Chronic rejection, or more accurately, chronic allograft nephropathy (CAN), is a major impediment to long-term renal allograft survival and an important cause of graft loss in the first posttransplant decade (1, 2). CAN is diagnosed by a progressive deterioration in renal allograft function, reflected by increasing levels of serum creatinine (SCr) (“creeping creatinine”), with characteristic but nonspecific histopathology (3). Both immunologic and nonimmunologic factors contribute to the development of CAN, including the adverse effects of chronic cyclosporine A (CsA) use, such as nephrotoxicity, hypertension, and hyperlipidemia (2,4). Biopsy studies of the natural history of CAN demonstrate that by 5 years posttransplant, 93.5% of patients have evidence of CsA nephrotoxicity and 65.9% have evidence of CAN Banff Grade II or III (5). No treatment has been proven to be effective in patients with CAN, although a number of small, nonrandomized, uncontrolled studies have suggested a benefit of CsA withdrawal or dose reduction in the presence of mycophenolate mofetil (MMF) for patients with deteriorating renal function (6–10).
The goal of the present study was to evaluate, in a prospective, controlled manner, the efficacy and safety of MMF as a potential replacement for CsA in patients with chronic progressive deterioration in graft function (“creeping creatinine”) caused by CAN.
Study Design and Patients
This was an open, randomized, controlled, multicenter, international study conducted at 24 centers between November 1998 and April 2002. The study was performed in accordance with the Declaration of Helsinki, and all patients gave written informed consent before study entry.
Patients 18 to 65 years old, at least 6 months posttransplant, and maintained on a CsA-based regimen with an SCr in the range of 100 to 400 μmol/L and a calculated creatinine clearance (CCr) (11) of more than 20 mL/min were identified. Eligible patients were required to have a documented deterioration of transplant function, as evidenced by a negative slope of the reciprocal of SCr plotted against time. Measurements of SCr were taken over at least the previous 3 months but not more than 18 months before study entry. A regression analysis, performed using at least the last six 1/SCr values, had to show a negative slope with a significance value of P less than 0.05 and an adjusted r2 of more than 0.35 (3). Representative plots for eligible and ineligible patients are shown in Figures 1A and B, respectively. Eligible patients were required to have had a renal allograft biopsy performed to investigate deteriorating renal function within 12 months before study entry. Patients with de novo or recurrent renal disease, transplant glomerulopathy, or acute rejection were to have been excluded from the study, as were patients in whom other causes of graft dysfunction were identified (e.g., obstruction, renal artery stenosis). Patients were also excluded if they had experienced a biopsy-proven acute rejection within 3 months before study entry or were taking MMF, sirolimus, or tacrolimus before recruitment. Additional exclusion criteria included pregnancy, history of gastrointestinal disorder, active infection, malignancy (except adequately treated nonmetastatic basal or squamous cell carcinoma of the skin), participation in another study, white blood cell count less than 2.5×109/L or hemoglobin less than 5 g/dL, and the use of bile acid sequestrants. Patients receiving CsA monotherapy had to be at a stable dose of CsA for the past 2 to 3 months.
Randomization and Treatment
Patients were randomized, after successful completion of the screening phase, by a computerized touch-tone system administered by S-Clinica, Brussels, Belgium. Deterministic balanced (1:1) minimization, stratified for center, was used for treatment allocation. The study design is shown in Figure 1C. Total treatment duration was 58 weeks. Patients randomized to group A had MMF introduced incrementally to a dose of 2 g/day over 4 weeks, and steroids were introduced if not already prescribed. Azathioprine (AZA) was discontinued on the addition of MMF when previously used. The CsA dose was then reduced in three consecutive steps over a 6-week period. At the end of this 10-week period (phase 1, Fig. 1C), patients randomized to group A were maintained on MMF and corticosteroids alone, the latter at a dose of not less than 10 mg/day. Patients randomized to group B (control arm) remained on their immunosuppressive regimen per normal center practice (CsA monotherapy, CsA/steroids, or CsA/AZA/steroids). CsA dose reduction was permitted, but patients were required to maintain a 12-hr CsA trough of not less than 80 ng/mL. Groups A and B were then monitored for 6 months (phase 2, Fig. 1C) for the primary endpoint. Follow-up continued for a further 6 months (phase 3, Fig. 1C) for evaluation of secondary endpoints. After the first 12 weeks of the study, group A patients were permitted to taper (but not discontinue) their corticosteroid dose over 4 weeks, per center practice. Patients in group A had to remain on MMF at a minimum of 2 g/day unless an adverse event warranted dose reduction. On resolution of an adverse event, the investigator was required to attempt to reinstate the 2 g/day dose. MMF could also be increased at the discretion of the investigator to 3 g/day.
