A randomized, open label study of sirolimus versus CSA

A Randomized, Open-Label Study of Sirolimus VersusCyclosporine in Primary De Novo Renal

Allograft Recipients

Stuart M. Flechner,1,9 Alihan Gurkan,2 Anders Hartmann,3 Christophe M. Legendre,4

Graeme R. Russ,5 Josep M. Campistol,6 Francesco P. Schena,7 Carolyn M Hahn,8 Huihua Li,8

Joan M. Korth-Bradley,8 Sandi See Tai,8 and Seth L. Schulman8

Background. Despite a decreased incidence of acute rejection and early renal allograft loss due to calcineurin in-hibitors (CNIs) in transplant recipients, nephrotoxicity associated with long-term CNI use remains an importantissue. This study evaluated whether a CNI-free regimen, including sirolimus, mycophenolate mofetil, corticosteroids,and antiYinterleukin-2 receptor antibody induction, results in improved long-term renal function.Methods. This open-label, randomized, parallel group, comparative study in primary de novo renal transplant re-cipients was planned for 48 months but terminated early because of high acute rejection rates in the sirolimus arm.Results. Enrollment was stopped after ,12 months, with 475 transplanted patients randomized (2:1) to sirolimus(n=314) or cyclosporine A (CsA) treatment (n=161). Mean length of follow-up after transplantation was 190 days;this article focuses on available data through 6 months. MeanTSD on-therapy Nankivell-calculated glomerular fil-tration rate was not significantly different between the sirolimus (69.1T18.7 mL/min) and CsA (66.0T15.2 mL/min)treatment groups. Occurrence and length of delayed graft function was not significantly different between groups.Patients in the sirolimus group experienced numerically lower survival rates (96.9% vs. 99.4%; P=0.14), with ninedeaths reported with sirolimus and one with CsA; higher rates of biopsy-confirmed acute rejection (21.4% vs. 6.1%;PG0.001); and higher rates of discontinuations due to adverse events (17.4% vs. 6.8%; P=0.001).Conclusion. A sirolimus-based, CNI-free immunosuppressive regimen, when used with mycophenolate mofetil,corticosteroids, and antiYinterleukin-2 receptor antibody induction, was associated with high rates of biopsy-confirmed acute rejection compared with CsA-based immunosuppression and is not recommended.

Keywords: Acute allograft rejection, Delayed graft function, Sirolimus, Cyclosporine, Mycophenolate mofetil.

(Transplantation 2013;95: 1233Y1241)

Immunosuppressive agents such as the calcineurin in-hibitors (CNI) cyclosporine A (CsA) and tacrolimus have

reduced the incidence of acute rejection and early renal

allograft loss. A frequently used protocol for human leuko-cyte antigen (HLA)Ymismatched recipients consists of anantiYinterleukin-2 receptor antibody induction agent followedby a maintenance regimen of a CNI plus an antiproliferative

CLINICAL AND TRANSLATIONAL RESEARCH

Transplantation & Volume 95, Number 10, May 27, 2013 www.transplantjournal.com 1233

ClinicalTrials.gov Identifier: NCT00137345.This study was sponsored by Wyeth Pharmaceuticals, which was acquired by

Pfizer in October 2009. Medical writing support was provided by Bina J.Patel, PharmD, at Peloton Advantage, and was funded by Pfizer.

No author received an honorarium or other form of financial support related tothe development of this article. S.M.F. has participated on scientific advisoryboards for Novartis, Pfizer, and Tolera, is a member of a speaker program forNovartis, and has received research support from Pfizer and Tolera. A.G. hasserved on advisory boards for Novartis, Roche, and Wyeth (Pfizer). A.H. hasserved as a speaker for Novartis and Roche and has received research grantsfrom Roche. C.M.L. has served as a speaker for Alexion, Novartis, Pfizer, andRoche. G.R.R. has received honoraria from Astellas Pharma, Bristol-MyersSquibb, Novartis, and Wyeth (Pfizer). J.M.C. has no potential conflicts ofinterest to report. F.P.S. has participated on scientific advisory boards forNovartis and Wyeth (Pfizer) and has received research support from Astellasand Novartis. C.M.H., H.L., J.M.K.-B., S.S.T., and S.L.S. are employees ofPfizer (formerly Wyeth).

1 Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, OH.2 Currently, Acibadem University, Istanbul, Turkey; formerly, Akdeniz

University, Antalya, Turkey.3 Oslo University Hospital, Rikshospitalet, Oslo, Norway.4 Universite Paris Descartes and Hopital Necker, Paris, France.

5 Queen Elizabeth Hospital, Woodville South, Adelaide, Australia.6 Hospital Clinic i Provincial, Barcelona, Spain.7 University of Bari, Bari, Italy.8 Pfizer, Collegeville, PA.9 Address correspondence to: Stuart M. Flechner, M.D., Glickman Uro-

logical and Kidney Institute, Cleveland Clinic, and Cleveland ClinicLerner College of Medicine, Mail Code Q10-1, 9500 Euclid Avenue,Cleveland, OH 44195.

E-mail: [email protected]. and S.L.S. participated in the study conception and design. S.M.F.,

A.G., A.H., C.M.L., G.R.R., J.M.C., and F.P.S. are the study investigatorsand participated in the provision of patients and collection and assemblyof data. S.M.F., C.M.H., H.L., J.M.K.-B., S.S.T., and S.L.S. participated inthe data analysis and interpretation. S.M.F., C.M.H., and S.L.S. partic-ipated in the article preparation. All authors participated in the articlereview and revisions and final approval of article.

Received 23 January 2013.Accepted 16 February 2013.Copyright * 2013 by Lippincott Williams & WilkinsISSN: 0041-1337/13/9510-1233DOI: 10.1097/TP.0b013e318291a269

Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

mailto:[email protected]

agent and corticosteroids (CS). This treatment regimen hasresulted in a 1-year graft survival rate approaching 90% andacute rejection rates of 20% or less (1). However, tacrolimusand CsA are associated with complications, such as acute andchronic nephrotoxicity, as well as long-term effects on renalperfusion and glomerular and tubular function (2, 3).

