The Top Cardiology Trials of 2018 in Review

; C. Michael Gibson, MD


December 18, 2018

Editor's note: Due to technical difficulties, the audio file is not available for this episode of The Bob Harrington Show. Below is an edited transcript of the discussion. We apologize for any inconvenience.

Robert A. Harrington, MD: Hi. This is Bob Harrington from Stanford University, on and Medscape Cardiology. We have an annual tradition on this podcast to have an end-of-year wrap-up discussion with my good friend and colleague, Mike Gibson. We will talk a bit about what has gone on in cardiology in 2018 and discuss trials and observational studies. We may try to forecast where things might be heading in 2019.

Again, my guest is Mike Gibson, from Beth Israel Deaconess in Boston, where he is an interventional cardiologist and a professor of medicine at the Harvard Medical School. He is also the CEO of the not-for-profit academic organization Baim Institute for Clinical Research.

Mike, thank you for joining us on Medscape Cardiology.

C. Michael Gibson, MD: Thank you for having me on again this year, Bob.

Harrington: Mike, you and I have had a lot of fun over the past few years reflecting about what we have observed in cardiology. This year, I'd like to lead us through a discussion of lipids, aspirin for primary prevention, atrial fibrillation (AF), structural heart disease, hypertension, and cardiomyopathies. Are you ready for all of that?

Gibson: I'm ready.

Harrington: If it's a trial that either one of us has been specifically involved in, we will make sure to let the readers know.

Developments in Lipid Management

Harrington: Some big things happened with lipids this year. ODYSSEY (I was a member of the executive committee) and REDUCE-IT came out. New guidelines on lowering elevated cholesterol were also released. Mike, would you start with ODYSSEY?

Gibson: ODYSSEY[1] was a great study. It differs from the preceding FOURIER study[2] in important ways. In ODYSSEY, patients had a prior acute coronary syndrome (ACS) event within 1 month to 1 year before enrollment and there was uptitration of the dose, whereas a fixed dose of a proprotein convertase subtilisin-kexin type 9 (PCSK9) inhibitor was used in FOURIER. If one didn't achieve a low-density lipoprotein cholesterol (LDL-C) below 50 mg/dL, then the dose was doubled. This is important because epidemiologists have pushed back on the concept of treating to goal, saying that no trial had been done that really uptitrated the agent to reach a target LDL-C. Here we have somewhat of a test to that hypothesis. Finally, the duration of FOURIER was only 2.2 years. ODYSSEY was longer, out to 3 years, where you had a little more time for some of the benefits to emerge.

ODYSSEY enrolled almost 19,000 patients who had had an ACS, who were on maximal-intensity statin therapy or the maximally tolerated dose. Patients were randomized to biweekly injections of the PCSK9 inhibitor alirocumab or placebo. After 2.8 years, alirocumab significantly reduced cardiovascular (CV) death, myocardial infarction (MI), stroke, or hospitalization, down from 11.1% to 9.5%. To prevent one major adverse cardiac event (MACE) event, you would have to treat 49 patients for about 4 years.

Importantly, mortality was reduced from 4.1% down to 3.5%, but it is an exploratory endpoint (nominal P value = .026). This is different from FOURIER where there was no mortality advantage. Again, I want to emphasize that this is exploratory.

The benefits were even greater in those people who had a baseline LDL-C at or above 100 mg/dL, where there was a 3.4% absolute risk reduction. In that subgroup, you would only have to treat about 30 people to prevent a MACE event and there was a 1.7% reduction in mortality.

We now have agents with three separate mechanisms of LDL-C-lowering that are all associated with an improvement in CV outcomes: statins, ezetimibe, and PCSK9 inhibitors.

I think you need to be at least below 50 mg/dL [LDL-C] if you've had an MI.

Harrington: That is an incredibly helpful summary to validate the PCSK9 story. I'm completely with you that now we have three distinct classes of drugs that, interestingly, show a pretty linear relationship between dropping LDL-C and reducing events. In the very early statin trials we lowered LDL-C to around 200 mg/dL; now we're all the way down into the 30 to 40 mg/dL range. It is pretty linear and predictable, at least with these three classes of drugs.

