Author + information
- Received October 21, 2005
- Revision received November 24, 2005
- Accepted December 13, 2005
- Published online April 18, 2006.
- Alfredo E. Rodriguez, MD, PhD, FACC⁎,⁎ (, )
- Juan F. Granada, MD†,
- Máximo Rodriguez-Alemparte, MD⁎,
- Cesar F. Vigo, MD⁎,
- Juan Delgado, MD†,1,
- Carlos Fernandez-Pereira, MD⁎,
- Antonio Pocovi, MD⁎,
- Alfredo M. Rodriguez-Granillo, BS⁎,
- Daryl Schulz, RTN†,
- Albert E. Raizner, MD, FACC†,
- Igor Palacios, MD, FACC‡,
- William O’Neill, MD, FACC§,
- Grzegorz L. Kaluza, MD, PhD†,
- Gregg Stone, MD, PhD, FACC‖,
- ORAR II Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Alfredo E. Rodriguez, Otamendi Hospital, Callao 1441 4 B, Buenos Aires, Argentina 1024.
Objectives The purpose of this study was to assess the role of oral rapamycin in decreased restenosis after bare metal stent implantation.
Background Small observational studies suggest that the administration of oral rapamycin reduces angiographic restenosis after bare metal stent implantation.
Methods Between September 2003 and September 2004, 100 patients were randomized to either oral rapamycin (6-mg loading dose given 2.7 h before intervention followed by 3 mg/day for 14 days) plus diltiazem 180 mg/day or no therapy after the implantation of a coronary bare metal stent design. The primary study end point was incidence of angiographic binary restenosis and late loss at nine months. The secondary end points were target lesion revascularization, target vessel revascularization, and incidence of major adverse cardiovascular events at 1 year.
Results Angiographic follow-up was completed in 87% of patients. In the rapamycin group, the drug was well tolerated (26% minor side effects) and was maintained in 96% of patients. At 9 months, the in-segment binary restenosis was reduced by 72% (11.6% rapamycin vs. 42.8% no-therapy group, p = 0.001) and the in-stent binary restenosis was reduced by 65% (12% rapamycin vs. 34.6% no-therapy group, p = 0.009). The in-segment late loss was also significantly reduced with oral therapy (0.66 vs. 1.13 mm, respectively; 43% reduction, p < 0.001). At 1 year, patients in the oral rapamycin group also showed a significantly lower incidence of target vessel revascularization (8.3% vs. 38%, respectively, p < 0.001), target lesion revascularization (7.6% vs. 37.2%, respectively, p < 0.001), and major adverse cardiovascular events (20% vs. 44%, respectively, p = 0.018).
Conclusions This randomized, controlled, and unblinded study showed that the administration of oral rapamycin during 14 days after stent implantation significantly reduces angiographic and clinical parameters of restenosis.
Since the introduction of coronary angioplasty, restenosis of the target lesion has been the main limitation of this procedure. Acute vessel recoil, chronic remodeling, and intimal hyperplasia are the mechanisms involved in this process (1–4). However, after the introduction of stents in daily practice during interventional procedures, intimal hyperplasia became the mechanism associated with the pathophysiology of in-stent restenosis (5–9). Therefore, its prevention should be related to therapies that inhibit smooth muscle cell proliferation.
In recent years, drug-eluting stents (sirolimus, Johnson & Johnson, Miami Lakes, Florida; and paclitaxel, Boston Scientific, Natick, Massachusetts) have been associated with significant reduction of in-stent restenosis (10–17). Although clinical and angiographic parameters of restenosis were significantly improved, long-term safety data of these stents needs to be addressed (18,19). Sirolimus (Rapamune, Wyeth, Roses Point, New York) is a potent immunosuppressive and antiproliferative agent that was approved by the U.S. Food and Drug Administration for use in patients after renal transplantation (20–24). Systemic use of rapamycin and its analog were associated in animal data with a significant reduction of intimal hyperplasia (22–25), but only recently were clinical studies with oral administration reported (26–31).
