|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Renal Transplantation |

* Department of Medicine, Division of Nephrology, and
Department of Pathology, University of Washington, Seattle, Washington
Correspondence: Dr. Nicolae Leca, University of Washington Medical Center, 1959 NE Pacific Street, Box 356521, Seattle, WA 98195. Phone: 206-543-0690; Fax: 206-685-8661; E-mail: nleca{at}u.washington.edu
| Abstract |
|---|
|
|
|---|
Design, setting, participants, & measurements: Leflunomide was used in 21 patients with BK nephropathy. These patients were divided into two groups on the basis of the leflunomide levels achieved: Low-level group (<40 µg/ml) and high-level group (>40 µg/ml).
Results: During 13 mo of follow-up, there was no difference in the rate of serum BK viral clearance between the groups. There were three graft losses in the low-level group and one in the high-level group; however, creatinine levels were higher at the time of starting leflunomide in the low-level group. Leflunomide was also used in six patients with chronic allograft injury. No graft loss was observed during the follow-up period of 16 mo. Treatment with leflunomide seemed to be associated with a new toxicity, hemolysis, seen in four of the 27 patients so treated. Patients with hemolysis had high leflunomide levels (81.4 ± 14 µg/ml) and worsening allograft function. Two patients had histologic evidence of thrombotic microangiopathy, which led to graft loss in one patient.
Conclusions: The clinical correlation between leflunomide levels and outcomes needs to be further refined. This study described a possible association of leflunomide with thrombotic microangiopathy, especially at higher levels.
| Introduction |
|---|
|
|
|---|
The immunosuppressive effects of leflunomide are exerted through multiple mechanisms, and the drug has been used in experimental and clinical applications to prevent or treat rejection (3–6). Leflunomide blocks lymphocytic proliferation by reversible blockade of the mitochondrial dihydro-orotate dehydrogenase, an enzyme required for de novo pyrimidine synthesis (7–10). It also alters intracellular signaling of the T and B lymphocytes by interference with the function of multiple protein kinases, NF-
B, and other factors (5,11–16).
The mechanisms accounting for the antiviral effects of leflunomide have not been well delineated, and they seem to be independent of the effects on dihydro-orotate dehydrogenase. In cytomegalovirus (CMV) infection, leflunomide inhibits the viral tegumentation, but in BK infection, this does not apply, because the BK virus is not enveloped (17,18).
The targets of leflunomide dosing and/or level are not well established for clinical practice for either BK nephropathy (BKN) or rejection. The usual prescribed dosage for rheumatoid arthritis is 10 to 20 mg/d. Leflunomide is metabolized to its active ingredient, A77,1726 (teriflunomide), a malonitrilamide that is measured for the reported leflunomide level. The leflunomide dosage for the antiviral effect in renal transplantation has been inferred from its in vitro IC50 against viral replication. The current clinical practice is to target leflunomide levels in excess of 40 µg/ml. Published data suggest a possible worse outcome for patients who do not achieve such levels during the BKN treatment (19,20). For achieving levels higher than 40 µg/ml, leflunomide dosages that substantially exceed those used for rheumatic disorders are required for most patients. For rheumatoid arthritis, leflunomide level monitoring is not recommended and is not routinely performed.
The reported adverse effect profile of leflunomide includes predominantly diarrhea and rash, with potential for severe reactions including hepatotoxicity, pneumonitis, neurotoxicity, and bone marrow suppression in rare cases (21–28). No hemolysis-associated syndromes have been reported to date in humans in relation to the use of leflunomide.
In this article, we review our experience with the use of leflunomide, define the adverse effect profile that we have observed, and compare the outcomes of patients who had BKN and achieved levels higher or lower than 40 µg/ml. We describe for the first time a concerning association between the treatment with leflunomide and thrombotic microangiopathy (TMA).
| Materials and Methods |
|---|
|
|
|---|
|
The diagnosis of BKN was made on the basis of the histologic evidence of characteristic viral inclusions in tubular cell nuclei, confirmed by staining for SV40. BK viral DNA was detected by a PCR assay that amplifies a region of the large T antigen gene (29).
