Conversion From Calcineurin Inhibitors to Sirolimus Maintenance Therapy in Renal Allograft Recipients: 24-Month Efficacy and Safety Results From the CONVERT Trial : Transplantation

Journal Logo

Clinical and Translational Research

Conversion From Calcineurin Inhibitors to Sirolimus Maintenance Therapy in Renal Allograft Recipients: 24-Month Efficacy and Safety Results From the CONVERT Trial

Schena, Francesco P.1,11; Pascoe, Michael D.2; Alberu, Josefina3; del Carmen Rial, Maria4; Oberbauer, Rainer5; Brennan, Daniel C.6; Campistol, Josep M.7; Racusen, Lorraine8; Polinsky, Martin S.9; Goldberg-Alberts, Robert9; Li, Huihua9; Scarola, Joseph9; Neylan, John F.9 for the Sirolimus CONVERT Trial Study Group

Author Information
Transplantation 87(2):p 233-242, January 27, 2009. | DOI: 10.1097/TP.0b013e3181927a41
  • Free


Despite concerted efforts over the last 20 years, improvements in short-term renal allograft survival have not translated into increased graft survival in the long term, in part because of calcineurin inhibitor (CNI) nephrotoxicity (1–3). Although CNIs have long been the standard of care for immunosuppression after solid organ transplantation, concerns about their nephrotoxicity have prompted investigations into strategies that minimize or eliminate CNI use.

Sirolimus (SRL) is first in a novel class of immunosuppressants that inhibit the mammalian target of rapamycin, a key serine-threonine kinase involved in regulation of cell growth and proliferation (4, 5). Phase 2 and 3 multicenter clinical trials have demonstrated that cyclosporine (CsA) could be withdrawn safely from a combined SRL, CsA, and corticosteroid regimen at 2 to 4 months after transplantation, with resultant improvements in long-term patient and graft survival, biopsy-confirmed preservation of renal parenchyma, sustained improvement in renal function and blood pressure, improved quality of life, and less malignancy (6–10).

Most renal allograft recipients currently receive CNI-based immunosuppression as initial therapy. On the basis of the favorable outcomes from the phase 3 CsA-withdrawal study, a large, multicenter, clinical trial was conducted to evaluate the safety and efficacy of conversion from CNI- to SRL-based maintenance of immunosuppression in renal allograft recipients.


Study Design

The Sirolimus Renal Conversion Trial (CONVERT) is a prospective, open-label, comparative study in which maintenance renal allograft recipients were randomly assigned (2:1) to undergo conversion from CsA- or tacrolimus-based immunosuppression to SRL within 24 hours of randomization or to continue receiving CNI-based therapy for 104 (subsequently extended to 208) weeks. The enrollment period began on February 5, 2002, and ended on March 1, 2004. The computerized randomization/enrollment system of automatic transtelephonic randomization was used to assign study treatment groups. Allocation was determined by a blinded, randomized block design within each stratum level and center.

This study was conducted according to the ethical principles of the Declaration of Helsinki. Approval for the study was obtained from the institutional review board or independent ethics committees of each participating center. All patients provided written informed consent before enrollment.

Eligibility Criteria

Patients eligible for enrollment were aged more than or equal to 13 years and recipients of living or deceased donor renal allografts within 6 to 120 months before randomization. Eligible patients had been receiving a CNI (CsA or tacrolimus) after transplantation along with corticosteroids, and azathioprine (≥50 mg/day) or mycophenolate mofetil (MMF, ≥500 mg/day) for at least 12 weeks before randomization. Additional requirements included a functioning allograft with calculated Nankivell glomerular filtration rate (GFR) more than or equal to 20 mL/min. Patients were stratified into two groups according to their baseline-calculated GFR: 20 to 40 mL/min or more than 40 mL/min. Enrollment in the GFR 20 to 40 mL/min stratum was halted on August 19, 2003. Baseline biopsies were required within 16 weeks before randomization; all biopsies were evaluated by a central pathology laboratory to assess the type and severity of preexisting renal parenchymal injury using Banff 97 criteria (11). Patients were excluded from participation if they had been treated for biopsy-confirmed or clinically diagnosed (presumptive) acute rejection (AR) within 12 weeks of enrollment.


The primary efficacy endpoint was Nankivell GFR in the intent-to-treat (ITT) population 12 months after randomization, stratified by baseline GFR 20 to 40 mL/min or more than 40 mL/min. Secondary endpoints included Nankivell GFR in the on-therapy population, incidence and severity of biopsy-confirmed AR (BCAR), patient and allograft survival, withdrawal from assigned therapy, and blood pressure. The primary safety endpoint was the composite rate of the first occurrence of BCAR, graft loss, or death 12 months after randomization.

Post Hoc Analyses

A change from baseline GFR of 5.0 mL/min was defined as the minimum increment of improvement that would be considered clinically meaningful. On the basis of this definition, binary analyses were performed for patients in baseline GFR more than 40 mL/min stratum to determine the percentages of patients who showed a clinically meaningful benefit from SRL conversion and CNI continuation. Additional analyses were performed using changes from baseline GFR of more than or equal to 7.5 mL/min and more than or equal to 10.0 mL/min at 12 and 24 months.

