FOURIER Presenter, Dr Marc Sabatine, Interviewed

Interviewer: John M. Mandrola, MD; Interviewee: Marc S. Sabatine, MD, MPH


March 24, 2017

This feature requires the newest version of Flash. You can download it here.

John M. Mandrola, MD: Hi, everyone. This is John Mandrola from on Medscape, and I am here at the 2017 American College of Cardiology (ACC) meeting. I am very pleased to be talking with Dr Marc Sabatine, who is the lead author of the FOURIER trial,[1] a major trial here at ACC. Welcome, Dr Sabatine.

Marc S. Sabatine, MD, MPH: Hello, John. Thank you so much for having me.

Dr Mandrola: The FOURIER trial is a randomized controlled trial of the PCSK9 inhibitor evolocumab. Give us some background of the trial.

Dr Sabatine: Sure. It has been known that evolocumab reduced LDL cholesterol (LDL-C) by about 60%, and that is true whether given as monotherapy[2] or on top of statins.[3,4] What we did not have was information for cardiovascular (CV) outcomes. We had hints from some extension studies from phase 2 and phase 3 trials that I had presented at ACC about 2 years ago,[5] but we needed a definitive cardiovascular outcomes trial.

Dr Mandrola: FOURIER was a huge trial.

Dr Sabatine: Yes, we enrolled over 27,000 patients. They were high-risk, stable patients. They had to have a history of myocardial infarction (MI), stroke, or symptomatic peripheral arterial disease (PAD), and they needed to be on an optimized lipid-lowering regimen—a high-intensity statin if they could tolerate it; otherwise, moderate-intensity statin therapy with or without ezetimibe.

Dr Mandrola: This is really a secondary prevention trial—patients with established heart disease, vascular disease.

Dr Sabatine: Yes, that is exactly right. We had over 1200 sites in 49 countries, so it was a global trial.

Dr Mandrola: A fair number of conference calls.

Dr Sabatine: A lot of conference calls and a lot of meetings.

FOURIER: Why so Big?

Dr Mandrola: How did you decide on how big the trial should be?

Dr Sabatine: That was driven by the statistics. It was an event-driven trial based on the key secondary endpoint, the composite of CV death, MI, and stroke. That is how the trial was powered. We thought the event rate was going to be about 2% per year. We were deliberately conservative. It turns out that the event rate was higher. Our enrichment criteria worked better than we anticipated. Because the event rate was a little bit higher, the trial duration was a little shorter than planned. With the 27,000 and an expected 2% event rate, we expected the trial to enroll and accrue the sufficient number of events in about 4 years.

Dr Mandrola: How did you go about picking the endpoints? First, I am interested in how one gets primary and secondary endpoints. What is the thinking?

Dr Sabatine: This differs from some other trials. We knew that the most important outcomes to physicians and patients would be CV death, MI, and stroke. That is how the trial was powered. We also knew that we were going into somewhat unchartered waters with really low levels of LDL-C. We know that people would be interested in looking at a variety of subgroups. As you do the subgroups, you want to make sure that they are adequately powered, given that we would be the first in these unchartered waters.

We were overpowered for the primary endpoint, which included softer outcomes for hospitalization, for unstable angina and coronary revascularization, which had looked good in older trials of statins, so we wanted to look at them as well. We powered it for the key secondary endpoint, but we also had the primary endpoint, which accrued even more events.

Rock-Steady LDL-C Lowering

Dr Mandrola: Tell us about how it came out.

Dr Sabatine: It came out great. The first thing is that evolocumab reduced LDL-C by 59%. Patients started with levels of 92 mg/dL (2.4 mmol/L) and evolocumab got it down to 30 mg/dL (0.78 mmol/L). The amazing part is that if you look at the interquartile range, a quarter of the patients achieved an LDL-C < 20 mg/dL (0.52 mmol/L). It is hard to imagine in our lifetime for a big trial that you have a quarter of the patients with LDLs under 20 mg/dL.

Most interesting was that the benefit was consistent across the quartiles of baseline LDL-C.

The LDL-C reduction was great, but it was also rock-steady over the duration of the trial, and this is important given other drugs in this class. There was no attenuation of effect. We have a similar paper out to 4 years, based on OSLER-1, that just came out in JAMA Cardiology online.[6]

For clinical events, evolocumab reduced the risk for the primary endpoint by 15%. It reduced the risk for the key secondary endpoint even more, a 20% risk reduction. We are very happy with that. There was a 2% absolute difference at the end of 3 years. The results were entirely consistent in all of the subgroups that we examined. In addition to the demographic subgroups, there was consistency regardless of the type of vascular disease, whether you were on high-intensity statin therapy or moderate-intensity statin therapy. I think most interesting was that the benefit was consistent across the quartiles of baseline LDL-C.

Dr Mandrola: Whether you started with a high or low LDL-C?

Dr Sabatine: Yes. It was equally effective, even in the lowest quartile—those are people who started the trial with an LDL-C of 74 mg/dL (1.91 mmol/L). If you saw a patient with that LDL-C, most physicians would say, "Fantastic. You are all set; leave it alone." Evolocumab got that LDL-C down to 22 mg/dL (0.57 mmol/L) and was associated with a significant benefit. In fact, if you take the event rates for CV death, MI, and stroke and you plot it based on achieved LDL-C, and you stratify patients into eight groups based on quartile of baseline LDL and randomized treatment arm, those eight dots form a very nice line all the way down to the low 20s mg/dL.

Dr Mandrola: With these low LDL-C's, there is going to be a concern about safety.

Dr Sabatine: You are right; people have fretted about that. It is hard to know what to make of the observational studies. We did not see anything in FOURIER. The overall rates of adverse events were similar in both treatment groups. There was no difference in allergic reactions. Injection-site reactions were slightly more common with evolocumab (0.5% over 2 years; 90% of those were classified by investigators as mild). For the side effects that we worry about with statins or low LDL-C—myalgias, cataracts, new-onset diabetes, neurocognitive issues—there were similar rates in the two arms.

Importantly, evolocumab is a fully human monoclonal antibody. There are no other animal species residues in the agent. We did not see any neutralizing antibodies, which is different from what was presented for bococizumab.[7,8]

Possibly Even Better With Time

Dr Mandrola: What is your general take? It sounds pretty positive.

Dr Sabatine: We are very happy. Another point I'd like to mention is that just as we saw in the statin trials, the benefit takes time to accrue. If you lower someone's LDL-C today, you do not see the benefit tomorrow, but your arteries get healthier over time. That 20% reduction [in the secondary endpoint] was averaged over the trial. If you look beyond the first year, there was a 25% reduction. In fact, if you look specifically at fatal or nonfatal MI and stroke, over the long term there was a 33% reduction. We were really happy about that. We find the data quite compelling.

Dr Mandrola: The Kaplan-Meier curve continues to separate over time. It seems like the benefit was accruing, which is what you would expect from the Mendelian randomization studies, on longer exposure to lower LDL-C levels.[9]

Dr Sabatine: Yes. I think there are a couple of points in there. It certainly is true in the statin trials that when we make an intervention, the curves typically separate after 3, 6, or 9 months—somewhere around there. If you think about 4S,[10] those curves did not diverge for about a year and a half. We see the same thing in FOURIER. The curves start to diverge sometime in the first year, but then they really start separating after the first year. That makes good biologic sense. It also fits with the point you raised that even relatively small differences in LDL-C that are genetically mediated turn out to have huge benefits long-term—if that affects someone over the course of their life rather than just over 1 year, 2 years, or 3 years like we do in a trial.

Dr Mandrola: Let's talk about the duration of the trial. It was positive at 2 years, but what are your plans for longer-term follow-up? There was a state-of-the-art paper in JACC about how many cardiovascular therapies are only studied for 2, 3, 4, or 5 years.[11] Patients might be taking these drugs for longer.

Dr Sabatine: Yes—for most therapies, one gets a pretty good sense after several years of how it is going to play out. For the lipid lowering, there are a couple of points. One is that there is greater benefit over time—at least for the statin trials after the first year, and then you are in a steady state for the relative risk reduction for each year. I think we have a good handle on that. Nonetheless, there is a plan for an open-label extension involving over 6000 patients from FOURIER. That will be very useful to get additional safety data, observational data. We are hoping to get additional follow-up in other subjects who are not in the open-label extension.

You will recall from the statin trials that just as there is a bit of a lag before the full clinical benefit kicks in, there is also a bit of a legacy effect even after patients come off therapy. The fact that their arteries were marinating in lower levels of LDL-C continues to give them benefit in the years afterwards. We expect to see that effect. We are hoping to gather top-line cardiovascular data. That is a work in progress.

Guidelines and Costs

Dr Mandrola: Two more questions. One is that the 2013 AHA/ACC guidelines for lipid management[12] changed our framing; we were told to stop titrating to LDL-C levels and to use statins based on how they were used in the clinical trials. Does this change that in your mind?

Dr Sabatine: I think it does. Many people grappled with the 2013 guidelines. There were many good features for reminding physicians that they should be prescribing high-intensity statin therapy whenever tolerated. I think that was a big plus.

I understand the limitations that the writers faced, given how the questions were framed in their methodologic approach, but it was a fairly radical departure from the prior guidelines, which took a more holistic biological view that LDL-C is a risk factor. We want to get it down. Based on the data from the trials, we had accrued evidence for when we would see benefit. The most recent update before the 2013 guidelines was in 2004,[13] and it noted—based on PROVE IT-TIMI 22[14]—that you could now target not just 100 mg/dL (2.59 mmol/L) but 70 mg/dL (1.81 mmol/L), because PROVE IT showed that 70 mg/dL was better than 100 mg/dL.

Maybe it is only a $14,000 question, not a $64,000 question.

Since the guidelines have come out, we have had IMPROVE IT,[15] showing that mid-50s mg/dL is better than 70 mg/dL, and now we have data from FOURIER that even down to the 20s mg/dL is beneficial. I hope this will redirect us, now armed with new data, to say that LDL-C is a risk factor. So far we have not found a threshold below which does not accrue benefit. We were worried that the relationship could have been a hockey stick [–shaped curve] and level off. We have not seen that.

I hope the guidelines will take an approach to reduce levels of LDL-C and then ask, what is the right tool for the patient? Which drug should we use?

Dr Mandrola: Final question, the $64,000 question: How is this going to be translated? Right now, if you look at cost, it is a big issue. What are your thoughts?

Dr Sabatine: Maybe it is only a $14,000 question, not a $64,000 question.

Dr Mandrola: Good point.

Dr Sabatine: Cost is a fair point. There are a lot of expensive drugs out there. There needs to be a careful discussion among drug manufacturers, insurers, and pharmacy benefit managers. Those discussions have been taking place and will continue to take place. I think they will be informed by rigorous cost-effective analysis, and now that we have the data, those are under way.

My perspective as a cardiologist, though, is that my primary duty is for the patient in front of me. If I can significantly and safely lower his or her risk for MI or stroke, that is what I want to do. I hope that these data and an update in the guidelines will make it easier for physicians to prescribe the drug, get their patients' LDL-C down, and reduce their risk for events.

Dr Mandrola: Great. Thank you for joining us. I really appreciate it.

Dr Sabatine: Thank you for having me.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.