BK Nephropathy in Kidney Transplant Recipients Treated with a Calcineurin Inhibitor-Free Immunosuppression Regimen
Abstract
Recently, polyomavirus-associated nephropathy (PVAN) has been reported more frequently and is emerging as an important cause of renal allograft dysfunction and graft loss. Susceptibility appears to be related to the type and intensity of pharmacologic immunosuppression but some reports have suggested a link among the development of PVAN, the treatment of rejection or maintenance with a tacrolimus-based immunosuppressive regimen. We report three cases of PVAN in patients who never received immunosuppression with calcineurin inhibitors (CNIs). Two patients received induction immunosuppression consisting of an IL-2 receptor antagonist while 1 received thymoglobulin. These 3 patients were maintained on prednisone, sirolimus and mycophenolate mofetil (MMF) and none was treated for rejection. All three patients presented with an elevated serum creatinine and demonstrated polyomavirus infection on biopsy and by blood PCR. These cases demonstrate that, unlike reports linking tacrolimus and PVAN, polyomavirus infection may develop in patients maintained on CNI-free immunosuppressive regimens and have not had episodes of rejection.
Introduction
Polyomavirus infection in kidney transplant recipients has emerged as an important cause of renal allograft loss. Infection in the transplanted kidney may result in non-anastomotic ureteral strictures or in tubulointerstitial nephritis compromising kidney allograft function and ultimately resulting in loss. Polyomavirus-associated nephropathy (PVAN), caused by BK virus (BKV), may be diagnosed by immunohistochemistry, in situ hybridization, PCR for polyoma virus from blood, urine and transplant biopsy tissue and the detection of decoy cells in urine (1–5). Recently there has been an increase in the reported incidence of PVAN, due to several reasons. Improvement in screening methods has allowed for differentiation of PVAN from allograft rejection (6). In addition, some have suggested that the development of more potent immunosuppression has led to an increase in the prevalence of PVAN. Some reports have suggested that PVAN is linked to the use of immunosuppression regimens containing tacrolimus (6–8). However, PVAN has not been exclusive to patients who received tacrolimus as it has been seen in patients maintained on the other calcineurin inhibitor (CNI), cyclosporine. Others believe it is related to a state of general overimmunosuppression (9).
The purpose of the present communication is to report the development of three cases of PVAN in patients who had not received CNIs as part of their immunosuppressive regimen. In part we hope that this will serve to support the concept that the risk of PVAN is not associated with a single immunosuppressive agent but instead with the overall state of immunosuppression.
Case Reports
Case 1
A 68-year-old native American female with end-stage renal disease secondary to hypertension presented for a deceased donor kidney transplant. Her CMV serology was D+R+. The donor and recipient were matched at 1 HLA locus. In the operating room she received 500 mg of methylprednisolone and 91 mg of daclizumab. On post-operative day 4, the patient was discharged with a creatinine of 4.3 mg/dL. Discharge immunosuppression consisted of a prednisone taper, mycophenolate mofetil (MMF) 1000 mg b.i.d. and sirolimus 5 mg/day. On follow-up 4 days later, the patient's serum creatinine was 2.0 mg/dL and 3 months later 0.9 mg/dL.