On the basis of the earlier definition of clinically meaningful GFR improvement, a subgroup of patients was identified whose risk-benefit profile after conversion was found to be more favorable than that for the overall SRL conversion cohort. These patients were defined as having GFR more than 40 mL/min and a baseline urinary protein-to-creatinine ratio (UPr/Cr) of less than or equal to 0.11.

Treatment Regimens

Patients randomly assigned to SRL conversion received a single loading dose ranging from 12 to 20 mg between 4 and 24 hours after the last dose of CNI. On day 2 of study, a daily dose of 4 to 8 mg SRL was given until results of the first SRL trough level were available (study days 5–7). Thereafter, SRL was concentration controlled to 8 to 20 ng/mL (chromatographic method). On achieving this range, the maximum allowable doses of MMF and azathioprine were reduced to 1.5 g/day and 75 mg/day, respectively; thereafter, either MMF or azathioprine could be discontinued as per investigator’s discretion.

Patients randomly assigned to CNI continuation remained on CsA (trough concentrations 50–250 ng/mL) or tacrolimus (trough concentrations 4–10 ng/mL) for the duration of the study. Switching of CNIs was permitted. Both groups received corticosteroid doses of 2.5 to 15 mg/day or equivalent regimen with every-other-day of dose administration.

Laboratory Determinations

Study drug concentrations were determined by high-performance liquid chromatography (SRL) or by standard monoclonal assays (CsA, tacrolimus). All other variables were determined by standard laboratory methodology. Urinary protein excretion was determined by UPr/Cr from single-voided specimens other than first morning void.

Statistical Analyses

Analysis of covariance was used to evaluate the primary efficacy endpoint, with Nankivell GFR at month 12 as the dependent variable, treatment and center as fixed effects, and baseline Nankivell GFR as covariate. Patients who had missing values for the primary endpoint, who experienced graft loss before month 12 or who died had GFR set to zero. Differences in AR rates were assessed by 95% confidence intervals (CIs) and Fisher’s exact test; severity of rejection, as determined by Banff 97 criteria (11), was assessed by a row mean score rank test. Kaplan-Meier methodology and log-rank test were used to assess patient and graft survival.

Safety analyses included all patients who received at least one dose of study medication. Analysis of covariance was used to analyze changes from baseline in blood pressure. Fisher’s exact test was used to analyze treatment failure. A post hoc, multiple regression analysis was conducted to identify factors contributing to the observed change from baseline GFR at 24 months after randomization; 34 historical, clinical, and histopathologic variables were included. The study was designed to randomly assign a sufficient number of patients such that approximately 600 would complete 12 months of therapy. Five hundred twenty-eight patients, 80% of the 660 randomly assigned patients, were expected to complete 12 months of therapy. For these 528 patients, 352 patients in the SRL conversion group and 176 in the CNI continuation group, a two-tailed t test of the difference in means, with a significance level of α equal to 0.05, would have a greater than 90% power to detect a difference of 7.5 mL/min in the changes from baseline Nankivell GFR at 52 weeks.


Patient Demographics

A total of 830 patients were enrolled at 111 centers in Asia, Australia, Europe, the Middle East, North America (Canada, Mexico, United States), South Africa, and South America (Argentina, Brazil, Chile). Baseline demographic characteristics were similar between treatment groups within each GFR stratum (Table 1). Figure 1 delineates the disposition of patients.

Baseline demographic characteristics (intent-to-treat population) and severity grades of baseline chronic allograft nephropathy
Patient disposition.

Mean time from transplantation to randomization was higher for GFR 20 to 40 mL/min versus GFR more than 40 mL/min patients (44.5 vs. 37.3 months, respectively). In addition, baseline GFR 20 to 40 mL/min patients were older (46 years vs. 43 years), had longer mean total organ ischemia times (17.4 hours vs. 12.2 hours), and a higher percentage had received deceased donor allografts (83.9% vs. 60.7%).

Baseline biopsies were obtained at a mean of 3.2 years after transplantation and presented a broad range and severity of renal parenchymal injury. Banff 97 severity scores were significantly higher for SRL conversion versus CNI continuation patients within each stratum (Table 1).

Primary and Secondary Efficacy Outcomes

Seven hundred forty-three patients (89.5%) were assigned to the GFR greater than 40 mL/min stratum. For these patients, the primary (ITT) analysis at 12 months showed mean Nankivell GFRs of 59.0 and 57.7 mL/min (SRL conversion and CNI continuation, respectively); the difference was 1.3 mL/min in favor of SRL conversion (95% CI, −1.06–3.69; P=0.278), but it was not statistically significant. In the 24-month ITT analysis, mean Nankivell GFRs were 53.7 and 52.1 mL/min, respectively, a difference of 1.6 mL/min in favor of SRL conversion that was not significant (95% CI, −1.43–4.6; P=0.301).

