The common premise for immunosuppressive

therapy update Current trends in immunosuppressive therapies for renal transplant recipients The common premise for immunosuppressive therapies in kidney transplantation is to use multiple agents to work on different immunologic targets. The use of a multidrug regimen allows for pharmacologic activity at several key steps in the T-cell replication process and lower dosages of each individual agent, thereby producing fewer drug-related toxicities. In general, there are three stages of clinical immunosuppression: induction therapy, maintenance therapy, and treatment of an established acute rejection episode. Only immunosuppressive therapies used for maintenance therapy are discussed in detail in this review. It is important to understand the three-signal model of T-cell activation and cell proliferation, because multidrug regimens act on several key targets. Briefly, signal 1 is an antigenspecific signal that is provided by the triggering of T-cell receptors by antigen-presenting cells (APCs) and is transduced through the CD3 complex. 1 Signal 2 is activated by the binding of costimulatory molecules and their ligands. These two signals Ruth-Ann Lee and Steven Gabardi activate the intracellular pathways that lead to the expression of interleukin 2 (IL-2) and other growth factors. Signal 3 triggers cell proliferation via stimulation of the IL-2 receptor. 1,2 Purpose. Current trends in immunosuppressive therapies for renal transplant recipients are reviewed. Summary. The common premise for immunosuppressive therapies in renal transplantation is to use multiple agents to work on different immunologic targets. The use of a multidrug regimen allows for pharmacologic activity at several key steps in the T-cell replication process and lower dosages of each individual agent, thereby producing fewer drug-related toxicities. In general, there are three stages of clinical immunosuppression: induction therapy, maintenance therapy, and treatment of an established acute rejection episode. Only immunosuppressive therapies used for maintenance therapy are discussed in detail in this review. The most common maintenance immunosuppressive agents can be divided into five classes: (1) the calcineurin inhibitors (CNIs) (cyclosporine and tacrolimus), (2) costimulation blockers (belatacept), (3) mammalian target of rapamycin inhibitors (sirolimus and everolimus), (4) antiproliferatives (azathioprine and mycophenolic acid derivatives), and (5) corticosteroids. Immunosuppressive regimens vary among transplantation centers but most often include a CNI and an adjuvant agent, with or without corticosteroids. Selection of appropriate immunosuppressive regimens should be patient specific, taking into account the medications pharmacologic properties, adverse-event profile, and potential drug drug interactions, as well as the patient s preexisting diseases, risk of rejection, and medication regimen. Conclusion. Advancements in transplant immunosuppression have resulted in a significant reduction in acute cellular rejection and a modest increase in long-term patient and graft survival. Because the optimal immunosuppression regimen is still unknown, immunosuppressant use should be influenced by institutional preference and tailored to the immunologic risk of the patient and adverse-effect profile of the drug. Am J Health-Syst Pharm. 2012; 69:1961-75 Maintenance immunosuppression is generally achieved by combining two or more medications from different drug classes. This maximizes the regimen s effectiveness by targeting unique components Ruth-Ann Lee, Pharm.D., CPP, is Solid Organ Transplant Clinical Specialist, Department of Pharmacy, and Clinical Assistant Professor, Department of Pharmacy Practice and Experiential Education, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill. Steven Gabardi, Pharm.D., BCPS, is Abdominal Organ Transplant Clinical Specialist, Renal Division, Departments of Transplant Surgery and Pharmacy Service, Brigham and Women s Hospital, Boston, MA, and Instructor in Medicine, Department of Medicine, Harvard Medical School, Boston. Address correspondence to Dr. Lee at the Eshelman School of Pharmacy, University of North Carolina, 301 Pharmacy Lane, CB# 7574, Bearch Hall 202 A, Chapel Hill, NC 27599-7574 (rlee2@unch.unc.edu). The authors have declared no potential conflicts of interest. Copyright 2012, American Society of Health-System Pharmacists, Inc. All rights reserved. 1079-2082/12/1102-1961$06.00. DOI 10.2146/ajhp110624 1961

of the immune response. The goals of maintenance immunosuppression are to prevent acute rejection episodes and optimize long-term patient and graft survival. Careful selection of antirejection medications and vigilant dosage adjustment are required to balance the risk of rejection with the risk of toxicities. Some of the most notorious adverse posttransplantation events (e.g., infection, malignancy) are associated with the net state of immunosuppression; therefore, it is essential that the degree of immunosuppression can be gradually reduced in the absence of rejection episodes. Between the 1960s and 1980s, azathioprine and corticosteroids were the cornerstones of maintenance suppression therapy for renal transplantation. The development of cyclosporine in the early 1980s revolutionized renal transplantation by dramatically reducing rejection rates and improving one-year graft survival rates. The evolution of maintenance immunosuppressants in the 1990s resulted in several new drug options to further prevent rejection and improve outcomes. These agents have made it possible to attain acute rejection rates of 10% and increase one-year graft survival rates to over 90%. The most common maintenance immunosuppressive agents can be divided into five classes: (1) the calcineurin inhibitors (CNIs) (cyclosporine and tacrolimus), (2) costimulation blockers (belatacept), (3) mammalian target of rapamycin (mtor) inhibitors (sirolimus and everolimus), (4) antiproliferatives (azathioprine and mycophenolic acid derivatives), and (5) corticosteroids. Immunosuppressive regimens vary among transplantation centers but most often include a CNI and an adjuvant agent, with or without corticosteroids. Selection of appropriate immunosuppressive regimens should be patient specific, taking into account the medications pharmacologic properties, adverse-event profile, and potential drug drug interactions, as well as the patient s preexisting diseases, risk of rejection, and medication regimen. CNIs Pharmacology. The CNIs cyclosporine and tacrolimus elicit their immune response by forming a complex with cytoplasmic proteins (cyclosporine with cyclophilin, and tacrolimus with FK binding protein-12 [FKBP-12]). 