Abstract
ABSTRACT. Cyclosporine (CsA) is the current primary immunosuppressant for the prevention of renal allograft rejection. Its chronic use is associated with various adverse effects like hypertension, hyperlipidemia, and nephrotoxicity, which in turn may contribute to chronic allograft nephropathy and cardiovascular mortality. This study compares a CsA-free maintenance regimen of mycophenolate mofetil (MMF) and corticosteroids with CsA and corticosteroids after early conversion from triple drug therapy. Eighty-four renal transplant recipients who had stable graft function on triple drug therapy with MMF, CsA, and steroids were randomly assigned to be withdrawn from either CsA (n = 44) or MMF (n = 40) at 3 mo posttransplantation. Kidney function at 1 yr was the primary endpoint. Secondary parameters of efficacy were patient and graft survival, incidence of acute rejection episodes, BP, and lipids. At study entry, the alternative treatment groups were similar with respect to demographics, renal function, dosage of CsA, BP, and concomitant medication. Both the creatinine clearance (71.7 versus 60.9 ml/min) and calculated GFR (73.2 versus 61.9 ml/min) were significantly better in MMF-treated patients at 1 yr. Conversion to MMF was associated with a decline of systolic and diastolic BP (128/76 versus 139/82 mmHg) and with a more favorable lipid profile. There was no difference in patient survival (100%) and graft survival (97.7% versus 100%). Acute rejection episodes occurred more frequently after withdrawal of CsA (11.3% versus 5.0%), but the difference was NS. Early tapering of CsA can safely be accomplished in renal transplant recipients who are stable on a triple drug regimen with MMF, thereby resulting in improved renal function, a more favorable lipid profile, and beneficial effects on posttransplant hypertension.
Cyclosporine (CsA) used in a dual or triple therapy regimen is the current primary immunosuppressant for the prevention of renal allograft rejection. Although the introduction of CsA into clinical practice has resulted in a 10 to 15% increase of the 1-yr graft function rate, little has been gained to improve long-term patient and graft survival (1,2). Minimizing the exposure to drug-related adverse side effects has the potential to improve morbidity and mortality. The practice of indefinitely using the same immunosuppressants as were being administered during the first few months after transplantation is to be called into question, and the reliance on long-term use of the calcineurin inhibitors needs to be reevaluated. CsA is associated with several adverse effects, such as hypertension, hyperlipidemia, and nephrotoxicity, factors that may contribute to cardiovascular mortality and chronic allograft dysfunction (3,4). Histologic changes of chronic allograft nephropathy consisting of obliterative vasculopathy and tubulointerstitial fibrosis may be difficult to distinguish from advanced CsA nephrotoxicity (5,6). Ongoing concerns about chronic CsA nephrotoxicity have prompted a large number of clinical trials in which attempts were made to switch to immunosuppression with azathioprine at a given time after transplantation. Several of those studies have reported improvements in renal function, lipid profile, and hypertension (7–9), but others have failed to confirm these findings (10,11), owing to the fact that the reduction in CsA toxicity was offset by the occurrence of acute rejection episodes (12). At least one randomized trial has demonstrated improved renal function parameters at 5 yr and a trend toward a reduced cardiovascular mortality in patients who had been withdrawn from CsA (13).
Mycophenolic acid, the active metabolite of mycophenolate mofetil (MMF), inhibits lymphocyte proliferation by blocking the de novo synthesis pathway of guanosine nucleotides (14,15). Three large, randomized, double-blind, multicenter parallel group trials have shown efficacy in the prevention of acute rejection episodes when MMF was given as part of a combination therapy with CsA (16–18). Following the drug’s mechanism of action, it was hoped that MMF would also prevent or delay progression of chronic allograft nephropathy, and preliminary clinical experience appears consistent with this hypothesis (19–21). Accumulating evidence from single-center studies suggests that CsA withdrawal under maintenance therapy with MMF results in a favorable graft function and a better control of hypertension and hyperlipidemia (22–25). However, controlled clinical data comparing MMF and CsA for maintenance therapy after kidney transplantation are not available to date. This study investigates the early conversion from a triple drug regimen of CsA, MMF, and steroids to either MMF or CsA with steroids at 3 mo after transplantation.
Materials and Methods
This open-label randomized controlled study was conducted in Groningen, Leiden, and Mannheim, Germany. The trial protocol conforms to the Declaration of Helsinki and its amendments. Formal approval from the institutional ethics committee was obtained at each participating site, and written informed consent was given before enrollment in the trial.
Patients
Eligible patients were renal transplant recipients >18 yr of age who had received a first or second graft and who were stable under triple drug therapy consisting of CsA, MMF, and corticosteroids. The serum creatinine concentration was <200 μmol/L at 3 mo after transplantation in all patients. We excluded sensitized patients (panel reactive antibodies >50%) and patients with more than one rejection episode posttransplant or acute rejection during the last month before study entry. Additional exclusion criteria were evidence of a systemic infection or a malignancy, severe gastrointestinal disorders interfering with the ability to absorb oral medication, a white cell count of <2.5 × 109/L or hemoglobin <6 g/dl, and treatment with investigational drugs or other prohibited medication, e.g., azathioprine, during the previous 4 wk. Women of childbearing age with a negative pregnancy test were required to use adequate contraception during and for 6 wk after the conclusion of treatment with MMF. None of the enrolled transplant recipients had received induction therapy.
Trial Design
Between April 1997 and June 2000, 84 renal transplant recipients were randomly assigned for withdrawal from either CsA or MMF. Patient enrollment was stratified for the transplant centers at 3 mo posttransplant. At this time, all patients were taking 1 g of MMF twice daily, using 500-mg tablets (Cellcept; Roche Pharmaceuticals, Mannheim, Germany), an oral microemulsion formulation of CsA (Neoral or Sandimmun Optoral; Novartis, Basel, Switzerland) administered in 25-, 50-, and 100-mg soft gelatin capsules dosed toward a target trough blood concentration of 150 to 250 ng/ml, and corticosteroids prescribed according to the center’s routine practice. CsA concentrations were measured at the investigational site by using a monoclonal technique (TDX [Abbott Labs, Pomezia, Italy]; CYCLOTrac SP [INCSTAR/Sorin, Dietzenbach, Germany]; or EMIT [Dade Behring, Deerfield, IL]).
Patients in group A were continued on the same dose of 1 g of MMF twice daily throughout the rest of the study period. CsA was reduced by 33% for the first 3 wk, lowered again by 33% of the initial pre-taper dose for another 3 wk, and then discontinued. In the event of severe drug-related leukopenia, MMF was temporarily interrupted and/or the maintenance dose was reduced to 1.5 g/d.
Patients in group B were tapered from MMF by using a decremental schedule of 500 mg every 2 wk, being off MMF after 6 wk. Further reduction of the CsA dose according to a desired trough concentration between 100 to 250 ng/ml was possible at the center’s discretion after steroids had been decreased to baseline again.
The oral steroid dose was temporarily increased to 25 mg/d prednisone in both study arms to cover the period of CsA and MMF taper, respectively, which was followed by a stepwise dose reduction of 5 mg every 4 wk. Baseline doses of 7.5 to 10 mg were achieved during 14 wk after randomization and continued during the entire study period.
Treatment of Rejection
When an acute rejection episode was suspected, a percutaneous graft biopsy was done. The renal core specimen was histologically graded by a local pathologist according to Banff classification (26). Patients experiencing acute rejection were treated with high-dose corticosteroids and/or antilymphocyte antibodies according to the local practice and, if the response was favorable, continued on study medication or were withdrawn from the study for safety considerations and converted back to triple drug therapy.
Efficacy Parameters and Safety Assessment
Kidney function was the primary efficacy parameter. Serum urea, creatinine, and uric acid concentrations were determined at study entry and at 6 and 12 mo. Endogenous creatinine clearance was calculated from 24-h urine collection. Urinary protein concentration was also measured. GFR was estimated by using the formula of Nankivell et al. (27) to check for the reliability of the clearance measurement, which is highly dependent on the completeness of the 24-h urine collection, especially in an outpatient setting.
Secondary efficacy parameters included the following: patient and graft survival, frequency and histologic grade of the first biopsy, confirmed acute rejection after study enrollment, and lipids and BP at 1 yr posttransplant. Body weight and vital signs, including pulse and BP taken in the supine and upright position, were recorded during each study visit, and concomitant medication was monitored. Use of lipid-lowering agents (statins) and the number of antihypertensive drugs prescribed at the physician’s discretion were also implemented in the evaluation for outcome analysis. Commonly used antihypertensive drugs were β-blockers, calcium entry blockers, diuretics, and angiotensin-converting enzyme inhibitors.
For the safety-analysis reporting of adverse effects, frequency and severity of infections and occurrence of malignancy was mandatory. Standard safety evaluation included physical examination, serial blood counts, and blood chemistry studies.
Statistical Analyses
The primary objective of this study was the evaluation of renal function at 1 yr posttransplant after CsA or MMF had been withdrawn in patients who had been initially treated with a triple therapy regimen. We calculated that 38 evaluable patients per group would result in an 80% power to detect a difference of creatinine clearance of at least 15% after CsA withdrawal, assuming that the common SD was 20% using a two-sided t test with a 5% two-sided significance level. Patient and graft survival and incidence of acute rejection episodes were also documented as a secondary endpoint and included in the analyses, even when occurring after study discontinuation (intention-to-treat). Numerical data were compared among groups by using the two-sided t test. Fisher’s exact test was applied to compare adverse events and other categorical variables, such as history of delayed graft function or a rejection episode before enrollment in the study. A two-tailed P < 0.05 was considered to be significant. Statistical analyses of the data were performed with Stata Statistical Software for Microsoft Windows (release 5.0; Stata Corp., College Station, TX).
Results
Patient Characteristics
Forty-four patients were randomized for conversion to MMF and 40 for continuation with CsA. The two treatment groups were well matched for gender, underlying disease, HLA compatibility, and certain aspects concerning the early postoperative course, including acute rejection episodes and occurrence of delayed graft function. However, recipients and donors were significantly younger in the MMF group. Patient characteristics at study entry are summarized in Table 1.
Demographic and baseline characteristicsa
Renal Function
There were no major differences in any of the renal function parameters of the two study groups at study entry (Table 2). At 1 yr, both serum creatinine and urea nitrogen were significantly lower in the MMF-treated group. The difference in the endogenous creatinine clearance was of 10.9 ml/min (95% CI, 2.4 to 19.2 ml/min). Using the Nankivell et al. (27) formula, a similar elevation in GFR in favor of the MMF treatment was found (11.3 ml/min [95% CI, 5.3 to 17.3 ml/min]). Furthermore, tapering of CsA was associated with consistently lower levels of uric acid. No differences were seen in the amount of proteinuria during the entire study period.
Renal function parameters at 3, 6, and 12 mo after transplantation
Blood Pressure
Both patient groups were similar regarding BP measurement or the number of prescribed antihypertensive drugs at time of randomization (Figure 1). Discontinuation of CsA was associated with a significant beneficial effect on the BP. At 12 mo, average differences of 11.2 mmHg (95% CI, 2.9 to 19.5) in systolic pressure and 6.0 mmHg (95% CI, 1.5 to 10.4) in diastolic pressure were found. BP remained unchanged in the CsA group, although the dose had been reduced during the study period, as permitted by protocol, with a subsequent decrease of the mean trough levels from 198 ng/ml at study entry to 153 ng/ml at 12 mo (P = 0.003).
Figure 1. BP and antihypertensive medication at study entry, 6 and 12 mo after transplantation. Data are given as mean ± SD. Line graph: ⧫, MMF; &U2591;, CsA. Bar graph: █, MMF; &U2591;, CsA.
Lipids
Comparing the groups at study entry, there was no difference in the number of patients who used lipid-lowering drugs (Figure 2). Despite the randomization procedure, the mean serum cholesterol level was significantly higher in patients assigned to continue with CsA (7.26 versus 6.52 mmol/L; P = 0.015). In both treatment arms of the study, mean cholesterol concentrations were decreased (10.8% [95% CI, 6.0 to 15.5] and 13.9% [95% CI, 8.6 to 19.6] from baseline in MMF- and CsA-treated patients, respectively). However, a significantly higher proportion of patients on CsA used HMG-CoA reductase inhibitors at 1 yr posttransplant (46.2% versus 21.9%; P = 0.033). Therefore, comparing the two treatment arms, the differences in serum cholesterol were NS (mean cholesterol, 6.23 versus 5.79 mmol/L; P = 0.087). Hypertriglyceridemia appeared to respond favorably to CsA withdrawal, and serum triglycerides were significantly better in MMF-treated patients at 12 mo, reaching almost the normal range. This was not the case in patients who were treated with CsA (mean triglycerides, 2.11 versus 3.23 mmol/L; P = 0.027). Average daily steroid dose was very similar between the groups at any time during the observation period. According to protocol, prednisone or prednisolone was administered at mean doses of 11, 15, and 9 mg at 3, 6, and 12 mo posttransplant, respectively.
Figure 2. Serum cholesterol, triglycerides, and lipid-lowering drugs at study entry (█)and 12 mo (&U2591;)after transplantation. Data are given as mean ± SD.
