Kidney transplantation in adults: BK polyomavirus-associated nephropathy

INTRODUCTION — BK polyomavirus (BKPyV) is a small DNA virus that establishes lifelong infection in the renal tubular and uroepithelial cells of most of the world's population. For the majority, infection is quiescent and benign. However, in immunocompromised patients, BKPyV can reactivate, and in some, lead to BKPyV-associated nephropathy (BKPyVAN).

Among kidney transplant recipients, BKPyV reactivation is common. Reactivation is frequently subclinical, although it may manifest with acute kidney injury (AKI), and is a risk factor for premature allograft loss. Thus, screening for reactivation is recommended for all kidney transplant recipients after transplantation. For those with clinically significant reactivation, reduction of immunosuppression is the cornerstone of management. There is no specific antiviral or immunomodulatory therapy sufficiently effective for routine use.

This topic will review the epidemiology, microbiology, pathogenesis, clinical manifestations, screening, diagnosis, and management of BKPyV infection in kidney transplant recipients. The general evaluation of kidney allograft dysfunction and other infections in solid organ transplant recipients are discussed separately:

●(See "Kidney transplantation in adults: Evaluation and diagnosis of acute kidney allograft dysfunction".)

●(See "Infection in the solid organ transplant recipient".)

EPIDEMIOLOGY

Prevalence — BK polyomavirus (BKPyV) is a ubiquitous virus with a worldwide seroprevalence of approximately 80 to 90 percent [1-3]. Primary infection is typically acquired during childhood, possibly via fecal-oral or respiratory transmission [1,4,5]. Following primary infection, the virus establishes lifelong infection in renal tubular and uroepithelial cells. For most individuals, both primary infection and persistent infection are clinically silent and not associated with any known adverse effects.

Among kidney transplant recipients, reactivation of latent infection or transmission of new infection via the donor kidney can lead to viruria (detection of intact virus or virus components in urine), viremia (typically by detection of viral DNA in blood), or allograft nephropathy (demonstration of virus or virus components in allograft tissue) [6,7]. Viral replication most commonly occurs during the first year after transplantation when cellular immunity is most suppressed. Viruria and viremia are detected in approximately 25 to 30 and 12 percent of kidney transplant recipients, respectively [8,9], and almost invariably precede the development of nephropathy in a stereotypical pattern of viruria, followed by viremia, followed by BKPyV-associated nephropathy (BKPyVAN).

Approximately 1 to 10 percent of kidney transplant recipients will develop BKPyVAN [10-15]. Historically, BKPyVAN was associated with graft loss rates that exceeded 50 percent [16]. However, with the implementation of standardized screening protocols, rates of short-term graft loss have fallen substantially [8].

Risk factors for viral replication — The intensity of immunosuppression (particularly suppression of cellular immunity) appears to be a dominant risk factor for BKPyV replication and disease. Consistent with this, replication rates are higher in the early posttransplant period and following treatment for allograft rejection when immunosuppression intensity is highest.

No specific immunosuppressive drug or regimen has been definitively associated with clinically significant BKPyV infection [9,13,17-21]. Several studies have suggested that certain drugs (particularly tacrolimus) may be associated with an increased relative risk [13,22,23], while others (eg, mammalian [mechanistic] target of rapamycin [mTOR] inhibitors) may be associated with a lower relative risk [24-27]. However, BKPyV replication and BKPyVAN have occurred in patients receiving nearly all immunosuppressive drugs and their combinations [9,12,14,17,28-34].

Based on the observation that BKPyVAN occurs predominantly in the kidney allograft rather than in native kidneys, factors other than profound immunosuppression alone are likely important for disease development. Other important risk factors include high risk serostatus (ie, kidney transplant from a BKPyV-seropositive donor to a seronegative recipient) [35,36], impaired immune response to BKPyV [36,37], and donor BKPyV viruria prior to transplant [38-40]. The last two factors suggest that the donor is an important source of transmission and that viral serotype may play a role in pathogenesis [38,41].

Other risk factors associated with an increased risk of BKPyVAN or disease severity include older age [23,42], ureteral stent placement [7], ABO incompatibility [43], rejection or ischemia of the transplanted kidney [44], delayed graft function [7], HLA mismatch [43,45-48], specific HLA-C alleles [41], BKPyV polymorphisms [49-51], and transplantation from an HCV-positive donor [52].

By contrast, some factors have been associated with decreased risk of BKPyVAN. In one cohort of 407 living kidney donor-recipient pairs, recipient HLA-B51 positivity was associated with an approximate fivefold reduction in BKPyVAN (hazard ratio [HR] 0.18, 95% CI 0.04-0.73) [53]. Polycystic kidney disease has also been associated with a lower risk of BKPyVAN [54].

In one cohort study of over 20,000 mate kidney pairs, factors associated with BKPyVAN included use of an antibody-depleting agent for induction; age <18 or ≥60 years; male sex; ≥4 HLA-A, -B, or -DR mismatches; and acute rejection [55]. Treatment for BKPyVAN was associated with an increased risk of allograft failure (HR 2.01, 95% CI 1.63-2.48). The leading cause of graft loss was BKPyVAN, occurring in 45 percent, followed by acute and chronic rejection in 25 and 12 percent, respectively.

VIROLOGY — BK polyomavirus (BKPyV) is a small (30 to 45 nm), icosahedral, nonenveloped, double-stranded, closed circular DNA virus [5]. The 5 kb genome encodes six viral proteins. There are two "early" nonstructural or enzymatic proteins, an agnoprotein, and three "late" proteins. The early proteins are the large tumor antigen, or "T antigen," and the small tumor antigen, or the "t antigen." The T antigen is responsible for cell immortalization and the establishment of latent infection. The agnogene, which transcribes the agnoprotein, appears to aid in the assembly of viral particles. The "late" genes encode three viral capsid proteins, VP-1, VP-2, and VP-3, which mediate cell entry and progeny virion assembly.