Primary and Secondary Endpoints
The primary endpoint was the change in renal function over the 6 months up to the end of phase 2. A “responder” was defined as a subject experiencing a clinically significant improvement in renal function, as reflected by a flat or positive slope of the 1/SCr plot, with no change to the randomized therapy and no graft loss. Regression analysis was also used to compare the 1/SCr plots between the two treatment groups during screening and phase 2. Secondary endpoints included graft and patient survival, acute rejection incidence, calculated CCr, blood pressure, antihypertensive and lipid-lowering medication use, laboratory parameters (urea, cholesterol, triglycerides, hemoglobin, white blood cell count, platelets), responder and nonresponder status, and the slopes of the regression line plot (1/SCr) between the two groups at 1 year. Baseline and study visit assessments included CsA blood level and doses (baseline and until week 10 only in group A), rejection episodes and treatment, adjustment in MMF and/or steroid doses, adverse events, and opportunistic infections.
The sample size of 184 evaluable patients was based on expected response rates of 40% and 20% in the two treatment groups, with 80% power for a two-sided test using an alpha of 0.045 (adjusted to allow two interim analyses using alphas of 0.0005 and 0.0140). The strict recruitment criteria made it difficult to reach the planned sample size. The actual sample size (n=143) resulted in a power of approximately 75%.
The response rates in group A and group B were compared using a chi-square test (with continuity correction). Three populations were studied:
- The primary intent-to-treat (ITT) population was composed of all patients randomized into the study who had taken at least one dose of study medication and had a minimum of three values for SCr during phase 2 (n=122), following the specifications in the protocol.
- The full ITT population was composed of all patients randomized into the study who had taken at least one dose of study medication (n=143). For the purposes of determining response rate in this population, all patients who had less than three values for SCr during phase 2, or who withdrew from the study for any reason before the end of phase 2, were considered as nonresponders.
- The per-protocol population was composed of all patients who did not have major protocol violations (n=107). These violations were noncompliance (n=7), violation of inclusion/exclusion criteria (n=5), or withdrawal before the end of phase 2 (i.e., 6-month endpoint; n=24).
Laboratory parameters and blood pressure (secondary endpoints) were analyzed for treatment differences by analysis of covariance, fitting the terms of treatment and the baseline value. Because of two planned interim analyses, the overall P value of the study was set at 0.045 for a two-sided test.
Role of the Funding Source
The study was designed by the investigators, supported by representatives from Roche Pharmaceuticals. Data were collected and analyzed by medical and statistical representatives from Roche Pharmaceuticals in conjunction with the investigators. All participating institutions received grant support for the conduct of the study.
Patients and Renal Histology
Figure 1D is a flow chart showing patient disposition during the study. Demographic characteristics of the full ITT population (n=143; group A, n=73; group B, n=70) are shown in Table 1. Baseline biopsies were reported as CAN in 78% and 77% of patients in groups A and B, respectively, and as CsA toxicity in 16% and 11% of patients in groups A and B, respectively. No patient in either group showed evidence of acute rejection on biopsy. In group B, one patient demonstrated de novo or recurrent disease, and one patient demonstrated transplant glomerulopathy; both patients were withdrawn as inappropriate enrollments during phase 1 but are included in the full ITT analysis. The groups were well matched for baseline SCr and calculated CCr (Table 1).
