• CNI nephrotoxicity;
  • cyclosporine;
  • renal function;
  • sirolimus;
  • renal transplantation


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Sirolimus (SRL) allows to minimize the use of cyclosporine (CsA), but de novo administration after transplantation is associated with various complications. We report a prospective, open-label, multicenter randomized study to evaluate conversion from a CsA-based regimen to a SRL-based regimen 3 months after transplantation. One hundred ninety-two of a total of 237 patients were eligible at 3 months to be converted to SRL (n = 95) or to continue CsA (n = 97). All patients were also given mycophenolate mofetil (MMF) and oral steroids, planned to be discontinued at month 8. The primary endpoint, the clearance estimated according to Cockcroft and Gault at week 52, was significantly better in the SRL group (68.9 vs. 64.4 mL/min, p = 0.017). Patient and graft survival were not statistically different. The incidence of acute rejection episodes, mainly occurring after withdrawal of steroids, was numerically but not statistically higher in the SRL group (17% vs. 8%, p = 0.071). Sixteen patients discontinued SRL, mainly for adverse events (n = 11), and seven patients discontinued CsA for renal failure or acute rejection. Significantly, more patients in the SRL group reported aphthous, diarrhea, acne and high triglyceride levels. Conversion CsA to SRL 3 months after transplantation combined with MMF is associated with improvement in renal function.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Since the early 1980s, the standard approach to immunosuppression in transplant recipients has involved the use of calcineurin inhibitors (CNI) such as cyclosporine (CsA) and tacrolimus. CNIs reduce the number of acute rejection episodes and enhance short-term allograft survival (1). However, the greater benefit appears to be a decrease in the number of acute rejections during the first months after transplantation. Progressive chronic nephrotoxicity is a major toxic effect of CNI in the long term and is associated with mild-to-moderate renal dysfunction. Death from cancer or cardiovascular diseases with a functioning graft is also a major cause of late graft loss (2–5). The balance between preventing immunological allograft failures, avoiding overimmunosuppression and managing nephrotoxicity is therefore still an unresolved issue. The introduction of immunosuppressive drugs such as mycophenolate mofetil (MMF) and sirolimus (SRL) has strengthened the case for minimizing the use of CsA (6–9).

SRL is an antifungal macrolide that displays potent antiproliferative activities that produce immunosuppressive effects. Binding of SRL to the intracellular immunophilin FKBP12 blocks the activity of the mammalian target of rapamycin (mTOR) with potent inhibition of downstream signaling and progression from the G1 to the S phase of the cell-cycle. SRL has been shown to reduce the incidence of acute rejection after renal transplantation, without appearing to cause significant inherent nephrotoxicity in most animal and human studies (10–12). However, when combined with CNI therapy, renal function often worsens as a result of potentiated nephrotoxicity (13). SRL has therefore been used as a primary immunosuppressant in CNI-free regimens. Together with anti-CD25 antibody or antithymocyte globulins, MMF and steroids, these regimens have provided comparable 1-year patient and graft survival, and incidence of acute rejection episodes (14–16). Furthermore, renal function was also significantly better in the CsA-free group and did not tend to worsen during the first year after transplantation. However, early use of SRL after transplantation has been demonstrated to be associated with various complications including lymphocele, prolonged delayed graft function (DGF), delayed wound healing and by a slight increased incidence of acute rejection (16–18).

One approach to overcoming this problem may be to delay the introduction of SRL, as trials have shown that withdrawal of CsA 3 months after transplantation improves renal function for up to 48 months (19,20). This strategy could avoid the early complications associated with SRL use and thus limit the long-term effects of CNI.