Noncoding control regions of the genome may be important in pathogenesis. Mutations in these regions can be progressively acquired during the course of an infection, leading to altered cell tropism, permissivity (ie, cell types that can support viral replication), and replication rates [49-51]. In kidney transplant recipients with clinically significant BKPyV replication, mutations in noncoding control regions have been associated with high BKPyV viral loads [50].

Virulence may also vary among serotypes, which differ genetically and in their cellular tropism [5,56,57]. Serotype I is the most prevalent and is responsible for most human disease. Neutralizing antibodies to one serotype do not appear to confer protection against others.

PATHOGENESIS — Insufficient cellular immune control is presumed to be an important part of BK polyomavirus (BKPyV)-associated nephropathy (BKPyVAN) pathogenesis [18,58]. After primary infection, which typically occurs in childhood, BKPyV maintains persistent infection in the renal and uroepithelium (transitional epithelium, renal tubular epithelium, and parietal epithelium of Bowman's capsule) of most individuals [59-61]. Control of this persistent infection is dependent on CD4+ and CD8+ T cell immunity [58]. When immune control is disrupted (as with immunosuppressive drugs), BKPyV can begin to actively replicate.

As viral replication persists, injury to the renal tubular epithelium results from direct viral replication and cell turnover. Subsequent inflammation and fibrosis lead to further injury and ultimately tubular atrophy, necrosis, and nephron loss [62]. The resultant inflammatory milieu may further promote viral replication and perpetuate injury [48]. Transcriptional analysis of kidney allograft tissue in BKPyVAN has demonstrated upregulation of multiple inflammatory pathways that overlap considerably with those seen in acute allograft rejection [63,64].

The development of alloimmunity and acute rejection (following reduction of immunosuppression in response to BKPyV replication) may further contribute to allograft dysfunction and potentially graft loss [65]. Preliminary data suggest that the kidney allograft inflammatory profile in BKPyVAN is similar to that seen in acute cellular rejection and involves both virus- and allograft-directed immune injury [63,66].

CLINICAL MANIFESTATIONS — In kidney transplant recipients, BK polyomavirus (BKPyV) replication typically develops in stages: viruria followed by viremia and then, if viral replication persists, nephropathy can ensue (figure 1).

Viruria and viremia — Viruria is the earliest manifestation of BKPyV infection in kidney transplant recipients, affecting approximately one-quarter to one-third of patients during the first year following transplantation (figure 1) [8,9]. For most, viruria is asymptomatic, detected only by screening, and does not progress to viremia (see 'Screening and diagnosis' below). While viruria is a sensitive marker for progression to BKPyV-associated nephropathy (BKPyVAN), it is nonspecific [67]. Shedding of BKPyV in the urine (as detected by highly sensitive polymerase chain reaction [PCR] methods) is common among otherwise healthy older adult patients, pregnant women, and others with suppressed cellular immunity and is generally without clinical consequence [42,68]. Urine decoy cells (renal tubular or uroepithelial cells containing intranuclear viral inclusions), which typically represent higher-level viruria, may be present at this stage (picture 1 and image 1), but they do not have a significantly greater prognostic value for progression to viremia than detection of viruria by more sensitive methods such as PCR. (See 'Urine quantitative PCR' below and 'Urine cytology' below.)

Viremia may follow viruria in a few weeks and occurs most frequently in those with high urine viral loads and sustained viruria [67,69,70]. Viremia is detected in 10 to 30 percent of recipients in the first six months posttransplantation and in 5 to 10 percent of recipients thereafter. As with viruria, viremia is typically asymptomatic. However, viremia has a greater predictive value than viruria for progression to BKPyVAN [8,70]. Viremia is present in nearly all patients with BKPyVAN and has a positive predictive value of approximately 40 to 65 percent for the development of BKPyVAN [67,71]. Because BKPyVAN can quickly follow viremia (eg, within one to two weeks) and damage to the graft can be irreversible, viremia is a generally accepted indication to reduce immunosuppression in kidney transplant recipients. As with viruria, higher viral loads and sustained viremia have greater predictive value for concomitant or progression to biopsy-confirmed BKPyVAN. (See 'Plasma quantitative PCR' below and 'Treatment' below.)

BKPyV-associated nephropathy (BKPyVAN) — Asymptomatic viruria, viremia, and/or a slow progressive rise in serum creatinine are typically the only indicators of BKPyVAN. The incidence of BKPyVAN is highest in the first two to six months posttransplant. While the majority of cases occur in the first posttransplant year, BKPyVAN can occur years after transplantation [72]. The incidence of late BKPyVAN appears to be highest in patients with multi-organ transplants and is possibly related to the more intensive immunosuppressive regimens used for these patients.

Without resolution of infection, progressive kidney allograft dysfunction and graft loss can ensue over a period of months [62]. Within the allograft, early infection triggers interstitial inflammation, which then progresses to fibrosis and tubular injury. Accordingly, urinalysis may reveal pyuria, hematuria, and/or cellular casts consisting of renal tubular cells and inflammatory cells, or may be normal. (See 'Histologic findings' below.)

Reduction of immunosuppression is the primary means of restoring immunity and immune control of BKPyV replication. With this comes the risk of rejection, which can be difficult to distinguish clinically from progressive BKPyVAN [73]. (See 'Distinguishing BKPyVAN from rejection' below.)

Other manifestations — Hemorrhagic cystitis is a rare manifestation of BKPyV infection in kidney transplant recipients but is the most commonly reported manifestation of BKPyV infection among hematopoietic cell transplant recipients [74,75]. (See "Overview of infections following hematopoietic cell transplantation", section on 'Hemorrhagic cystitis'.)

There is a putative link between BKPyV and the development of genitourinary cancers, largely based upon viral-associated oncogenesis in animal models and the ability of the virus to transform cells in vitro [76-78]. However, a causal role for BKPyV and human malignancies has not been definitively established.

SCREENING AND DIAGNOSIS

Posttransplant screening — We recommend routine screening for BK polyomavirus (BKPyV)-associated nephropathy (BKPyVAN) for all kidney transplant recipients in the early posttransplant period. Observational data suggest that screening and preemptive reduction in immunosuppression for patients with clinically significant BKPyV viremia prevent progression to BKPyVAN in the majority of patients [22,79,80]. The optimal screening strategy has not been determined, and approaches vary among transplant centers [81-83].

We screen patients with a quantitative plasma BKPyV polymerase chain reaction (PCR; ie, viral load) at the following time points:

●Monthly for the first six months following transplant, then every three months until two years posttransplant, and then annually until five years posttransplant

●Whenever kidney allograft dysfunction occurs or when an allograft biopsy is performed for allograft dysfunction

Clinical response varies and is dependent upon the degree of viremia, the patient's immunosuppressive regimen, and presence of allograft dysfunction. The threshold plasma BKPyV viral load that is considered positive or clinically significant varies according to the particular assay used. In general, levels >1000 copies/mL are considered positive in most assays, and levels >10,000 copies/mL correlate with biopsy-confirmed BKPyVAN. However, there is wide variability in the quantitation of BKPyV DNA across assays. (See 'Plasma quantitative PCR' below.)

●For patients with viremia (eg, viral loads >1000 copies/mL) and normal allograft function, we typically reduce immunosuppression and monitor the viral load every two to four weeks thereafter to ensure that it is downtrending. (See 'Reduction of immunosuppression' below.)

●For patients with viremia and new-onset allograft dysfunction, we generally reduce immunosuppression and monitor viral loads every two to four weeks thereafter when the clinical picture suggests that BKPyVAN is the most likely cause. We consider a kidney allograft biopsy if the cause for kidney allograft dysfunction is uncertain or if kidney dysfunction and/or viremia fail to resolve despite reducing immunosuppression. (See "Kidney transplantation in adults: Evaluation and diagnosis of acute kidney allograft dysfunction", section on 'Patients presenting with an elevated serum creatinine'.)

Our approach is generally consistent with the Kidney Disease: Improving Global Outcomes (KDIGO) and American Society of Transplantation Infectious Diseases Community of Practice (AST-IDCOP) guidelines [84,85]. The AST-IDCOP takes a slightly more conservative approach to screening and recommends an initial monthly screening period of nine months rather than six months. Because approximately 85 percent of kidney transplant recipients who are going to become viremic develop BKPyV viremia within the first four months of transplantation, the utility of extending the screening period beyond six months is uncertain [9,86]. Other variations in practice include screening methods used and thresholds for reducing immunosuppression and obtaining a kidney allograft biopsy. As examples, some centers screen using a urine BKPyV PCR because of its high sensitivity and proceed to plasma PCR for those with viruria [87]; others may screen with urine cytology (for decoy cells) if PCR is not available. (See 'Urine quantitative PCR' below and 'Urine cytology' below.)

Several studies have evaluated the efficacy of a screening and preemptive strategy for BKPyVAN [9,10,79,88,89]. As examples:

●This approach was systematically evaluated in a trial of 200 kidney transplant recipients who were randomly assigned to receive either tacrolimus or cyclosporine in combination with an antimetabolite (mycophenolate or azathioprine) and low-dose prednisone for maintenance immunosuppression [9]. All patients were monitored with PCR of blood and urine to detect early BKPyV viruria and viremia. At one year posttransplant, viruria (defined as >2000 copies/mL of urine) and viremia (defined as >2000 copies/mL of plasma) were noted in 35 and 12 percent, respectively. Upon detection of viremia, but not viruria, the antimetabolite (either mycophenolate or azathioprine) was discontinued; if viremia failed to clear within four weeks, the calcineurin inhibitor dose was reduced. This protocol resulted in resolution of viremia in 95 percent of patients without an increased risk of acute rejection, kidney dysfunction, or allograft loss. No patients developed biopsy-proven BKPyVAN. At five years, patient and graft survival were 91 and 84 percent, respectively, and there were no differences between those with BKPyV viruria detected in the first year posttransplant compared with those without BKPyV viruria [17].

●Similar findings were reported in a prospective study of 62 consecutive pediatric kidney transplant recipients who underwent screening for BKPyV with blood and urine BKPyV PCR at months 1, 3, 6, 9, 12, 18, 24, 36, and 48 after transplantation [88]. At a median of 24 months, 39 patients (63 percent) developed BKPyV viruria, and 13 (21 percent) developed BKPyV viremia. Patients with viremia were preemptively treated with reduction in immunosuppression (stepwise reduction in calcineurin inhibitor dose followed by stepwise reduction in mycophenolate dose), which cleared viremia and prevented the development of BKPyVAN in all patients, without increasing the risk for rejection or graft dysfunction.

Testing methods

Plasma quantitative PCR — Quantification of plasma BKPyV DNA by real-time polymerase chain reaction (PCR) is the preferred screening test for BKPyVAN at most transplant centers (table 1) [84,85]. The detection of BKPyV viremia by plasma quantitative PCR is both highly sensitive (100 percent) and specific (88 percent) for the diagnosis of BKPyVAN and has a higher positive predictive value for BKPyVAN than the detection of viruria by urine quantitative PCR or urine cytology (50 to 60 percent versus 40 and 29 percent, respectively) [10,29]. The suggested frequency of screening after transplantation is described elsewhere in this topic. (See 'Posttransplant screening' above.)

In addition, following the initial detection of viremia, plasma PCR can be used to monitor the patient's response to therapy since a decrease in BKPyV viremia usually occurs soon after a reduction in immunosuppression and precedes a decrease in viruria by weeks to months [9]. Importantly, persistent high-grade viremia even in the absence of histologic evidence of BKPyVAN is associated with worse allograft function and risk of allograft loss [62].

There are no clearly established threshold levels for BKPyV viremia that predict BKPyVAN. However, most experts agree that a BKPyV viral load of ≥10,000 copies/mL, particularly when sustained for more than three weeks' duration, is highly suggestive of BKPyVAN ("presumptive" BKPyVAN) [12,84].

BKPyV DNA levels may vary substantially among assays [90], with variations of a magnitude of 1 to 2 log-fold in sequential weekly analyses in a commercially available quantitative PCR assay. This interassay variability complicates the interpretation of results and precludes the establishment of universally accepted thresholds for the management of BKPyV infection. To address this issue, the World Health Organization (WHO) introduced an international standard in 2016 to standardize viral load values among different laboratory assays when results are expressed as international units/mL [91]. Initial studies evaluating the utility of the WHO international standard have showed improved agreement in reporting across laboratories [92,93]. However, variability among laboratories still exists due to differences in DNA extraction techniques, amplicon size, and PCR primer design [94].

Urine quantitative PCR — We do not use quantitative PCR of the urine for BKPyV DNA to screen for BKPyVAN. Some transplant centers prefer to screen with urine polymerase chain reaction (PCR), given the high sensitivity and less invasive nature of this test, and proceed to plasma PCR for those with viruria (table 1) [87]. However, patients who are found to have viruria require confirmation with quantitative plasma PCR, since approximately one-half of patients with BKPyV viruria will not develop viremia or BKPyVAN. As such, the cost effectiveness of this approach has been questioned [87]. Furthermore, urine PCR for BKPyV DNA is not as useful as plasma PCR for monitoring the response to therapy, since changes in the urine viral load lag behind changes in the plasma viral load upon lowering immunosuppression. However, some studies suggest that the detection of high-grade viruria might be helpful to predict clinically significant viremia [70,95].

Urine cytology — Cytologic examination of the urine, which may reveal BKPyV-infected cells, is infrequently used to screen for BKPyVAN. Although the presence of characteristic cytopathologic changes in infected cells (which have been called decoy cells (picture 1) due to their resemblance to renal carcinoma cells) is strongly suggestive of BKPyV infection [10,96], urine cytology is less sensitive and specific for the diagnosis of BKPyVAN compared with plasma quantitative PCR (table 1) [10,29]. In addition, the diagnostic utility of urine cytology is further limited by interobserver variability and the qualitative, rather than quantitative, nature of this test.

Kidney allograft biopsy — Kidney allograft biopsy is the gold standard for diagnosing BKPyVAN, assessing its severity, and evaluating for concomitant processes. However, because biopsy is invasive and sampling error can occur, a presumptive diagnosis is often made based upon the presence of significant viremia (plasma BKPyV viral load ≥10,000 copies/mL). (See 'Plasma quantitative PCR' above.)

A definitive diagnosis of BKPyVAN requires the following findings on kidney biopsy [12,97,98]:

●Characteristic cytopathic changes. (See 'Histologic findings' below.)

plus

●Positive immunohistochemistry tests using antibodies directed specifically against BKPyV or against the cross-reacting SV40 large T antigen. Positive SV40 staining is useful as it is associated with a specificity of almost 100 percent for polyomavirus nephropathy (PVN), although it does not distinguish between BKPyV- and JC virus (JCV)-associated cases.

Because of the focal nature of early BKPyVAN, the diagnosis may be missed on one-third of biopsies. As a result, at least two biopsy cores, preferably including medulla, should be examined [97,99,100]. Medullary tissue should be included because BKPyV is more likely to be present in the medulla. If the initial biopsy does not confirm BKPyVAN, a repeat biopsy should be considered.

Histologic findings — BKPyVAN is associated with characteristic histologic findings on kidney biopsy. Since viral cytopathic changes can also can be observed with cytomegalovirus (CMV), adenovirus, and herpes simplex virus (HSV) infections, the morphologic changes may not be pathognomonic. They include the following:

●Intranuclear basophilic viral inclusions without a surrounding halo [11,59,101,102]. CMV has cytoplasmic inclusions, as noted above, and HSV has both intranuclear and cytoplasmic inclusions.

●Anisonucleosis, hyperchromasia, and chromatin clumping of infected cells [11,59,102].

●Interstitial mononuclear or polymorphonuclear cell infiltrates in the areas of tubular damage [11,59,102].

●Tubular injury, which is characterized by tubular cell apoptosis, cell drop out, desquamation, and flattened epithelial lining [11,59,102].

●Tubulitis, which is manifested by lymphocyte permeation of the tubular basement membrane. When extensive, it is difficult to differentiate BKPyVAN from allograft rejection [11,59,101,102]. (See 'Distinguishing BKPyVAN from rejection' below.)

●With electron microscopy, intranuclear viral inclusions (with a diameter size of 30 to 50 nm) and tubular damage characterized by tubular cell necrosis, prominent lysosomal inclusions, and luminal protein and cellular casts [11,59,101,102].

In 2017, the Banff Working Group on Polyomavirus Nephropathy established a classification system that defined three histologic classes of definitive (biopsy-proven) BKPyVAN on the basis of two morphologic variables: intrarenal polyomavirus replication/load levels (pvl) and Banff interstitial fibrosis (ci) scores [103]. The pvl score is based upon the extent of virally induced tubular changes. A tubule with intranuclear viral inclusion bodies (type 1 or 2) and/or a positive immunohistochemical reaction for SV40 large 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 (ie, all available cores and all tubules/ducts in the cortex and medulla) and three levels of pvl are defined: pvl 1 = <1 percent, pvl 2 = 1 to 10 percent, pvl 3 = >10 percent positive tubules/ducts. The three histologic classes of PVN are as follows:

●PVN class 1 – pvl 1, ci ≤1

●PVN class 2 – pvl 1, ci ≥2 or pvl 2, any ci score or pvl 3, ci ≤1

●PVN class 3 – pvl 3, ci ≥2

In BKPyVAN classes 1 to 3, interstitial inflammation and tubulitis can vary from Banff scores ti 0 to ti 3/t0 to t3. BKPyVAN class 1 often lacks a significant inflammatory reaction. BKPyVAN and rejection (acute, chronic, cell mediated, and/or antibody mediated) can occur concomitantly. BKPyVAN 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 establish or exclude a diagnosis of definitive BKPyVAN, two biopsy cores including portions of medulla in at least one of the two cores are required.

In a retrospective analysis of 178 patients with biopsy-proven BKPyVAN, the Banff classification system was shown to correlate with clinical outcomes [103]. Twenty-five, 63, and 12 percent of patients were categorized as PVN class 1, 2, and 3, respectively. At 24 months after the initial diagnostic kidney allograft biopsy, the median increase in serum creatinine was 0.4 mg/dL, 1.0 mg/dL, and 4.8 mg/dL for patients with PVN class 1, 2, and 3, respectively. Graft failure occurred in 16, 31, and 50 percent of patients with PVN class 1, 2, and 3, respectively. Further validation of the Banff classification system has been reported [104]; however, this classification system has not been shown to directly correlate with clinical outcomes across all studies [105].

Two studies examined the longitudinal histologic evolution of BKPyVAN in sequential allograft biopsies and made several important observations [62,106]. SV40 T antigen-negative interstitial inflammation preceded a histologic diagnosis of BKPyVAN in up to 23 percent of cases with BKPyV viremia and was also present in cases of resolving BKPyVAN, suggesting that the SV40 T antigen stain may be falsely negative due to sampling error and/or limited staining sensitivity. Persistent high-level viremia was the most significant risk factor for destructive chronic infection and eventual allograft loss, but no single clinical or pathologic feature could predict the clinical course of BKPyVAN in an individual patient. A high proportion of cases eventually evolved to interstitial fibrosis and tubular atrophy, and graft failure was associated both with BKPyV and rejection resulting from attempts to treat BKPyVAN, highlighting the importance of prevention over treatment strategies.

Distinguishing BKPyVAN from rejection — Allograft rejection may closely resemble BKPyVAN on kidney biopsy [63,101]. Distinguishing BKPyVAN from allograft rejection is important since treatment for presumed rejection with increased immunosuppression (without concomitant reduction in maintenance immunosuppression) may result in allograft loss if BKPyVAN is present.

BKPyVAN is generally distinguished from rejection by the presence of BKPyV inclusions and immunohistologic or in situ hybridization evidence of virally infected cells, which are usually tubular epithelial cells, rather than podocytes or endothelial cells [11]. It is important to correlate the histologic findings with PCR evidence of viremia. (See 'Plasma quantitative PCR' above.)

A separate issue is identifying coexisting lesions of BKPyVAN and acute rejection [11,107,108]. Whether BKPyVAN and T cell-mediated, antibody-mediated, or mixed rejection can exist concomitantly is controversial. Establishing a diagnosis of concomitant T cell-mediated rejection in a biopsy that has BKPyVAN is difficult since both the histologic features and transcriptional profiles of these two disorders are similar [63,101]. In general, the presence of extensive tubulitis in areas remote from the viral cytopathic changes suggests that acute rejection is present, in addition to BKPyVAN. The combined presence of endarteritis, fibrinoid vascular necrosis, glomerulitis, and C4d deposits along peritubular capillaries is conclusive evidence of concurrent rejection, although some patients with BKPyVAN without concurrent rejection may have C4d deposits in the tubular basement membrane [12,109]. (See 'Concurrent BKPyVAN and acute rejection' below.)

Among some patients, it may only be possible to distinguish the effects of BKPyV viral infection from those of rejection by empirically altering the immunosuppressive regimen and observing the clinical response. (See 'Concurrent BKPyVAN and acute rejection' below.)

Other diagnostic methods — Negative-staining electron microscopy of the urine of patients with BKPyVAN often reveals the presence of cast-like, three-dimensional polyomavirus aggregates, termed Haufen (picture 2) [110-112]. Haufen form in injured tubules with BKPyV replication and a high intratubular uromodulin concentration and are excreted into the urine similar to other urinary casts. In one cohort study of >300 kidney transplant recipients, the detection of Haufen in voided urine had a sensitivity, specificity, negative predictive value, and positive predictive value for biopsy-proven BKPyVAN of greater than 95 percent [110], suggesting that this may be a noninvasive way to diagnose BKPyVAN. However, the urinary Haufen test requires electron microscopy. Thus, it is not a widely used screening test, but it may be used in certain clinical scenarios, such as in pediatric patients or when a kidney allograft biopsy cannot be safely performed.

TREATMENT

Overview of treatment — Since there are no specific antiviral therapies for BK polyomavirus (BKPyV)-associated nephropathy (BKPyVAN), the cornerstone of management is to decrease immunosuppressive medications [12,84,85]. In general, this approach applies to both the prevention of BKPyVAN in patients with BKPyV viremia detected by routine screening (see 'Posttransplant screening' above) and the treatment of patients with established BKPyVAN. The optimal approach to reducing immunosuppression has not been defined; protocols vary among transplant centers and are often individualized. (See 'Reduction of immunosuppression' below.)

The initial approach of decreasing immunosuppressive medications is effective in most patients. For patients who have progressive allograft dysfunction, despite a maximal decrease in immunosuppressive therapy for a period of several weeks to months, we may try agents that may have antiviral and/or immunomodulatory activity, such as intravenous immune globulin (IVIG). However, the efficacy of this approach has not been proven. We do not use leflunomide, cidofovir, or quinolone antibiotics. (See 'Adjunctive therapies' below.)

Reduction of immunosuppression — We recommend reducing maintenance immunosuppression for most kidney transplant recipients with detectable BKPyV viremia or biopsy-proven BKPyVAN. The goals of reducing immunosuppression are to restore immunity against BKPyV without triggering rejection. Approaches to reducing immunosuppression vary among transplant centers, and there are no randomized controlled trials that directly compare different protocols.

Patients should be closely monitored while immunosuppression is being decreased since additional interventions are based upon the clinical and virologic response. The complete resolution of viremia may take several months. Prior to reducing immunosuppression, we obtain a plasma BKPyV quantitative polymerase chain reaction (PCR) (see 'Plasma quantitative PCR' above) and subsequently monitor the plasma quantitative PCR every one to two weeks until BKPyV DNA is undetectable for two consecutive tests obtained at least one week apart. In addition, we monitor the serum creatinine level weekly. If the serum creatinine level increases by ≥25 percent from baseline at any time while immunosuppression is being reduced, the patient should be evaluated for the possibility of acute rejection. (See 'Acute rejection after reducing immunosuppression' below.)

In patients who do not have concurrent acute rejection (see 'Concurrent BKPyVAN and acute rejection' below), our approach to reducing immunosuppression is as follows:

●In patients who are on a triple immunosuppression therapy consisting of a calcineurin inhibitor (tacrolimus or cyclosporine), an antimetabolite (mycophenolate mofetil/sodium or azathioprine), and prednisone, we initially reduce the dose of the antimetabolite by 50 percent. If the BKPyV viral load does not decrease within two to four weeks, we completely discontinue the antimetabolite. If there is still no decrease in viral load after another two weeks, we decrease the dose of the calcineurin inhibitor by 25 to 50 percent, targeting a whole blood tacrolimus trough level of 4 to 6 ng/mL or a whole blood cyclosporine trough level of 60 to 100 ng/mL.

An alternative approach that is used by others is to first decrease the dose of the calcineurin inhibitor by 25 to 50 percent in one or two steps, followed by reducing the antimetabolite by 50 percent, followed by discontinuing the antimetabolite [84].

●In patients who are on a glucocorticoid-free maintenance regimen with a calcineurin inhibitor and an antimetabolite without prednisone, a similar approach as described above for patients on triple immunosuppression therapy may be used. However, transplant centers may feel less inclined to use monotherapy with either a calcineurin inhibitor or antimetabolite. An alternative approach is to reduce both the calcineurin inhibitor and the mycophenolate, which allows both the targeting of two pathways and lower total immunosuppression.

●Following resolution of BKPyV viremia or biopsy-proven BKPyVAN, the decision to increase the level of maintenance immunosuppression should be individualized, taking into consideration the risk of acute rejection as well as the risk for recurrent BKPyV. Evidence from one randomized trial with long-term follow-up suggests that patients with BKPyV viremia can be safely maintained on reduced immunosuppression for at least 10 years without an increase in acute rejection or graft loss [9,17,113].

Although the cornerstone of management is to decrease immunosuppressive medications, the effectiveness of this strategy has not been evaluated systematically in randomized trials. In one meta-analysis of observational data from 40 studies, graft loss rates were similar when comparing reduction in immunosuppression alone versus reduction of immunosuppression plus either cidofovir or leflunomide [114]. Similar findings have been reported in subsequent cohort studies [79,115,116].

Adjunctive therapies — Several agents have been shown to have in vitro anti-BKPyV activity. However, we do not routinely use any of these agents for the treatment of BKPyV infection, given that the efficacy of these agents has not been established and use of these therapies has not been clearly shown to be superior to reduction in immunosuppression alone [114]:

●Intravenous immune globulin – Commercially available IVIG preparations contain BKPyV-neutralizing antibodies against all major genotypes [117]. We do not routinely administer IVIG for the treatment of BKPyVAN. However, the adjunctive use of IVIG may be considered in patients with established BKPyVAN who do not respond to a reduction in immunosuppression and who also have severe hypogammaglobulinemia (ie, immunoglobulin G [IgG] <400 mg/dL). (See "Overview of intravenous immune globulin (IVIG) therapy".)

If used, IVIG is administered at a dose of 300 mg/kg every three weeks in conjunction with a reduction in immunosuppression. We typically administer intravenous hydration with normal saline (10 to 20 mL/kg) prior to starting the infusion to mitigate the risk of acute kidney injury (AKI). We repeat 21-day trough IgG levels after three months of therapy with the goal of maintaining an IgG level >400 mg/dL. (See 'Reduction of immunosuppression' above and "Intravenous immune globulin: Adverse effects", section on 'Acute kidney injury' and "Intravenous immune globulin: Adverse effects", section on 'Strategies for reducing adverse events'.)

Limited data are available concerning the efficacy of IVIG in patients with BKPyVAN. Some observational studies have reported clearance of BKPyV viremia following IVIG therapy [118-122]. However, the findings of these studies are difficult to evaluate, since other antiviral interventions were administered concurrently.

●Leflunomide – Leflunomide is a prodrug; its active metabolite, A77 1726, has both immunosuppressive and antiviral activity [123,124]. We do not use leflunomide for the treatment of BKPyV infection given its uncertain efficacy, long half-life, the potential for hematologic toxicity and hepatotoxicity, the wide interpatient A77 1726 level variability in metabolism, and the inability to easily monitor A77 1726 levels.

Studies on the efficacy of leflunomide have yielded mixed results. In an initial case series, there was improvement or stabilization in 23 of 26 patients with BKPyVAN after discontinuing mycophenolate mofetil, targeting tacrolimus levels to 4 to 6 ng/mL, maintaining prednisone at 5 to 10 mg/day, and replacing mycophenolate with leflunomide with target levels of 50 to 100 mcg/mL. However, other observational studies have not found a benefit to treatment with leflunomide [123,125,126].