1,2 These drug protein complexes bind to and inhibit calcineurin phosphatase, which prevents the dephosphorylation and translocation of the nuclear factor of activated T cells. Inhibition of the passage of these cells through the nuclear membrane hampers the expression of several key cytokine genes that promote T-cell activation and expansion. The end result of calcineurin inhibition is a reduction in cytokine synthesis and a subsequent decline in lymphocyte proliferation. 1-3 Drugs and dosages. Cyclosporine is available in oral and injectable formulations with markedly variable bioavailability. 1-3 The original, oilbased formulation, first available in 1983, had bile-dependent absorption resulting in erratic bioavailability. Therefore, a modified microemulsion became available in 1994, allowing an improved pharmacokinetic profile and more consistent drug exposure. Cyclosporine (modified) is the preferred formulation for most transplantation centers that continue to use cyclosporine-based maintenance therapy. 3 It is important to note that the two formulations are not interchangeable; however, conversion from the unmodified to the modified formulation is safe with proper therapeutic drug monitoring (TDM). The labeled initial oral dosage of cyclosporine for adults after renal transplantation is dependent on the formulation, ranging from 6 to 18 mg/kg daily; however, the higher dosage is rarely used. 1-4 The biliary pathway is the primary elimination route for more than 25 identified metabolites of cyclosporine; however, the parent compound provides almost all of the immunosuppressive effects. 1 Although the unmodified formulation is rarely prescribed in practice, it is not recommended to alter the formulation for patients who are stable on this agent. Tacrolimus was approved by the Food and Drug Administration (FDA) in 1994 to prevent graft rejection in transplant recipients and is commercially available in oral and injectable formulations. 1,2,5 The recommended oral dosage of tacrolimus for adults after renal transplantation is 0.2 mg/kg daily (range, 0.1 0.3 mg/ kg daily) given in two divided doses. In rare instances, tacrolimus may be given sublingually if gut absorption is questionable. 6,7 More than 95% of tacrolimus metabolites are eliminated via the biliary route; therefore, biliary obstruction increases the concentration of tacrolimus metabolites in the blood. 1 Two once-daily, extended-release tacrolimus formulations (LCP-Tacro, Veloxis Pharmaceuticals, and Advagraf, Astellas Pharma Europe Ltd.) are currently under review in the United States. 2,8 The i.v. formulation of tacrolimus has the potential to cause anaphylaxis, most likely attributable to the polyoxyethylated hydrogenated castor oil delivery vehicle, and to increase the risk of nephrotoxicity and neurotoxicity because of overexposure to the vehicle; thus, the i.v. formulation is generally used only if the oral route is unavailable. 9 I.V. CNIs are commonly administered as a continuous infusion; alternatively, the total daily dose can be divided into two doses and infused over 4 hours. Conversion from oral to i.v. administration of cyclosporine or tacrolimus requires a reduction to one third or one fifth of the total daily oral dose, respectively, with close TDM. When converting from the i.v. to oral 1962

route, the first oral dose should be administered within 8 12 hours after discontinuation of i.v. therapy. Polyvinyl chloride containers and tubing may compromise drug stability and should thus be avoided. 4,5,9 CNIs should be initiated as soon as possible. There is no evidence that delaying CNI initiation will benefit renal function, improve delayed graft function, or prevent acute rejection. 10 The earlier that a CNI reaches target serum levels, the more effective the CNI will be in preventing acute rejection. Generic formulations of cyclosporine (unmodified and modified) and tacrolimus are considered, by FDA standards, to be bioequivalent to their innovator products. The initial dosage must take into account the patient s immunologic risk profile, preexisting disease states, history of common transplantation complications, and use of adjuvant immunosuppressive agents. 1,2,10 TDM. Constant monitoring of serum cyclosporine and tacrolimus levels is imperative, especially in the period immediately following transplantation, given the great variation in interpatient and intrapatient absorption and metabolism of these agents. 11-13 Although not the focus of this review, it is important to note that contributing factors influencing this variation may be due to the method used to measure blood concentrations of these agents, including radioimmunoassay, highperformance liquid chromatography (HPLC), and enzyme-multiplied immunoassay technique. 1 Identifying the institution s preferred method is essential, because HPLC measures only the parent compound, while other methods will detect all metabolites. 1 Trough serum levels have been the standard monitoring parameter for the management of patients taking both cyclosporine and tacrolimus. 11,12 However, for cyclosporine, trough concentrations may not provide optimal correlation with either efficacy or toxicity. 12,13 The medical literature suggests that total exposure to cyclosporine as determined by the area under the serum concentration time curve (AUC) is best predicted by a two-hour-postdose concentration (C 2 ) level and that a desirable C 2 range is 800 1500 ng/ml. 12-14 The long-term benefits of C 2 monitoring are unclear; however, initial improvements in renal function and reductions in the frequency and severity of cyclosporine-induced hypertension have been observed when such monitoring is employed. 12-14 Manufacturer recommendations for cyclosporine and tacrolimus trough concentrations range from 100 to 400 ng/ml and from 5 to 20 ng/ml, respectively, 4,5 depending on several factors, including the patient s immunologic risk profile, the time elapsed since transplantation, adverse events, and institution-specific protocols. 1,2,10-14 Trough concentrations toward the upper limit of the target range are generally avoided due to drug-related toxicities, and several studies have found excellent efficacy and tolerability with tacrolimus trough concentrations of 12 ng/ml. 12 Adverse events. One of the major drawbacks of CNIs is their extensive adverse-effect profile (Appendix A), with most toxicities being dose dependent. One of the most challenging effects of CNIs is their ability to cause acute and chronic nephrotoxicity, as well as neurotoxicity, hypertension, hyperlipidemia, and posttransplantation diabetes. 1,2,21 However, slight differences between cyclosporine and tacrolimus may influence the use of one agent over another. Comparative efficacy. The use of cyclosporine has significantly decreased since the advent of tacrolimus in 1994. However, it remains unclear which CNI should be preferentially used in renal transplant recipients. 