Patients and Methods

Study population, samples, and definitionsCase patients included 4 kidney transplant recipients with histopathologically diagnosed BK virus nephropathy. Control patients included 16 transplant recipients from the same center who did not develop BK virus nephropathy. The control group included patients with allograft dysfunction attributable to a variety of causes, including cytomegalovirus nephritis, cellular rejection, vascular rejection, acute tubular necrosis, focal segmental glomerulosclerosis, and glomerulonephritis. BK virus nephropathy was diagnosed by the presence of intranuclear viral inclusion bodies in epithelial cells and was confirmed by immunohistochemical detection of SV40 antigen in renal biopsy specimens. Serum samples were collected from patients at various times (at the time of transplantation, during the initial hospitalization after transplantation, and during readmissions to the hospital) and were stored at −70°C until analyzed

PCR methodsAll serum samples were analyzed without knowledge of the patient’s clinical status (i.e., case or control). A real-time fluorescent probe–based PCR (TaqMan; Applied Biosystems) method was used to quantitate BK virus in serum samples. DNA was extracted from 400 μL of serum. One-tenth of the total extracted DNA was used for each PCR reaction. The primers Pep-1 (AGT CTT TAG GGT CTT CTA CC) and Pep-2 (GGT GCC AAC CTA TGG AAC AG) and probe BKTAQ1 (GCA ACA GCA GAT TCT CAA CAC TCA ACA) [13] were used in the Taqman assay. BKTAQ1 was labeled at the 5′ end with 6-carboxyfluorescein and at the 3′ end with 6-carboxytetramethylrhodamine (Synthegen). The PCR conditions were as described elsewhere [14]. The specificity of the BK virus PCR was verified by testing against JC virus and human genomic DNA

The threshold for detection of the PCR assay was ⩾1 DNA copy per reaction (equivalent to 25 copies of BK DNA/mL serum). Each specimen was run in duplicate and was paired with a negative control, to coprocess through the entire procedure. Negative control reactions without DNA also were included in every run. To ensure that negative results were not due to nonspecific inhibition of the PCR, each PCR reaction was spiked with 50 copies of internal control jellyfish gene DNA (EXO), primers EXO186F and EXO314R (30 nM each), and probe PiMP-242T (50 nM), as described elsewhere [15]. All negative PCR results required the detection of EXO DNA to be considered to be valid

Statistical analysisThe Mann-Whitney&rank sum test and Fisher’s exact test were used as appropriate

Results

Patient characteristicsAs shown in table 1, there were no significant differences between the case and control patients with regard to age, sex, underlying disease leading to transplantation, transplant type, cytomegalovirus serostatus, immunosuppression, or antiviral prophylaxis (P>.05 for all comparisons)

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Figure 1

Time course of serum creatinine level, serum BK virus load, and kidney biopsy histopathology in 4 patients with BK virus nephropathy. Data points are represented by □ (serum BK DNA) and ⋄ (serum creatinine). + and −, Presence and absence, respectively, of specified histologic finding on renal biopsy; arrow transplant nephrectomy

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Table 1

Characteristics for 4 transplant recipients with documented BK virus nephropathy (case patients) and for 16 transplant recipient patients (control patients)

Serum PCR resultsTwenty-eight serum samples from 4 case patients (mean, 7 serum samples per patient; range, 4–9 serum samples) and 76 serum samples from the 16 control patients (mean, 5 serum samples per patient; range, 4–7 serum samples) were analyzed (P=.064, Mann-Whitney&amp;rank sum test). The timing of obtaining serum samples after transplantation was similar for the case and control patients (data not shown). Case patients were more likely than were control patients to have BK virus DNA detected in serum samples. This result was the same when either the total number of patients or the total number of specimens were considered (4 of 4 case patients compared with 0 of 16 control patients, and 11 of 28 specimens compared with 0 of 76 specimens, respectively; P<.0001 for both comparisons, Fisher’s exact test)

Figure 1 shows the time course of serum PCR results, serum creatinine levels, and renal biopsy results for each case patient

Patient 1BK virus DNA was first detected in the blood of patient 1 at 25 weeks after transplantation, which was concomitant with an increase in the serum creatinine level. However, renal biopsy at this time showed only nonspecific inflammation without evidence of rejection or BK virus nephropathy. About 18 weeks later, after progressive deterioration in renal function, a renal biopsy demonstrated persistent inflammation and BK virus nephropathy. Stepwise reduction in immunosuppression was subsequently associated with a reduction in serum BK virus load, clearance of BK viropathic effects from the renal biopsy, and some improvement in renal function

Patient 2Patient 2 had an increase in serum creatinine level 20 weeks after transplantation. Renal biopsy demonstrated acute cellular rejection and nonspecific inflammation, and the patient was treated with pulse-dose corticosteroid therapy. BK virus was first detected in serum at this time (530 copies/mL); however, no evidence of BK nephropathy was noted on 3 separate renal biopsy specimens, and renal function initially improved. About 68 weeks after transplantation, renal function again deteriorated. Renal biopsy at 80 weeks demonstrated persistent nonspecific inflammation, and the patient was treated with pulse-dose corticosteroids for possible rejection. BK virus nephropathy was documented on renal biopsy 89 weeks after transplantation. Progressive deterioration in renal function led to the institution of hemodialysis. Immunosuppression was continued, despite the loss of renal function, to preserve pancreas transplant function. About 140 weeks after transplantation, the patient died of sepsis. A serum sample at that time showed persistent BK virus DNA at 440 copies/mL