Mike, what are you doing for your patients with MI? How low are you going?

Gibson: I think you should go as low as you can go where they can tolerate the drugs. That is my goal.

Harrington: Yes, I'm starting with statins and pushing them down. If they are right at 70 mg/dL, now that we have both of the PCSK9 inhibitor trials and IMPROVE-IT[3] with ezetimibe, I am pushing it lower. I think you need to be at least below 50 mg/dL if you've had an MI. What do you think?

Gibson: Absolutely.


Harrington: Along the lines of lipids, we also have now REDUCE-IT,[4] which got a lot of buzz at the American Heart Association (AHA) meeting. It enters into the controversy of what to do with patients with elevated triglycerides. Certainly, the whole issue of fish oils to lower triglycerides has been up and down in CV medicine because the trials have largely been disappointing.

Could you provide an overview of REDUCE-IT?

Gibson: It is confusing. There were two simultaneous but contradictory trials: REDUCE-IT and VITAL. Both of them were released into the lay press at the same time and created a lot of confusion.

The VITAL study[5] was very different from REDUCE-IT. It was a large primary-prevention study that studied two forms of fish oil, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). It was negative.

That should not be confused with the results of REDUCE-IT,[4] which was a study of pure EPA fish oil (Vascepa®, made by Amarin Pharma) given at a high dose of 4 g/day for secondary prevention. I was chairman of the clinical events committee for this study, by way of disclosure.

REDUCE-IT studied 8000 patients who had established CV disease or diabetes. They were on statins and had triglycerides that were elevated, between 135 and 499 mg/dL, and LDL-C levels between 41 and 100 mg/dL.

Over the 4.9 years of follow-up, Vascepa reduced the primary endpoint of CV death, MI, stroke, or coronary events that required hospitalization, from 22% down to 17%, which was highly significant. My recollection is that the P value had about eight zeros in it.

If you look at the secondary endpoint of CV death, MI, or stroke, it was reduced from 14.8% down to 11.2%, which again was highly significant. The prespecified endpoint of CV death was reduced from 5.2% down to 4.3% (P = .03).

There were some side effects, though. In the treatment group, there was more AF that required hospitalization and a trend towards more bleeding, which went up from 2.1% to 2.7% (P = .06). Importantly, gastrointestinal disturbances were present in about 30% of patients and diarrhea in about 10%. Keep in mind that the control arm received mineral oil as the matching placebo to the fish oil. This might account for the high rate of diarrhea seen in both the fish oil and mineral oil arms. This turns out to be very important. People have expressed concern that the mineral oil placebo was not inert and may have actually diminished the absorption of the statin. In keeping with this, there was a 10% rise in LDL-C to 84 mg/dL in the placebo group. That is 6% more of a rise than was seen over in the Vascepa arm. There were also rises in apolipoprotein B and C-reactive protein (CRP). CRP rose over 30% in the control group. What does this mean? Physicians have speculated that the 6% rise in LDL would reduce the relative risk reduction down from the 25% that was observed to be something closer to 20% to 21%, which would still be statistically significant.

The other concern is that no one really knows the mechanism of benefit. The triglycerides were lowered 22% to 170 mg/dL, but that likely does not account for the magnitude of the benefit. Fish oil is known to have some antiplatelet effects, which might explain some of the benefit.

Dr Alexander Leaf was the chairman of medicine at Mass General. I was familiar with his work, where he showed that fish oil stabilized cardiac membranes and reduced the inducibility of arrhythmias in dogs.[6] I think it's notable that the risk for sudden death was reduced by about 31% in the Vascepa group, which is at least provocative.

A lot of people have wondered, is it just a fluke? There was a Japanese study called JELIS,[7] where the events were reduced 19% with Vascepa. It was a 5-year study with pretty impressive results. What are your thoughts, Bob?

Harrington: Let me first give my clinical overview.

