Author + information
- Received May 18, 2007
- Revision received July 27, 2007
- Accepted July 30, 2007
- Published online November 6, 2007.
- Christopher P. Cannon, MD, FACC⁎,⁎ (, )
- Steen Husted, MD†,1,
- Robert A. Harrington, MD, FACC‡,2,
- Benjamin M. Scirica, MD⁎,
- Håkan Emanuelsson, MD, PhD§,3,
- Gary Peters, MD∥,4,
- Robert F. Storey, MD¶,5,
- DISPERSE-2 Investigators
- ↵⁎Reprint requests and correspondence:
Dr. Christopher P. Cannon, TIMI Study Group, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115.
Objectives Our goal was to compare the safety and initial efficacy of AZD6140, the first reversible oral adenosine diphosphate receptor antagonist, with clopidogrel in patients with non–ST-segment elevation acute coronary syndromes (NSTE-ACS).
Background AZD6140 achieves higher mean levels of platelet inhibition than clopidogrel in patients with stable coronary artery disease.
Methods A total of 990 patients with NSTE-ACS, treated with aspirin and standard therapy for ACS, were randomized in a 1:1:1 double-blind fashion to receive either twice-daily AZD6140 90 mg, AZD6140 180 mg, or clopidogrel 300-mg loading dose plus 75 mg once daily for up to 12 weeks.
Results The primary end point, the Kaplan-Meier rate of major or minor bleeding through 4 weeks, was 8.1% in the clopidogrel group, 9.8% in the AZD6140 90-mg group, and 8.0% in the AZD6140 180-mg group (p = 0.43 and p = 0.96, respectively, vs. clopidogrel); the major bleeding rates were 6.9%, 7.1%, and 5.1%, respectively (p = 0.91 and p = 0.35, respectively, vs. clopidogrel). Although not statistically significant, favorable trends were seen in the Kaplan-Meier rates of myocardial infarction (MI) over the entire study period (MI: 5.6%, 3.8%, and 2.5%, respectively; p = 0.41 and p = 0.06, respectively, vs. clopidogrel). In a post-hoc analysis of continuous electrocardiograms, mostly asymptomatic ventricular pauses >2.5 s were more common, especially in the AZD6140 180-mg group (4.3%, 5.5%, and 9.9%, respectively; p = 0.58 and p = 0.01, respectively, vs. clopidogrel).
Conclusions This initial experience with AZD6140 in patients with ACS showed no difference in major bleeding but an increase in minor bleeding at the higher dose with encouraging results on the secondary end point of MI. This agent is currently being studied in a large outcomes trial in 18,000 patients with ACS.
Antiplatelet therapy with aspirin, thienopyridines (e.g., clopidogrel), or their combination has resulted in major reductions in cardiovascular events (1–6). This latter class inhibits platelet activation and aggregation by blocking the P2Y12receptor, 1 of 2 adenosine diphosphate (ADP) receptors on platelets (7). However, clopidogrel has several limitations: it is a pro-drug that requires hepatic conversion, leading to delay in onset of action, and there is wide interindividual variability in levels of platelet inhibition, with up to one-third of patients exhibiting minimal platelet inhibition (often referred to as “clopidogrel nonresponders”) (8–10). At steady state, clopidogrel achieves only 30% to 40% mean inhibition of platelet aggregation response to ADP, and its active metabolite binds irreversibly to P2Y12receptors, such that recovery of platelet function is precluded.