22-25 Early trials demonstrated lower one-year frequency of biopsy-proven acute rejection (BPAR) and lower serum creatinine levels with tacrolimus compared with both cyclosporine formulations; however, long-term follow-up revealed minimal differences in patient and graft survival and function. Gonwa et al. 25 evaluated 30 randomized controlled trials including over 4000 patients in an attempt to differentiate the relative efficacy and safety of cyclosporine and tacrolimus in renal transplantation. The metaanalysis revealed that tacrolimus was associated with a significantly reduced rate of graft loss, independent of the type of cyclosporine formulation used, at 6 months after transplantation (relative risk [RR], 0.56; 95% confidence interval [CI], 0.36 0.86). This benefit persisted three years after transplantation but was diminished in patients in whom higher tacrolimus concentrations were targeted. Tacrolimus was found to have a 12-month benefit on the frequency of rejection (RR, 0.69; 95% CI, 0.60 0.79) and corticosteroidresistant rejection (RR, 0.49; 95% CI, 0.37 0.64). A comparison of the drugs safety profiles revealed that tacrolimus was associated with a higher risk of posttransplantation diabetes and neurologic and gastrointestinal effects compared with cyclosporine. Cyclosporine-treated patients complained of more cosmetic-related adverse events and had higher rates of hyperlipidemia. Recent consensus guidelines have suggested a benefit in using tacrolimus over cyclosporine in preventing early acute rejection; however, no differences in patient survival, graft function, infection, malignancy, or blood pressure have been found. 10 CNI minimization strategies. CNIs have undeniably improved short-term outcomes in renal transplantation. 10 Attention has recently shifted toward preventing CNIrelated toxicities while improving overall graft function and survival. Several comprehensive reviews of CNI-sparing regimens have been published. 26-30 The larger trials ex- 1963

ploring these strategies are reviewed below. Vincenti et al. 27 explored de novo CNI avoidance in 98 renal transplant recipients with low immunologic risk. Immunosuppression consisted of daclizumab induction, mycophenolate mofetil, and prednisone. The rates of one-year patient and graft survival were excellent (97% and 96%, respectively); however, the frequency of acute rejection at 6 months was 48%. This high rate of acute rejection prompted the CAESAR trial, 28 designed to investigate the efficacy and safety of de novo CNI reduction versus cyclosporine withdrawal. This prospective, multicenter trial randomized 536 renal transplant recipients into three groups: (1) standard-dose cyclosporine, (2) daclizumab induction with low-dose cyclosporine (target cyclosporine trough, 50 100 ng/ml), and (3) daclizumab induction with low-dose cyclosporine, which was withdrawn at 6 months. All patients received mycophenolate mofetil and prednisone. The primary endpoint was renal function; secondary endpoints included rates of BPAR and graft survival. The 12-month mean glomerular filtration rate (GFR) did not significantly differ among the groups; however, the frequency of BPAR at 12 months was significantly higher in the low-dose cyclosporine withdrawal group (38%) versus the low-dose (25.4%) and standarddose (27.5%) groups (p < 0.05). The investigators concluded that (1) cyclosporine withdrawal was associated with a higher frequency of BPAR, potentially contributing to further organ damage, and (2) lowdose cyclosporine was not inferior to standard-dose cyclosporine in terms of BPAR and showed no impact on reducing adverse effects. The same group of investigators further explored the use of low-dose CNI in the largest randomized controlled trial involving patients who underwent de novo renal transplantation to date. 29 The ELITE-SYMPHONY trial randomized 1645 renal transplant recipients to one of four treatment groups: (1) standard-dose cyclosporine (target cyclosporine trough, 150 300 ng/ml for 3 months, followed by 100 200 ng/ml), (2) daclizumab induction with low-dose tacrolimus (target tacrolimus trough, 3 7 ng/ml), (3) daclizumab induction with low-dose sirolimus (target sirolimus trough, 4 8 ng/ml), or (4) daclizumab induction with low-dose cyclosporine (target cyclosporine trough, 50 100 ng/ml). All patients received mycophenolate mofetil and prednisone. The primary endpoint of this study was renal function, and the secondary endpoints were frequency of BPAR and graft survival at 12 months. The mean calculated GFR at 12 months was higher in the tacrolimus group (65.4 ml/min) compared with the other three groups (range, 56.7 59.4 ml/min), and the frequency of BPAR was lower in the tacrolimus group (12.3%) than in the other three groups (range, 24 37.2%). The rate of graft survival was significantly higher in the tacrolimus group (94.2%) compared with the other three groups (range, 89.3 93.1%; p = 0.02). Serious adverse events were most commonly reported in the sirolimus group (53.2%) compared with the other groups (range, 43.4 44.3%). Three-year observational follow-up continued to show superior GFR and graft survival in the tacrolimus group compared with the other groups. 30 Based on this trial, the Ekberg et al. concluded that low-dose tacrolimus provided effective immunosuppression while potentially avoiding negative effects on renal function. A meta-analysis of 19 trials including 3321 patients was conducted to evaluate a CNI-sparing regimen with mycophenolate mofetil only with a median follow-up time of 12 months. 31 Five subgroups were analyzed and compared with standard-dose or higher-dose CNI therapy: (1) de novo CNI minimization, (2) elective CNI minimization, (3) elective CNI elimination (2 months after transplantation), (4) CNI minimization in renal dysfunction, and (5) CNI elimination in renal dysfunction. Overall, the risk of acute rejection did not significantly differ among groups; however, a higher GFR was apparent among all CNI-sparing trials versus controls, potentially affecting long-term graft survival. Significant heterogeneity between subgroups was evident, as there was an increased risk of acute rejection between subgroups, particularly with elective CNI elimination regimens in patients with normal renal function compared with patients in any CNI minimization or elimination subgroup with renal dysfunction. De novo CNI minimization and early withdrawal may be a safe alternative for patients with declining renal function; however, late CNI withdrawal to stabilize or improve renal function may result in irreversible kidney damage before CNI discontinuation. The authors concluded that CNI withdrawal should be done carefully to minimize the risk of acute rejection. Overall, de novo CNI minimization, rather than CNI avoidance, may be the best regimen for sparing renal function and maintaining acceptable acute rejection rates. Furthermore, identifying a population with a low immunologic risk that may benefit from this regimen is essential. Longterm data are needed to confirm that the short-term benefits will persist. Costimulation blockade Pharmacology. Belatacept is a newly approved biological indicated for long-term maintenance immunosuppressive therapy in renal transplant recipients. 32,33 This agent works by binding to CD80 and CD86 receptors on APCs and blocking the CD28-mediated costimulation of T cells. Costimulation is a critical com- 1964

