Experts discuss FDA advisory against aspirin for primary prevention of MI, stroke

Issue: June 2014

The FDA in May issued a public health advisory announcing that an evidence review shows no support for the general use of aspirin for primary prevention of MI or stroke. At the same time, the agency informed Bayer HealthCare that it had rejected the firm’s citizen petition to amend the rule on professional labeling for aspirin, which would have allowed marketing of aspirin for MI prevention to people with no history of CVD. The agency stated that current evidence for primary prevention of MI was not strong enough to outweigh bleeding risks.

Cardiology Today asked its Editorial Board and other experts in the field to provide perspective on the FDA’s decision.

Carl J. Pepine, MD, Cardiology Today Chief Medical Editor

Carl J. Pepine

Estimates suggest that more than one-third of the US population uses aspirin regularly. Most of these users have manifest CVD, although a large cohort do not. Thus, this medication has important public health implications and the recent FDA advisory against general use of aspirin for primary prevention is likely not to be embraced by many.There are many concerns with this old drug. Although the bleeding risks are well known, we do not know the optimal dose and it is established that aspirin-related bleeding is dose-related. This lack of specific dose issue, in my opinion, also applies to those who have already experienced an event. It is unlikely that doses used in older secondary prevention trials represent truly optimal doses for secondary prevention in today’s setting.

However, relative to the recommendation against aspirin use for primary prevention, it seems to me that although we lack randomized trial evidence, there is likely a high-risk subgroup that could benefit. Many cardiologists, including myself, would have a discussion with the patient about aspirin use if the risk for adverse outcome was high (eg, 10-year risk for death, MI or stoke was ≥10%). This discussion would include the lack of randomized trial data, the bleeding risk and the lack of optimal dose information. Nevertheless, the majority of active, apparently healthy men aged older than 65 years in my practice will fall into this risk category, and most will not have bleeding risks. Reducing the risk of MIs and/or strokes in this rapidly growing segment of our population is clearly important and most bleeding events in this cohort are neither fatal nor disabling.

JoAnn E. Manson, MD, DrPH, FAHA, Brigham and Women's Hospital and Harvard Medical School

JoAnn E. Manson

The FDA’s decision doesn’t override or invalidate the guidelines of numerous professional societies regarding the appropriate use of aspirin for CVD prevention. The key is that any decisions about regular aspirin use should be based on the balance of benefits and risks for the individual patient, and this requires input and supervision by the patient’s health care provider who knows the patient’s medical history and risk factor status. Although aspirin is available over the counter, it is not free of risks, and the bleeding risks can be very serious. Many patients perceive OTC medications as being relatively free of risks and much safer than medications requiring prescriptions. Thus, if aspirin had a label for prevention of first MI or first CV event, it’s possible that many patients would begin taking aspirin regularly on their own for this purpose without the guidance of their clinician. When taken regularly by people at low risk for CVD and without any indications for treatment, the risks of aspirin can far outweigh the benefits. Thus, clinical decision making about regular aspirin use needs to be individualized and personalized, with active collaboration between the patient and health care provider.For women, the most relevant primary prevention trial on this topic is the Women’s Health Study, which showed that aspirin lowered the risk for stroke in women, but did not lower their risk for MI or CV death and increased their risk for bleeding. However, aspirin was much more effective at preventing MI and total CVD events, and had a more favorable benefit-risk profile, in women aged 65 years and older than in younger women. (Ridker PM. N Engl J Med. 2005;352:1293-1304.) For men aged 40 to 84 years, the Physicians’ Health Study found a 44% reduction in risk for MI (RR=0.56; 95% CI, 0.45-0.7) for those assigned aspirin vs. those assigned placebo, but no benefit for reduction of risk for stroke or CV death and an increased risk for bleeding. (Steering Committee of the Physicians’ Health Study Research Group. N Engl J Med. 1989;321:129-135.)

PAGE BREAK

Haitham Ahmed, MD, MPH, Johns Hopkins Hospital; Roger S. Blumenthal, MD, Cardiology Today Section Editor

Haitham Ahmed

Roger S. Blumenthal

CVD medications are often categorized by their intent to reduce MI or stroke risk in either the primary or secondary prevention setting. For decades, aspirin has been used and indicated for secondary prevention of MI and stroke based on thorough data showing benefit in the literature. The FDA has recently published a public health advisory stating it does not support general use of aspirin for the primary prevention of MI or stroke due to the risk of bleeding outweighing potential benefit.To think of CVD prevention in binary terms, such as ‘primary’ and ‘secondary,’ is not only outdated but also flawed. Atherosclerotic vascular events do not develop overnight. They are an end outcome that develops after decades of plaque build-up. To think that a coronary artery that is 100% occluded should be treated one way, but one that is only 60% occluded (prior to infarction) be treated another way, does not make much sense.

Rather than thinking of CVD in primary or secondary prevention terms, we prefer to look at CVD risk as a continuum. Atherosclerosis develops over decades across this risk continuum, slowly and progressively increasing risk of clinical CVD. To reserve aspirin use for only those at the extreme end of this continuum after they have had a coronary event would be too late.

For these reasons, we do not agree with a blanket statement discouraging aspirin use for all primary prevention purposes. There is no one-size-fits-all here. Patients deserve a tailored assessment and personalized risk-benefit discussion about aspirin with their providers. We agree with the American Heart Association/American College of Cardiology recommendation that high-risk primary prevention patients should be considered for aspirin therapy. One of the ways to start that discussion is using the 2013 AHA/ACC risk assessment (by pooled cohort equation). Other tools such as coronary artery calcium may help further refine risk and guide decisions to treat. Our group has recently published data from the Multi-Ethnic Study of Atherosclerosis (MESA) showing probable net benefit with aspirin treatment for patients with a CAC score >100 and net harm for those with scores of 0 (Miedema M. Circ Cardiovasc Qual Outcomes. 2014;7:453-460). Further research is needed to help give our patients the most information and provide them with the personalized assessments that they truly deserve.

Rita F. Redberg, MD, MSc, Cardiology Today Editorial Board

Rita F. Redberg

I don’t know of any new data that have come out on aspirin that would have led to this advisory. The FDA stated they are issuing it now because they denied the request by Bayer, but I don’t see this as a public health emergency. I think aspirin has been very complex for some time in terms of its risks and benefits. There may be some benefit for high-risk people. The data have been somewhat confusing because they look different for men and women as well as for MI and stroke, and indicate a reduction in CV events but not in mortality. A meta-analysis of aspirin for primary prevention of CVD found pretty consistently a reduction of 10% for CV events, but no reduction for CVD mortality (Seshasai SRK. Arch Intern Med. 2012;172:209-216).

The reason I found it a little surprising is that the data on aspirin benefits seem similar to the data on statins for primary prevention, and the FDA hasn’t issued any warnings like this on statins. Statins for primary prevention have not been shown to reduce all-cause mortality. For some people, there’s a slight reduction in events, but you have to weigh that against what seems to me a much greater adverse event profile compared with aspirin. Yet the FDA issued a warning on aspirin. I find that mystifying.

PAGE BREAK

Joseph S. Alpert, MD, Cardiology Today Editorial Board

Joseph S. Alpert

I agree with the FDA’s stance, but in every clinical situation there is a judgment call. If you’ve got a patient who has never had a coronary event, but has a nasty family history, diabetes and hyperlipidemia, that’s a person that I already know has atherosclerosis, although they’re theoretically classified as primary prevention, and I’m going to put that person on aspirin. But for the patient who comes in at age 30 and says, ‘My uncle had a heart attack in his 40s,’ but doesn’t have much in the way of risk factors, I’m not going to put them on aspirin. The data for women are even weaker, so I really want to be convinced before I start aspirin on a woman. This will probably not impact clinical practice in a huge way. Most doctors are already prescribing aspirin for secondary prevention, but not primary prevention. Of course, a lot times what happens is that the patients start aspirin themselves, because you can buy it OTC for almost no money. The risk for bleeding is very small, but the data say that risk is greater than the risk for a coronary event. If people are concerned about family history and their doctor has told them not to take aspirin, it can’t hurt for them to carry an aspirin in their pocket in case they have chest pain.

Americans love the idea of taking a pill to prevent a disease, but we know that’s not true. There are a whole bunch other things that need to be done: managing weight, eating right, exercising regularly, controlling BP and preventing diabetes if there's a family history.

Michael H. Davidson, MD, FACC, FNLA, Cardiology Today Editorial Board

Michael H.
Davidson

The FDA recently denied the Citizen’s Petition filed by Bayer HealthCare in 2003 to amend the professional labeling of aspirin to include the use of 75 mg to 325 mg per day for the primary prevention of CVD for patients with a Framingham 10-Year Risk Score >10% or a positive risk benefit as assessed by a health professional. After 11 years of review, the FDA denied the petition based on analysis of six landmark aspirin primary prevention trials. The FDA agreed there was an overall significant reduction in nonfatal MI in these trials, but this endpoint was not the primary endpoint; the benefit was inconsistent across the studies and the trials did not exclude patients with a potential history of silent MI. Furthermore, a recent trial (AAA) in patients with decreased ankle-brachial index without known CVD demonstrated no benefit with aspirin use.

In addition, the FDA reviewed the safety of low-dose aspirin and noted a slight absolute increase in risk for gastrointestinal bleeding or cerebral hemorrhage, but not fatal. Lastly, the FDA commented that there are five ongoing primary prevention trials with aspirin and will reconsider the evidence if requested once these trials report the data over the next few years. The denial of the petition is discordant with recommendations by the US Preventive Task Force, AHA/ACC and American Diabetes Association guidelines that encourage aspirin utilization for specific high-risk populations without CVD. The FDA’s more cautious view is also due to the concern that if approved, this could lead to significant promotion directly to the consumer by aspirin manufacturers and although health provider advice would be recommended, there could be the potential for inappropriate use.

As a preventive cardiologist, I will continue to recommend low-dose (81 mg) aspirin use for patients at a high lifetime CV risk. I believe the nonfatal MI reduction is compelling in the first five aspirin primary prevention trials (HR=0.73; P<.0001) although the benefit is attenuated if the Women’s Health Study is added to the analysis (HR=0.77). In addition, nonfatal stroke appears to be reduced (HR=0.81; P=.04). Therefore, I prefer to review the data with the patient and provide a balanced risk-benefit analysis, keeping in mind the patient’s preference. Take myself as an example. My father died of a MI at age 47 and I have a coronary artery calcium score of 30 (I wish it were zero). I take a coated 81 mg dose of aspirin daily, and for my patients with a similar risk profile, I will give them the same advice.

Disclosures: Davidson is the chief medical officer for Omthera Pharmaceuticals and reports speaking for Merck and advising/consulting for AbbVie, Aegerion, Amgen, Esperion, Lipimedix, Pfizer, Sanofi and Vindico. Ahmed, Alpert, Blumenthal, Manson, Pepine and Redberg report no relevant financial disclosures.