COMMIT-CCS2 logo ClOpidogrel and Metoprolol in Myocardial Infarction Trial
FACT SHEET 3
Questions and Answers about the design and conduct of COMMIT/CCS-2
Q What is COMMIT/CCS-2?
A It is the ClOpidogrel and Metoprolol in Myocardial Infarction Trial/Second Chinese Cardiac Study – the largest clinical study ever conducted in China and the world's second largest clinical study of the emergency treatment of heart attacks (acute myocardial infarction or MI). This randomised trial has investigated the effects of adding the oral anti-platelet drug clopidogrel to aspirin, and of starting the beta-blocker metoprolol, within 24 hours of the onset of a heart attack.
Q When did the study start and when did it end?
A The first patients were recruited in October 1999. Within a year more than 800 patients a month were entering the study, and this recruitment level was maintained until the study closed on 9 February 2005.
Q Where was the study undertaken and how many patients took part?
A The study was carried out in China, and 45,852 patients from 1,250 collaborating hospitals took part in the study.
Q Who carried out the study?
A It was a joint Chinese-British venture. The international coordinating centre was at the Clinical Trial Service Unit (CTSU) of Oxford University, UK, which worked in close collaboration with the national coordinating centre at the Fuwai Hospital, Chinese Academy of Medical Sciences, Beijing, China.
Q Why was clopidogrel studied?
A The use of aspirin for the emergency treatment of heart attacks had been shown previously by the Oxford research team to reduce the risk of death and repeat heart attacks by about one quarter. As a consequence of that research, aspirin is now used routinely during, and after, a heart attack. The objective of COMMIT/CCS-2 was to see whether adding clopidogrel to aspirin during a heart attack might be even more effective at preventing death or a repeat heart attack than is aspirin alone.
Q What do aspirin and clopidogrel do?
A Aspirin is an anti-platelet drug that prevents blood clots. Specifically, it irreversibly inhibits an enzyme complex called cyclooxygenase, blocking the formation of thromboxane A2, a potent agent that makes platelet cells in the blood stick together. Clopidogrel is also an anti-platelet drug, introduced in 1998, which works through a different mechanism from aspirin. It inhibits an energy storage molecule called adenosine diphosphate, which also makes platelets stick together. The study's objective is to discover whether blocking these two pathways simultaneously by using an aspirin/clopidogrel combination produces a greater clinical effect than blocking the aspirin pathway alone.
Q What doses of aspirin and clopidogrel were studied?
A All patients received 162mg oral aspirin daily. Half were randomly allocated to receive 75mg oral clopidogrel daily for up to 4 weeks (or prior hospital discharge), and the other half were allocated to receive matching placebo (“dummy”) tablets.
Q Why was metoprolol studied?
A The use of beta-blockers in the years after a heart attack had been shown previously to reduce the risk of death and repeat heart attacks, and such treatment is now used routinely. But, despite the use of intravenous beta-blockers for the emergency treatment of heart attacks having been investigated in about two dozen clinical studies during the 1970s and 1980s, its value remains uncertain (especially on top of current standard therapy). So, such use of beta-blockers is limited, even though current guidelines advise their prompt administration during a heart attack unless there are contraindications. COMMIT/CCS-2 aimed to establish whether starting intravenous and then oral metoprolol during a heart attack could cut deaths and reduce the risk of repeat heart attacks and cardiac arrest.
Q What does metoprolol do?
A Metropolol is a beta-blocker, introduced in 1975, which blocks the action of adrenaline (epinephrine) and noradrenaline (norepinephrine), two chemicals that increase the heart rate and raise the blood pressure. Metoprolol has been shown to be of value for treating hypertension, certain arrhythmias and heart failure, as well as for preventing further heart attacks during the years after a heart attack. The study's objective is to discover whether slowing the heart rate and reducing the force of the heartbeat, thus reducing the workload on the heart and preventing life-threatening arrhythmias early after the onset of a heart attack, is beneficial.
Q What dose of metoprolol was studied?
A Half of the patients were randomly allocated to receive up to 15 mg intravenous metoprolol over about 15 minutes followed by 200mg oral metoprolol daily for up to 4 weeks (or prior hospital discharge), and the other half were allocated to receive matching placebo injections and tablets.
Q What type of study was it?
A A randomised double-blind placebo-controlled clinical trial – that is, one in which patients were randomly allocated to receive the study treatments or matching “dummy” placebo (in addition to all the standard treatments for a heart attack), but neither the patients nor their doctors knew which they were receiving.
Q How was the study able to assess clopidogrel as well as metoprolol?
A It used what is known as a factorial (2X2) design:
  • 1/4 of patients received active clopidogrel and active metoprolol
  • 1/4 of patients received active clopidogrel and placebo metoprolol
  • 1/4 of patients received placebo clopidogrel and active metoprolol
  • 1/4 of patients received placebo clopidogrel and placebo metoprolol
The assessment of clopidogrel involved comparison of outcome in the two quarters that received active clopidogrel versus that in the two quarters that received placebo clopidogrel. Similarly, the assessment of metoprolol involved comparison of the two quarters that received active metoprolol versus the two quarters that received placebo metoprolol. (N.B. All patients were allocated active aspirin, with all other treatments at the discretion of the patients’ own doctors.)
