Prevention of Acute Rejection and Allograft Vasculopathy by Everolimus in Cardiac Transplants Recipients: A 24-Month Analysis

ALLOGRAFT VASCULOPATHY

Prevention of Acute Rejection and Allograft Vasculopathy by Everolimus in Cardiac Transplants Recipients: A 24-Month Analysis Mario Viganò, MD,a Murat Tuzcu, MD,b Raymond Benza, MD,c Pascale Boissonnat, MD,d Axel Haverich, MD,e James Hill, MD,f Guenther Laufer, MD,g Robert Love, MD,h Jayan Parameshwar, MD,i Luis Alonso Pulpón, MD,j Dale Renlund, MD,k Kamal Abeywickrama, PhD,l Nathalie Cretin, MD,l Randall C. Starling, MD,b and Howard J. Eisen, MD,m for the RAD B253 Study Group Background: Everolimus is an immunosuppressive agent that reduces cardiac allograft vasculopathy. This report presents the 24-month results of a multicenter trial of everolimus vs azathioprine in heart transplantation. Methods: A total of 634 patients were randomized to receive 1.5 mg everolimus, 3 mg everolimus or azathioprine, with cyclosporine and steroids. A 12-month, double-blind, double-dummy period was followed by a 12-month open-label period. Results: At 24 months, the percentage of patients reaching the composite efficacy end-points was significantly lower with everolimus (1.5 mg: 45.9%, p ⫽ 0.016; 3 mg: 36.0%, p ⬍ 0.001) than with azathioprine (57.5%). The change in maximal intimal thickness from baseline to 24 months was significantly smaller with everolimus 1.5 mg (0.07 mm, p ⫽ 0.014) and 3 mg (0.06 mm, p ⫽ 0.004) compared with azathioprine (0.15 mm). The 24-month incidence of vasculopathy was 33.3% with everolimus 1.5 mg, 45.5% with everolimus 3 mg and 58.3% with azathioprine ( p ⫽ 0.017 vs everolimus 1.5 mg). Incidence of cytomegalovirus infection was 3-fold lower in patients receiving everolimus compared with azathioprine (7.2% and 7.1% in the 1.5-mg and 3-mg everolimus cohorts, respectively, and 21% in the azathioprine group; p ⬍ 0.0001). Median serum creatinine levels at 24 months were higher with everolimus than with azathioprine, but decreased when cyclosporine exposure was reduced (everolimus 1.5 mg: baseline 167 ␮mol, after 6 months 157.5 ␮mol; everolimus 3 mg: baseline 185.6 ␮mol, after 6 months 160 ␮mol; azathioprine: baseline 123.3 ␮mol, after 6 months 127 ␮mol). Conclusions: Everolimus significantly reduced acute rejection and limited the progression of allograft vasculopathy at 24 months compared with azathioprine. Although graft and patient survival was comparable at 24 months, everolimus therapy may improve longer-term outcomes after heart transplantation. J Heart Lung Transplant 2007;26:584 –92. Copyright © 2007 by the International Society for Heart and Lung Transplantation.

Long-term outcomes after heart transplantation are limited by cardiac allograft vasculopathy, which is the major cause of death in patients surviving the first year post-transplant.1,2 Vasculopathy is characterized by myFrom the aDepartment of Cardiac Surgery, San Matteo Hospital, Pavia, Italy; bDepartment of Cardiovascular Medicine, Cleveland Clinic Foundation, Cleveland, Ohio; cDivision of Cardiology, University of Alabama, Birmingham, Alabama; dService de Chirurgie et Cardiovasculaire, Hôpital Louis Pradel, Bron, France; eDivision of Thoracic and Cardiovascular Surgery, Hannover Medical School, Hannover, Germany; fCardiovascular Center, Shands Hospital, University of Florida, Gainesville, Florida; gDepartment of Cardiac Surgery, University Klinik für Chirurge, Innsbruck, Austria; hDivision of Cardiothoracic Surgery, Clinical Science Center, Madison, Wisconsin; iCardiopulmonary Transplant Unit, Papworth Hospital, Cambridge, UK; jDepartment of Cardiology, Clinica Puerto de Hierro, Madrid, Spain; kDivision of Cardiology, University of Utah, Salt Lake City, Utah; lNovartis Pharma AG, Basel, Switzerland; and mDrexel University College of Medicine, Philadelphia, Pennsylvania.

584

ointimal proliferation and vascular remodeling, leading to ischemia and myocardial dysfunction.3 Intravascular ultrasonography, a sensitive diagnostic technique for detecting early vascular lesions, has demonstrated the occurrence of intimal thickening in a majority of car-

Submitted November 12, 2005; revised February 26, 2007; accepted March 9, 2007. Presented at 24th annual scientific session of the International Society for Heart and Lung Transplantation, April 2004, San Francisco, California. Supported by grant from Novartis Pharma AG, Basel, Switzerland. Reprint requests: Howard J. Eisen, MD, Division of Cardiology, Drexel University College of Medicine, 245 North 15th Street, MS #1012, Philadelphia, PA 19102. Telephone: 215-762-3829. Fax: 215762-4503. E-mail: [email protected] Copyright © 2007 by the International Society for Heart and Lung Transplantation. 1053-2498/07/$–see front matter. doi:10.1016/ j.healun.2007.03.005

The Journal of Heart and Lung Transplantation Volume 26, Number 6

diac recipients within 1 year of transplantation.4 Moreover, the extent of intimal thickening at 1 year is predictive of 5-year survival.5–7 Recently introduced, potent immunosuppressive agents have enhanced short-term results after transplantation but have failed to influence allograft vasculopathy and long-term graft survival.8 There is therefore a pressing need for immunosuppressive strategies that prevent both acute rejection and vascular proliferation leading to late graft loss. Everolimus (Certican, Novartis Pharma AG, Basel, Switzerland) is a proliferation signal inhibitor that blocks growth factor-driven proliferation of hematopoietic and non-hematopoietic cells. The potential of everolimus to target allograft vasculopathy has been demonstrated in pre-clinical studies, where everolimus was shown to inhibit proliferation of vascular smooth muscle cells and prevent myointimal thickening.9 –11 Everolimus and cyclosporine display immunosuppressive synergy, possibly because they suppress lymphocyte proliferation by different mechanisms.12–14 The use of everolimus in heart transplantation has been evaluated in a 2-year, multi-center, randomized clinical trial that compared everolimus with azathioprine in de novo cardiac transplant recipients. At 12 months, everolimus was superior to azathioprine with respect to incidence of acute rejection and incidence of allograft vasculopathy.15 The trial protocol was amended after all patients had completed 12 months because a high prevalence of renal dysfunction was unexpectedly observed in patients who received everolimus. Investigators were unblinded to optimized patient management and reduced cyclosporine exposure was targeted in transplant recipients with altered renal function. The 24-month results from the trial are presented. METHODS Study Design In this 2-year, multicenter, randomized, parallel group study we compared the efficacy and safety of two oral doses of everolimus (1.5 mg/day or 3 mg/day) vs azathioprine in combination with cyclosporine microemulsion (Neoral) and corticosteroids in de novo heart transplant recipients. At the time of the study design, azathioprine was the standard adjunctive agent used in cardiac transplantation (mycophenolate mofetil was not available in all countries participating in the study). The first year was a double-blind, double-dummy period, which was followed by a 1-year, open-label period. During the second year, an amendment to the protocol was required because of unanticipated renal toxicity, prompting a reduction of cyclosporine trough levels in patients with altered renal function. This was done because previous renal transplant studies demonstrated

Viganò et al.

585

improved renal function in patients on everolimus and reduced-exposure cyclosporine and comparable outcomes with full-exposure cyclosporine regimens.16,17 The 24-month analysis addressed the following objectives: (1) to compare the efficacy of everolimus with azathioprine at 24 months, as measured by the incidence of composite efficacy end-points, defined as death, graft loss/re-transplant, biopsy-proven acute rejection episode of ISHLT Grade ⱖ3A, clinical rejection associated with hemodynamic compromise, loss to follow-up, and acute rejection episodes treated with anti-lymphocyte antibodies; (2) to evaluate the incidence of allograft vasculopathy at 24 months as assessed by intravascular ultrasound in patients receiving everolimus or azathioprine; (3) to assess the safety and tolerability of everolimus vs azathioprine; and (4) to evaluate renal function in patients treated under the amended protocol. Study Medication Within 72 hours after transplantation, patients were randomly assigned to receive everolimus 0.75 mg twice daily, everolimus 1.5 mg twice daily or azathioprine (1 to 3 mg/kg/day) in a double-blind, double-dummy format for 12 months and in an open-label format for another 12 months. Cyclosporine doses were adjusted to maintain cyclosporine trough blood levels at within target ranges (Weeks 1 to 4: 250 to 400 ng/ml; Months 1 to 6: 200 to 350 ng/ml; Months 7 to 24: 100 to 300 ng/ml). Corticosteroids were initiated at 0.5 to 1.0 mg/kg/day and tapered to no less than 0.1 mg/kg/day by 6 months. Centers were designated as induction or non-induction sites based on their preference for use of induction therapy; induction sites were required to treat all patients with anti-thymocyte globulin (ATG) or muromonab CD3 (anti-CD3 murine antibody, OKT3) for up to a maximum of 3 days. Each site used either prophylactic- or antigenemia-based approaches for cytomegalovirus (CMV) therapy and was then required to treat all patients accordingly. All patients received statin therapy (pravastatin, simvastatin or fluvastatin administered to maintain a low-density lipoprotein [LDL] level ⬍130 mg/dl) irrespective of baseline LDL level. Patient Population Heart transplant recipients from 16 to 65 years of age were enrolled at 52 centers in North and South America and Europe.15 All patients gave written informed consent. The study was approved by local institutional review boards and was conducted in accordance with the Declaration of Helsinki. Exclusion criteria included: receipt of a donor heart ⬎60 years old or with known heart disease or cold ischemia time ⬎8 hours; previous transplant or ABO-incompatible transplant; hypercho-

586

Viganò et al.

lesterolemia ⱖ350 mg/dl or hypertriglyceridemia ⱖ750 mg/dl; panel-reactive antibodies ⱖ20%; white blood cell count ⬍5,000/mm3; or platelet count ⱕ70,000/mm3. Evaluation Efficacy and safety evaluations were performed during the baseline period before study medication was initiated, at several follow-up visits during the first year, and at 18 and 24 months. Composite efficacy end-point (defined by incidence of death, graft loss/re-transplant, biopsy-proven acute rejection episode ISHLT Grade ⱖ3A, clinical rejection associated with hemodynamic compromise, and loss to follow-up) was determined at 24 months. Graft loss included graft failure, death arising from a cardiac event, or unexplained sudden death. Hemodynamic compromise was defined as a left ventricular ejection fraction of ⱕ30% or ⱖ25% lower than baseline, fractional shortening ⱕ20% or ⱖ25% lower than baseline, or treatment with inotropes. An endomyocardial biopsy was obtained by protocol at 18 and 24 months and to confirm any suspected rejection episode. An echocardiogram was performed to determine whether rejection was associated with hemodynamic compromise. Allograft vasculopathy at 24 months was assessed by measuring coronary artery intimal proliferation at baseline (i.e., within 6 weeks after transplantation) and at 24 months at matched sites using intravascular ultrasonography with automatic pull-back.15 Only patients who had had baseline assessment underwent 24 months examination. Analysis of registration was performed by core lab investigators blinded to patient immunosuppressive treatment. For the intravascular ultrasound studies, efficacy variables included the mean change in maximal intimal thickness from baseline to Month 24 and the incidence of allograft vasculopathy (defined as an increase in maximal intimal thickness from baseline of ⱖ0.5 mm in one matched slice). Protocol Amendment for Management of Renal Function In everolimus-treated patients with evidence of renal dysfunction (elevated serum creatinine levels or other renal function assessments), trough levels of cyclosporine were reduced progressively to approximately 100 ng/ml during the open-label phase of the study. The definition of renal dysfunction was mainly based on single-investigator clinical judgment and patients were enrolled in the amendment protocol when their creatinine levels had increased significantly compared with pre-transplant levels. Doses of everolimus were adjusted to maintain trough concentrations of ⱖ3 ng/ml, especially before reduction of cyclosporine exposure. Immunosuppression remained unchanged in everoli-