The severity of tubulointerstitial inflammation and fibrosis in BKN was graded semiquantitatively by two independent observers on a 0- to 3-point scale: 0 points, <10% tissue core area; 1 point, 10 to 25%; 2 points, 25 to 50%; and 3 points, >50%. The amount of immunohistochemical staining for SV40 was quantified by counting the number of positive epithelial cell nuclei in the biopsy and reported as number of positive cells per square millimeter of tissue.
Patients who received leflunomide for BKN were divided into two groups on the basis of their highest leflunomide levels during the entire duration of the treatment. The cutoff value for inclusion in the higher level group was 40 µg/ml. Clinical characteristics specifically determined for each patient included age, ethnicity, gender, cause of ESRD, retransplant status, type of transplant, HLA matching, CMV donor/recipient status, history of delayed graft function, creatinine levels, immunosuppression protocol, and history of rejection. The data collected at the time of BKN diagnosis included time after transplantation, serum creatinine, and BK viral load determined by PCR (29). The follow-up data included length of follow-up, treatment, serum creatinine, complete blood count values and peripheral smear, LDH, haptoglobin, CMV, graft and patient outcome, leflunomide levels, leflunomide dosage, and any documented adverse reaction during the treatment. Leflunomide assays were performed using HPLC/mass spectrometry (National Medical Services, Willow Grove, PA).
Treatment
The leflunomide dosage and target levels were individually determined for each patient by their transplant nephrologist. For both BKN and chronic allograft injury (CAI), leflunomide replaced MMF on the immunosuppressive regimen. Starting dosage for leflunomide included a loading dose (60 mg/d for 3 d). Maintenance was started with a dosage of 20 mg/d, which was then titrated on the basis of the levels. Leflunomide levels were checked monthly. Maintenance dosage did not exceed 60 mg/d. The targeted FK level was lowered to 5 ng/ml. The prednisone dosage was unchanged (5 to 10 mg/d).
In addition to leflunomide, some patients received other adjuvant therapies for BKN. These were initiated immediately upon histologic diagnosis of BKN and before starting leflunomide and included three to four infusions of cidofovir 0.35 mg/kg (n = 8) or low-sucrose IVIg 2 g/kg (n = 3).
Statistical Analyses
Categorical clinical characteristics and outcomes were compared using Fisher exact test. Continuous variables were compared using the t test (for normally distributed data) or the Wilcoxon rank-sum test (for skewed data). A significance level of P < 0.05 was used for all tests.
A Kaplan-Meier plot was used to summarize time to clearance of BK from serum, comparing the two groups of patients. Patients were considered at risk from the time starting leflunomide until serum BK virus levels were undetectable or until study end, with patients who experienced graft loss before viremia clearance considered to be at risk until study end. The time to the clearance of BK virus in the serum was compared by the attained leflunomide level using the log-rank test. Statistical comparisons were conducted using Stata statistical software (Stata 9; Stata Corp, College Station, TX).
| Results |
|---|
|
|
|---|
|
|
|
No CMV reactivation or disease was observed in any group before or after the diagnosis of BKN. Two patients in the high-level group had episodes of mild rejection with negative staining for SV40 at the time of clearing the BK virus in the serum. No rejection episodes were observed in the low-level group.
Leflunomide Use for CAI
Leflunomide was used in six renal transplant recipients with CAI changes represented by moderate to severe tubular atrophy and interstitial fibrosis in association with persisting interstitial inflammation. All six patients had preceding episodes of rejection; one patient had persistent antibody-mediated rejection treated with multiple cycles of plasmapheresis, IVIg, and rituximab. All six patients were receiving steroids, and leflunomide was used in combination with FK (mean level 6.4 ng/ml; range 5 to 9) for five patients and cyclosporine (level 150 ng/ml) for one patient. BK virus was negative in all patients by urine PCR.
