The Top Cardiology Trials of 2017 in Review


; C. Michael Gibson, MD


December 21, 2017

Robert A. Harrington, MD: Hi. This is Bob Harrington from Stanford University, here with Medscape Cardiology and Over the past 3 or 4 years, Mike Gibson and I have spent some time at the end of the year reflecting on some of the big news stories that hit the airwaves in cardiology over the course of the year.

This year, we have an incredible list of clinical trials that have informed practice (and maybe even changed practice), or have at least forced us in new directions in clinical cardiology.

I am pleased that my good friend Mike Gibson has agreed to be with us today to have this conversation. Mike is a professor of medicine at Harvard Medical School. He is an interventional cardiologist at Beth Israel Deaconess Medical Center in Boston and the newly named president and CEO of the Baim Institute for Clinical Research. Mike, thanks for joining us here on Medscape Cardiology.

C. Michael Gibson, MD: Thanks for having me again, Bob.

Dr Harrington: Mike, it's always fun to have this conversation with you. The clinical trials I want to go through are FOURIER, CANTOS, COMPASS, CULPRIT-SHOCK, and ORBITA. Does that sound like a pretty good list?

Dr Gibson: It's a great line-up.

FOURIER: PCSK9 Inhibitors

Dr Harrington: Let's start with FOURIER.[1] Give us a sentence or two on what FOURIER was, and then we will get into some of the issues around the data and how the community is responding.

Dr Gibson: The trial was highly anticipated. There have been dramatic reductions in low-density lipoprotein (LDL) cholesterol with the proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors. This trial studied one of them : evolocumab. People had anticipated about a 30% event rate reduction, but there was about a 15% risk reduction in cardiovascular death, myocardial infarction (MI), stroke, and hospitalization for unstable angina. That 1.5% or 15% relative risk reduction kind of underperformed relative to what people had anticipated. There was no mortality reduction.

The US Food and Drug Administration (FDA) label just came out the other day, and I thought it was very interesting that mortality is not in the label —just MI, stroke, and hospitalization.

Probably the big punch line was the cost issue. Here is a drug that costs about $14,000 a year. You would have to treat 70 people to prevent one nonfatal event. That would be about $1 million per year, $2 million per 2 years, to prevent that event. People have questioned the cost- effectiveness of this.

Mendelian randomization is how God does randomized trials : Some people have the genetic variant, and some do not.

Dr Harrington: FOURIER is an example of real scientific achievement. Here was an observation made about the genetics of low LDL cholesterol that was then quickly, from a discovery perspective, translated into human observation, then a therapy, to clinical development, and finally a clinical trial. As I understand from the group at University of Texas Southwestern, all of this happened pretty quickly. That a genetically defined medicine came through a clinical trial successfully in a pretty short period is maybe one of the big pieces of news from FOURIER. Do you agree, or do you think something else about the whole concept of PCSK9 inhibition?

Dr Gibson: No, I agree. Mendelian randomization is how God does randomized trials : Some people have the genetic variant, and some do not, and we look at what happens. We do not look at what happens over a year or two; we look at what happens over the course of the patient for a lifetime. It's an amazing way to guide drug development.

One of the big issues here was the duration of treatment and exposure. The trial has been criticized because the duration of exposure was somewhat short. People have wondered if the trial had gone on longer, whether bigger risk reductions would have been seen. The upcoming trial, ODYSSEY, that is an 18,000-patient trial of patients with acute coronary syndrome (ACS) , does have a slightly longer duration of exposure and will be presented in 2018. There are more data to come in this area.

Dr Harrington: Yes, I agree with those comments. FOURIER was a scientific achievement ; in some ways it was a predictable achievement, because we had so many things lining up that would portend or give us a sense that the trial was going to be positive.

I was disappointed with the length of follow-up. I thought, if you look at the length of follow-up in the original statin trials that demonstrated a mortality finding, there was much longer length of follow-up. I also know that in these clinical trials, in some ways, it's a race to get to the results. In retrospect, it seems pretty clear that they should have gone longer, unfortunately, the investigators did not.

As you said, there will be other trials coming. In the spirit of disclosure, I'm involved with the ODYSSEY outcomes trial in the ACS setting, and there will be longer follow-up.

The cost issue is not trivial. A lot was written over the course of the past year about cost- effectiveness. This is a great example. It has almost been a tutorial in the literature about cost- effectiveness instead of just cost. There is no question that whether we look at absolute cost or cost effectiveness, it just does not meet the bar at the current state, does it?

Dr Gibson: Not at this point, so people have to roll up their sleeves, sharpen their pencils, and see where we end up.

Dr Harrington: You and I have been involved with the Medicines Company around this notion of small interfering RNA (siRNA) used to interfere with PCSK9 and inhibit it. It is exciting that we might be able to deliver a therapy to lower LDL cholesterol only once or twice a year. Certainly, I'm waiting for the clinical trials on that to start. How about you?

Dr Gibson: Yes, that is a great concept. Moving from twice a month or once a month to every 6 months is a major advance.

The other major advance is that the cost of goods perhaps will be much lower with this kind of agent. Many people are speculating —and I want to use that word, "speculating "—that the cost of this agent might be closer to $4000 a year and could be much more cost- effective. The price point of $3000 -$4000 a year is more in line with that of some of the oral agents at present.

Dr Harrington: When you start getting into that sort of overall cost for the product and you look at the benefit with the therapies thus far, that would start to move us into a level of cost- effectiveness that, as a society, we have deemed acceptable.

CANTOS and the Inflammatory Hypothesis

Dr Harrington: We then see another clinical trial, CANTOS,[2] which explored a new avenue of biology in inflammation as a way to attack the atherosclerotic process. Would you like to talk about what CANTOS was designed to do and give your quick observations on it?

Dr Gibson: It was very exciting. Paul Ridker has championed this notion that inflammation plays a role in atherosclerosis for a decade or two now. I have been proud to watch him continue that campaign.

Canakinumab is a human monoclonal antibody that has been used for some orphan diseases, and it lowers interleukin (IL)-6 and lowers C-reactive protein (CRP) dramatically. Dr Ridker randomly assigned about 10,000 patients with prior MI who had a persistently elevated CRP level—greater than 2 mg/dL despite high-intensity statins—to receive different regimens of this drug versus placebo. He found a reduction in nonfatal MI, stroke, or cardiovascular death by about 15%. This was with no change in LDL cholesterol and a 30%-40% reduction in IL-6 and CRP.

He has said that this event rate is similar in magnitude to the PCSK9 inhibitors. Peter Libby commented and said that this does validate the inflammatory hypothesis.[3] People debated that, but I think it is clear that we now have a whole new range of inflammatory targets for future drug development. Most of it driven by reduction in MI. It is very costly, as you pointed out, and costs $200,000 per year. One strategy might be to give a single dose of the drug and see whether the patient responds, and then continue therapy.

Fascinating observations from this study were an interesting increase in death from sepsis, although low, and a reduction in cancer mortality, particularly in lung cancer. Again, it's been debated as to whether that is real or not. So, Paul is following on with the 7000-patient trial of methotrexate, a much less expensive anti-inflammatory agent. This science is very exciting, and we may have some new targets.

CANTOS is a scientific achievement.

Dr Harrington: You hit the key points that I wanted to bring up. Like with the PCSK9 inhibitors, I think that CANTOS is a scientific achievement. I'm with Peter, in that I think it moves forward our understanding of the inflammatory hypothesis as a mechanism in the pathogenesis and the progression of atherosclerosis. Kudos to Paul and Peter and others who have persevered in this arena for many years now in trying to understand the role of inflammation in atherosclerosis. Now, a therapy that suppresses inflammation through this mechanism results in improved clinical benefit. This is a tremendous scientific achievement.

A lot of interesting things came out of the trial. The infection issue has to be understood better. The positive effect on lung cancer has to be understood better. Although some would say the effect was not unexpected, it was at least understood that it was a possibility based on the biology. But the magnitude of the effect is so profound that it needs to be followed up in other studies.

The responder analysis that you indicated (ie, give someone a shot of the drug and see whether they respond) is an attractive hypothesis. Whether or not that is a way forward is worth some more reflection and some more thought.

Finally, just as you said, are the cost issues. This drug is already approved for rare diseases, so its cost has been established like it's being used to treat rare diseases. What do you do now that it is being used for a common disease? In that case, the cost is not an acceptable by current measures of cost- effectiveness. I would say there is a lot more to come in CANTOS, but wow, what a positive step forward for the field of biology in atherosclerosis.


Dr Harrington: Let's go to an area that you championed (probably more than anyone I know), and that is the role of rivaroxaban, one of the new or direct oral anticoagulants (NOACs/DOACs), with or without antiplatelet therapy in the setting of stable cardiovascular disease (CVD). Do you want to talk a little bit about the COMPASS trial?[4] A lot of your work with the ATLAS ACS2 trial set the stage for what went on in COMPASS.

Dr Gibson: We often think of ACS and its treatment and prevention as an antiplatelet problem. But one of the learnings is that platelets calm down after ACS. We are seeing in multiple studies that thrombin generation is a chronic problem and persists over time. It's a viable biological target, as you alluded to. Almost 10 years ago , we studied 3500 patients[5] to look at the right dose of rivaroxaban in ACS, and it turned out that the very lowest dose, 2.5 mg twice a day, was quite effective.

In a phase 3 study of ACS patients,[6] we saw a mortality reduction of over 30%. So, it was very interesting to see a follow-on study[4] of chronic coronary artery disease (CAD) where they randomly assigned over 27,000 patients to that very low dose of rivaroxaban, 2.5 mg twice a day, with aspirin versus aspirin alone. Again, the addition of the antithrombin reduced the risk for cardiovascular death, MI, and stroke from 5.4% down to 4.1% at 2 years. On the other hand, bleeding was increased by 1.2% over those 2 years, with a 0.6% per year increase in major bleeding, similar to what we saw in ATLAS ACS. To put that in perspective, both prasugrel and ticagrelor both also had a 0.6% increase in major bleeding per year. Any time you have effectiveness, you are going to pay a price in bleeding.

The big punch line was that all-cause death was reduced significantly, which kind of confirmed what we had seen in the ATLAS trial. Also, major adverse limb events, amputation, and those types of events were reduced. Dr Braunwald stated[7] that he felt these results should change guidelines, and we will have to see. There was an announcement in the media the other day that this drug at this dose is now being evaluated by the FDA for the COMPASS indication.

Dr Harrington: It has been an interesting story. When I reflect on it, almost unexpectedly it does go back over a decade, as you have pointed out. There has always been this notion that the combination of an anticoagulant and an antiplatelet therapy, or a thrombolytic and an antiplatelet therapy, does offer an advantage over either one alone. So, this is not a new concept biologically, but these data do sort of cement the notion that the combination of an anticoagulant and antiplatelet agent offers additional benefit. Without there being real symmetry in the arms, I am disappointed that the rivaroxaban alone arm did not offer enough from a clinical perspective. Did that surprise you?

I wish life were easy ; I wish we could give one pill, but maybe the right thing is to create polypills.

Dr Gibson: The platelet pathway still is important. It made me think a lot of the hypertension studies[8] where a combination of a fraction of the dose of three drugs was better than the full dose of any drug. We have multiple axes here, biologically. I think we are seeing that inhibiting multiple axes at a low dose where you have efficacy, but you are not on that part of the dose/response curve where you are having a lot of safety problems, might make a lot of sense biologically. I wish life were easy ; I wish we could give one pill, but maybe the right thing is to create polypills that combine these very low doses of drugs to achieve inhibition of multiple targets.

Dr Harrington: That is well said. In an era where we have an increasing focus on personalized medicine, might there be periods in the atherosclerosis syndromes, if you will, where people are more platelet- responsive or people are more coagulation- responsive? And, might you be able to, in a more refined way, predict which group of patients at which period might respond to the anticoagulant versus the antiplatelet? For now, I'll accept that the combination of the two of them at low doses looks to be a pretty good strategy. Is that fair?

Dr Gibson: Yes, that is fair. We have also learned from ticagrelor[9] that the 60- mg dose might be a very viable dose later on. So, the dose of the drugs may vary over time depending upon the activation of the platelets and the coagulation cascade.

Dr Harrington: To reference the ticagrelor story is a reasonable one. It is very reasonable that at some point, you down-titrate the ticagrelor dose from the 90 mg to the 60 mg, or you move from a PLATO strategy to a PEGASUS strategy. That appears to have some rationality attached to it.

CULPRIT-SHOCK Practice Changing

Dr Harrington: Let's move now to the cath lab, where both you and I have spent a lot of time over the years. You continue to spend a lot of time there. Would you like to describe CULPRIT-SHOCK[10]?

Dr Gibson: This was the largest trial of cardiogenic shock to date, with 706 patients who had multivessel disease, acute MI, and shock. Patients who underwent culprit percutaneous coronary intervention (PCI) only, with the potential for later- stage procedures, had a lower 30-day risk for death or severe renal failure compared with those who had immediate multivessel PCI (45% vs 55%, respectively). Of these, 62% were ST-segment elevation myocardial infarction (STEMI) patients and 28% were on hemodynamic support.

One of the key findings (and the finding that, in my mind, changes practice) was that mortality was 43% for culprit-only and 51% for multivessel PCI. Clearly, it is safer to do the culprit-only practice. I do think this changes practice immediately—most people do. I write the chapter on primary PCI for UpToDate. We had always been hesitant to recommend multivessel PCI because we were never compelled by the data, which were mostly driven by revascularization in previous studies. Revascularization did not count as an event in CULPRIT-SHOCK. The European guidelines[11] actually had stated that you can consider multivessel PCI in cardiogenic shock, but you will see a movement back to a more conservative approach now.

The radial evangelists said, "The multivessel strategy might have done better had it been a radial approach." I'm not sure I agree with that. There was more contrast here and longer duration of cases. Having looked at many thousands of STEMI cases as a core lab for years, I've seen all of the misadventures that people get into in these multivessel cases and I never believed it was a good idea, to be perfectly honest.

Dr Harrington: For years, the strategy for cardiogenic shock that many of us advocated was to get the main vessel open and then only go after the other vessels if you had persistent shock state. In some ways, that became a conservative strategy. People more and more were advocating to do it all at one time for multivessel disease patients.

It always interested me that in cardiogenic shock, your mortality was not tied necessarily to your left ventricular (LV) dysfunction (ie, ejection fraction). If you survive the initial insult of cardiogenic shock (which is not a trivial statement, because about 50% of patients do die) and you get out to 30 days, your 10-year mortality is the same as that of people who did not have shock. This suggests that there is something else going on in the shock state that is worth thinking about. The additional time in the lab, fluids, contrast, and ischemic time while other lesions are being treated all adds up to probably not being good for patients. I'm totally with you that this is a practice-changing clinical trial.

Dr Gibson: Whether we like to believe it or not, we are causing some obligate distal embolization when we do these procedures. In the nonculprit vessels, you are sending some more things downstream and causing some more distal embolization when you run around and do this. It's changed my practice, and I think it will change most people's practice.

Dr Harrington: It fits with what a lot of us previously had done in shock—which is get in, open up the culprit vessel, and get out of there.

ORBITA and the Editorial That Went Too Far

Dr Harrington: The last one is ORBITA.[12] I think this trial generated more tweets than there were patients in the trial. There is a lot of controversy in the community and a lot of really great discussions about this. Unfortunately, some of it has devolved into some not- so- great discussions. But it was a good trial and is an important part of the debate about how best to treat patients with stable angina. Could you give a real brief highlight on this trial?

Dr Gibson: This was a modest-sized study in 200 stable angina patients with single-vessel disease who were randomly assigned to undergo PCI or a sham procedure where patients had headphones on and the staff was blinded after the procedure to the care. There was no difference in the improvement of exercise capacity or angina symptoms. There was only 6 weeks of follow-up, so the durability of the difference is not really established.

I think [ORBITA] generated more tweets than there were patients in the trial.

PCI did improve exercise duration by 28 seconds, but medical management did not, at only 11 seconds. It has not been widely acknowledged that there was greater variability in the exercise improvement than was anticipated—kind of undercutting the power of the study. And there was a greater improvement in objective ischemia on dobutamine stress echo in the PCI patients.

One thing I've always noted (and many interventional people are not aware of this ) is that when you do a PCI, you still have microvascular disruption for many weeks. That is why people do not get exercise tests early on after PCI—because nuclear studies show a persistent abnormality in the microvasculature.

And, as Marty Leon pointed out, if you have monthly angina, it's hard to detect an improvement in a 1.5-month follow-up study. We are all eagerly awaiting ISCHEMIA, <<coder please link underlined to >> a study with more than 5000 patients with documented ischemia and stable angina at 300 centers in 40 countries, which will be a much more definitive test of the hypothesis.

Dr Harrington: You have hit a lot of the highlights of ORBITA. First off, kudos to the investigators for doing a sham-controlled trial. They have been given a lot of credit for this. This was a small to modest- sized trial. If anything, it was a proof of concept that starts to address some of the issues around stable angina, but by no means definitive. A push back from the pro- ORBITA camp has been that it was designed to detect a certain change in angina. "It was well-powered for that ; how do you say it's too small?" It's too small because you cannot tell all of the other things you want to know about outcomes in stable angina. We are doing ISCHEMIA, so we can try and sort out the hard clinical outcomes.

I'm going to enter into an area that has been a little more controversial. I thought that the article itself was well presented, but I had some real issues with the editorial,[13] which I thought bordered on the irresponsible. Then, the pick-up of the media was definitely over the top.

Dr Gibson: Yes, I'll defend anyone's right to speak their mind and state their views, but you are right. I do think the interpretation of this study was overstated, and we have to be more measured in our interpretation as a community.

Dr Harrington: That is well said. Clearly, in an editorial you should get to say what you want to say—that is the point of an editorial. I thought it just went too far in interpretation of a very limited amount of data, which is a very complex data set to understand. And I think it got a lot of conversation going, which is good ; it stated a viewpoint, which is certainly good and appropriate ; but I thought it was a little bit over the top.

Mike, this has been a fantastic discussion of some of the really big trials that have generated a lot of conversation over the course of the past year and I think are going to point to new directions in 2018, so I want to thank you for joining me.

I want to remind our viewers that our guest today was Mike Gibson, interventional cardiologist at Beth Israel hospital ; master of the Twittersphere; professor of medicine at Harvard; and, most recently, the new president and CEO of the Baim Research Institute. Mike, thanks for joining me.

Dr Gibson: Thanks for having me, Bob.

For more Cardiology, follow us on Twitter: @theheartorg


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.