The purpose of the present controlled, randomized study was carried out to determine whether short oral administration of the drug was associated with a reduction of restenosis in patients treated percutaneously with bare metal stent therapy of de novo lesions.
Patient population and study design
From September 2003 to September 2004, 100 patients with severe stenosis in de novo coronary arteries were enrolled and included in this protocol. All of the percutaneous coronary interventions (PCIs) were performed at the Catheterization Laboratories at Otamendi Hospital and Sanatorio Las Lomas in Buenos Aires, Argentina. For a list of study participants, please see the Appendix.
Patients were considered for randomization if they meet the following criteria: clinical indication of percutaneous coronary revascularization, age >18 years, de novo severe stenosis in a native coronary artery, lesion suitable for stent, reference vessel size between 2.5 and 4.0 by visual estimation, and candidate for coronary bypass surgery. Enrollment was also permitted after the successful treatment of one additional non-study lesion in a non-study vessel before randomization.
Patients were excluded if they had acute myocardial infarction 48 h before randomization, rapamycin allergy, clopidogrel or aspirin intolerance, significant bleeding in the last 6 months, stroke or transient ischemic attack in the last 12 months, severe concomitant illness, recent major bleeding requiring transfusion, major blood dyscrasias, participation in another trial that did not allow a follow-up angiogram, hyperlipidemia of difficult treatment, thrombocytopenic disease, or chronic total occlusion or in-stent restenosis lesions, or if they were not amenable to signing the informed consent form allowing a follow-up angiogram. Lesion length was not an exclusion criteria, and multiple stents in the same vessel as well as overlapping stents were allowed.
The protocol of this non–industry-sponsor study was approved by the ethics committee of the participating centers of the study, by the ethics committee of the Argentine Society of Cardiac Angiography and Interventions, and by the Argentina National Regulatory Agency for Drug, Food, and Medical Technology. During the study, an Independent Safety Monitoring Committee adjudicated the clinical adverse events. The study was conducted according to the principles of the Declaration of Helsinki, and all patients signed a written inform consent form for participation in this trial.
Randomization, medication, and coronary procedures
All eligible patients were randomized in the catheterization laboratory, immediately after the diagnostic angiogram was performed, to either the control group or the oral rapamycin group. The randomization process in each center was performed in a blinded manner for the coordinating center with the use of an Internet system containing the block randomization sequence for each participating center. In the oral rapamycin arm, patients received a loading dose of 6 mg at 2.71 ± 0.9 h before stent implantation, followed by 3 mg/day for a total of 14 days. Sustained-release diltiazem 180 mg/day was added to the sirolimus regimen to achieve higher sirolimus blood concentrations (21). Patients in the control group did not receive either a placebo or additional therapy. Blood samples were drawn to measure sirolimus blood levels and were taken at seven days after an oral loading dose of sirolimus in a central core laboratory (27,31). In addition, serum creatine, cholesterol, triglycerides, red and white blood cells, and platelet counts were measured before and at the end of sirolimus treatment. A clinical interview was required each week for the first month of treatment, then monthly for six months, and at one year after the intervention. Coronary angiography was scheduled between six to nine months after the initial PCI procedure.
PCI and stent procedure
The PCI was performed using standard techniques described elsewhere (27). All 100 patients received one or more identical closed-cell stent design; an appropriate-sized stent (Gryphus stent delivery system, Endovascular Devices Inc., Wilmington, Delaware) with a ratio of stent diameter to distal reference vessel diameter of 1:1 to 1.1:1 was implanted at a pressure of at least 13 atm guided by on-line quantitative coronary angiography. The stent used in this study is a CE-marked, clinically tested, closed-cell design with a strut thickness and width of 110 μm and 120 μm, respectively. The same stent design was used to avoid potential bias with stent selection in both groups (32). The available stent lengths were 15, 18, and 23 mm, and the stent diameters were 2.5, 3.0, 3.5, and 4.0 mm. All patients received 325 mg/day of aspirin indefinitely, and clopidogrel at a loading dose of 300 mg on the day of the procedure and 75 mg/day thereafter for one month. Statins were given to all patients indefinitely.