The remaining 87 of 830 patients (10.5%) had a GFR of 20 to 40 mL/min at randomization. Enrollment in this stratum was halted by the Drug Safety Monitoring Board when, during a protocol-specified review of data, the primary safety endpoint of AR, graft loss, or death was reached by 8 of 48 (16.7%) of SRL conversion patients and 0 of 25 of CNI continuation patients (P=0.045). In this stratum, the primary (ITT) analysis at 12 months showed mean Nankivell GFRs of 24.6 and 27.2 mL/min (SRL conversion and CNI continuation, respectively); a numeric difference (−2.6 mL/min) in favor of CNI continuation (95% CI, −12.27–6.91; P=0.575). At 24 months, mean GFRs were 21.7 and 17.9 mL/min, respectively; a 3.8 mL/min difference in favor of SRL conversion (95% CI −8.66–16.35; P=0.538).

For patients with baseline GFR more than 40 mL/min who remained on assigned therapy, mean GFR was significantly higher after SRL conversion at all time points between 12 and 24 months (Fig. 2).

Mean Nankivell GFR (mL/min) in on-therapy patients with baseline GFR more than 40 mL/min.*

Figure 3(A) shows the binary ITT analysis (percentage of patients achieving ≥5.0, ≥7.5, or ≥10.0 mL/min increment in Nankivell GFR from baseline) at 24 months in patients with baseline GFR more than 40 mL/min. At 24 months, a significantly higher percentage of SRL conversion patients achieved the designated GFR improvement compared with those allocated to CNI continuation. This analysis was also significant at 12 months (data not shown).

Percentage of patients showing clinically meaningful improvements in GFR at 24 months. (A) ITT analysis of patients with Nankivell GFR more than 40 mL/min. (B) ITT analysis of subgroup of patients with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11.

In the subgroup of patients with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11, mean Nankivell GFR was significantly higher in SRL conversion versus CNI continuation cohorts at both 12 months (66.2 vs. 60.1 mL/min, 95% CI, 2.02–10.19; P=0.004) and 24 months (63.8 vs. 59.0 mL/min, 95% CI, 0.01–9.57; P=0.049). In this subgroup, binary ITT analysis of Nankivell GFR also demonstrated clinically meaningful improvement in significantly higher percentages of SRL conversion versus CNI continuation patients at 24 months (Fig. 3B).

Multiple linear regression analysis showed that baseline glomerular disease (primarily focal segmental glomerulosclerosis; P<0.001), baseline GFR (P<0.001), donor age more than 55 years (P<0.001), UPr/Cr more than 0.11 (P<0.001), baseline total sum score (P<0.001), black race (P=0.001), and delayed graft function (P=0.023) were independent predictors of change from baseline GFR at 2 years. Of these, black race and UPr/Cr more than 0.11 were specific for SRL conversion whereas donor age more than 55 years, delayed graft function, or presence of glomerular disease or Banff Total Sum Score in baseline biopsies were treatment-independent effects. Despite its incorporation into the dependent variable, GFR at baseline still had an effect on change from baseline GFR at 24 months.

Primary Safety Endpoint

The overall rates for the primary safety endpoint (composite of first occurrence of BCAR, graft loss, or death at 12 months) were not significantly different (95% CI, −0.6–5.6; P=0.135; Table 2). In general, these rates were higher for patients with baseline GFR 20 to 40 mL/min and lower for GFR stratum more than 40 mL/min and for the subgroup with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11. This trend was similar at 24 months (Table 2).

Number (%) of patients who met composite and individual safety endpoints at 12 and 24 months after randomization

Biopsy-Confirmed Acute Rejection

The incidence of BCAR was similar for SRL conversion and CNI continuation cohorts at 12 and 24 months after randomization (Table 2). In each treatment group, BCAR rates were lower with baseline GFR more than 40 mL/min and also for the subgroup with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11 than in the stratum with baseline GFR 20 to 40 mL/min.

Graft loss after BCAR occurred in one SRL conversion patient with a protocol violation who failed to return for required follow-up. Graft loss also followed presumptive AR from suspected noncompliance in one CNI continuation patient. Four (4) patients whose protocol-mandated biopsies revealed Banff grade IA or IB AR responded to increased immunosuppression: increased corticosteroids were prescribed for one SRL conversion and two CNI continuation patients; MMF dosage was increased in a second SRL conversion patient. In the remaining patients for whom BCAR was identified in the protocol-mandated week 104 biopsies, immunosuppression was not altered.

Graft and Patient Survival

No significant differences in graft survival or patient survival were observed between treatment groups at both 12 and 24 months (Table 2). In each cohort, graft survival was higher with baseline GFR more than 40 mL/min and also in the subgroup with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11 compared with baseline GFR 20 to 40 mL/min.

Patient survival was not significantly different through 2 years (Table 2). In each cohort, survival was higher for baseline GFR more than 40 mL/min and in the subgroup of patients with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11 (Table 2).


At 12 months, a significantly higher percentage of SRL conversion versus CNI continuation patients discontinued treatment (15.7% vs. 9.5%, respectively, P=0.013). However, by 24 months the difference was no longer statistically significant (25.8 vs. 20.0%, P=0.070). Discontinuation rates among the subgroup of patients with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11 were similar through 24 months (12.8% vs. 13.7%, P=0.857, SRL conversion vs. CNI continuation, respectively).