Additionally, in a phase II randomized trial, an investigational agent derived from an active metabolite of leflunomide (FK778) provided no clinically significant benefit compared with the reduction of immunosuppressive agents [127]. In this trial, 30 patients were administered FK778 in addition to having mycophenolate discontinued completely. In the control group of 16 patients, mycophenolate dose was decreased by 40 to 60 percent. In both groups, calcineurin inhibitors were decreased by 40 to 60 percent. At six months, there was a slightly greater reduction of viremia but no improvement in kidney function in the FK778 group compared with control.

●Cidofovir – Cidofovir is a nucleotide analog of cytosine that is active against various DNA viruses and is approved for both HIV-associated cytomegalovirus (CMV) retinitis and the topical treatment of genital warts. Cidofovir has modest in vitro activity against polyomaviruses [128]. It should only be considered for treatment of BKPyVAN when other interventions have failed; it has been used anecdotally, in combination with antinucleoside therapy, for HIV-infected patients with progressive multifocal leukoencephalopathy (PML) resulting from polyomavirus infection.

Only a few uncontrolled studies have evaluated the use of cidofovir in BKPyVAN [129-131]:

•In a retrospective study of 21 patients with BKPyVAN, eight were administered weekly adjuvant low-dose cidofovir plus a reduction in immunosuppressive therapy, while 13 were treated with reduced immunosuppressive therapy alone [131]. At a median period of 25 months, all allografts of those administered cidofovir survived. By comparison, 9 of 13 allografts of those not given cidofovir were lost at a median period of eight months.

•In one report of four patients, including two children, cidofovir was effective in all four [129].

However, a small randomized trial of kidney transplant recipients with BKPyVAN found that treatment with cidofovir, compared with placebo, had minimal effect on the BKPyV viral load in plasma or urine [132].

Cidofovir is potentially highly nephrotoxic, resulting in proteinuria and kidney failure in 20 percent of patients [128,133]. In addition, this agent has caused at least one case of subacute interstitial nephritis, which led to end-stage kidney disease (ESKD) [128]. It is also associated with other adverse effects.

●Quinolone antibiotics – We do not use quinolone antibiotics as an adjunctive therapy to treat BKPyV infection. Although quinolone antibiotics were initially reported to have anti-BKPyV activity [134-138], two randomized trials showed no benefit of levofloxacin given either prophylactically immediately following transplantation or as treatment for active BKPyV viremia [139,140]:

•One trial included 154 allograft recipients transplanted between 2011 and 2013 [140]. Recipients were randomly assigned to receive three months of either levofloxacin (500 mg per day) or placebo beginning within five days after transplantation. At a mean follow-up of 46 weeks, there was no difference between the levofloxacin and placebo groups in the incidence of BKPyV viruria, viremia, peak viral load, rejection, or patient or allograft survival. There was an increased incidence of infections caused by resistant bacterial isolates that are usually sensitive to quinolones in the levofloxacin versus the placebo group (58 versus 33 percent, respectively [risk ratio 1.75, 95% CI 1.01-2.98]).

•A placebo-controlled, randomized trial including 39 transplant recipients with BKPyV viremia showed no benefit of levofloxacin (given at a dose of 500 mg daily for 30 days) over placebo for either reduction of viral load at one, three, or six months, or allograft function [139]. However, participating centers used different immunosuppressive protocols, which may have affected the rate of BKPyV viremia, and different assays for BKPyV viremia [141]. Despite these limitations, this study suggests that one month of levofloxacin at this dose does not clear BKPyV viremia after transplantation.

Experimental therapies — Based upon the important role of cellular and humoral immune mechanisms in control of BKPyV infection, and the absence of other proven treatments, immune-based therapies are being actively assessed in human clinical trials:

●Virus-specific T cells (VSTs) against BKPyV are being assessed for safety, tolerability, and antiviral effect in an ongoing multicenter trial of kidney transplant recipients with BKPyV viremia, based upon encouraging preliminary results of VSTs for BKPyV-associated hemorrhagic cystitis or nephropathy in hematopoietic cell transplant recipients (ClinicalTrials.gov identifier: NCT04605484).

●BKPyV-specific monoclonal antibodies are being assessed for safety, tolerability, and antiviral effect in an ongoing multicenter trial of kidney transplant recipients with BKPyV viremia (ClinicalTrials.gov identifier: NCT04294472).

Special considerations

Concurrent BKPyVAN and acute rejection — The coexistence of BKPyVAN and acute rejection in a kidney allograft biopsy remains controversial [22,84,142,143]. (See 'Distinguishing BKPyVAN from rejection' above.)

There are no data to guide the optimal management of patients with concurrent BKPyVAN and acute rejection. Some experts advocate for treating the acute rejection first (eg, with pulse glucocorticoids) and then subsequently reducing immunosuppression as a second step once the patient has had a clinical response to antirejection treatment (ie, a decrease in serum creatinine level) [84]. By contrast, other experts avoid augmented immunosuppression and favor a reduction in maintenance immunosuppression alone [9,79,115,144,145]. If immunosuppression is augmented, more frequent monitoring of BKPyV viremia may be warranted. (See "Kidney transplantation in adults: Prevention and treatment of antibody-mediated rejection", section on 'Approach to initial therapy' and "Kidney transplantation in adults: Treatment of acute T cell-mediated (cellular) rejection", section on 'Approach based on histologic severity'.)

Acute rejection after reducing immunosuppression — Acute rejection can occur in 8 to 12 percent of kidney transplant recipients with BKPyV viremia or established BKPyVAN following a reduction in immunosuppression [17,79,116]. Acute rejection should be suspected in patients whose serum creatinine levels increase after immunosuppression has been decreased. Obtaining a kidney allograft biopsy in this setting may be helpful to establish the diagnosis of rejection. However, as discussed above, the histologic findings of acute rejection can be difficult to distinguish from those of BKPyVAN. (See 'Distinguishing BKPyVAN from rejection' above.)

