ESC 2017: Updates in Antithrombotic Strategies


Sanjit S. Jolly, MD, MSc, FRCPC; Christopher B. Granger, MD


September 20, 2017

Sanjit S. Jolly, MD, MSc, FRCPC: Welcome, everyone. I'm Sanjit Jolly, an interventional cardiologist at McMaster University in Canada. I have with me Chris Granger from Duke University in the United States. It's my pleasure to have you here today.

Christopher B. Granger, MD: Thanks.


Dr Jolly: We're going to talk about three studies today. One of the biggest studies that has received a lot of press is the COMPASS trial.[1]  The COMPASS trial had three arms: aspirin alone (which, of course, is standard secondary preventive strategy), aspirin plus low-dose rivaroxaban 2.5 mg twice daily, and rivaroxaban 5 mg twice daily.

Many people looked at the ATLAS-ACS trial[2]  to see whether there was a benefit from low-dose rivaroxaban 2.5 mg twice daily and a mortality reduction, and were a little hesitant to take that up. Obviously that dose was not approved in the United States.

Chris, why don't you take us through the results of COMPASS.

I think the most exciting opportunity to apply the results [of COMPASS] will be in patients with peripheral arterial disease.

Dr Granger: COMPASS had 27,000 patients and was stopped early. It had a 24% relative risk reduction in the primary outcome of cardiovascular death, myocardial infarction (MI), and stroke which was highly statistically significant. That was really important. These are patients with coronary disease and/or peripheral arterial disease (PAD), although PAD here was also defined as carotid stenosis and so it was a vascular disease.

Not surprisingly, both rivaroxaban arms—including the aspirin plus low-dose rivaroxaban 2.5 twice-a-day arm with the 24% relative risk reduction—had more bleeding than aspirin alone.

There was a trade-off, not surprisingly, between fewer thrombotic events and more bleeding events. There were two striking findings to me, Sanjit: There was an 18% relative risk reduction in all-cause mortality. However, the P value of .01 did not meet the multiple adjusted, prespecified P value. But nonetheless I think it is a convincing reduction in all-cause mortality and means that the aggregate, balance effect is beneficial to patients who were enrolled in the trial. It excluded patients at high bleeding risk. The second striking finding was about a 50% reduction in stroke. There was not as much of a reduction in MI as I might have expected.

Even more impressive was low-dose rivaroxaban plus aspirin in patients with PAD where, as you know, we have very little effective treatment. This population is very high-risk. There was substantially reduced critical limb ischemia and substantially reduced vascular and cardiovascular events in this population where we really need additional therapy. I think the most exciting opportunity to apply these results will be in patients with PAD.

Dr Jolly: The PAD population is the kind of population where we have not had good therapies. Reducing amputations by more than two thirds is a very important outcome for these patients.

The other important thing is that stroke, as a nonfatal outcome, is probably one of the most important outcomes. Reducing stroke by 50% is something important for a patient. I would prefer to have a gastrointestinal bleed than a stroke, where I could become hemiplegic permanently.

Dr Granger: I think there are two important issues in applying these results in practice. One is that we will have to be thoughtful about assessing thrombotic and bleeding risks. This may not be appropriate in patients at high bleeding risk and low thrombotic risk. Cost is an issue. The comparator was aspirin, and maybe in PAD, clopidogrel is better than aspirin, based on the CAPRIE trial.[3] There are some twists that will force us to be thoughtful in the application of the results in practice.

Dr Jolly: I enjoyed reading Eugene Braunwald's editorial,[4] and at age 88 I think it's fabulous that he is still going strong. One of the interesting things that he talks about is the choice of dual antiplatelet therapy (DAPT) versus aspirin plus low-dose rivaroxaban. You have a patient who is 12, 14, 16 months post-MI and had a stent. Which regimen do you choose? Do you choose aspirin and continue with ticagrelor? Do you choose aspirin-clopidogrel? Or do you choose aspirin with low-dose rivaroxaban? What do you think of this issue?