Immunosuppression in the full ITT population is shown in Table 2. There was a trend toward reduction in CsA dose and trough levels over the course of the study in group B. Forty-eight patients in group B were taking AZA, which was discontinued in six patients at the 6-month endpoint. No patient started AZA in group B during the study. At the 6-month endpoint, two patients in group A were receiving 3 g/day of MMF, 42 were receiving 2 g/day, and 16 were receiving less than 2 g/day. At 12 months, two patients were receiving 3 g/day, 38 patients were receiving 2 g/day, and 14 patients were receiving less than 2 g/day.
Primary Endpoint: Response Rate
The response rate was determined for the three populations as described previously. For all three populations, the proportion of responders was significantly greater in group A versus group B at the 6-month endpoint (primary ITT population, 36/62 or 58% vs. 19/60 or 32%, P=0.0060; full ITT population, 36/73 or 49% vs. 18/70 or 26%, P=0.0062; per-protocol population, 36/60 or 60% vs. 12/47 or 26%, P=0.0008)(Fig. 2A).
Renal function (as assessed by SCr, calculated CCr, and serum urea) improved significantly in group A but worsened in group B (Table 3). These differences were significant at the 6- and 12-month endpoints in the full ITT population (SCr, P=0.0003, CCr, P=0.0066 at 6 months; SCr, P=0.0024, CCr, P=0.0044 at 12 months) (Table 3). Similar results were observed in the per-protocol population (results not shown). Figures 2B and C show least squares mean plots for SCr and calculated CCr over the course of the study in the full ITT population.
Figure 2D shows a regression line analysis of 1/SCr for patients in the full ITT population over the course of the study. Patients in groups A and B showed a similar negative slope of 1/SCr during the screening period (Fig. 2D). During phase 2, renal function stabilized or improved in group A but continued to deteriorate in group B, and these differences persisted to the end of phase 3 (Fig. 2D) (P=0.0037).
Response Rate at 12 Months
According to the categoric analysis at the 12-month endpoint, in the primary ITT population (n=122), 48% (30/62) were responders and 52% (32/62) were nonresponders in group A versus 35% (21/60) and 65% (39/60), respectively, in group B (P=0.1885).
Changes in Laboratory Parameters and Blood Pressure
Changes from baseline in relevant laboratory parameters are shown in Table 3 for the full ITT population. There was a statistically significant difference in total cholesterol between the groups at both the 6- and 12-month endpoints, with greater decreases in group A (P=0.0330 and P=0.0077, respectively). Although the decrease in triglycerides was greater in group A at 12 months, there were no statistically significant differences between the groups in this parameter at either endpoint. There were also no statistically significant differences in the use of lipid-lowering medications between the groups at baseline, 6 months, or 12 months.
Changes in hematologic parameters are shown in Table 3. There was a transient decrease in hemoglobin levels in group A that began shortly after patient entry into phase 1 of the study. This decline reached a maximum of 9.2 g/L (least squares mean) and resolved by the 6-month endpoint, remaining stable thereafter. Decreases in systolic and diastolic blood pressure occurred in both groups, with a trend toward greater decreases in group A (Table 3); however, these differences did not reach statistical significance at either endpoint. Antihypertensive use was not different between the groups at baseline, 6 months, or 12 months. Similar results for metabolic and hematologic parameters were observed in the per-protocol population.
Adverse Events, Patient and Graft Survival, and Acute Rejection
Adverse events are listed in Table 4. The overall incidence of adverse events was 85% in group A and 67% in group B. Patient survival in the full ITT population was 95.9% and 100% in groups A and B, respectively. The three deaths in the study occurred in group A. One death, from infective ascites associated with chronic hepatitis B-associated liver disease, occurred on day 108. A second death occurred in a patient with polycystic kidney disease at day 392 secondary to septic shock complicated by acute respiratory distress syndrome, pseudomembranous colitis, and myocardial infarction. A third death, from pneumococcal pneumonia, occurred on day 279. There were six grafts lost, two in group A and four in group B, all occurring after the 6-month primary endpoint. Graft survival (including death with a functioning graft) was thus 93.2% in group A and 94.3% in group B at the 12-month endpoint. There were no acute rejections in either group.