Three and a half months after transplantation, she presented with a serum creatinine of 1.9 mg/dL and an ultrasound demonstrated hydrouteronephrosis. She had not suffered any episodes of rejection during this time. Immunosuppression consisted of prednisone 5 mg/day, MMF 250 mg b.i.d. and sirolimus 3 mg/day (level 11.5 ng/mL; target range 8–15 ng/mL). A pyelogram demonstrated a narrow stricture proximal to the ureterovesical junction. Urine PCR for BKV was 7.1 × 107 DNA copies/mL. Repeat urine PCR for BKV 2 weeks later demonstrated 1.1 × 107 copies/mL while PCR from blood demonstrated 5.7 × 104 copies/mL (quantitative PCR for all specimens performed by Viracor, Lee's Summit, MO). A kidney transplant biopsy demonstrated tubulointerstitial nephritis with cytopathic change consistent with polyoma virus infection. The patient's immunosuppression was changed to microemulsion cyclosporine and prednisone. Three weeks later the patient's creatinine peaked at 4.9 mg/dL. Intravenous cidofovir was begun at this time (0.25 mg/kg dose, monthly, for 2 doses, without probenecid) when urine BKV was 1.1 × 108 copies/mL with PCR of the blood demonstrating 1.7 × 104 copies/mL. The patient began to respond 3 weeks later as her serum creatinine declined to 2.6 mg/dL and BKV in the blood declined to 9.6 × 103 copies/mL. Two and a half months after beginning therapy and immunosuppression reduction, blood BKV declined to 2.1 × 103 copies/mL and became undetectable 2.5 months later.
Case 2
A 62-year-old white male with end-stage renal disease secondary to polycystic kidney disease received a live donor kidney transplant from his nephew. His CMV serology was D−R+. The donor and recipient were matched at 3 HLA loci. In the operating room he received 500 mg of methylprednisolone and 20 mg of basiliximab. He received a second dose of 20 mg of basiliximab on the 4th post-operative day. His course was unremarkable and he was discharged on a prednisone taper, MMF 1000 mg b.i.d. and sirolimus 5 mg/day. His sirolimus was concentration controlled to 10–12 ng/mL the first 6 months. His serum creatinine was 1.3 mg/dL after the first month. During the 4th month he grew 3000 copies/mL of CMV from whole blood, which cleared after valganciclovir, and was undetectable 2 months later.
Between 6 and 8 months after the transplant, the patient experienced an elevated serum creatinine of 1.8–2.3 mg/dL, and two renal biopsies were Banff borderline (10). At month 10, a third biopsy demonstrated tubular nuclear enlargement, BKV by in situ hybridization and over 2 × 105 copies/mL of BKV was detected from the blood by quantitative PCR (performed by Focus Technologies, Cypress, CA). Immunosuppression was reduced to sirolimus 2 mg/day and prednisone 5 mg/day, and the patient was treated with biweekly doses of cidofovir for 3 months (0.33 mg/kg dose without probenecid). The treatment was not successful, renal function deteriorated and he returned to hemodialysis 14 months after transplantation. After cidofovir therapy at the time he returned to hemodialysis, his blood viral load was at its peak (3 × 106 copies/mL). Two months later, and after discontinuing immunosuppression, blood BKV declined to 7.6 × 102 copies/mL.
Case 3
A 61-year-old African American male with end-stage renal disease secondary to focal segmental glomerulosclerosis presented for a deceased donor kidney transplant. His CMV serology was D−R+. The donor and recipient were matched at 2 HLA loci. The graft was initially slow to function but the recipient did not require hemodialysis. He was given 3 doses of thymoglobulin (1.5 mg/kg per dose) on post-transplant day 0, 2 and 4. Methylprednisolone 500 mg was introduced on day 0 and tapered to prednisone 20 mg/day by day 4. He was also given MMF 1500 mg b.i.d. starting on day 0. Sirolimus 4 mg/day was started on day 1 and the trough level was maintained at 10–15 ng/mL. On day 5, his serum creatinine was 3.5 mg/dL and he was discharged home. His serum creatinine stabilized at 2.3 mg/dL.
Five months later a biopsy was performed for a rising serum creatinine (3 mg/dL) and BKV was diagnosed by intranuclear inclusions on microscopy and positive staining for SV40 T antigen on immunohistochemistry. His immunosuppression was sirolimus 7 mg/day (trough level 11.8 ng/mL), MMF 1000 mg b.i.d. and prednisone 10 mg/day. Blood and urine were also positive (blood 1.3 × 103 copies/mL; urine 2.2 × 106 copies/mL) (both performed by Viracor, Lee's Summit, MO). Sirolimus was discontinued and prednisone was increased to 20 mg/day. The patient's serum creatinine decreased to 2.4 mg/dL after 3 weeks. One month later, he developed a CMV infection characterized by fever, generalized weakness and malaise. CMV antigen was positive in 310 out of 300 000 leukocytes, which was successfully treated with ganciclovir.