Dr Granger: It's a great point, and what we have now is options. We can use a variety of treatments which will further reduce thrombotic risk. Each one has some bleeding risk and we don't really know which one is most effective. But we have a toolkit now. We have a variety of options. I think there is real benefit from continuing ticagrelor in a patient who has done well on ticagrelor for a year after acute coronary syndrome (ACS). It's a class IIb recommendation in the guidelines,[5] but if you look in PEGASUS,[6] those patients 1 year after their MI seemed to have real benefit from continuing ticagrelor. I think that is a good option. Certainly, the COMPASS data are also fairly compelling.

Dr Jolly: Yes. I think this is a dilemma for physicians—cardiologists, general practitioners. They will have to decide whether to stop ticagrelor and switch to a COMPASS-type regimen, or whether they continue aspirin-ticagrelor. It's interesting. We do not have head-to-head trials, so all we can do is look at PEGASUS on its own and COMPASS on its own.

It's apparent to me that both agents are associated with higher risks of bleeding. The second thing that is interesting is that ticagrelor appears to be very effective at preventing stent thrombosis and MI. Rivaroxaban, on the other hand, seems to reduce stroke, peripheral vascular outcomes, and mortality. As a clinician, I'll have to look at what a patient's MI risk is and what their other risks are and make a distinction.

I think you're right that there is probably a rebound effect. You come off your DAPT and you are at an increased risk for a certain period of time. I think patients who have been off DAPT for a period of time are perfect COMPASS patients. Then you have to make a decision on which regimen to use, based on how far they are from their ischemic event and potential type of revascularization.

A coronary artery bypass graft (CABG) substudy in COMPASS is looking at graft patency, so we will have more information on the three different regimens in CABG patients. This will probably be presented at the American Heart Association meeting.

Dr Granger: Of course, the rivaroxaban 2.5 mg twice-a-day dose is not yet available in the United States. We will have some time to further analyze and distill these results, and figure out practical tools to assess residual risk for thrombosis and risk of bleeding and how we can make the best decisions on which regimens to use. It's really good news that we have additional tools.

Dr Jolly: Let me ask you a question. The rivaroxaban 2.5 mg twice-daily dose was not approved by the US Food and Drug Administration (FDA). When you go through the meeting minutes, a comment pointed out that there was no dose relationship. In other words, the higher dose did not appear to have a further reduction in events versus the 2.5-mg dose. Now, with COMPASS, how do you feel about the ATLAS results? Do you feel that things are different? Obviously, the FDA may revisit ATLAS now.

Dr Granger: Yes, I do think things are different because mortality was a secondary outcome and the lack of consistency across doses made me wonder whether that was an exaggerated finding. It probably was. The reality is probably more in line with what we saw in COMPASS.

I don't know if the FDA will revisit. Perhaps. I think it would be reasonable because it's on that continuum and it may well be reasonable to use rivaroxaban earlier.


Dr Jolly: Okay. Excellent. COMPASS was a very exciting trial. We're going to move from one novel oral anticoagulant (NOAC) trial to another NOAC trial. RE-DUAL PCI[7] is really addressing a thorny issue. You have a patient with atrial fibrillation (AF), you do percutaneous coronary intervention (PCI), and the vast majority of patients are getting drug-eluting stents. Now what do you do? Do you treat them with triple therapy with warfarin? Or can you use dual therapy with a single P2Y12 inhibitor (which, for the most part, is clopidogrel) and a NOAC?

In retrospect, it looks like the FDA maybe made the right decision in encouraging the use of the 150-mg dose of dabigatran.

Here they had three arms: dabigatran 110 mg twice daily with a P2Y12 inhibitor; 150 mg twice daily with a P2Y12 inhibitor; and triple therapy with warfarin, aspirin, and a P2Y12 inhibitor. No surprise, there was a reduction in major bleeding when you got rid of aspirin and used a NOAC. What do you think the messages are from this trial?

Dr Granger: RE-DUAL was underpowered and this is a challenge now in the field. This comes on the heels of PIONEER-AF-PCI,[8] the studying of rivaroxaban, and now RE-DUAL. Neither was powered to show whether the strategy of avoiding aspirin—and having a somewhat lower dose, especially in PIONEER—would still adequately protective against stent thrombosis and stroke.

Because both doses of dabigatran have already been shown to reduce stroke in AF, one should have somewhat more confidence in RE-DUAL. First of all, the overall findings in RE-DUAL really reinforced the PIONEER findings. The two major take-home messages for me are: It's okay to use a NOAC, including dabigatran, in patients with AF who have ACS and/or stents; and it's okay to stop the aspirin fairly early. Most of these patients got aspirin for a few days because it took a couple of days to get them enrolled after their ACS, but they seemed to do well with relatively lower rates of stent thrombosis.