Sirolimus, a mammalian target of rapamycin immu-nosuppressant, is indicated for the prophylaxis of organrejection in combination with CsA and CS, with CsA with-drawal recommended 2 to 4 months after transplantation inpatients with low to moderate immunologic risk (4). Severalstudies have indicated that CNI-free regimens comprisingsirolimus, mycophenolate mofetil (MMF), and CS mainte-nance therapy may be beneficial for long-term renal allograftfunction (5Y11). This study was designed to assess whether aCNI-free regimen, including sirolimus, MMF, CS, and anantiYinterleukin-2 receptor antibody, improves long-termrenal function without negatively impacting safety or immu-nosuppressive efficacy in de novo allograft recipients.

RESULTS

Trial TerminationThe study began in June 2005. Upon detection of an im-

balance in the incidence of acute rejection, the data monitoringcommittee met ad hoc in December 2005. A significantly greater

rate of rejection was noted in sirolimus-treated patients, andsirolimus trough levels were below the target range in 39% ofpatients 2 weeks after transplantation, prompting implementa-tion of Amendment 2, which increased the loading dose ofsirolimus. The imbalance in acute rejection rates continued de-spite implementation of Amendment 2, and in June 2006, thesponsor and data monitoring committee concluded that it wasinadvisable to continue the study. Patients were advised to re-sume best local immunosuppressive therapy, and data for themajority of patients beyond ,6 months could not be collected.

Patient DemographicsIn total, 487 patients were enrolled and randomized in a

2:1 ratio to sirolimus or CsA treatment; however, 12 patientswere excluded from all analyses because they were ineligible forinclusion (were not transplanted or withdrew consent beforetransplantation; Fig. 1). The remaining 475 patients comprisedthe intent-to-treat (ITT) population (defined as all patientswho were randomly assigned to study therapy and underwenttransplantation), with 314 patients receiving sirolimus and161 receiving CsA treatment. The safety population included471 patients who received at least one dose of study medication;of these, 151 patients receiving sirolimus and 87 patients re-ceiving CsA treatment were enrolled after the implementationof Amendment 2. In total, 127 patients receiving sirolimusand 79 receiving CsA completed 6-month treatment. Patients

FIGURE 1. Patient disposition.

1234 www.transplantjournal.com Transplantation & Volume 95, Number 10, May 27, 2013


remained on study for a mean of 190 days (range, 5Y441 days);thus, this article focuses on available data through 6 months.

Between-group patient characteristics were similar inthe ITT population. Mean age was 42.9 years (range, 14Y76days), 70% were males, and 78% were white (Table 1). Themost common etiologies for kidney disease were glomeru-lonephritis, hypertension, polycystic kidney disease, andimmunoglobulin A nephropathy. The overall distribution ofetiologies of end-stage renal disease did not differ signifi-cantly between treatments.

Mean donor age in both groups was approximately41 years. In both cohorts, the majority were white males,mean organ ischemia time was approximately 11 hr, andsources of most grafts were deceased donors (58.9% vs.41.1% living donors).

Before study termination, 31.3% of sirolimus and16.8% of CsA patients discontinued therapy (Fig. 1). Adverseevents (AEs) and unsatisfactory response (efficacy) were the

reported causes for discontinuation in significantly morepatients receiving sirolimus versus CsA (P=0.001 and 0.033,respectively).

Renal FunctionThe primary efficacy analysis of Nankivell-calculated

glomerular filtration rate (GFR) at 52 weeks was not per-formed owing to early study termination. Secondary analysisof Nankivell-calculated GFR at 6 months was performedbased on patients completing the assessment. No difference inmeanTSD calculated GFR was observed on ITT analysis (in-cluding on-therapy value and off-therapy value; sirolimus[n=185], 67.0 T19.0 mL/min; CsA [n=103], 67.2T14.7 mL/min;P=0.874). Although a numerical difference (3.13 mL/min) fa-voring sirolimus was observed, no significant difference in themeanTSD calculated GFR was observed between treatmentsfor on-therapy analysis (sirolimus [n=108], 69.1T18.7 mL/min;CsA [n=63], 66.0T15.2 mL/min; P=0.261).

TABLE 1. Baseline patient demographics (ITT population)

Characteristics Sirolimus (n=314) Cyclosporine (n=161) Total (n=475) P

Male sex, n (%) 218 (69.4) 116 (72.1) 334 (70.3) 0.554a

Ethnic origin, n (%)

White 245 (78.0) 127 (78.9) 372 (78.3) 0.291a

Black 25 (8.0) 13 (8.1) 38 (8.0)

Asian 16 (5.1) 6 (3.7) 22 (4.6)

Hispanic 13 (4.1) 4 (2.5) 17 (3.6)

Other 15 (4.8) 11 (6.8) 26 (5.5)

Mean (SE) age, years 42.9 (0.8) 42.7 (1.1) 42.9 (0.6) 0.876b

Donor organ source, n (%)

Deceased 187 (59.6) 93 (57.8) 280 (58.9) 0.826a

Living related 92 (29.3) 47 (29.2) 139 (29.3)

Living unrelated 35 (11.1) 21 (13.0) 56 (11.8)

Primary etiology of renal failure, n (%)

Autoimmune disease, systemic 5 (1.6) 5 (3.1) 10 (2.1) 0.450a

Diabetes mellitus 23 (7.3) 13 (8.1) 36 (7.6)

Glomerulonephritis 64 (20.4) 27 (16.9) 91 (19.2)

Hypertension 38 (12.1) 18 (11.3) 56 (11.8)

Immunoglobulin A nephropathy (Berger disease) 37 (11.8) 12 (7.5) 49 (10.3)

Interstitial nephritis/pyelonephritis 18 (5.7) 6 (3.8) 24 (5.1)

Obstructive uropathy/reflux 20 (6.4) 12 (7.5) 32 (6.8)

