Journal Logo

Clinical and Translational Research

Results of a Prospective Randomized Trial of Sirolimus Conversion in Kidney Transplant Recipients on Early Corticosteroid Withdrawal

Heilman, Raymond L.1,4; Younan, Kerrie2; Wadei, Hani M.2; Mai, Martin L.2; Reddy, Kunam S.3; Chakkera, Harini A.1; Gonwa, Thomas A.2

Author Information
doi: 10.1097/TP.0b013e31822805d7


Advances in immunosuppression after kidney transplantation have resulted in remarkably low rates of acute rejection during the first posttransplant year but long-term graft survival continues to be disappointing (1). The causes of long-term graft loss include patient death, recurrent primary disease, poor patient compliance, and progressive interstitial fibrosis with tubular atrophy (2). The cause of interstitial fibrosis with tubular atrophy is multifactorial and includes chronic calcineurin inhibitor (CNI) nephrotoxicity (3). As a result, one strategy is the minimization or elimination of CNIs. Most of these protocols have substituted a mammalian target of rapamycin (mTOR) inhibitor for the CNI. In an early multicenter trial of cyclosporine withdrawal at 3 months posttransplant followed by continued maintenance with sirolimus, the study showed a significant improvement in estimated glomerular filtration rate (eGFR) at 1 year in the CNI withdrawal group (4). Other studies of CNI minimization in combination with sirolimus were encouraging (5), but data from these studies suggested that the combination of CNI and sirolimus resulted in a lower GFR (4, 6).

Results of complete CNI avoidance studies have been mixed. The studies by Flechner et al. (7, 8) demonstrated improved GFR and kidney allograft histology in sirolimus treated patients. However, the study by Larson et al. (9) failed to show a benefit in GFR or graft histology in the patients treated with sirolimus (10). A large multicenter trial of immunosuppression minimization demonstrated that a CNI-free group, treated with reduced dose sirolimus, experienced more rejection and a lower GFR compared with CNI-treated patients, but these findings may have been related to suboptimal immunosuppression resulting from inadequate sirolimus exposure (11). All of these studies have included chronic maintenance corticosteroids.

Corticosteroids contribute to many chronic clinical problems after kidney transplant; therefore, several groups have studied early steroid withdrawal (12–15). A recent report of a 5-year multicenter randomized placebo-controlled trial showed a higher risk of mild acute rejection, but the 5-year patient and graft survival and GFR were similar between the groups (16).

This study is a prospective trial of kidney transplant recipients treated with a rapid corticosteroid withdrawal protocol in combination with tacrolimus and mycophenolate mofetil (MMF) for 1 month followed by randomization to switch to sirolimus and MMF or to continue tacrolimus and MMF. All patients were followed for a minimum of 2 years. The primary outcome was the measured GFR groups at 1-year posttransplant using an intention-to-treat analysis. We hypothesized that CNI-free immunosuppression would result in a better GFR at 1 year.


The patient recruitment and disposition throughout the study is shown in Figure 1. One hundred seventy-five patients were recruited from two transplant centers between 2005 and 2008. Fifty-three patients failed to randomize (30%) for various reasons as detailed in Figure 1. Sixty-two patients were randomized to the sirolimus group and 60 to the tacrolimus group.

Patient flow diagram for the study. One hundred seventy-five patients were consented for the study and 53 patients failed to randomize for the various reasons indicated. The patients who were withdrawn from the study at various time points are shown.

The baseline characteristics of the treatment groups and the nonrandomized patients are given in Table 1. There were no significant differences in patient or transplant characteristics between the study groups. African Americans were 14% of the enrolled patients but a larger proportion of African Americans failed to randomize (54% compared with 26% of all other racial groups, P=0.012). This was primarily related to low graft function in 21% and rejection in 13%.