ponent of the three-signal transplant model of T-cell activation. Belatacept dosage. Belatacept represents a major paradigm shift in immunosuppression for transplant recipients, being the first FDAapproved i.v. maintenance immunosuppressant. This medication is administered in two distinct phases: the initial phase and the maintenance phase. 15,33 In the initial phase, belatacept is administered at 10 mg/ kg i.v. on postoperative days 0 and 4, followed by postoperative weeks 2, 4, 8, and 12. The maintenance phase begins at the end of postoperative week 16, with a belatacept dose of 5 mg/kg, which is administered every 4 weeks (within three days) thereafter. The manufacturer recommends that doses should be calculated using the patient s actual body weight at the time of transplantation and should not be modified during the course of therapy, unless there is a change in body weight of greater than 10%. 15 The manufacturer also recommends that the final dose be evenly divisible by 12.5 mg in order for the dose to be prepared accurately. Each dose should be infused over 30 minutes. TDM. One of the benefits of belatacept is that TDM is not required. In clinical trials, belatacept trough concentrations remained constant 3 36 months after renal transplantation. 15,33 Moreover, belatacept clearance is not affected by patient age, sex, or race; renal or hepatic function; or the presence or absence of diabetes. Adverse events. The adverseevent profile of belatacept is similar to or better than that of CNI therapy. 34-36 However, belatacept s labeling carries a black-box warning regarding the risk of posttransplantation lymphoproliferative disorder, predominantly involving the central nervous system. 15,33 Because the risk of this disorder is particularly high in potential transplant recipients who have not been exposed to the Epstein-Barr virus (EBV), the manufacturer recommends using belatacept only in patients testing seropositive for the virus. 15 The most common adverse events associated with belatacept are listed in Appendix A. Although infusion-related reactions have occurred infrequently, belatacept infusions have been generally well tolerated; therefore, premedication is not required before the infusion. 34,35,37 Two cases of progressive multifocal leukoencephalopathy have been reported 15 ; however, in both cases, the patients were receiving doses much higher than those currently approved. 34,35,37 Comparative efficacy. A Phase III, randomized, active-controlled, partially blind, parallel-group study of belatacept-based immunosuppression regimens versus cyclosporine in renal transplant recipients (BENEFIT) was conducted over three years in 100 centers worldwide. 34 Two belatacept-based regimens (a moreintensive regimen [10 mg/kg during months 0 6, followed by 5 mg/kg during months 7 12] and a lessintensive regimen [10 mg/kg during months 0 3, followed by 5 mg/kg during months 3 12] were compared with a cyclosporine-based regimen. Three primary outcomes were evaluated: (1) a composite of patient and graft survival, (2) composite renal impairment, and (3) the frequency of acute rejection at 12 months. All patients received basiliximab induction, maintenance therapy with mycophenolate mofetil (2 g orally daily), and corticosteroids in addition to the study drug. Both belatacept regimens met the noninferiority margin of 10% as well as a more rigorous threshold of 5% for patient and graft survival compared with the cyclosporine regimen. Renal function was superior in patients who received belatacept versus cyclosporine. The frequency of acute rejection at 12 months was 22% in patients receiving the more-intensive belatacept regimen, 17% in those receiving the less-intensive belatacept regimen, and 7% in the cyclosporine-treated group. The less-intensive belatacept regimen was deemed noninferior for preventing acute rejection compared with cyclosporine by satisfying a 20% margin for comparison. The mean measured GFR was 13 15 ml/min higher in the belatacept groups despite a higher frequency of BPAR. Cardiovascular and metabolic outcomes were considerably better in both belatacept groups versus the cyclosporine group. Blood pressure was lower in both belatacept groups versus the cyclosporine group (p 0.0273), and cyclosporine-treated patients required more antihypertensive therapies. Concentrations of non-high-density lipoproteins increased in all treatment groups, and triglyceride levels decreased in both belatacept groups compared to increases that were observed in the cyclosporine group. The frequency of new-onset diabetes mellitus was not statistically different among the treatment groups. No significant differences were noted in the frequencies of the most common adverse events (anemia, urinary tract infection, hypertension, constipation, diarrhea, and nausea) among treatment groups. Frequencies of bacterial, viral, and fungal infections were similar in all groups. A one-year, randomized, openlabel, Phase II, multicenter trial was conducted to evaluate the efficacy and safety of tacrolimus-based immunosuppression versus two belatacept-based regimens with which corticosteroid and CNI avoidance was attempted. 36 Eighty-nine EBV-seropositive adults who received a renal transplant from a primary living donor or standardcriteria deceased donor were randomized in a 1:1:1 fashion to one of three treatment groups: belatacept and mycophenolate mofetil (n = 33), belatacept and sirolimus (n = 26), or tacrolimus plus mycophenolate mofetil (n = 30). All patients received rabbit-antithymocyte globulin induc- 1965