The Journal of Heart and Lung Transplantation June 2007

mus-treated patients with satisfactory renal and allograft function. Patients on azathioprine who had elevated serum creatinine levels were treated as considered appropriate by the investigator. Statistical Analysis Efficacy analyses were conducted on the intent-to-treat (ITT) population, which consisted of all randomized patients who had at least one assessment after administration of study medication. Safety and tolerability analyses were conducted on all randomized patients who received at least one dose of study medication with at least one safety assessment. Adjustments for multiple comparisons of the primary efficacy failure proportions (Z-test) only were made using Hochberg’s modified Bonferroni18 procedure to maintain an overall 2-sided Type I error rate of 0.05. A sample size of 210 per treatment arm was estimated to have a power of 80% to detect a 15% difference in primary efficacy end-point. All other analyses were unadjusted for multiple comparisons and at a 0.05 (2-sided) level. Secondary efficacy variables included the proportions and Kaplan–Meier estimates of the probability of primary efficacy failure within 12 and 24 months as well as its individual components, rejections treated with antibodies, and the presence of vasculopathy at 12 and 24 months.19 The primary intravascular ultrasound efficacy variable was mean change over the matched slices in maximum intimal thickness from baseline (Wilcoxon rank-sum test); secondary end-points included incidence of vasculopathy (Fisher’s Exact test). Safety data were analyzed using the Wilcoxon ranksum test for continuous variables and Fisher’s Exact or chi-square tests for categorical data for between-group comparisons. Role of the Funding Source Novartis Pharma AG developed the protocol with the investigators, performed the data analysis, and commented on the manuscript before submission for publication. RESULTS Patient Population In total, 634 patients were randomized to everolimus 1.5 mg (n ⫽ 209), everolimus 3 mg (n ⫽ 211) or azathioprine (n ⫽ 214). Patients in the three groups had similar baseline demographic and clinical characteristics, as described previously,15 with no significant between-group differences in terms of age, gender, race, indication for transplantation, CMV status, use of induction therapy, donor age or cold ischemia time. Five hundred forty-seven patients (86%) completed the 24-

The Journal of Heart and Lung Transplantation Volume 26, Number 6

Viganò et al.

587

Table 1. Patient Disposition (ITT Population) at 24 Months

Receiving study medication Discontinued study medication Adverse event(s) Abnormal laboratory value(s) Unsatisfactory therapeutic effect Protocol violation Withdrawn consent Lost to follow-up Administrative problems Death Discontinued study Withdrawn consent Lost to follow-up Death a

p ⫽ 0.038 vs azathioprine;

b

Everolimus 1.5 mg (n ⫽ 209) 127 (60.8%) 82 (39.2%) 43 (20.6%) 9 (4.3%) 15 (7.2%) 2 (1.0%) 6 (2.9%) 0 0 7 (3.3%) 23 (11.0%) 2 (1.0%) 0 21 (10.0%)

Everolimus 3 mg (n ⫽ 211) 107 (50.7%) 104 (49.3%)a 58 (27.5%) 18 (8.5%) 3 (1.4%) 4 (1.9%) 11 (5.2%) 1 (0.5%) 0 9 (4.3%) 33 (15.6%) 3 (1.4%) 1 (0.5%) 29 (13.7%)

Azathioprine (n ⫽ 214) 131 (61.2%) 83 (38.8%) 40 (18.7%) 10 (4.7%) 18 (8.4%)b 2 (0.9%) 3 (1.4%) 2 (0.9%) 1 (0.5%) 7 (3.3%) 31 (14.5%) 5 (2.3%) 2 (0.9%) 24 (11.2%)

p ⫽ 0.0011 vs everolimus 3 mg.

month study, with ⬎50% of patients in each group continuing to receive study medication (Table 1). Immunosuppression At 24 months, mean daily dose was close to the planned dose in the everolimus 1.5 mg group (1.3 mg) and the azathioprine group (1.5 mg/kg), whereas in the everolimus 3 mg group the mean dose was 2.3 mg. Over the 24-month period, the mean daily dose of cyclosporine was significantly lower in both everolimus-treated groups (3.3 mg/kg in both groups) than in the azathioprine-treated group (4.0 mg/kg, p ⬍ 0.001 for both 1.5 mg and 3 mg everolimus vs azathioprine). Efficacy The 24-month incidence of composite efficacy endpoints was significantly lower in patients who received everolimus 1.5 mg (45.9%, p ⫽ 0.016) or everolimus 3 mg (36.0%, p ⬍ 0.001) than in patients who received azathioprine (57.5%) (Table 2). Of the individual efficacy components, the incidence of acute rejection episodes of Grade ⱖ3A was significantly reduced by treatment with everolimus. The frequency of rejec-

tion associated with hemodynamic compromise was not statistically different with everolimus compared with azathioprine. Patient and graft survival rates were similar in all groups. The frequency of composite efficacy end-points after the first 12 months (between Days 351 and 810) was low and comparable among the three groups (everolimus 1.5 mg, n ⫽ 11 [5.8%]; everolimus 3 mg, n ⫽ 10 [5.3%]; azathioprine, n ⫽ 12 [6.2%]). Allograft Vasculopathy The increase in average maximum intimal thickness from baseline to 24 months was significantly smaller in patients treated with everolimus (1.5 mg: 0.07 mm; 3 mg: 0.06 mm) compared with patients treated with azathioprine (0.15 mm) (p ⫽ 0.014 and p ⫽ 0.004, respectively). The 24-month incidence of allograft vasculopathy, defined as an increase in maximal intimal thickness of ⱖ0.5 mm, was significantly lower after treatment with everolimus 1.5 mg (33.3%) than after treatment with azathioprine (58.3%) (p ⫽ 0.017 for everolimus 1.5 mg vs azathioprine). The 24-month incidence of vasculopathy was not significantly lower

Table 2. Efficacy Outcomes at 24 Months (Includes All Events to Day 810)

Composite efficacy end-points Acute rejection ISHLT Grade ⱖ3A Acute rejection associated with HDC Graft loss Death Lost to follow-upd

Everolimus 1.5 mg (n ⫽ 209) 96 (45.9%) 73 (34.9%) 19 (9.1%) 10 (4.8%) 21 (10.0%) 0

Everolimus 3 mg (n ⫽ 211) 76 (36.0%) 48 (22.7%) 17 (8.1%) 14 (6.6%) 29 (13.7%) 0

Azathioprine (n ⫽ 214) 123 (57.5%) 103 (48.1%) 28 (13.1%) 13 (6.1%) 24 (11.25%) 2 (0.9%)

p-value 0.016,a ⬍0.001,b 0.038c 0.005,a ⬍0.001,b 0.005c NS NS NS NS

HDC, hemodynamic compromise; NS, not statistically significant. a Everolimus 1.5 mg versus azathioprine; bEverolimus 3 mg versus azathioprine; cEverolimus 1.5 mg versus everolimus 3 mg. d Lost to follow-up for the primary efficacy variable includes all patients who had no efficacy evaluations after Day 365 and who had not experienced acute rejection of ISHLT Grade ⱖ3A or associated with HDC, graft loss or death.

588

Viganò et al.

The Journal of Heart and Lung Transplantation June 2007

Similarly, from baseline to 24 months, the mean change in intimal area over matched segments was significantly smaller with everolimus 1.5 mg (0.79 mm2 [n ⫽ 45]) and everolimus 3 mg (0.83 mm2 [n ⫽ 44]) compared with azathioprine (1.52 mm2 [n ⫽ 60]) (p ⫽ 0.011 for everolimus 1.5 mg vs azathioprine, p ⫽ 0.012 for everolimus 3 mg vs azathioprine). The mean change in percentage cross-sectional area of stenosis from baseline to 24 months was also significantly lower with everolimus (1.5 mg: 4.32% [n ⫽ 45]; 3 mg: 4.22% [n ⫽ 44]) vs azathioprine (8.62% [n ⫽ 60]) (p ⫽ 0.007 for everolimus 1.5 mg vs azathioprine, p ⫽ 0.003 for everolimus 3 mg vs azathioprine). Tolerability

Figure 1. Intravascular ultrasound findings at 12 months (open bars) and 24 months (filled bars). (a) Mean change in maximal intimal thickness from baseline. (b) Incidence of vasculopathy (ⱖ0.5 mm increase in maximal intimal thickness). (a) Everolimus 1.5 mg vs azathioprine: 12 months, p ⫽ 0.014; 24 months, p ⫽ 0.014. Everolimus 3 mg vs azathioprine: 12 months, p ⫽ 0.003; 24 months, p ⫽ 0.004. Everolimus 1.5 mg vs everolimus 3 mg: 12 months, p ⫽ 0.491; 24 months, p ⫽ 0.673. (b) Everolimus 1.5 mg vs azathioprine: 12 months, p ⫽ 0.045; 24 months, p ⫽ 0.017. Everolimus 3 mg vs azathioprine: 12 months, p ⫽ 0.010; 24 months, p ⫽ 0.235. Everolimus 1.5 mg vs everolimus 3 mg: 12 months, p ⫽ 0.590; 24 months, p ⫽ 0.282.

with 3 mg everolimus 20 (45.5%) vs azathioprine. A comparison of findings at 12 and 24 months (Figure 1) shows that patients treated with everolimus continued to exhibit significantly less intimal thickening than patients treated with azathioprine throughout the course of the study.

Adverse events were experienced by all but two patients in the study by 24 months. The incidence of adverse events considered to be drug related was similar in both everolimus groups (1.5 mg: 70%; 3 mg: 73%) and lower in patients treated with azathioprine (63%). The most frequent adverse events related to study treatment were anemia, leukopenia, thrombocytopenia, hypercholesterolemia, hypertriglyceridemia, renal impairment and CMV infection. Non-fatal serious adverse events were reported by 150 patients (71.8%) in the everolimus 1.5 mg group, 162 patients (76.8%) in the everolimus 3 mg group, and 140 patients (65.4%) in the azathioprine group (p ⫽ 0.01 for everolimus 3 mg vs azathioprine), and the 3 mg everolimus-treated patients were more likely than the azathioprine-treated patients to discontinue study medication because of adverse events (Table 1, p ⫽ 0.038). Renal impairment, anemia and leukopenia were the most common reasons for discontinuation. The incidence of infection was similar with everolimus 1.5 mg and azathioprine, but higher for everolimus 3 mg (Table 3). Bacterial pneumonia occurred significantly more frequently with both doses of everolimus compared with azathioprine. Conversely, azathioprine treatment was associated with a 3-fold, significant increase in the rate of CMV infection (Table 3).

Table 3. Infections Occurring in ⱖ10% of Patients in Any Treatment Group (ITT Population)

All infections Pneumonia Upper respiratory tract infection Urinary tract infection Herpes simplex virus Cytomegalovirus a

Everolimus 1.5 mg (n ⫽ 209) 160 (76.6%) 29 (13.9%) 22 (10.5%) 18 (8.6%) 17 (8.1%) 15 (7.2%)

Everolimus 3 mg vs azathioprine (Fisher’s Exact test). b Everolimus 1.5 mg vs azathioprine (Fisher’s Exact test).

Everolimus 3 mg (n ⫽ 211) 169 (80.1%) 20 (9.5%) 22 (10.4%) 28 (13.3%) 12 (5.7%) 15 (7.1%)

Azathioprine (n ⫽ 214) 154 (72.0%) 6 (2.8%) 29 (13.6%) 23 (10.7%) 23 (10.7%) 45 (21.0%)

p value 0.023a ⬍0.0001b 0.0043c NS NS NS ⬍0.0001,b ⬍0.0001a

The Journal of Heart and Lung Transplantation Volume 26, Number 6

Table 4. Mean Serum Creatinine in Patients With 6-Month Data in the Amendment Protocol Serum creatinine Everolimus 1.5 mg Everolimus 3 mg Azathioprine (␮mol/liter) (n ⫽ 38) (n ⫽ 27) (n ⫽ 24) Baseline 164.4 182.7 138.0 At 6 months 162.9 189.5 135.2

Levels of total cholesterol, LDL-cholesterol, highdensity lipoprotein (HDL)-cholesterol and triglycerides increased in all three groups from baseline to 24 months, despite ⬎90% of patients receiving statin therapy. The simple incidence of hypercholesterolemia and hypertriglyceridemia was 13% and 4% (1.5 mg), 14% and 5% (3 mg), respectively, in the everolimus groups, compared to 4% and 1% with azathioprine (p ⱕ 0.01). Mean maximum LDL-cholesterol levels over the 24month study period were: everolimus (1.5 mg) 3.9 mmol/liter; everolimus (3 mg) 3.8 mmol/liter; and azathioprine 3.6 mmol/liter (p ⫽ not significant [NS]). At 24 months, mean levels of HDL-cholesterol were 1.2 mmol/liter, 1.1 mmol/liter and 1.2 mmol/liter in the everolimus 1.5 mg, everolimus 3 mg and azathioprine groups, respectively (p ⫽ NS). Renal Function In the ITT population, median serum creatinine levels were higher in the everolimus-treated groups at 24 months (177 ␮mol/liter with everolimus 1.5 mg, 168 ␮mol/liter with everolimus 3 mg, 133 ␮mol/liter in the azathioprine group; p ⬍ 0.001 for both everolimus groups vs azathioprine). In comparison, median serum creatinine at 12 months was 167 ␮mol/liter in the everolimus 1.5 mg group, 170 ␮mol/liter in the everolimus 3 mg group, and 141 ␮mol/liter in the azathioprine group. Mean serum creatinine at 12 and 24 months was 182 ␮mol/liter and 180 ␮mol/liter with everolimus 1.5 mg, 186 ␮mol/liter and 180 ␮mol/liter with everolimus 3 mg, and 148 ␮mol/liter at both time-points with azathioprine (p ⬍ 0.001 at 12 and 24 months for both everolimus doses vs azathioprine), respectively. A total of 170 patients participated in the amendment protocol that induced reduction of cyclosporine trough levels in patients with signs of renal dysfunction. Paired serum creatinine taken prior to cyclosporine dose reduction and 6 months afterwards were available for 89 patients and showed a reduction in mean serum creatinine level in patients receiving everolimus 1.5 mg as well as those receiving azathioprine (Table 4). The associated reduction in median cyclosporine trough level was 26 ng/ml, 5 ng/ml and 25 ng/ml for the everolimus 1.5 mg, everolimus 3 mg and azathioprine groups, respectively.

Viganò et al.

589

DISCUSSION This study has compared the efficacy and safety of two oral doses of everolimus (1.5 mg/day or 3 mg/day) vs azathioprine in combination with cyclosporine microemulsion (Neoral) and corticosteroids in de novo heart transplant recipients. When this trial was designed, azathioprine was used as the comparator drug, because, at that time, it was the standard adjunctive agent used in cardiac transplantation. Furthermore, mycophenolate mofetil (MMF) was not available in all countries participating in the trial at the time of its inception. Over the past few years, MMF has evolved as a more popular adjunctive agent in heart transplantation, but azathiopine continues to be used as an alternative immunosuppressive agent. The results presented herein show that an everolimus-based immunosuppressive regimen resulted in significantly fewer patients reaching efficacy end-points and reduced the severity of allograft vasculopathy in cardiac transplant recipients at 24 months after transplantation. Furthermore, everolimus 1.5 mg was also associated with a reduced incidence of allograft vasculopathy. These findings confirm and extend the 12month results from the trial.15,20,21 At 24 months, as at 12 months, both doses of everolimus were significantly superior to azathioprine in preventing the composite efficacy end-point and acute rejection Grade ⱖ3A, maintaining a continued benefit from everolimus through the second year. At 24 months, the reduction in efficacy failure was 37% and 20%, respectively, for everolimus 1.5 mg and 3 mg compared with azathioprine. Rates of rejection associated with hemodynamic compromise, graft loss and death were comparable for the three treatments. The potential for everolimus to influence allograft vasculopathy has been demonstrated in experimental models, in which everolimus inhibited growth factor– driven cell proliferation, vascular remodeling and intimal proliferation.9,10,12 The 12-month results of the current trial provided the first clinical evidence that everolimus could limit the development of allograft vasculopathy in heart transplant recipients.15 Because allograft vasculopathy has an increasing prevalence up to 3 to 5 years after heart transplantation, we believed that the clinical relevance of this initial finding deserved a further evaluation with a longer follow-up. The present report has shown that this benefit is sustained for another 12 months and that continued treatment with everolimus limits the progression of intimal thickening and lowers the incidence of allograft vasculopathy at 24 months when compared with azathioprine. The difference is significant for patients receiving everolimus 1.5 mg, whereas statistical significance was not reached for patients treated with everolimus 3 mg.

590

Viganò et al.

The elevated number of drop-outs from the intravascular ultrasound (IVUS) study, which restricted the procedure to about 25% of patients, may have played a role as a limitation of the analysis. Despite this statistical consideration, one should bear in mind that, unlike previous studies,22 this trial had a very strict IVUS study protocol with paired-set analysis and baseline evaluation. According to these criteria, this is the first demonstration of a single drug acting as a vascular smooth muscle cell proliferation inhibitor after solid-organ transplantation. Intimal thickening has been linked to an increased risk of subsequent adverse clinical events5–7,23–25 and, accordingly, a reduction in IVUSdefined vasculopathy with use of everolimus has the potential to improve long-term outcomes. Although azathioprine was used as the comparator arm in this study, a recent analysis of azathioprine vs MMF suggested that patients receiving MMF have less intimal thickening than those receiving azathioprine at 1 year post-transplant, suggesting that MMF may also provide benefits with regard to allograft vasculopathy after heart transplantation.26 In addition, in this study, the reduction of allograft vasculopathy frequency may also be partially explained by a lower incidence of CMV infection in the everolimus-treated groups. Indeed, this viral infection has been linked to acute rejection and development of allograft vasculopathy, but its actual clinical significance remains to be determined. Rates of discontinuation and of serious adverse events were similar in patients treated with everolimus 1.5 mg and azathioprine, but were increased in patients receiving the higher dose of everolimus. At 24 months, everolimus treatment increased cholesterol and triglyceride levels in fewer than 15% and 5% of patients, respectively (⬎90% of patients were receiving statin therapy). The incidences of major cardiac events and malignancy were comparable in the three groups. Renal function stabilized between 12 and 24 months in everolimus-treated patients in the larger study population, although serum creatinine levels remained significantly higher in the everolimus groups than in the azathioprine group at 24 months. The renal dysfunction observed in the everolimus cohorts is likely to have been caused by a toxic effect of cyclosporine, which was synergistically amplified by everolimus. Everolimus alone has not shown any renal toxicity in pre-clinical studies or in human subjects receiving everolimus without cyclosporine. Everolimus with reduced-exposure cyclosporine in renal transplant patients resulted in good efficacy and improved renal function.16,17 Furthermore, increasing clinical experience with everolimus in heart transplantation suggests that the use of lower levels of CsA can lead to improved renal

The Journal of Heart and Lung Transplantation June 2007

function while maintaining adequate immunosuppression.27,28 The efficacy of everolimus in combination with lowdose CsA is also retained in heart transplant recipients during the maintenance phase.29 Separately, there was convincing evidence that everolimus trough levels should be maintained at ⬎3 ng/ml for optimal efficacy.30 Hence, in the amendment protocol for the current study, cyclosporine trough levels were lowered in patients with renal dysfunction, whereas everolimus exposure was maintained at ⬎3 ng/ml. This strategy improved renal function in the cardiac transplant population within 6 months, requiring only a minor decrease in cyclosporine exposure without any adverse sequelae from the dose reduction. A further demonstration of a synergistic nephrotoxic effect of cyclosporine and everolimus is the fact that a similar cyclosporine trough level reduction in the azathioprine group did not result in any improvement in renal function. In conclusion, this analysis has shown that the superiority of everolimus over azathioprine in preventing efficacy failure and limiting the progression of allograft vasculopathy in heart transplant recipients is maintained at 24 months of follow-up. Everolimus was also significantly more effective than azathioprine in reducing the risk of acute rejection and decreasing the frequency of CMV infection. The findings presented herein suggest that an optimum immunosuppressive combination for heart transplant patients may be a starting dose of 1.5 mg everolimus, titrated to maintain blood levels of at least 3 ng/ml, and then progressively reducing cyclosporine exposure, targeting maintenance trough levels of 100 ng/ml. The incidence of graft and patient survival were comparable between groups at 24 months post-transplant; however, this immunosuppressive regimen appears to be safe and effective, and the potential to influence allograft vasculopathy may improve longer-term outcomes after heart transplantation. The authors acknowledge the contributions of the other RAD B253 study investigators: Novartis Pharma, Basel, Switzerland, and Novartis Pharmaceuticals, Summit, NJ—P. Bernhardt, K. H. Abeywickrama, J. Lind, N. Cretin, S. Le Breton, J. Kabir, J. Murphy; Data and Safety Monitoring Board—A. Laupacis (Toronto), G. A. Wells (Ottawa, ON, Canada), R. Mills (Cleveland, OH), G. Parry (Newcastle upon Tyne, UK). Canada: Institut de Cardiologie de Montreal, Montreal—M. Carrier, D. Normandin, J. Vézina; University of Ottawa Heart Institute, Ottawa, ON—R. Masters, R. Davies; Toronto Hospital, Toronto—H. Ross, C. Cardella, D. Delgado, C. O’Grady; New Halifax Infirmary, Halifax, NS—H. Haddad, J. Howlett, G. Hirsch, C. Kells, B. J. O’Neill, K. Giddens; Argentina: Fundacion Favaloro, Buenos Aires—S. V. Perrone, L. Favaloro, R. Favaloro, E. Kapplinsky, A. Natello; Austria: Allgemeines Krankenhaus Universitaet, Vienna—M. Grimm,

The Journal of Heart and Lung Transplantation Volume 26, Number 6

G. Laufer, G. Wieselthaler, A. Zuckermann, E. Deviatko; Belgium: Universitats Ziekenhuis Gasthuisberg, Leuven—J. Vanhaecke, J. Van Cleemput, W. Droogne, A. Strijckmans; Cliniques Universitaires St. Luc, Brussels—M. Goenen, T. Timmermans; Onze Lieve Vrouw Ziekenhuis, Aalst—F. Wellens, M. Goethals, W. Tack; Switzerland: Universitats Spital, Zurich—W. Kiowski, E. Oechslin, H. Brunner, R. Schindler; Germany: Kliniken der Medizinischen Hochschule, Hannover—A. Haverich, K. Pethig, C. Bara, I. Scheibner; Deutsches Herzzentrum, Berlin—R. Hetzer, M. Hummel, S. Kapell, E. Wenzel; Denmark: Skejby Sygehus, Aarhus—K. Soerensen, H. Moelgaard, H. Egeblad, J. E. Nielsen-Kudsk, H. Eiskjær, E.-M. Tram; Spain: Hospital Reina Sofia, Cordoba— J. M. Arizon, M. Concha, F. Vallés, A.L. Granados; Hospital Juan Canalejo, La Coruna—M. G. Crespo, M. J. PanianguaMartín, T. Tabuyo, A. Juffé, J. A. Rodriguez; Clinica Puerta de Hierro, Madrid—L. A. Pulpon, J. Segovia; France: Hôpital La Pitié Salpêtrière, Paris—I. Gandjbakhch, R. Dorent, P. Léger, J.-P. Levasseur, E. Vaissier; Hôpital Foch, Suresnes—P. De Lentdecker, G. Dreyfus; Hôpital Cardiologique de Lyon, Lyons—G. Dureau, P. Boissonnat, L. Sebbag, A. Roussouliere; UK: Papworth Hospital, Cambridge—J. Parameshwar, P. Schofield, L. Steel, V. Beresford; Wythenshawe Hospital, Manchester—N. Yonan, R. Martyszczuk, J. Reader; Italy: Azienda Ospedale Niguarda Ca’ Granda, Milan—M. Frigerio, G. Masciocco, M. Grassi, M. Garbellini, F. Oliva; Policlinico San Matteo–Instituto di Ricovero è Cura a Carattere Scientifico, Pavia—M. Viganó, C. Pellegrini, M. Rinaldi, A. M. D’Armini; Policlinico, Università degli Studi, Padua—D. Casarotto, A. Gambino, T. Luca, G. Feltrin, G. Gerosa; Azienda Ospedaliera Ospedali Riuniti di Bergamo, Bergamo—R. Fiocchi, A. Gamba, C. Mammana, L. Iamele, G. Guagliomi; Ospedale Civile Maggiore Borgo Trento, Verona—G. Faggian, A. Mazzucco, A. Forni; Norway: Rikshospitalet, Oslo—S. Simonsen, O. Geiran, A. Relbo, I. Grov; Poland: Klinika Kardiochirurgii Collegium Medicum Uniwersytetu Jagiello, Krakow—M. Garlicki; USA: Rush–Presbyterian–St. Luke’s Medical Center, Chicago, IL—W. Kao, C. Downer; Temple University, Philadelphia, PA—H. J. Eisen, S. Furukawa, K. B. Margulies, P. J. Mather, G. O. Berman, S. Rubin, J. Garcia, L. Lavelle, J. Wong; Shands Hospital Health Science Center, Gainesville, FL—J. Hill, J. Aranda, D. Leach, A. Ebling; UCLA Medical Center, Los Angeles, CA—J. Kobashigawa, J. Chait, E. Wang, B. Cole, J. Patel, J. Moriguchi, H. Laks, J. Tobis, M. Espejo; Columbia Presbyterian Medical Center, New York, NY—D. Mancini, M. E. Cordisco, L. Donchez; Texas Heart Institute, Houston, TX—O. H. Frazier, J. H. Connelly, C. D. Thomas; Utah Transplantation Affiliated Hospitals Cardiac Transplant Program, Salt Lake City, UT—D. Renlund, S. A. Moore, D. Taylor, A. G. Kfoury, B. Campbell, M. Eidson; Emory Clinic, Atlanta, GA—A. Smith, W. Book, G. Snell; Stanford University School of Medicine, Stanford, CA—H. A. Valantine-von Kaeppler, K. Woodman, K. Kari, R. Majem; Heart Failure–Transplant Service, Northern California Kaiser Permanente, San Jose—H. Parekh, D.

Viganò et al.

591

Weishaar, J. Trammell; Tulane University Medical Center, New Orleans, LA—F. Smart, S. Bowers; Cleveland Clinic Foundation, Cleveland, OH—R. C. Starling, B. Gus, J. B. Young, P. McCarthy, K. James; University of Pittsburgh Medical Center, Pittsburgh, PA—K. McCurry, D. Zaldonis; University of Alabama at Birmingham, Birmingham, AL—R. Benza, B. Rayburn, R. Bourge; University of Michigan Medical Center, Ann Arbor, MI—K. Aaronson, F. D. Pagani, T. M. Koelling, D. B. Dyke, R. Baliga, D. McLean; Saint Louis University Health Science Center, St. Louis, MO—P. J. Hauptman, M. Rawlings; Brigham and Women’s Hospital, Boston, MA—J. Jarcho, B. Symko; University of Minnesota, Minneapolis, MN—L. W. Miller, K. K. Schafer; Johns Hopkins Hospital, Baltimore, MD—J. Hare, E. Kasper; University of Colorado University Hospital, Denver, CO—J. Lindenfeld, L. Clegg; University of Wisconsin Hospital, Madison, WI—R. B. Love, L. Jacobs; Vanderbilt University Medical Center, Nashville, TN—S. F. Davis, P. A. Thole; University of Pennsylvania Medical Center, Philadelphia, PA—L. R. Goldberg, S. Chambers, C. Twomey; Massachusetts General Hospital, Boston, MA—G. W. Dec, D. Cocca-Spoffard, J. Camuso; St. Luke’s Physicians Office, Milwaukee, WI—J. Hosenpud, A. J. Tector, J. Courch, D. Hudson. The authors also thank Dr Carlos Macaya, Dr Fernando Alonso and Dr Javier Segovia, Hospital Clínico San Carlos de Madrid, Spain, and the Cleveland Clinic Intravascular Sonography Core Laboratory, Cleveland, OH, for providing expert intravascular ultrasound services.

REFERENCES 1. Young JB. Allograft vasculopathy: diagnosing the nemesis of heart transplantation. Circulation 1999;100:458 – 60. 2. Weiss M, von Scheidt W. Cardiac allograft vasculopathy: a review. Circulation 1997;96:2069 –77. 3. Young JB. Perspectives on cardiac allograft vasculopathy. Curr Atheroscler Rep 2000;2:259 –71. 4. Kapadia S, Nissen S, Tuzcu M. Impact of intravascular ultrasound in understanding transplant coronary artery disease. Curr Opin Cardiol 1999;14:140 –50. 5. Kobashigawa JA. First-year intravascular ultrasound results as a surrogate marker for outcomes after heart transplantation. J Heart Lung Transplant 2003;22:711– 4. 6. Kobashigawa JA, Tobis JM, Starling RC, et al. Multicenter intravascular ultrasound validation study among heart transplant recipients: outcomes after five years. J Am Coll Cardiol 2005;45: 1532–7. 7. Tuzcu EM, Kapadia SR, Sachar R, et al. Intravascular ultrasound evidence of angiographically silent progression in coronary atherosclerosis predicts long-term morbidity and mortality after cardiac transplantation. J Am Coll Cardiol 2005;45:1538 – 42. 8. Dumont FJ. Treatment of transplant rejection: are the traditional immunosuppressants good enough? Curr Opin Invest Drugs 2001;2:357– 63. 9. Cole OJ, Shehata M, Rigg KM. Effect of SDZ RAD on transplant arteriosclerosis in the rat aortic model. Transplant Proc 1998;30: 2200 –3. 10. Schuurman HJ, Pally C, Weckbecker G, Schuler W, Bruns C. SDZ RAD inhibits cold ischemia-induced vascular remodeling. Transplant Proc 1999;31:1024 –5.

592

Viganò et al.

11. Nishimura T, Faul J, Berry GJ, Veve I, Pearl R, Kao PN. 40-0-(2hydroxylethyl)-rapamycin attenuates pulmonary arterial hypertension and neointimal formation in rats. Am J Resp Crit Care Med 2001;163:498 –502. 12. Schuler W, Sedrani R, Cottens S, et al. SDZ RAD, a new rapamycin derivative: pharmacological properties in vitro and in vivo. Transplantation 1997;64:36 – 42. 13. Schuurman HJ, Cottens S, Fuchs S, et al. SDZ RAD, a new rapamycin derivative: synergism with cyclosporine. Transplantation 1997;64:32–5. 14. Hausen B, Boeke K, Berry G, Segarra IT, Christians U, Morris RE. Suppression of acute rejection in allogeneic rat lung transplantation: a study of the efficacy and pharmacokinetics of rapamycin derivative (SDZ RAD) used alone and in combination with a microemulsion formulation of cyclosporine. J Heart Lung Transplant 1999;18:150 –9. 15. Eisen HJ, Tuzcu M, Dorent R, et al. Everolimus for the prevention of allograft and vasculopathy in cardiac-transplant recipients. N Engl J Med 2003;349:847–58. 16. Nashan B, Curtis J, Ponticelli C, et al. Everolimus and reducedexposure cyclosporine in de novo renal-transplant recipients: a three-year phase II, randomized, multicenter, open-label study. Transplantation 2004;78:1332– 40. 17. Vitko S, Tedesco H, Eris J, et al. Everolimus with optimized cyclosporine dosing in renal transplant recipients: 6-month safety and efficacy results of two randomized studies. Am J Transplant 2004;4:626 –35. 18. Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika 1988;75:800 –2. 19. Kaplan EL, Meier P. Nonparametric estimation from incomplete observation. J Am Stat Assoc 1958;53:457– 81. 20. Valantine H. Prevention of cardiac allograft vasculopathy with Certican (everolimus): the Stanford University experience within the Certican phase III clinical trial. J Heart Lung Transplant 2005;24(suppl 4):S191–5.

The Journal of Heart and Lung Transplantation June 2007

21. Valantine H, Zuckermann A. From clinical trials to clinical practice: an overview of Certican (everolimus) in heart transplantation. J Heart Lung Transplant 2005;24(suppl 4):S185–90. 22. Keogh A. Long-term benefits of mycophenolate mofetil after heart transplantation. Transplantation 2005;79(suppl):S45– 6. 23. Mehra M, Ventura H, Stapleton D, Smart F, Collins T, Ramee S. Presence of severe intimal thickening by intravascular ultrasonography predicts cardiac events in cardiac allograft vasculopathy. J Heart Lung Transplant 1995;14:632–9. 24. Rickenbacher P, Pinto F, Lewis N, et al. Prognostic importance of intimal thickness as measured by intracoronary ultrasound after cardiac transplantation. Circulation 1995;92:3445–52. 25. Liang D, Gao S, Botas J, Pinto FJ, Schroeder JS, Alderman EL, Yeung AC. Prediction of angiographic disease by intracoronary ultrasonographic findings in heart transplant recipients. J Heart Lung Transplant 1996;15:980 –7. 26. Kobashigawa JA, Tobis JM, Mentzer RM, et al. Mycophenolate mofetil reduces intimal thickness by intravascular ultrasound after heart transplant: reanalysis of the multicenter trial. Am J Transplant 1006;6:993–7. 27. Lehmkuhl H, Hetzer R. Clinical experience with Certican (everolimus) in de novo heart transplant patients at the Deutsches Herzzentrum, Berlin. J Heart Lung Transplant 2005; 24(suppl 4):201–5. 28. Hummel M. Recommendations for the use of Certican (everolimus) after heart transplantation: results from a German and Austrian Consensus Conference. J Heart Lung Transplant 2005; 24(suppl 4):196 –200. 29. Zuckermann A. Clinical experience with Certican (everolimus) in maintenance heart transplant patients at the Medical University of Vienna. J Heart Lung Transplant 2005;24(suppl 4):206 –9. 30. Kovarik JM, Kaplan B, Tedesco Silva H, et al. Exposureresponse relationships for everolimus in de novo kidney transplantation: defining a therapeutic range. Transplantation 2002;73:920 –5.