Polycystic kidney disease 28 (8.9) 28 (17.5) 56 (11.8)

Other 79 (25.2) 35 (21.7.0) 114 (24)

HLA mismatches, n (%)

0 12 (3.8) 5 (3.1) 17 (3.6) 0.663a

1 16 (5.1) 11 (6.9) 27 (5.7)

2 51 (16.2) 32 (20.0) 83 (17.5)

3 101 (32.2) 46 (28.8) 147 (31.0)

4 56 (17.8) 28 (17.5) 84 (17.7)

5 43 (13.7) 27 (16.9) 70 (14.8)

6 34 (10.8) 11 (6.9) 45 (9.5)

Mean (SE) PRA status, n 0.9 (0.3) 2.8 (1.4) 1.5 (0.5) 0.089b

a Fisher’s exact test P value (two-tailed).b One-way ANOVA with treatment as factor.HLA, human leukocyte antigen; ITT, intent-to-treat; PRA, panel-reactive antibodies; SE, standard error.

* 2013 Lippincott Williams & Wilkins Flechner et al. 1235


Graft and Patient SurvivalAt 6 months, graft survival was 94.7% and 96% in the

sirolimus and CsA groups, respectively; death-censored graftsurvival was 96.2% and 96.6%, respectively (Fig. 2A). Themost common graft loss causes were graft vessel thrombosisand rejection.

Six-month patient survival was 96.9% in sirolimus-treated patients and 99.4% in CsA-treated patients (P=0.14;Fig. 2B). During the trial, nine deaths occurred in thesirolimus group and one in the CsA group. All but one death(in the sirolimus group) occurred in the first 6 months.Causes in the sirolimus group included pulseless electricalactivity (n=1; day 211), cardiac arrest (n=1; day 43), sepsis(n=1; day 80), cardiovascular insufficiency (n=1; day 31),cerebral abscess (n=1; day 106), organ failure/sepsis (n=2;days 32 and 140), feculent peritonitis (n=1; day 67), andunknown (n=1; day 79). Cause of death in the CsA-treatedpatient was thrombotic thrombocytopenic purpura.

All deaths (nine in the sirolimus group and one in theCsA group) occurred in patients who received deceased-donor organs. Only two graft losses occurred in patients onsirolimus who received living-donor organs. All others oc-curred in deceased-donor kidney recipients.

Delayed Graft FunctionThe incidence of delayed graft function (DGF; defined

as the need for dialysis within 7 days after transplantation) wascomparable between the sirolimus (20.4%) and CsA (22.4%)groups. Median time to recovery from DGF was similar insirolimus-treated patients (7.5 days) and CsA-treated patients(7.0 days).

Biopsy-Confirmed Acute RejectionOverall biopsy-confirmed acute rejection (BCAR)Yfree

survival was significantly greater in CsA-treated patients versussirolimus-treated patients (PG0.001; Fig. 2C). At 6 months,BCAR-free survival was 78.6% and 93.9% in the sirolimus andCsA groups, respectively. Sixty patients in the sirolimus groupand 8 in the CsA group experienced BCAR during the first6 months after transplantation. Another sirolimus grouppatient experienced BCAR after 6 months. PostamendmentBCAR-free survival remained significantly greater with CsAversus sirolimus (P=0.008) despite an increase in the sirolimusload dose to therapeutic trough level; 17 of 314 (5.4%)sirolimus-treated patients and 2 of 161 (1.2%) CsA-treatedpatients developed moderate or severe BCAR while on study.

BCAR rates appeared to be higher in the sirolimusgroup versus the CsA group regardless of whether or notpatients received living-donor or deceased-donor kidneys.

Adverse EventsA significantly greater proportion (P=0.001) of sirolimus-

treated patients (n=54 [17.4%]) discontinued because ofAEs versus CsA-treated patients (n=11 [6.8%]). Through6 months, 460 (97.7%) patients reported treatment-emergentAEs (TEAEs) during the study (sirolimus, 306 [98.7%]; CsA,154 [95.7%]; P=0.051; Table 2). A significantly greater per-centage of sirolimus-treated patients experienced albumin-uria, diarrhea, hyperlipidemia, hypokalemia, thrombocytopenia,and transplant rejection, whereas a significantly greater percent-age of CsA-treated patients reported gum hyperplasia, hirsutism,hyperuricemia, and hypomagnesemia.

At 6 months, infection-related TEAEs occurred in 238(51%) patients, the most common being general infection(sirolimus, 17%; CsA, 24%) and urinary tract infection(sirolimus, 21%; CsA, 21%; Table 2).

FIGURE 2. Death-censored graft survival (A), patientsurvival (B), and BCAR-free survival (C) through 6 months(ITT population).



Three (0.6%) patients developed a treatment-emergentmalignancy (one sirolimus-treated patient [bladder carcinoma]and two CsA-treated patients [epithelioid hemangioendothe-lioma and basal/squamous cell carcinomas]). Wound healingcomplications occurred more frequently with sirolimus (15.2%)versus CsA (8.2%; P=0.033; ITT population).

Serious AEs (SAEs) were reported more often insirolimus-treated patients than CsA-treated patients (57.1% vs.47.2%; P=NS). The most common SAEs were abnormalhealing, diarrhea, fever, increased creatinine, infection, kidneytubular necrosis, lymphocele, transplant rejection, and urinarytract infection. Transplant rejection (18.1% vs. 4.3%; PG0.001)

TABLE 2. TEAEs with an incidence of Q10% and infection-related AEs with an incidence of Q2% in either group (6-monthposttransplantation data; safety population)

Sirolimus (n=310) Cyclosporine (n=161) Total (n=471)

TEAEs occurring in Q10% of patients

Fever 58 (18.7) 22 (13.7) 80 (17.0)

Headache 35 (11.3) 11 (6.8) 46 (9.8)

Pain 41 (13.2) 11 (6.8) 52 (11.0)

Transplant rejectiona 59 (19.0) 9 (5.6) 68 (14.4)