AZD6140 is the first reversible oral P2Y12receptor antagonist in a new chemical class of antiplatelet agents termed cyclopentyl-triazolo-pyrimidines with a half-life of approximately 12 h (11,12). Distinct from the thienopyridines, this agent does not require metabolic conversion to an active form; it binds directly to the P2Y12receptor, and it can more completely inhibit the sustained aggregation response to ADP. A study in patients with stable atherosclerosis found a dose-dependent increase in the level of inhibition with AZD6140, with levels being significantly higher than those achieved with clopidogrel (12). The DISPERSE (Dose confIrmation Study assessing anti-Platelet Effects of AZD6140 vs. clopidogRel in non–ST-segment Elevation myocardial infarction)-2 trial was a randomized, double-blind, double-dummy trial conducted to assess the safety, tolerability, and initial efficacy of AZD6140 plus aspirin in comparison with clopidogrel plus aspirin in patients with non–ST-segment elevation (NSTE) acute coronary syndromes (ACS).
Between October 3, 2004 and April 23, 2005, patients with NSTE-ACS were entered into the trial from 152 participating sites in 14 countries (Online Appendix 1). The study protocol was approved by the relevant institutional review boards, and written informed consent was obtained from all patients before the initiation of trial procedures.
Detailed entry criteria are in Online Appendix 2. Briefly, patients were eligible if they were hospitalized for NSTE-ACS within the preceding 48 h, experienced ischemic symptoms of ≥10 min duration at rest, with either biochemical marker evidence of myocardial infarction (MI) or electrocardiographic evidence of ischemia.
Patients received standard medical (anti-ischemic and antithrombotic) and interventional treatment for ACS, including aspirin at an initial dose of up to 325 mg followed by 75 to 100 mg daily with or without a glycoprotein IIb/IIIa inhibitor. Patients who had received clopidogrel before randomization were permitted in the study, but open-label clopidogrel was discontinued after randomization and replaced with study drug.
Eligible patients were randomized in a 1:1:1 double-blind fashion to receive AZD6140 90 mg twice daily, AZD6140 180 mg twice daily, or clopidogrel 300 mg followed by 75 mg once daily for up to 3 months. Patients in the AZD6140 group were subrandomized to receive or not receive an initial loading dose of 270 mg. Patients were scheduled to receive 1, 2, or 3 months of study drug, depending on when during the trial period they were enrolled. For patients undergoing percutaneous coronary intervention (PCI) within 48 h post-randomization, an additional 300 mg of clopidogrel study drug or placebo (clopidogrel for patients in the clopidogrel group or placebo for patients in the AZD6140 group) could be administered at the discretion of the treating physician. Patients returned monthly for follow-up visits, and were contacted by telephone 7 days after stopping study drug for evaluation of any adverse events.
The primary objective was to assess the safety and tolerability of the different doses of AZD6140 plus aspirin, versus clopidogrel plus aspirin, in patients with NSTE-ACS by evaluating total bleeding events (major plus minor bleeding, but excluding minimal bleeds as adjudicated by the Independent Clinical Adjudication Committee) observed within the first 4 weeks of treatment (see Online Appendix 2for definitions). Additional objectives of the trial were to assess: 1) individual and composite incidence of MI (including silent MI), death, stroke, and severe recurrent ischemia; and 2) incidence of recurrent ischemia with AZD6140 plus aspirin and clopidogrel plus aspirin, using total duration of ischemia as detected by continuous Holter electrocardiogram monitoring (LifeCard CF, DelMar Reynolds, Irvine, California) during the first 4 to 7 days after randomization.
Statistical analyses were performed using SAS version 8.1 (SAS Institute, Inc., Cary, North Carolina), with details provided in Online Appendix 2. An independent data safety monitoring board comprising clinical experts and a statistician monitored safety data on an ongoing basis to ensure that patient safety was maintained. Comparisons between AZD6140 and clopidogrel were made using the Cox proportional hazards model (time-to data), Fisher exact test (count data), and the t test (continuous data). The Thrombolysis In Myocardial Infarction (TIMI) study group has a copy of the database and has independently confirmed the data presented here.
A total of 1,018 patients were screened and provided informed consent, and 990 patients from 132 sites were randomized. After randomization, 984 patients received at least 1 dose of study drug and were included in the safety analysis dataset. A total of 491 (50%) were scheduled to receive study drug for 12 weeks, 243 (25%) for 8 weeks, and 250 (25%) for 4 weeks. The median duration of study drug treatment was 56 days.