I do wonder in RE-DUAL about the low dose of dabigatran (110 mg), although it was a relatively small sample size, because there were numerically higher rates of both stroke and stent thrombosis. I think it reinforces the strongest dabigatran data, which are with the 150-mg dose, which the FDA really likes. In retrospect, it looks like the FDA maybe made the right decision in encouraging the use of the 150-mg dose of dabigatran. By the way, dabigatran 150 mg with mostly clopidogrel had very low rates of intracranial hemorrhage in RE-DUAL, confirming the concept that dabigatran is very safe—even at 150 mg—with respect to intracranial hemorrhage.

Dr Jolly: You have a patient, let's say next week, who has PCI with a drug-eluting stent. They have AF and are on warfarin. What would you do?

Dr Granger: First of all, If they had been on warfarin, I would be willing to enroll them in a trial like the ongoing AUGUSTUS trial, which we are participating in, because I think we still are uncertain with warfarin. For example: Should we stop the aspirin with warfarin? These other two trials, PIONEER and RE-DUAL, used triple therapy with warfarin. We know from the WOEST trial[9] that you may get a similar reduction in bleeding by dropping the aspirin, and it would still be effective as an antithrombotic strategy.

I think it would be reasonable at this point to use either rivaroxaban or dabigatran in that patient. What I would say, Sanjit, is that if it's a patient who comes in already on dabigatran or on rivaroxaban, I would definitely continue that agent and then use the strategies tested in PIONEER and RE-DUAL as my guidance for what to do with them subsequently.

Dr Jolly: All three datasets together will be important because each trial individually was not powered for ischemic outcomes—stroke, stent thrombosis. Once we put all of the trials together in a meta-analysis, that will certainly be much more useful.

Dr Granger: I completely agree. The AUGUSTUS trial is testing apixaban at the full 5 mg twice-a-day dose unless patients have dose-reduction criteria. Patients are assigned to warfarin or apixaban, and then there is a factorial randomization to continuing aspirin or having no additional aspirin. That will be very important because we will get a more direct comparison with warfarin versus a NOAC and find out whether aspirin is important for the antithrombotic effect with warfarin or the NOAC.

Dr Jolly: Of the three trials, I think that design is probably going to be the most informative in terms of the factorial design, so I'm looking forward to it. Do you know when AUGUSTUS is going to be completed and presented?

Dr Granger: It's going to be at least another year or so before it will be completed.

Dr Jolly: What is the sample size of AUGUSTUS?

Dr Granger: It’s 4000 or 5000 patients.

Dr Jolly: That is an important piece of data that we will await, but I agree. In the current period, my approach has been, for the most part, stopping aspirin. I continue aspirin in patients who have a left main stent because a stent thrombosis would be catastrophic and potentially deadly.


Dr Jolly: Why don't we wrap up and spend some time on VALIDATE-SWEDEHEART.[10] Bivalirudin did not seem to beat heparin. What do you think?

The message is: There is really no significant advantage of bivalirudin. And there is no disadvantage.

Dr Granger: In VALIDATE-SWEDEHEART, the Swedish group once again did a registry-based randomized trial. A large portion of all the acute MIs in Sweden—about 6000 patients—were entered into this pragmatic study testing heparin (I think appropriately dosed) versus bivalirudin. The message is: There is really no significant advantage of bivalirudin. And there is no disadvantage; in fact, there were numerically slightly fewer bleeding events and early ischemic events. Basically, it looks as though bivalirudin is not providing any substantial advantage in a more general, inclusive population.

Dr Jolly: I think that registry-based trials, whether in the United States or Sweden, are probably the wave of the future, and it's fantastic that they were able to do such a trial.

To summarize, it was very exciting that we had the results of a large trial like COMPASS and that we have a new option, particularly for patients with peripheral vascular disease. RE-DUAL PCI again provides more information on how to treat these complex patients with AF and PCI. I think VALIDATE-SWEDEHEART provides another trial supporting either bivalirudin or heparin. Thank you, Chris.

Dr Granger: Thank you.


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.