Treatment-Emergent Adverse Events

By 24 months, at least one investigator-reported treatment-emergent adverse event (TEAE) had been reported for most patients: 98.2% and 94.9% of SRL conversion and CNI continuation patients, respectively (P=0.014). Events that occurred significantly more often among SRL conversion than CNI continuation patients included hyperlipemia (hypertriglyceridemia), hypercholesterolemia, hyperglycemia, diarrhea, anemia, peripheral edema, fever, albuminuria, acne, thrombocytopenia, leukopenia, skin rash, and increased lactate dehydrogenase (Table 3). Hyperuricemia occurred significantly more often among CNI continuation patients. The overall pattern of investigator- reported TEAEs was consistent with the known safety profile of SRL and the CNIs, CsA and tacrolimus, except for the rate of hyperglycemia after conversion. As was true for all TEAEs, hyperglycemia was reported at the discretion of the investigator; no numerical or other criteria were prospectively defined for this event. Moreover, the frequency of diabetes mellitus as a TEAE (4.7% vs. 4.4%, P=1.000), and adjusted mean fasting blood glucose concentrations (5.5 vs. 5.4 mM/L, P=0.327) were not significantly different between SRL conversion and CNI continuation groups, respectively. Finally, the higher rate of TEAEs was only observed in SRL conversion patients during the first 6 months after randomization (Fig. 4). Thereafter and through 24 months, the overall rate of de novo TEAEs was lower after SRL conversion (Fig. 4).

Treatment-emergent adverse events in the safety population
Rates of investigator-reported TEAEs.

The rate of infection-related TEAEs was significantly higher at 24 months after SRL conversion (77.5% vs. 67.0%, P=0.002). These events included pneumonia, acne, presumptive herpes simplex, fever, and aphthous stomatitis, and stomatitis both of presumed infectious origin (Table 3). The higher infection rate in SRL conversion patients occurred primarily during the first few months after randomization. Specifically, between 0 and 6 months rates were significantly higher (59.3% vs. 39.6%, SRL conversion vs. CNI continuation, respectively, P<0.001); thereafter, rates were nearly identical through 24 months (58.8% vs. 56.4%, respectively, P=0.549).

The incidence of stomatitis was significantly higher for SRL conversion versus CNI continuation from randomization through 24 months (34.1% vs. 9.9%, P<0.001). However, most events were reported during the first 6 months (29.6% vs. 3.7%, P<0.001); from 6 through 24 months, stomatitis rates for SRL conversion patients were markedly lower (4.5%) and numerically lower versus CNI continuation (6.2%, P=0.315).

Overall malignancy rates, including the incidence of skin carcinomas, were significantly lower in SRL conversion versus CNI continuation through 24 months (ITT analysis, P<0.001, Table 3); the incidence was also significantly lower when patients with a prior history of malignancy at baseline were excluded (2.8% vs. 8.0%, P=0.002).

Laboratory Determinations

Of 743 patients randomized with baseline GFR more than 40 mL/min, baseline UPr/Cr was reported for 620 patients (74.7%). Mean and median UPr/Cr values for patients remaining on therapy are shown in Table 4. Baseline mean and median UPr/Cr values were similar between SRL conversion and CNI continuation patients. After randomization, these values increased among SRL conversion compared with CNI continuation patients (Table 4), primarily between baseline and 6 months, whereas median values were only slightly above the upper limit of normal (≤0.2). A categorical analysis stratified by baseline UPr/Cr showed that among SRL conversion and CNI continuation patients with UPr/Cr less than or equal to 0.2 at baseline, 72.9% and 87.3%, respectively, had UPr/Cr less than or equal to 0.5 at 24 months; among those with baseline UPr/Cr more than 1.0, a higher percentage of SRL conversion patients had UPr/Cr more than 1.0 at 24 months (66.7% vs. 57.1%, respectively, P<0.001 for stratified analysis). An unstratified analysis among patients with baseline and month 24 UPr/Cr revealed a significantly lower percentage of SRL conversion than CNI continuation patients had UPr/Cr less than or equal to 0.5 at 24 months (64.2% vs. 78.8%, respectively) and a higher percentage had UPr/Cr more than 1.0 (20.9% vs. 10.3%, respectively; P=0.001).

Urinary protein/creatinine ratios (UPr/Cr) by treatment in patients with baseline GFR >40 mL/min

Mean on-therapy fasting lipid profiles were similar between groups at baseline. At one month, baseline-adjusted mean fasting serum total cholesterol levels were significantly higher among SRL conversion patients, peaking at month 2, then declining through month 24 such that corresponding differences between groups decreased from 1.31 mmol/L to 0.96 mmol/L (50.7 mg/dL to 33.3 mg/dL). Similar patterns were observed for low- and high-density lipoprotein cholesterol. At 24 months, adjusted mean low-density lipoprotein levels remained significantly higher (3.19 vs. 2.63 mmol/L), as did those for high-density lipoprotein (1.43 and 1.32 mmol/L, SRL conversion and CNI continuation, respectively [both P<0.001]). A similar pattern was observed for fasting triglyceride levels; at 24 months, adjusted mean values were 2.61 versus 1.77 mmol/L (SRL conversion vs. CNI continuation, respectively, P<0.001). In the GFR more than 40 mL/min stratum, fasting lipid profiles were similar to those for the overall treatment groups (data not shown).

A higher percentage of SRL conversion versus CNI continuation patients received lipid-lowering therapy; at 24 months, rates were 77.7% and 54.6%, respectively (P<0.001). The most frequently prescribed drugs were HMG-CoA reductase inhibitors (74.4% and 50.9%, SRL conversion and CNI continuation patients, respectively, P<0.001).

Mean hemoglobin concentrations were similar between groups at baseline but decreased after SRL conversion and remained significantly lower from 1 to 6 months. Thereafter, the differences were no longer significant; at 24 months, mean hemoglobin concentrations were 131.9 and 132.0 g/L (SRL conversion and CNI continuation, respectively, P= 0.955). Erythropoietic drug use was similar at baseline (4.7% and 4.8%, SRL conversion and CNI continuation, respectively), increased after SRL conversion to 14.7% at month 8, then remained relatively stable to 24 months (12.7% vs. 5.6% for CNI continuation).

Blood Pressure

Systolic blood pressure was significantly lower among SRL conversion patients at month 1 (132.4 vs. 135.5 mm Hg, P=0.008), as was diastolic blood pressure at 1, 2, 3 (P<0.001), and 19 (P=0.041) months after conversion. At 24 months, mean systolic blood pressure was 132 mm Hg in both groups (P=0.922); mean diastolic blood pressure was 79.9 and 80.9 mm Hg for SRL conversion and CNI continuation patients, respectively (P=0.225).


This is the first large, prospective, randomized, clinical trial assessing the safety and efficacy of converting maintenance renal allograft recipients from CNI- to SRL-based immunosuppression. The protocol eligibility criteria were designed to be broadly inclusive (patient age, 13 to 75 years; time since transplantation, 5 to 162 months; baseline GFR, 20 to more than 80 mL/min; preexisting histopathology, ranging from no CAN to severe [grade III] disease; no exclusion for de novo or recurrent glomerular disease and CNI toxicity).

The study results at 1 and 2 years showed no significant differences in primary safety outcomes. Enrollment in the GFR 20 to 40 mL/min stratum was halted prematurely, however, because of a higher incidence of safety endpoints among the SRL conversion patients in this stratum.

ITT analysis of the primary efficacy endpoint—change from baseline Nankivell GFR one year after randomization—failed to demonstrate a statistically significant advantage to SRL conversion over CNI continuation. This lack of difference may be partly attributable to the broad spectrum of severity of preexisting renal parenchymal injury in the study population. Additionally, the stabilization or improvement in renal function observed among CNI continuation patients was unexpected, an observation attributable to the closer monitoring of renal function and CNI blood levels than might otherwise have been the case had they not been participating in a controlled, clinical trial.

The 40-mL/min cutoff for the prospectively defined baseline GFR stratum was based on the consensus recommendation of participating study investigators and was intended to distinguish patients whom they considered more or less likely to benefit from conversion to SRL-based immunosuppression. Nonetheless, the higher baseline GFR stratum still included patients with a broad spectrum of preexisting injury and levels of residual renal function and diverse comorbid conditions and causes of end-stage renal disease, some of which had recurred before enrollment in this study. Thus, it is not surprising that some patients benefited from conversion, whereas others did not. Although ITT analyses failed to show a significant improvement for GFR more than 40-mL/min stratum, a subset of SRL conversion patients did show improvement from baseline GFR, as demonstrated by the retrospectively defined binary outcome analysis of clinically meaningful GFR improvements for the overall stratum (Fig. 3A) and for the subgroup with baseline UPr/Cr less than or equal to 0.11 (Fig. 3B). Moreover, patients who continued to receive SRL as assigned therapy showed significantly greater improvements in renal function that persisted from 12 through 24 months after conversion (Fig. 2).

The post hoc multiple regression analysis showed a significant interaction between baseline urinary protein excretion and SRL conversion. Accordingly, in the SRL conversion subgroup with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11, the change in baseline-adjusted GFR was greater at 12 months than in the corresponding subgroup of CNI continuation patients and at a level (6 mL/min) that was both statistically significant and clinically important. Further, this significant treatment difference persisted through 24 months of follow-up at a level approaching 5 mL/min. Finally, in this subset of 297 patients, representing 40% of all patients with baseline GFR more than 40 mL/min, the safety profiles were similar for SRL conversion and CNI continuation. However, even in this subset, proteinuria increased and to a much greater extent than in those who remained on CNIs. The reasons for this may be hemodynamic but also may reflect podocyte injury and antagonism of vascular endothelial growth factor associated with SRL (12, 13).

Among patients converted to SRL, the pattern of adverse events was consistent with the known safety profile of SRL. During the first 6 months after randomization, a significantly higher rate of investigator-reported TEAEs was observed in SRL conversion versus CNI continuation patients that did not continue during the ensuing 18 months of follow-up (Fig. 4). This early imbalance was most likely attributable to the fact that subjects randomly assigned to SRL conversion underwent a major change in immunosuppression, whereas those allocated to CNI continuation did not. Significantly fewer treatment-emergent malignancies were reported after SRL conversion, a finding that emerged early in the study and persisted through 24 months. Others have reported similar findings in SRL-treated patients (6, 14). Although the reasons for this difference are not known, it may be speculated that the similarity in observed AR rates between CNI continuation and SRL conversion cohorts, and the slightly higher rates of infection after SRL conversion, suggest that the difference in malignancies is more consistent with a mammalian target of rapamycin-specific effect on tumor biology than a reduction in net immunosuppression.

An unexpected finding in this study was the development of de novo proteinuria and progression of preexisting proteinuria after SRL conversion. Patients enrolled with more severe renal parenchymal injury (i.e., higher UPr/Cr) were more likely to experience (death-censored) graft loss or death. Moreover, in both treatment groups, graft loss after randomization was associated with significantly higher baseline UPr/Cr and significantly lower baseline GFR. These findings are consistent with earlier studies in which lower GFR (15) and abnormally increased urinary protein excretion (16) were shown to be risk factors for increased morbidity and mortality in renal allograft recipients. Further, conversion has been shown to be more successful in patients with less proteinuria at baseline (17). To the extent that proteinuria and reduced GFR are surrogate markers for renal parenchymal injury, those patients with the greatest degree of preexisting injury appeared to be at higher risk for suffering further damage, regardless of whether they underwent conversion to SRL. In this regard, patients in the subgroup with baseline GFR more than 40 mL/min and UPr/Cr less than or equal to 0.11 showed substantially less of an increment in UPr/Cr after SRL conversion, had significantly better renal function, and experienced fewer investigator-reported TEAEs, deaths, and graft losses than did the overall SRL conversion cohort. Thus, the overall safety profile associated with conversion from CNI- to SRL-based immunosuppression appears to be more favorable when conversion is initiated before or at an early stage in the progression of chronic renal allograft damage (GFR more than 40 mL/min and normal urinary protein excretion). Whether or not conversion can be accomplished safely and effectively in patients with higher UPr/Cr must await further study.

In conclusion, patients with more severely injured renal allografts were less capable of tolerating this major change in immunosuppression than those with healthier transplants at the time of conversion. These observations support the argument for conversion from CNI-based immunosuppression to SRL before a renal allograft has sustained substantial, permanent renal parenchymal injury, as imputed by decreasing GFR and the development of a persistent pattern of proteinuria. The results from this clinical trial suggest the target population for conversion should be those with baseline GFR more than 40 mL/min and urinary protein excretion that is well within normal limits.


The authors recognize Christine Innes (Wyeth Research, Collegeville, PA) and Karine Rutault (Wyeth Research, Paris, France) for their contributions to study coordination and data review; Bernadette Maida (Wyeth Research, Collegeville, PA) for her contribution to study design and data review; and Susan A. Nastasee (Wyeth Research, Collegeville, PA) for assistance in preparation of the manuscript.


The members of Sirolimus CONVERT Trial Study group are as follows: J. Alberu, Instituto Nacional de Ciencias, Mexico City, Mexico; R.R. Alloway, University Hospital, Cincinnati, OH; E. Ancona, Clinica Chirurgica General III, University of Padua, Padua, Italy; M. Arias, Hospital Marques de Valdecilla, Santander, Spain; P. Bachleda, Fakultni Nemocnice, Olomouc, Czech Republic; J.A. Bertolatus, University of Iowa Hospitals and Clinics, Iowa City, IA; J.L. Bosmans, University Hospital Antwerpen, Edegem, Belgium; B. Bourbigot, CHU La Cavale Blanche, Brest, France; D.C. Brennan, Washington University School of Medicine, St. Louis, MO; K. Brinker [deceased], Dallas Transplant Institute, Dallas, TX; S. Campbell, Princess Alexandra Hospital, Wooloongabba, Australia; J.M. Campistol, Hospital Clinic i Provincial, Barcelona, Spain; D.B. Monteiro de Carvalho, Hospital Geral de Bonsucesso, Rio de Janeiro, Brazil; J.A. Castillo-Lugo, Dallas Transplant Institute, Dallas, TX; S. Cho, Boston Medical Center, Boston, MA; S. Cockfield, University of Alberta Hospitals, Edmonton, Alberta, Canada; D.J. Cohen, Columbia University Medical Center, New York NY; D.J. Conti, Albany Medical Center, Albany, NY; F.L. de Carvalho Contieri, Hospital Universitario Evangelico de Curitiba, Curitiba, Brazil; F.G. Cosio, Rochester Methodist Hospital, Rochester, MN; P. Daloze, Hopital Notre Dame, Montreal, Quebec, Canada; M. Davalos Michel, CEMIC, Buenos Aires, Argentina; C.L. Davis, University of Washington Medical Center, Seattle, WA; D. del Castillo, Hospital Universitario Reina Sofia, Cordoba, Spain; M. del Carmen Rial, Instituto de Nefrologia, Buenos Aires, Argentina; H. Diliz, Centro Medico Nacional 20 de Noviembre, Colonia del Valle, Mexico; J. Dominguez, Hospital Dr. Sotero del Rio, Santiago, Chile; D. Durand, Hopital Rangueil, Toulouse, France; F. Egidi, Methodist University Hospital, Memphis, TN; J. Eris, Royal Prince Alfred Hospital, Camperdown, Australia; B. Fellstrom, Njurmedicinska Kliniken, Uppsala, Sweden; A. Flores, Hospital de Especialidades Num. 1 Leon, Guanajuato, Mexico; L. Gaite, Clinica de Nefrologia y Urologia y Sante Fe, Argentina; V.D. Garcia, Santa Casa de Porto Alegre, Porto Alegre, Brazil; M Glyda, Szpital Wojewodzki, Poznan, Poland; I. Gonzalez, Hospital Fray Junipero Serra ISSSTE de Tijuana, Tijuana Baja California, Mexico; J.M. Gonzalez Posada, Hospital Universitario de Canarias–La Laguna, Sta. Cruz de Tenerife, Spain; P.F. Gores, Carolinas Medical Center, Charlotte, NC; M.V. Govani, Clarian Health Partners, Indiana University School of Medicine, Indianapolis, IN; C. Gracida, Hospital de Especialidades Centro Medico Nacional Siglo XXI, Mexico; J.M. Grinyo, C.S.U. Bellvitge, Barcelona, Spain; G.C. Groggel, University of Nebraska Medical Center, Omaha, NE; J Halkett, Groote Schuur Hospital, Cape Town, South Africa; K. Hamawi, King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia; A Holm, Hospital General Centro Medico Nacional La Raza, Mexico; I. Houde, Centre Hospitalier Univerisitaire de Quebec, Quebec, Quebec, Canada; D.E. Hricik, University Hospitals of Cleveland, Cleveland, OH; B. Hutchison, Sir Charles Gairdner Hospital, Nedlands, Australia; R. Jaramillo, Hospital Regional 1 de Octubre, Mexico; T.D. Johnston, University of Kentucky, Lexington, KY; J.F Juarez, Hospital de Especialidades 71 IMSS, Torreón Coahuila, Mexico; J. Kanellis, Monash Medicla Centre, Clayton, Australia; M. Klinger, Akademia Medyczna we Wroclawiu, Wroclaw, Poland; G. Knoll, Ottawa Hospital, Ottawa, Ontario, Canada; B. Kraemer, Klinikum der Universitat Regensburg, Regensburg, Germany; H. Kreis, Hopital Necker, Paris, France; M.R. Laftavi, Buffalo General Hospital, Buffalo, NY; D. Landsberg, St. Paul’s Hospital, Vancouver, British Columbia, Canada; C. Legendre, Hopital Saint-Louis, Paris, France; M. Lorber, Yale-New Haven Hospital, New Haven, CT; F-L. Luan, University of Michigan Hospital, Ann Arbor, MI; D. Ludwin, St. Joseph’s Healthcare, Hamilton, Ontario, Canada; E.C. Maggiora, Sanatorio de la Trinidad Mitre, Buenos Aires, Argentina; E. Mancilla, Instituto Nacional de Cardiologia, Tlalpan, Mexico; R. Manfro, Hospital das Clinicas de Porto Alegre, Porto Alegre, Brazil; R. Mann, UMDNJ/Robert Wood Johnson University Hospital, New Brunswick, NJ; P.U. Massari, Hospital Privado-Centro Medico de Cordoba, Cordoba, Argentina; A.J. Matas, Fairview University Medical Center, Minneapolis, MN; G. Mayer, Universitatsklinik fur Innere Medizin, Innsbruck, Austria; J. Morales, Clinica Las Condes, Santiago, Chile; J.M. Morales, Hospital 12 de Octubre, Madrid, Spain; A. Mota, Hospitais da Univerisdade de Coimbra, Coimbra, Portugal; N. Muirhead, London Health Sciences Centre, London, ON, Canada; S. Naicker, Johannesburg Hospital, Parktown, South Africa; G. Nanni, Cattedra di Chirurgia Sostitutiva e dei Trapianti, Rome, Italy; I.L. Noronha, Hospital Sao Joaquim da Beneficencia Portuguesa, Sao Paulo, Brazil; R. Oberbauer, Allgemeines Krankenhaus der Stadt Wien, Wien, Austria; P.J. O’Connell, Westmead Hospital, Westmead, Australia; S.A. Ojeda, Centro Médico Nacional de Occidente, Hospital de Pediatria, Guadalajara, Jalisco, Mexico; M. Ostrowski, Klinika Chirurgii Ogolnej i Transplantacyjnej Pomorskiej, Szczecin, Poland; L. Paczek, Klinica Immunologii, Warszawa, Poland; S. Palekar, Newark Beth Israel Medical Center, Newark, NJ; G.E. Palti, Hospital Aleman, Buenos Aires, Argentina; M.D. Pascoe, Groote Schuur Hospital, Cape Town, South Africa; J. Paul, Hospital Miguel Servet, Zaragoza, Spain; J.R. Pinto, Hospital Curry Cabral, Lisboa, Portugal; C. Ponticelli, Ospedale Maggiore di Milano, Milano, Italy; B. Pussell, Prince of Wales Hospital, Randwick, Australia; P.R. Rajagopalan, Medical University of South Carolina, Charleston, SC; R. Reyes, Clinica San Cosme, Aguascalientes, Mexico; D. Roth, University of Miami/Jackson Memorial Medical Center, Miami, FL; G. Russ, Queen Elizabeth Hospital, Woodville, Australia; J. Cavaliere Sampaio, Hospital Universitario Pedro Ernesto, Rio de Janeiro, Brazil; J. Sanchez-Plumed, Hospital Universitario La Fe, Valencia, Spain; F.P. Schena, Univeristy of Bari, Policlinico, Bari, Italy; R.O. Schiavelli, Hospital General de Agudos Dr. Cosme Argerich, Buenos Aires, Argentina; S. Schwartz-Sorensen, Nefrologisk afdelning, Kobenhavn, Denmark; N. Shah, Saint Barnabas Medical Center, Livingston, NJ; A. Shoker, St. Paul’s Hospital, Saskatoon, Saskatchewan, Canada; H. Tedesco Silva, Jr., Hospital do Rim e Hipertensao-Fundacao Oswaldo Ramos, Sao Paulo, Brazil; J.P. Squifflet, University Hospital St. Luc, Brussels, Belgium; S. Steinberg, Sharp Memorial Hospital, San Diego, CA; B. Suwelack, Medizinische Klinik und Poliklinik D Universitatsklinikum Muenster, Muenster, Germany; A. Vathsala, Singapore General Hospital, Singapore, Singapore; J. Vella, Maine Medical Center, Portland, ME; S. Vitko, IKEM Transplantcenter, Prague, Czech Republic; R. Wali, University of Maryland Medical System, Baltimore, MD; R. Walker, Royal Melbourne Hospital, Parkville, Australia; S. Weinstein, Lifelink Transplant Institute, Tampa, FL; J. Welchel, Piedmont Hospital, Atlanta, GA; J.A. Yamamoto, Hospital de Pediatria del Centro Medico Nacional Siglo XXI, Mexico; H.C. Yang, Pinnacle Health Systems, Harrisburg Hospital, Harrisburg, PA; M. Zand, University of Rochester-Strong Memorial Hospital, Rochester, NY.


1. Meier-Kriesche HU, 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. Nankivell BJ, Borrows RJ, Fung CL, et al. The natural history of chronic allograft nephropathy. N Engl J Med 2003; 349: 2326.
3. Paul LC. Chronic allograft nephropathy: An update. Kidney Int 1999; 56: 783.
4. Sehgal SN. Rapamune (RAPA, rapamycin, sirolimus): Mechanism of action immunosuppressive effect results from blockade of signal transduction and inhibition of cell cycle progression. Clin Biochem 1998; 31: 335.
5. Terada N, Lucas JJ, Szepesi A, et al. Rapamycin blocks cell cycle progression of activated T cells prior to events characteristic of the middle to late G1 phase of the cycle. J Cell Physiol 1993; 154: 7.
6. Campistol JM, Eris J, Oberbauer R, et al. Sirolimus therapy after early cyclosporine withdrawal reduces the risk for cancer in adult renal transplantation. J Am Soc Nephrol 2006; 17: 581.
7. Gonwa TA, Hricik DE, Brinker K, et al.; Sirolimus Renal Function Study Group. Improved renal function in sirolimus-treated renal transplant patients after early cyclosporine elimination. Transplantation 2002; 74: 1560.
8. Mota A, Arias M, Taskinen EI, et al. Sirolimus-based therapy after early cyclosporine withdrawal results in significantly better renal histology and function at 3 years following kidney transplantation. Am J Transplant 2004; 4(suppl 8): 566.
9. Oberbauer R, Hutchison B, Eris J, et al. Three-year assessment of health-related quality of life in sirolimus-treated kidney transplant patients after cyclosporine elimination. Am J Transplant 2003; 3(suppl 5): 215.
10. Oberbauer R, Segoloni G, Campistol JM, et al. Early cyclosporine withdrawal from a sirolimus-based regimen results in better renal allograft survival and renal function at 48 months after transplantation. Transpl Int 2005; 18: 22.
11. Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int 1999; 55: 713.
12. Izzedine H, Brocheriou I, Frances C. Post-transplantation proteinuria and sirolimus. N Engl J Med 2005; 353: 2088.
13. Mathieson PW. How much VEGF do you need [comment]? J Am Soc Nephrol 2006; 17: 602.
14. Kauffman HM, Cherikh WS, Cheng Y, et al. Maintenance immunosuppression with target-of-rapamycin inhibitors is associated with a reduced incidence of de novo malignancies. Transplantation 2005; 80: 883.
15. Meier-Kriesche HU, Baliga R, Kaplan B. Decreased renal function is a strong risk factor for cardiovascular death after renal transplantation. Transplantation 2003; 75: 1291.
16. Roodnat JI, Mulder PG, Rischen-Vos J, et al. Proteinuria after renal transplantation affects not only graft survival but also patient survival. Transplantation 2001; 72: 438.
17. Diekmann F, Budde K, Oppenheimer F, et al. Predictors of success in conversion from calcineurin inhibitor to sirolimus in chronic allograft dysfunction. Am J Transplant 2004; 4: 1869.

Sirolimus; Immunosuppression; Calcineurin inhibitor; Kidney transplantation

© 2009 Lippincott Williams & Wilkins, Inc.