One year after the transplant, the patient has continued to do well with a serum creatinine at baseline (2.2 mg/dL). He does not have viremia but has low-level viruria (1.3 × 104 copies/mL). He is maintained on MMF 1000 mg b.i.d. and prednisone 5 mg/day.
Discussion
PVAN has been associated with early graft loss following kidney transplantation. It is generally of low incidence (1) but the diagnosis has been increasing recently (11). A relationship between this increase in incidence and the use of more potent immunosuppressive agents appears likely. Since these agents have greatly reduced the rates of acute cellular rejection, as a consequence they may have created an environment that increases the incidence of polyomavirus reactivation (1,12). It is, however, more likely that it is the intensity of the immunosuppression rather than the specific immunosuppressant agent that is responsible for the development of PVAN.
Reports have concluded that PVAN is linked to the use of immunosuppression regimens containing tacrolimus (6–8), and not just to a state of overimmunosuppression (9). Some have suggested this relationship because the incidence of BKV nephropathy appeared to increase after the introduction of tacrolimus. This link, as has been suggested by some, may instead simply represent the widespread use of tacrolimus in most transplant centers today (13). It is generally believed that tacrolimus has greater immunosuppressive efficacy than cyclosporine. However, no study has in fact addressed this in a controlled manner. In a recent report (14), though, a randomized study of 50 kidney transplant patients were evaluated prospectively after treatment with either tacrolimus or cyclosporine, the authors found no difference in the incidence of polyoma virus infection between the two groups during 36 weeks after transplantation. Currently, a large worldwide-randomized trial of cyclosporine versus tacrolimus (the DIRECT Trial) is being conducted; it will assess the risk of developing BKV infection as one of its endpoints.
Regardless of whether tacrolimus or cyclosporine plays a greater role, the patients described in this series did not receive any CNIs as part of their immunosuppressive regimen and thus evidence to suggest a causal link between CNIs and the development of BKV nephropathy appears to be lacking. A fourth case does exist in the literature (15). In a multicenter trial initiated at the Mayo clinic, one patient in one arm of a cyclosporine avoidance trial utilizing sirolimus, MMF and steroids (after Thymoglobulin induction), also developed BKV nephropathy (15). Whether the development of PVAN with the combination of MMF and sirolimus is a reflection of the immunosuppression intensity of this regimen or the result of the additive effects of two antiproliferative agents is unclear.
Other risk factors may play a role in the development of PVAN, such as the number of HLA mismatches (3), which may decrease the efficiency of MHC-linked antiviral immunity (16). An increased incidence of PVAN was also found in recipients who were male (11,17) and older (11). Additional risk factors may include the presence of CMV disease as BKV has the potential to increase the amount of regulatory proteins involved in the pathogenesis of CMV (18); whether there is reciprocal transactivation is unclear. Clinical reports in patients have been conflicting.
The surgical techniques of transplantation result in warm ischemia, reperfusion injury and acute tubular necrosis resulting in the release of proinflammatory cytokines, cellular injury and expression of virus-specific cell surface receptors (16) all promoting an invasive viral state. However, one report found no statistically significant relationship between allograft type (deceased or living donor) or duration of cold ischemia and PVAN (3). In the present report, 2 out of 3 patients were male, all were older than 60, the average HLA mismatch rate was high (4 loci) and 2 developed CMV disease. Together these factors along with general pharmacologic immunosuppression, not related to a single immunosuppressive agent, may have played a combined role in the development of PVAN.
