Transplant Immunology 20 (2008) 32–42

Contents lists available at ScienceDirect

Transplant Immunology

j ourna l homepage: www.e lsev ie r.com/ locate / t r im

Comparison of four different immunosuppression protocols without long-termsteroid therapy in kidney recipients monitored by surveillance biopsy:Five-year outcomes

Mysore S. Anil Kumar a,⁎, M. Irfan Saeed a, Karthik Ranganna b, Gregory Malat a, Nedjema Sustento-Reodica c,Arjun M.S. Kumar a, William C. Meyers a

a Department of Surgery and Division of Transplantation, Drexel University College of Medicine and Hahnemann University Hospital Philadelphia PA 19102, United Statesb Department of Medicine and Division of Nephrology, Drexel University College of Medicine and Hahnemann University Hospital Philadelphia PA 19102, United Statesc Department of Pathology, Drexel University College of Medicine and Hahnemann University Hospital Philadelphia PA 19102, United States

⁎ Corresponding author. Drexel University College oFeinstein Bldg, Broad and Vine, Philadelphia PA 19102,1856; fax: +1 215 762 1621.

E-mail address: akumar01@drexelmed.edu (M.S. An

0966-3274/$ – see front matter © 2008 Elsevier B.V. Aldoi:10.1016/j.trim.2008.08.005

a b s t r a c t

a r t i c l e i n f o

Article history:

Induction and maintenance i Received 27 May 2008Received in revised form 1 August 2008Accepted 8 August 2008

Keywords:Immunosuppression protocolKidney transplantSteroid withdrawal

mmunosuppression protocols with or without long-term steroid therapy in kidneytransplant recipients are variable and are transplant center-specific. The aim of this prospective randomizedpilot study was to compare 5-year outcomes in kidney recipients maintained on 4 different calcineurin inhibitor(CNI)-based immunosuppression protocolswithout long-term steroid therapy. Two hundred consenting patientswho received kidney transplants between June 2000 and October 2004 were enrolled in 4 immunosuppressionprotocol groups, with 50 patients in each group: cyclosporine (CSA)/mycophenolate mofetil (MMF), CSA/sirolimus (SRL), tacrolimus (TAC)/MMF, and TAC/SRL. Induction therapy was done with basiliximab andmethylprednisolone. Steroids were withdrawn on post-transplant day 2, and long-term steroid therapy was notused. Demographic characteristics among the four groupswere comparable; approximately 50% of the recipientswere AfricanAmerican and ≥80% of the kidneys transplantedwere fromdeceased donors. Clinical acute rejection(CAR) was confirmed by biopsy and treated with intravenous pulse steroid therapy. Steroid-unresponsive CARwas treated with Thymoglobulin. Surveillance biopsies were performed at 1, 6, 12, 24, 36, 48, and 60 months toevaluate subclinical acute rejection (SCAR), chronic allograft injury (CAI), and other pathological changes per theBanff 2005 schema. The primary end point was CAR, and secondary end points were 5-year patient and graftsurvival rates, renal function, SCAR, CAI, and adverse events. In thefirst year post-transplant, the incidence of CARwas 18% in the CSA/MMFgroup, 8% in the CSA/SRL group,14% in the TAC/MMFgroup, and 4% in the TAC/SRL group(CSA/MMF vs. TAC/SRL; p=0.05). The incidence of SCARwas 22% in the CSA/MMF group, 8% in the CSA/SRL group,16% in the TAC/MMF group, and 6% in the TAC/SRL group (CSA/MMF vs. CSA/SRL and TAC/SRL; p=0.05). After thefirst year, the incidences of CAR and SCAR decreased and were comparable in all 4 groups. At 5 years post-transplant, cumulative CAI due to interstitial fibrosis/tubular atrophy (IF/TA), hypertension (HTN), and chroniccalcineurin inhibitor (CNI) toxicitywas observed in 54%, 48%, and 8% of the CSA/MMF group vs.16%, 36%, and 12%of the CSA/SRL group vs. 38%, 24% and 6%of the TAC/MMFgroup vs.14%, 25% and 12%of the TAC/SLR group (IF/TA:CSA/MMF vs. CSA/SRL and TAC/SRL; p=0.04, HTN: CSA/MMF vs. TAC/MMF and TAC/SRL; p=0.05, CNI toxicity:TAC/SRL and CSA/SRL vs. TAC/MMF; p=0.05). Five-year patient and graft survival rates were 82% and 60% in theCSA/MMF group, 82% and 60% in the CSA/SRL group, 84% and 62% in the TAC/MMF group, and 82% and 64% in theTAC/SRL group (p=0.9). Serum creatinine levels and creatinine clearances at 5 yearswere comparable among thegroups. Our data show that the rates of CAR and SCAR in the first year post-transplantwere significantly lower inthe CSA/SRL and TAC/SRL groups and that cumulative CAI rates due to IF/TA andHTN at 5 yearswere significantlylower in the TAC/MMF, TAC/SRL, and CSA/SRL groups than in the CSA/MMF group. Despite significant differencesin the incidences of CAR and SCAR and prevalence of different types of CAI at 5 years, renal function and patientand graft survival rates at 5 years were comparable among kidney recipients maintained on 4 differentimmunosuppression protocols without long-term steroid therapy.

© 2008 Elsevier B.V. All rights reserved.

f Medicine, MS 417, 5th Floor,United States. Tel.: +1 215 762

il Kumar).

l rights reserved.

1. Introduction

The present standard-of-care immunosuppression protocols inmost transplant programs in the United States consist of inductionwith IL-2R or a lymphocyte-depleting antibody andmaintenance with

Table 1Demographic characteristics of donors and recipients

Patient characteristics CSA/MMFgroup

CSA/SRLgroup

TAC/MMFgroup

TAC/SRLgroup

p value

RecipientsAge in years (mean±SD) 51±14 56±13 48±14⁎ 59±12⁎ ⁎0.05Male gender (%) 35 (70%) 37 (74%) 34 (68%) 34 (68%) 0.08African American (%) 25 (50%) 25 (50%) 27 (54%) 26 (52%) 0.9Diabetes (%) 12 (24%)⁎ 26 (52%)⁎ 13 (26%) 23 (46%)⁎ ⁎0.03BMI (mean±SD) 28.0±4.8 28.2±5.8 29.8±8.4 27.1±5.3 0.88HLA antigen mismatch(mean±SD)

4.0±1.9 4.1±2.0 4.0±2.1 4.1±1.8 0.9

Donor demographyDeceased donors (%) 41(82%) 43 (86%) 44(88%) 43 (86%) 0.90Expanded criteria donors (%) 12 (24%) 11 (22%) 11 (22%) 13 (26%) 0.89Cold ischemia time inhours (mean±SD)

14±8⁎ 15±6 16±8 17±8⁎ ⁎0.04

Age in years (mean±SD) 42±18 47±11⁎ 38±19⁎ 45±25 ⁎0.04BMI (mean±SD) 26.1±6.3 28.0±6.1 28.6±15.3 27.2±13.6 0.88Male gender (%) 29 (58%) 26 (52%)⁎ 33 (66%) 34 (68%)⁎ ⁎0.05

* Significant p value.

