Michelle L. O’Donoghue, MD, MPH; David J. Cohen, MD


January 04, 2018

Michelle L. O'Donoghue, MD, MPH: Hi. I'm Michelle O'Donoghue of Harvard Medical School and Brigham and Women's Hospital in Boston. I'm at the American Heart Association's (AHA's) Scientific Sessions in Anaheim, where I'm pleased to be joined by David Cohen, who's now the newly appointed vice chairman of medicine and research at Beth Israel Deaconess in Boston. You're going to be a neighbor. Welcome.

David J. Cohen, MD: Exactly. Back to where I started.

Dr O'Donoghue: A lot of interesting research has been presented at the Scientific Sessions this year looking at different combinations of antiplatelets and antithrombotics. There are a few different topics we can touch upon. There was a cost analysis presented for COMPASS. Perhaps let's just remind the viewers what COMPASS was studying.

COMPASS Background

Dr Cohen: COMPASS was a very large, practical trial in patients with documented coronary artery disease (CAD) or peripheral artery disease (PAD). There were three arms in the trial; one was a standard-therapy arm with just low-dose aspirin. The second arm was 5 mg twice daily of rivaroxaban compared with aspirin. The third arm was the combination arm of a very low dose of rivaroxaban, 2.5 mg twice daily, and low-dose aspirin.

The results of the trial, which were presented earlier this year at the European Society of Cardiology (ESC) annual meeting and published in the New England Journal of Medicine,[1] showed that the benefit was all in the low-dose rivaroxaban arm. The rivaroxaban 2.5 mg twice daily plus low-dose aspirin had about a 20%-25% relative risk reduction for severe cardiovascular events (cardiovascular death, myocardial infarction [MI], stroke, and limb-related events) compared with the low-dose aspirin group. There was about a 70% relative increase in bleeding. There was a tradeoff like we see so many times when we're giving anticoagulants or other antithrombotic agents.

The most striking thing in that main analysis was the reduction in mortality—that trumps everything—and there was a reduction in stroke. The two most worrisome endpoints, the two worst cardiovascular endpoints that we have, were lowered. There were also reductions in things like amputations and revascularization procedures. There was a real, strong signal of benefit in that trial.

Dr O'Donoghue: Certainly the mortality benefit is the piece that got many people discussing the results and seriously considering the addition of rivaroxaban on a background of aspirin for these patients.

Dr Cohen: Exactly.

Dr O'Donoghue: It's supportive of what we saw post-acute coronary syndrome (ACS) in the ATLAS ACS2 trial,[2] where the same dose, that very low dose of rivaroxaban at 2.5 mg twice daily, appeared to translate into a mortality benefit.

Dr Cohen: Yes, exactly. That was more controversial because there were some issues with the trial, and it was unexpected. At this point, I think we have to acknowledge that there's something really there.

Dr O'Donoghue: As we think about cost—that's what you were the discussant for the presentation today—what are your thoughts?

Dr Cohen: Cost is very important. We just had a session where we were discussing the cost of the PCSK9 inhibitors. It's important there, and it's important with COMPASS for a couple of reasons; the first one being the population size. In Western societies like the United States, the potential population is very large for these drugs. The second thing is because there were tradeoffs. When there are tradeoffs, we need some way to arbitrate whether those tradeoffs are beneficial or not. People use net clinical benefit, but net clinical benefit is actually an overly simplistic way to look at these tradeoffs because it simply says okay, we'll consider a bleeding event to be just as bad as a stroke or death or just as bad as an MI, and who knows?

One way we can put these things on the same scale is to add costs. Then, of course, the other reason is that these therapies are not going to be just given for 23 months as they were in the COMPASS trial; they're going to be given for a much longer period of time, and those costs add up.

Cost and cost-effectiveness are helpful to try to understand whether there is value for society, which is very important these days. If there isn't value perhaps in the overall population, are there subgroups where there's particular value?

COMPASS Cost Analysis Incomplete

Dr O'Donoghue: Thinking about the cost-analysis data[3] that were presented for COMPASS, what are your thoughts, and what's the takeaway message?

Dr Cohen: There were a couple of really interesting findings from that analysis, which was an incomplete analysis, as I acknowledged during my discussion. It only looked at the cost consequences, just the cost offsets, because we don't yet know the price of the drug at this dose. They hadn't completed any of the effectiveness analyses looking at the life expectancy of those things.

Just on the cost side, there were a couple of things. First of all, there were real cost offsets, and they were modest. There were about $350 a year on average. Just like we heard about the PCSK9 inhibitors, there's a big gradient of cost reductions depending on the population. The group that got the least benefit by far was the patients who just had a history of coronary disease. Their cost reductions were on the order of $200 a year, very little.

Once you started to get into PAD and polyvascular disease again, the cost offsets got much more substantial, on the order of $700-$800 per patient per year, which obviously does add up and helps to offset whatever the cost of the drug might be. That's the main message that came out of the study.

The other thing that I thought was very interesting is where those costs reductions came from, and the vast majority of the cost reduction came from reduction in stroke. First of all, there was a real benefit on stroke.

Dr O'Donoghue: Ischemic stroke?

Dr Cohen: Yes, ischemic stroke. The issue with stroke is that it's not just the hospitalization costs because many of these patients are disabled. They have extended stays and rehabilitations; some of them end up in chronic care facilities. All of those things add substantially to the costs. That accounted for about half of the cost savings in the overall trial. The other biggest things that drove cost savings were reduction in procedures, particularly percutaneous coronary intervention (PCI) procedures but also peripheral artery procedures.

