The Banff Working Group Classification of Definitive Polyomavirus Nephropathy: Morphologic Definitions and Clinical Correlations : Journal of the American Society of Nephrology

Journal Logo

Advanced Search
Clinical Research

The Banff Working Group Classification of Definitive Polyomavirus Nephropathy: Morphologic Definitions and Clinical Correlations

Nickeleit, Volker1; Singh, Harsharan K.1; Randhawa, Parmjeet2; Drachenberg, Cinthia B.3; Bhatnagar, Ramneesh3; Bracamonte, Erika4; Chang, Anthony5; Chon, W. James6; Dadhania, Darshana7; Davis, Vicki G.1; Hopfer, Helmut8; Mihatsch, Michael J.8; Papadimitriou, John C.3; Schaub, Stefan9; Stokes, Michael B.10; Tungekar, Mohammad F.11; Seshan, Surya V.12;  on behalf of the Banff Working Group on Polyomavirus Nephropathy

Collaborators

Nickeleit, Volker; Singh, Harsharan K.; Randhawa, Parmjeet; Drachenberg, Cinthia B.; Bhatnagar, Ramneesh; Bracamonte, Erika; Chang, Anthony; Chon, W. James; Dadhania, Darshana; Davis, Vicki G.; Hopfer, Helmut; Mihatsch, Michael J.; Papadimitriou, John C.; Schaub, Stefan; Stokes, Michael B.; Tungekar, Mohammad F.; Seshan, Surya V.

Author Information

1Division of Nephropathology, Department of Pathology and Laboratory Medicine, The University of North Carolina School of Medicine, Chapel Hill, North Carolina;

2Division of Transplantation Pathology, Department of Pathology, University of Pittsburgh Medical Center–Montefiore, Pittsburgh, Pennsylvania;

3Department of Pathology, School of Medicine, University of Maryland, Baltimore, Maryland;

4Department of Pathology, The University of Arizona College of Medicine, Tucson, Arizona;

5Department of Pathology, The University of Chicago, Chicago, Illinois;

6Renal Transplant Program, University of Missouri–Kansas City School of Medicine/Saint Luke's Health System, Kansas City, Missouri;

7Division of Nephrology and Hypertension, Department of Medicine, New York Presbyterian Hospital–Weill Cornell Medical Center, New York, New York;

8Institute for Pathology and

9Transplantation Immunology and Nephrology, University Hospital of Basel, Basel, Switzerland;

10Department of Pathology, Columbia Presbyterian Medical Center, New York, New York;

11Histopathology Department, St. Thomas’ Hospital, Guy’s and St. Thomas Foundation Trust and King’s College London, London, United Kingdom; and

12Department of Pathology, Weill Cornell Medicine, New York, New York

Correspondence: Dr. Volker Nickeleit, Division of Nephropathology, University of North Carolina School of Medicine at Chapel Hill, Brinkhous-Bullitt Building, Room 409, Campus Box 7525, Chapel Hill, NC 27599-7525. Email: [email protected]

Journal of the American Society of Nephrology 29(2):p 680-693, February 2018. | DOI: 10.1681/ASN.2017050477
  • Free
  • SDC
  • SDC

Abstract

Polyomavirus nephropathy (PVN), usually caused by the BK-virus strain, is a significant infectious complication affecting renal allografts.1–8 Post kidney transplantation PVN evolves gradually with initial reactivation of latent polyomaviruses in the uro-genital tract and viruria in approximately 30%–50% of patients, followed in 30%–40% of affected individuals by viremia and in a subgroup of 20%–40% of viremic patients by morphologically apparent viral nephropathy. 6,7,9 The term “definitive” PVN has been coined for patients with biopsy proven viral nephropathy in comparison to “presumptive” PVN for patients with high viremia titers but no histologic confirmation of renal disease. Characterization of early events during the pathogenesis of PVN and the clinical overlap between “presumptive” and “definitive” disease has led to preemptive therapeutic strategies in order to prevent disease progression and the development of morphologically apparent virally induced kidney injury. Patient screening and preemptive treatment protocols are well established and part of recommendations for patient management post renal transplantation.6,10–16

Clinical indications for a kidney biopsy that can result in a histologic diagnosis of definitive PVN are multifold and dependent on the local standard of care practice and available resources. Biopsies in patients with viremia/patients with presumptive PVN are obtained routinely in some centers to establish or exclude a diagnosis of definitive PVN, assess the degree of acute and chronic tissue injury, and search for other renal diseases that might affect allograft function and patient management.6,11,17 This practice is fueled by the fact that quantitative PCR tests to assess BK viremia are not standardized, and BK viremia titers only imperfectly reflect the degree of viral renal injury, thereby leaving diagnostic uncertainty.7,8,17–23 In the vast majority of patients, renal allograft biopsies are obtained in the setting of clinically unexplained deterioration of graft function that can be due to unsuspected PVN. Definitive PVN can also be unexpectedly observed in surveillance biopsies of stable grafts8,18,19 or occasionally, in patients with polyomavirus infections other than BK virus.20,21 Thus, renal allograft biopsies obtained in various clinical settings and for different indications can show morphologic signs of PVN. In developed countries, the incidence of biopsy-confirmed definitive PVN is approximately 5%–6%, with broad transplant center variations.22,23 The highest incidence of definitive PVN is found in ABO-incompatible grafts (18%) and highly sensitized allograft recipients after desensitization (20%).22,24

Although morphologic changes and the clinical presentation in patients with definitive PVN vary,8 consensus on how to subclassify viral nephropathy is lacking. Previous classification attempts developed from expert opinion have not gained broad acceptance.25–29 The objective of developing a consensus histologic classification of definitive PVN was broadly shared by over 100 renal pathologists and nephrologists responding to a questionnaire circulated by the Renal Pathology Society, the Banff Group on Allograft Pathology, and the Nephropathology Working Group of the European Society of Pathology. The needs to standardize terminology and improve diagnostic communication, comparability of clinical data, and outcome results between centers were broadly acknowledged. In response, the “Banff community” created a multicenter working group of nephrologists and pathologists with the goal of “developing a clinically relevant morphologic classification for PVN.”30,31 The Banff Working Group on Polyomavirus Nephropathy collected the largest cohort of patients with definitive PVN that has been systematically analyzed thus far. This study on the basis of findings in diagnostic index biopsies resulted in the development of a morphologic classification scheme reflecting three clinical parameters: (1) presentation at time of diagnosis, (2) renal function after index biopsy, and (3) graft failure. This classification of definitive PVN, discussed at previous Banff meetings, is presented here.

Results

Study Cohort

A cohort of 192 patients with biopsy-proven definitive PVN transplanted between 1996 and 2008 (50% between 2006 and 2008) was collected by the working group: median age of recipients was 53 years, 74% were men, 49% were white, 32% were black, 20% were other ethnicities, and 64% of organs were of deceased donor source (Table 1).