Like you, I think that the trial was a winner. It showed a reduction in important clinical events with the therapy. It's been difficult to show whether it might matter clinically. This is highly purified EPA, so it's different from some of the agents studied in other trials. That is worth noting.

The sudden cardiac death piece is interesting. There was an observed small but real reduction in sudden cardiac death. This whole issue of stabilizing cellular membranes is certainly provocative.

I'm not bothered by the lack of clear-cut mechanism. Mechanism is nice when you have it, but I don't think you need it if you have a well-done trial. Oftentimes, drugs work in a myriad of ways and we are not quite sure what the specific mechanism of benefit is. That does not disturb me. Teasing out mechanisms opens up whole areas of interest.

However, the mineral oil issue bothers me. You see the elevated inflammatory markers and see LDL-C going up. I agree with you: It does not explain the magnitude of the benefit, but it probably does diminish it some.

Overall, it's a winner. While maybe not a grand slam, Mike, could we settle on it being a home run?

Gibson: Sure. Absolutely.

New Cholesterol Guidelines

Harrington: Let's finalize the lipid discussion with a talk of the new cholesterol guidelines[8] that came out at AHA just a few weeks ago. I think the guideline writers were waiting to see if the ODYSSEY results were going to be published. They were, and those data were included in the guidelines. What is your takeaway from the new cholesterol guidelines?

Gibson: They seem to pivot back a little bit to "lower is better" and reintroducing a target, at least in secondary prevention. For secondary prevention, it's now back to being a class I recommendation to get your LDL-C down by over 50% using statins.

For those people who have multiple events who have an LDL-C over 70 mg/dL, [the guidelines] indicated that you should add ezetimibe. If the LDL-C remains above 70 mg/dL, they gave a class II recommendation that a PCSK9 inhibitor should be added to those patients.

What I thought was interesting is that they gave consideration to cost in shaping these statements. Some questions come to mind: Should we be going back to target? I think we should. Is an LDL-C of 70 mg/dL the right target? Should a PCSK9 be used before ezetimibe? Should price be a consideration? Keep in mind that these prices have now changed. What are some of your thoughts?

Harrington: First off, I'll say that the whole concept of risk assessment using scores like the atherosclerotic cardiovascular disease (ASCVD) risk assessment tool becomes very important. Since the last iteration of the guidelines, we've all gotten very comfortable. Many of our electronic health records have the ability to calculate ASCVD right into it. To me, that's a positive. Let's assess risk. Let's have a conversation with our patients using those data.

Second, let's be aggressive with secondary prevention. I like Dr Braunwald's comment, that you can't be too thin, have too low an LDL-C, or be too rich. He likes to put them all there together. I think that there is some wise advice in that.

Finally, you are spot-on about the cost equation. More and more, you are going to see the guidelines consider cost, such as in the case of lipid lowering when you have formal cost-effectiveness analyses. I think cost-effectiveness is here to stay and aggressive LDL-C lowering is here to stay. Risk assessment will continue to be refined as more work is done.

Gibson: What do you think of the introduction of using calcium scoring? If you have a calcium score of zero and you are in this intermediate group of risk, they say that could put you over the edge of saying don't give someone a statin. What do you think?

Harrington: I've been using calcium score the other way, which is that when I have an intermediate-risk score in a patient, particularly a patient in whom I'm trying to encourage lifestyle changes, I've been getting calcium scores and using it in the positive direction, meaning that if it was positive, to be more aggressive with treatment.

When it's negative, I think we have to be careful that, although the risk for an event is quite low in those people, it's not forever. Some people have said, "I can do whatever I want because my calcium score is zero." I don't think that's the case at all. It might be at that point in time, yes, but it doesn't obviate the need for lifestyle changes.

Gibson: You should check again in 5 years and be careful. Someone could be a smoker and have a family history even though they have a calcium score of zero.

Aspirin in Primary Prevention

Harrington: I agree with that 100%. Let's turn to primary prevention. We have almost 50,000 patients in aspirin primary prevention trials—ASCEND,[9]ARRIVE,[10] and ASPREE [11,12,13] which are remarkably consistent in terms of the message. What is the take-home message for you?