Q Previous studies have already shown that clopidogrel and metoprolol are effective in treating heart disease. How could you justify using placebo?
A Adding clopidogrel to aspirin had previously been shown to be beneficial for people who had unstable angina or were undergoing percutaneous coronary intervention [PCI] balloon angioplasty, but there was no good evidence for its use during a heart attack. For metoprolol, although over two dozen studies had previously assessed the effects of starting intravenous then oral beta-blocker therapy during a heart attack, substantial uncertainties remained about the balance of any benefits and hazards of such therapy when added to current standard treatments for a heart attack. Entry to the study was guided by the “uncertainty principle”: that is, each patient was to be considered for inclusion in the study only when their responsible doctors were substantially uncertain as to whether the study treatments were clearly indicated or clearly contraindicated for that particular patient.
Q What percentage of acute MI patients receive these drugs already?
A Clopidogrel is becoming increasingly used as long-term therapy for unstable angina or during coronary artery procedures (such as PCI or stenting), but its use during a heart attack is low, presumably because there is no clear evidence about the balance of benefits and risks in this setting. The use of intravenous beta-blocker during a heart attack varies widely between countries: for example, there is about 100-fold variation in Europe, ranging from less than 1% in the UK to over 50% in Sweden.
Q Who was eligible for the study?
A Patients were eligible if they presented with:
  • suspected acute heart attack within 24 hours of the onset of symptoms;
  • ST elevation or other ischemic abnormality on their electrocardiogram; and
  • no clear indication for, or contraindication to, any of the study treatments
and they then provided their informed consent to participate.
Q Who might have been excluded?
A Patients were to be excluded if there was thought to be:
  • an increased risk of possible adverse effects of the study treatments; or
  • only a small likelihood of worthwhile benefit from the study treatments (e.g. unconvincing evidence of a heart attack or undergoing primary percutaneous coronary intervention [PCI] balloon angioplasty)
Q What possible adverse effects of study treatment might warrant exclusion?
A Examples would include:
  • previous allergy to aspirin or any of the study drugs
  • active bleeding or history of bleeding disorder
  • persistent very low blood pressure (e.g. systolic pressure <100mmHg)
  • persistent very slow heart beat (e.g. <50 beats/min)
  • third degree heart block(complete loss of electrical signals between upper and lower heart chambers) or need for a pacemaker
  • cardiac shock (a condition in which the heart muscle is so severely damaged that it is unable to pump sufficient blood to the body)
Q What was the age range of the patients?
A There was no age limit for inclusion in the study. The youngest participant was 15 years of age and the oldest was 100, with a mean age of 61 years, and about 12,000 patients were aged 70 or over. So, the findings should be relevant not only for middle-aged heart attack victims but also for older people who are often not included in clinical studies, and are less likely to be given other emergency treatments (such as “clot-busting” fibrinolytic therapy) for their heart attacks.
Q How many men and how many women?
A There were 33,082 men (72%) and 12,760 women (28%).
Q Why were there many more men than women in the study
A Because women tend to have a much lower incidence of heart attack in early and late middle age than do men.
Q What were other characteristics of patients in the study?
A By age:
Under 60 42%
60-69 32%
70 or over 26%
By time from onset of symptoms to study entry
0-6 hrs 34%
7-12 hrs 33%
13-24 hrs 33%
By systolic blood pressure (mmHg)
under 120 34%
120-139 35%
140-159 20%
160 & over 11%
By heart rate (beats/min)
under 70 22%
70-89 49%
90-109 22%
110 and over 7%
Q How long were patients in the study?
A Following admission to hospital with a suspected heart attack, patients were scheduled to continue study treatment and follow-up for up to four weeks or until prior hospital discharge. In practice, the average duration of the hospital stay and study treatment was 16 days.
Q Did all randomised patients complete their allocated study treatment?
A Both study treatments are well tolerated. For the clopidogrel comparison, 93% of patients completed their scheduled study clopidogrel/placebo tablets and there was no difference between the clopidogrel and placebo groups in the numbers who stopped these tablets prematurely. For the metoprolol comparison, intravenous study treatment was stopped more commonly in the metoprolol group than the placebo group (10% versus 4%), and so too was the oral treatment (14% versus 8%), chiefly due to persistent lower blood pressure and slower heart rate.
Q Did you have to unblind any patients? If so, why?
A A 24-hour telephone service was available to deal with any relevant medical queries, but the study treatment of only about a dozen patients needed to be “unblinded” (i.e. active or placebo nature of the allocated study treatment disclosed) for various medical conditions throughout the whole study period.
Q Were patients also given other therapies while taking part in the study?
A Yes, the patients’ doctors were free to use any other treatments in hospital that they considered necessary. So, as well as all patients being given aspirin, they received:
fibrinolytic agents 55%
anticoagulants 68%
ACE inhibitors 74%
anti-arrhythmics 23%
diuretics 23%
calcium antagonists 12%
nitrates (oral or iv) 94%
The level of use of these treatments in this study in China was similar to that found in similar situations in the US or Europe. Moreover, when patients who presented within 12 hours of symptom onset with definite ischaemic ECG changes (ST segment elevation) are considered separately, the use of fibrinolytic therapy increased to 68% (and it increased further with restriction to patients aged under 70). Typically in China, the fibrinolytic agent used was urokinase, whereas tissue plasminogen activator in the US and streptokinase in Europe are more commonly used.
Q Didn't these other therapies affect the assessment of the study treatments?
A No, because the use of these other therapies tended to be balanced between the study treatment and placebo groups in this double-blind study. So, in principle, it provides an unbiased assessment of the effects of adding clopidogrel and metoprolol to currently standard treatments for a heart attack.
Q Are the findings from this study in China applicable to other populations?
A Yes. There is no evidence that the nature of heart attacks is materially different in China than elsewhere, or that Chinese patients react differently to these treatments than do other populations. Furthermore, the concomitant therapies used in hospital in the study were very similar to those typically used in other countries.
Q Why did the study have to be so large?
A The study sought to determine the effects of treatment on hospital mortality and other serious outcomes (such as repeat heart attack, stroke or cardiac arrest) for which even quite small improvements could be very worthwhile. Studying large numbers of patients reduces the play of chance in the results. By involving 40-50,000 patients, the study has good statistical power to detect reductions in the risks of death or of other major vascular events as small as 10 per 1000 (e.g. 10% dead reduced to 9% dead).
Q If the treatments produce only modest benefits is this worthwhile?
A Yes. Acute MI is a very common condition, with about 10 million heart attacks every year worldwide. Moreover, both clopidogrel and metoprolol are simple to use, and 2-weeks of treatment is not expensive. So, the reliable demonstration of modest benefits in this common disease with such widely applicable treatments could mean that thousands of patients avoid death or serious non-fatal vascular events each year.
Q Was it difficult to recruit patients?
A Successful recruitment depended on the widespread collaboration of many doctors and nurses throughout China. In order to make it easier for them to collaborate, the study procedures were streamlined by the coordinating centre, with only essential data collected, no extra investigations and no interference with the routine care of patients. Apart from giving the study treatments, all other aspects of patient management were entirely at the discretion of the patients’ responsible doctors.
Q This huge study involved 1,250 hospitals thousands of miles apart. How did you manage to ensure the study was well-run and the results reliable?
A Careful design at the beginning aimed to make it straightforward for collaborating doctors and nurses to participate in the study, with minimal extra work and without disruption of their routine practice. The study treatments were conveniently packaged, using calendar packing to facilitate compliance, and were provided in packs with clear instructions and brief study forms. A number of centralised quality control checks were implemented throughout the course of the study, including random testing of drug packs, extensive computerised checks and validation of data, and regular monitoring of recruitment patterns and patient characteristics in each hospital. In addition, on-site auditing of medical notes was performed at 350 of the fastest recruiting hospitals which accounted for about two thirds of all randomised patients.
Q How did you manage to persuade 1,250 hospitals to take part in the study?
A The key factor to persuading a large number of hospitals to participate was to make the study attractive to clinicians, with streamlined patient entry involving little or no extra testing or examination, with collection only of essential data at entry and follow up, with reimbursement for the small amount of extra work involved, and with publication of the main study findings in the name of the whole collaborative group (rather than just the study organisers). Potential collaborating hospitals were identified in various ways, including from personal contacts of the principal investigators and experience in previous studies, as well as national and regional hospital directories.
Q What approval and patient consent procedures were there for the study?
A The study was approved by the Chinese Ministry of Health and China State Drug Administration Bureau, as well as by national and regional ethics or health research boards. Written informed consent had to be obtained either from the patient or (if too ill to give consent) from a close family member prior to entry of each patient into the study.
Q Were study doctors paid or did they receive any other incentive to take part?
A Collaborating doctors had their travel costs covered for attending investigators’ meetings. They were also reimbursed with a small amount of money (about $12 per patient) to cover the costs of the minimal extra clinical time spent on recruiting, treating and managing patients enrolled in the study. No other fee or financial incentive was given to doctors or their hospitals for participating in the study.
Q How much did the study cost and who funded it?
A It cost US$3 million [UK£1.6 million or 24.6 million Chinese Yuan]. This cost was split between the manufacturers of clopidogrel (Bristol-Myers Squibb and Sanofi-Aventis) and of metoprolol (AstraZeneca). The British Heart Foundation and UK Medical Research Council also provided core staff and infrastructure support to the CTSU.
Q Is the study independent of the funders?
A Yes. The idea for this study came from the principal investigators in CTSU, following close discussion with Chinese colleagues. The funding was provided as grants to Oxford University, and the pharmaceutical companies had no say in the running of the study, the analysis of the data, or the way the results are presented, published or publicised.
Q Were any other steps taken to ensure the independence of the research?
A Yes. An independent data monitoring committee, which did not include anyone involved in the study, has been examining interim results in confidence during recruitment to ensure that there were no safety concerns. As no such issues arose during the study, it continued until its scheduled end.