Table 1. - Descriptive statistics for study patients with definitive PVN
Demographics and Baseline Characteristics a , b Measurement Value
Age, n=192 b Median (IQR) c 52.5 (40–63.5)
Men, n=192 b N (%) 142 (74)
Race, n=190 b
 White N (%) 93 (49)
 Black N (%) 60 (32)
 Latino N (%) 21 (11)
 Asian N (%) 13 (7)
 Other N (%) 3 (2)
Donor source, n=191 b
 Deceased N (%) 122 (64)
 Living: related N (%) 39 (20)
 Living: unrelated N (%) 30 (16)
Renal biopsy before PVN index biopsy, n=192 b N (%) 112 (58)
Biopsy-proven acute allograft rejection before PVN, n=192 b N (%) 47 (24)
Week of PVN index biopsy post-transplantation, n=188 b Median (IQR) c 28 (16–71)
At PVN index biopsy
 >15% Increase in S-Cr over baseline, n=175 b N (%) 142 (81)
 Hematuria, minimal, n=130 b N (%) 25 (19)
 Proteinuria, minimal, n=147 b N (%) 70 (48)
 Plasma PCR readings (×104), n=126 b Median (IQR) c 14.0 (1.2–70.7)
PVN index biopsy lacking viral inclusion bodies, n=188 b N (%) 36 (19)
PVN index biopsy with at least two biopsy cores N (%) 137 (71)
 Diagnostic PVN changes limited to one core, n=137 b N (%) 39 (28)
 Diagnostic PVN changes only in medulla, n=137 b N (%) 13 (9)
Follow-up of 24 mo post-PVN index biopsy
 Allograft failure, n=189 b N (%) 56 (30)
 PVN clearance, n=159 b N (%) 117 (74)
 Time to clearance, wk, n=114 b Median (IQR) c 17 (12–44)
aMedians and IQRs are given for continuous variables; counts and percentages are given for categorical variables. Percentages are on the basis of the number of subjects with data available.
bSample size (N) for the complete cohort is 192 subjects, with changes in sample size due to either subgroup analyses or missing data.
cIQR (25th percentile to 75th percentile).

PVN was diagnosed in the index biopsy between 4 and 582 weeks post-transplantation (median of 28 weeks). During the time before diagnosis of PVN, 58% of patients had undergone renal biopsy with a diagnosis of acute allograft rejection in 24% (median interval of 14 weeks between rejection episode and index biopsy; interquartile range [IQR], 10–35). At time of diagnostic PVN index biopsy, 81% of patients presented with elevated serum creatinine (S-Cr) levels (>15% above lowest baseline reading noted within 4 months before diagnosis), 19% of patients had microscopic hematuria, and 48% had minimal proteinuria (≤1 g/24 hours; no patient presented with an active urine sediment or nephrotic-range proteinuria). Diagnostic changes were limited to one of several index biopsy cores in 28% of patients or the medulla (sparing the renal cortex) in 9%. In 19% of biopsies lacking characteristic intranuclear viral inclusion bodies, diagnostic confirmation was obtained by immunohistochemistry (IHC) with a positive SV40-T staining reaction.32 During 24 months of follow-up, PVN cleared (repeat biopsy examination lacking evidence of PV replication and/or negative plasma PCR testing/clearance of BK viremia) in 74% of patients within a median of 17 weeks (IQR, 12–44 weeks) after index biopsy. Graft failure (return to dialysis, retransplantation, and/or S-Cr≥7 mg/dl) occurred in 30% of patients within 24 months. No patient death occurred due to PVN.

In patients from whom data were available before index biopsy, 98% (184 of 188) were on an immunosuppressive protocol containing calcineurin inhibitors (tacrolimus, 94% or cyclosporin, 4%), 92% (171 of 186) received mycophenolate mofetil, 53% (99 of 186) were on triple immunosuppressive therapy (tacrolimus, mycophenolate mofetil, and steroids), 7% (14 of 188) received sirolimus, and 32% (49 of 153) were on a steroid-free protocol. After index biopsy, immunosuppression was lowered in 96% (174 of 182) of patients (including switching from mycophenolate mofetil or sirolimus to leflunomide); 68% (127 of 186) received leflunomide, cidofovir, or intravenous Igs.

Definition of PVN Disease Classes

Banff activity and chronicity scores, polyomavirus replication/load level (pvl), the presence/absence of intranuclear polyomavirus inclusion bodies types 1 and 2, age, race, sex, graft type, plasma BK-PCR readings at the time of index biopsy, and week of index biopsy/PVN diagnosis after grafting were evaluated as potential predictors of allograft function and graft failure over 24 months. Individual mixed model repeated measures (MMRM) analyses in a first step revealed that five of the parameters tested were significantly associated at α=0.05 with function/increases of S-Cr levels over time: pvl (P<0.001), the presence of viral inclusion bodies (P<0.001), Banff ci scores (P=0.004), plasma PCR levels at index biopsy (P=0.01), and Banff ti scores (P=0.04) (Supplemental Table 1, Serum Creatinine Change). Using a forward selection approach with an MMRM model starting with the most significant effect from step 1, it was determined that, after pvl was included in the model (along with the covariates baseline S-Cr, center, and visit), only the Banff ci score showed a marginal trend to the model fit (P=0.06) (Supplemental Table 2, Serum Creatinine Change). When the same parameters were individually evaluated relative to graft failure within 24 months with forward logistic regression adjusting for center (Supplemental Table 1, Graft Failure within 24 mo), three were significantly associated: Banff ci score (P=0.02), pvl (P=0.03), and Banff ti score (P=0.04). After inclusion of the Banff ci score in a forward selection approach, neither pvl nor the Banff ti score explained additional variation in the graft failure rate at 24 months (Supplemental Table 2, Graft Failure within 24 mo). Notably, although an examination of race effects suggested a trend for higher S-Cr levels among patients of African descent versus other races (median increase at 24 months: 1.3 versus 0.9 mg/dl) as well as higher rates of graft failure (37% versus 27%), neither difference was statistically significant after covariate adjustment in the models described above (P=0.21 and P=0.17, respectively). Thus, with pvl and Banff ci scores accounting for the most variation in both allograft function/changes in S-Cr levels and graft failure over 24 months, respectively, the combination of these two parameters formed the basis for the PVN disease classification.

Using the three pvl scores as defined (Concise Methods or the information in parentheses below) and dividing Banff ci scores into two groups (zero or one versus two or three), six preliminary categories were generated, whose allograft function over 24 months is presented in Supplemental Figure 1. (Histologic scoring of intrarenal pvl: pvl was semiquantitatively scored on the basis of the overall percentage of tubules with morphologic evidence of polyomavirus replication. According to this definition, a tubule with intranuclear viral inclusion bodies [type 1 or 2] and/or a diagnostic IHC staining reaction for SV40-T antigen in one or more tubular epithelial cells per tubular cross-section was considered “one positive tubule.” Note that tubules with intranuclear viral inclusion bodies can, on occasion, be SV40-T “negative”; because inclusion bodies clearly mark viral replication, such tubules [even in the absence of SV40-T positivity] were scored as “positive.”17,32 In general, the pvl score is mainly influenced by the degree of SV40-T expression by IHC. The overall percentage of all positive tubular cross-sections [SV40-T expressing and occasionally only showing intranuclear viral inclusion bodies] was estimated in the entire biopsy sample. The percentage positive was on the basis of evaluation of all available cores and all tubules/ducts in the cortex and medulla, including all negative tubules in zones/biopsy cores lacking diagnostic evidence of polyomavirus replication using ×10 scanning magnification and confirming the presence of positive tubules at ×20 or ×40. On the basis of expert input during the initial study design, three levels of pvl were defined: pvl 1: <1%; pvl 2: 1%–10%; pvl 3: >10% positive tubules/ducts.) The depicted six categories allowed for grouping patients into those with best, worst, and intermediate overall outcome after further refinement, including the exclusion of 14 patients (14 of 192; 7%) with concurrent acute rejection (C4d positivity and/or Banff type 2 rejection) in the index biopsy. Accordingly, three morphologic PVN classes were defined that were significantly different from each other in both overall function/S-Cr levels and graft survival (see below) (Table 2):