I don't think this was a good year for aspirin in primary prevention.

Gibson: ASPREE[11,12,13] had over 19,000 patients > 70 years (> 65 years for black or Hispanic patients). After 4.7 years of low-dose aspirin, there was no difference in death, dementia, or physical disability. There was nothing there in terms of cardiac events. Actually, mortality was a little higher in the aspirin group: 12.7 versus 11.1 events per 1000 person-years. Cancer-related deaths were higher in the aspirin group. Aspirin has always increased bleeding.

That has certainly rained on the parade for aspirin in primary prevention. On the other hand, the ASCEND trial[9] looked at people with diabetes who had no evidence of heart disease and found that 7 years of aspirin reduced MI, stroke, and transient ischemic attack, but it was associated with more bleeding.

I don't think this was a good year for aspirin in primary prevention.

Harrington: Certainly not for routine primary prevention. I think the jury is still out on the intermediate-risk patient, maybe the patient with diabetes; but overall, as a routine strategy, I agree with you. Aspirin as a primary preventive strategy in low-risk individuals is certainly not indicated for most or for routine use.

Percutaneous Mitral Valve Repair

Harrington: I'm going to switch away from drugs to technologies and devices. There were two really interesting trials dealing with percutaneous treatment of mitral regurgitation: MITRA-FR[14] and COAPT.[15] Do you want to give your impression as to where we are in percutaneous mitral valve repair?

Gibson: Yes. The COAPT study[15] was big news at Transcatheter Cardiovascular Therapeutics (TCT). For a decade now, we have used the MitraClip device in Europe and the United States to treat primary mitral regurgitation (MR). This study was about secondary MR in those people with bad left ventricular function, where there is no guideline recommendation.

There was a preceding trial called MITRA-FR,[14] presented in Europe, that showed no reduction in rehospitalization or death at a year with the MitraClip. One month later, there were the results of the COAPT study, presented at TCT. The MitraClip in this study lowered the rate of heart failure hospitalizations tremendously, down from 67.9% to 35.8%, and, importantly, reduced all-cause mortality down from 46.1% to 29.1%. You would only need to treat six people to prevent one death. Quality of life and left ventricular and diastolic volumes were better in the MitraClip group.

There has been a lot of debate and speculation about why these two trials yielded such different results. A lot of it may have been the patients that were studied. The COAPT patients may have hit the sweet spot. They had moderate to severe MR, but not such bad MR where their left ventricle was dilated, which is different from what was in the MITRA-FR study.

In COAPT, patients were on maximally tolerated medications, whereas in MITRA-FR, patients got their drugs titrated up. It was a very select group of people. It took 8 years to enroll 610 patients. It's estimated that only about 10% of our patients would really be COAPT patients.

Finally, it's been speculated that the operators were quite good in COAPT. There was only residual MR in 5% of the COAPT patients, whereas it was present in about 17% of the MITRA-FR patients.

The big question is whether for the reductions in MR there would be even better outcomes, and whether replacement devices that totally get rid of the MR will actually be better than the repair devices. Very interesting stuff.

This was a good year for mitral valve percutaneous repair.

Harrington: This was a good year for mitral valve percutaneous repair. We have been doing them with great evidence that it improved clinical outcomes, but certainly nowhere near the clinical outcome data that we saw associated with the percutaneous replacement of the aortic valve for aortic stenosis.

You described the two trials well. They get at the efficacy-versus-effectiveness question. If you go after efficacy and very carefully screen your patients and investigators, and very carefully monitor the technology usage and the outcome of the procedure, you can then look at efficacy. COAPT shows that you can get good efficacy in a select group of patients.

When you start getting more broad usage with MitraClip, maybe the effectiveness of the therapy as a routine treatment is not quite so clear. I think this says that it clearly works from an efficacy perspective. Now we need to further refine technique, including with newer technology.

I took this as a win for structural heart disease, the mitral valve, but we clearly have more work to do.