Patient 3Patient 3, within the first month of and then again at 25 weeks after transplantation, was diagnosed with renal allograft rejection and was treated with pulse-dose corticosteroids, which led to an initial improvement in renal function. BK virus DNA was detected in 2 separate blood samples shortly after treatment of the first rejection episode, but no evidence of BK virus nephropathy was noted on renal biopsy specimens at that time. Rapid deterioration in renal function recurred ∼60 weeks after transplantation, and 3 separate renal biopsy specimens showed BK virus nephropathy. Progressive renal dysfunction ensued, and hemodialysis was begun ∼115 weeks after transplantation. Further reduction in immunosuppression was precluded because of concerns about rejection of the functioning pancreatic allograft. A transplant nephrectomy was done 168 weeks after transplantation, and a serum specimen 195 weeks after transplantation was negative for BK virus DNA

Patient 4Patient 4 had initially undergone renal transplantation 10 years earlier. Progressive allograft deterioration ensued, which led to a second kidney transplantation. About 5 weeks after transplantation, progressive renal dysfunction occurred, and a renal biopsy 10 weeks after transplantation showed BK virus nephropathy. BK virus DNA was detected in 2 separate serum samples obtained 6 and 5 weeks before the histopathologic diagnosis of BK virus nephropathy. Some improvement in renal function initially occurred. with reduction in immunosuppression, but significant, persistent renal dysfunction remained. A serum sample ∼84 weeks after transplantation showed no evidence of BK virus DNA

Discussion

In the present study, we demonstrated the potential clinical utility of a newly developed, highly sensitive, quantitative, real-time PCR method to detect BK virus DNA in serum of transplant patients with histopathologically documented BK virus nephropathy. In all 4 patients, BK virus DNA was detected in serum samples before the histologic diagnosis of BK nephropathy, and a reduction in virus load was demonstrated after reduction in immunosuppression and/or transplant nephrectomy. Furthermore, BK virus DNA was not detected in the serum samples of 16 matched transplant recipients without BK nephropathy who were similarly immunosuppressed, which demonstrates the specificity of the assay

Our findings are similar to those of Nickeleit et al. [4], who also found BK virus DNA in plasma samples of renal transplant recipients with BK virus nephropathy. Of interest, in both studies, prolonged viremia (in some cases >50 weeks) was found. In 3 of our patients (patients 1–3), persistent, unexplained, inflammatory changes were evident in renal biopsy specimens weeks to months before characteristic histopathologic changes consistent with BK virus nephropathy became evident. In all of these patients, BK virus DNA was detected in blood coincident with these nonspecific inflammatory changes, which suggests the possibility that BK virus–mediated inflammation was occurring in the kidney, despite the absence of characteristic histopathologic abnormalities attributable to BK virus [6]

This discordance between detection of BK virus DNA in blood and the absence of inclusions in renal biopsy specimens might be explained by sampling error due to patchy renal involvement or insensitivity of histologic evaluation, or the nonspecific inflammation could represent an antecedent to BK virus nephropathy from a variety of renal injuries, as has been suggested by Atencio et al. [16]. Additional studies to determine whether active BK virus infection was present within these areas of nonspecific inflammation in the kidney (e.g., in situ hybridization and immunocytochemistry) might be helpful in this regard

Treatment of BK virus nephropathy is problematic, since no antiviral agents have been demonstrated to be of benefit. Treatment currently is limited to the reduction of immunosuppression, which can be problematic, particularly in patients who have received multiple organ transplants [2–4, 7]. Response to therapy has been assessed by monitoring serum creatinine levels and by serial renal biopsy specimens. Nickeleit et al. [4] demonstrated that the reduction of immunosuppression and/or transplant nephrectomy led to the clearance of BK virus from the blood. Having a precise, quantitative method of determining response to therapy (i.e., reduction in immunosuppression) could be particularly important in patients who underwent combined kidney/pancreas transplantations, since excessive reduction in immunosuppression may trigger rejection in the functioning pancreatic allograft. In this regard, we demonstrated that a reduction in serum BK virus load followed reduction in immunosuppression and/or nephrectomy in 3 of 3 patients. In patient 2, immunosuppression could not be significantly reduced because of rejection of the pancreatic allograft, and progressive renal dysfunction leading to hemodialyis ultimately ensued. In this patient, low levels of BK virus DNA remained detectable in blood for >100 weeks, at which time the patient died of fungal sepsis

The present study had limitations that should be noted. The number of patients was small, and further validation with larger numbers of patients will be necessary to confirm our findings. Also, since this was a retrospective study, the timing of samples, while similar for case and control patients, was not necessarily identical. However, the high specificity of the assay reported in the present study is in accord with that reported elsewhere [4]

In summary, we have demonstrated the potential clinical utility of a highly sensitive, quantitative assay for BK virus DNA in blood for both diagnosing BK virus nephropathy and monitoring the patient’s response to therapy. Additional prospective studies with larger numbers of patients will be required to further define the appropriate role of such an assay in the diagnosis and treatment of patients at risk for BK virus nephropathy