The dosage of leflunomide was on average 29.2 ± 10.2 mg/d, and the average leflunomide level was 51.6 ± 28.9 µg/ml. The treatment with leflunomide was started on average 53 ± 45 mo after transplantation (range 21 to 144), and the duration of follow-up was on average 16 ± 10 mo (range 6 to 31).
Creatinine levels were statistically higher at the time of starting leflunomide (2.05 ± 0.68 mg/dl) when compared with the nadir level after transplantation (1.4 ± 0.49 mg/dl; P < 0.05). At the end of the follow-up period, the creatinine levels were higher (2.43 ± 0.74 mg/dl) but did not reach statistical significance when compared with the time of initiation of leflunomide therapy. No acute rejection episodes and no graft losses were observed during the duration of therapy.
Adverse Reactions Observed during Leflunomide Therapy
Adverse reactions were observed in nine of the 27 patients during the treatment with leflunomide. These included rash (n = 1), neuropathic-type pain (n = 1), mild elevation of transaminases (n = 1), pancytopenia (n = 1), and alopecia and malaise (n = 1). Hemolysis was present in four patients; two of them were receiving leflunomide for BKN and the other two for CAI. All patients with hemolysis had worsening allograft function and leflunomide levels >40 µg/ml (average 81.4 ± 14) at the time of hemolysis diagnosis. The incidence of hemolysis was 14.8% overall for the 27 patients who were treated with leflunomide and 28.6% for the 14 patients with leflunomide levels >40 µg/ml. Hemolysis was diagnosed on the basis of the presence of undetectable haptoglobin (<31 mg/dl), elevated LDH (718.7 ± 508 U/L), decreasing platelet numbers (83.7 ± 25 x 103/ml), anemia (hematocrit 29 ± 1.6%), and the finding of erythrocyte fragments in the peripheral smear for all four patients. Mean FK levels in the four patients with hemolysis was 6 ng/ml (range 5 to 7 ng/ml). Coombs test was negative. Bruising was the most frequent clinical symptom; however, no bleeding or thrombotic episodes were observed. Liver function test abnormalities did not occur. Three patients underwent a renal transplant biopsy at the time of diagnosis of the hemolysis. TMA lesions were observed in two of these patients.
Hemolysis was diagnosed in all four patients within the first 6 mo of leflunomide therapy. Specifically, these patients had no evidence of a positive C4d stain on biopsy. There was no evidence of other concurrent infections (viral or bacterial) at the time of hemolysis diagnosis. Leflunomide was stopped for two patients and reduced for the other two. Graft loss occurred in one patient despite discontinuation of the drug. One other patient had persistent renal dysfunction despite leflunomide discontinuation. Renal function improved in the other two patients upon reduction of leflunomide dosage (from 1.7 and 2.6 to 1.2 and 2 mg/dl, respectively). Laboratory abnormalities consistent with the diagnosis of hemolysis resolved upon discontinuation or reduction of the drug in all cases. The cause of renal disease of patients with hemolysis was polycystic renal disease (n = 2), type 1 diabetes (n = 1), and IgA nephropathy (n = 1). None of the patients had viral hepatitis. There was no history of recurrent thrombosis, anti-phospholipid antibody syndrome, or lupus. BP was well controlled in all patients.
| Discussion |
|---|
|
|
|---|
The beneficial effects of leflunomide in BKN are thought to be delivered by the antiviral as well as anti-inflammatory properties of the drug. To date, no clear correlation has been made between these effects and the in vivo leflunomide levels in humans. A target leflunomide level of 40 µg/ml has been suggested by extrapolating the in vitro effective concentrations against the CMV virus (17); however, it is possible that a lower level may be sufficient in vivo for the antiviral effect. In addition, the immunomodulatory effects of leflunomide, which may be particularly important in BKN, may be seen at drug exposures <40 µg/ml.
