In 1971, BK virus (BKV), a new member of the papovavirus family, was isolated from the urine of a 39-year-old living related renal transplant recipient who displayed ureteric obstruction (1). Today, this organism is recognized to be an important cause of renal dysfunction in the kidney transplant population (2). With increasing interest in the disease and the introduction of more potent immunosuppressive regimens, its detection rate has grown from 1.09% in 1995 to 5.09% in 2001 (3–9). However, three reports of patients treated with the combination of tacrolimus (TRL) and mycophenolate mofetil (MMF) have described incidences of 3.14%, 6.44%, and 8.00% (8–10).
The 60–80% prevalence of antiviral antibodies suggests that polyoma viruses are widely distributed among humans (11, 12). However, only immunocompromised individuals appear to display clinical disease. In renal transplant recipients, BKV may cause tubulointerstitial nephritis and ureteral stenosis. In contrast, bone marrow transplant patients have been reported to experience hemorrhagic cystitis (13), although 90% of these episodes appear to be mild (14).
The relationship of sirolimus (SRL)-based immunosuppressive therapy to polyoma virus–associated nephropathy (PVAN) is controversial. Hirsch et al. (15) reported a single case who cleared the BK viral load after discontinuation of SRL, the third drug in the patient’s immunosuppressive regimen. In three patients, Wali et al. (16) observed reduction in plasma content of BK virus by DNA polymerase chain reaction (PCR) estimates accompanying an improved creatinine clearance after rescue therapy by conversion from TRL/MMF/prednisone (Pred) to SRL. Due to these limited experiences, we sought to describe the incidence, clinical characteristics, treatment, and outcomes of PVAN among 378 recipients treated with SRL-based regimens.
MATERIALS AND METHODS
This retrospective study included 378 patients who received SRL-based immunosuppressive regimens after kidney alone (KTX; n=344) or simultaneous pancreas-kidney (SPK; n=34) transplantations between June 2000 and December 2004. Their mean overall follow-up was 43.3 months (range: 12.9 to 72.4). At 1 year posttransplant, the regimens included SRL-Pred or SRL alone (n=21), or SRL associated with cyclosporine (CsA; n=315), CsA/MMF (n=17), MMF only (n=16), TRL only (n=7), or TRL/MMF (n=2). Generally, recipients were seen at 3- to 6-month intervals after the first year. Patients who displayed renal dysfunction as evidenced by a >20% increase above the serum baseline creatinine value underwent a kidney transplant biopsy. This report compared the nine individuals who displayed histopathologic features of PVAN with the other 369 recipients in terms of demographic variables, renal function, maintenance immunosuppressive regimen, and rejection history. In addition, we have described the evolution of the viral load based upon PCR determinations, the treatment regimens, and the outcomes of the afflicted subjects.
The induction therapy for 125 patients included rabbit antithymocyte globulin (rATG; 1.5 mg/kg; Thymoglobulin, Sang Stat Medical Corp, Menlo Park, CA) based upon their presumed high immunological risk: African Americans, a panel reactive antibody level (PRA) above 20%, a history of previous transplantation, or receipt of an SPK transplant. An alternate regimen employing basiliximab (20 mg iv on days 0 and 4 posttransplantation; Novartis, Basel, SZ) was administered to 117 patients, whereas the other 136 recipients, primarily including those with living donor grafts, did not receive induction antibody treatment. The level of CsA exposure de novo varied from 0.3 to 7 mg/kg/day (mean: 1.80; median: 1.83) with a corresponding 1-month mean trough value (C0) of 87.1±54.6 (range: 25–262) ng/mL. Recognizing the adverse renal interactions between CsA and SRL, CsA exposure was rapidly tapered at 6 months to a mean C0 of 58.8±40.2 (range: 24–287) ng/mL. These levels were 60–70% lower than those which are targeted for patients not receiving SRL.
The initial SRL doses were adjusted to achieve target trough concentrations of 10–15 ng/mL and were reduced to 10±3 ng/mL within 3 months. The SRL C0 values were similar among patients irrespective of the combination of immunosuppressants. Prednisone was routinely tapered to 5 mg daily (or less) by 3 months and withdrawn in 50% of the patients within 2 years.
Two core kidney biopsies obtained from patients displaying renal dysfunction were fixed in formalin, embedded in paraffin, sectioned at 4 μm, and stained with hematoxylin and eosin as well as periodic acid-Schiff. The diagnosis of PVAN was based on the presence of tubular intranuclear viral inclusions typical of polyoma virus infection. The irregular and enlarged epithelial nuclei showed peripheral margination of chromatin along the nuclear basement membrane and a smudgy, lavender color central clearing. Glomerular parietal epithelial cells sometimes showed similar viral inclusions. The histological findings were confirmed by immunohistochemical stains for polyoma/BK virus based upon SV40 large T antigen, which shows 70% homology with human polyoma virus and JC virus. The results were reported according to the semiquantitative classification proposed by Hirsch et al. (17): PVAN A was minimal to mild viral cytopathic changes (cy1+ to cy2+), insignificant inflammatory infiltrates (<i1+), tubular atrophy (<ct1+), or fibrosis (<ci1+); PVAN B was viral cytopathic changes scoring from mild to severe (cy2+ to cy4+) and significant inflammatory infiltrates scoring from moderate to severe (i1+ to i3+), but only mild tubular atrophy (≤ct2+) and mild fibrosis (≤ci2+); PVAN C was moderate to severe tubular atrophy (ct3+) and interstitial fibrosis (ci2+ to ci3+), and variable scores for cytopathic changes (cy1+ to cy4+) and inflammatory infiltrates (i1+ to i3+).
After histopathologic confirmation of PVAN, the viral load was quantified by PCR of BKV DNA in plasma and urine (Viracor, Lee’s Summit, MO); these tests were repeated every 2 weeks during the treatment course. The level of virus in plasma (copies/mL) was used to assess treatment efficacy.
The initial therapeutic approach was a reduction in the intensity of maintenance immunosuppression. MMF was immediately terminated and calcineurin antagonists withdrawn over a 4-week period. Prednisone was tapered to 5 mg daily or less. SRL exposure was not changed from a C0 target of 10±3 ng/mL. Following vigorous hydration, intravenous antiviral treatment with cidofovir was started at 0.25 mg/kg every 2 weeks. In the presence of an increasing viral load, as assessed by PCR, the dose was increased by 0.1 mg/kg to a maximum of 0.5 mg/kg.
The cohort was described using estimates of central tendency (means, medians) and spread (standard deviation, range) for continuous data; frequencies and percentages were derived for categorical data. Differences between the groups were assessed by univariate analysis using the exact chi-square test for categorical variables and analysis of variance (ANOVA) for numerical variables. Significant covariates in the univariate analysis were included in a multiple logistic regression model. For all analyses, the significance level for the P value was set at 0.05.
Characteristics of the PVAN Cohort
At a mean follow-up of 43.3 months, 6 of 344 (1.7%) KTX and 3 of 34 (9.0%) SPK patients were diagnosed with PVAN (overall incidence: 2.4%). The demographic features and selected clinical metrics of the four women and five men with PVAN are shown in Table 1. The mean time to diagnosis after transplantation was 18.2 months (range: 3.5–31.1 months). The mean age at the time of PVAN diagnosis was 44.5±9.4 years (range: 25–57 years). All patients had received kidneys from deceased donors with four to six HLA mismatches and four patients had a pretransplant diagnosis of type I diabetes mellitus. The mean SRL trough level at the time of diagnosis was 11.2±2.4 ng/dL (range: 8.8 to 16.5). Among these nine patients, the distribution of maintenance immunosuppressive regimens was SRL alone or SRL-Pred (0 of 21), CsA-SRL (6 of 315, 1.9%), TRL-SRL (1 of 7, 14.2%), MMF-SRL (1 of 16, 6.2%), or CsA-MMF-SRL (1 of 17, 5.9%). The small sizes of the various cohorts obviated a rigorous statistical analysis. Three patients had experienced a previous rejection episode and eight of nine patients had been exposed to rATG: three patients for rejection and seven patients for induction treatment. The mean number of doses of rATG used during induction was 7.2 and the mean dose was 1.3 mg/kg/day.