Table 2. - Histologic classification system of PVN: Definitions
Biopsy-Proven PVN a Class 1 Biopsy-Proven PVN a Class 2 Biopsy-Proven PVN a Class 3
pvl Banff ci Score pvl Banff ci Score pvl Banff ci Score
1 0–1 1 2–3 — —
— — 2 0–3 — —
— — 3 0–1 3 2–3
—, na.
aHistologic classes of definitive PVN are defined by the morphologic degree of intrarenal pvl and Banff ci scores. The pvl scoring is on the basis of the extent of virally induced tubular changes. A tubule with intranuclear viral inclusion bodies (type 1 or 2) and/or a positive IHC reaction for SV40-T antigen in one or more cells per tubular cross-section is considered “a positive tubule.” The overall percentage of positive tubular cross-sections is estimated in the entire biopsy sample (all available cores, cortex, and medulla): pvl 1: ≤1% of all tubules/ducts with viral replication; pvl 2: >1% to ≤10% of all tubules/ducts with viral replication; pvl 3: >10% of all tubules/ducts with viral replication. In PVN classes 1–3, interstitial inflammation and tubulitis can vary from Banff scores ti 0 to ti 3/t 0 to t 3. PVN class 1 often lacks a significant inflammatory reaction. PVN and rejection (acute, chronic, cell mediated, and/or antibody mediated) can occur concomitantly. PVN classes are recorded in the context of the Banff classification scheme and established guidelines for Banff lesion scoring that should be recorded in parallel. To adequately establish or exclude a diagnosis of definitive PVN, two biopsy cores including portions of medulla in at least one of the two cores are required.

  • PVN Class 1: pvl 1, ci≤1; (Figure 1);
    fig1
    Figure 1.:
    PVN class 1 (pvl 1, Banff ci 0). (A and B) Patient 1. The renal medulla only shows three tubules with polyomavirus replication in an SV40-T stain (arrows in B), representing <1% of all tubules (cortex and medulla) found in a total of two biopsy cores (pvl 1); there are no diagnostic intranuclear viral inclusion bodies, no interstitial fibrosis, and no inflammation. (C–F) Patient 2. The renal cortex shows seven tubules with SV40-T expression (arrows in D), representing <1% of all tubules (cortex and medulla) in a total of two biopsy cores (pvl 1); (C) there are no diagnostic intranuclear viral inclusion bodies, there is no fibrosis (ci 0), and there is only patchy minor inflammation. Note that parenchyma and tubules lacking significant changes by light microscopy (E) show viral replication by IHC (F) with a crisp diagnostic intranuclear staining pattern. (A, C, and E) Hematoxylin and eosin–stained sections and (B, D, and F) IHC on formalin-fixed and paraffin-embedded tissue sections to detect the polyomavirus large T antigen (simian virus large T; mouse mAb, clone 416, DP02; Calbiochem). Original magnification, ×10 (A, B), ×20 (C, D), and ×40(E, F).
  • PVN Class 2: pvl 1, ci≥2 OR pvl 2, any ci score OR pvl 3, ci≤1 (Figure 2); and
    fig2
    Figure 2.:
    PVN class 2 (pvl 3, Banff ci 0). The cortex shows several areas with polyomavirus replication both by light microscopy (intranuclear viral inclusion bodies; arrows in A, C, and E) and in an SV40-T stain with crisp intranuclear staining (B, D, and F), representing >10% of all tubules (cortex and medulla) found in a total of two biopsy cores (pvl 3); (A) there is only very minimal fibrosis (Banff ci 0) and patchy interstitial inflammation with edema. (A, C, and E) Hematoxylin and eosin–stained sections and (B, D, and F) IHC on formalin-fixed and paraffin-embedded tissue sections to detect the polyomavirus large T antigen (simian virus large T; mouse mAb, clone 416, DP02; Calbiochem). Original magnification, ×10 (A, B), ×20 (C, D), and ×40 (E, F).
  • PVN Class 3: pvl 3, ci≥2 (Figure 3).
    fig3
    Figure 3.:
    PVN class 3 (pvl 3, Banff ci 3). The cortex shows several areas with polyomavirus replication mainly in an SV40-T stain (with intranuclear staining of different intensities; B, D, and F), representing >10% of all tubules (cortex and medulla) found in a total of two biopsy cores (pvl 3); (A) there is diffuse interstitial fibrosis (Banff ci 3), tubular atrophy, and inflammation. (A) Trichrome-stained section, (C and E) hematoxylin and eosin–stained sections, and (B, D, and F) IHC on formalin-fixed and paraffin-embedded tissue sections to detect the polyomavirus large T antigen (simian virus large T; mouse mAb, clone 416, DP02; Calbiochem). Original magnification, ×10 (A, B), ×20 (C, D), and ×40 (E, F).
In the study cohort, 25% of patients (44 of 178) were categorized as PVN class 1, 63% (112 of 178) were categorized as class 2, and 12% (22 of 178) were categorized as class 3. Intranuclear viral inclusion bodies (types 1 and 2) by light microscopy were absent in 48% (20 of 42) of patients in class 1, 9% (ten of 110) of patients in class 2, and 5% (one of 22) of patients in class 3 (P<0.001). In PVN class 1, 55% (18 of 33) of patients only showed morphologic evidence of PV replication (viral inclusions and/or SV40-T positivity) in one of several biopsy cores; in class 2, 22% (18 of 81) and in class 3, 0% (zero of 14; P<0.001). PVN was only found in the medulla (sparing the renal cortex) in 24% (eight of 33) in class 1, 2% (two of 82) in class 2, and 0% (zero of 14) in class 3 (P<0.001). Significant inflammation and tubulitis were less common in PVN class 1: Banff ti score 0: class 1, 34% (15 of 44); class 2, 9% (ten of 112); class 3, 0% (zero of 22) (P<0.001); Banff t score 0: class 1, 39% (17 of 44); class 2, 19% (21 of 112); class 3, 0% (zero of 21) (P<0.001).

PVN Disease Classes and Clinical Presentation at the Time of Index Biopsy

Within 4 months before index biopsy, all PVN disease classes 1–3 had similar baseline renal function, with lowest S-Cr readings (median) of 1.4 mg/dl (class 1), 1.5 mg/dl (class 2), and 1.8 mg/ dl (class 3; P=0.13) (Table 3). At the time of diagnostic index biopsy, significant differences in clinical presentation were noted. Class 1 was diagnosed early: 18 weeks (median) after grafting in patients with a modest rise in S-Cr levels (median change from baseline to time of index biopsy, 0.3 mg/dl; 33% of patients with only minor S-Cr rises not exceeding 15% above baseline levels). In comparison, classes 2 and 3 were diagnosed later (medians of 30 and 54 weeks after transplantation, respectively; P<0.01), with more pronounced increases in S-Cr levels compared with baseline (class 2 increase: 0.6 mg/dl, median value; class 3: 0.8 mg/dl, median value; P<0.001). In contrast to class 1, only a few patients in class 2 (17%) and no patients in class 3 presented with <15% above baseline S-Cr rises at time of index biopsy (P=0.002). Plasma PCR levels in PVN disease classes at time of index biopsy showed vast overlap with statistically significant differences only noted for PVN class 1 compared with class 3 (Table 3).

Table 3. - Clinical presentation at time of diagnostic PVN index biopsy
Clinical Parameters PVN P Value a
Class 1 Class 2 Class 3
Time between transplant and PVN diagnosis, wk
 Median 18 30 54 <0.01
 IQR b 14–52 19–78 30–100
 N 44 110 20
Plasma PCR readings (×104) at index biopsy
 Median 3.9 9.5 25.0 0.05
 IQR b 0.1–27.5 1.2–75.0 14.9–120.0
 N 33 73 11
Baseline S-Cr, mg/dl within 4 mo preindex biopsy
 Median 1.4 1.5 1.8 0.13
 IQR b 1.2–1.7 1.3–1.8 1.2–2.1
 N 41 105 21
Peak S-Cr, mg/dl at index biopsy
 Median 1.8 2.1 2.7 <0.001
 IQR b 1.5–2.3 1.8–2.6 2.0–4.0
 N 42 111 19
Change in S-Cr, mg/dl baseline to peak
 Median 0.3 0.6 0.8 <0.001
 IQR b 0.1–0.7 0.3–1.0 0.6–2.0
 N 39 104 19
Change in S-Cr baseline to peak, %
 Median 21 39 62 0.001
 IQR b 13–56 21–64 33–100
 N 39 104 19
Patients with <15% change in S-Cr baseline to peak
 N (%) 13 (33) 18 (17) 0 (0) 0.002
aP values for the medians were on the basis of the Kruskal–Wallis test (one-way ANOVA of the rank scores). P value for the percentages was on the basis of the Cochran–Mantel–Haenszel chi-squared test for nonzero Spearman correlation. P values for all individual pairwise comparisons were significant at α=0.05 for all variables, except baseline S-Cr (with no significant pairwise comparisons) and plasma PCR readings (class 1 versus 2, P=0.15; class 1 versus 3, P=0.02; class 2 versus 3, P=0.08).
bIQR (25th percentile to 75th percentile).

PVN Classes and Outcome

Renal Function over 24 Months of Follow-Up after Index Biopsy

During follow-up, PVN class 1 fared well, with a median increase of S-Cr readings by 0.4 mg/dl at 24 months (IQR, 0.0–1.6 mg/dl) (Figure 4, Supplemental Figure 2). In comparison, PVN class 3 showed progressive functional deterioration with a median increase of S-Cr by 4.8 mg/dl at 24 months (IQR, 1.3–5.3 mg/dl). PVN class 2 had a median increase of 1.0 mg/dl at 24 months (IQR, 0.4–4.8 mg/dl). A linear MMRM on the ranks, controlling for baseline S-Cr and study center, revealed significantly increasing S-Cr levels with worsening PVN class at month 12 (1 versus 2: P<0.001; 1 versus 3: P<0.001; 2 versus 3: P=0.02) and month 24 (1 versus 2: P=0.01; 1 versus 3: P<0.001; 2 versus 3: P=0.02). Differences were most pronounced between classes 1 and 3. Of note, differences remained after limiting the follow-up analysis to 86 patients by excluding individuals with rejection or known additional complications that might have skewed the analysis (i.e., patients with recurrence of renal disease, generalized infections/sepsis, cardiovascular complications, malignancies; in this subgroup of 86 patients, the small sample sizes, in particular in PVN class 3, precluded statistical testing).

fig4
Figure 4.:
Median change in S-Cr from baseline (y axis) is plotted by visit (x axis) for PVN classes 1–3. A linear MMRM on the ranks, controlling for baseline S-Cr and study center (n=178 study subjects), revealed significantly increasing S-Cr levels with increasing PVN class over time, highlighted here at month 12 (1 versus 2: P<0.001; 1 versus 3: P<0.001; 2 versus 3: P=0.02) and month 24 (1 versus 2: P=0.01; 1 versus 3: P<0.001; 2 versus 3: P=0.02).

Graft Failure within 24 Months after Index Biopsy

During follow-up, 16% of grafts failed in PVN class 1, 31% of grafts failed in PVN class 2, and 50% of grafts failed in PVN class 3 (P=0.004) (Table 4). A subanalysis excluding patients with presumed rejection episodes or other severe complications (e.g., generalized infections/sepsis, recurrence of renal disease, or malignancies) after index biopsy as a significant contributing factor to graft loss showed lower overall failure rates with maintained differences between PVN disease classes (class 1: 7%; class 2: 25%; class 3: 38%; P=0.01).

Table 4. - Graft failure rates at 24 mo
PVN Disease Classification Graft Failure at 24 mo, N (%) a
All Patients, n=175 Patients without Other Complications after Index Biopsy, n=92 b
Class 1 7/44 (16) 2/28 (7)
Class 2 34/109 (31) 13/51 (25)
Class 3 11/22 (50) 5/13 (38)
P value c 0.004 0.01
aThe percentage is on the basis of the number of patients with graft failure within 24 mo (N) of the total number of patients who fall into that PVN disease class.
bExcluded are patients with presumed rejection episodes, generalized infections/sepsis, recurrence of renal disease, or malignancies.
cP value was on the basis of the Cochran–Mantel–Haenszel chi-squared test for nonzero Spearman correlation using the midrank scores.

ARF at the time of index biopsy with S-Cr levels >2.5 mg/dl increased the odds of graft failure 5.3 times (95% confidence interval for the odds ratio, 2.6 to 10.9; P<0.001) with maintained relative differences between PVN classes. Within the PVN classes, the degree of inflammation judged by the Banff ti score did not add additional prognostic information (Supplemental Table 3).

PVN Resolution

Within 24 months, PVN resolved in 110 of 149 (74%) patients (on the basis of first repeat biopsy without PVN and/or first repeat plasma PCR testing without detectable BK viremia): resolution in 75% of patients in class 1, 78% of patients in class 2, and 50% of patients in class 3 (differences between groups P=0.17) (Table 5). Median time to clearance (n=104 patients): 12 weeks in PVN class 1, 24 weeks in PVN class 2, and 14 weeks in PVN class 3 (P=0.26). A subanalysis limited to those patients presenting with only repeat negative plasma PCR testing showed similar findings.

Table 5. - PVN resolution
Clinical Parameters PVN P Value a
Class 1 Class 2 Class 3
All subjects
 PVN resolution b
  N (%) 27/36 (75) 74/95 (78) 9/18 (50) 0.17
 Time to PVN resolution, wk
  Median 12 24 14 0.26
  IQR c 8–42 12–44 12–32.5
   N 26 70 8
Subjects with negative plasma PCR readings during follow-up
 PVN resolution d
  N (%) 20/29 (69) 54/73 (74) 5/14 (36) 0.15
 Time to clearance, wk
  Median 12 24 48 0.23
  IQR c 12–48 12–48 12–48
   N 20 54 5
aP values for resolution were on the basis of the Cochran–Mantel–Haenszel chi-squared test for nonzero Spearman correlation using the midrank scores. P values for time to PVN resolution and clearance were on the basis of the Kruskal–Wallis test (one-way ANOVA of the rank scores).
bPVN resolution for all subjects was on the basis of the earliest occurrence of either negative repeat biopsy or negative BK viremia by PCR during follow-up; data are available from n=149 subjects.
cIQR (25th percentile to 75th percentile).
dPVN resolution for only subjects with negative plasma PCR readings during follow-up were on the basis of negative BK viremia by PCR during follow-up; data are available from n=116 subjects.

Discussion

The objective put forward by the Banff renal transplant community was to “develop a clinically relevant morphologic classification for definitive PVN.”30,31 Because previous expert opinion–based single-center classification attempts failed to gain broad acceptance,25–27 this task was assigned to a multicenter Banff Working Group on Polyomavirus Nephropathy that assembled the largest cohort of patients with biopsy-proven PVN ever studied systematically. Morphologic changes, including the histologic assessment of intrarenal pvl, were scored, and the findings correlated with clinical parameters. On the basis of comprehensive statistical analyses, including mixed effects statistical models to identify predictive histologic features, the study resulted in the development of a morphologic grading scheme of definitive PVN reflecting three clinical parameters: (1) clinical presentation at time of initial index diagnosis, (2) renal function after index biopsy, and (3) graft failure. These PVN disease classes add value to clinical information in the context of established patient management protocols after kidney transplantation (Figure 5).6,11

fig5
Figure 5.:
PVN. Flow chart illustrating recommendations for risk assessment and patient management after renal transplantation. The novel morphologic Banff Working Group on Polyomavirus Nephropathy classification of definite PVN with three disease classes (blue box) adds prognostic value and enhances clinical communication. Urinary PV-Haufen Test.7 , 46 , 47 *Negative viremia/viruria. Differential diagnosis includes false negative PCR result due to mutant BK virus strain, presence of JC virus, or presence of Simian Virus 40. †Other renal diseases can concur with PVN, including calcineurin inhibitor toxicity, recurrent or de novo renal disease, and rejection.

Our approach of histologically classifying definitive PVN is similar to existing classification schemes for lupus nephritis, IgA nephropathy, or rejection events in solid organ transplants, all of which aim to identify clinically significant morphologic changes in renal biopsies from patients with complex clinical presentations.30,33–36 Because specific and effective antipolyomavirus therapy is not available, patients with PVN have commonly been treated by reduction of the overall maintenance immunosuppression, and consequently, outcome data seem to be less affected by differences in therapy.6

In this study, assessing pvl added a new histologic approach to lesion scoring in the context of established “Banff” criteria, and the combination of pvl with the Banff interstitial fibrosis ci score formed the basis for the proposed PVN classification scheme. A low pvl and fibrosis score, defining PVN class 1, was seen comparatively early after grafting. Class 1 presumably reflects the initial disease stage, in which virally induced tissue injury is very focal and is not fully developed (i.e., in our analysis, intranuclear polyomavirus inclusion bodies and significant inflammation [Banff ti score ≥1] were absent in approximately 50% and 30% of biopsies, respectively). Not surprisingly in this context, 33% of patients with PVN class 1 presented with stable graft function at the time of diagnosis, and overall outcome was favorable. In contrast, more pronounced interstitial scarring and/or higher pvl scores, defining PVN classes 2 and 3, were noted significantly later after transplantation. PVN classes 2 and 3 represent more advanced disease stages, with intratubular viral replication, tubular injury, and often, graft scarring affecting function and graft survival. The predictive value of the three PVN disease classes was maintained in patient populations of various backgrounds. Severe acute allograft failure at the time of PVN index diagnosis increased the overall probability of graft loss with maintained relative differences between disease classes. Overall graft failure in all PVN classes was affected by rejection episodes and other severe illnesses, underscoring the complexity of outcome analyses. Interestingly, patients in all PVN classes showed similar rates of disease resolution of >50% as early as 2–3 months after diagnosis.

Our study did not include specific reproducibility analyses. The Banff interstitial fibrosis score was grouped in a binary fashion (i.e., >25% or <25%), thereby simplifying the assessment and making it “more robust” (compare with ref. 37). Interobserver agreement rates for interpreting IHC SV40-T staining results have been reported in reproducibility studies to range from fair to substantial, and it was previously concluded that “…BKV immunohistochemistry is reproducible between observers…(especially if)…simplified to a single-feature schema.”38 Of note, one previous study specifically addressing agreement rates on pvl scoring found a high concordance coefficient of 0.72.39 These reports are in agreement with our own unpublished working group observations: in an online survey circulated during the initial phase of our efforts, we found agreement rates for pvl scoring between 60% and 80% for the three-tier scoring approach—agreement rates similar to those reported for the detection of/agreement on the presence of transplant endarteritis/Banff type 2 acute rejection.40 Thus, the Banff Working Group on Polyomavirus Nephropathy approach on classifying PVN on the basis of ci and pvl scores seems to be robust and suited for routine diagnostic workup.

Attempts to morphologically classify definitive PVN have been reported previously.17,25–29,41 All reports, including the current working group classification, agree on emphasizing the predictive diagnostic value of certain histologic features. Not surprisingly, however, major differences in morphologic PVN disease class definitions exist, such as the prognostic significance of interstitial inflammation stressed in some reports but not stressed in our analysis.17,26–29 There are various reasons for these discrepancies: (1) arbitrarily set definitions of “expert morphologic criteria”17,26–28; (2) differences in inclusion criteria for definitive PVN (i.e., requiring the presence of detectable intranuclear viral inclusion bodies in most studies,27,29 which would have led to exclusion of a significant number of patients with early PVN cases only expressing SV40-T antigen in our working group report); (3) single-center versus multicenter analyses; (4) small study cohorts; (5) differences in the era of patient collections27,29; (6) timing of diagnostic allograft biopsies after transplantation (reported in some studies late, with effects on PVN morphology and outcome29); (7) differences in study end points; (8) differences in the statistical approach; and (9) differences in exclusion criteria (such as rejection in this study). We found fibrosis rather than inflammation to constitute a significant negative histologic prognosticator in PVN. This observation is in agreement with previous molecular studies describing urinary levels of PAI-1 mRNA as markers of graft failure.42 In this context, it is tempting to speculate that possibly, in PVN, reconstitution type of inflammation can create a “benign” rather than destructive inflammatory microenvironment, facilitating healing and a favorable outcome—thus, in PVN, not all inflammation is detrimental.43 The here-described multicenter approach defining morphologic disease classes of definitive PVN on the basis of in-depth statistical analysis is advancing previous classification attempts.

Limitations of this study should be recognized. (1) First and foremost, this is a retrospective observational analysis with study material collected by participating centers on the basis of local standards of care guidelines and laboratory test results. (2) Analyses were focused on findings at the time of diagnostic index biopsy; our study was not designed to specifically follow the dynamics of PVN class changes in repeat biopsies. (3) Some critique might also be raised that PVN classes 2 and 3 are defined by active (i.e., pvl) and chronic lesions (i.e., interstitial scarring). (4) From an individual patient perspective, PVN classes 1 (“best outcome”) and 3 (“worst outcome”) carry the strongest predictive values; renal function and outcome in PVN class 2 can vary most. In the future, the PVN classification proposed here by the Banff Working Group on Polyomavirus Nephropathy will need to be further validated in a mixed population, independent of the cohort in which this model was originally derived and tested. Specifically, it would be desirable to use uniformly and prospectively collected datasets controlling for tissue adequacy and comorbidities, including rejection.

An important issue in renal (allograft) pathology, especially in patients with cases of focally accentuated diseases, is the question of sample adequacy and how much tissue is required for a reliable diagnosis. Diagnosing definitive PVN is no exception, because morphologic changes, in particular in disease class 1, can be limited to few nephrons or one of several biopsy cores or even confined to the renal medulla as reported here and elsewhere.17,27 For optimal diagnostic workup, we recommend that two biopsy cores including renal medulla and IHC staining results for SV40-T antigen are available. IHC for SV40-T antigen expression seems to be superior to other assays, including IHC with antibodies directed against viral capsid proteins or in situ hybridization, because early polyomavirus genes comprising large T antigen are expressed first when late gene products or intranuclear viral inclusion bodies are still undetectable. We recommend that, if an interpretation is “negative for definitive PVN” and if the PVN adequacy criteria are not met, a specific comment as to the limitation of the interpretation should be rendered. This practice becomes crucial in the setting of an elevated risk profile with BK viremia and viruria (Figure 5).

In conclusion, we have described a systematic, multicenter approach to developing a morphologic classification scheme for definitive PVN on the basis of the largest cohort of patients studied thus far. The here-described “evidence- and statistic-based” approach to define PVN classes differs from previous “expert opinion–based” classification attempts. It is our contention that the proposed, readily applicable PVN classification provides prognostic information during routine biopsy workup worldwide, especially at the time of index diagnosis. The Banff Working Group classification of definitive PVN should be conducted in the framework of established clinical algorithms for patient risk assessment (Figure 5) as well as “Banff” lesion scoring.34,36 Standardized morphologic assessment and reporting will add value to clinical information. It is a crucial cornerstone to improve comparability of clinicopathologic studies, outcome results, and therapeutic trials. A simple diagnosis of PVN in a biopsy report no longer suffices.

Concise Methods

The Multicenter Banff Working Group on Polyomavirus Nephropathy

The Banff Working Group on Polyomavirus Nephropathy consists of nine transplant centers in the United States and Europe (Columbia Presbyterian Medical Center, New York, NY; Guy’s and St. Thomas Foundation Trust and King’s College, London, United Kingdom; University of Arizona College of Medicine, Tucson, AZ; University Hospital Basel, Basel, Switzerland; University of Chicago, Chicago, IL; University of Maryland, Baltimore, MD; University of North Carolina, Chapel Hill, NC; University of Pittsburgh Medical Center, Pittsburgh, PA; and Weill Cornell Medicine, New York, NY). Each center retrospectively contributed patients with cases of definitive PVN (between nine and 41; total n=192) who were diagnosed histologically and managed according to local standard of care guidelines. Locally collected demographic and clinical data and all biopsy results were systematically analyzed up to 24 months after initial histologic diagnosis of definitive PVN (i.e., time of index biopsy). S-Cr measurements were collected within 4 months before diagnosis (with the lowest reading recorded as “baseline level”), at the time of index biopsy (“peak level”), and subsequently, at 1, 3, 6, 12, and 24 months of follow-up. All S-Cr values in the clinical setting of graft failure (or values recorded >7 mg/dl) were imputed with a value of 7 mg/dl at time of graft failure and all preset time points thereafter.

Histologic Analyses

A histologic diagnosis of PVN was established on the basis of a typical IHC staining reaction for SV40-T (simian virus large T) antigen in epithelial cell nuclei that was often but not always accompanied by the detection of typical intranuclear viral inclusion bodies.6,10–15,27,32,44 All index biopsies, defined for study purposes as the first allograft biopsy with definitive PVN and fulfilling Banff adequacy criteria, were initially scored locally and subsequently centrally (by V.N., who was blinded to the clinical dataset) according to Banff guidelines.34,45 The degree of interstitial fibrosis was semiquantitatively scored in trichrome stains.

Histologic Scoring of pvl

The pvl was semiquantitatively scored on the basis of the overall percentage of tubules with morphologic evidence of polyomavirus replication. According to this definition, a tubule with intranuclear viral inclusion bodies (type 1 or 2 as the most typical variants) by light microscopy10,32 and/or a diagnostic IHC staining reaction for SV40-T antigen in one or more tubular epithelial cells per tubular cross-section was considered “one positive tubule.” (Note that tubules with intranuclear viral inclusion bodies can, on occasion, be SV40-T negative; because inclusion bodies clearly mark viral replication, such tubules [even in the absence of SV40-T positivity] were scored as “positive.”17,32 In general, the pvl score is mainly influenced by the degree of SV40-T expression by IHC.) The overall percentage of all positive tubular cross-sections (SV40-T expressing and occasionally only showing intranuclear viral inclusion bodies) was estimated in the entire biopsy sample. The percentage positive was on the basis of evaluation of all available cores and all tubules/ducts in the cortex and medulla, including all negative tubules in zones/biopsy cores lacking diagnostic evidence of polyomavirus replication using ×10 scanning magnification and confirming the presence of positive tubules/cells at ×20 or ×40. On the basis of expert input during the initial study design, three levels of pvl were defined: pvl 1: ≤1% positive tubules/ducts; pvl 2: 1%–10% positive tubules/ducts; pvl 3: >10% positive tubules/ducts.

Statistical Analyses

The central review histologic scoring results were used for statistical analysis. Descriptive statistics are presented, with medians and IQRs for continuous variables and counts and percentages for categorical variables. Analysis of graft function over time was performed using an MMRM. The model included fixed categorical effects for center and visit, with a continuous fixed covariate for the baseline S-Cr reading. The association of various parameters with S-Cr levels was determined with separate models. Significant predictors of allograft function over time were then analyzed using a forward approach, starting with the parameter explaining the most variance in function and adding additional predictors to test for further significant improvements to the model fit. Because of the skewed nature of the S-Cr levels, a rank transformation was used in the MMRM for both the change from baseline reading and the baseline covariate; median S-Cr values are displayed in the figures. The same set of predictors was also evaluated with respect to graft failure using a logistic regression with forward selection, adjusting for study center. Finally, patients were categorized into PVN classes on the basis of the values of parameters identified as independently explaining variation in either graft function or failure. An MMRM that included effects for baseline S-Cr reading, study center, visit, PVN class, and PVN class by visit interaction was then used to compare function at 12 and 24 months between the PVN classes. All statistical tests were two sided, with α=0.050 (Supplemental Material has more detailed information on statistics).

Disclosures

None.

Published online ahead of print. Publication date available at www.jasn.org.

See related editorial, “Banff Classification of Polyomavirus Nephropathy: A New Tool for Research and Clinical Practice,” on pages .

This article contains supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2017050477/-/DCSupplemental.

The authors thank all local transplant teams for their generous support and dedication; numerous hours were spent by many colleagues to provide crucial information. The Banff Working Group on Polyomavirus Nephropathy also thanks the Banff community for providing a supportive platform that promoted fruitful discussions and facilitated fine tuning of the proposed polyomavirus nephropathy classification scheme.