33M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

a calcineurin inhibitor (CNI) in combination with mycophenolatemofetil (MMF) or sirolimus (SRL) and long-term prednisone therapy[1–6]. Recent studies show that early withdrawal of steroids, that is,between 2 and 7 days after kidney transplant, is safe and equallyefficacious compared with long-term steroid therapy but has theadvantage of avoiding the wide-ranging side effects associated withlong-term steroid administration [7–9]. Few prospective studies inkidney recipients compare different immunosuppression protocolswith or without long-term steroid therapy [10–12]. The current CNI-based combination immunosuppression protocols are potent indecreasing acute rejections but are ineffective in preventing thedevelopment and progression of chronic allograft injury (CAI), whichpresently is one of the major causes of graft loss after a kidney istransplanted. Surveillance biopsy is a useful tool to monitor CAI inkidney transplant recipients. Additionally, CNI-based therapy hassignificant morbidity due to nephrotoxicity and other side effects thatinclude malignancy, infection, and metabolic diseases [13,14]. BothMMF and SRL have antiproliferative properties and may be beneficialin delaying or minimizing CAI [15,16]. Prospective studies comparingthe long-term outcomes of combination CNI-based therapy withoutlong-term steroid therapy are lacking.

2. Objective

The aim of this pilot study was to prospectively compare andevaluate long-term outcomes in 200 kidney transplant recipientsrandomized into four different CNI-based immunosuppression pro-tocols with early steroid withdrawalmonitored by surveillance biopsy.

3. Patients and methods

This study was approved by the institutional review board of DrexelUniversity College of Medicine and Hahnemann University Hospital(Drexel university project # 1000182). This study was registered as arandomized trial with the Cochrane Renal Group (# CRG 1106000095).

The protocol was designed to randomly assign 200 de novo kidneytransplant recipients to one of 4 groups, with 50 patients per group andwith each group receiving one of four immunosuppressive regimens:cyclosporine (CSA)withMMF, CSAwith SRL, tacrolimus (TAC)withMMF,and TAC with SRL. Randomization was completed online using the firstgenerator plan at Randomization.com (accessed at http://www.rando-mization.com). The study was open for randomization until 50 de novokidney recipients were enrolled in each of the four groups for a total of200 patients. All recipients in this study received transplants betweenJune2000andOctober 2004. Selection criteriawere as follows: recipientswere older than 20 years of age, were able to sign an informed consentform,wereHIVandHBVnegative, haddeceasedand livingdonorkidneys,and were of any ethnicity. This report is an analysis of the subgroup ofpatients randomized into one of four groups of immunosuppressionprotocols treatedwithout long-term steroid therapy. A cohort of patientsin this study was reported previously to compare ethnicity [AfricanAmerican (AA) vs. non-AfricanAmerican (non-AA)kidney recipients] andto compare early steroid withdrawal and long-term steroid therapy inkidney transplant recipients but without comparing the outcomes ofthe four different immunosuppression protocols [9,17]. A cohort of theCSA/MMF group from the present study was described in a previouspublication comparing early steroid withdrawal with long-term steroidtherapy in kidney transplant recipients [18].

The present study is the first report of results comparing kidneytransplant recipients randomized to four different CNI-based immu-nosuppression protocols with early steroid withdrawal.

3.1. Immunosuppression protocols and infection prophylaxis

All recipients were given induction therapy with 2 doses of 20 mgbasiliximab on days 0 and 4 and 2 doses of methylprednisolone, 250mg

on day 0 and 125 mg on day 1. Steroid therapy was discontinuedcompletely after the second dose of methylprednisolone [9].

TAC or CSA was initiated on day 1. TAC was initiated at 0.02 mg/kgbody weight per day and the dose was increased rapidly to achievetrough blood levels of 15–18 ng/ml by day 4; these levels weremaintained for 1 month. Beginning with the second month, the TACblood levels were gradually reduced to 10 ng/ml by the end of 1 year.CSA was initiated at 3 mg/kg body weight per day, and the dose wasrapidly increased to maintain C2 blood levels (cyclosporine bloodlevels 2 h after the dose) of 1000–1200 ng/ml at 1 month and thengradually tapered to maintain 700 ng/ml at 1 year. MMF was initiatedon day 1 at 2 g/day in divided doses, and trough blood levels ofmycophenolic acid were maintained between 1 and 3 µ/ml. SRL wasinitiated on day 4 at 2mg/day; the dosewas then adjusted to maintaintrough blood levels of 5 to 10 ng/ml. During the initial study period, inthe first 10 patients, clinical observations and surveillance biopsyresults showed that with the above-described CNI blood levels,recipients in the CSA/SRL and TAC/SRL groups showed lower renalfunction and acute CNI toxicity confirmed by biopsy. After theseclinical and biopsy findings were noted in the first 10 patients in eachof the CSA/SRL and TAC/SRL groups, beginning 1 month aftertransplantation, the protocol was modified in subsequent patientsto maintain lower blood levels of CSA and TAC. In the CSA/SRL andTAC/SRL groups, the daily doses of CSA and TAC were adjusted tomaintain C2 blood levels of CSA of between 500 and 800 ng/ml andtrough levels of TAC of between 5 and 9 ng/ml of blood.We did not usea loading dose of SRL, and initiation of SRL was delayed to day 4 tominimize wound-healing complications. Steroid therapy was discon-tinued after the second dose of methylprednisolone, and long-termsteroid therapy was not used in these patients.

All participants except donor/recipient seronegative pairsreceived valganciclovir for prophylaxis for cytomegalovirus (CMV)infection. Trimethoprim-sulfamethoxazole was given for prophylaxisfor Pneumocystis carinii. Both of these prophylactic agentswere given for100 days post-transplant.

3.2. Diagnosis and treatment of clinical acute rejection

Clinical acute rejection (CAR) was diagnosed as a persistentlyelevated level of serum creatinine that was 15% or more above baseline.All CARs were confirmed by percutaneous biopsy of the transplantedkidney. Patients with CAR confirmed by biopsy were treated with pulsedoses of methylprednisolone: 1 g for 2 days or 500 mg for 4 days [9].Recipients with steroid unresponsive rejections were re-biopsied andtreated with Thymoglobulin (Genzyme Corp., Cambridge MA).

34 M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

3.3. Surveillance biopsy

Surveillance biopsies were completed at 1, 6, 12, 24, 36, 48, and60 months. The renal tissue was evaluated for subclinical acuterejection (SCAR) and CAI per the Banff 2005 schema [19]. Thebiopsies were performed percutaneously in the transplant out-patient office using an 18-gauge needle (Easy Core Biopsy Sys-tems™). Usually one biopsy core was obtained, but in many cases 2cores were obtained to permit a detailed examination of renal tissuewith hematoxylin and eosin, Masson's trichrome, periodic Schiff,

Fig. 1. Number of kidney recipients treated for 1 to 5 years with CSA/MMF vs. CSA/SRL vs. TACCNI toxicity, acute pyelonephritis, and acute tubular necrosis.

Jones metanamine silver, and C4D stains. After the biopsy, patientswere kept under observation and on absolute bed rest for 4 h andwere discharged home the same day. Acute CNI toxicity wasprimarily a pathological diagnosis based on the results of renalbiopsies showing isometric vacuolization of tubular cells. Biopsieswere performed on demand in recipients with graft dysfunction.Acute CNI toxicity was also diagnosed from the results of surveil-lance biopsies from patients without renal dysfunction in thepresence of isometric vacuolization of tubular cells and in theabsence of other pathological features.

/MMF vs. TAC/SRL who had indicated biopsies, incidence of clinical acute rejection, acute

35M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

3.4. Diagnosis and treatment of SCAR

SCAR was defined as acute rejection of Banff grade 1 or more in theabsence of concurrent functional deterioration [20] and was obser-vable at biopsy. SCAR was treated with methylprednisolone in amanner similar to that used for acute rejection. Thymoglobulin wasnot used to treat SCAR.

Fig. 2. Patient and graft survival rates in kidney recipients treated with CSA

3.5. CAI

Surveillance biopsies were monitored for CAI and other incidentalpathological changes. CAI was diagnosed and reported as per the Banffschema 2005 [19]. In recipients of extended criteria donor (ECD)kidneys, the pre-existing structural changes observed in pre-transplantbaseline biopsy material were taken into consideration while scoring

/MMF, CSA/SRL, TAC/MMF, and TAC/SRL immunosuppression therapy.

36 M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

CAI in surveillance biopsy. The final CAI score in ECD kidneys wasderived by comparing and subtracting the pre-existing structuralchanges seen prior to transplant from the overall changes seen insurveillance biopsies [12].

3.6. Steroid protocol and maintenance immunosuppression in recipientswith CAR and SCAR

After pulse doses of steroids were given to recipients with CAR andSCAR, steroid therapywas discontinued completely. These patients didnot receive long-term steroid therapy. In addition to treating CAR andSCARwith a short course of pulse steroid therapy, the daily dose of TACor CSA was increased to maintain elevated blood levels. TAC bloodlevels were increased by 3 to 5 ng/ml, and CSA blood levels wereincreased by 50 to 100 ng/ml from the pre-rejection baseline levels.These higher levels were maintained for 3 months after CAR and SCARresolved as described previously [9].

3.7. Delayed graft function

Delayed graft function (DGF) was diagnosed if the patient's urinevolume was less than 1000 ml, if serum creatinine levels decreasedslowly by less than 25% from pre-transplant levels in the first 24 hafter transplant, or if the patient required dialysis in the first weekafter the transplant [21]. A biopsy was performed at 7 to10 days in allrecipients with DGF to rule out associated acute rejection.

3.8. ECD kidneys

ECD kidneys were transplanted according to our previouslydescribed protocol [22]. All ECD kidneys were biopsied prior totransplant into recipients matched by donor age and bodymass index.The pre-transplant biopsy specimens from ECD kidneys wereevaluated for the presence of glomerulosclerosis, interstitialfibrosis, arterial injury, and interstitial inflammation. The wedgebiopsy performed at the time of donor nephrectomywas large enoughto contain at least 50 glomeruli or more. In cases with insufficientspecimens, wedge biopsies were repeated to obtain specimensadequate for pre-transplant evaluation of ECD kidneys. Biopsy resultsusually showed 60 to 100 glomeruli; the injury was evaluated

Fig. 3. Kidney recipients completing surveillance biopsies and incidence of s

by counting all glomeruli and assessing all arterioles and interstitiumfor fibrosis in at least 2 sections of the wedge biopsy. Only thosekidneys with less than 15% structural damage were accepted fortransplant.

3.9. Monitoring of kidney function

Kidney function was assessed by regular monitoring of serumcreatinine levels and estimated creatinine clearances using theCockcroft–Gault formula.

3.10. Surveillance of proteinuria

Recipients were monitored for proteinuria at surveillance and atindicated biopsy points of time. Random urine samples were collectedprior to biopsy, and the protein:creatinine ratio was measured;creatinine was measured using the Jaffe reaction and protein, by thepyrogallol red method [23,24]. The definition of proteinuria for thisstudy is protein loss of N300 mg/day [25].

3.11. Subgroup analysis

In our center, approximately 50% of kidney recipients are of AfricanAmerican (AA) descent. The long-term outcomes in AA kidneyrecipients are generally inferior compared with those from non-AArecipients [26]. We compared AA with non-AA recipients in each ofthe four groups to assess the effect of race on long-term outcomeswith different immunosuppression protocols.

3.12. Conversion of maintenance immunosuppression

This study intended to continue the baseline randomized immu-nosuppression protocol and to treat patients to obtain the bestpossible outcomes. However, recipients who were unable to toleratetheir randomized immunosuppression protocols due to severe anddistressing side effects, to persistent acute rejections, or to severe CAIinjury were converted to immunosuppression protocols that bestmatched their tolerance and response. All recipients receivingconversion therapy were included in their baseline randomizedimmunosuppression group for final analysis of outcomes.

ubclinical acute rejection in groups treated with CSA/MMF vs. CSA/SRL.

37M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

3.13. Definition of adverse events

Adverse events monitored in this study were new onset diabetesmellitus (NODM), gastrointestinal side effects, hyperlipidemia, com-plications of wound healing, complications of biopsy, severe infectionsrequiring hospitalization, de novomalignancies, and anemia requiringprolonged erythropoietin therapy. NODM was defined according tothe guidelines of the American Diabetes Association, with de novoelevation of fasting blood sugar of N125 mg/dl and/or requirement fororal hypoglycemic agents or insulin after transplant [27].

3.14. Patient and graft survival and statistical analysis

Patient and graft survival rates were calculated by the Kaplan–Meierestimate, Student's t-test was used to compare continuous variables,and chi-square analysis was used to compare categorical demographicvariables and outcome differences among patient groups. Statisticalanalyses were performed using SPSS® (Statistical Package for the SocialServices, SPSS, Inc., Chicago, IL, version 6.14) software.

4. Results

All surviving patients have completed 4 years post-transplant, and 170 patients(85%) have completed 5 to 8 years post-transplant. Table 1 shows the demographiccharacteristics of patients and donors in the four groups. All recipients including thosewith CAR and SCAR have remained free of long-term steroid therapy.

4.1. CAR and other incidental pathological findings in indicated biopsies

Clinically acute increase in the serum creatinine level was an indication forbiopsy. We performed 164 indicated biopsies: 71 were performed in the firstyear; 32, in the second year; 23 each, in the third and fourth years; and 15, in thefifth year post-transplant. In the first year, CAR was diagnosed in 18% ofrecipients in the CSA/MMF, 8% in the CSA/SRL, 14% in the TAC/MMF, and 4% in the

Fig. 4. CAI identified by surveillance biopsies in kidney recipients treat

TAC/SRL groups (CSA/MMF vs. TAC/SRL, p=0.05). In the subsequent 2 to 5 years,6% or fewer patients had CAR in each of the 4 groups annually. In mostrecipients, other pathological conditions caused elevation of serum creatininelevels. Fig. 1 shows the incidence of CAR and other pathological conditionsdiagnosed from the results of indicated biopsies in each of the four groups from1 year through 5 years.

During the 5-year transplant period, the total number of indicated biopsiesperformed was 54 in the CSA/MMF group, 38 in the CSA/SRL group, 45 in the TAC/MMFgroup, and 27 in the TAC/SRL group (CSA/MMF vs. TAC/SRL, p=0.04).

4.2. Patient and graft survival rates

The patient and graft survival rates in the 4 groups were actual at 4 yearsand actuarial at 5 years. At 5 years, the patient survival rates were 82%,82%, 84%, and 82%; graft survival rates were 60%, 60%, 62%, and 64% in theCSA/MMF, CSA/SRL, TAC/MMF, and TAC/SRL groups, respectively (p=0.9). Fig. 2shows the patient and graft survival rates for all four groups for 1 through5 years.

4.3. Surveillance biopsy and SCAR

Surveillancebiopsywas completed inmostof thepatients at the scheduled intervals. Afew patients could not be biopsied at the scheduled intervals because they were receivinganticoagulation therapyor because they refused tohave theprocedure. Surveillancebiopsyat 1monthwas completed in100%of recipients. The incidence of SCARwashigh in thefirstyear and decreased over the 5-year period. At each biopsy point, from 1 through 5 years,97% ormore of the surviving patients with functioning grafts had surveillance biopsies. Inthe first year, the incidence of SCAR was significantly lower in the TAC/SRL and CSA/SRLgroups than in the CSA/MMF group (p=0.05). In years 2 through 5, the incidence of SCARdecreased in all 4 groups. At 5 years, SCAR was diagnosed in 10% or fewer of thesurviving recipients. Fig. 3 shows the number of recipients with functioning grafts, thenumber and percent of recipients completing surveillance biopsies at 1, 2, 3, 4, and 5 years,and the incidence of SCAR in the four groups of recipients.

4.4. CAI

Surveillance biopsies were performed in all patients at 1, 2, 3, 4, and 5 years tomonitor for CAI. Banff 2005 criteria were used for pathological classification and

ed for 1 to 5 years with CSA/MMF, CSA/SRL, TAC/MMF, or TAC/SRL.

Fig. 4 (continued).

38 M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

grading of CAI (19). Chronic hypertensive changes were based on pathological findingsassociated with long-standing mean arterial pressures ≥85; chronic CNI toxicity wasbased on pathological findings associated with mean arterial pressures ≤80. Interstitialfibrosis and tubular atrophy (IF/TA) levels were scored and reported separatelyregardless of etiology. Fig. 4 shows CAI due to hypertension (chronic hypertensivechanges included arterial and/or fibrointimal thickening with reduplication of elasticausually affecting small artery and arteriolar hyaline changes); chronic CNI toxicity andIF/TA; viral infections; pyelonephritis; chronic obstruction; chronic T-cell-mediatedrejection and chronic active antibody-mediated rejection; recurrent disease observedin material obtained from surveillance biopsies at 1 to 5 years; and severity of IF/TA.

4.5. DGF

DGF was diagnosed in 46%, 84%, 42%, and 74% of patients from the CSA/MMF,CSA/SRL, TAC/MMF, and TAC/SRL groups respectively (CSA/MMF and TAC/MMF vs.CSA/SRL and TAC/SRL, p=0.05). The mean length of DGF was 6.9±3.3, 10.8±2.6, 6.7±2.9, and 9.8±2.1 days in the CSA/MMF, CSA/SRL, TAC/MMF, and TAC/SRL groups,respectively (CSA/MMF and TAC/MMF vs. CSA/SRL and TAC/SRL groups p=0.05).

4.6. Change in maintenance immunosuppression

In each of the 4 groups, some of the recipients did not tolerate their randomizedcombination of immunosuppressive agents. These patients were converted to one ofthe other combinations, depending on the side effects of one or both of theirrandomized immunosuppressive combinations. Table 2 shows the number of recipientsin each group with a change in immunosuppressive agents and the number ofrecipients able to continue the original randomized immunosuppression therapy.

4.7. Daily dose and blood levels of CSA, TAC, MMF, and SRL

Per the protocol, the targeted blood levels of CSA, TAC, MMF, and SRL were similarin all 4 groups. However, both indicated and surveillance biopsies showed that, withcomparable blood levels of CSA and TAC, both short- and long-term CNI nephrotoxicitysignificantly increased in recipients maintained on combinations containing SRL (CSA/SRLand TAC/SRL) compared to those containingMMF (CSA/MMFand TAC/MMF). To overcomethe CNI nephrotoxicity, the daily doses of CSA and TAC were reduced to maintain

Fig. 4 (continued).

39M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

significantly lower blood levels of CSA and TAC combined with SRL compared with thoserecipients receiving combinations containing MMF.

Six months after the transplant, the mean daily dose of CSA was 375±35 mg inthe CSA/MMF group and 189±23 mg in the CSA/SRL group (p=0.05); the mean dailydose of TAC was 11.8±2.9 mg in the TAC/MMF group and 6.5±1.3 mg in the TAC/SRLgroup (p=0.04). Themean daily dose ofMMFwas 1.82±0.4 g in the CSA/MMF group and1.15±0.2 g in the TAC/MMF group (p=0.05). The mean daily dose of SRL was 3.8±1.0 mgin the CSA/SRL group and 3.7±0.9mg in the TAC/SRL group (p=0.9). At 5 years, themeanCSAdosewas 225±22mg/day in the CSA/MMFgroup and 142±15mg/day in the CSA/SRLgroup (p=0.02); the mean TAC dose was 8.9±1.6 mg/day in the TAC/MMF group and4.9±0.9mg/day in theTAC/SRL group (space) (p=0.01); themeanMMFdosewas1.74±0.3 g/dayin the CSA/MMF group and 1.39±05 g/day in the TAC/MMF group (p = 0.05); and themeanSRL dose was 2.1±0.5 mg/day in the CSA/SRL group and 1.8±0.3 mg/day in the TAC/SRLgroup (p=0.08), respectively.

At 6 months post-transplant in the CSA/MMF group, the mean C2 blood levels ofCSA were 1062±121 ng/ml and trough MPA levels were 2.8±09 µg/ml; in the CSA/SRLgroup, C2 blood levels were 756±65 ng/ml and SRL was 6.9±2.2 ng/ml of blood. In theTAC/MMF group, trough TAC levels were 11.3±2.8 ng/ml and MPA levels were 1.9±0.8 µg/ml. In the TAC/SRL group, TAC blood levels were 6.1±2.7 ng/ml and SRL levelswere 6.8±1.7 ng/ml (C2 levels: CSA/MMF vs. CSA/SRL; p=0.03, TAC levels: TAC/MMF vs.TAC/SRL; p=0.04, MPA levels: CSA/MMF vs. TAC/MMF; p=0.05 and SRL levels: CSA/SRLvs. TAC/SRLp=0.6). At 5 years, themeanCSAC2 levelswere 524±87ng/ml in theCSA/MMFgroup and 389±26ng/ml in theCSA/SRL group (p=0.03); themean trough TAC levelswere7.9±1.6 ng/ml in the TAC/MMF group and 4.8±1.3 ng/ml in the TAC/SRL group (p=0.04);themean trough levels ofMPAwere 1.9±0.9 µg/ml in the CSA/MMF group and 1.6 µg/ml inthe TAC/MMF group (p=0.05); and the mean trough SRL levels were 4.8±0.8 ng/ml in theCSA/SRL group and 4.5±0.6 ng/ml in the TAC/SRL group (p=0.6).

Both theMMF dose andMPA blood levels were lower in the TAC/MMF group than inthe CSA/MMF group because patients in the TAC/MMF group could not tolerateequivalent doses of MMF compared to recipients in the CSA/MMF group because ofsevere upper and lower gastrointestinal side effects.

4.8. Kidney function

Kidney function was monitored by serial measurements of serum creatinine levelsand estimated creatinine clearances at monthly intervals in the first year and thenquarterly for 5 years. Fig. 5 shows the mean serum creatinine levels and meancreatinine clearances at 1, 2, 3, 4, and 5 years in the 4 groups of kidney recipients. At5 years, a significant difference existed in the serum creatinine levels (p=0.04) andcreatinine clearances (p=0.05) of the CSA/MMF and TAC/SRL groups.

4.9. Subgroup analysis of AA and non-AA recipients

In each of the four groups, approximately 50% were AA recipients. The outcomes ofthe AA and non-AA recipients in each of the four groups were analyzed. Table 3 showsthe incidence of CAR, SCAR, the prevalence of interstitial fibrosis/tubular atrophy,hypertensive changes, and patient and graft survival rates at 5 years. In each of the 4groups, subgroup analysis of AA and non-AA recipients showed the following: theincidence of CAR was comparable; the incidence of SCAR was higher in AA recipientsthroughout the 5-year period; the prevalence of CAI due to hypertension and interstitial

Table 2Conversion of baseline immunosuppression due to intolerance, persistent side effects, and

Recipient group Persistent side effectsand intolerance

Unresponsive acuterejections

Total co

CSA/MMF CSA in 2, MMF in 4 3 9CSA/SRL CSA in 3, SRL in 3 2 8

TAC/MMF TAC in 3, MMF in 4 2 9

TAC/SRL TAC in 2, SRL in 2 1 5

fibrosis/tubular atrophy was significantly higher at 5 years in AA recipients than in non-AA recipients in all 4 groups. Five-year patient and graft survival rates were comparablebetween AA and non-AA recipients in all 4 groups.

4.10. Surveillance proteinuria

During the 5 years post-transplant, proteinuria was present in 34 (17%) patients: 6in the CSA/MMF group, 11 in the CSA/SRL group, 6 in the TAC/MMF group, and 11 in theTAC/SRL group. The proteinuria was in the nephrotic range of N3500 mg/day in 2patients in the CSA/MMF group (both had pre-transplant diabetes); in 9 patients in theCSA/SRL group (4 had pre-transplant diabetes, 2 had recurrent renal disease, 3 had nopredisposing factors); in 3 patients in the TAC/MMF group (2 patients had recurrentdisease and 1 had diabetes); and in 8 patients (4 had diabetes, 1 had recurrent disease,and 3 had no predisposing factors) in the TAC/SRL group (CSA/SRL and TAC/SRL vs.CSA/MMF and TAC/MMF; p=0.05). A nephrotic range of proteinuria developed atvarious time points after transplantation; in the CSA/MMF group, both recipientsdeveloped proteinuria between 2 and 4 years; in the CSA/SRL group, 3 recipientsdeveloped proteinuria at 1 year, 1 at 2 years, and 5 at 3 years; in the TAC/MMF group,1 recipient developed proteinuria at 2 years and 2, at 3 years; and in the TAC/SRLgroup, 3 patients developed proteinuria at 2 years, 2 recipients at 3 years, and 3 at4 years. In recipients of all 4 groups with a nephrotic range of proteinuria, biopsyfindings included podocyte injury, chronic active antibody-mediated rejec-tion, and peritubular capillary basement multilayering. Most of these patients hadpre-transplant diabetes mellitus or recurrent renal disease. All recipients weretreated with angiotensin receptor blocking drugs to reduce proteinuria. Eightrecipients who received SRL treatment lost the grafts between 1 and 4 years. Theremaining 9 recipients with reduced proteinuria continue to have graft function at5 years. In the MMF-treated group, 3 patients lost the grafts between 1 and 4 yearsand 2 continue to have graft function at 5 years.

4.11. Adverse events

The mean body mass index from pre-transplant baseline to 3 years post-transplantincreased by approximately 2% to 3% in each of the four groups of patients. NODM wasseen in 6% of patients in the TAC/MMF group; in 4% of those in the TAC/SRL group, in 0%of those in the CSA/MMF group, and in 4% of those in the CSA/SRL group at 5 years post-transplant (p=ns). NODMwas observedwithin the first 12months after transplant in allgroups of recipients. There was no correlation between NODM and pulse doses ofmethylprednisolone used to treat CAR or SCAR.

Severe infections diagnosed included bacterial sepsis, CMV infection, and Candidainfection. Severe bacterial infectionswere seen in 18% of patients in the TAC/MMF group,in 24% in the TAC/SRL group, in 10% in the CSA/MMF group, and in 16% in the CSA/SRLgroup (CSA/MMF vs. TAC/SRL groups p=0.05). CMV infections were seen in one patienteach in the TAC/MMFand CSA/MMF groups, and Candida pyelonephritis was seen in onepatient in the TAC/SRL group. Mouth ulcers were seen in 3 (6%) patients in the CSA/SRLgroup and in 2 (4%) in the TAC/SRL group.

Endoscope-proven refluxgastroesophagitiswas seen in1 (2%)patient in theCSA/MMFgroup and in 5 (10%) patients in the TAC/MMF group. Severe diarrhea causing dehydrationand acute renal failurewas observed in 3 (6%) patients in the CSA/MMF group and in 4 (8%)in the TAC/MMF group.

unresponsive acute rejection

nversions Conversion therapy Number of recipientsmaintained on original therapy

TAC/MMF in 5, CSA/SRL in 4 41 (82%)TAC/MMF in 6 42 (84%)CSA/MMF in 2CSA/MMF in 3 41 (82%)TAC/SRL in 6CSA/SRL in 3 45 (90%)TAC/MMF in 2

Fig. 5. Renal function: Serum creatinine levels and estimated creatinine clearances at 1, 2, 3, 4, and 5 years in kidney recipients receiving CSA/MMF, CSA/SRL, TAC/MMF, TAC, or SRL.

40 M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

Erythropoietin treatment was given to patients with a hemoglobin level of less than11g/dl: 6% of patients in theTAC/MMFgroup, 22% in theTAC/SRL group, 4% in the CSA/MMFgroup, and 8% in the CSA/SRL group (p=0.02; TAC/SRL vs. TAC/MMF, p=0.02; TAC/SRL vs.CSA/MMF and p=0.05; TAC/SRL vs. CSA/SRL groups).

Table 3Incidence of CAR and SCAR, prevalence of interstitial fibrosis/tubular atrophy,hypertensive changes, and 5-year patient and graft survival rates in African Americanvs. non-African American recipients in each of 4 groups of immunosuppressionprotocols

Recipient group CSA/MMFgroup

CSA/SRLgroup

TAC/MMFgroup

TAC/SRLgroup

Ethnicity AA Non-AA AA Non-AA AA Non-AA AA Non-AA

Number of recipients 25 25 25 25 27 23 26 24

Recipients with CAR1 year 11 7 5 3 7 7 3 12 years 3 3 2 2 1 3 1 13 years 1 3 2 2 2 2 1 14 years 4 2 2 0 3 1 1 35 years 2 2 0 0 2 0 2 0

Recipients with SCAR1 year 16 6 6 2 12 4 5 12 years 10 4 7 3 9 3 9 13 years 8 2 8 2 9 1 6 24 years 7 3 6 2 6 2 7 15 years 6 2 9 1 5 1 5 1

Hypertensive changes1 year 14 8 6 4 6 4 4 42 years 18 10 12 6 10 6 8 63 years 26 10 14 6 10 6 10 64 years 29 12 14 8 14 8 12 65 years 38 14 22 14 16 8 18 8

Interstitial fibrosis/tubular atrophy1 year 12 4 4 4 6 2 4 42 years 16 6 4 4 12 6 4 43 years 20 8 4 4 18 6 4 44 years 28 14 8 4 22 10 8 45 years 36 18 14 6 28 14 14 6

Patient survival1 year 93 95 92 94 94 94 94 962 years 90 93 89 94 90 90 90 923 years 88 89 86 90 88 90 88 884 years 82 83 80 84 83 86 85 885 years 79 80 78 82 80 82 82 84

Delayed wound healing or infection of the initial transplant surgery incision wasseen in 2 (4%) patients in the CSA/MMF group, in 3 (6%) in the CSA/SRL group, in 3 (6%)in the TAC/MMF group, and in 3 (6%) in the TAC/SRL group.

The lipid levels were measured prior to transplant in all recipients and includedserum cholesterol, triglycerides, and LDL and HDL fractions. Approximately 20% ofpatients in each group were maintained on lipid-lowering agents prior to transplant. Inthefirst year after transplant, lipid-lowering agentswere used in 80%,100%, 74%, and 82%in the CSA/MMF, CSA/SRL, TAC/MMF, and TAC/SRL groups, respectively (TAC/MMF vs.CSA/SRL, p=0.05).

In the post-transplant period ranging from 1 to 8 years, de novo malignancies wereseen in 23 recipients: 9 in the CSA/MMF group (2 each of post-transplantlymphoproliferative disease and colon and thyroid cancers, one each of leukemia,renal cell carcinoma, and malignant parotid tumor); 2 in the CSA/SRL group (1 each ofthyroid and lung cancers); 10 in the TAC/MMF group (2 each of thyroid andbreast cancers, 1 each of bladder, colon, prostate, lung, and renal cell cancer and 1post-transplant lymphoproliferative disease); and 2 in the TAC/SRL group (1 each ofbreast cancer and post-transplant lymphoproliferative disease) ( CSA/SRL and TAC/SRLvs. CSA/MMF and TAC/MMF; p=0.04).

5. Discussion

The calcineurin inhibitors have remained the mainstay ofimmunosuppression therapy in patients undergoing kidney trans-plants despite their nephrotoxicity and the chronic attrition oftransplanted kidneys due to CAI [20]. In non-sensitized patients,most transplant centers in the United States use a single immuno-suppression protocol consisting of inductionwith an IL-2R antibody ora lymphocyte-depleting antibody and maintenance with a calcineurininhibitor in combination with either MMF or SRL and long-termsteroid therapy. Long-term steroid therapy has wide-ranging sideeffects andmay potentiate themetabolic and hypertensive side effectsof calcineurin inhibitors [13,28]. Previous reports show that earlysteroid withdrawal after kidney transplant is feasible in kidneytransplant recipients [7–9]. Earlier reports show that transplantedkidneys undergo structural changes despite stable graft function [20].CARs associated with graft dysfunction are widely reported by alltransplant centers. On the other hand, SCAR is diagnosed only fromthe results of surveillance biopsies and is reported by few centerspracticing surveillance biopsy as standard-of-care management.Previous studies have shown that SCAR may affect short- andlong-term graft function and survival [29,30]. Continuous monitor-ing of kidney transplant recipients by periodic surveillance biopsy atpre-determined intervals is a useful tool for the diagnosis of SCARand to study the prevalence of CAI [17,29]. Randomized studiesshow that treatment of SCAR is beneficial in prolonging survival

41M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

[29,30]. Previous studies compared 1- to 3-year outcomes in kidneyrecipients maintained on two to three different calcineurin-basedimmunosuppression protocols with or without long-term steroidtherapy but without continuously monitoring structural changes byserial surveillance biopsies soon after transplantation [10,11]. At present,no prospective studies compare long-term outcomes of early steroidwithdrawal in 4 different calcineurin-based immunosuppression pro-tocols in kidney recipients monitored for renal structural changes bysurveillance biopsy.

This prospective randomized study with predominantly deceaseddonor kidney recipients and approximately 50% AA recipientsevaluated 5-year outcomes in kidney transplant recipients withearly steroid withdrawal and maintained on four calcineurin-basedimmunosuppression protocols: CSA/MMF, CSA/SRL, TAC/MMF, andTAC/SRL. All recipients were continuously monitored for incidence ofCAR, long-term patient and graft survival, incidence of SCAR,incidence and prevalence of CAI, and renal function. The primaryend point of this study was CAR confirmed by biopsy. More than 80%of recipients continued with their assigned immunosuppressionregimen. All recipients have remained free of long-term steroidtherapy. In this study, the incidence of CAR, SCAR, type of CAI, adverseevents, and daily dose and blood levels of CNIs varied in the 4 immu-nosuppression regimens. The rate of DGF was significantly higher andthe duration longer in the CSA/SRL and TAC/SRL groups than in theCSA/MMF and TAC/MMF groups. This finding may be explained by theincreased nephrotoxicity of CNI with the SRL combination. Wedelayed initiation of SRL to 4 days post-transplant and started witha low initial dose of 2 mg/day without a loading dose. Despite thisprecaution, DGF was significantly higher in the CSA/SRL and TAC/SRLgroups. The incidence of CAR and SCAR was highest in the CSA/MMFgroup of recipients in the first year after transplant. The incidence ofCAR in the first year after transplant in kidney recipients treated withCSA/MMF in this study is comparable and consistent with that inprevious reports [18,31]. CAR was significantly higher in the CSA/MMFgroup than in the TAC/MMF group because the daily dose of MMF was2 g. It is possible that many AA recipients in the CSA/MMF group mayhave experienced a higher incidence of CAR becausewe did not use 3 gof MMF per day. Our initial experience showed that most of ourpatients, including AA recipients, could not tolerate 3 g/day of MMFdue to the severe gastrointestinal side effects. Five-year patientand graft survival rates were comparable among the four groups.Evaluation of CAI showed a significantly increased prevalence ofchanges due to hypertension and IF/TA in the CSA/MMF group.Chronic CNI toxicity was high in the CSA/SRL and TAC/SRL groups.Because of higher incidence of CNI toxicity, both daily doses andblood levels of CSA and TAC were significantly lower in the CSA/SRLand TAC/SRL groups than in the CSA/MMF and TAC/MMF groups. Aprevious report shows that chronic allograft nephropathy eventuallydevelops in all transplanted kidneys at 10 years in recipientsmaintained on CSA-based immunosuppression [20]. In our study, itis possible that the lower prevalence of CAI due to hypertension andIF/TA in the CSA/SRL and TAC/SRL groups may be due to lower dosesand lower blood levels of CSA and TAC. Calcineurin inhibitors havebeen implicated for some of the chronic structural changes observedin transplanted kidneys, and SRL has been reported to limit chronicvascular changes in native kidneys and ameliorate arteriolar injury[32,33]. A recent report of a randomized study shows that thevascular changes of CAI were significantly less in kidney recipientsmaintained on a non calcineurin inhibitor-based immunosuppres-sion regimen consisting of SRL, MMF, and prednisone compared witha TAC, MMF, and prednisone regimen [34]. Progressive chronicallograft injury results in decreased graft function and predicts earlygraft loss [20,35]. This study monitored the chronic structuralchanges due to toxicity, immunological, and non-immunologicalinjuries with different combinations of immunosuppressive drugsfrom 1 through 5 years.

Subgroup analysis of the AA and non-AA recipients in each of thefour groups showed that CAR rates were comparable, SCAR rates weresignificantly higher, and CAI changes due to hypertension andinterstitial fibrosis/tubular atrophy were significantly greater in AArecipients than in non-AA recipients in their respective immunosup-pression protocols. The mean arterial pressures were ≥85 in most ofthe AA recipients compared with those in the non-AA recipients; thisdifference may explain the higher incidence of chronic hypertensivechanges in AA recipients. The higher incidences of SCAR and CAI didnot affect long-term patient and graft survival rates in AA recipientscompared with non-AA recipients in their respective regimens.Subgroup analysis shows that, despite the small numbers, AA recipientsin each of the four immunosuppression groups had comparable rates ofCAR but that the rates of side effects from SCAR and CAI, hypertensionand interstitial fibrosis/tubular atrophy, were significantly higher in theAA than in the non-AA recipients. Although the number of recipientsanalyzed in each of the four groups in this study is too small to provideany meaningful statistical significance, these results are consistent withthose from our previous report on a larger group of recipients in whichwe compared AA with non-AA recipients (17).

The incidence of severe infections was significantly higher in theTAC/SRL group compared to that in the CSA/MMF group, suggestingthat dose reduction of one or both drugs in the TAC/SRL combinationmay be beneficial in lowering the number of serious infections.Delayed wound healing was comparable in all groups. It has beenreported that SRL causes delayed wound healing [36]. We initiatedSRL on the 4th transplant day, and we did not use a loading doseof SRL. It is possible that we did not observe a higher incidence ofwound complications in SRL-treated patients because of earlysteroid withdrawal, lack of loading dose of SRL, and delayed initia-tion of SRL after transplant. A significantly higher proportion ofpatients in the TAC/SRL group required erythropoietin treatment inthe post-transplant period than in the other 3 groups of patients. SRLhas been shown to affect hematopoietic cell growth and survival[37]. It is possible that antiproliferative and bone marrow suppres-sion properties of SRL were more severely expressed when used incombination with TAC to cause anemia.

The incidence of NODM in kidney recipients treated with TAC incombination with MMF and prednisone is reported to be as high as21% [38]. Calcineurin inhibitors and long-term steroid therapy areindependent risk factors for NODM in kidney transplant recipients,and a combination of these 2 has the potential to increase the riskfor NODM. Early steroid withdrawal significantly reduces NODM[9]. In this series NODM was 6% or less at 5 years in all 4 groups.The incidence of de novo malignancies was significantly higher inthe CSA/MMF and TAC/MMF groups compared with the CSA/SRL andTAC/SRL groups. CNIs are reported to increase malignancy [14]. On theother hand, sirolimus has known antitumor effects [39,40]. Twofactors may have decreased de novo malignancies in the CSA/SRL andTAC/SRL groups: (1) the direct antitumor effect of sirolimus in thesepatients and (2) the fact that recipients were maintained onsignificantly lower daily doses and lower blood levels of CSA andTAC compared with recipients in the CSA/MMF and TAC/MMF groups.

This study has certain limitations: It was a single-center trial with asmall number of patients in each of the four groups. Despite a longerfollow-up period after transplant, a significant difference among thegroups may not become apparent when one has a small cohort ofpatients in each group. In each of the four groups, approximately20% to 25% of recipients were transplanted with ECD kidneys withage-associated structural changes. These kidneys may be moresusceptible to preservation injury, CNI toxicity, and immunologicalinjuries and may have affected the outcomes of CAI, renal function,and survival compared to standard criteria donor kidneys.

The results of this pilot prospective randomized trial in kidneyrecipients treated with four different calcineurin inhibitor-basedtherapies, CSA/MMF, CSA/SRL, TAC/MMF, and TAC/SRL and early

42 M.S. Anil Kumar et al. / Transplant Immunology 20 (2008) 32–42

steroid withdrawal 2 days post-transplant, show that 5-year patientand graft survival rates are comparable among the four groups. Theoverall incidence of NODM remains low in all groups. The incidence ofacute rejection, the prevalence of CAI, and the adverse events vary ineach of the immunosuppression protocols analyzed in this study.

Despite significant differences in the incidence of CAR, SCAR,severity of CAI, and adverse events, 5-year renal function and patientand graft survival rates were comparable in all four groups of patients.

References

[1] Calne RY,WhiteDJG, EvansDB, et al. CyclosporinA in cadaveric organ transplantation.Br Med J 1981;282:934–6.

[2] LaskowDA, Vincenti F, Neylon J, et al. An open-label, concentration-ranging trial ofFK506 in primary kidney transplantation: a report of the United States MulticenterFK506 Kidney Transplant Group. Transplantation 1996;62:900–5.

[3] Sollinger HW. Mycophenolate mofetil for the prevention of acute rejection inprimary cadaveric renal allograft recipients. Transplantation 1995;60:225–32.

[4] Kahan BD, Podbielski J, Napoli KL, et al. Immunosuppressive effects and safety of asirolimus/cyclosporine combination regimen for renal transplantation. Transplan-tation 1998;66:1040–6.

[5] Brennan DC, Flavin K, Lowell JA, et al. A randomized, double-blinded comparisonof Thymoglobulin versus Atgam for induction immunosuppressive therapy inadult renal transplant recipients. Transplantation 1999;67(7):1011–8.

[6] Nashan B, Moore R, Amlot P, et al. Randomized trial of basiliximab versus placebo forcontrol of acute cellular rejection in renal allograft recipients. Lancet 1997;350:1193–8.

[7] Matas AJ, Kandaswamy R, Humar A, et al. Long-term immunosuppression withoutmaintenance prednisone after kidney transplantation. Ann Surg 2004;240:510–7.

[8] Hricik DE, Knauss TC, Bodziak KA, et al. Withdrawal of steroid therapy in AfricanAmerican kidney transplant recipients receiving tacrolimus and sirolimus.Transplantation 2003;76:938–42.

[9] Kumar MSA, Heifets M, Moritz MJ, et al. Safety and efficacy of steroid withdrawal2 days after kidney transplantation: analysis of results at 3 years. Transplantation2006;81:832–9.

[10] Ciancio G, Burke GW, Gaynor JJ, et al. A randomized long-term trial of tacrolimus/sirolimus versus tacrolimus/mycophenolate mofetil versus cyclosporine/sirolimusin renal transplantation: three year analysis. Transplantation 2006;81(6):845–52.

[11] Kandaswamy R, Melancon JK, Dunn TY, et al. A prospective randomized trial ofsteroid-free maintenance regimens in kidney transplant recipients: an interimanalysis. Am J Transplant 2005;5:1529–36.

[12] Anil Kumar MS, Heifets, Fyfe B, et al. Comparison of steroid avoidance in tacrolimus/mycophenolatemofetil and tacrolimus/sirolimus combination inkidney transplantationmonitored by surveillance biopsy. Transplantation 2005;80:807–14.

[13] Kahan BD, Flechner SM, Lorber MI, et al. Complications of cyclosporine-prednisoneimmunosuppression in 402 renal allograft recipients exclusively followed at a singlecenter for from one to five years. Transplantation 1987;43:197–204.

[14] HojoM, MorimotoT, Maluccio M, et al. Cyclosporine induces cancer progression bya cell-autonomous mechanism. Nature 1999;397:530–4.

[15] Sollinger HW. Mycophenolate mofetil for the prevention of acute rejection inprimary cadaveric renal allograft recipients. Transplantation 1995;60:225–32.

[16] Kahan BD, Podbielski J, Napoli KL, et al. Immunosuppressive effects and safety of asirolimus/cyclosporine combination regimen for renal transplantation. Transplantation1998;66:1040–6.

[17] Kumar MSA, Khan S, Ranganna K, et al. Long-term outcome of early steroidwithdrawal after kidney transplantation in African American recipients monitoredby surveillance biopsy. Am J Transplant 2008;8:574–5.

[18] Anil Kumar MS, Xiao SG, Fyfe B, et al. Steroid avoidance in renal transplantationusing basiliximab induction, cyclosporine-based immunosuppression and proto-col biopsy. Clin Transplant 2005;19:61–9.

[19] Solez K, Colvin RB, Racussen LC, et al. Banff '05 meeting report: Differential diagnosisof chronic allograft injury and elimination of chronic allograft nephropathy (‘CAN’).Am J Transplant 2007;7:518–26.

[20] Nankivell BJ, Borrows RJ, Fung CL, O'Connel PJ, Allen RD, Chapman JR. The naturalhistory of chronic allograft nephropathy. N Engl J Med 2003;349:2326–33.

[21] Halloran PF, Aprile MA, Farewell V, et al. Early function as the principal correlate ofgraft survival. Amultivariate analysis of 200 cadaveric renal transplants treatedwithaprotocol incorporating antilymphocyte globulin and cyclosporine. Transplantation1988;46:223–8.

[22] Kumar MSA, Panigrahi D, Dezii CM, et al. Long-term function and survival of elderlydonor kidneys transplanted into young adults. Transplantation 1998;65:282–5.

[23] Taussky HH. Amicrometric determination of creatinine in urine by the Jaffe reaction.J Biol Chem 1954;208(2):853–61.

[24] Orsonnaeu JL, Douet P, Massoubre C, et al. An improved pyrogallol red-molybdatemethod for determining total urinary protein. Clin Chem 1989;35(11):2233–6.

[25] Price CP, Newall RG, Boyd JC. Use of protein:creatinine ratio measurements on randomurine samples for prediction of significant proteinuria: a systematic review. Clin Chem2005;51(9):1577–86.

[26] U.S. Renal Data System USRDS 2006 Annual Data Report. Atlas of End Stage RenalDisease in the United States. Bethesda MD: National Institutes of Health, NationalInstitute of Diabetes and Digestive and Kidney Diseases; 2006.

[27] American Diabetes Association. Screening for type 2 diabetes. Diabetes Care2004;27(suppl 1):S11–4.

[28] Veenstra DL, Best JH, Hornberger J, et al. Incidence and long-term cost of steroid-relatedside effects after renal transplantation. Am J Kidney Dis 1999;33:829–39.

[29] Rush D, Nickerson P, Gough J, et al. Beneficial effects of treatment of earlysubclinical rejection: a randomized study. J Am Soc Nephrol 1998;9:2129–34.

[30] Miyagi M, Ishikawa Y, Mizuiri S, et al. Significance of subclinical rejection in earlyallograft biopsies for chronic allograft dysfunction. ClinTransplant 2005;19(4):456–65.

[31] Vincenti F, Schena S, Paraskevas IA, et al. A randomized multicenter study ofsteroid avoidance, early steroid withdrawal and standard steroid therapy in kidneytransplant recipients. Am J Transplant 2008;8:307–16.

[32] Bestetti R, Theodoropoulos TAD, Burdmann EA, et al. Switch from calcineurininhibitors to sirolimus-induced renal recovery in heart transplant recipients in themidterm follow-up. Transplantation 2006;81(5):692–6.

[33] Yi T, Cuchara L, Wang Y, et al. Human allograft arterial injury is ameliorated bysirolimus and cyclosporine and correlateswith suppression of interferon-[gamma].Transplantation 2006;81(4):559–66.

[34] Larson TS, Dean PG, Stegall MD, et al. Complete avoidance of calcineurin inhibitorsin renal transplantation: a randomized trial comparing sirolimus and tacrolimus.Am J Transplant 2006;6:514–22.

[35] Cosio FG, Grande JP, Larson TS, et al. Kidney allograft fibrosis and atrophy earlyafter living donor transplantation. Am J Transplant 2005;5:1130–6.

[36] Valente JF, Hricik D, Weigel K, et al. Comparison of sirolimus versus mycophenolatemofetil on surgical andwoundhealing in adult kidney transplantation.Am J Transplant2003;3:1128–34.

[37] Hammerman PS, Fox CJ, Birnbaum MJ, Thompson CB. Pim and Akt oncogenes areindependent regulators of hematopoietic cell growth and survival. Blood2005;105:4477–83.

[38] Pirsch JD, Miller J, Deierhoi MH, et al. for the FK506 kidney transplant study group.A comparison of tacrolimus (FK506) and cyclosporine for immunosuppressionafter cadaveric renal transplantation. Transplantation 1997;63:977–83.

[39] Luan FL, Hojo M, Maluccio M, Yamagi K, Suthanthiran M. Rapamycin blocks tumorprogression: unlinking immunosuppression from antitumor efficacy. Transplantation2002;73:1565–72.

[40] Huang S, Houghton PJ. Inhibitors of mammalian target of rapamycin as novelantitumor agents: from bench to clinic. Curr Opin Investig Drugs 2002;3:295–304.