Gibson: Everyone is holding their breath to see what happens in the 420-patient Reshape-HF2 trial, which will be released soon, to see if it confirms the results.

The other lesson is on the pathophysiology. I think everyone thought that MR was just an innocent bystander to heart failure. What this shows is that if you reduce the MR, you favorably improve LV remodeling. It's more than an innocent bystander.

Catheter Ablation for AF

Harrington: That is a great description. Another device trial, CABANA, was long in coming. I was at the Duke Clinical Research Institute when the trial was first awarded from the National Heart, Lung, and Blood Institute. It took a long time to get the trial done, looking at catheter ablation in an elderly group of patients versus medical treatment in the initial therapy of AF. How do you interpret the CABANA results?

Gibson: CABANA was a 2200-patient trial with 5 years of follow-up. The rate of death, disabling stroke, serious bleeding, or cardiac arrest was similar in the ablation group and drug therapy groups, at 8% and 9.2%. There was a lot of mining of the data. They looked for some subgroups with benefit. The people below 65 years of age had a hazard ratio (HR) of 0.52 with ablation, whereas those over 75 had an HR of 1.46. There was a borderline interaction with age (P = .074). Some of the effect might be modulated by age.

Ablation did lower the secondary endpoint of CV hospitalization or death from 58.1% down to 51.7%. One thing that was controversial was the analysis of the data on an actual treatment-received basis, which showed a 33% reduction in the endpoint down from 10.9% to 7%. However, there were very high rates of crossover in the study, with 9.2% of patients randomized to ablation [crossing over to drug] and 27.5% of patients randomized to drug therapy crossing over to ablation. Given that kind of crossover, it's hard to interpret the as-treated analysis. You have to be careful.

Harrington: You brought up one of the key concerns when you do device trials. It is very different from drugs where virtually everybody gets their assigned therapy. In this case, you have a group of patients who do not get the therapy they were assigned to and a group of patients who get the therapy when they were assigned to the control . One has to take some caution with those analyses, and also learn something from the analyses as to why patients crossed over and how they did when the technology was applied to them.

CABANA tells me that you do not have to ablate everyone as an initial therapeutic strategy in AF, but if you have persistent AF that particularly is bothersome from a symptomatic perspective, it is quite reasonable to go ahead and ablate it. This trial gives us some comfort in that strategy. Do you agree with that, Mike?

Gibson: Yes, I agree.

Miscellaneous Trials That 'Grabbed Me'

Harrington: I'm going to list three trials that also grabbed me this year and then let you comment.

First, there was a terrific trial out of the Los Angeles region looking at the barbershop as the setting to treat hypertension in a group of African American men.[16] Men were randomized essentially to receive advice in their barber shop versus receive pharmacist-directed management of high blood pressure in their barber shop. It showed that there is an advantage to both using the barber shop as a setting for the treatment of a common chronic disease, and also that the pharmacist-therapy strategy could be well deployed in that setting. To me, that was a big win for practical trials and practical questions.

The next trial that grabbed me was mSTOPS.[17] This interesting trial, run out of the Scripps clinic, demonstrated that there is a fair bit of background AF that can be detected with patch ECG in an asymptomatic population. To me, the bigger question is: What do you do with that? But I thought mSTOPS was a great example of how you can deploy technology on a large scale and understand how it might play out in the wild, if you will.

Finally, a trial called ATTR-ACT[18] looked at a new therapy, tafamidis, to treat a disease which is becoming increasingly commonly diagnosed, amyloid cardiomyopathy. The drug demonstrated potentially transformative results by interfering with amyloid deposition in patients with amyloid cardiomyopathy and improving cardiac outcomes, including mortality.

These are three very different but interesting trials that I think move the ball down the field. Closing out some of those three, Mike?

Gibson: ATTR-ACT and its approach are revolutionary and could be transformative. A lot of patients have amyloid, and I think this is going to change things.

Harrington: Mike, as always, it's been a fun end-of-year go-through. Thank you for joining us on Medscape Cardiology.

Gibson: Thanks for having me, Bob.


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.
Post as: