and HLA-Incompatible Kidney Transplant Recipients - Semantic Scholar

Article

Incidence and Outcomes of BK Virus Allograft Nephropathy among ABO- and HLA-Incompatible Kidney Transplant Recipients Adnan Sharif,* Nada Alachkar,† Serena Bagnasco,‡ Duvuru Geetha,† Gaurav Gupta,† Karl Womer,† Lois Arend,‡ Lorraine Racusen,‡ Robert Montgomery,§ and Edward Kraus†

Summary Background and objectives ABO-incompatible kidney transplant recipients may have a higher incidence of BK virus allograft nephropathy (BKVAN) compared with ABO-compatible recipients. It is unclear whether HLAincompatible recipients share this risk or whether this phenomenon is unique to ABO-incompatible recipients. Design, setting, participation, & measurements This study analyzed adult incompatible kidney transplant recipients from 1998 to 2010 (62 ABO-incompatible and 221 HLA-incompatible) and identified patients in whom BKVAN was diagnosed by biopsy (per protocol or for cause). This was a retrospective analysis of a prospectively maintained database that compared BKVAN incidence and outcomes between ABO- and HLA-incompatible recipients, respectively. BKVAN link to rejection and graft accommodation phenotype were also explored. The Johns Hopkins Institutional Review Board approved this study. Results Risk for BKVAN was greater among ABO-incompatible than HLA-incompatible patients (17.7% versus 5.9%; P=0.008). Of BKVAN cases, 42% were subclinical, diagnosed by protocol biopsy. ABO-incompatibility and age were independent predictors for BKVAN on logistic regression. C4d deposition without histologic features of glomerulitis and capillaritis (graft accommodation-like phenotype) on 1-year biopsies of ABO-incompatible patients with and without BKVAN was 40% and 75.8%, respectively (P=0.04). Death-censored graft survival (91%) and serum creatinine level among surviving kidneys (1.8 mg/dl) were identical in ABO- and HLAincompatible patients with BKVAN (median, 1399 and 1017 days after transplantation, respectively). Conclusions ABO-incompatible kidney recipients are at greater risk for BKVAN than HLA-incompatible kidney recipients. ABO-incompatible recipients not showing the typical graft accommodation-like phenotype may be at heightened risk for BKVAN, but this observation requires replication among other groups. Clin J Am Soc Nephrol 7: 1320–1327, 2012. doi: 10.2215/CJN.00770112

Introduction BK polyoma virus has emerged as an important human pathogen in immunocompromised kidney transplant recipients. BK viremia is prevalent among 13%–22% of the kidney transplant population, and BK virus allograft nephropathy (BKVAN) is found in 1%–10% of kidney transplant biopsies (1,2). These observations are important because BKVAN is associated with a 50% risk for graft loss, although contemporaneous outcome data are lacking (2). Myriad donorand recipient-related risk factors contribute to the complex interplay of the relationship between the BK virus and the immune system, although overall immunosuppression burden is considered to be the most important risk factor for BKVAN (3,4). ABO-incompatible kidney transplantation has established itself as an important component of livingdonor transplant programs (5). Risks and benefits of performing transplantation across the blood-group barrier are well documented (6–9), but a link between 1320

Copyright © 2012 by the American Society of Nephrology

*Renal Institute of Birmingham, Queen Elizabeth Hospital, Edgbaston, Birmingham, United Kingdom, and † Department of Medicine, ‡ Department of Pathology, and § Department of Surgery, Johns Hopkins Medical Institutions, Baltimore, Maryland Correspondence: Dr. Edward S. Kraus, Division of Nephrology, Johns Hopkins Bayview Medical Center, Suite 2500, 301 Mason Lord Drive, Baltimore, MD 21224. Email: [email protected]

ABO-incompatible kidney transplantation and risk for BKVAN is contentious. Some groups have observed a higher risk for BK viremia or nephropathy (10–12), whereas others have not (8,13). The implication is that ABO-incompatible kidney recipients may be at greater risk for development of BK viremia or nephropathy because of increased intensity of induction protocols and subsequent immunosuppressant burden (both maintenance and treatment of cellularand antibody-mediated rejections). Whether this risk also applies to patients with incompatibility to HLAantigens, who undergo similar desensitization protocols, is unknown. Here we report the incidence and outcome of BKVAN among ABO- and HLA-incompatible kidney transplant recipients, diagnosed by protocol and indication biopsies, to investigate this relationship. In addition, we explored the relationship between BKVAN with episodes of rejection and development of the graft-accommodation phenotype to determine its effect www.cjasn.org Vol 7 August, 2012

Clin J Am Soc Nephrol 7: 1320–1327, August, 2012

on incidence and outcome of BKVAN development among incompatible kidney transplant recipients.

Materials and Methods This is a retrospective analysis of a prospectively maintained database for patients who underwent transplantation at the Johns Hopkins Hospital within the Incompatible Kidney Transplant Program. All patients received a desensitization treatment protocol approved by the Johns Hopkins Institutional Review Board. Initially the protocol was considered to be innovative therapy, and then the institutional review board granted approval to convert the prospective clinical database to a research database reviewed annually. Sixty-two ABO-incompatible and 221 HLA-incompatible (transplantation across preformed donorreactive HLA antibodies) consecutive kidney transplant recipients who underwent transplantation from 1998 to 2010 were analyzed (28 simultaneous ABO-/HLA-incompatible recipients were excluded). Treatment Protocol Incompatible kidney transplant patients received plasmapheresis using the COBE Spectra apheresis system (Gambro BCT, Lakewood, CO), as described elsewhere (14). One plasma volume was removed per treatment and replaced using 5% albumin solution or fresh frozen plasma. Intravenous immunoglobulin (Cytogam; MedImmune Inc., Gaithersburg, MD) was administered at a dose of 100 mg/kg after each plasmapheresis treatment. Escalating numbers of treatments were performed before and after transplantation on the basis of the level of blood group isohemagglutinin or donor-specific anti-HLA antibody at baseline and serial monitoring after surgery. Immunosuppression regimens have not altered since commencement of the incompatible program at Johns Hopkins in 1998. Mycophenolate mofetil (at a dosage of 2 g/d) and tacrolimus (target level, 8–12 ng/ml) starts with plasmapheresis. Induction therapy consisted of an IL-2 receptor antagonist or antithymocyte globulin. As part of preconditioning, some patients underwent preoperative splenectomy or received rituximab (administered at 375 mg/m2 on the eve of transplantation) depending on perceived immunologic risk (baseline antibody titer, repeat mismatch, repeat transplant). In recent years, this step has been removed from ABO-incompatible protocols but retained for high-risk HLA-incompatible recipients. Glucocorticoids were administered perioperatively and then tapered to 5 mg/d by 3 months. Goal tacrolimus trough levels were reduced to 8–10 ng/ml at 1 month, 7–8 ng/ml at 3 months, 5–8 ng/ml at 6 months, and 4–7 ng/ml at 1 year. Average tacrolimus plasma levels and mycophenolate dosage calculations were made by the mean level/dosage from time of transplant up to (but censored for) BKVAN diagnosis and up to the median day of BKVAN diagnosis for incompatible patients with or without BKVAN, respectively (means calculated over this time period as total sum divided by number of monthly levels and dosages, respectively). Diagnostic biopsies were performed for evaluation of graft dysfunction, defined by an increase in serum creatinine .20% above a patient’s baseline or for development

BKVAN and Incompatible Kidney Transplantation, Sharif et al.

1321

of proteinuria. In addition, protocol surveillance biopsies were performed before reperfusion and at 1, 3, 6, and 12 months. In most instances, cellular rejection with Banff grades (15) of 1a or 1b was treated with corticosteroids; for Banff scores of 2a, 2b, and 3, patients would receive antithymocyte globulin. Antibody-mediated rejection was treated with reinitiation of plasmapheresis/intravenous immunoglobulin until clinical improvement was achieved or rejection resolved, as demonstrated by follow-up biopsies. We did not distinguish between treatment of clinical rejection versus that of subclinical rejection (the latter diagnosed by protocol biopsy in the absence of evidence of allograft dysfunction) among our incompatible patients. All patients with pathologically proven rejection, clinical or subclinical, received the same therapy, as indicated. During the time span of this study, the protocol for management of BKVAN changed, reflecting increasing awareness of and published experience with this entity. In general, however, the strategy used has been based on cautious minimization of immunosuppression (i.e., reduction and, in some instances, elimination of mycophenolate mofetil dosage and reduction of tacrolimus exposure [target level, 3–5 ng/ml]). In some instances, leflunomide was substituted for mycophenolate or patients received a course of cidofovir. Patients continued to receive prednisone, 5 mg/d. Histologic Evaluation The diagnosis of BKVAN was based on demonstration of characteristic pathologic features, including cytopathic changes (intranuclear viral inclusions in tubular epithelial cells), cell enlargement with polymorphic nuclei, interstitial inflammation, and varying degrees of tubular atrophy or fibrosis, confirmed by immunohistochemistry (staining for SV40 large T antigen was done routinely). Classification of rejection was based on Banff criteria, as was development of the graft accommodation-like phenotype. The latter was defined as C4d deposition in the absence of glomerulitis (g score,1), transplant glomerulopathy (cg score,1) and peritubular capillaritis (ptc score,1) (15). We sought to compare incidence of rejection episodes and development of graft accommodation phenotype between ABO- and HLA-incompatible kidney transplant recipients to ascertain any link with BKVAN development. Because of the histologic mimicry between cellular rejection and BKVAN, we excluded, as indicated, Banff grades 1a and 1b from certain analyses to ensure clarity of data. Statistical Analyses Statistical analysis was performed using SPSS software, version 20 (SPSS Inc., Chicago, IL). Normality of data was assessed using the Kolmogorov–Smirnov tests. Descriptive statistics were used to estimate the frequencies, means, and medians of study variables. For continuous variables, a t test and a Mann–Whitney test were used for parametric and nonparametric data, respectively. Difference between groups was assessed with a two-sided Fisher exact test or Pearson chi-squared test for categorical variables as appropriate. Multivariable logistic regression analysis was used to determine independent risk factors for BKVAN development. The following potential risk factors for BKVAN were evaluated for inclusion according to univariate analysis: age (dichotomized

1322

Clinical Journal of the American Society of Nephrology

at median of 46 years), ethnicity, cause of ESRD, number of transplants, induction therapy, antibody-incompatibility type, and number of rejection episodes (pre-median day of BKVAN diagnosis). Variables with a significant univariate test (based on the Wald test and P value cutoff of 0.25) were entered into a multivariable logistic regression model. A P value , 0.05 was considered to represent a statistically significant difference.

Results Clinical, biochemical, and pathology data were reviewed for 62 ABO-incompatible and 221 HLA-incompatible consecutive kidney transplant recipients who underwent transplantation between 1998 and 2010. Demographic details for incompatible patients are given in Table 1. Incidence of BKVAN in Incompatible Patients ABO-incompatible patients were three times more likely to develop BKVAN diagnosed by protocol or indication biopsies after transplantation compared with HLA-incompatible patients (incidence of BKVAN, 17.7% versus 5.9%, respectively; P=0.008) (see Figure 1). This compares to a BKVAN rate of 3% among all kidney transplant recipients at the Johns Hopkins Hospital from 1997 to 2008 (16). There was no significant trend for earlier recognition of BKVAN among ABO-incompatible patients compared with HLAincompatible patients (median post-transplantation day for BKVAN diagnosis of 134 days versus 183 days). The diagnosis of BKVAN was subclinical and was made by protocol surveillance biopsies for 45% and 38% of cases among ABO-incompatible and HLA-incompatible patients, respectively. ABO-incompatible patients with BKVAN tended to be older (59 versus 48 years; P=0.03), to be male (91% versus 46%; P=0.03), and to be more likely to have

received first transplant (91% versus 38%; P=0.01) compared with HLA-incompatible patients with BKVAN. Comparisons within ABO-Incompatible Group Compared with ABO-incompatible patients without evidence of BKVAN (ABOi/BKVAN2), those with BKVAN (ABOi/BKVAN+) were older (59 versus 48 years; P=0.009) and more were of nonblack ethnicity (percentage of blacks in ABOi/BKVAN+ patients versus ABOi/BKVAN2 patients, 0% versus 30%, respectively, P=0.01). The ABOi/ BKVAN+ group had a trend toward being exclusively male, but this was not statistically different from the ABOi/ BKVAN2 group (91% versus 58%, respectively; P=0.08). No difference in tacrolimus plasma levels or mycophenolate mofetil dosages were observed at any monthly time point (or as the overall mean) between ABO-incompatible recipients who did or did not develop BKVAN. A full comparison of demographic variables is shown in Table 2. ABOi/BKVAN+ patients were twice as likely to have had a Banff category 2/3 cellular rejection episode as ABOi/ BKVAN2 patients (mean episodes per person, 0.36 versus 0.16, respectively), but this difference failed to achieve statistical significance. In the ABOi/BKVAN+ group, 29% of all rejection episodes occurred before the diagnosis of BKVAN, 12% occurred simultaneously, and 59% occurred after BKVAN diagnosis. These episodes include, however, Banff 1a and 1b rejections, which BKVAN can mimic. Thus, these findings require cautious interpretation. Among all ABO-incompatible patients, 61% of rejection episodes occurred before the median day of BKVAN diagnosis (134 days). Probing this further, we found that 45% of rejection episodes among ABOi/BKVAN+ patients, compared with 70% of rejection episodes in ABOi/BKVAN2 patients, occurred before this median day of BKVAN diagnosis. Therefore, the propensity for rejection in

Table 1. Demographic comparison of ABO- and HLA-incompatible patients

Demographic variable

ABO-Incompatible Group

HLA-Incompatible Group

P Value

Patients (n) Median year of transplantation (range) Median age (range) (yr) Men (%) Ethnicity (%) white black other First transplant (%) Cause of ESRD (%) diabetes glomerular disease tubulointerstitial disease polycystic kidney disease congenital/familial or metabolic disorder renovascular disease hypertensive disease other unknown Antithymocyte globulin induction (%) Mean tacrolimus plasma level 6 SD (ng/nl) Mean mycophenolate mofetil dose 6 SD (mg)

62 2006 (1999–2010) 51 (18–77) 65

221 2005 (1998–2010) 45 (18–74) 34

— 0.82 0.08 ,0.001 0.10

73 26 1 81 25 31 6 13 11 3 3 0 8 8.3 8.663.4 18926651

81 14 5 48 10 37 8 14 16 4 4 3 4 28.5 8.664.1 18516588

,0.001 0.15

,0.001 0.98 0.90



1323

Figure 1. | Incidence of BK virus allograft nephropathy in incompatible kidney transplant recipients (diagnosed by protocol and for-cause biopsies). Cumulative incidence was 17.7% for ABO-incompatible kidney recipients and 5.9% for HLA-incompatible kidney recipients; the difference was statistically significant (P=0.008).

ABO-incompatible patients with BKVAN was to occur after the diagnosis of BKVAN (in the context of immunosuppression tapering). We subsequently reviewed all available 1-year protocol biopsies for ABO-incompatible patients (n=48). We classified patients on the basis of the presence or absence of a predominant ABO-incompatible graft phenotype that has been reported in the literature, which includes C4d staining (C4d+) in the absence of other pathologic features of rejection (g, cg, and ptc scores , 1) (17,18). We observed that 40% of ABOi/BKVAN+ patients had this predominant phenotype at 1 year versus 75.8% of ABOi/BKVAN2 recipients (P=0.04) (see Figure 2). Comparisons within HLA-Incompatible Group Compared with HLA-incompatible patients without evidence of BKVAN (HLAi/BKVAN2), those with BKVAN (HLAi/BKVAN+) were more likely to be of nonblack ethnicity (percentage of blacks in the HLAi/BKVAN+ versus HLAi/BKVAN2 groups, 31% versus 18%, respectively; P=0.01). No differences in tacrolimus plasma levels or mycophenolate mofetil dosages were observed at any monthly time point (or as the overall mean) between HLA-incompatible recipients who did or did not develop BKVAN. A full comparison of demographic variables between HLA-incompatible patients is shown in Table 2. There was a nonsignificant trend for more Banff category 2/3 cellular rejections among HLAi/BKVAN+ patients than HLAi/BKVAN2 patients, with no difference in episodes of antibody-mediated or mixed rejection between the groups. In the HLAi/BKVAN+ group, 54% of rejection episodes occurred before the diagnosis of BKVAN, 8% occurred simultaneously, and 38% occurred after BKVAN diagnosis. However, as with the ABOi patients, these episodes include Banff 1a and 1b rejections, which BKVAN can mimic. Therefore, these findings also require cautious interpretation.

Among all HLA-incompatible patients, 80% of rejection episodes occurred before the median day of BKVAN diagnosis (183 days). Further evaluation indicates that 46% of rejection episodes among HLAi/BKVAN+ patients, compared with 69% among HLAi/BKVAN2 patients, occurred before this median day of BKVAN diagnosis. These results thus mirror those observed among ABOincompatible patients and support the notion that immunosuppression tapering, the treatment of choice for BKVAN, is associated with an increased risk for subsequent rejection. Multivariable Logistic Regression Analysis After the incompatible patients were pooled, putative determinants of BKVAN development were analyzed in a multivariable logistic regression model. ABO-incompatibility and older age were independently associated with development of BKVAN (see Table 3). Because of the preponderance of men in the ABO-incompatible group, male sex was removed from multivariable analysis to ensure that such inborn bias did not influence the logistic regression model output. Patient and Graft Outcomes in Antibody-Incompatible Recipients with BKVAN Finally, we analyzed long-term patient and graft outcome for all incompatible patients who developed BKVAN. All ABOi/BKVAN+ patients are alive at a median of 1399 days after transplantation (range, 199–2570 days). One graft was lost 808 days after transplantation because of recurrence of IgA nephropathy. Two patients in the HLAi/BKVAN+ group died: one of bacterial and fungal sepsis 340 days after transplantation and the other of gastric malignancy 891 days after transplantation. The most recent mean serum creatinine concentration in both groups for surviving kidneys was 1.8 mg/dl, which is higher than that for recipients without BKVAN.

1324


Table 2. Comparison of incompatible patients with versus those without histologic evidence of BK nephropathy

ABO-Incompatible (n=62) Demographic

Median year of transplantation (range) Mean age (yr) Men (%) Ethnicity (%) white black other First transplant (%) Cause of ESRD (%) diabetes glomerular disease tubulointerstitial disease polycystic kidney disease congenital/familial or metabolic disorder renovascular disease hypertensive disease other unknown Plasmapheresis sessions (n) Pretransplant Post-transplant Rituximab (%) Splenectomy (%) Antithymocyte globulin (%) Mean tacrolimus plasma level 6 SD (ng/ml) Mean mycophenolate mofetil dose 6 SD (mg) Rejection episodes (mean episode per patient) cellular (Banff 2/3) antibody mixed

BKVAN (n=11) 2007 (2002–2010) 59 91

No BKVAN (n=51) 2007 (1999–2010) 48 58

91 0 7 91

70 30 0 77

46 27 9 18 0

20 30 6 12 14

0 0 0 0

4 4 0 10

HLA-Incompatible (n=221) P Value 0.92 0.009 0.08 0.01

0.58 0.49

BKVAN (n=13) 2006 (2001–2010) 48 46

No BKVAN (n=208) 2005 (1998–2010) 45 33

69 16 15 38

82 13 5 48

16 38 8 0 23

10 37 8 14 15

15 0 0 0

4 4 3 5

P Value 0.69 0.49 0.25 0.01

0.42 0.44

4.6 3.2 18 18 8.1 8.664.4

4.7 4.2 9 21 8.4 8.663.8

0.91 0.19 0.16 0.91 0.82 0.93

4.8 5.8 8 2 29.2 7.563.0

3.4 5.3 111 30 28.2 8.663.3

0.21 0.65 0.43 0.55 0.74 0.53

19076554

18916603

0.77

19656734

18816655

0.66

0.36 0.00 0.09

0.16 0.22 0.16

0.001 0.13 0.62

0.08 0.31 0.77

Discussion Recipients of ABO-incompatible kidney transplants are at significantly greater risk of developing BKVAN compared with recipients of HLA-incompatible kidneys. This finding suggests that the increased rate of BKVAN among ABO-incompatible kidney transplant recipients is not secondary to desensitization protocols and heavy maintenance immunosuppression alone. Despite the higher incidence of BKVAN and higher rates of post-BKVAN rejection with immunosuppression reduction, the rate of graft loss was low. These findings require confirmation among other incompatible patient cohorts, but this work introduces a novel insight into risk for BKVAN after kidney transplantation. Case reports of BKVAN in ABO-incompatible kidney transplantation began to emerge from Japan over the past decade (17,18). Recently, Habicht et al. (12) retrospectively compared outcomes between 21 ABO-incompatible kidney recipients against a control group of 47 ABO-compatible patients. Although short-term graft outcomes were equivalent, ABO-incompatible transplantation was associated with more infectious complications (60% for

0.01 0.56 0.65

0.05 0.35 0.68

ABO-incompatible versus 30% for ABO-compatible; P,0.05). Specifically, there was a three-fold increase in incidence of BKVAN in ABO-incompatible versus ABOcompatible recipients (25% versus 8.5%, respectively), but this difference failed to reach statistical significance. In contrast, Genberg et al. (8) reported no BKVAN in their series of 15 ABO-incompatible recipients, although one case of BKVAN was identified from their five pediatric ABO-incompatible cases. Flint et al. (13) recently reported their prospective, observational experience with 37 ABOincompatible transplant recipients (52 contemporaneous ABO-compatible recipients acted as a comparator group). They found a BK viremia rate of 13% but no cases of BKVAN in the ABO-incompatible group (compared with 19% and one case of BKVAN in the ABO-compatible group). Oettl et al. (11) detected BK viremia in 7 of their 10 ABO-incompatible recipients in a prospective analysis, but no BKVAN was seen on serial protocol biopsy. Large series of ABO-incompatible patients have not reported on the incidence of BK viremia or nephropathy, including the Japanese, who have the most experience with ABO-incompatible



1325

Figure 2. | Banff scoring for ABO-incompatible patients differential by presence or absence of BK virus allograft nephropathy. C4d+ patients with no evidence of glomerulitis, transplant glomerulopathy, or peritubular capillaritis (g, cg, and ptc scores ,1) are classified as demonstrating the typical graft accommodation phenotype associated with ABO-incompatible transplantation.

transplantation (19,20). However, lack of reporting does not imply lack of association. Why ABO-incompatible kidney transplant recipients should be at greater risk for developing BKVAN is speculative. One possibility is an interaction between BK virus and the immune system, which occurs at many different levels (21) and may be altered in the milieu of blood group incompatibility. The occurrence of a graft accommodation phenotype, an acquired resistance of an organ to immune-mediated damage, is more frequently observed in ABO-incompatible kidney recipients than in HLAincompatible recipients (22–25). The results of this study suggest that ABO-incompatible kidney recipients who do not develop the accommodation phenotype are more likely to develop BKVAN than are HLA-incompatible and ABOcompatible patients. Other mechanisms are also possible. Although preconditioning and treatment of antibodymediated rejection could theoretically attenuate the immune response to BK replication and foster BKVAN, the use of intravenous immunoglobulin may protect against BK virus because of its anti-inflammatory or anti-idiotypic properties (26). A variety of intravenous immunoglobulin preparations have been shown to have neutralizing properties against BK virus (27) and are often used to treat BKVAN (2,3). Our protocol, however, uses low-dose intravenous immunoglobulin, which should be cleared from patients long before the development of BKVAN. Plasma levels of tacrolimus or mycophenolate mofetil dosages also

Table 3. Independent predictors for development of BK nephropathy by multivariable logistic regression analysis

Variable

Odds Ratio

P Value

ABO incompatibility Age (median $ 46 yr)

2.32 3.27

0.04 0.01

did not differ between incompatible kidney recipients who did or did not develop BKVAN. Mengel et al. (28) reported a tacrolimus plasma level .8 ng/ml as being associated with a greater risk for BKVAN. When we replicated that trough level in our analysis, it was not associated with a greater risk for BKVAN among ABO- or HLAincompatible kidney transplant recipients. Despite higher rates of BKVAN among incompatible kidney transplant recipients, we paradoxically observed no evidence of graft loss as a direct consequence of BKVAN. This is surprising because tapering of immunosuppression among incompatible recipients was associated with an increased risk for rejection. In a 5-year retrospective study, Hardinger et al. (29) reported the safety and efficacy of preemptive immunosuppression reduction (antimetabolite withdrawal with or without calcineurin inhibitor reduction) in the presence of BK viremia. Early BKVAN detection with our frequent protocol biopsies is also likely to be relevant. Buehrig et al. (30) demonstrated graft benefits when protocol biopsies depicted early BKVAN, with subsequent immunosuppression tapering, in compatible kidney transplant recipients. Our practice of reducing both antimetabolite and calcineurin inhibitor exposure appears to be successful, although at the cost of a higher incidence of late rejection. Categorization of BKVAN by the proposed Banff classification system (A, early changes without tubular epithelial cell necrosis; B, active nephropathy with virally induced tubular necrosis; C, late sclerosing changes) (31), modified from Drachenberg et al. (32), may be useful to risk-stratify such patients. However, investigations into the reproducibility of this scoring system are awaited, and the utility of this system in the context of subclinical BKVAN with protocol biopsies is uncertain. In addition, this categorization focuses on cell injury rather than inflammation, which may limit its diagnostic utility (33). The limitations of this study include the inability to ascertain BK viremia and viruria rates in patients who did

1326


not develop BKVAN. Until recent years, protocol screening for BK viremia was not part of standard post-transplantation practice. We are therefore unable to ascertain whether BK viremia is more prevalent among ABO-incompatible kidney transplant recipients than among HLA-incompatible recipients or whether, despite having a similar BK viremia rate, ABO-incompatible patients have a greater likelihood of progression to BKVAN. Sampling error may contribute to a significant false-negative rate in the diagnosis of BKVAN, reported to be as high as 36.5% in the literature (32); the incidence of BKVAN could be an underestimate, especially if no medullary tissue is obtained for histologic testing. The association between the graft accommodation phenotype and protection from BKVAN observed in the study among ABO-incompatible recipients does not automatically imply a causal relationship (which could also be reciprocal), and further work will be required to investigate these relationships and putative mechanisms. The ability to differentiate between BKVAN and rejection is a challenge for both transplant clinicians and histopathologists, and the Banff committee on this topic remains interested (31,34). In the context of incompatible transplantation, a cohort at increased risk for rejection, the ability to clearly differentiate between the two histopathologic diagnoses is crucial. In a retrospective analysis of compatible kidney transplant recipients, Singh et al. (35) reported excellent ability of cast-like, three-dimensional polyomavirus aggregates (termed Haufen crystals) to qualitatively predict the presence of BKVAN (97% and 100% positive and negative predictive values), a putative future option. Finally, we should acknowledge the small numbers of incompatible recipients with BKVAN in our study, which limits any clear determination of risk. The failure to observe this link in other studies may relate to small numbers, heterogeneous patients and protocols, lack of protocol biopsies, or simply a failure to include BKVAN as an outcome measure. In conclusion, we observed in this hypothesis-generating study that ABO-incompatible kidney transplant recipients have a higher risk of developing BKVAN compared with HLA-incompatible patients that is independent of other variables. Further investigation is required to shed light on which intrinsic attributes of ABO-incompatible recipients contribute to a greater risk for BKVAN. Answering this question may contribute to our understanding of not just ABO-incompatible patients but all kidney transplant recipients at risk for BKVAN. Acknowledgments A.S. gratefully acknowledges funding from the St. John Ambulance Airwing Transplantation Travelling Award. The study was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant RC1 DK086731 and the Charles T. Bauer Foundation (R.M.). Disclosures None. References 1. 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, 2002

2. Ramos E, Drachenberg CB, Wali R, Hirsch HH: The decade of polyomavirus BK-associated nephropathy: state of affairs. Transplantation 87: 621–630, 2009 3. Dall A, Hariharan S: BK virus nephritis after renal transplantation. Clin J Am Soc Nephrol 3[Suppl 2]: S68–S75, 2008 4. Prince O, Savic S, Dickenmann M, Steiger J, Bubendorf L, Mihatsch MJ: Risk factors for polyoma virus nephropathy. Nephrol Dial Transplant 24: 1024–1033, 2009 5. Garonzik Wang JM, Montgomery RA, Kucirka LM, Berger JC, Warren DS, Segev DL: Incompatible live-donor kidney transplantation in the United States: Results of a national survey. Clin J Am Soc Nephrol 6: 2041–2046, 2011 6. Montgomery RA, Locke JE, King KE, Segev DL, Warren DS, Kraus ES, Cooper M, Simpkins CE, Singer AL, Stewart ZA, Melancon JK, Ratner L, Zachary AA, Haas M: ABO incompatible renal transplantation: a paradigm ready for broad implementation. Transplantation 87: 1246–1255, 2009 7. Takahashi K, Saito K, Takahara S, Okuyama A, Tanabe K, Toma H, Uchida K, Hasegawa A, Yoshimura N, Kamiryo Y; Japanese ABOIncompatible Kidney Transplantation Committee: Excellent long-term outcome of ABO-incompatible living donor kidney transplantation in Japan. Am J Transplant 4: 1089–1096, 2004 8. Genberg H, Kumlien G, Wennberg L, Berg U, Tydeń G: ABOincompatible kidney transplantation using antigen-specific immunoadsorption and rituximab: A 3-year follow-up. Transplantation 85: 1745–1754, 2008 9. Sonnenday CJ, Warren DS, Cooper M, Samaniego M, Haas M, King KE, Shirey RS, Simpkins CE, Montgomery RA: Plasmapheresis, CMV hyperimmune globulin, and anti-CD20 allow ABO-incompatible renal transplantation without splenectomy. Am J Transplant 4: 1315–1322, 2004 10. Haidinger M, Schmaldienst S, Ko¨rmo¨czi G, Regele H, Soleiman A, Schwartz D, Derfler K, Steininger R, Mu¨hlbacher F, Bo¨hmig GA: Vienna experience of ABO-incompatible living-donor kidney transplantation. Wien Klin Wochenschr 121: 247–255, 2009 11. Oettl T, Halter J, Bachmann A, Guerke L, Infanti L, Oertli D, Mihatsch M, Gratwohl A, Steiger J, Dickenmann M: ABO blood group-incompatible living donor kidney transplantation: A prospective, single-centre analysis including serial protocol biopsies. Nephrol Dial Transplant 24: 298–303, 2009 12. Habicht A, Bro¨ker V, Blume C, Lorenzen J, Schiffer M, Richter N, Klempnauer J, Haller H, Lehner F, Schwarz A: Increase of infectious complications in ABO-incompatible kidney transplant recipients—a single centre experience. Nephrol Dial Transplant 26: 4124–4131, 2011 13. Flint SM, Walker RG, Hogan C, Haeusler MN, Robertson A, Francis DM, Millar R, Finlay M, Landgren A, Cohney SJ: Successful ABO-incompatible kidney transplantation with antibody removal and standard immunosuppression. Am J Transplant 11: 1016–1024, 2011 14. Montgomery RA, Zachary AA, Racusen LC, Leffell MS, King KE, Burdick J, Maley WR, Ratner LE: Plasmapheresis and intravenous immune globulin provides effective rescue therapy for refractory humoral rejection and allows kidneys to be successfully transplanted into cross-match-positive recipients. Transplantation 70: 887–895, 2000 15. 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, 2008 16. Geetha D, Parkhie S, Nadkarni GN, He C, Shafi T: Polyomavirusassociated nephropathy: A comparison of 2 different strategies for immunosuppression reduction. Medicine (Baltimore) 90: 296–302, 2011 17. Matsumoto A, Namba Y, Yazawa K, Ichimaru N, Miyagawa Y, Takahara S, Okuyama A, Kim M, Kojima Y, Kokado Y, Kyo M: [A case report of BK virus nephropathy after an ABO-incompatible living kidney transplantation]. Nippon Hinyokika Gakkai Zasshi 94: 685–688, 2003 18. Itabashi Y, Sakai K, Kawamura T, Hyodo Y, Muramatsu M, Arai K, Aikawa A, Mizuiri S, Ohara T, Hasegawa C, Ishikawa Y,


19.

20. 21. 22. 23.

24.

25.

26. 27. 28.

Hasegawa A: BK virus nephropathy in a patient with ABOincompatible renal transplantation. Clin Transplant 18[Suppl 11]: 39–43, 2004 Fuchinoue S, Ishii Y, Sawada T, Murakami T, Iwadoh K, Sannomiya A, Koyama I, Kubota K, Tojimbara T, Nakajima I, Teraoka S: The 5-year outcome of ABO-incompatible kidney transplantation with rituximab induction. Transplantation 91: 853–857, 2011 Crew RJ, Ratner LE: ABO-incompatible kidney transplantation: Current practice and the decade ahead. Curr Opin Organ Transplant 15: 526–530, 2010 Babel N, Volk HD, Reinke P: BK polyomavirus infection and nephropathy: The virus-immune system interplay. Nat Rev Nephrol 7: 399–406, 2011 Lynch RJ, Platt JL: Accommodation in renal transplantation: Unanswered questions. Curr Opin Organ Transplant 15: 481– 485, 2010 Haas M, Segev DL, Racusen LC, Bagnasco SM, Locke JE, Warren DS, Simpkins CE, Lepley D, King KE, Kraus ES, Montgomery RA: C4d deposition without rejection correlates with reduced early scarring in ABO-incompatible renal allografts. J Am Soc Nephrol 20: 197–204, 2009 Zachary AA, Montgomery RA, Ratner LE, Samaniego-Picota M, Haas M, Kopchaliiska D, Leffell MS: Specific and durable elimination of antibody to donor HLA antigens in renal-transplant patients. Transplantation 76: 1519–1525, 2003 Haas M, Montgomery RA, Segev DL, Rahman MH, Racusen LC, Bagnasco SM, Simpkins CE, Warren DS, Lepley D, Zachary AA, Kraus ES: Subclinical acute antibody-mediated rejection in positive crossmatch renal allografts. Am J Transplant 7: 576–585, 2007 Randhawa PS, Schonder K, Shapiro R, Farasati N, Huang Y: Polyomavirus BK neutralizing activity in human immunoglobulin preparations. Transplantation 89: 1462–1465, 2010 Jordan SC, Toyoda M, Vo AA: Regulation of immunity and inflammation by intravenous immunoglobulin: relevance to solid organ transplantation. Expert Rev Clin Immunol 7: 341–348, 2011 Mengel M, Marwedel M, Radermacher J, Eden G, Schwarz A, Haller H, Kreipe H: Incidence of polyomavirus-nephropathy in renal allografts: Influence of modern immunosuppressive drugs. Nephrol Dial Transplant 18: 1190–1196, 2003


1327

29. 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, 2010 30. 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, 2003 31. Sis B, Mengel M, Haas M, Colvin RB, Halloran PF, Racusen LC, Solez K, Baldwin III WM, Bracamonte ER, Broecker V, Cosio F, Demetris A, Drachenberg C, Einecke G, Gloor J, Glotz D, Kraus E, Legendre C, Liapis H, Mannon RB, Nankivell BJ, Nickeleit V, Papadimitriou JC, Randhawa P, Regele H, Renaudin K, Rodriguez ER, Seron D, Seshan S, Suthanthiran M, Wasowska BA, Zachary A, Zeevi A: Banff ’09 meeting report: Antibody mediated graft deterioration and implementation of Banff Working Groups. Am J Transplant 10: 464–471, 2009 32. 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, 2004 33. Matsutani 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, 2012 34. 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, 2011 35. 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, 2009 Received: January 22, 2012 Accepted: April 30, 2012 Published online ahead of print. Publication date available at www. cjasn.org.