Acute Rejection Episodes, Study Discontinuation, and 12-mo Survival Data
Acute rejection episodes occurred more frequently after withdrawal from CsA compared with withdrawal from MMF (11.3% versus 5.0%). The difference was, however, NS (Table 3). The histologic grading by Banff classification (26) appeared to reveal more severe rejection episodes in patients with MMF, which could not be confirmed statistically due to the very limited number of any events observed. One patient from each group required anti-T cell globulin treatment for the reversal of rejection. In three instances, calcineurin inhibitors were reintroduced; one patient on CsA was switched back to MMF because a biopsy had demonstrated CsA nephrotoxicity. Two acute rejection episodes in the MMF treated group were diagnosed along with cytomegalovirus viremia. Another two patients experienced acute diarrhea after the ingestion of spoiled food. One patient in the MMF group was discontinued on day 22 for persistent leukopenia despite dose reduction. In the MMF treatment arm, the patient and graft survival at 12 mo were 100% and 97.7%, respectively. In the CsA arm, patient and graft survival were both 100% at 12 mo. One graft loss occurred by month 12 due to rejection that was associated with reactivated cytomegalovirus disease (Table 3).
Acute biopsy-proven rejection episodes and 12-mo survival dataa
Adverse Events
The number of patients experiencing opportunistic infections was similar in both groups. Bacterial infections involving the urinary tract were reported more frequently with CsA, although these differences were NS (P = 0.10). New-onset diabetes during the phase of steroid coverage was seen in both treatment arms and ranged from 9 to 15%. Table 4 provides a survey of spontaneously reported adverse events exceeding an incidence of 5% in at least one of the study groups.
Principal spontaneously reported adverse events with an incidence of at least 5% in one group from time of randomization until study completion at 12 moa
Discussion
Population-based studies provide evidence that chronic allograft dysfunction results from both immunologic and nonimmunologic events (28–31). Medical management of the transplant recipient should take into consideration the time-dependency of the various risks affecting patient and graft (32), which may lead to a modification in the ranking of the priorities: maintain the current low level of acute rejection during the early months after transplantation and get superior graft function with reduced patient risk during long-term follow-up. Short-term survival benefits of the novel immunosuppressants cannot be measured anymore in improved short-term graft survival; therefore, it becomes relevant to use surrogate endpoints in which the incidence of acute rejection is viewed together with secondary endpoints. Among these, graft function, metabolic abnormalities, and drug-related side effects have emerged as additional benchmarks for the assessment of the newer therapies. Renal function at 1 yr was the primary endpoint of the present study, and it was significantly better in MMF-treated patients compared with the recipients who remained on CsA treatment. The mean difference in creatinine clearance between the two study arms (MMF versus CsA) was 17.7%. This beneficial effect on renal function may be somewhat tempered by the fact that the CsA-treated patients, both donor and recipient, were older than the MMF-treated patients (Table 1); albeit the strength of the age difference is unlikely to have affected the overall results of the present study. There is substantial evidence that the serum creatinine concentration after the first year as well as graft histology correlate with the ultimate graft prognosis (33–36). It should be noted that the improvements of the renal function parameters after CsA discontinuation do not necessarily prove the presence of drug-related structural nephrotoxicity because protocol biopsies were not done in our study. Functional effects on renal hemodynamics (37,38) may not be the only consequences of CsA in the kidney. CsA mediates its immunosuppressive capacity not only through the inhibition of calcineurin phosphatase, but also through the expression of cytokines like transforming growth factor–β (TGF-β), which enhances the renal scarring process (39,40). Irreversible morphologic lesions of chronic CsA toxicity typically present with afferent arteriolopathy, tubulointerstitial fibrosis, and tubular atrophy (5). These lesions appear to be less dose-dependent but attributable to an elevated individual susceptibility, which is consistent with the observation of genetic differences in the quantitative liberation of TGF-β after the administration of CsA (41,42). Thus, eliminating the drug exposure in the early posttransplantation period is likely to favorably affect the long-term graft prognosis. The severity of posttransplant hypertension is significantly correlated with the incidence of chronic allograft failure (29,30), albeit that the interrelation between cause and effect may be less clear (43). This study indicates that switching to CsA-free immunosuppression with MMF is associated with a substantial decline in both systolic and diastolic BP through the entire observation period. The magnitude of the BP lowering effect is in accordance with previous experience (23,25,44). Both better renal function and better control of BP argue against the development of unrecognized chronic rejection in the MMF-treated group. Along with preclinical data showing that MMF reduces vascular smooth muscle proliferation in vitro (15), it could contribute to a reduction of late graft deterioration. Indeed, recent data from the US renal transplant scientific registry indicate that MMF therapy decreases the risk of developing chronic allograft nephropathy independent of acute rejection (45). It should be noted that no change in BP was seen in CsA-treated patients even though the dosage was lowered in accordance with the protocol, which was followed by a 25% reduction of the mean trough levels.