Risk Factors for Occurrence of PVAN
Sex, age, ethnicity, presence of diabetes mellitus, cold ischemia time, occurrence of delayed graft function, and previous rejection episode did not correlate with the occurrence of nephropathy (Table 2). Furthermore, there was no association with the level of drug exposure. The CsA trough levels (C0) at 1 year were similar between affected versus nonaffected patients, namely 60.1±41.3 ng/mL versus 61.3± 51.9 ng/mL, respectively. In addition, the mean SRL C0 was not significantly different among affected versus nonaffected subjects over the first year: namely, at 6 months (11.9±1.9 ng/mL versus 12.5±5.0 ng/mL, P=0.74) and 12 months (11.7±3.5 ng/mL versus 11.5±3.9 ng/mL, respectively, P=0.99).
Three patients had experienced a previous rejection episode and eight out of nine had been exposed to rATG. Patients exposed (4.2%) versus not exposed (0.5%) to rATG for either induction or antirejection treatment showed an increased incidence of PVAN (odds ratio [OR]: 8.35; 95% confidence interval [CI]: 1.03–67.47; P= 0.019). Upon multivariate analysis, induction treatment with rATG correlated with a higher incidence of PVAN (3.53% versus 1.44%; one-sided OR: 4.38; CI: 1.011–∞; P=0.048), while exposure for antirejection therapy showed only a trend toward a higher incidence of PVAN compared with patients treated with other drugs, namely, 4.76% versus 3.13% (P=0.723).
A significantly greater incidence of PVAN was observed among patients receiving SPK (8.82%) versus KTX (1.74%; P=0.009) which was confirmed upon multivariate analysis (OR: 5.43; CI: 1.29–22.8; P=0.038). Using a multiple logarithmic regression model among SPK patients, the higher incidence of PVAN was not explained by diabetes mellitus or by SRL C0 levels. However, because all SPK patients received rATG induction, this variable could not be excluded as contributing to PVAN in multiple logarithmic regression models.
Trends toward a greater incidence of PVAN were observed among recipients of deceased (3.18%) versus living donor organs (0.0%; P=0.080), and among hosts with greater numbers of HLA mismatches with their donors (5.0 versus 4.26; P=0.160).
Presentation and Clinical Course of PVAN
Deterioration of kidney function signaled the need for a transplant biopsy, which revealed PVAN. The observed serum creatinine concentration at the time of diagnosis had increased by 41% to 268% from the baseline value. PVAN Hirsch grade B was diagnosed in six patients and Hirsch grade C in the other three patients. The average BK virus load in plasma and urine at the time of diagnosis was 3.11E+06 and 1.18E+09 copies/mL, respectively (Table 3).
The average SRL trough level during antiviral treatment was 10.2±2.7 ng/dL. All patients were started on cidofovir at the time of diagnosis except the first, who initially showed a good response to decreased immunosuppression. The second and fifth cases showed marked reductions and clearance of the plasma viral load after the 22nd and 11th doses of cidofovir, respectively, allowing discontinuation of the antiviral medication. Unfortunately, after 2 months, they experienced a relapse with increased numbers of viral copies in plasma despite stable kidney function. The seventh case presented with an aggressive course requiring dialysis within 2 weeks (Table 4).
At a mean postdiagnosis followup of 17.5 months, the graft survival of patients with diagnosis of PVAN was 44.4%, with five patients requiring dialysis at 2, 29, 30, 43, and 56 weeks after diagnosis. The mean graft-survival time was 42.1 weeks (range: 2–140 weeks). Graft survival of patients with histological grade B of PVAN (50%) was similar to that of subjects with grade C (66%; Table 3).
One SPK patient experienced a venous thrombosis of the pancreas graft at 1 week after transplantation and lost the kidney due to BKV nephropathy at 31 weeks after the diagnosis. The other two patients have adequate kidney function with good pancreatic function despite discontinuation of the calcineurin inhibitor and anti-viral treatment.