References

1. Purighalla R, Shapiro R, McCauley J, Randhawa P: BK virus infection in a kidney allograft diagnosed by needle biopsy. Am J Kidney Dis 26: 671–673, 19957573026
2. Nickeleit V, Hirsch HH, Binet IF, Gudat F, Prince O, Dalquen P, Thiel G, Mihatsch MJ: Polyomavirus infection of renal allograft recipients: From latent infection to manifest disease. J Am Soc Nephrol 10: 1080–1089, 199910232695
3. Binet I, Nickeleit V, Hirsch HH, Prince O, Dalquen P, Gudat F, Mihatsch MJ, Thiel G: Polyomavirus disease under new immunosuppressive drugs: A cause of renal graft dysfunction and graft loss. Transplantation 67: 918–922, 199910199744
4. Drachenberg CB, Beskow CO, Cangro CB, Bourquin PM, Simsir A, Fink J, Weir MR, Klassen DK, Bartlett ST, Papadimitriou JC: Human polyoma virus in renal allograft biopsies: Morphological findings and correlation with urine cytology. Hum Pathol 30: 970–977, 199910452511
5. Hirsch HH: Polyomavirus BK nephropathy: A (re-)emerging complication in renal transplantation. Am J Transplant 2: 25–30, 200212095052
6. Hirsch HH, Randhawa P; AST Infectious Diseases Community of Practice: BK polyomavirus in solid organ transplantation. Am J Transplant 13[Suppl 4]: 179–188, 201323465010
7. Nickeleit V, Singh HK: Polyomaviruses and disease: Is there more to know than viremia and viruria? Curr Opin Organ Transplant 20: 348–358, 201525933251
8. Nankivell BJ, Renthawa J, Sharma RN, Kable K, O’Connell PJ, Chapman JR: BK virus nephropathy: Histological evolution by sequential pathology. Am J Transplant 17: 2065–2077, 201728371308
9. Hirsch HH, Knowles W, Dickenmann M, Passweg J, Klimkait T, Mihatsch MJ, Steiger J: Prospective study of polyomavirus type BK replication and nephropathy in renal-transplant recipients. N Engl J Med 347: 488–496, 200212181403
10. Nickeleit V, Hirsch HH, Zeiler M, Gudat F, Prince O, Thiel G, Mihatsch MJ: BK-virus nephropathy in renal transplants-tubular necrosis, MHC-class II expression and rejection in a puzzling game. Nephrol Dial Transplant 15: 324–332, 200010692517
11. Hirsch HH, Brennan DC, Drachenberg CB, Ginevri F, Gordon J, Limaye AP, Mihatsch MJ, Nickeleit V, Ramos E, Randhawa P, Shapiro R, Steiger J, Suthanthiran M, Trofe J: Polyomavirus-associated nephropathy in renal transplantation: Interdisciplinary analyses and recommendations. Transplantation 79: 1277–1286, 200515912088
12. Brennan DC, Agha I, Bohl DL, Schnitzler MA, Hardinger KL, Lockwood M, Torrence S, Schuessler R, Roby T, Gaudreault-Keener M, Storch GA: Incidence of BK with tacrolimus versus cyclosporine and impact of preemptive immunosuppression reduction. Am J Transplant 5: 582–594, 200515707414
13. Schaub S, Hirsch HH, Dickenmann M, Steiger J, Mihatsch MJ, Hopfer H, Mayr M: Reducing immunosuppression preserves allograft function in presumptive and definitive polyomavirus-associated nephropathy. Am J Transplant 10: 2615–2623, 201021114642
14. Kuypers DR: Management of polyomavirus-associated nephropathy in renal transplant recipients. Nat Rev Nephrol 8: 390–402, 201222508181
15. Hardinger KL, Koch MJ, Bohl DJ, Storch GA, Brennan DC: BK-virus and the impact of pre-emptive immunosuppression reduction: 5-Year results. Am J Transplant 10: 407–415, 201020055811
16. Seifert ME, Gunasekaran M, Horwedel TA, Daloul R, Storch GA, Mohanakumar T, Brennan DC: Polyomavirus reactivation and immune responses to kidney-specific self-antigens in transplantation. J Am Soc Nephrol 28: 1314–1325, 201727821629
17. Nickeleit V, Mihatsch MJ: Polyomavirus nephropathy in native kidneys and renal allografts: An update on an escalating threat. Transpl Int 19: 960–973, 200617081225
18. Renoult E, Coutlée F, Pâquet M, St Louis G, Girardin C, Fortin MC, Cardinal H, Lévesque R, Schürch W, Latour M, Barama A, Hébert MJ: Evaluation of a preemptive strategy for BK polyomavirus-associated nephropathy based on prospective monitoring of BK viremia: A kidney transplantation center experience. Transplant Proc 42: 4083–4087, 201021168633
19. Buehrig CK, Lager DJ, Stegall MD, Kreps MA, Kremers WK, Gloor JM, Schwab TR, Velosa JA, Fidler ME, Larson TS, Griffin MD: Influence of surveillance renal allograft biopsy on diagnosis and prognosis of polyomavirus-associated nephropathy. Kidney Int 64: 665–673, 200312846764
20. Delbue S, Ferraresso M, Ghio L, Carloni C, Carluccio S, Belingheri M, Edefonti A, Ferrante P: A review on JC virus infection in kidney transplant recipients. Clin Dev Immunol 2013: 926391, 201323424601
21. Milstone A, Vilchez RA, Geiger X, Fogo AB, Butel JS, Dummer S: Polyomavirus simian virus 40 infection associated with nephropathy in a lung-transplant recipient. Transplantation 77: 1019–1024, 200415087764
22. Sharif A, Alachkar N, Bagnasco S, Geetha D, Gupta G, Womer K, Arend L, Racusen L, Montgomery R, Kraus E: Incidence and outcomes of BK virus allograft nephropathy among ABO- and HLA-incompatible kidney transplant recipients. Clin J Am Soc Nephrol 7: 1320–1327, 201222626962
23. Boan P, Hewison C, Swaminathan R, Irish A, Warr K, Sinniah R, Pryce TM, Flexman J: Optimal use of plasma and urine BK viral loads for screening and predicting BK nephropathy. BMC Infect Dis 16: 342, 201627448566
24. Gabardi S, Townsend K, Martin ST, Chandraker A: Evaluating the impact of pre-transplant desensitization utilizing a plasmapheresis and low-dose intravenous immunoglobulin protocol on BK viremia in renal transplant recipients. Transpl Infect Dis 15: 361–368, 201323647907
25. Drachenberg RC, Drachenberg CB, Papadimitriou JC, Ramos E, Fink JC, Wali R, Weir MR, Cangro CB, Klassen DK, Khaled A, Cunningham R, Bartlett ST: Morphological spectrum of polyoma virus disease in renal allografts: Diagnostic accuracy of urine cytology. Am J Transplant 1: 373–381, 200112099383
26. Colvin RB, Nickeleit V: Renal transplant pathology. In: Heptinstall’s Pathology of the Kidney, 6th Ed., edited by Jennette JC, Olson JL, Schwartz MM, Silva FG, Philadelphia, Lippincott Williams & Wilkins, 2007, pp 1345–1490
27. Drachenberg CB, Papadimitriou JC, Hirsch HH, Wali R, Crowder C, Nogueira J, Cangro CB, Mendley S, Mian A, Ramos E: Histological patterns of polyomavirus nephropathy: Correlation with graft outcome and viral load. Am J Transplant 4: 2082–2092, 200415575913
28. Hirsch HH, Randhawa P; AST Infectious Diseases Community of Practice: BK virus in solid organ transplant recipients. Am J Transplant 9[Suppl 4]: S136–S146, 200920070673