Baseline characteristics

Sixty-three percentage of the patients in the sirolimus group withdrew during the 2-year period of the study as compared with 18% of the tacrolimus group (P<0.0001) (Table 2). Rejection and medication side effects were the most common reason for withdrawal and both were more frequent in the sirolimus group. Withdrawal after rejection occurred in 13% of the sirolimus group and 2% of the tacrolimus group (P=0.04) while medication side effect resulted in withdrawal in 37% of the sirolimus group and 2% in the tacrolimus group (P<0.0001).

Withdrawal from therapy

The drug side effects prompting withdrawal from study drug are shown in Table 3. Oral ulcerations (7), proteinuria (4), and hyperlipidemia (4) were the most common, and all occurred exclusively in the sirolimus group. Two sirolimus-treated patients experienced pneumonitis.

Withdrawal for drug side effect (may be more than 1 per patient)

There were no significant differences in the dose of MMF between the study groups at 4 months (1350±589 mg/day sirolimus group vs. 1658±437 mg/day tacrolimus group) or 1 year (1276±647 mg/day sirolimus group vs. 1438±485 mg/day tacrolimus group). The sirolimus level was 11.0±4.4 ng/dL at 4 months and 9.8±3.6 ng/dL at 1 year. The tacrolimus level was 7.6±3.8 ng/dL at 4 months and 6.9±4.6 ng/dL at 1 year. At 1 year, 15% of patients in the sirolimus group 3% in the tacrolimus group were on corticosteroids (P=0.07).

Rejection, defined as Banff grade 1A or greater, occurred in 10.2% of all consented patients during the first year. Rejection during the first year occurred in 13.2% of the nonrandomized patients mostly in the first month. Rejection during the first year occurred in 5% of the tacrolimus group and 13% of the sirolimus group, occurring primarily after 1 month (P=0.15). If Banff borderline classification was included, 3 (5%) in the tacrolimus group and 14 (23%) in the sirolimus group experienced rejection during the first year (P<0.01). Biopsy-proven rejection (including Banff borderline changes) occurred during the first month in 12 patients; 3 (5%) in the tacrolimus group, 3 (5%) in the sirolimus group, and 6 (11%) in the group who were not randomized. The six patients with early rejection who were subsequently randomized had a follow-up 1 month biopsy, which was negative for rejection.

Actual 2-year patient survival was 98.4% for sirolimus group, 96.7% for the tacrolimus group (Table 4). Actual 2-year graft survival (including death) was 98.4% in the sirolimus group, 96.7% in the tacrolimus group, and 81% in the nonrandomized group (P<0.0007). The most common causes of graft loss were death (8) and graft thrombosis (4). There were no graft losses from rejection. There were two deaths related to cancer (1 sirolimus and 1 nonrandomized) but no cases of posttransplantation lymphoproliferative disorder. During the first posttransplant year, cytomegalovirus (CMV) viremia occurred in 13% of patients in both treatment groups and BK nephropathy occurred in 2% of the sirolimus group and 4% of the tacrolimus group (not significant).

Graft and patient survival

The primary outcome was the differences in the measured GFR (by iothalamate clearance or creatinine clearance) at 1 year between study groups by an intention-to-treat analysis. The measured GFR was 57.4±20.7 mL/min/1.73 m2 in the sirolimus group and 62.7±26.5 mL/min/1.73 m2 in the tacrolimus group (P=0.54, 95% CI −3.7 to 14.4) (Table 5). At 1-year posttransplant, measured GFR was available in 96% of the tacrolimus group (39% by iothalamate) and 92% of the sirolimus group (44% by iothalamate). The estimated GFR using the Chronic Kidney Disease Epidemiology Collaboration formula (17) was 63.0±19.1 mL/min/1.73 m2 in the sirolimus group and 59.8±18.9 mL/min/1.73 m2 in the tacrolimus group (P=0.36, 95% CI −10.2 to 3.8). The GFR by iothalomate clearance was higher in the tacrolimus group at 1 year for the tacrolimus group, by intention-to-treat or on-therapy analysis.

Renal function outcomes

For patients who remained on therapy at 1 year, the measure and estimated GFR values were not significantly different for the study groups (Table 5). At 2 years, for the patients on therapy, the measured GFR was numerically higher in the sirolimus group (73.4±34.6 and 62.9±22.2), but this difference was not significant (P=0.21).

After randomization, the total cholesterol and triglycerides were significantly higher in the sirolimus group at 4 and 12 months. At 4 months, the cholesterol level was 212±55 mg/dL for sirolimus group and 179±34 mg/dL for tacrolimus group (P<0.001) and the triglyceride level was 238±147 mg/dL for sirolimus group and 168±89 mg/dL for tacrolimus group (P=0.003). At 1 year, the cholesterol was 202±60 mg/dL for the sirolimus group and 178±44 mg/dL for tacrolimus group (P=0.018) and the triglyceride level was 229±136 mg/dL for sirolimus group and 154±103 mg/dL for tacrolimus group (P=0.001).

There were no significant differences in the measurements of glucose metabolism, blood pressure values or number of antihypertensive agents between the groups at any time point during the study (data not shown).


We conducted a randomized clinical trial of CNI elimination at 1 month posttransplant in kidney transplant recipients on a rapid corticosteroid withdrawal protocol expecting to see improved GFR in the group converted to sirolimus. Based on our power calculations, we initially intended to randomize 230 patients. However, we stopped enrollment early after enrolling only 175 patients for three reasons. First, the high dropout rate in the sirolimus group related to poor tolerance of the study protocol made it difficult to achieve our target enrollment. Second, the interval data analysis failed to show even a trend in the outcome to support the study hypothesis. Third, when we included the borderline rejections in the analysis, there was a higher risk of rejection in the sirolimus-treated patients. The data safety monitoring board (DSMB) evaluated these factors and agreed with the investigators to end the study early. However, we continued all patients to the 2-year time point for study analysis and safety monitoring.

The primary outcome of the study, which was the difference in the measured GFR at 1 year between the study groups by an intention-to-treat analysis, was not different between the study groups (Table 5). In addition, there was a trend toward more acute rejection during the first posttransplant year in the sirolimus group.

Previous studies of early use or conversion to mTOR inhibitor therapy on GFR have been mixed. Buchler et al. (18) reported the results of a multicenter randomized trial of 145 kidney transplant recipients comparing sirolimus and cyclosporine in patients receiving a deceased donor kidney transplant. The primary outcome was the eGFR using Nankivell formula at 1 year. There was no significant difference in eGFR between the groups by intention-to-treat analysis (60±27 mL/min in the sirolimus group and 57±21 mL/min in the cyclosporine group), although for the patients remaining on study drug, the eGFR was higher in the sirolimus group 69±19 mL/min compared with 60±14 mL/min. Discontinuation from study drug due to adverse events was higher in the sirolimus group (28.2%) then in the cyclosporine group (14.9%).

Lebranchu et al. (19) randomized 192 kidney transplant recipients to switch to sirolimus at 3 months posttransplantation or stay on cyclosporine along with MMF and steroids. The primary outcome was the eGFR by the Cockcroft-Gault formula at 1 year by an intention-to-treat analysis. The eGFR was higher in the sirolimus group (68.9 vs. 64.4 mL/min). Adverse events were more common in the sirolimus group primarily related to anemia, oral ulcers, and rash. Eleven patients were withdrawn from the sirolimus and one from the cyclosporine group. Rejection occurred in 17% of the sirolimus group compared with 8% of the cyclosporine group (P=0.07).

Weir et al. (20) randomized 299 patients between 30 and 120 days posttransplant to switch to sirolimus-MMF-prednisone or stay on CNI-MMF-prednisone. The primary endpoint, which was the change in measure GFR from 1 to 12 months, was higher in the sirolimus group but by 24 months, there was no difference between the groups. In this study, 19% of the sirolimus group was withdrawn due to adverse events.

Finally, Budde et al. (21) recently reported the results of a multicenter-randomized trial of 300 kidney transplant recipients with CNI elimination at 4.5-month posttransplant followed by everolimus-based immunosuppression, showing a significantly better eGFR at 1 year in the CNI elimination group. Withdrawal due to drug intolerance occurred in 24% of the everolimus-treated group. All of these trials included maintenance corticosteroids.

Sirolimus conversion was poorly tolerated in our study. This was primarily related to medication side effects. Thirty-seven percentage of the sirolimus group was withdrawn from study drug related to medication side effects. This is generally higher than the previous referenced studies. These differences may be related to the absence of corticosteroids in our study which may modulate some of the toxicities of the mTOR inhibitor.

There are several limitations to our study. The measured GFR was a combination of iothalamate clearance or 24 hr creatinine clearance. Creatinine clearance as a method for measuring GFR has several technical limitation. In addition, neither the patients nor investigators were blinded to the study drug. The decision to withdraw therapy was not protocol driven. Both of these limitations could have introduced a bias to the decision to withdraw patients from the study drug. Finally, the decision to end enrollment prematurely may have underpowered the results to show a difference in GFR between the study groups (type 2 error).

However, there are several interesting observations generated by this study. We relied on actual measured GFR, whether by iothalamate or creatinine clearance, rather than estimated GFR. This allows us to know the true GFR at each time point. The excellent renal function, graft survival, and overall low incidence of rejection were achieved with the lymphocyte depleting induction agent and rapid steroid withdrawal demonstrating the overall success and safety of that strategy. Patients who remained on tacrolimus had the same renal function at 1 and 2 years compared with those who remained on sirolimus. We conclude, therefore, that sirolimus conversion at 1 month posttransplant in kidney recipients on a rapid steroid withdrawal protocol is poorly tolerated and does not seem to improve the measured GFR at 1 year.


Study Design

The trial was a 2-year, prospective, randomized, nonblinded trial of CNI elimination at 1 month posttransplantation in kidney transplant recipients on rapid corticosteroid withdrawal. Patients were enrolled at Mayo Clinic in Florida or Mayo Clinic in Arizona. The study was approved by the Mayo Clinic Institutional Review Board and was registered with (identifier NCT00170053). The trial was an investigator initiated study, funded by the Pfizer (formerly Wyeth) and Genzyme Corporations. (Cambridge, MA). The study design, collection of the data, data storage, data analyses, and writing of this manuscript were all performed by the investigators independent of the funding sources.

Patients were eligible if they were receiving a deceased or living donor kidney transplant, were at least 18 years of age, and could give informed consent. Exclusions included loss of a previous transplant from rejection or recurrent primary disease, combined organ transplant, history of malignancy other than nonmelanoma skin cancer within 3 years, infection with hepatitis C, HIV, chronic corticosteroid use within 6 months of transplant, panel reactive antibodies more than 20%, serum cholesterol more than 300 mg/dL or triglycerides more than 400 mg/dL, positive flow crossmatch at the time of transplant, extended criteria donor, donation after cardiac death donor, pretransplant WBC less than 3.0×103 cells/mm3 or platelet count less than 100,000 cells/mm3.

All consented patients received induction with rabbit-antithymocyte globulin (Thymoblobulin; Genzyme) for a total dose of 6 mg/kg starting with 1.5 mg/kg on the day of transplant and usually given in four divided doses. Tacrolimus was started when the serum creatinine dropped by at least 30%, or by postoperative day 4. Trough tacrolimus levels were 10 to 12 ng/mL for the first 30 days, 8 to 10 ng/mL between days 30 to 90, and 5 to 8 ng/mL after 90 days. MMF was started at 2000 mg/day, in divided doses, and adjusted according to the individual patient's tolerance. The protocol for rapid corticosteroid withdrawal was as follows: 500 mg of methylprednisolone (MP) administered intravenously (IV) intraoperatively, MP 250 mg IV on postoperative day (POD) 1, MP 125 mg IV on POD 2, prednisone (P) 60 mg orally on POD 3, and P 30 mg orally on POD 4 and no corticosteroids thereafter.

CMV prophylaxis was valgancyclovir 450 mg daily for 3 months if the recipient's CMV serology was positive or if the donor was positive and recipient was negative. All recipients received trimethoprim-sulfamethoxazole (80 mg/400 mg daily) for total of 6 months posttransplant.

Consent was obtained before transplantation and randomization occurred at 1 month if the patient had adequate wound healing and a 1 month biopsy negative for rejection. Patients with rejection early after transplant were still eligible if the 1 month biopsy was negative. Patients were recruited sequentially and assigned to the tacrolimus or sirolimus groups at random in a one-to-one ratio. Patients were stratified according to donor type and race. Treatment allocation was assigned by using a computer random number generator. Protocol biopsies were performed at 1, 4, 12, and 24 months. All biopsies were interpreted using the Banff '97 criteria (22).

The process of sirolimus conversion changed during the trial. For the initial few months subjects were given a loading dose for 3 days, the MMF dose was decreased by 50%, the tacrolimus dose was decreased by 50% and sirolimus was dosed at 10 mg in single daily dose. Once the sirolimus level was more than 8 ng/mL, the tacrolimus was stopped. This sirolimus conversion protocol was changed after the initial few months due to large number of adverse events, including oral ulcers, rash, and edema. The sirolimus conversion protocol was changed by eliminating the initial loading dose and starting with 3 mg daily dose.

Except for the management of leucopenia, any decision to withdrawal patients from the study drug was not protocol driven. This decision was based on the local investigator's clinical assessment and the patient's desire. Patients removed from sirolimus were switched to tacrolimus and vise versa.

Statistical Methods

Power calculations were performed with the aim of detecting a 20% increase in mean GFR with 80% power at the 5% significance level in the GFR in the sirolimus group compared with the tacrolimus group at 12 months posttransplant. Sample size calculations were used to estimate the need to enroll 180 patients to show a 20% difference in GFR at 1 year with confidence of 80%. A 20% increase from 62 would be approximately 12 mL/min/1.73 m2. With a standard deviation of 30 mL/min/1.73 m2, 100 patients will be needed per arm. To allow for potential unavailability of GRF measures at 1 year on up to 10% of patients, we intended to randomize a total of 230 patients.

Patient accrual began in June 2005 and was stopped in September 2008. The DSMB reviewed outcomes at defined intervals with predefined stop points. No patient safety stop points were reached. A decision was made by the investigators (supported by the DSMB) to stop enrollment prematurely because of the high drop out rate in the sirolimus group and because interval data analysis which failed to show even a trend in the outcome to support the study hypothesis.

Descriptive analysis of the baseline characteristics was performed using two-sided t test or chi-square testing. The primary outcome was the difference in measured GFR at 1 year using an intention-to-treat analysis. GFR was measured by iothalamate clearance or 24 hr creatinine clearance. Secondary outcomes included GFR at 2 years, biopsy-proven acute rejection during the first year, patient and graft survival, percentage of patients who remained steroid free at 1 year, lipid parameters, difference in glucose metabolism, and difference in blood pressure between the study groups. Acute rejection was defined as the occurrence of the first biopsy-proven rejection with Banff Classification of Grade 1A (minimum of i2, t2) or greater (23).


1. Meier-Kriesche H-U, Schold JD, Kaplan B. Long-term renal allograft survival: Have we made significant progress or is it time to rethink our analytic and therapeutic strategies? Am J Transplant 2004; 4: 1289.
2. El-Zoghby ZM, Stegall MD, Lager DJ, et al. Identifying specific causes of kidney allograft loss. Am J Transplant 2009; 9: 527.
3. Cosio FG, Grande JP, Larson TS, et al. Kidney allograft fibrosis and atrophy early after living donor transplantation. Am J Transplant 2005; 5: 1130.
4. Gonwa TA, Hricik DE, Brinker K, et al. Improved renal function in sirolimus-treated renal transplant patients after early cyclosporine elimination. Transplantation 2002; 74: 1560.
5. Johnson RW, Kreis H, Oberbauer R, et al. Sirolimus allows early cyclosporine withdrawal in renal transplantation resulting in improved renal function and lower blood pressure1,2,10. Transplantation 2001; 72: 777.
6. Gonwa T, Mendez R, Yang HC, et al. Randomized trial of tacrolimus in combination with sirolimus or mycophenolate mofetil in kidney transplantation: Results at 6 months. Transplantation 2003; 75: 1213.
7. Flechner SM, Goldfarb D, Modlin C, et al. Kidney transplantation without calcineurin inhibitor drugs: A prospective, randomized trial of sirolimus versus cyclosporine. Transplantation 2002; 74: 1070.
8. Flechner SM, Goldfarb D, Solez K, et al. Kidney transplantation with sirolimus and mycophenolate mofetil-based immunosuppression: 5-year results of a randomized prospective trial compared to calcineurin inhibitor drugs. Transplantation 2007; 83: 883.
9. Larson TS, Dean PG, Stegall MD, et al. Complete avoidance of calcineurin inhibitors in renal transplantation: A randomized trial comparing sirolimus and tacrolimus. Am J Transplant 2006; 6: 514.
10. Dean PG, Grande JP, Sethi S, et al. Kidney transplant histology after one year of continuous therapy with sirolimus compared with tacrolimus. Transplantation 2008; 85: 1212.
11. Ekberg H, Tedesco-Silva H, Demirbas A, et al. Reduced exposure to calcineurin inhibitors in renal transplantation. N Engl J Med 2007; 357: 2562.
12. Anil Kumar MS, Heifets M, Fyfe B, et al. Comparison of steroid avoidance in tacrolimus/mycophenolate mofetil and tacrolimus/sirolimus combination in kidney transplantation monitored by surveillance biopsy. Transplantation 2005; 80: 807.
13. Matas AJ, Kandaswamy R, Gillingham KJ, et al. Prednisone-free maintenance immunosuppression. A 5-year experience. Am J Transplant 2005; 5: 2473.
14. Opelz G, Dohler B, Laux G. Long-term prospective study of steroid withdrawal in kidney and heart transplant recipients. Am J Transplant 2005; 5: 720.
15. Vincenti F, Schena FP, Paraskevas S, et al. A randomized, multicenter study of steroid avoidance, early steroid withdrawal or standard steroid therapy in kidney transplant recipients. Am J Transplant 2008; 8: 307.
16. Woodle ES, First MR, Pirsch J, et al. A prospective, randomized, double-blind, placebo-controlled multicenter trial comparing early (7 day) corticosteroid cessation versus long-term, low-dose corticosteroid therapy. Ann Surg 2008; 248: 564.
17. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150: 604.
18. Buchler M, Caillard S, Barbier S, et al. Sirolimus versus cyclosporine in kidney recipients receiving thymoglobulin, mycophenolate mofetil and a 6-month course of steroids. Am J Transplant 2007; 7: 2522.
19. Lebranchu Y, Thierry A, Toupance O, et al. Efficacy on renal function of early conversion from cyclosporine to sirolimus 3 months after renal transplantation: Concept study. Am J Transplant 2009; 9: 1115.
20. Weir MR, Mulgaonkar S, Chan L, et al. Mycophenolate mofetil-based immunosuppression with sirolimus in renal transplantation: A randomized, controlled spare-the-nephron trial. Kidney Int 2011; 79: 897.
21. Budde K, Becker T, Arns W, et al. Everolimus-based, calcineurin- inhibitor-free regimen in recipients of de-novo kidney transplants: An open-label, randomised, controlled trial. Lancet. 2011; 377: 837.
22. Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999; 55: 713.
23. Solez K, Colvin RB, Racusen LC, et al. Banff '05 meeting report: Differential diagnosis of chronic allograft injury and elimination of chronic allograft nephropathy (‘CAN′). Am J Transplant 2007; 7: 518.

Kidney transplantation; Steroid avoidance; Sirolimus; Tacrolimus; Randomized trial

© 2011 Lippincott Williams & Wilkins, Inc.