The increased incidence of PVAN seems to reflect the more potent immunosuppressive effects of new agents and the improved diagnostic tools. The overall intensity of the immunosuppressive regimen has been reported to be the most important risk factor. In contrast to previous experience with TRL/MMF regimens reporting rates of up to 8% (8), our experience revealed a lower incidence of PVAN among kidney-only transplant patients treated with SRL-based regimens (1.7%), and particularly the absence of infections among a limited cohort of patients on SRL-alone or SRL combined with prednisone. However, to establish such a protective effect will require a large prospective study. It seems possible that the less potent immunosuppression produced by SRL-only regimens compared with those including a calcineurin antagonist and/or MMF may explain this observation.
As in a previous report (5), we observed a higher rate of BKV nephropathy among patients who had undergone SPK compared with kidney transplantation alone (9.0% versus 1.7%, respectively; P=0.038). The difference may relate to the more intense regimen used with SPK, including uniform antithymocyte antibody induction. In a parallel clinical setting, T-cell dysfunction has been considered to be a key factor leading to reactivation of JC virus among human immunodeficiency virus (HIV) patients (18). In addition, Jin et al. (19) reported a negative correlation between CD4 cell count and BKV prevalence among HIV patients. However, the role of antilymphocyte preparations as risk factors for PVAN in kidney transplantation has been controversial (20–22). In our analysis, the use of rATG for induction or during rejection treatment did not reach statistical significance as separate variables. However, the combination of patients exposed to rATG during either induction or rejection treatment showed a significant association with PVAN (OR: 8.35; 95% CI: 1.03–67.47; P= 0.019).
A history of previous rejection episodes has not been consistently regarded to be a risk factor (22, 23). We found a trend toward PVAN among patients who experienced a previous rejection episode, especially when they received rATG treatment. The association with rejection might reflect either the renal injury itself, possibly exacerbated by the nephrotoxic effects of calcineurin antagonists, or the augmented immunosuppression required for its treatment. As in previous reports (20, 24), our analysis showed a weak association with HLA mismatches, probably reflecting a higher incidence of rejection episodes and the consequent use of antilymphocyte therapy.
The 18.2-month mean time to PVAN diagnosis after transplantation was slightly beyond that observed in previous studies, namely, 8 to 13 months (6–10). However, our lack of routine surveillance for BKV by PCR and of a uniform follow-up protocol beyond 1 year mitigates any conclusion about the effects of sirolimus to delay the onset or retard the progression of the disease.
Some authors have argued that prolonged cold ischemic time (CIT) may promote virus replication in the injured kidney. However, there has not been a consistent relationship between the outcome of deceased donor kidneys and the occurrence of PVAN (20, 25, 26). Although our analysis failed to show a significant relationship between the occurrence of PVAN and CIT, the incidence of PVAN tended to be higher among recipients of organs from deceased versus living donors (3.18% versus 0%, respectively, P=0.080), suggesting the possibility of an association. However, as reported in other studies, we did not observe even a weak association between the presence of diabetes mellitus and PVAN (5, 27).
Our experience showed that reduced immunosuppression after diagnosis of BKV did not eventuate in allograft rejection even among SPK recipients. Seven of nine patients were solely treated with SRL and modest (≤5 mg) doses of prednisone. One other patient was off prednisone at the time of diagnosis, whereas the last one was withdrawn from prednisone after the diagnosis of PVAN. No patient experienced a rejection episode during the treatment course, suggesting that they had been excessively immunosuppressed prior to the diagnosis. Pancreatic function continued to be excellent among the SPK patients.
Although the use of cidofovir, an acyclic nucleoside phosphonate, has not been approved for the treatment of BKV by the U.S. Food and Drug Administration, initial studies have demonstrated in vitro activity against the virus (28) and some initial clinical experiences have been positive (29, 30). In this analysis, patients did not display extrarenal side effects related to the medication. However, most patients experienced adverse outcomes suggesting either cidofovir nephrotoxicity, lack of drug therapeutic efficacy, prior calcineurin antagonist injury, or, more likely, a delay in diagnosis that permitted the slow advance of this graft-threatening disease. The last alternative is consistent with the relatively aggressive histological features displayed by our cohort: grade B in six patients and grade C in three patients.
An alternative therapeutic drug, leflunomide (31), has been proposed for treatment of BKV nephropathy. Its active metabolite, A77 1726, has been claimed to inhibit the synthesis of the viral envelope. The in vitro activity using embryonic lung fibroblast and a Gardner strain of BKV showed a modest and low selectivity index against the virus (32). Nevertheless, some initial clinical experiences have been positive (33, 34).
Finally, studies utilizing a screening protocol for BKV diagnosis have reported uniform graft survival (20, 22), documenting that early detection avoids the late outcomes of nephropathy and fibrosis. In contrast, prior studies without surveillance showed wide differences among graft-loss rates due to PVAN, namely, 11–71% (5, 8–10, 35), a range that includes our 55% observed loss. Whether reduced SRL exposures from C0 ≈ 10±3 ng/mL would improve the outcomes needs to be evaluated in future studies.
Compared to studies without routine screening for BK virus, our experience suggested that patients treated with SRL-containing immunosuppressive regimens displayed a relatively low incidence of PVAN. The disease appeared to be virulent in course, suggesting the need for surveillance in patients at increased risk (particularly after rATG treatment) for new therapies and for novel algorithms of maintenance immunosuppression. Larger studies are needed to evaluate the cost-benefit yield of minimizing the occurrence of BKV infections by routine surveillance versus the utility of SRL therapy alone or with modest doses of prednisone.
The authors are indebted to Charles Green, PhD, of The Center for Clinical Research and Evidence-Based Medicine at the University of Texas Medical School at Houston for the statistical analysis, to the nurses and nurse coordinators of the Renal Transplant Center, and Ms. Maude Veech for editorial assistance.
1. Gardner SD, Field AM, Coleman DV, Hulme B. New human papovavirus (B.K.) isolated from urine after renal transplantation. Lancet
1971; 1: 1253.
2. Randhawa P, Brennan DC. BK virus
infection in transplant recipients: an overview and update. Am J Transplant
2006; 6: 2000.
3. Buehrig CK, Lager DJ, Stegall MD, et al. Influence of surveillance renal allograft biopsy on diagnosis and prognosis of polyomavirus-associated nephropathy
. Kidney Int
2003; 64: 665.
4. Ramos E, Drachenberg CB, Portocarrero M, et al. BK virus
nephropathy diagnosis and treatment: Experience at the University of Maryland Renal Transplant Program. Clin Transpl
5. Trofe J, Gaber LW, Stratta RJ, et al. Polyomavirus in kidney and kidney-pancreas transplant recipients. Transpl Infect Dis
2003; 5: 21.
6. Howell DN, Smith SR, Butterly DW, et al. Diagnosis and management of BK polyomavirus interstitial nephritis in renal transplant recipients. Transplantation
1999; 68: 1279.
7. Rahamimov R, Lustig S, Tovar A, et al. BK polyoma virus
nephropathy in kidney transplant recipient: The role of new immunosuppressive agents. Transplant Proc
2003; 35: 604.
8. Mengel M, Marwedel M, Radermacher J, et al. Incidence of polyomavirus-nephropathy in renal allografts: Influence of modern immunosuppressive drugs. Nephrol Dial Transplant
2003; 18: 1190.
9. Ramos E, Drachenberg CB, Papadimitriou JC, et al. Clinical course of polyoma virus
nephropathy in 67 renal transplant patients. J Am Soc Nephrol
2002; 13: 2145.
10. Rocha PN, Plumb TJ, Miller SE, et al. Risk factors for BK polyomavirus nephritis in renal allograft recipients. Clin Transplant
2004; 18: 456.
11. Shah KV, Daniel RW, Warszawski RM. High prevalence of antibodies to BK virus
, an SV40-related papovavirus, in residents of Maryland. J Infect Dis
1973; 128: 784.
12. Flaegstad T, Ronne K, Filipe AR, Traavik T. Prevalence of anti BK virus
antibody in Portugal and Norway. Scand J Infect Dis
1989; 21: 145.
13. Leung AY, Suen CK, Lie AK, et al. Quantification of polyoma BK viruria in hemorrhagic cystitis complicating bone marrow transplantation. Blood
2001; 98: 1971.
14. Coleman DV, Gardner SD, Field AM. Human polyomavirus infection in renal allograft recipients. BMJ
1973; 3: 371.