Hyperlipidemia is associated with the development of atherosclerosis and is frequently found in renal transplant recipients. Abnormalities in lipid metabolism are thought to add to the posttransplant cardiovascular mortality (3,46). Pretransplant lipoprotein elevations were shown to correlate with a higher graft damage score in renal biopsies taken later (47). This study provides evidence that MMF in combination with steroids for maintenance immunosuppression is associated with a more favorable lipid profile as compared with a CsA-based regimen. By chance, serum cholesterol concentrations at study entry were slightly higher in patients assigned to maintain on CsA. Prescription of 3-hydroxy-3-methylglutaryl-coenzyme-A (HMG-CoA) reductase inhibitors was permitted by protocol according to the local practice; therefore, serum cholesterol decreased significantly within both study groups. However, more CsA-treated patients used cholesterol-lowering drugs at 1 yr after transplantation (46.2% versus 21.9%; P = 0.033). Plasma triglycerides were markedly decreased in the MMF-treated group after withdrawal from CsA, which was not the case in the patients who remained on CsA. The drop in fasting triglycerides was similar as previously reported (23,25) and may be of importance with respect to long-term graft survival. Hypertriglyceridemia has been identified as an independent risk indicator of chronic allograft nephropathy, although the pathogenetic mechanism regarding its molecular basis is still poorly understood (48,49).
The relative risk for acute rejection after CsA reduction has been examined in a number of studies and attributed to a variety of factors (7–11). Previous clinical trials in which CsA-based regimens were converted to azathioprine for maintenance therapy have reported acute rejection episodes occurring in 10 to 40% of cases. A meta-analysis of these studies concluded that, despite an increased incidence of acute rejection episodes after CsA withdrawal, short-term patient and graft survival were not affected. Rejection rate was independent of variables such as indication, speed, or timing of withdrawal (50). However, it remains unclear whether CsA discontinuation outweighs the long-term consequences of a relatively high rate of acute rejection episodes, which is a risk factor for the development of chronic allograft nephropathy. A single-center trial from the Netherlands (44) has investigated the conversion from CsA to either MMF or azathioprine at 1 yr posttransplant, and significantly less acute rejection episodes were observed in the MMF group (4 of 34) compared with the azathioprine group (11 of 30). A more recent multicenter trial has reported a higher incidence of acute rejection episodes occurring in 14 of 63 patients when CsA was withdrawn from triple drug therapy with MMF and steroids at 6 mo after transplantation. In contrast with our protocol, steroid coverage during CsA taper was limited to a dose of 0.15 mg/kg prednisone per day, and a reduced kidney function at study entry was not an exclusion criteria (51). The present study provides evidence that tapering of CsA can safely be accomplished even at 3 mo posttransplant in renal transplant recipients who are stable on a triple drug regimen with MMF. Acute rejection attributable to CsA withdrawal occurred in 11.3% of cases in the MMF group when doing the analysis on an intention-to-treat basis. In particular, two patients experienced an acute rejection episode after CsA had been discontinued due to malabsorption after severe acute diarrhea, which was not drug related. Another two patients presented with acute rejection in conjunction with reactivated cytomegalovirus disease.
In summary, MMF in combination with CsA and steroids has emerged as a standard immunosuppressive regimen that prevents acute rejection in the early phase after kidney transplantation in most patients. The clinician facing up to the long-term risks of immunosuppression wants to know how to balance toxicities against rejection after a given time-period. This controlled randomized trial provides evidence that either CsA or MMF can safely be tapered from triple drug therapy in stable renal transplant recipients 3 mo after transplantation. A low but persisting risk of acute rejection episodes was observed in both treatment arms, which was a little higher after the discontinuation of CsA, although it did not affect short-term patient and graft survival. It should be noted that the presence of rejection is not a completely benign event, and 1-yr graft survival is not necessarily a surrogate for long-term allograft survival. Rejection episodes occurring more frequently after CsA withdrawal may be important if they translate into impaired graft function in the long-term. On the other hand, patients who were successfully converted to MMF presented with better renal function, better control of hypertension, and beneficial effects on lipid abnormalities at 1 yr after transplantation. Another important feature is that the discontinuation of CsA early after transplantation will eliminate its effects on chronic nephropathy. It remains to be seen if all these salutary effects on the cardiovascular risk profile will ultimately result in improved patient and graft survival.
- © 2002 American Society of Nephrology