The present study is the first randomized, controlled trial to demonstrate that the withdrawal of CsA in the presence of MMF in patients with a progressive deterioration in renal allograft function (“creeping creatinine”) secondary to CAN results in a significant improvement in renal function. Inspection of the regression lines (Fig. 2D) shows a significant difference in allograft function between each group that is maintained up to 12 months after CsA withdrawal. Furthermore, the stepwise removal of CsA can be achieved safely over a 6-week period without the risk of acute rejection. This is the first therapeutic strategy proven to be effective in CsA-treated patients with chronic allograft dysfunction. These results confirm a number of uncontrolled studies in which improved graft function was maintained during follow-up periods of 6 months to 1.8 years after CsA withdrawal in the presence of MMF (6–10).
The exact mechanism by which MMF substitution for CsA results in improved renal function remains uncertain. The initial improvement in renal function observed early after CsA withdrawal in phase 2 suggests that loss of CsA-mediated vasoconstriction of the preglomerular afferent arteriole may be an important mechanism (12,13). However, inhibition of lymphocyte and macrophage-derived cytokines, together with remodeling of renal blood vessels because of inhibition of smooth muscle cell proliferation and migration as suggested by MMF use in animal models of graft vascular disease, may be responsible for a more enduring improvement (14–17).
Although there are differences at baseline between the two treatment groups with respect to prior rejections and type of transplant, neither factor influenced the analysis of response rate or SCr. There were no obvious features that distinguished between patients who did and did not respond to CsA withdrawal. At baseline, there were no significant differences in SCr, CCr, CsA dose, or CsA trough level between responders and nonresponders, or in the screening value for the slope of 1/SCr. In addition, at the 6-month endpoint, in patients who remained on CsA (group B), there was no significant difference between responders and nonresponders in CsA total dose, dose corrected for body weight, or trough levels. Angiotensin-converting enzyme inhibitors and angiotensin-II receptor antagonists were used equally in each group.
Better methods are needed to identify, at an early stage, patients with CAN who are likely to benefit from CsA withdrawal. These methods include the use of protocol biopsies, with better histologic classification, possibly using peritubular capillary C4d staining and/or cytokine gene expression profiles (18–20). Although it seems likely that early intervention, even before creatinine has started to increase, would result in a greater response, such an approach needs to be tested.
In addition to the improvement in renal function, changes in a number of other laboratory parameters were also observed in patients in whom CsA was withdrawn. Total cholesterol levels were significantly reduced at both the 6- and 12-month endpoints, consistent with the known hyperlipidemic effect of CsA. A similar decrease in serum cholesterol after CsA dose reduction or withdrawal has been documented in other studies (7,9,21). There was a trend toward lower blood pressure in group A compared with group B, but the differences did not reach statistical significance. Nonetheless, improvement in these metabolic parameters may aid in reducing cardiovascular risk in this population.
A transient but significant decrease in hemoglobin was observed during the conversion to MMF treatment and withdrawal of CsA (phase 1) in group A, which resolved by 6 months (end of phase 2) and was not clinically significant. This phenomenon is most likely because of a pharmacokinetic interaction between CsA and mycophenolic acid (MPA), the active metabolite of MMF. CsA is known to inhibit biliary excretion of the MPA-glucuronide conjugate (22). Withdrawal of CsA increases the enterohepatic recirculation of MPA-glucuronide conjugate, leading to higher MPA plasma levels (and concomitantly more suppression of erythropoiesis), which return to baseline by 6 months (23). A similar proportion of patients in each group required erythropoietin treatment during the study (Table 3).
The adverse event profile was as expected for patients taking MMF, with a predominance of gastrointestinal side effects. Although the three deaths in group A were associated with infections, two of the deaths occurred late in the study (days 279 and 392, respectively), when patients were taking only MMF and low-dose prednisolone. An increased mortality rate has not been associated with MMF use in a large number of controlled studies of solid-organ transplantation (24,25).
Long-term follow-up will determine whether the treatment responses will be maintained or even improve. A recent analysis of registry data has shown that MMF treatment is associated with a lower risk of chronic allograft failure than treatment with AZA, and that this effect is independent of acute rejection incidence (26). Whether this reflects specific effects of MMF, which may ameliorate the renal dysfunction secondary to CAN, is unclear and will require further study.
This study has established the safety and efficacy of complete CsA withdrawal in the presence of MMF and steroids as a treatment option for renal transplant recipients with a documented, progressive decline in renal function (“creeping creatinine”) secondary to CAN. The intervention is safe and well tolerated, and results in a measurable improvement in renal function and a significant reduction in serum cholesterol levels.
The authors thank Neva Coello and Michael Pickering of BIOP Biometrical Practice, Basel, Switzerland, for guidance on statistical analysis of the data. C. Dudley, P. Wijngaard, and C. Sutter designed the study. C. Dudley, E. Pohanka, H. Silva, J. Dedochova, and H. Riad were study investigators. C. Dudley, E. Pohanka, H. Riad, J. Dedochova, P. Wijngaard, C. Sutter, and H. Silva contributed to the data analysis and participated in the drafting and revision of the article.
In addition to the authors, the following investigators participated in the MMF Creeping Creatinine Study: Med. Univ. Klinik Graz, Graz, Austria, H. Holzer; Afdeling Nefrologie Universitaire Ziekenhuis, Gasthuisberg, Leuven, Belgium, Y. Vanrenterghem; Hospital Beneficencia Portuguesa, São Paulo, Brazil, I. Noronha; Hospital de Clinicas de Porto Alegre, Porto Alegre, Brazil, L.F. Gonçalves; Hospital Geral do Bonsucesso, Rio de Janeiro, Brazil, D de Boni Montes de Carvalho; St. Paul's Hospital, Vancouver, Canada, D. Landsberg; Queen Mary Hospital, Hong Kong, China, T.M. Chan; Princess Margaret Hospital, Hong Kong, China, W.K. Tsang; Faculty Hospital, Olomouc, Czech Republic, P. Bachleda; Institute for Clinical and Experimental Medicine, Prague, Czech Republic, S. Vitko; CHU Le Kremlin Bicètre, Le Kremlin Bicètre, France, B. Charpentier; Georg-August-Universität, Göttingen, Germany, C. Grupp; Medizinische Hochschule Hannover, Hannover, Germany, R. Brunkhorst, B. Nashan; KfH Kuratorium für Dialyse und Nierentransplantation eV, Jena-Drackendorf, Germany, H. Sperschneider; Semmelweis Medical University, Budapest, Hungary, F. Perner; Rikshospitalet, Oslo, Norway, P. Fauchal; Hospital Ramón y Cajal, Madrid, Spain, J. Ortuño; St. Mary's Hospital, Portsmouth, UK, S. Sadek; National Institute of Transplantation, Los Angeles, California, R. Mendez.
1. Cecka JM. The UNOS Scientific Renal Transplant Registry. Clin Transpl
2. Paul LC. Chronic allograft nephropathy: an update. Kidney Int
1999; 56: 783.
3. Paul LC, Hayry P, Foegh M, et al. Diagnostic criteria for chronic rejection/accelerated graft atherosclerosis in heart and kidney transplants: joint proposal for the Fourth Alexis Carrel Conference on Chronic Rejection and Accelerated Arteriosclerosis in Transplanted Organs. Transplant Proc
1993; 25: 2022.
4. Halloran PF, Melk A, Barth C. Rethinking chronic allograft nephropathy: the concept of accelerated senescence. J Am Soc Nephrol
1999; 10: 167.
5. Nankivell BJ, Borrows RJ, Fung CL, et al. The natural history of chronic allograft nephropathy. N Engl J Med
2003; 349: 2326.
6. Weir MR, Anderson L, Fink JC, et al. A novel approach to the treatment of chronic allograft nephropathy. Transplantation
1997; 64: 1706.
7. Weir MR, Ward MT, Blahut SA, et al. Long-term impact of discontinued or reduced calcineurin inhibitor in patients with chronic allograft nephropathy. Kidney Int
2001; 59: 1567.
8. Hueso M, Bover J, Seron D, et al. Low-dose cyclosporine and mycophenolate mofetil in renal allograft recipients with suboptimal renal function. Transplantation
1998; 66: 1727.
9. Houde I, Isenring P, Boucher D, et al. Mycophenolate mofetil, an alternative to cyclosporine A for long-term immunosuppression in kidney transplantation? Transplantation
2000; 70: 1251.
10. Ducloux D, Fournier V, Bresson-Vautrin C, et al. Mycophenolate mofetil in renal transplant recipients with cyclosporine associated nephrotoxicity: a preliminary report. Transplantation
1998; 65: 1504.
11. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron
1976; 16: 31.
12. Curtis JJ, Luke RG, Dubovsky E, et al. Cyclosporin in therapeutic doses increases renal allograft vascular resistance. Lancet
1986; 2: 477.
13. Weir MR, Klassen DK, Shen SY, et al. Acute effects of intravenous cyclosporine on blood pressure, renal hemodynamics, and urine prostaglandin production of healthy humans. Transplantation
1990; 49: 41.
14. Gregory CR, Huang X, Pratt RE, et al. Treatment with rapamycin and mycophenolic acid reduces arterial intimal thickening produced by mechanical injury and allows endothelial replacement. Transplantation
1995; 59: 655.
15. Klupp J, Dambrin C, Hibi K, et al. Treatment by mycophenolate mofetil of advanced graft vascular disease in non-human primate recipients of orthotopic aortic allografts. Am J Transplant
2003; 3: 817.
16. Allison AC, Eugui EM. Preferential suppression of lymphocyte proliferation by mycophenolic acid and predicted long-term effects of mycophenolate mofetil in transplantation. Transplant Proc
1994; 26: 3205.
17. Nadeau KC, Azuma H, Tilney NL. Sequential cytokine expression in renal allografts in rats immunosuppressed with maintenance cyclosporine or mycophenolate mofetil. Transplantation
1996; 62: 1363.
18. Mauiyyedi S, Pelle PD, Saidman S, et al. Chronic humoral rejection: identification of antibody-mediated chronic renal allograft rejection by C4d deposits in peritubular capillaries. J Am Soc Nephrol
2001; 12: 574.
19. Benigni A, Bruzzi I, Mister M, et al. Nature and mediators of renal lesions in kidney transplant patients given cyclosporine for more than one year. Kidney Int
1999; 55: 674.
20. Stegall M, Park W, Kim D, et al. Gene expression during acute allograft rejection: novel statistical analysis of microarray data. Am J Transplant
2002; 2: 913.
21. Schrama YC, Joles JA, van Tol A, et al. Conversion to mycophenolate mofetil in conjunction with stepwise withdrawal of cyclosporine in stable renal transplant recipients. Transplantation
2000; 69: 376.
22. van Gelder T, Klupp J, Barten MJ, et al. Comparison of the effects of tacrolimus and cyclosporine on the pharmacokinetics of mycophenolic acid. Ther Drug Monit
2001; 23: 119.
23. Shipkova M, Armstrong VW, Kuypers D, et al. Effect of cyclosporine withdrawal on mycophenolic acid pharmacokinetics in kidney transplant recipients with deteriorating renal function: preliminary report. Ther Drug Monit
2001; 23: 717.
24. Mathew TH. A blinded, long-term, randomised multicenter study of mycophenolate mofetil in cadaveric renal transplantation: results at three years. Tricontinental Mycophenolate Mofetil Renal Transplantation Study Group. Transplantation
1998; 65: 1450.
25. European Mycophenolate Mofetil Cooperative Study Group. Mycophenolate mofetil in renal transplantation: 3-year results from the placebo-controlled trial. Transplantation
1999; 68: 391.
26. Ojo AO, Meier-Kriesche HU, Hanson JA, et al. Mycophenolate mofetil reduces late renal allograft loss independent of acute rejection. Transplantation
2000; 69: 2405.
1 All of the authors have received honoraria, travel, and accommodation expenses incurred by attending committee meetings related to this study. C. Dudley, E. Pohanka, and H. Tedesco have received travel and accommodation expenses, payment for lecturing, or funding for research from Hoffmann-La Roche, Wyeth, Novartis, and Fujisawa Ltd. H. Riad has received unrestricted educational grants from Hoffmann-La Roche to attend international transplant conferences. C. Dudley has served as a consultant for Hoffmann-La Roche, Wyeth, and Novartis. P. Wijngaard and C. Sutter are employees of Hoffmann-La Roche Ltd., and C. Dudley is a member of the Transplantation editorial board.