Hypertension 84 (27.1) 45 (28.0) 129 (27.4)

Constipation 46 (14.8) 28 (17.4) 74 (15.7)

Diarrheaa 93 (30.0) 21 (13.0) 114 (24.2)

Liver function tests abnormal 31 (10.0) 11 (6.8) 42 (8.9)

Nausea 46 (14.8) 26 (16.1) 72 (15.3)

Vomiting 34 (11.0) 26 (16.1) 60 (12.7)

Anemia 133 (42.9) 54 (33.5) 187 (39.7)

Leukopenia 39 (12.6) 11 (6.8) 50 (10.6)

Thrombocytopeniaa 36 (11.6) 4 (2.5) 40 (8.5)

Creatinine increased 67 (21.6) 34 (21.1) 101 (24.4)

Healing abnormal 45 (14.5) 14 (8.7) 59 (12.5)

Hypercholesteremia 79 (25.5) 35 (21.7) 114 (24.2)

Hyperglycemia 37 (11.9) 22 (13.7) 59 (12.5)

Hyperlipemiaa 134 (43.2) 46 (26.1) 176 (37.4)

Hypokalemiaa 45 (14.5) 9 (5.6) 54 (11.5)

Hypophosphatemia 52 (16.8) 18 (11.2) 70 (14.9)

Lactate dehydrogenase increased 39 (12.6) 18 (11.2) 57 (12.1)

Peripheral edema 87 (28.1) 46 (28.6) 133 (28.2)

Acne 49 (15.8) 15 (9.3) 64 (13.6)

Albuminuriaa 67 (21.6) 18 (11.2) 85 (18.0)

Dysuria 34 (11.0) 12 (7.5) 46 (9.8)

Hematuria 37 (11.9) 14 (8.7) 51 (10.8)

Kidney tubular necrosis 35 (11.3) 12 (7.5) 47 (10.0)

Local reaction 71 (22.9) 36 (22.4) 107 (22.7)

Treatment-emergent infection-related AEs occurring in Q2% of patients

General infection 54 (17.4) 39 (24.2) 93 (19.7)

Urinary tract infection 65 (21.0) 34 (21.1) 99 (21.0)

Pneumonia 8 (2.6) 0 (0) 8 (1.7)

Herpes simplex 15 (4.8) 7 (4.3) 22 (4.7)

Upper respiratory infection 12 (3.9) 6 (3.7) 18 (3.8)

Sepsis 12 (3.9) 2 (1.2) 14 (3.0)

Fever 7 (2.3) 2 (1.2) 9 (1.9)

Diarrhea 7 (2.3) 1 (0.6) 8 (1.7)

Vaginitis 2 (2.1) 0 (0) 2 (1.4)

Oral moniliasis 6 (1.9) 5 (3.1) 11 (2.3)

Fungal dermatitis 3 (1.0) 4 (2.5) 7 (1.5)

aPe0.005 for sirolimus versus cyclosporine treatment groups.AE, adverse effects; TEAE, treatment-emergent adverse effects.



and diarrhea (6.8% vs. 1.2%; P=0.006) were SAEs that weresignificantly more common in sirolimus-treated patients.

Laboratory Determinations, Vital Signs, andConcomitant Treatments

MeanTSD sirolimus (14.9T15.9 ng/mL) and CsA(244T88 ng/mL) time-normalized minimum concentration(Cmin,TN) values were within recommended target ranges, al-though there was considerable variability, and many patientshad individual trough concentration values outside the targetrange. In CsA-treated patients versus sirolimus-treated patients,meanTSD mycophenolic acid (MPA) concentrations were lower(2.25T3.37 vs. 3.65T3.13 ng/mL, respectively) and mean myco-phenolic acid glucuronide (MPAG) concentrations were higher(95.1T60.4 vs. 72.9T61.2 ng/mL, respectively). Table 3 shows nodifference in sirolimus or CsA Cmin,TN values for those whohad rejection, graft loss, or death compared with those whodid not, and neither sirolimus nor CsA Cmin,TN was a signifi-cant factor for rejection, graft loss, or death. Similarly, nodifference was noted in the MPA or MPAG Cmin,TN valuesin those who experienced rejection, graft loss, or death com-pared with those who did not, and MPA or MPAG Cmin,TN

was not a significant factor for rejection, graft loss, or death.Fasting blood levels of total, high-density lipoprotein,

and low-density lipoprotein cholesterol and triglycerides werehigher among sirolimus-treated patients versus CsA-treatedpatients at most study visits; a significantly greater percent-age of sirolimus-treated patients received lipid-loweringagents after transplantation. Overall, 288 (61.1%) patientsreceived lipid-lowering agents on study, with 64.8% receivingsirolimus versus 54.0% receiving CsA (P=0.028).

At 6 months, mean systolic blood pressure was com-parable between groups; however, diastolic blood pressurewas significantly higher in CsA-treated patients. A compa-rable number of patients in the sirolimus (88.1%) and CsA(93.2%) groups received at least one antihypertensive med-ication on study.

At 6 months, mean blood hemoglobin and hematocritvalues were similar between groups. Recombinant erythro-poietin use decreased after transplantation in both treat-ment groups but was significantly greater with sirolimus(32.9%) versus CsA (16.1%; PG0.001).

After transplantation, the percentage of patients receiv-ing insulin increased and was similar between the sirolimus

(24.2%) and CsA groups (24.8%). Although post transplan-tation diabetes mellitus incidence, defined as 30 consecutivedays of new insulin use, was not assessed, the investigator-reported rate of treatment-emergent diabetes mellitus wassimilar between the sirolimus and CsA groups (5.2% and5.0%, respectively), as was the rate of hyperglycemia (12.6%and 13.7%).

DISCUSSIONEarly in the study, a significantly increased BCAR rate

was observed in the group receiving sirolimus. Despite at-tempts to enhance immunosuppression by increasing theloading dose, no improvement was seen and enrollment wasterminated. The recent ORION study also reported increasedacute rejection rates in patients treated with sirolimus andMMF, leading to sponsor termination of that group (12). Inthe ORION study, these outcomes may be attributable in partto sirolimus trough levels being below the target range in 56%of patients in the affected group. In this study, 39% of pa-tients had below-target sirolimus trough levels 2 weeks aftertransplantation. Despite amending the protocol to achievetarget trough levels, increased rates of BCAR persisted. Therelationship between sirolimus trough concentrations andoutcome is not easily interpreted. Although many patientshad below-target individual trough concentrations, a com-parison of Cmin,TN that included all observed trough concen-trations showed no difference in Cmin,TN between rejectionand nonrejection, nor was Cmin,TN a risk factor associatedwith increased rejection rates. Sirolimus-based immunosup-pression was also associated with numerically lower rates ofpatient survival.

In this study, patients in the sirolimus group experi-enced numerically higher GFR during on-therapy periods.The lack of superior renal function in the ITT populationmay be attributable to the replacement of a CNI drug in mostof the sirolimus discontinued group. Furthermore, althoughprevious studies have shown an increase in the duration ofDGF associated with the use of sirolimus (13Y15), the use ofsirolimus-based immunosuppression was not associated witha definitive impact on the frequency or duration of DGFcompared with those receiving CNI treatment in this study.Other findings included impaired wound healing during theearly perioperative period, lower mean hemoglobin and he-matocrit levels, and higher fasting lipid levels consistent with

TABLE 3. Mean time-normalized trough concentration of drug 6 months after transplantation in the presence or absenceof an event

Rejection Graft Loss Death

Yes No Yes No Yes No

Drug MeanTSD Cmin,TN (ng/mL)

Sirolimus 13.9T4.9 (61)a 14.4T4.9 (249) 15.3T6.2 (14) 14.2T4.6 (296) 14.0T4.9 (9) 14.3T4.7 (301)

Cyclosporine 246T35 (8) 225T75 (154) 183T32 (5) 228T74 (156) 126 (1) 227T74 (161)

MPA 3.32T4.05 (45) 3.03T4.64 (295) 2.82T3.40 (9) 3.02T4.41 (336) 1.98T1.08 (6) 3.04T4.41 (340)

MPAG 65.9T52.0 (45) 78.5T81.6 (291) 138T117 (9) 66.2T54.3 (332) 107T69 (6) 67.6T57.4 (336)

a Number of observations used to calculate the mean.MPA, mycophenolic acid; MPAG, mycophenolic acid glucuronide; SD, standard deviation.



the known sirolimus safety profile (6, 8Y10). The study wasnot long enough to permit effective evaluation of cancer ratesin either treatment group. Some of the TEAEs recordedduring the study (e.g., anemia, diarrhea, and thrombocyto-penia) may have been exacerbated by the concomitant use ofMMF, as reported in previous studies with sirolimus andMMF (7, 16).

Lower-than-expected rates of rejection seen in CsA-treated patients may be attributed to the immunologic pro-file of the study population, which was of low to moderaterisk, or to the use of 2-hr CsA concentration monitoring,which was permitted in the trial but was not measured, aswell as the open-label nature of the study. This may haveprompted investigators to perform renal biopsies less fre-quently in patients receiving the standard of care. In addi-tion, the higher-than-expected CsA trough levels (6-monthCmin,TN of 244T88 ng/mL) may have provided more protec-tion against acute rejection than seen in previous studies (17).

CNI withdrawal has been documented in previousstudies to be accompanied by an increased acute rejectionrate due to preexisting subclinical rejection and/or low MPAexposure in previous studies (18Y20). In the present study,however, there was no difference between the MPA troughvalues in those who had rejection and who did not.

Early phase II data from a study examining CNI avoid-ance with the use of sirolimus+MMF and CS in primary kid-ney transplant recipients showed a 6.4% BCAR rate at 1 year insirolimus-treated patients when sirolimus trough levels wereoptimized to 10 to 12 ng/mL (5). A BCAR rate of approxi-mately 10% was confirmed by subsequent trials (7, 21Y23).However, similar to those in the present study, higher-than-expected BCAR rates were seen in the Efficacy Limiting Toxic-ity EliminationYSymphony study when sirolimus was targetedto initial blood levels of 4 to 8 ng/mL (24). One-year BCAR ratesin the low-dose sirolimus group (37.2%) were significantlyhigher than in other treatment groups (standard-dose CsA, low-dose CsA, and low-dose tacrolimus) (24). Although the rate ofBCAR in this trial (19%) was higher with sirolimus than in thecontrol arm, it was nevertheless lower than the 22% reported in aCNI avoidance trial using belatacept, MMF, and steroids withbasiliximab induction (25). In the Belatacept Evaluation ofNephroprotection and Efficacy as First-Line Immunosuppres-sion Trial, long-term renal function was superior despite higherinitial BCAR rates. It is not known whether superior renalfunction would have been detected beyond 1 to 2 years in thepresent study.

In summary, when used from the time of transplan-tation in association with basiliximab, MMF, and CS, asirolimus-based, CNI-free immunosuppressive regimen wasassociated with a high rate of BCAR and one that was sig-nificantly higher when compared with CsA-based immu-nosuppression and is not recommended.

MATERIALS AND METHODS

Study DesignThis open-label, randomized, parallel group, comparative study was

conducted in primary de novo renal transplant recipients at 68 centers

worldwide. Patients who underwent transplantation were stratified prospec-

tively based on race and donor source; screening and baseline evaluations

were performed within 7 days before transplantation. Patients were randomly

assigned (2:1) by a computerized randomization/enrollment system to receive

sirolimus or CsA, in combination with basiliximab, MMF, and CS, stratified

by race and donor origin. The study was planned for a 104-week treat-

ment period with 104 weeks of follow-up. Because of an imbalance in

BCAR rates between treatment groups, an amendment was implemented

in December 2005, changing the loading dose and overall dosing of

sirolimus for the sirolimus group. Despite this change, the study was

terminated early owing to additional events of acute rejection in

sirolimus-treated patients. Patients were followed for 2 additional months

to collect and report AEs.

The protocol for this study, as well as subsequent amendments, received

institutional review board and independent ethics committee approval. The

study was conducted according to the principles of the Declaration of

Helsinki and the International Conference on Harmonisation guidelines for

Good Clinical Practice in the European Community.

Eligibility CriteriaEligible patients (age 913 years and weight 940 kg) included those with

end-stage renal disease who received a primary renal allograft from a de-

ceased donor, living unrelated donor, or HLA-mismatched living related

donor. Eligible patients had white cell count 4000/mm3 or more, platelet

count 100,000/mm3 or more, cholesterol less than 300 mg/dL, and tri-

glycerides less than 350 mg/dL. Excluded patients received kidneys from

deceased donors older than 60 years, living donors older than 65 years, or

HLA-identical living related donors. Patients with previous solid organ

transplant, active infection, unstable angina, malignancy, multiple-organ

transplants, donor with HLA panel-reactive antibody more than 20%, do-

nor kidney ischemia time more than 30 hr, donation after cardiac death,

positive B-cell or T-cell crossmatch, ABO blood group incompatibility with

the allograft, or body mass index more than 30 kg/m2 were also ineligible.

EndpointsThe primary efficacy endpoint was renal function measured by Nankivell-

calculated GFR, planned for 12 months using ITT analysis. Because the study

was terminated early, this analysis was not performed. Efficacy endpoints an-

alyzed were renal function at 6 months, as measured by calculated GFR

(Nankivell), severity of BCAR, and incidence of BCAR. A subanalysis was

performed examining the incidence of BCAR before and after the implemen-

tation of the study’s second protocol amendment. Acute rejection was confirmed

by biopsy (Banff ’97 grades IYIII or antibody mediated), as interpreted by the

local pathologist.

The primary safety endpoint was the composite of the incidence of the first

occurrence of graft loss or death. Graft loss was defined as more than 56 days of

continuous dialysis, nephrectomy, retransplantation, or death with a function-

ing graft. Additional safety endpoints reported here include graft survival, pa-

tient survival, mean systolic and diastolic blood pressures, incidence of infection,

malignancy, DGF, wound-healing complications, TEAEs, posttransplantation

diabetes mellitus, anemia and use of recombinant erythropoietic agents, lipid-

lowering agents, and antihypertensive medications.

Treatment RegimensIn the original study design, patients in the sirolimus group received a 10

to 15 mg oral loading dose within 48 hr after transplantation followed by

4 to 8 mg daily until whole-blood sirolimus trough levels were 10.0 ng/mL

or more (as measured by chromatographic methods). Subsequent doses

were adjusted to maintain trough levels within the protocol-specified range

(study start to week 13, 10Y15 ng/mL; weeks 14Y26, 8Y12 ng/mL; weeks

27Y104, 5Y12 ng/mL).

Following Amendment 2, patients in the sirolimus group received two

15 mg oral loading doses within 48 hr after transplantation followed by 10 mg

daily until whole-blood sirolimus trough levels were 10.0 ng/mL or more.

Patients who began sirolimus dosing before transplantation received a 10 mg

dose of sirolimus daily beginning within 24 hr after transplantation until

whole-blood sirolimus trough levels were 10.0 ng/mL or more. Subsequent



doses were adjusted to maintain trough levels within the protocol-specified

range (study start to week 26, 10Y15 ng/mL; weeks 27Y104, 8Y15 ng/mL).

Patients in the CsA group received 6 to 8 mg/kg CsA as an oral loading

dose in divided doses, initiated between 7 days before and 48 hr after

transplantation. Whole-blood CsA trough concentrations were monitored

at designated intervals, and the twice-daily CsA maintenance dose was ad-

justed to maintain the desired trough concentration ranges (study start to

week 13, 150Y300 ng/mL; weeks 14Y26, 50Y200 ng/mL; weeks 27Y104,

50Y150 ng/mL).

A 20 mg dose of basiliximab was administered intravenously within 2 hr

before transplantation, and a second 20 mg dose was administered intrave-

nously on day 4 after transplantation. MMF administration was initiated

within 48 hr after transplantation (total dose e2 g/day, administered once or

in divided doses) and continued throughout the treatment period. In the

sirolimus group, the MMF dose was reduced to 1.5 g per day once the

sirolimus trough level was attained. CS was administered starting with intra-

venous methylprednisolone 500 mg for 2 days and tapered to a minimum of

7.5 mg per day through 6 months and 5 mg per day thereafter.

Laboratory DeterminationsSirolimus and CsA blood levels were determined using previously vali-

dated high-performance liquid chromatography or liquid chromatography-

tandem mass spectroscopy immunoassays performed at a local or designated

central laboratory to guide dosing. Trough concentrations were determined

within 48 hr after transplantation, at each study visit, and at the time of a

suspected drug-related AE or rejection episode. Although not used to guide

dosing, MPA/MPAG trough plasma levels were determined and assayed at a

central laboratory using a liquid chromatography-tandem mass spectroscopy

method at regular intervals.

Data Analysis and MonitoringA sample size of 500 patients was planned for analysis of efficacy and

safety endpoints. Continuous variables were analyzed using the one-way

analysis of variance with treatment as a factor. Categorical variables were

analyzed using Fisher’s exact test. Time-to-event data (graft, patient, and

BCAR-free survival) were analyzed using KaplanYMeier methods and

compared using the log-rank test for treatment difference. Ordinal data

(severity of BCAR) was compared using the Cochran-Mantel-Haenszel row

mean tests. The ITT analysis of Nankivell-calculated GFR at 6 months was

performed based on all randomized patients who had 6-month GFR data

(either on-therapy or off-therapy). A GFR value of 0 was assigned in pa-

tients who died or experienced graft loss before 6 months. Missing values

owing to early withdrawal from the study were not imputed. The on-

therapy analysis of Nankivell-calculated GFR was performed based on all

randomized patients who remained on assigned therapy up to 6 months or

more and who had a GFR assessment at the 6-month visit. No imputation

was made for missing values.

ACKNOWLEDGMENTSThe authors thank the patients and their families, the

study nurses, and clinical staff who cared for patients while inthe trial and the subinvestigators, study coordinators, researchsupport staff, and research and data management personnelfor their participation in this study. This study was sponsoredby Wyeth Pharmaceuticals, which was acquired by Pfizer inOctober 2009. Medical writing support was provided by Bina J.Patel, PharmD, at Peloton Advantage and was funded by Pfizer.

The authors also acknowledge the following leadinvestigators, in alphabetical order, for their efforts inthe execution of this study: Enver Akalin, Mount SinaiSchool of Medicine, Racanati/Miller Transplantation Insti-tute, New York, NY; Paolo Altieri, Ospedale S Michele,Cagliari, Italy; Laszlo Asztalos, Debreceni Egyetem Orvos-Es Egeszsegtudomanyi Centrum Aok I. Sz. Sebeszeti Klinika,Derecen, Nagyerde, Hungary; Prabhakar Baliga, MedicalUniversity of South Carolina, Charleston, SC; Rajendra

Baliga, Tampa General Hospital, Tampa, FL; Marcelo O.Baran, Universidad Nacional de La Plata, Buenos Aires,Argentina; Kenneth A. Bodziak, University Hospitals ofCleveland, Cleveland, OH; Ionannis Boletis, Laikon GeneralHospital, Athens, Greece; Scott Campbell, Princess AlexandraHospital, Woolloongabba, Australia; Steven Chadban, RoyalPrince Alfred Hospital, Camperdown, Australia; GiovanniCivati, Ospedale Niguarda Ca’ Granda, Milano, Italy; PierreDaloze, C.H.U.M., Hopital Notre-Dame, Montreal, Quebec,Canada; Maria Del Carmen Rial, Instituto de NefrologiaBuenos Aires, Buenos Aires, Argentina; Javier Dominguez,Pontificia Universidad Catolica de Chile, Santiago, Chile;George Francos, Thomas Jefferson University, Philadelphia, PA;Michael J. Germain, Western New England Renal and Trans-plant Associates, Springfield, MA; Maciej Glyda, SzpitalWojewodzki W Poznaniu, Poznan, Poland; Reginald Gohh,Rhode Island Hospital, Providence, RI; Duck J. Han, AsanMedical Center, Seoul, Korea; Jeno Jaray, Semmelweis EgyetemAok Transzplantacios Es Sebeszeti Klinika, Budapest, Hungary;Lynt B. Johnson, Georgetown University Hospital, Washington,DC; Delawir Kahn, Groote Schuur Hospital, Observatory,South Africa; Yu S. Kim, Yonsei University Medical Cen-ter, Seoul, Korea; Mysore S. Anil Kumar, Hahnemann UniversityHospital, Philadelphia, PA; Paul C. Kuo, Duke Univer-sity Medical Center, Durham, NC; George K. Kyriakides,Paraskevaidion Surgical and Transplantation Center, Nicosia,Cyprus; Gyorgy Lazar, Albert Orvos-Es GyogyszeresztudomanyiCentrum Sebeszeti Klinika, Pecsi, Hungary; Marc I. Lorber, YaleUniversity School of Medicine, New Haven, CT; Eduardo C.Maggiora, Asistencia Nefrologica Integral en Ciudad, BuenosAires, Argentina; Pablo U. Massari, Hospital PrivadoYCentroMedico de Cordoba, Cordoba, Argentina; Tsai Meng-Kun,National Taiwan University Hospital, Taipei, Taiwan; DeepakMital, Rush University Medical Center, Chicago, IL; Jose M.Morales, Hospital Doce de Octubre, Madrid, Spain; AlfredoMota, Hospitais da Universidade de Coimbra, Coimbra,Portugal; George Mourad, C.H.U. Lapeyronie, MontpellierCedex, France; Norman Muirhead, London Health Sci-ences Centre, London, Ontario, Canada; Saraladevi Naicker,Johannesburg Hospital, Parktown, South Africa; MichaelNicholson, Leicester General Hospital, Leicester, UK; Pablo A.Novoa, Universidad Catolica Cordoba Cordoba, Argentina;Philip O’Connell, Westmead Hospital, Westmead, New SouthWales, Australia; Ravi Parasuraman, Henry Ford TransplantInstitute, Detroit, MI; Pamela Patton, University of Florida,Gainesville, FL; Jaime Sanchez Plumed, Hospital UniversitarioLa Fe, Valencia, Spain; Henry Randall, Baylor UniversityMedical Center, Dallas, TX; Hany Riad, Manchester Royal In-firmary, Manchester, UK; Juan C. Ruiz, Hospital Marquesde Valdecilla, Santander, Spain; Ruben O. Schiavelli, HospitalGeneral de Agudos, Buenos Aires, Argentina; Giuseppe Segoloni,Azienda ospedaliera S Giovanni Battista, Torino, Italy; StevenSteinberg, California Institute of Renal Research, San Diego,CA; Dimitris Takoudas, Aristotelian University Thessaloniki,Ipokratio General Hospital, Thessaloniki, Greece; LorenzoToselli, CRAI Norte, Buenos Aires, Argentina; Huseyin Toz, EgeUniversitesi Tip Fakultesi, Izmir, Turkey; Aydin Turkmen,Istanbul Universitesi, Istanbul, Turkey; Charles T. Van Buren,Baylor Clinic, Houston, TX; Anantharaman Vathsala, SingaporeGeneral Hospital, Singapore; John Vella, Maine MedicalCenter, Portland, ME; Maria M. Vidas, Hospital Clınico San



Carlos, Madrid, Spain; Overlakare J. Wadstrom, KirurgklinikenAkademiska Sjukhuset, Uppsala, Sweden; Rowan Walker,The Royal Melbourne Hospital, Parkville, Victoria, Australia;John D. Whelchel, Piedmont Hospital, Atlanta, GA.

REFERENCES1. Ciancio G, Burke GW, Suzart K, et al. Daclizumab induction,

tacrolimus, mycophenolate mofetil and steroids as an immunosup-pression regimen for primary kidney transplant recipients. Trans-plantation 2002; 73: 1100.

2. Nankivell BJ, Borrows RJ, Fung CL, et al. The natural history of chronicallograft nephropathy. N Engl J Med 2003; 349: 2326.

3. Naesens M, Kuypers DR, Sarwal M. Calcineurin inhibitor nephro-toxicity. Clin J Am Soc Nephrol 2009; 4: 481.

4. Rapamune [Package Insert]. Philadelphia, PA: Wyeth Pharmaceuticals;2011.

5. Flechner SM, Goldfarb D, Modlin C, et al. Kidney transplantationwithout calcineurin inhibitor drugs: a prospective, randomized trialof sirolimus versus cyclosporine. Transplantation 2002; 74: 1070.

6. Martinez-Mier G, Mendez-Lopez MT, Estrada-Oros J, et al. Conver-sion from calcineurin inhibitor to sirolimus for renal function dete-rioration in kidney allograft recipients. Arch Med Res 2006; 37: 635.

7. Buchler M, Caillard S, Barbier S, et al. Sirolimus versus cyclosporine inkidney recipients receiving thymoglobulin, mycophenolate mofetiland a 6-month course of steroids. Am J Transplant 2007; 7: 2522.

8. Flechner SM, Goldfarb D, Solez K, et al. Kidney transplantationwith sirolimus and mycophenolate mofetil-based immunosuppression:5-year results of a randomized prospective trial compared to calcineurininhibitor drugs. Transplantation 2007; 83: 883.

9. Hamdy AF, Bakr MA, Ghoneim MA. Long-term efficacy and safety ofa calcineurin inhibitor-free regimen in live-donor renal transplantrecipients. J Am Soc Nephrol 2008; 19: 1225.

10. Glotz D, Charpentier B, Abramovicz D, et al. Thymoglobulin induc-tion and sirolimus versus tacrolimus in kidney transplant recipientsreceiving mycophenolate mofetil and steroids. Transplantation 2010;89: 1511.

11. Schena FP, Pascoe MD, Alberu J, et al. Conversion from calcineurininhibitors to sirolimus maintenance therapy in renal allograft re-cipients: 24-month efficacy and safety results from the CONVERTtrial. Transplantation 2009; 87: 233.

12. Flechner SM, Glyda M, Cockfield S, et al. The ORION study: com-parison of two sirolimus-based regimens versus tacrolimus andmycophenolate mofetil in renal allograft recipients. Am J Transplant2011; 11: 1633.

13. McTaggart RA, Tomlanovich S, Bostrom A, et al. Comparison ofoutcomes after delayed graft function: sirolimus-based versus othercalcineurin-inhibitor sparing induction immunosuppression regi-mens. Transplantation 2004; 78: 475.

14. McTaggart RA, Gottlieb D, Brooks J, et al. Sirolimus prolongs recoveryfrom delayed graft function after cadaveric renal transplantation. Am JTransplant 2003; 3: 416.

15. Simon JF, Swanson SJ, Agodoa LY, et al. Induction sirolimus and de-layed graft function after deceased donor kidney transplantation inthe United States. Am J Nephrol 2004; 24: 393.

16. Lebranchu Y, Thierry A, Thervet E, et al. Efficacy and safety of earlycyclosporine conversion to sirolimus with continued MMFVfour-yearresults of the postconcept study. Am J Transplant 2011; 11: 1665.

17. Gonwa T, Johnson C, Ahsan N, et al. Randomized trial of tacrolimus +mycophenolate mofetil or azathioprine versus cyclosporine + myco-phenolate mofetil after cadaveric kidney transplantation: results atthree years. Transplantation 2003; 75: 2048.

18. Abramowicz D, Manas D, Lao M, et al. Cyclosporine withdrawal froma mycophenolate mofetil-containing immunosuppressive regimen instable kidney transplant recipients: a randomized, controlled study.Transplantation 2002; 74: 1725.

19. Hazzan M, Labalette M, Copin MC, et al. Predictive factors of acuterejection after early cyclosporine withdrawal in renal transplant re-cipients who receive mycophenolate mofetil: results from a prospec-tive, randomized trial. J Am Soc Nephrol 2005; 16: 2509.

20. Ekberg H, Grinyo J, Nashan B, et al. Cyclosporine sparing withmycophenolate mofetil, daclizumab and corticosteroids in renal al-lograft recipients: the CAESAR Study. Am J Transplant 2007; 7: 560.

21. Larson TS, Dean PG, Stegall MD, et al. Complete avoidance ofcalcineurin inhibitors in renal transplantation: a randomized trialcomparing sirolimus and tacrolimus. Am J Transplant 2006; 6: 514.

22. Lo A, Egidi MF, Gaber LW, et al. Comparison of sirolimus-basedcalcineurin inhibitor-sparing and calcineurin inhibitor-free regimensin cadaveric renal transplantation. Transplantation 2004; 77: 1228.

23. Hamdy AF, El-Agroudy AE, Bakr MA, et al. Comparison of sirolimuswith low-dose tacrolimus versus sirolimus-based calcineurin inhibitor-free regimen in live donor renal transplantation. Am J Transplant 2005;5: 2531.

24. Ekberg H, Tedesco-Silva H, Demirbas A, et al. Reduced exposure tocalcineurin inhibitors in renal transplantation. N Engl J Med 2007;357: 2562.

25. Vincenti F, Charpentier B, Vanrenterghem Y, et al. A phase III study ofbelatacept-based immunosuppression regimens versus cyclosporinein renal transplant recipients (BENEFIT study). Am J Transplant 2010;10: 535.



Health & Medicine

A randomized, open label study of sirolimus versus CSA