29. Masutani K, Shapiro R, Basu A, Tan H, Wijkstrom M, Randhawa P: The Banff 2009 working proposal for polyomavirus nephropathy: A critical evaluation of its utility as a determinant of clinical outcome. Am J Transplant 12: 907–918, 201222390378
30. Mengel M, Sis B, Haas M, Colvin RB, Halloran PF, Racusen LC, Solez K, Cendales L, Demetris AJ, Drachenberg CB, Farver CF, Rodriguez ER, Wallace WD, Glotz D; Banff Meeting Report Writing Committee: Banff 2011 meeting report: New concepts in antibody-mediated rejection. Am J Transplant 12: 563–570, 201222300494
31. Haas M, Sis B, Racusen LC, Solez K, Glotz D, Colvin RB, Castro MC, David DS, David-Neto E, Bagnasco SM, Cendales LC, Cornell LD, Demetris AJ, Drachenberg CB, Farver CF, Farris AB 3rd, Gibson IW, Kraus E, Liapis H, Loupy A, Nickeleit V, Randhawa P, Rodriguez ER, Rush D, Smith RN, Tan CD, Wallace WD, Mengel M; Banff Meeting Report Writing Committee: Banff 2013 meeting report: Inclusion of c4d-negative antibody-mediated rejection and antibody-associated arterial lesions. Am J Transplant 14: 272–283, 201424472190
32. Nickeleit V, Mengel M, Colvin RB: Renal transplant pathology. In: Heptinstall’s Pathology of the Kidney, 7th Ed., edited by Jennette JC, Olson JL, Silva FG, D’Agati VD, Philadelphia, Wolters Kluwer, 2015, pp 1321–1459
33. Weening JJ, D’Agati VD, Schwartz MM, Seshan SV, Alpers CE, Appel GB, Balow JE, Bruijn JA, Cook T, Ferrario F, Fogo AB, Ginzler EM, Hebert L, Hill G, Hill P, Jennette JC, Kong NC, Lesavre P, Lockshin M, Looi LM, Makino H, Moura LA, Nagata M: The classification of glomerulonephritis in systemic lupus erythematosus revisited. J Am Soc Nephrol 15: 241–250, 200414747370
34. Racusen LC, Solez K, Colvin RB, Bonsib SM, Castro MC, Cavallo T, Croker BP, Demetris AJ, Drachenberg CB, Fogo AB, Furness P, Gaber LW, Gibson IW, Glotz D, Goldberg JC, Grande J, Halloran PF, Hansen HE, Hartley B, Hayry PJ, Hill CM, Hoffman EO, Hunsicker LG, Lindblad AS, Yamaguchi Y: The Banff 97 working classification of renal allograft pathology. Kidney Int 55: 713–723, 19999987096
35. Cattran DC, Coppo R, Cook HT, Feehally J, Roberts IS, Troyanov S, Alpers CE, Amore A, Barratt J, Berthoux F, Bonsib S, Bruijn JA, D’Agati V, D’Amico G, Emancipator S, Emma F, Ferrario F, Fervenza FC, Florquin S, Fogo A, Geddes CC, Groene HJ, Haas M, Herzenberg AM, Hill PA, Hogg RJ, Hsu SI, Jennette JC, Joh K, Julian BA, Kawamura T, Lai FM, Leung CB, Li LS, Li PK, Liu ZH, Mackinnon B, Mezzano S, Schena FP, Tomino Y, Walker PD, Wang H, Weening JJ, Yoshikawa N, Zhang H; Working Group of the International IgA Nephropathy Network and the Renal Pathology Society: The Oxford classification of IgA nephropathy: Rationale, clinicopathological correlations, and classification. Kidney Int 76: 534–545, 200919571791
36. Loupy A, Haas M, Solez K, Racusen L, Glotz D, Seron D, Nankivell BJ, Colvin RB, Afrouzian M, Akalin E, Alachkar N, Bagnasco S, Becker JU, Cornell L, Drachenberg C, Dragun D, de Kort H, Gibson IW, Kraus ES, Lefaucheur C, Legendre C, Liapis H, Muthukumar T, Nickeleit V, Orandi B, Park W, Rabant M, Randhawa P, Reed EF, Roufosse C, Seshan SV, Sis B, Singh HK, Schinstock C, Tambur A, Zeevi A, Mengel M: The Banff 2015 kidney meeting report: Current challenges in rejection classification and prospects for adopting molecular pathology. Am J Transplant 17: 28–41, 201727862883
37. Furness PN, Taub N; Convergence of European Renal Transplant Pathology Assessment Procedures (CERTPAP) Project: International variation in the interpretation of renal transplant biopsies: Report of the CERTPAP Project. Kidney Int 60: 1998–2012, 200111703620
38. Adam B, Randhawa P, Chan S, Zeng G, Regele H, Kushner YB, Colvin RB, Reeve J, Mengel M: Banff Initiative for Quality Assurance in Transplantation (BIFQUIT): Reproducibility of polyomavirus immunohistochemistry in kidney allografts. Am J Transplant 14: 2137–2147, 201425091177
39. Sar A, Worawichawong S, Benediktsson H, Zhang J, Yilmaz S, Trpkov K: Interobserver agreement for Polyomavirus nephropathy grading in renal allografts using the working proposal from the 10th Banff conference on allograft pathology. Hum Pathol 42: 2018–2024, 201121733554
40. Colvin RB, Cohen AH, Saiontz C, Bonsib S, Buick M, Burke B, Carter S, Cavallo T, Haas M, Lindblad A, Manivel JC, Nast CC, Salomon D, Weaver C, Weiss M: Evaluation of pathologic criteria for acute renal allograft rejection: Reproducibility, sensitivity, and clinical correlation. J Am Soc Nephrol 8: 1930–1941, 19979402096
41. Drachenberg CB, Hirsch HH, Ramos E, Papadimitriou JC: Polyomavirus disease in renal transplantation: Review of pathological findings and diagnostic methods. Hum Pathol 36: 1245–1255, 200516311117
42. Dadhania D, Snopkowski C, Muthukumar T, Lee J, Ding R, Sharma VK, Christos P, Bang H, Kapur S, Seshan SV, Suthanthiran M: Noninvasive prognostication of polyomavirus BK virus-associated nephropathy. Transplantation 96: 131–138, 201323778650
43. Menter T, Mayr M, Schaub S, Mihatsch MJ, Hirsch HH, Hopfer H: Pathology of resolving polyomavirus-associated nephropathy. Am J Transplant 13: 1474–1483, 201323721552
44. Randhawa PS, Finkelstein S, Scantlebury V, Shapiro R, Vivas C, Jordan M, Picken MM, Demetris AJ: Human polyoma virus-associated interstitial nephritis in the allograft kidney. Transplantation 67: 103–109, 19999921805
45. Solez K, Colvin RB, Racusen LC, Haas M, Sis B, Mengel M, Halloran PF, Baldwin W, Banfi G, Collins AB, Cosio F, David DS, Drachenberg C, Einecke G, Fogo AB, Gibson IW, Glotz D, Iskandar SS, Kraus E, Lerut E, Mannon RB, Mihatsch M, Nankivell BJ, Nickeleit V, Papadimitriou JC, Randhawa P, Regele H, Renaudin K, Roberts I, Seron D, Smith RN, Valente M: Banff 07 classification of renal allograft pathology: Updates and future directions. Am J Transplant 8: 753–760, 200818294345
46. Singh HK, Andreoni KA, Madden V, True K, Detwiler R, Weck K, Nickeleit V: Presence of urinary Haufen accurately predicts polyomavirus nephropathy. J Am Soc Nephrol 20: 416–427, 200919158358
47. Singh HK, Reisner H, Derebail VK, Kozlowski T, Nickeleit V: Polyomavirus nephropathy: Quantitative urinary polyomavirus-Haufen testing accurately predicts the degree of intrarenal viral disease. Transplantation 99: 609–615, 201525136849
Keywords:

Polyomavirus Nephropathy; Intra renal polyomavirus load/replication level; Simian Virus 40 large T antigen; mixed-effects repeated measures model; Interquartile range; immunohistochemistry

Copyright © 2018 by the American Society of Nephrology