The mean age of patients enrolled was 63 years (range 30 to 93 years), more than a third of were women, 24% had diabetes mellitus, and 48% had ST-segment depression ≥0.5 mm and 62% were coded as non–ST-segment elevation MI (Table 1).Ninety-eight percent of patients received aspirin, 85% received heparin (51% unfractionated heparin, 40% low-molecular-weight heparin, not mutually exclusive), 31% received glycoprotein IIb/IIIa inhibitors, 85% received beta-blockers, and 86% received statins, without differences between the treatment groups. A total of 67% of patients underwent coronary angiography, 42% had PCI (2% balloon angioplasty, 40% stenting, of which 48% had drug-eluting stents), and 9% underwent post-randomization coronary artery bypass grafting (CABG). The median time to PCI was 20 h after randomization (interquartile range [IQR] <2 to 72 h), and the median time to CABG was 10 days after randomization (IQR <7 to 19 days).
Primary end point
The Kaplan-Meier rate of the primary end point, protocol-defined major or minor bleeding at 4 weeks, was not different among the 3 groups, with 26 patients (8.1%) experiencing a bleeding episode in the clopidogrel group, compared with 32 (9.8%) in the AZD6140 90-mg group and 25 (8.0%) in the 180-mg group (p = 0.43 and p = 0.96 vs. clopidogrel, respectively) (Table 2).The rates of “major—fatal/life-threatening” bleeding and “major—other” bleeding were not different among the AZD6140 and clopidogrel groups (Fig. 1).There were 2 fatal bleeds, both in the AZD6140 90-mg group. Protocol-defined minor bleeding occurred in 4 patients (1.3%) in the clopidogrel group, 9 (2.7%) in the AZD6140 90-mg group, and 12 (3.8%) in the AZD6140 180-mg group over 4 weeks (p = 0.18 and p = 0.05 vs. clopidogrel, respectively). A similar pattern of bleeding was seen up through 12 weeks.
The most common type of bleeding was epistaxis, followed by periprocedural hemorrhage or hematoma. Seventy-three percent of all bleeds in the clopidogrel group were procedure related versus 53% and 52% in the AZD6140 90- and 180-mg groups, respectively. Gastrointestinal bleeding occurred in 3 clopidogrel patients (0.9%), 7 AZD6140 90-mg patients (2.1%), and 4 AZD6140 180-mg patients (1.2%). Bleeding episodes that led to discontinuation of study treatment occurred in 3 clopidogrel patients (0.9%), 8 patients (2.4%) in the AZD6140 90-mg group, and 5 (1.5%) in the AZD6140 180-mg group. For the entire study period, blood transfusions were required by 22 patients (6.7%) in the clopidogrel group, 24 (7.2%) in the AZD6140 90-mg group, and 15 (4.6%) in the AZD6140 180-mg group; the median number of units transfused was 3 (IQR 2 to 4 U). Protocol-defined minimal bleeding (nonadjudicated) occurred in 70 (21%) patients in the clopidogrel group and in 89 (27%) and 100 (31%) patients in the AZD6140 90- and 180-mg groups, respectively, over the entire study period (p = 0.12 and p = 0.006, vs. clopidogrel, respectively).
Assessment of bleeding by AZD6140 loading dose showed that major bleeding within the first 48 h occurred in 8 clopidogrel patients (2.4%), compared with 3 (1.8%), 2 (1.3%), and 6 (1.8%) patients in the AZD6140 90-, 180-, and 270-mg groups, respectively (Fig. 2).Protocol-defined minor bleeding within the first 48 h occurred in 1 (0.3%) clopidogrel patient and in 1 (0.6%), 0, and 6 (1.8%) patients in the AZD6140 90-, 180-, and 270-mg groups, respectively.