This prospective multicenter trial was undertaken in order to determine whether conversion from a CsA-based regimen to a SRL-based regimen 3 months after transplantation is associated with a positive risk–benefit balance, that is, improvement in renal function without increased incidence of acute rejection.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References


The CONCEPT study was an open-label, randomized, multicenter study conducted in 16 French centers between November 2004 and October 2006 (Eudract-N° 2004–002987-62). The protocol was reviewed and approved by the Institutional Review Board of Tours on February 22, 2005. Recipients of a first renal allograft, aged over 18 years and less than 75 years, were included after providing informed written consent. Noninclusion criteria were: living and donation after cardiac death, previous renal allograft, multiple organ transplantation, cold ischemia time > 36 h, donor age >65 years, peak panel reactive antibodies (PRA) >30%, active major infection (HBV, HCV, HIV), history of recent malignancy, white blood cells <2500 mm3 and Hb < 9g/dL.


All patients received a humanized anti-IL2 receptor monoclonal antibody intravenously (daclizumab, Zenapax®, Roche) delivered at 2 mg/kg on day 0 and 1 mg/kg on day 14, combined with 2 g MMF (CellCept®, Roche, Neuilly/Seine, France) daily, adjusted according to clinical events and not to MPA levels, and CsA (Neoral®, Novartis, Rueil-Malmaison, France), adjusted to maintain blood levels 2 h after intake (C2) in the range 1000–1500 ng/mL until the end of month 1 and 800–1200 ng/mL until the end of month 3. All patients received steroids at an initial dose of 500 mg at day 0, then 0.5 mg/kg/day between days 1 and 7, 0.25 mg/kg/day between days 8 and 14, followed by a progressive decrease to 10 mg/day until month 8. Oral steroids were planned to be completely discontinued at month 8.

Patients were randomized at week 12 between continuation of CsA with targeted C2 of 500 to 800 ng/mL (CsA group) or receiving SRL (SRL group). The SRL dose consisted of a loading dose of 10 mg/day for 2 consecutive days, followed by 6 mg daily adjusted to maintain C0 blood levels between 8 ng/mL and 15 ng/mL from week 12 to week 39, then between 5 ng/mL and 10 ng/mL after week 39.

Eligibility for randomization was assessed at week 12 and patients presenting one of the following criteria were not randomized: episode of acute rejection ≥ grade I according to Banff classification, estimated clearance (Cr Cl) using the Cockcroft–Gault (C–G) formula <40 mL/min at week 12, serum creatinine variations >30% during the 15 days before randomization, proteinuria >1 g/24 h, mean MMF daily dose <1.5 g during the week before randomization.

Prophylactic treatments

Prophylaxis against Pneumocystis carinii using cotrimoxazole was required for all patients. All cytomegalovirus (CMV)-negative patients who received a kidney from a CMV-positive donor received prophylaxis for CMV infection. CMV prophylaxis was given according to the center practice for a minimum of 12 weeks.

Study endpoints

The primary study endpoint was renal function at week 52 assessed by adjusted Cr Cl estimated according to the Cockcroft and Gault (C–G) formula. In model one, Cr Cl was adjusted on the following factors: center, C–G at randomization and donor age. In model two, Cr Cl was adjusted on the following factors: center, donor age and gender, EBV status (D+/R−), creatinine clearance, BMI and SBP at randomization.

Secondary efficacy endpoints included graft and patient survival, incidence, timing and severity of biopsy-proven acute rejection (BPAR), eGFR estimated according to simplified MDRD (21) and GFR measured at week 52 using iohexol clearance in patients of 13 of the 16 centers (22).

Biopsies were performed in all suspected episodes of acute rejection, centrally reviewed and scored according to Banff criteria (23). Treatment of acute rejection episodes consisted of intensified steroid therapy according to local practice, with rATG or OKT3 given at the discretion of the investigator if clinically indicated.

Safety was assessed by adverse events (especially gastrointestinal, hematological and cutaneous disorders, opportunist infections and cancers). Biochemical and hematological parameters were analyzed at fixed time points.

In accordance with the protocol, patients were prematurely withdrawn from the study at the moment they died, lost their graft, withdrew their consent or for any other reason at the investigator's discretion. No further data were collected after premature withdrawal.

Statistical analysis

Randomization at week 12 was centralized and balanced (1:1). (A minimization method was used on the following factors: center, Cr Cl C–G at randomization and donor age). Data collections were ensured by an electronic case report form and the centralized randomization was ensured via internet. Statistical analyses were performed with a two-sided significance level of 5% using SAS software, version 9.1 (SAS Institute, Cary, NC).

With a two-sided alpha of 5% and a common standard deviation of 25, 74 patients per arm were needed to achieve 80% power and to detect a difference of 10 mL/min at week 52 on Cr Cl calculated according to C–G formula. Assuming that 25% of patients would not be randomized at week 12 and 10% of randomized patients were not assessable at week 52, a total of 220 patients had to be recruited.

Mean Cr Cl were compared between treatment arms by covariance analysis adjusted in a first time on randomization factors: center, Cr Cl C–G at week 12 and donor age. Then, mean Cr Cl were compared using a second covariance analysis adjusted on the following factors: center, donor age and gender, EBV status (D+/R−), creatinine clearance, BMI and SBP at randomization.

Graft and patient survival were analyzed using the Kaplan–Meier method, and the incidence of BPAR was compared using a chi-square test. Efficacy analyses were performed on the evaluable (Figure 1) and on the intent to treat (ITT) population, defined as all randomized patients with at least one assessment of serum creatinine after randomization (LOCF analysis). Safety was assessed on all patients having received at least one study drug dose.


Figure 1. Enrollment and outcomes: 237 patients underwent assessment. Ten patients prematurely withdrew before randomization and 34 did not meet randomization criteria. At week 12, 193 patients were randomized. One patient was wrongly randomized and did not receive any study drug. This patient was excluded from intent to treat (ITT) population that consisted in 192 patients (95 in the SRL group and 97 in the CsA group). The ITT population was defined as all randomized patients with at least one assessment of serum creatinine after randomization. Safety was assessed on all patients having received at least one study drug dose.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Baseline characteristics

A total of 237 patients were included in 16 centers. Ten patients prematurely withdrew before randomization and 34 did not meet randomization criteria. The reasons for nonrandomization are presented in Figure 1. One patient was wrongly randomized and did not receive any study drug. This patient was excluded from ITT population that consisted in 192 patients (95 in the SRL group and 97 in the CsA group). Eleven patients were withdrawn prematurely after randomization (Figure 1). Ten patients in the SRL group, mainly due to withdrawal of consent (n = 6) and one patient in the CsA group for graft loss.

At baseline, there was no statistical difference between the groups for recipient characteristics except for BMI (p = 0.015) and proportion of male donors (which was slightly higher for CsA patients (p = 0.047)) (Table 1).

Table 1.  Characteristics of the population intent to treat (ITT) population (N = 192) and of the evaluable population (N = 181)
  SRL group (n = 95)1CsA group (n = 97)1SRL group (n = 85)2CsA group (n = 96)2
  1. 1ITT population.

  2. 2Evaluable population.

  3. *p < 0.05.

Recipient characteristics
 Age (years)Mean ± SD 46.5 ± 12.047.3 ± 10.646.5 ± 12.147.3 ± 10.6
 BMIMean ± SD23.8 ± 3.425.2 ± 4.2*24.0 ± 3.325.2 ± 4.2*
 Initial nephropathy%Polycystic kidney28.418.6 28.218.8
Glomerulopathy16.827.8 17.728.1
Interstitial nephritis
Other32.527.8 31.828.1
Donor characteristics
 Age (years)Mean ± SD 43.9 ± 13.744.9 ± 13.143.3 ± 14.044.8 ± 13.2
 Cause of death%Cardio vascular accident48.444.3 48.243.8
Brain trauma26.322.7 27.122.9
Other25.333.0 24.733.3
Transplant characteristics
 Preemptive transplantationn4343
 Cold ischemia time (h)mean ± SD18.8 ± 6.318.4 ± 6.1 18.4 ± 5.918.4 ± 6.1
 Panel reactive antibodies > 0%n2010
 Total HLA mismatch (n)Mean ± SD 3.9 ± 1.23.7 ± 1.33.9 ± 1.23.7 ± 1.3
 CMV status%D+/R−25.317.5 24.716.7
D−/R+25.324.7 25.925.0
D+/R+22.122.7 21.222.9
D−/R−27.335.1 28.235.4
 Delayed graft function% 13.714.6 10.614.6
 Slow graft function% 24.225.0 23.524.0

Immunosuppressive regimen

Mean targeted blood levels were achieved for both SRL and CsA (Table 2). In the SRL group, 16 patients discontinued SRL for adverse events (n = 11), acute rejection (n = 4) and treatment inefficacy (n = 1). In the CsA group, seven patients discontinued CsA for renal failure or acute rejection (p = 0.043)

Table 2.  Drug doses and blood levels (C2 CsA and C0 SRL) after randomization (ITT population)
 Drug daily dose (mg/d)Blood levels (ng/mL)MMF daily dose (g/d)Steroids (%pts)Daily steroid dose (mg/d)1
  1. 1For patients on steroids.

  2. *p < 0.001.

SRL group
 Week 14 5.2 ± 1.416.4 ± 8.2 1.9 ± 0.310010.3 ± 3.1 
 Week 26 3.7 ± 1.49.9 ± 3.41.7 ± 0.4 9.2 ± 2.6
 Week 39 3.3 ± 1.59.7 ± 3.51.7 ± 0.439.512.3 ± 16.7
 Week 52 3.2 ± 1.49.6 ± 4.31.7 ± 0.428.28.1 ± 2.6
CsA group
 Week 14270 ± 66951 ± 3401.9 ± 0.310010.7 ± 5.7 
 Week 26237 ± 52816 ± 3051.9 ± 0.2 9.6 ± 3.9
 Week 39231 ± 52758 ± 2431.9 ± 0.332.37.4 ± 3.7
 Week 52226 ± 49749 ± 233 1.9 ± 0.3*21.98.0 ± 2.3

As soon as week 26, the mean MMF dose was slightly lower in the SRL group than in the CsA group (p < 0.001) (Table 2). At week 52 corticosteroids were withdrawn in 72% of the SRL group and in 78% of the CsA group (p = 0.32).

Renal function

Adjusted Cr Cl at week 52 estimated according to the C–G formula was the primary endpoint of the study. At randomization, mean Cr Cl was 60.7 ± 14.2 mL/min in the SRL group compared to 60.3 ± 14.6 in the CsA group (p = 0.851). At the end of the study, the mean (95%CI) of Cr Cl, from covariance analysis (adjusted on the randomization factors) was significantly higher in the remaining patients in the SRL group (68.9 [65.9;71.8] mL/min, n = 85) compared to the CsA group (64.4 [61.7;67.1] mL/min, n = 96) (p = 0.017). A second covariance analysis taking into account additional factors showed similar results (66.5 [60.5;72.5] mL/min in the SRL group vs. 62.5 [55.5;68.4] mL/min in the CsA group, p = 0.028). The nonadjusted mean values were 69.6 and 64.8 mL/min, respectively. Covariance analysis also showed that the Cr Cl level at randomization and donor age were statistically correlated with Cr Cl at week 52 (p < 0.001, respectively). As shown in Figure 2, eGFR was better in the SRL group as early as 4 weeks after conversion.


Figure 2. Renal function estimated by Cockcroft and Gault 9 months after randomization (evaluable population).

Download figure to PowerPoint

Similar results were observed for eGFR calculated according to the MDRD formula, measured GFR with iohexol and serum creatinine (Table 4). A LOCF analysis including the patients who prematurely discontinued the study also showed a better renal function in the SRL group (68.0 [65.2–70.8] mL/min, n = 95) compared to the CsA group (64.4 (61.7–67.1) mL/min, n = 96) (p = 0.048). Finally the results of an ‘on treatment’ analysis demonstrated also a better renal function in the SRL group (70.8 [67.4–74.2] mL/min vs. 65.2 [62.6–68.0] mL/min), respectively (p = 0.006).

Table 4.  Renal function at week 52 (ITT population)
 SRL group (n = 85)CsA group (n = 96)p-Value
  1. Covariance analysis with adjusted mean (adjustment on randomization factors: center, eGFR C–G at week 12 and donor age).

  2. 1GFR measured using iohexol was performed in 13 out of 16 centers.

Cr Cl C–G formula (mL/min)68.9 [65.9;71.8]64.4 [61.7;67.1]0.017
eGFR simplified MDRD formula (mL/min)61.2 [58.2;64.1]53.9 [51.2;56.7]0.002
Creatininemia μM/L 117.4 [110.7; 124.2] 132.3 [126.1; 138.5]<0.001 
 SRL group (n = 59)CsA group (n = 75)p-Value
Measured GFR using iohexol1 (mL/min)67.3 [63.7;71.0]60.3 [57.0;63.5]0.004

Acute rejection episodes

Patients with clinically suspected acute rejection underwent a renal graft biopsy. After randomization, 43 biopsies (27 in SRL group and 16 in CsA group) were performed and centrally reassessed. The incidence of patients with BPAR was higher in the SRL group although not significantly different (17% vs. 8%, p = 0.071). All BPAR were mild in the SRL group (Table 3). A majority of the BPAR in the SRL group occurred late after randomization while steroids were withdrawn (Figure 3). At the time of rejection, mean serum creatinine was higher and Cr Cl lower in the CsA group than in SRL group (174.1 ± 25.9 μmol/L vs. 141.8 ± 54.5 μmol/L and 51.6 ± 7.8 mL/min vs. 62.4 ± 19.9 mL/min, respectively).

Table 3.  Biopsy-proven acute rejection after randomization in kidney allograft recipients treated with SRL or CsA (ITT population)
 SRL group (n = 95)CsA group (n = 97)p-Value
Number of biopsies2716 
Patients with BPAR, n (%)16 (16.8%)8 (8.2%)0.071
Number of episodes of BPAR1710 
Acute rejection episodes (Banff 2005)
 Grade I175 
 Grade II4 
 Grade III1 

Figure 3. Kaplan–Meier analysis of biopsy-proven acute rejection-free after randomization (ITT population).

Download figure to PowerPoint

One patient received antithymocyte globulins in the SRL group compared to three in the CsA group. After the rejection episodes, steroids were withdrawn in 3 patients in the SRL group and 1 in the CsA group, 6 patients in the SRL group and 2 in the CsA group were switched to tacrolimus. Interestingly, Cr Cl at week 52 was similar in patients with or without previous episodes of acute rejection in the SRL group (67.2 ± 20.5 mL/min vs. 69.3 ± 17.5).

Patient and graft survival

Before randomization, one patient died of septicemia 17 days after transplantation and three patients lost their grafts (two vascular thrombosis in the first 48 h and one transplantectomy at day 12 for graft failure). No deaths were reported after randomization and one patient of the CsA group lost his graft 36 weeks after randomization because of acute rejection. The total patient and graft survival in this study was thus 99.5 and 97.9%, respectively.

Adverse events

The percentages of patients reporting at least one adverse event after randomization were not different between the groups. Significantly more patients in the SRL group than in the CsA group reported aphthous stomatitis (46% vs. 5%, p < 0.001), diarrhea (30% vs. 9%, p < 0.001) and/or acne (19% vs. 5%, p = 0.004) (Table 5). Most of the aphthous stomatitis episodes (82%) occurred before steroids withdrawal. More patients presented serious adverse events in the SRL than in the CsA group (60% vs. 44%, p = 0.025). There was no difference between groups for CMV infection (4% vs. 6%) and not BK virus infection was reported. No case of post transplant lymphoproliferative disease occurred.

Table 5.  Adverse events after randomization (safety population) (as reported by investigators)
 SRL group (n = 96)CsA group (n = 97)p-Value
  1. *p < 0.01; **p < 0.01; ***p < 0.001.

TOTAL, nb events /nb patients613/94395/89ns
Aphthous stomatitis62/445/5***
Peripheral edema28/2724/22ns
CMV infection4/47/6ns
Diabetes (NODAT)32ns
Cutaneous carcinoma1ns
Prostate cancer1ns

At week 52 hemoglobin levels, total cholesterol and LDL cholesterol were similar in both groups (Table 6). Triglyceride levels were higher in the SRL group (p < 0.01). The percentages of patients with Hb < 12 g/dL were 35% and 30%, respectively. Proteinuria was not higher in the SRL group compared to the CsA group, and the number of patients with proteinuria above 0.5g/day was similar in both groups (12% and 9%, respectively).

Table 6.  Biological parameters after randomization (safety population)
 nSRL group (n = 96) Mean ± SDnCsA group (n = 97) Mean ± SD
  1. *p < 0.05; **p < 0.01.

Hemoglobin (g/dL)
 W129612.6 ± 1.7 9712.7 ± 2.0 
 W26**8712.3 ± 1.8 9713.1 ± 1.6 
 W528512.8 ± 2.0 9613.0 ± 2.0 
Cholesterol (g/L)
 W12782.1 ± 0.4862.2 ± 0.6
 W26**772.4 ± 0.5802.1 ± 0.4
 W52812.1 ± 0.8921.9 ± 0.4
LDL cholesterol (g/L)
 W12701.2 ± 0.4731.3 ± 0.5
 W26**681.4 ± 0.4711.2 ± 0.3
 W52721.1 ± 0.3861.1 ± 0.3
Triglycerides (g/L)
 W12781.7 ± 1.1861.7 ± 1.0
 W26*772.4 ± 2.8801.6 ± 1.1
 W52**801.9 ± 1.4921.4 ± 0.7
Proteinuria (g/24 h)
 W12690.2 ± 0.3700.3 ± 0.3
 W26*650.6 ± 1.4760.3 ± 0.4
 W52540.4 ± 0.5690.3 ± 0.8

At the end of the follow-up period, blood pressure was similar in both groups (135 ± 16/77 ± 11 vs. 137 ± 16/78 ± 10 for systolic and diastolic pressure in the SRL and CsA groups, respectively). Fewer patients needed antihypertensive therapy (ACE I, ARB) in SRL group compared to CsA group but the difference was not significant (49% vs. 62%, respectively) (p = 0.084). There was no statistical difference in the incidence of patients receiving lipid-lowering agents (33% in the SRL group vs. 24% in the CsA group) (p = 0.169) and EPO at the end of the study (8% vs. 5%, respectively).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Long-term maintenance of renal function has become a challenge in renal transplantation recipients. The use of CNIs is associated with chronic long-term graft dysfunction (5–7). It is therefore of major importance to define strategies of calcineurin inhibitor avoidance or elimination. We recently reported that a de novo SRL-based regimen combined with polyclonal antilymphocyte antibodies provided better renal function at 1 year but was associated with more adverse events and study withdrawals (16). Furthermore, the use of SRL from day 0 after transplantation is associated with a higher risk of surgical complications (e.g. lymphocele or wound healing) and with a longer period of DGF (17,18). These findings prompted us to study the introduction of SRL at 3 months. This is the first large randomized clinical trial assessing the efficacy and safety of such a conversion in a homogeneous population of kidney transplant patient.

The primary endpoint of this study was renal function 52 weeks after transplantation. Renal function, both estimated and measured, was significantly better in the SRL group. The targeted drug levels of CsA were in accordance with the current consensus report (24) and it is unlikely that the difference was due to too high a CsA exposure. The difference between GFR observed at week 52 was around 5 to 7 mL/min or 10 to 15%, depending the method used for calculation. This result is very similar to differences reported when CsA and SRL were compared directly (14,16) and when CsA was eliminated at 3 months (20). The improvement was observed soon after conversion and is therefore very likely to have been a direct consequence of eliminating the effects of CsA on blood pressure and renal blood flow.

One main question was the potentially increased risk of acute rejection after conversion. Compared with the CsA arm, there was a numerical increase in acute rejection rates after conversion. Although the difference was not statistically different, it may be clinically relevant. The increase did not occur soon after conversion but was noted after day 150. During this period of time, steroids were discontinued as planned in the protocol. It is possible that steroid withdrawal increases the risk of acute rejection in this specific population (25). However, we have recently reported that steroid withdrawal can be achieved in most patients in a CNI-free regimen without a high risk of rejection (16). The two main differences between these two studies were the use of SRL from day 0 and the biological induction. When SRL is used from day 0, it may be anticipated that patients with high risk of acute rejection will present an episode during the first few months and will not then be eligible for steroid withdrawal. In the present protocol, since patients were receiving conventional treatment during the early phase after transplantation, they were not assessed according to their immunological risk. The role of polyclonal antibodies may also be of importance in this context. Indeed, when polyclonal antithymocyte globulin is used, relative by long-term T-cell depletion is achieved, thus allowing safe steroid withdrawal (26–28). The use of induction with anti IL-2 R antibodies without long-term T-cell depletion may not allow safe steroid withdrawal in the context of CNI withdrawal. In a recent large multicenter trial, a 1-year rejection rate of 35% in renal graft recipients with a low dose SRL-based CNI-free treatment was reported after induction with daclizumab (29). Furthermore, only two doses of daclizumab were given instead of the usual five dose scheme. However, it must be emphasized that in this study, steroids were successfully withdrawn at 1 year in 72% of the patients in the SRL group compared to 78% in the CsA group, thus avoiding the long-term complications of steroids. This is similar to the results reported in our previous study (16). Interestingly, these episodes of acute rejection during steroid withdrawal were mild, steroid sensitive, and had no consequence on renal function at 1 year. The adverse event profile reported here is in accordance with those reported before (30–32). Delayed introduction of SRL allows avoidance of surgery-related complications or prolonged tubular necrosis (30–32). Furthermore, it makes possible better management of complications of SRL such as proteinuria and dyslipidemia with specific treatments (e.g. renin angiotensin blockers or statins, respectively). Cholesterol levels were similar in both groups at 12 months.

Interestingly, in contrast to the previous study of de novo use of SRL, we did not observe increased proteinuria (16). This could be due to the nonrandomization of patients with a previous episode of acute rejection or with proteinuria,

The major adverse events leading to drug discontinuation were skin lesions and mouth ulcers. The use of a loading dose may be related to the occurrence of these complications, as the mean trough level of SRL was above the expected target 2 weeks after the introduction of treatment and as a majority of these episodes occurred early, before steroid withdrawal. A recent report described a simple way to treat mouth ulcers without drug discontinuation (33), but a better strategy could avoid loading doses. Finally, as MMF daily dosage was not intentionally reduced but adapted to clinical events, and not to MPA levels, some of these events, such as diarrhea, could be due to MMF, as the bioavailability of mycophenolate acid is significantly higher in combination with SRL (34). In summary, we report here a new strategy to combine SRL with MMF in the early period after transplantation. The introduction of SRL after 3 months is associated with an improvement in renal function. Nevertheless, because of the increased percentage of acute rejection and the higher dropout rate due to adverse events, the benefit of such a strategy must be confirmed in the long term.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

We thank the Spiesser Group and Drs. N. Kossari, F. Glowacki, F. Bayle for their implication in the study. We thank Françoise Allano, clinical study manager and Cécile Hayem, statistician (Roche, France). This study was sponsored by a grant from Roche SAS, Neuilly sur Seine, France.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References