The optimal approach to managing patients who develop acute rejection after reducing immunosuppression is not well defined and frequently varies from center to center. In general, we avoid augmenting immunosuppression in such patients if they have biopsy-confirmed BKPyVAN and maintain immunosuppression at the same reduced level as it was when the patient developed rejection.

KIDNEY RETRANSPLANTATION — Retransplantation in patients with graft failure due to BK polyomavirus (BKPyV)-associated nephropathy (BKPyVAN) is a reasonable option and has been successfully performed [146-150]. In general, the absence of BKPyV replication should be confirmed prior to retransplantation [12], although successful preemptive, living, related kidney transplants during active BKPyVAN with viremia have been reported [151]. We typically wait until BKPyV viremia has completely resolved before proceeding with retransplantation. Because BKPyVAN appears to be donor derived and cytotoxicity mediated, this usually necessitates continued close screening following retransplantation. Some centers recommend avoiding intense immunosuppression and graft injury. We do not routinely perform nephrectomy of the failed allograft or of the native kidneys, which may serve as a reservoir and a source of reinfection, as there are no high-quality data to support this approach.

Limited information exists concerning the outcome of a kidney transplant among patients who lost their first allograft because of BKPyVAN [150,152,153]; however, available data suggest that retransplantation in these patients is generally associated with good graft outcomes. In the largest study to examine this issue, outcomes were compared between 341 patients who were retransplanted for first graft failure resulting from BKPyVAN and 13,260 patients who were retransplanted for first graft failure resulting from other causes [150]. Five-year, death-censored graft survival rates were 91 percent for the BKPyVAN group and 84 percent for the non-BKPyVAN group. There was no significant difference in the rates of acute rejection or patient survival at one year between the two groups.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: BK virus infection in kidney transplant recipients".)

SUMMARY AND RECOMMENDATIONS

●Important background – BK polyomavirus (BKPyV) is a small DNA virus that establishes lifelong persistent infection in the renal tubular and uroepithelial cells of most of the world's population. For the majority, this infection is quiescent and benign. However, in immunocompromised patients, BKPyV can reactivate, leading to clinically significant disease. (See 'Prevalence' above and 'Pathogenesis' above.)

●Risk factors for viral replication – In kidney transplant recipients, BKPyV-associated disease typically occurs in the early posttransplant period when immunosuppression is most intense. Other factors associated with BKPyV disease include high-risk serostatus (ie, transplant from a BKPyV-seropositive donor to a seronegative recipient), impaired donor immune response to BKPyV, and donor BKPyV viruria prior to transplant. (See 'Risk factors for viral replication' above.)

●Clinical manifestations – BKPyV replication typically progress through three stages. Asymptomatic viruria occurs in approximately one-quarter to one-third of patients during the first posttransplant year. Viremia follows viruria in approximately one-half of patients. In a subset of viremic patients, viral replication progresses leading to damage to renal tubular epithelium and BKPyV-associated nephropathy (BKPyVAN). Like viruria and viremia, BKPyVAN is typically asymptomatic, and a rise in serum creatinine may be the sole presenting sign. Without control of viral replication, allograft loss can ensue within a period of months. (See 'Clinical manifestations' above and 'Pathogenesis' above.)

●Screening and diagnosis

•We recommend routine screening for BKPyVAN for all kidney transplant recipients in the early posttransplant period. Screening and preemptive reduction in immunosuppression for patients with clinically significant BKPyV viremia prevent progression to BKPyVAN in the majority of patients. The optimal screening strategy has not been determined, and approaches vary among transplant centers. We screen patients with a quantitative plasma BKPyV polymerase chain reaction (PCR; ie, viral load) at the following time points (see 'Posttransplant screening' above):

-Monthly for the first six months following transplant, then every three months until two years posttransplant, and then annually until five years posttransplant

-Whenever kidney allograft dysfunction occurs or when an allograft biopsy is performed for allograft dysfunction

•Clinical response varies and is dependent upon the degree of viremia, the patient's immunosuppressive regimen, and presence of allograft dysfunction:

-For patients with viremia (eg, viral loads >1000 copies/mL) and normal allograft function, we typically reduce immunosuppression and monitor the viral load every two to four weeks thereafter to ensure that it is downtrending.

-For patients with viremia and new-onset allograft dysfunction, we generally reduce immunosuppression and monitor viral loads every two to four weeks thereafter, when the clinical picture suggests that BKPyVAN is the most likely cause. We consider a kidney allograft biopsy if the cause for kidney allograft dysfunction is uncertain or if kidney dysfunction and/or viremia fail to resolve despite reducing immunosuppression. (See 'Posttransplant screening' above.)

•Kidney allograft biopsy is the gold standard for diagnosing BKPyVAN, assessing its severity, and evaluating for concomitant processes. However, because biopsy is invasive and sampling error can occur, a presumptive diagnosis is often made based upon the presence of significant viremia (plasma BKPyV viral load ≥10,000 copies/mL). (See 'Kidney allograft biopsy' above.)

●Treatment

•In kidney transplant recipients with detectable BKPyV viremia or biopsy-proven BKPyVAN, we recommend reducing maintenance immunosuppression. The optimal approach to reducing immunosuppression has not been defined; protocols vary among transplant centers and are often individualized. (See 'Overview of treatment' above and 'Reduction of immunosuppression' above.)

•Several agents have been shown to have in vitro anti-BKPyV activity, including intravenous immune globulin (IVIG), leflunomide, cidofovir, and quinolone antibiotics. However, we do not routinely use any of these agents for the treatment of BKPyV infection, given that the efficacy of these agents has not been established and use of these therapies has not been clearly shown to be superior to reduction in immunosuppression alone. (See 'Adjunctive therapies' above.)