Among the 84 patients who underwent CABG, median times (IQRs) to the procedure from the last dose of study drug were 96 (40 to 168), 55 (30 to 124), and 60 (29 to 144) h in the clopidogrel, AZD6140 90-mg, and AZD6140 180-mg groups, respectively. One of 2 clopidogrel patients and 5 of 10 AZD6140 patients who underwent CABG within 1 day of receiving study drug had a major bleed. Among those undergoing CABG between 1 and 5 days after the last dose, the rate of major bleeding was numerically lower in the AZD6140 group (10 of 28 patients; 36%), compared with the clopidogrel group (9 of 14; 64%). Beyond 5 days, the rate of major bleeding was 60% (6 of 10) for clopidogrel patients and 50% (10 of 20) for AZD6140 recipients.
The rates of death or cardiovascular death were not different among groups (Table 3).There was a numerical trend toward lower rates of MI in the AZD6140 groups, with 15 (5.6%) patients in the clopidogrel group, 12 (3.8%) in the AZD6140 90-mg group, and 8 (2.5%) in the AZD6140 180-mg group experiencing an MI. The respective hazard ratios were 0.80 and 0.53, with confidence intervals (CIs) of 0.37 to 1.70 and 0.23 to 1.25. The Kaplan-Meier event rate of cardiovascular death, MI, or stroke, the combined end point used in the CURE (Clopidogrel in Unstable angina to prevent Recurrent Events) trial, showed similar rates between clopidogrel (6.2%) and AZD6140 90-mg (6.0%) but a numerically lower rate of 3.5% in the AZD6140 180-mg group, with a hazard ratio of 0.65 and a 95% CI of 0.30 to 1.38 (Fig. 3).Rates of periprocedural MI were 1.5% for the clopidogrel group and 2.1% and 0.9% for the AZD6140 90- and 180-mg groups, respectively. This accounted for 43% of the overall number of MI and was similar in each of the groups. No silent MIs or CABG-associated MIs were reported. After discontinuation of study medication, the number of major cardiovascular events in the clopidogrel, AZD6140 90- and 180-mg groups were 5, 12, and 6, respectively. The numbers of cardiovascular deaths were 1, 6, and 3, and MIs were 3, 3, and 2, respectively.
Table 4lists crude incidence rates of other investigator-reported adverse events. Nonspecific factors, such as headache, were common. There appeared to be a higher rate of nausea, dyspepsia, and hypotension reported among AZD6140 recipients. Dyspnea was more common in the AZD6140 groups (Table 4). Of those who reported dyspnea, 27% of the patients had resolution of this symptom within 24 h, 25% had resolution of the dyspnea after 24 h, and 48% experienced persistent symptoms during treatment (>15 days). The overall incidence of persistent dyspnea was approximately 2% for clopidogrel, and 6% for each of the AZD6140 groups. Discontinuation rates were low and similar between groups: 6% of patients receiving AZD6140 90 mg twice a day, 7% of patients receiving AZD6140 180 mg twice a day, and 6% of patients receiving clopidogrel 75 mg once a day.
Table 5presents the results of the arrhythmia analysis from 885 patients (89.4% of all patients enrolled). There was no difference in the rates of ventricular tachycardias among the 3 treatment groups. However, there were a greater number of mostly asymptomatic ventricular pauses lasting >2.5 s detected post hoc among patients receiving AZD6140 compared with those receiving clopidogrel. Of the patients who experienced pauses >5 s, 7 were due to sinus block or sinus node exit block and 4 were due to complete heart block.
This trial evaluated the first reversible oral ADP antagonist, AZD6140, in a patient population with ACS. As reported in the DISPERSE-2 substudy companion paper (13), the doses tested yielded a level of platelet inhibition that was nearly double that of clopidogrel, indicating that more effective P2Y12inhibition is feasible. With respect to the primary objective of assessing the overall safety and tolerability of AZD6140, the current study demonstrated no significant difference, despite the higher level of platelet inhibition, between this agent and clopidogrel in the rate of protocol-defined major bleeding (which is similar to the definition of TIMI major or minor bleeding). Slightly more protocol-defined minor bleeding was observed in the AZD6140 groups, but with numerically lower rates of major bleeding. Thus, the rate of overall protocol-defined major or minor bleeding was not different among the AZD6140 and clopidogrel groups. The bleeding rates also were not different among patients who had or had not previously received clopidogrel or among those who did or did not receive a loading dose of AZD6140 or did or did not receive glycoprotein IIb/IIIa inhibitors. Thus, in this initial experience, it appears that the higher level of inhibition of P2Y12is reasonably well tolerated, even in the face of multiple additional antithrombotic agents and with invasive procedures being performed in more than two-thirds of patients.
There was an intriguing, numerically lower rate of bleeding in AZD6140-treated patients undergoing CABG between 1 and 5 days after stopping study drug, which would be consistent with a recovery of platelet function due to the reversible binding of AZD6140 to the P2Y12receptor. However, further study in a larger number of patients is needed to confirm such a benefit with this agent, compared with the nonreversible effects of clopidogrel (or other thienopyridines). If these results were confirmed, however, the use of the reversible agent AZD6140, with a half-life of 12 h, would allow greater flexibility for use of P2Y12inhibitors. One could treat all patients at the time of presentation with ACS, but stop the treatment for patients who are found on coronary angiography to need CABG. This would avoid the need to make patients wait 5 days for CABG, as recommended in current American College of Cardiology/American Heart Association guidelines and supported by recent studies (14–16). Similarly, for patients treated chronically post-ACS, it would potentially allow greater flexibility for other types of elective surgery.
There were not sufficient numbers of clinical events to reliably determine the efficacy of AZD6140 versus clopidogrel. The overall rates of clinical events, however, were consistent with prior ACS trials, and there was an apparent dose-response in the MI rate among AZD6140 recipients, with the highest rate being observed among patients treated with clopidogrel, and lower rates among those who had received the 90- and 180-mg doses of AZD6140. The overall composite rate of cardiovascular death, MI, or stroke appeared to be numerically lower only at the higher dose, but the CIs for these clinical events are wide, and thus a large phase III trial (which is ongoing) will be required to evaluate the true efficacy of this agent. It is notable, however, that the benefit of AZD6140 in preventing MI appeared to be present both surrounding the time of PCI and also during the 1- to 3-month follow-up period, with 57% of the MIs being spontaneous in this study.
Judging by the rate of discontinuation of (blinded) study medication, the overall tolerability of AZD6140 was good, with a similar percentage of patients discontinuing study medication in each of the 3 groups. There were differences in the adverse events reported between drugs, however, with higher incidences of dyspnea, hypotension, and nausea among AZD6140 recipients. There also was a mild increase in uric acid levels with AZD6140. The number of patients who discontinued study treatment due to these reported side effects, however, was small (1% to 4% overall). Nonetheless, this observation highlights the fact that each new class of drugs has the potential for specific side effects. Thrombocytopenia or neutropenia was not observed with either AZD6140 or clopidogrel in this study.
The present study confirmed the dose-dependent increase in reported dyspnea in the AZD6140 groups as first reported in the DISPERSE study, but also reports a rate of dyspnea in the clopidogrel group. As in the DISPERSE study (12), the episodes were reported as mild or moderate in severity, were often self-limited, and rarely led to study drug discontinuation. There were not reported signs of fluid retention, congestive heart failure, or bronchospasm in the cases of AZD6140-related dyspnea, suggesting it does not relate to heart failure. Investigations are ongoing to determine if any physiologic basis might exist for this finding.
The greater number of mostly asymptomatic ventricular pauses detected post hoc in patients receiving AZD6140 was both unexpected and not previously reported in studies of AZD610. Although there is no known mechanism to explain this observation, it is possible that AZD6140, although itself not an ATP or adenosine analog, may affect adenosine metabolism. Several agents, such as dipyridamole, draflazine, and iodotubercidin, modulate intra- and paracellular adenosine concentration via interference with adenosine degradation and reuptake inhibition. If AZD6140 has similar actions, it is possible that the increased incidence of asymptomatic ventricular pauses and other observed side effects of AZD6140 such as dyspnea and hyperuricemia may also be mediated via modulation of adenosine metabolism (17,18). Further assessments of the effects of AZD6140 on adenosine metabolism and on cardiac conduction are required.
This study has demonstrated that AZD6140 shows similar safety and tolerability to clopidogrel. Because AZD6140 is a reversible inhibitor of the P2Y12receptor, it offers the potential for flexibility with respect to more rapid initiation of coronary bypass and other surgical procedures after discontinuation of the drug. With the lack of increased major bleeding and the encouraging trends in efficacy observed in the current study, this agent is currently being studied in PLATO (PLATelet inhibition and patient Outcomes), a large outcomes trial in 18,000 patients with ACS.
For a full list of investigators, and detailed entry criteria, definitions, and statistical analysis, please see the online version of this article.
Cardiosource Slide Set
Safety, Tolerability, and Initial Efficacy of AZD6140, the First Reversible Oral Adenosine Diphosphate Receptor Antagonist, Compared With Clopidogrel, in Patients With Non-ST-Segment Elevation Acute Coronary Syndromes Primary Results of the DISPERSE2 Trial and Inhibition of Platelet Aggregation by AZD6140, A Reversible Oral P2Y12 Receptor Antagonist, Compared With Clopidogrel in Patients With Acute Coronary Syndromes
↵1 Dr. Husted has received research grants from AstraZeneca.
↵2 Dr. Harrington has received research grants, consultancy fees, or honoraria from AstraZeneca, Bristol-Myers Squibb, Eli Lilly and Company, Sanofi-Aventis, and The Medicines Company.
↵3 Dr. Emanuelsson is an employee of AstraZeneca.
↵4 Dr. Peters is a former employee of AstraZeneca.
↵5 Dr. Storey has received research grants, consultancy fees, or honoraria from AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo, Eli Lilly and Company, Sanofi-Aventis, and The Medicines Company.
This study was funded by AstraZeneca. In addition, Dr. Cannon receives research funding from Schering Plough, and the TIMI Study Group has received research funding from Daichi-Sankyo, Eli Lilly and Company, Bristol-Myers Squibb, and Sanofi-Aventis, manufacturers of oral antiplatelet agents.
- Abbreviations and Acronyms
- acute coronary syndromes
- adenosine diphosphate
- coronary artery bypass grafting
- confidence interval
- interquartile range
- non–ST-segment acute coronary syndromes
- percutaneous coronary intervention
- Received May 18, 2007.
- Revision received July 27, 2007.
- Accepted July 30, 2007.
- American College of Cardiology Foundation
- Matetzky S.,
- Shenkman B.,
- Guetta V.,
- et al.
- Wiviott S.D.,
- Antman E.M.
- Gurbel P.A.,
- Bliden K.P.,
- Hiatt B.L.,
- O’Connor C.M.
- Husted S.,
- Emanuelsson H.,
- Heptinstall S.,
- Sandset P.M.,
- Wickens M.,
- Peters G.
- Storey R.F.,
- Husted S.,
- Harrington R.A.,
- et al.
- Braunwald E.,
- Antman E.M.,
- Beasley J.W.,
- et al.
- Kapetanakis E.I.,
- Medlam D.A.,
- Petro K.R.,
- et al.
- Bavry A.A.,
- Lincoff A.M.
- Dennis D.M.,
- Raatikainen M.J.,
- Martens J.R.,
- Belardinelli L.