tion and underwent corticosteroid withdrawal. The belatacept maintenance dose was 10 mg/kg i.v. for the first six months after transplantation and then reduced to 5 mg/kg thereafter. The primary endpoint was frequency of acute rejection at six months, which occurred at a rate of 12%, 4%, and 3% in the belatacept plus mycophenolate, belatacept plus sirolimus, and tacrolimus plus mycophenolate groups, respectively. Most rejection episodes occurred within the first three months after transplantation; however, one episode in the belatacept plus mycophenolate group occurred after month 6. The composite endpoint of patient and graft survival was similar among treatment groups (91%, 92%, and 100% in the belatacept plus mycophenolate, belatacept plus sirolimus, and tacrolimus plus mycophenolate groups, respectively). One patient in the belatacept plus mycophenolate group died during the study period. The calculated GFR for each group was 64 ± 27 ml/min, 62 ± 31 ml/ min, and 54 ± 15 ml/min for the belatacept plus mycophenolate, belatacept plus sirolimus, and tacrolimus plus mycophenolate groups, respectively. CNI and corticosteroid avoidance was successful in 73% of the belatacept plus mycophenolatetreated patients and 77% of the belatacept plus sirolimus-treated patients; corticosteroid avoidance alone was achieved in 93% of patients receiving tacrolimus plus mycophenolate. Frequencies of serious infection were similar in each treatment group: 21%, 15%, and 17% in the belatacept plus mycophenolate, belatacept plus sirolimus, and tacrolimus plus mycophenolate groups, respectively. Several studies have demonstrated that belatacept is as efficacious as CNIs in preventing acute rejection and is associated with improved outcomes in GFR for renal transplant recipients. 34-38 Substitution of belatacept for a CNI may spare renal transplant patients from the adverse nephrotoxic and cardiovascular effects of CNIs. A Phase II, randomized, open-label study was conducted to evaluate conversion from a CNI-based regimen to belatacept in renal transplant recipients. 39 A total of 171 patients who had undergone renal transplantation 6 36 months prior, were receiving CNI-based immunosuppression, and had stable renal function were randomized to receive one of two treatments: conversion to belatacept 5 mg/kg with CNI discontinuation (n = 83) or continued CNI therapy (n = 88). All patients remained on their maintenance immunosuppressive agents, including mycophenolate, sirolimus, azathioprine, and corticosteroids. At month 12, the primary outcome of mean ± S.D. change in GFR (as calculated with the Modification of Diet in Renal Disease formula) from baseline increased by 7.0 ± 11.99 ml/min in the belatacept group and by 2.1 ± 10.34 ml/min in the CNI group. Six patients (7%) who underwent belatacept conversion experienced an acute rejection episode within the first 6 months versus none in the CNI group. The frequency of patient and graft survival was 100% in the belatacept group and 99% in the CNI groups. The overall safety profile was similar in each treatment group. mtor inhibitors Pharmacology. The mtor inhibitors sirolimus and everolimus are macrolide antibiotics that inhibit lymphocyte activation and proliferation. 2,16,40 Intracellularly, both agents form a complex with FKBP-12, subsequently binding to and modulating the activity of mtor, a key regulatory kinase in cytokine-dependent T-cell proliferation. Inhibition of mtor results in the arrest of the cell-division cycle in the G 1 -to-s phase. Both mtor inhibitors also affect hematopoietic and nonhematopoietic cells. Drugs and dosages. Sirolimus is available only in an oral formulation. 16 When initiating sirolimus in de novo renal transplant recipients with a low-to-moderate immunologic risk, a sirolimus loading dose of 6 mg be given, followed by a maintenance dosage of 2 mg daily. 16,40 Although maintenance dosages of 5 mg/day, after a 15-mg loading dose, have been studied for this indication, no efficacy advantage over the 2-mg/day dosage has been established. For renal transplant recipients with a high immunologic risk, a sirolimus loading dose of 15 mg is recommended, followed by a maintenance dosage of 5 mg daily. 16 Despite the manufacturer s dosing recommendations, loading doses are rarely used in clinical practice, and loading and maintenance doses are often lower than those that appear in the drug s FDA-approved labeling. Everolimus was recently approved by FDA for the prevention of rejection in low-to-moderate-risk renal transplant recipients when given in conjunction with basiliximab induction, reduced-dosage cyclosporine, and corticosteroids. 17,41 The recommended starting dosage of everolimus is 0.75 mg twice daily, without the need for a loading dose. 17 TDM. An excellent correlation exists between whole-blood trough concentrations and the AUC for sirolimus. 42 The target sirolimus trough ranges from 5 to 24 ng/ml; however, in clinical practice, this target is highly dependent on institution-specific protocols. 16,42 In low-to-moderate immunologic risk patients, cyclosporine withdrawal is recommended, and the target sirolimus trough range is 16 24 ng/ml within the first year posttransplantation. 16 Thereafter, target trough sirolimus concentrations should be 12 20 ng/ml. In patients with a high immunologic risk, sirolimus doses should be adjusted to obtain troughs of 10 15 ng/ml in combination with cyclosporine and corticosteroids. Additional therapeutic ranges for various immunosuppression strategies can be found 1966

in the package insert. 16 Due to the dose-dependent adverse effects often associated with sirolimus, higher therapeutic ranges of sirolimus are not encouraged. Sirolimus has a half-life of approximately 62 hours, so patients receiving maintenance sirolimus who require a dosage adjustment will not achieve a new steady-state level for several days. Thus, trough serum levels should be monitored five to seven days after the dosage adjustment. The administration of a loading dose generally helps achieve a steady-state trough within 24 hours for de novo renal transplant recipients; however, it is not recommended to use sirolimus immediately after renal transplantation. Despite the approved TDM recommendations, sirolimus is often used with target trough concentrations below those recommended by the manufacturer, and its use is often dictated by center-specific protocols, adjuvant immunosuppression, and patient-specific immunologic risk. For everolimus, the recommended target trough is 3 8 ng/ml. 17,41 With a significantly shorter half-life, steady-state everolimus levels can be reached in 90 150 hours. 17 Adverse events. Some of the most frequently reported adverse events associated with mtor inhibitors are listed in Appendix A. Early posttransplantation complications of sirolimus, particularly the potential to prolong or increase the occurrence of delayed graft function, as well as poor wound healing, lymphocele formation, pneumonitis, and mucositis, have limited the de novo use of this agent. 39,40 Everolimus is indicated for de novo use despite the potential to have a toxicity profile similar to that of sirolimus. Proteinuria and glomerulonephropathy have been reported with mtor inhibitors, especially after conversion from a CNI. 43-48 Other important, yet rare, adverse events with these agents include hemolytic uremic syndrome, thrombotic microangiopathy, and liver dysfunction. 16,17,40 Clinical efficacy. The efficacy of sirolimus for primary maintenance immunosuppression after renal transplantation is well documented. Early studies focused on its use as an adjunct agent in combination with cyclosporine and found higher rates of impaired renal function. Newer studies have focused on its use as a primary immunosuppressant to replace CNIs. The clinical use of sirolimus has been investigated in de novo transplant recipients with CNI avoidance regimens and CNI conversion protocols. De novo utilization strategies. In a single-center, prospective trial, Flechner and colleagues 49 randomized 61 renal transplant recipients to receive either concentrationcontrolled sirolimus (target trough of 10 12 ng/ml for the first six months, then 5 10 ng/ml thereafter) or cyclosporine (target trough of 200 250 ng/ml) in addition to basiliximab induction, mycophenolate, and prednisone. There was significantly improved renal function at one, two, and five years in the sirolimus group versus the cyclosporine group (66.7 ml/min versus 50.7 ml/ min at five years, p = 0.0075). 49-51 In contrast, several retrospective analyses have found sirolimus-containing regimens to be associated with significantly worse long-term outcomes. 52,53 The largest of these studies was a retrospective analysis of the Scientific Registry of Renal Transplant Recipients that included over 2000 patients receiving immunosuppression with sirolimus and mycophenolate. 52 Compared with several regimens involving the combination of CNI with either mycophenolate or sirolimus, the combination of sirolimus and mycophenolate was associated with higher six-month acute rejection rates compared with other regimens (16% versus 11.2%, p < 0.01) and lower five-year graft survival (64% versus 78%, p = 0.001). For deceased-donor renal transplant recipients, delayed graft function was highest in the sirolimus and mycophenolate group (47% versus 27%, p < 0.001). Inconsistent outcomes may be attributable to several factors, including population size and trial design. However, several investigators have alluded to the importance of maintaining higher trough sirolimus levels for the first six months. In the SYMPHONY trial, low-dose sirolimus (target trough of 4 8 ng/ ml) resulted in the highest rates of acute rejection compared with the other treatment groups. 29,30 Most recently, the ORION study demonstrated that there was no benefit associated with sirolimus-based immunosuppression in regimens that either withdraw or completely avoid CNIs. 53 The ORION study was an open-label multicenter trial that randomized 469 renal transplant recipients to receive sirolimus and tacrolimus with a subsequent tacrolimus withdrawal at postoperative week 13 (group 1), sirolimus and mycophenolate (group 2), or tacrolimus and mycophenolate (group 3). All patients received IL-2 receptor antagonist induction and corticosteroids. Overall, there was no difference between patient or graft survival or renal function at one and two years. The frequency of acute rejection was highest in group 2, most of which occurred during the first six months of treatment and led to the termination of the study in that group. The rate of adverse effects that necessitated drug discontinuation was higher in the sirolimus-treated groups. Overall, these data, along with the potential for early posttransplantation complications with sirolimus, have limited the use of this agent in de novo renal transplant recipients at some centers. Several Phase II and III studies have demonstrated the efficacy and safety of everolimus when used with a cyclosporine-sparing regimen in de novo renal transplant recipients. 41,54 These studies also demonstrated that 1967

use of everolimus with reduced-dose cyclosporine preserves renal function without the loss of efficacy compared with full-dose cyclosporine regimens. Conversion strategies. Due to inconclusive results and an increased frequency of delayed graft function with de novo sirolimus regimens, recent studies evaluated newer regimens that entailed early (three to six months posttransplantation) and late (more than six months posttransplantation) CNI withdrawal and conversion to sirolimus. The Sparethe-Nephron trial was a two-year, open-label, multicenter trial evaluating preservation of renal function in a CNI-free regimen. 55 Two hundred ninety-nine patients were randomized to remain on standard immunosuppression, mycophenolate mofetil and a CNI (n = 151), or convert to mycophenolate and sirolimus, with a target sirolimus trough of 5 10 ng/ml (n = 148) between 30 180 days posttransplantation. At one year after the conversion, the mean change in the GFR was significantly higher in the sirolimus group (24.4% versus 5.2%, p = 0.012); however, the change was indistinguishable by two years. There was no difference in acute rejection rates. Of note, 41 patients were switched back from mycophenolate plus sirolimus to mycophenolate plus a CNI, which may be a reason why the change in the GFR did not persist at two years. Several randomized trials have evaluated late conversion to sirolimus in the setting of CNI toxicity or development of chronic graft nephropathy. The CONVERT trial was similar in design except conversion took place between 6 and 120 months posttransplantation. 56 Investigators identified that converted patients with a GFR of >40 ml/min and a urine protein:creatinine ratio of <0.11 at the time of conversion had significantly higher GFRs compared with the CNI continuation group (62.6 ml/min versus 59.9 ml/min, p = 0.009), although acute rejection rates were similar between the groups. Conversely, the investigators halted enrollment of patients with GFRs of 20 40 ml/min due to increased frequency of death and adverse events. Therefore, the timing of sirolimus conversion is crucial and may be most beneficial before the deterioration of kidney function. In addition to the renal-sparing properties of mtor inhibitors, there is mounting evidence that these agents inhibit angiogenesis and vascular endothelial growth factor, thus preventing cancer development or slowing cancer cell proliferation, particularly in skin cancers, lymphoma, and renal cell carcinoma. 10,41 Although these trials are not discussed in this review, sirolimus and everolimus may be considered in patients at high risk of cancer and de novo cancer after transplantation. 41 Practice guidelines recommend caution with the use of mtor inhibitors, as there are minimal data to support any net benefit over harm. 10 The use of mtor inhibitors should be avoided in patients with persistent urinary protein excretion of more than 500 1000 mg/day and in patients with significant dyslipidemias. Therefore, if conversion is necessary, a baseline spot urine sample to determine the protein:creatinine ratio and a fasting lipid panel are recommended to determine whether the patient is a candidate for an mtor-based regimen. Other potential roles for mtor inhibitor utilization include patients who have histologically proven CNI toxicity despite low levels of CNI and possibly in the setting of malignancy that is in remission or being actively treated. 10 Antiproliferatives The antiproliferative agents are generally considered to be adjuvant to the CNIs. The primary antiproliferative agents are azathioprine and the mycophenolic acid derivatives, though other antiproliferatives, such as cyclophosphamide and leflunomide, have been used in transplantation. 2 Azathioprine was first approved by FDA in 1968 as an immunosuppressant in renal transplant recipients. Before the advent of cyclosporine, the combination of azathioprine and corticosteroids was the mainstay of immunosuppressive therapy. 2 Over the past few decades, the use of azathioprine has declined markedly, largely due to the success of the mycophenolic acid derivatives, which are more specific inhibitors of T-cell proliferation. 2,10 Pharmacology. Azathioprine is a prodrug for the purine analogue 6-mercaptopurine (6-MP). 1,2,18 6-MP acts as an antimetabolite after its incorporation into the cellular DNA, where it alters the synthesis and function of RNA with a resultant reduction in T-cell proliferation. Azathioprine is also myelosuppressive and can inhibit the proliferation of promyelocytes, which can reduce the number of circulating monocytes. 1,2,18 Mycophenolate mofetil is a prodrug that is rapidly hydrolyzed to mycophenolic acid, mostly by the liver, after oral absorption. 1,2,19,57 Enteric-coated mycophenolic acid is absorbed in the intestine as the active moiety. Mycophenolic acid acts by inhibiting inosine monophosphate dehydrogenase, a vital enzyme in the de novo pathway of guanosine nucleotide synthesis. Inhibition of this enzyme prevents the proliferation of most cells that are dependent on the de novo pathway for purine synthesis, including lymphocytes. 19 Other rapidly dividing cell lines are capable of purine synthesis via the salvage pathway, which is not affected by mycophenolic acid. 1,2,19,57 Drugs and dosages. Azathioprine is available in oral and i.v. formulations. 18 The typical oral dosage of azathioprine for transplant recipients is 3 5 mg/kg once daily when used as the primary immunosuppressant. When used in conjunction with CNIs, lower dosages (1 2 mg/kg daily) are sufficient. Dosage reduc- 1968

tions may be necessary in patients with severely impaired renal function, since 6-MP and its metabolites are eliminated by the kidneys. Mycophenolate mofetil was approved by FDA in 1995, and entericcoated mycophenolate sodium was approved in 2004. Mycophenolate mofetil is available in oral and injectable formulations. 19 The recommended dosage of mycophenolate mofetil in renal transplant recipients is 2000 mg daily in two divided doses. The bioavailability of the oral dosage forms of mycophenolate mofetil exceeds 90%; therefore, conversion between oral and i.v. mycophenolate mofetil is 1:1. I.V. mycophenolate mofetil must be reconstituted and diluted to a concentration of 6 mg/ml using 5% dextrose injection only. Administration must be by slow i.v. infusion over no fewer than two hours, and rapid or bolus administration should be avoided. The recommended starting dose of enteric-coated mycophenolate sodium is 720 mg (of mycophenolic acid) given twice daily, as this dosage provides the equivalent equimolar mycophenolic acid concentrations seen with mycophenolate mofetil 1000 mg twice daily. 19,57 Enteric-coated mycophenolate sodium was developed in an attempt to decrease the occurrence of adverse upper-gastrointestinal-tract events associated with mycophenolate mofetil. 58 Both medications have been studied head-to-head in de novo and stable renal transplant recipients. 59-62 In both analyses, efficacy and tolerability outcomes were comparable between the formulations. Most notably, there were no significant differences in adverse gastrointestinal effects between the two formulations. An open-label study found distinct benefits in the relief of gastrointestinal symptoms with enteric-coated mycophenolate sodium; however, this study was not randomized and its data can only be regarded as indicative. 63 Patients complaining of significant gastrointestinal distress associated with mycophenolate mofetil were converted to an equimolar dose of enteric-coated mycophenolate sodium, resulting in significantly lower rates of gastrointestinal symptoms and an improved quality of life. Adverse events. The most common adverse events associated with these agents are listed in Appendix A. Myelosuppression (mainly leukopenia and thrombocytopenia) is a frequent, dose-dependent, and dose-limiting complication (>50% of patients) that often necessitates dosage reductions. 18,19 Importantly, pancreatitis and veno-occlusive disease of the liver occur in fewer than 1% of patients after chronic azathioprine therapy. 18 The mycophenolic acid derivatives are considered to have excellent tolerability compared with azathioprine. TDM. Mycophenolic acid has a very complex pharmacokinetic profile; therefore, interpretation of concentrations is difficult. 64 Results regarding actual improvement in efficacy and safety outcomes when using TDM for patients receiving mycophenolic acid are conflicting. Three large randomized prospective trials have attempted to address the role of TDM for mycophenolic acid. 65-67 The APOMYGRE study was a 12-month study that included 11 French centers and randomized 137 patients to either fixed-dose (2 g/day) or concentration-controlled (target mycophenolic acid AUC of 40 mg hr/l) oral mycophenolate mofetil in a quadruple immunosuppression regimen consisting of basiliximab, cyclosporine modified, and corticosteroids. 65 The primary endpoint was the frequency of treatment failure, defined by a composite score of death, graft loss, acute rejection, and discontinuation of mycophenolate mofetil therapy. Mean ± S.D. mycophenolate mofetil doses in the concentrationcontrolled group were 2969 ± 780 mg, 2279 ± 878 mg, and 1827 ± 654 mg at months 1, 3, and 12, respectively. The frequency of treatment failure was significantly lower in the concentration-controlled group compared with the fixed-dose group (29.2% versus 47.7%, p = 0.03). Interestingly, 7 of 10 rejection episodes occurring in the first 3 months of the concentration-controlled group were associated with a mycophenolic acid AUC value of <30 mg hr/l. However, 7 of 18 late rejection episodes (occurring after 3 months) were associated with a mycophenolic acid AUC of >45 mg hr/l. Rates of reporting at least one adverse effect were similar for patients reporting at least one adverse event in both groups (90% in the controlled-concentration group versus 97% in the fixed-dose group), and there were no differences in frequency or severity of infectious complications between the two groups. In contrast, the Fixed Dose versus Concentration Control (FDCC) study compared fixed-dose oral mycophenolate mofetil (2 g daily) with a concentration-controlled regimen (target mycophenolic acid AUC of 30 60 mg hr/l) with either cyclosporine or tacrolimus in 901 patients. 66 The primary endpoint was similar to that in the APOMYGRE trial. During the study, 54% of patients received cyclosporine, 46% received tacrolimus, and 46% of all patients received induction therapy. There was no significant difference in graft or patient survival, BPAR, treatment failure composite endpoint, rates of adverse effects, infection, or malignancy between the groups. This was expected due to the lack of difference in mycophenolic acid exposure between the groups. Nonetheless, the risk of developing BPAR in the first year after transplantation was 18.8% in patients with a day 3 mycophenolic acid AUC of <30 mg hr/l, stressing the importance of targeting optimal concentrations as early as possible. Mycophenolic acid exposure was below the target in 37.3% of patients by day 3 in both groups and in 34.8% of patients in 1969

the concentration-controlled group. As expected, mycophenolic acid exposure was lower in the cyclosporine subgroup, with only 51.2% of the cyclosporine subgroup reaching a mycophenolic acid AUC of >30 mg hr/l, compared with 76.2% in the tacrolimus group. Throughout all time points of the study, the cyclosporine subgroup had a lower mean mycophenolic acid AUC compared with the tacrolimus group, despite the higher mean daily doses of mycophenolate used. Disparities in outcomes between the two large trials may have been influenced by different methods used to calculate the AUC as well as failure to reach target AUC levels in the FDCC trial. The Opticept trial evaluated the safety and efficacy of mycophenolic acid TDM in patients receiving either cyclosporine or tacrolimus in whom mycophenolic acid levels were measured 12 hours before the administration of mycophenolate mofetil. 67 In this two-year, openlabel, prospective trial, 720 renal transplant recipients were randomized to receive one of the following regimens: concentration-controlled oral mycophenolate mofetil and reduced CNI (group A), concentrationcontrolled mycophenolate mofetil and standard CNI (group B), or fixed-dose mycophenolate mofetil (1 g twice daily) and standard-dose CNI (group C). In groups A and B, mycophenolate mofetil dosages were adjusted to achieve mycophenolic acid trough concentrations of 1.3 mg/ml if receiving cyclosporine and 1.9 mg/ml if receiving tacrolimus. The primary endpoint was treatment failure, a composite defined similarly to that in the previous trials. There was no difference between groups when standard-dose CNI was used; however, group A had fewer treatment failures than did the two other groups (22.6% versus 28 29%, p = 0.18). Of note, mycophenolic acid 12-hour predose concentrations of 1.6 mg/ml were associated with longer times to the first BPAR. The results of this trial suggest that TDM for mycophenolic acid may be suitable in CNI-minimization regimens. However, it is difficult to decipher whether the low-dose CNI or concentration-controlled mycophenolate mofetil influenced the positive outcome of this trial. Many of the studies evaluating the mycophenolic acid AUC have calculated the AUC using limitedsampling strategies (two to four samples at different times after dose administration). 68 It is important to note that given the differences in absorption characteristics between mycophenolate mofetil and mycophenolate sodium, the limited-sampling equations have not been validated in calculating AUCs resulting from the use of mycophenolate sodium. Recently updated guidelines from the 2008 Transplantation Society consensus meeting provided recommendations on the specific patient populations for whom TDM is appropriate. 69 Recommendations for the target mycophenolic acid AUC remain 30 60 mg hr/l for reduced risk of acute rejection. 68,69 Furthermore, the consensus statement recommends a trough mycophenolic acid concentration of 1.3 mg/l when cyclosporine is used and 1.9 mg/l when tacrolimus is used; these recommendations were based on the assumption that achieving these trough concentration would ensure that at least 80% of patients reached a mycophenolic acid AUC of >30 mg hr/l. Close estimates of the AUC can be obtained by using limited-sampling strategies over two to three hours. Van Gelder and colleagues 68 outlined possible acceptable algorithms. The use of mycophenolic acid and azathioprine TDM in renal transplant recipients is not routinely recommended; however, for patients taking regimens involving CNI minimization or high-risk patient populations, higher mycophenolic acid exposure may be necessary. 68,69 Comparative efficacy. The mycophenolic acid derivatives have largely replaced azathioprine as the anti proliferative agent of choice in most renal transplant centers. Most of the initial studies comparing mycophen olate mofetil with azathioprine found superior acute rejection rates with mycophenolate in combination with the original cyclosporine formulation. 70 A pooled efficacy analysis of early trials using the original cyclosporine formulation revealed a significant reduction in the frequency of acute rejection and an improvement in short-term renal function, though no benefit in patient or graft survival was observed when compared with azathioprine. 71 Despite earlier trials that have led to the replacement of azathioprine with mycophenolate, several trials have yielded conflicting data. 72-76 In the MYSS trial, Remuzzi and colleagues 72 compared mycophenolate mofetil with azathioprine in conjunction with cyclosporine (modified) and corticosteroids in 336 renal transplant patients. The frequency of acute rejection and graft loss was similar between the two groups in both phases. Renal function and adverse events were also comparable between myco phenolate- and azathioprine-treated patients at all time points. A five-year follow-up study of these patients was conducted and found a comparable rate of patient survival, graft loss, and late rejection. 73 In addition, renal function and the rate of adverse events were similar with both agents. The authors concluded that the previous trials demonstrated greater efficacy with mycophenolate in conjunction with the original cyclosporine formulation. 72,73 Due to the lower acquisition cost of azathioprine versus mycophenolate mofetil, the authors suggested the consideration of using azathioprine over mycophenolate for maintenance immunosuppression. Shah et al. 74 performed a pairedkidney analysis in the United Kingdom 1970