It really emphasizes how neglected a population the PAD group is. I think it's a very good time to be caring for those patients because we now have several therapies that have shown important benefits in that group; that's such a high-risk group. They keep having events, and we need better therapies for them.

Dr O'Donoghue: I think one of the things that came out of this analysis is, as you pointed out in the beginning, that the dose arm that had rivaroxaban alone without aspirin did not appear to do as well as the dose arm that had low-dose aspirin in combination with rivaroxaban 2.5 mg twice daily.

Dr Cohen: It looks like inhibiting both pathways is a good thing.

RE-DUAL PCI Subgroup Analysis

Dr O'Donoghue: You need a little bit of both, right? I guess that dovetails nicely into the question of RE-DUAL PCI. There were updated subgroup data[5] presented here at the AHA. There was a strategy of people who required triple therapy using either 150 mg or 110 mg twice daily of dabigatran with a P2Y12 inhibitor as opposed to standard triple therapy that includes aspirin.

Dr Cohen: This has been an issue for those of us who work in the cath lab and do coronary interventions for many, many years; the challenge of you should do in a patient who has a strong indication for oral anticoagulation, whether it be with warfarin or one of the newer agents.

Obviously, a number of years ago, 5 years ago or so, there was the WOEST trial[6] that came out of Europe, which was the first study to really take this on and ask the question: Can we pull back? Can we have the best of both worlds if, in the case of WOEST, we just took away the aspirin, continued to give patients warfarin and clopidogrel, and compared that with triple therapy? They saw a striking benefit—a major reduction in bleeding but also a reduction in things like stent thrombosis and a suggestion even of a mortality benefit in that trial.

Now we have the non-vitamin K oral anticoagulants (NOACs)/direct-acting oral anticoagulants (DOACs), and so several trials have been conducted looking at those drugs as well, asking if we can do something similar to what was done in the WOEST trial.

The most recent one is the RE-DUAL trial that was run by Chris Cannon; as you know, it came out at the ESC. This was a trial with the two doses of dabigatran you alluded to compared with warfarin. The key here was that both of the dabigatran doses (110 mg and 150 mg) were combined with a single agent—just clopidogrel or ticagrelor (mostly clopidogrel). Then the warfarin group got triple therapy with aspirin added on top.

What they saw, in the overall population, was very reminiscent of what had been seen previously in WOEST, which was that the group that didn't get aspirin, in this case the two dabigatran groups, had significant reductions in the various bleeding endpoints with, at least, no statistically significant increase in ischemic events like stent thrombosis. However, in the dabigatran 110 mg group, there was a trend that was a little bit concerning for that—that dose may not be enough in freshly stented patients—although it was not significant.

Those are the basic findings. Today, they presented several subgroup analyses to see if there was anything that looked strange in them. The subgroups they looked at were: ACS vs none; the chosen P2Y12 inhibitor (clopidogrel or ticagrelor); and then bare-metal vs drug-eluting stents.

Everything was consistent. It was a pretty simple, straightforward message, which was that the overall trial results held up in each of those three subgroups. The message that seems to be coming out from these trials is that taking away aspirin may be okay in these settings. That is, when you're adding on a full dose of an oral anticoagulant plus a full dose of a P2Y12 inhibitor, it's probably sufficient for these patients, and adding the aspirin just increases the [bleeding] risk.

Dr O'Donoghue: It does provide reassurance. Clinically we grappled with this quite a bit, the idea of whether or not to drop aspirin; and, if we do, do we do that for everyone or just our patients at high risk for bleeding? What is the timing? One interesting piece with RE-DUAL is that they had some patients on ticagrelor. One thing that I've actually sometimes worried about is that if I have a patient who is on clopidogrel with either warfarin or an NOAC or DOAC, and I drop the aspirin, what if they're one of the one third of people who don't respond to clopidogrel?

Dr Cohen: It is actually interesting because we know that those patients are out there—a third of them who respond less to clopidogrel, and then there's 2% or 3% who essentially don't respond at all. Nonetheless, we can't detect anything in these trials. That simply could be because all of them are underpowered to look at the stent thrombosis events. None of them have been powered to adequately assess that. We would need to have enrolled five to ten times as many patients in these trials to make that assessment. It may be that having the oral anticoagulant there provides some measure of background protection.

If you go way, way back—this is before your time, but I was practicing when we did the STARS trial back in the '90s with PCI.[7] In that trial, there were three arms. There was aspirin alone and an aspirin and ticlopidine arm, which was the big winner at that time.

Dr O'Donoghue: Huge relative risk reduction.

Dr Cohen: Yes. Huge relative risk reduction in stent thrombosis. There was also an aspirin plus warfarin arm. There was actually a little bit of a benefit of the warfarin at that time. I think we underappreciate that antithrombins do provide some protection from stent thrombosis as well.

Dr O'Donoghue: That's been seen with rivaroxaban as well.

Dr Cohen: Yes, exactly. The answer to exactly what's the right time to stop the aspirin in these patients is controversial. I can tell you that at the Mid America Heart Institute, where I practiced for the last 11 years, a lot of us would give at least a week of aspirin, often a month, just to bridge that highest-risk period for stent thrombosis and to cover against the issue of clopidogrel nonresponse, but there are no data on that.

Dr O'Donoghue: That's terrific. It was great to get your thoughts on this topic. I'm sure a lot of our viewers are grappling with the same clinical issues. Thanks for your insights.

Dr Cohen: My pleasure.


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.