15. Hirsch HH, Mohaupt M, Klimkait T. Prospective monitoring of BK virus
load after discontinuing sirolimus
treatment in a renal transplant patient with BK virus
nephropathy. J Infect Dis
2001; 184: 1494.
16. Wali RK, Drachenberg C, Hirsch HH, et al. BK virus
in renal allograft recipients: Rescue therapy by sirolimus
-based immunosuppression. Transplantation
2004; 78: 1069.
17. Hirsch HH, Brennan DC, Drachenberg CB, et al. Polyomavirus-associated nephropathy
in renal transplantation: Interdisciplinary analyses and recommendations. Transplantation
2005; 79: 1277.
18. Dubois V, Dutronc H, Lafon ME, et al. Latency and reactivation of JC virus in peripheral blood of human immunodeficiency virus type 1-infected patients. J Clin Microbiol
1997; 35: 2288.
19. Jin L, Pietropaolo V, Booth JC, et al. Prevalence and distribution of BK virus
subtypes in healthy people and immunocompromised patients detected by PCR-restriction enzyme analysis. Clin Diagn Virol
1995; 3: 285.
20. Hirsch HH, Knowles W, Dickenmann M, et al. Prospective study of polyomavirus type BK replication and nephropathy in renal-transplant recipients. N Engl J Med
2002; 347: 488.
21. Binet I, Nickeleit V, Hirsch HH, et al. Polyomavirus disease under new immunosuppressive drugs: A cause of renal graft dysfunction and graft loss. Transplantation
1999; 67: 918.
22. Brennan DC, Agha I, Bohl DL, et al. Incidence of BK with tacrolimus versus cyclosporine and impact of preemptive immunosuppression reduction. Am J Transplant
2005; 5: 582.
23. Nickeleit V, Hirsch HH, Binet IF, et al. Polyomavirus infection of renal allograft recipients: From latent infection to manifest disease. J Am Soc Nephrol
1999; 10: 1080.
24. Awadalla Y, Randhawa P, Ruppert K, et al. HLA mismatching increases the risk of BK virus
nephropathy in renal transplant recipients. Am J Transplant
2004; 4: 1691.
25. Kazory A, Ducloux D. Renal transplantation and polyomavirus infection: Recent clinical facts and controversies. Transpl Infect Dis
2003; 5: 65.
26. Priftakis P, Bogdanovic G, Tyden G, Dalianis T. Polyomaviruria in renal transplant patients is not correlated to the cold ischemia period or to rejection episodes. J Clin Microbiol
2000; 38: 406.
27. Hogan TF, Borden EC, McBain JA, et al. Human polyomavirus infections with JC virus and BK virus
in renal transplant patients. Ann Intern Med
1980; 92: 373.
28. Andrei G, Snoeck R, Vandeputte M, De CE. Activities of various compounds against murine and primate polyomaviruses. Antimicrob Agents Chemother
1997; 41: 587.
29. Vats A, Shapiro R, Singh RP, et al. Quantitative viral load monitoring and cidofovir therapy for the management of BK virus
in children and adults. Transplantation
2003; 75: 105.
30. Kadambi PV, Josephson MA, Williams J, et al. Treatment of refractory BK virus
with cidofovir. Am J Transplant
2003; 3: 186.
31. Chong AS, Huang W, Liu W, et al. In vivo activity of leflunomide: Pharmacokinetic analyses and mechanism of immunosuppression. Transplantation
1999; 68: 100.
32. Farasati NA, Shapiro R, Vats A, Randhawa P. Effect of leflunomide and cidofovir on replication of BK virus
in an in vitro culture system. Transplantation
2005; 79: 116.
33. Poduval RD, Kadambi PV, Javaid B, et al. Leflunomide-a potential new therapeutic agent for BK nephropathy. Am J Transplant
2003; 3 Suppl 5: 189.
34. Williams JW, Javaid B, Kadambi PV, et al. Leflunomide for polyomavirus type BK nephropathy. N Engl J Med
2005; 352: 1157.
35. Li RM, Mannon RB, Kleiner D, et al. BK virus
and SV40 co-infection in polyomavirus nephropathy. Transplantation
2002; 74: 1497.