Difficulties with the clinical use of leflunomide are represented by the long serum half-life (15 d) of teriflunomide and the lengthy turnaround time of laboratory reporting. These characteristics and the high pharmacokinetic intersubject variability explain the wide differences observed in the leflunomide levels. The correlation between leflunomide dosage and levels was only moderate in our patients (r = 0.45 by Spearman rank-sum analysis).
We believe that leflunomide remains a potential effective therapy for BKN, but its particular properties make difficult the assignment of a "therapeutic" level and may explain the differences between our BKN outcomes and other studies (19,20). Leflunomide use for CAI has not been associated with an increased risk for rejection; however, the outcomes of therapy are difficult to interpret because of a small number of patients treated with multiple confounding factors.
The adverse effects that we observed during the treatment with leflunomide are similar to what has been reported previously in the literature. At higher levels, however, we observed a new toxicity associated with the leflunomide therapy: Hemolysis accompanied by evidence of TMA. Hemolysis can be clinically silent, and making the diagnosis requires a high index of suspicion. It is interesting that anemia has been consistently reported in the literature in association with the use of leflunomide, but further analysis into the cause of anemia is rare. Our data suggest that hemolysis may be playing an undiagnosed role. To date, the only reported association of leflunomide with hemolysis was found in a canine transplant model, particularly in dogs that were administered a higher dosage of leflunomide (30). We recommend obtaining a hemolysis panel (peripheral smear for red blood cell fragments, LDH, and haptoglobin) for patients who exhibit anemia and/or thrombocytopenia during the leflunomide treatment. A renal biopsy may be indicated to rule out TMA in patients who exhibit worsening renal function.
Although it is premature to draw the conclusion that leflunomide causes hemolysis and TMA directly, this drug has proven antiangiogenic properties mediated by inhibition of tyrosine kinases, as seen with its effect on the PDGF receptor (31). Leflunomide levels of 40 µg/ml also exceed the in vitro equivalent concentration required for tyrosine kinase inhibition (11–14,31). In addition, leflunomide has been shown to inhibit endothelial cell proliferation and organization of endothelial cells into capillary-like networks, suggesting a potential direct effect of leflunomide on endothelial cells (32).
In general, the mechanisms of de novo TMA after renal transplantation are not completely understood but are related most commonly to immunosuppressive agents such as cyclosporine, tacrolimus, and sirolimus, particularly when used in combination or in high dosages (33,34). TMA has also been described in association with viral infections, including CMV and BK (35–37). It is possible that the association of leflunomide with TMA may stem from a possible combined toxicity with the use of a calcineurin inhibitor, particularly in patients with a background of existing renal vasculopathy, as seen in both BKN and CAI. In our report, the FK levels of patients with hemolysis were relatively low and not different from patients without hemolysis.
Our study is subject to a number of limitations because of its observational design. First, the results could be confounded by differences in clinical characteristics between the groups; however, no features favoring the group that achieved levels <40 µg/ml were observed. Second, there may be unmeasured characteristics of the patients (e.g., differences in drug metabolism) or the impact of the adjuvant BKN treatments that may affect clinical outcomes. Third, no set protocol was used for treatment, follow-up, or evaluation of adverse effects, and it cannot be excluded that some outcomes were ascertained differently.
| Conclusions |
|---|
|
|
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
See related editorial, "Leflunomide Therapy in Kidney Transplantation: Ready for Prime Time?" on pages 652–653.
Received September 19, 2007. Accepted February 13, 2008.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
J. K Wu and M. T Harris Use of Leflunomide in the Treatment of Polyomavirus BK-Associated Nephropathy Ann. Pharmacother., November 1, 2008; 42(11): 1679 - 1685. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. B. Mannon Leflunomide Therapy in Kidney Transplantation: Ready for Prime Time? Clin. J. Am. Soc. Nephrol., May 1, 2008; 3(3): 652 - 653. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |