi transplantation; BK-virus nephropathy; Torque Teno virus; kidney transplantation.

Graphical abstract

Figure 1

Flowchart illustrating patient inclusion. Flowchart illustrating patient selection based on prespecified inclusion criteria. ABOi, ABO-incompatible; BKPyV, BK polyomavirus; HLAi, human leukocyte antigen–incompatible; Tx, transplantation.

Figure 2

TTV levels over the course of 24 months. Box plots illustrating TTV levels over the course of 24 months. Whiskers represent the 10th to 90th percentile. Values below and above the whiskers are drawn as individual points. At each time point, P-value refers to comparisons between 2 groups. Numbers in brackets represent patients with available parameters at each time point. ABOi, ABO-incompatible; HLAi, human leukocyte antigen–incompatible; M, month; ns, not significant; TTV, Torque Teno virus. ∗P < 0.05; ∗∗P < 0.01.

Figure 3

Tacrolimus trough levels at M1, M3, and M6 after transplantation. Box plots illustrating tacrolimus trough levels at M1, M3, and M6 after transplantation. Whiskers represent the 10th to 90th percentile. Values below and above the whiskers are drawn as individual points. At each time point, P-value refers to comparisons between 2 groups. Numbers in brackets represent patients with available parameters at each time point. ABOi, ABO-incompatible; HLAi, human leukocyte antigen–incompatible; M, month; ns, not significant; TTV, Torque Teno virus. ∗P < 0.05; ∗∗P < 0.01; ∗∗∗P < 0.001; ∗∗∗∗P < 0.0001.

Figure 4

AUC of tacrolimus trough levels within the first 12 and 26 weeks after transplantation. AUCs comparing tacrolimus trough levels between the 3 groups until (a) week 12 and (b) 26 weeks. Trough level AUC after 12 weeks included 5064 tacrolimus trough level medians (derived from 8471 single measurements) and trough level AUC after 26 weeks included 8342 trough level medians (based on 15,533 single measurements). ABOi, ABO-incompatible; AUC, Area under the curve; HLAi, human leukocyte antigen–incompatible.

Figure 5

Analysis of cumulative event rates for BKPyV outcomes. Cumulative event rates comparing BKPyV end points between the 3 groups. Log-rank test revealed no significant difference between the groups regarding (a) BKPyV-DNAemia (P = 0.35) or (b) presumptive BKPyVAN (P = 0.25). (c) Definite BKPyVAN was significantly different with highest events in ABOi patients (P = 0.045). ABOi, ABO-incompatible; BKPyV, BK polyomavirus; HLAi, human leukocyte antigen–incompatible.

Figure 6

Peak BKPyV-DNAemia. Box plots illustrate peak BKPyV-DNAemia in all 3 groups. Whiskers represent the 10th to 90th percentile. Values below and above the whiskers are drawn as individual points. ABOi, ABO-incompatible; BKPyV, BK polyoma virus; HLAi, human leukocyte antigen–incompatible; TTV, Torque Teno virus. ∗P < 0.05.