Study end points
The primary end point of the study was to compare the angiographic binary restenosis rate and late loss determined by an independent core laboratory blinded to treatment allocation. Angiographic binary restenosis was defined as >50% residual stenosis in the target lesion in the follow-up angiography. In patients with multiple lesions, the lesions were counted separately. Secondary end points were target lesion revascularization (TLR), target vessel revascularization (TVR), target vessel failure, and major adverse cardiovascular events (MACEs). The TLR and TVR were performed in the presence of angiographic restenosis and symptoms and signs of myocardial ischemia, determined by an independent clinical events committee (clinically driven). A MACE was defined as death, myocardial infarction, stroke, and TVR at 1 year of follow-up. Target vessel failure was defined as death, non-fatal myocardial infarction, and TVR during the entire follow-up period. The TLR was counted by lesion, TVR was counted by vessel, and MACEs were counted by patient. The diagnosis of acute myocardial infarction was based on typical chest pain combined with either new pathological Q waves or an increase of creatine kinase to >3 times the upper limit of normal, with a concomitant increase in the MB isoenzyme. Sirolimus compliance and adverse side effects related to oral administration of sirolimus were also recorded. All events were adjudicated by an independent clinical events committee whose members were unaware of the patient’s assigned treatment.
Quantitative coronary angiography
Angiographies were analyzed in all patients at the angiographic core laboratory facilities of the Methodist Research Institute, Houston, Texas, by blinded operators, using an automatic edge detection system (CAASII Pie Medical Imaging, Maastrich, the Netherlands). The analysis segment comprised the stent segment and the proximal and distal stent edge, defined as 5 mm proximal or distal to the stent. Similar views (an average of two orthogonal projections) were selected for angiograms recorded before the intervention, immediately after stent deployment, and at follow-up. Each angiography sequence was preceded by an intracoronary injection of nitroglycerin. Acute gain was defined as the difference between minimal luminal diameter before and at the end of the PCIs and stent procedure. Late lumen loss was calculated as the difference in minimal luminal diameter between measurements noted immediately after the procedure and at follow-up. Net gain was defined as the differences between minimal luminal diameter at follow-up and before the interventional procedure.
The sample size of the study was determined on the basis of a test for a trend analysis based on the estimation for the primary end point of angiographic restenosis and late loss according to our previous pilot trials and from recent trials with drug-eluting stents; in the control group we assumed an incidence of binary restenosis of 35% in accordance with recently reported data (12–14,16,17) of in-segment restenosis with bare metal stents. If we average the restenosis rate in the control arm of those drug-eluting stents trials, the in-segment restenosis rate was over 35%. In the oral rapamycin group, we assumed a binary restenosis rate of 10% according our previous pilot studies (27,31). Using a two-sided test for differences in independent binomial proportions with an alpha level of 0.05, we calculated that 80 patients (40 for each group) would have to undergo randomization for the study to have 80% power to detect a difference in the binary restenosis rate among both groups; thus, we enrolled a total of 50 patients in each arm to accommodate patients in whom follow-up angiography could not be performed. Late loss was calculated assuming a loss around 0.60 mm in the rapamycin group and 1.10 mm in the control arm, which is in accordance with the late loss data obtained in our previous pilots trials with oral rapamycin (27,31). With the above number, we needed a sample size of 30 patients in each group for a power of 0.80%.
Continuous variables were compared using an unpaired two-sided Student ttest, and categorical variables were compared using the Fisher exact test. Continuous variables were expressed as mean ± SD, and categorical variables were expressed as percentages. Changes in blood measurements after rapamycin treatment were compared using the raw data of the ttest. Freedom from survival end points at follow-up were obtained using Kaplan-Meier curves and were compared by log-rank test. A multiple logistic regression analysis, backward stepwise method (Wald), was performed to correlate angiographic binary restenosis with clinical and angiographic variables, including treatment groups. Statistical significance was accepted for a value of p < 0.05.
Between September 2003 and September 4, 2004, 100 patients were randomized, 50 patients in the control group (55 arteries and 59 lesions) and 50 patients in the oral sirolimus group (60 arteries and 66 lesions). A total of 132 stents were deployed, 61 in the control and 71 in the oral sirolimus group; a small stent size (2.5 mm) was deployed in 44.7% of the lesions. In the control group, 8 patients with 17 lesions in 13 vessels were treated. In the oral rapamycin group, 15 patients with 31 lesions in 25 vessels were treated.
The baseline demographic, clinical, and angiographic characteristics among both groups are described in Table 1;treated diabetes was more frequent in the oral sirolimus group, p = 0.054 (Fisher exact test). Hospital and 30-day outcomes in both groups were similar (Table 2).During the course of treatment with oral sirolimus, 26% of patients had side effects; however, none of them were major. The most frequent side effect was mouth ulceration (16%). Only two patients (3.9%) discontinued treatment, at three and eight days after the first dose.
After the rapamycin treatment, during the first 30 days, white blood counts showed significant transient changes; however, severe leucopoenia was not seen. There was also a non-significant increase in triglycerides levels (Table 3).
Hospital and follow-up results are described in Table 2. One-year clinical follow-up (366 days) was obtained in all patients (100%) in both groups; after hospital discharge during the follow-up, there were two deaths (4%) in the control group (both cardiac), whereas two patients in the oral sirolimus group (4%) died during follow-up (one because of colon cancer and the other after an elective coronary bypass surgery). After hospital discharge, there was no documented non-fatal myocardial infarction or stroke in either group.
The rate of clinically driven TLR or TVR was significantly lower (Fisher exact test) in the oral sirolimus compared with the control group (Table 2). The TVR was 5 of 60 (8.3%) versus 21 of 55 (38%), respectively, p < 0.001; and TLR was 5 of 66 (7.6%) versus 22 of 59 (37.2%), respectively, p < 0.001. Target vessel failure and MACEs were also improved with oral sirolimus therapy (p = 0.009 and p = 0.018, respectively, Table 2). All surviving patients and those who did not have follow-up angiography were asymptomatic at 1 year of follow-up.
Figure 1shows survival curves of freedom from TVR (Fig. 1A), freedom from TLR (Fig. 1B), and freedom from MACEs (Fig. 1C) that have significantly better outcomes in those patients treated with oral sirolimus, with numbers that represent an 80% reduction of TLR and a 55% reduction of MACEs compared with the control group.
Follow-up angiographic data
Baseline and follow-up angiographic data are shown in Table 4.Reference vessel size, minimal luminal diameter before and after the procedure, and acute gain were similar in both groups.
Clinically driven or per-protocol follow-up angiography at nine months was completed in 87% of the population (87 patients and 99 vessels). At nine months, the binary in-stent restenosis rate per vessel was 12% for the rapamycin group and 34.6% for the control group (p = 0.009). The in-segment analysis showed a restenosis rate of 12% and 42.8% for the rapamycin and control groups, respectively (p = 0.001). As shown in Table 4, the use of oral rapamycin reduced the risk of binary restenosis by 65% within the stent and by 72% in the analysis segment. With the above numbers, the power of our study to detect differences between groups for restenosis was 0.81 per patient and 0.94 per vessel, with a value of p < 0.05. In the five patients in the oral rapamycin group in whom restenosis developed, the restenosis was diffuse but not proliferative or with total occlusion. The restenosis pattern in the control group showed that from 23 lesions with restenosis, only 3 were among 51% to 69%, 8 were between 69% to 80%, and 12 were >80%, including 4 with almost complete closure (Thrombolysis In Myocardial Infarction [TIMI] flow grade 1 or 2). Thus, this finding (by quantitative coronary angiography, severe restenosis in 87%) explains the rate of TLR in the control group. With the oral therapy, in-stent late loss was reduced from 1.41 mm in the control versus 0.73 mm in the oral rapamycin group, and in-segment from 1.13 mm in the control versus 0.66 mm in the oral rapamycin group, meaning a reduction of 48% and 43% in-stent and in-segment late loss, respectively. As mentioned previously, the degree of restenosis is in correlation with the amount of late loss in the control group.
Multivariate analysis showed that randomization to the control group was the only independent predictor of restenosis (odds ratio [OR] 6.01, 95% confidence interval 2.19 to 16.46, p < 0.0001). As shown in Table 4, compared with the control group, patients who received oral rapamycin had a significantly smaller amount of late loss (0.66 mm in the sirolimus group vs. 1.13 mm in the control group, p = 0.0002), resulting in greater luminal dimension and a smaller degree of stenosis at follow-up. The relative reduction in the risk of restenosis among patients who received oral rapamycin was independent of diabetes mellitus status, vessel location, and the length and diameter of the lesion or stent.
This prospective, randomized, controlled trial in patients with de novo lesions showed a significant reduction of angiographic binary restenosis and late loss when patients were allocated to the oral sirolimus group, and both end points were determined by blinded operators. Clinical safety and efficacy parameters of restenosis, such as TVR, TLR, and MACEs at follow-up were also significantly improved with oral sirolimus therapy.
The population sample analyzed in the present study represents a relatively high-risk population involving B2/C lesions (approximately 70%), small vessels (44.7%), and lesions longer than 18 mm (53%). Also, overlapping (7%) and multiple stent implantations per treated vessel (17%) were also allowed in the study. Finally, despite the short period of oral administration of the drug, minor side effects were present in 26% of patients.
Several pre-clinical studies supported the use of systemic administration of rapamycin or its analog everolimus in reducing smooth muscle cell proliferation (20–25). The anti-inflammatory and antiproliferative effects of rapamycin was based on its ability to inhibits the TOR kinase (target of rapamycin), an essential component in the pathway of the cell cycle progression (20–24). Drug-eluting stents have been extensity studied in several randomized studies (10–17); however, their long-term safety data are not well established (18,19).
Systemic use of oral sirolimus in patients undergoing PCI procedures has been recently tested. One randomized trial in patients with in-stent restenosis (29) showed a significant reduction of restenosis and late loss in the high-oral-sirolimus dose group. The Oral Rapamune to Inhibit Restenosis Trial (ORBIT) was a non-randomized trial in de novo lesions that showed a single-digit restenosis rate in a cohort of 60 patients treated with low and high oral doses of sirolimus (30), and when 2 mg/day as a maintenance dose was used, only minor side effects were reported.
We previously performed two observational studies (27,31) showing that a sirolimus blood concentration of >8 ng/ml was associated with a single-digit restenosis rate and lower late loss. In the present randomized study, the sirolimus oral administration scheme was different than we previously reported (27,31): the bolus was given two h before intervention, and the daily dose was 3 mg instead 2 mg and was used only for 14 days. It is not clear when would be the ideal moment to start oral sirolimus treatment; in one study (29) of patients with in-stent restenosis, the oral sirolimus loading dose was administrated two days before the procedure. This was supported by the concept that the immunosuppressive effects of sirolimus achieved optimal levels after 4 days of treatment, and that perhaps, with a high-risk population for restenosis such as patients with in-stent restenosis, a pre-intervention loading dose would obtain better results.
Finally, drug-eluting stents have been associated with a significant reduction of restenosis and late loss compared with bare-metal stents; in fact, late loss during the first year of follow-up showed only a minor increase in minimal luminal diameter with drug-eluting stent therapy, and those numbers (13,14) are lower than the results presented here, meaning that local therapy achieved high immunosuppressive effects.
In conclusion, this was the first randomized study using oral rapamycin in patients with de novo lesions treated with coronary bare metal stent therapy. We report a significant reduction of angiographic and clinical parameters of restenosis, suggesting that this treatment may be a cost-effective alternative (33) to drug-eluting stent therapy in a selected group of patients, such as those at moderate risk of restenosis, with a reference vessel size >2.5 mm, without diabetes, and unsuitable for long-term antiplatelet therapy.
First, the restenosis and late loss of the bare metal stent therapy in the control group was slightly higher than historical records reported with other stent designs, and the patient population of the trial could be in part responsible for the high restenosis rate. However, it is possible that these parameters would be lower with different stent designs. Nevertheless, in the drug-eluting stent era, stent binary restenosis in the control groups was similar to that we are reporting here in the control group (12–14). Furthermore, it is very important to emphasize that we used the same stent design in both groups, and these findings are in favor of the immunosuppressive effects of oral rapamycin. Both restenosis and late loss could be lower if we associated the oral therapy with a third generation of bare metal stent design. However, if a much lower restenosis rate were seen in the control group, such as that seen with the third generation of bare metal stents, the ability to show a difference might not have been possible with the sample size chosen in our study.
Second, we conducted an angiographic follow-up at nine months, and it is unknown whether these findings will be maintained during more prolonged follow-up. Furthermore, the optimal dosing, the need for pre-treatment, and the duration of the oral therapy have not been determined by this study. Minor side effects were present in 26% of the patients, and these numbers could be underestimated because of the sample size. Finally, more complex subsets of lesions were not analyzed by this study, and they would need further investigation.
For a list of the Steering Committee, Safety and Ethics Committee, Clinical Events Committee, Angiographic Core Laboratory, Participating Hospitals and Clinical Investigators, and Coordinating Center for the ORAR II trial, please see the online version of this article.
↵1 Delgado is now affiliated with the Corbic Research Foundation, Envigado, Columbia.
- Abbreviations and Acronyms
- major adverse cardiovascular event
- percutaneous coronary intervention
- target lesion revascularization
- target vessel revascularization
- Received October 21, 2005.
- Revision received November 24, 2005.
- Accepted December 13, 2005.
- American College of Cardiology Foundation
- Mintz G.,
- Popma J.,
- Pichard A.,
- et al.
- Kimura T.,
- Nosaka H.,
- Yokol H.,
- Iwabuchi M.,
- Nobuyoshi M.
- Rodriguez A.,
- Santaera O.,
- Larribau M.,
- et al.
- Rodriguez A.,
- Ayala F.,
- Bernardi V.,
- et al.
- Sousa J.E.,
- Costa M.A.,
- Abizaid A.C.,
- et al.
- Schampaert E.,
- Cohen E.A.,
- Schluter M.,
- et al.
- Grube E.,
- Silber S.,
- Hauptmann K.E.,
- et al.
- Colombo A.,
- Drzewiecki J.,
- Banning A.,
- et al.
- Virmani R.,
- Guagliumi G.,
- Farb A.,
- et al.
- McFadden E.,
- Stabile E.,
- Regar E.,
- et al.
- Groth C.G.,
- Backman L.,
- Morales J.M.,
- et al.,
- Sirolimus European Renal Transplantation Study Group
- Gallo R.,
- Padurean A.,
- Jayaraman T.,
- et al.
- Poon M.,
- Marx S.O.,
- Gallo R.,
- et al.
- Farb A.,
- Jhon M.,
- Acampado E.,
- et al.
- Brara P.,
- Moussavian M.,
- Grise M.,
- et al.
- Hausleiter J.,
- Kastrati A.,
- Mehilli J.,
- et al.
- Waksman R.,
- Ajani A.,
- Pichard A.,
- et al.
- Rodriguez A.,
- Rodriguez-Alemparte M.,
- Vigo C.,
- et al.
- Serruys P.W.,
- Kutryk M.J.B.
- Bagust A.,
- Grayson A.D.,
- Palmer N.D.,
- et al.