Happy 5th Birthday: How Has AF Stroke Prevention Changed

Christopher P. Cannon, MD: Hello and welcome. My name is Chris Cannon from the Harvard Clinical Research Institute.I want to welcome you to this program entitled: Happy 5th Birthday: How Has AF Stroke Prevention Changed?

I'm joined by several friends and colleagues, beginning with Stuart Connolly, who is the director of the Divisionof Cardiology at Hamilton Health Center at McMaster, University in Hamilton,Canada. Also, Hans-Christoph Diener, who is a professor of neurology,University Duisburg-Essen, Germany; Dr Michael Ezekowitz, who is a professor of medicine at the Sidney Kimmel Medical School at Thomas Jefferson University in Philadelphia, United States. Welcome to you all.

We’re here to discuss today how things have changed over the past five years. We’re in Barcelona, where exactly 5 years agoat this meeting, the RE-LY study results, with the first of the novel oral anticoagulants (NOACs), were presented and published simultaneously. A lot haschanged over the last 5 years; we have, now, several agents that are available,and wanted to take this opportunity to reflect on how thinga have changed,where more change is needed, and what might be coming in the future.

To begin, what was the need in this population? And perhaps Stuart -- who was one of many involved in the RE-LYtrial and its development -- can start us off. Warfarin was the only anticoagulant we had for 50 years, but it has lots of limitations, that became the impetus to get a simpler agent for our patients. This is where things sort of started for the NOACs.

Stuart J. Connolly, MD: I think the disadvantages of warfarin therapy are pretty well known. Just to reiterate, we know that warfarin is an effective treatment for prevention of stroke, but it was not being used nearly enough.There were many patients in which physicians were reluctant to use it because of an increased risk of bleeding, in particular gastrointestinal (GI) bleedingand intracranial hemorrhage (ICH). And, on top of that, it was a difficult drugto use. Patients did not like having their international normalized ratios(INRs) being checked on a frequent basis.

It was clear that we needed a better drug in order to increase the uptake of anticoagulant therapy to reduce the risk ofstroke. That better drug ideally was not going to require monitoring, would have a lower risk of hemorrhage; we didn’t really expect that we’d be able todo any better than warfarin. We thought warfarin was very effective; as effectiveas an anticoagulant could be. So we designed these trials as noninferiority trials.

Dr Cannon: Well, it’s very exciting, andthere have been lots of studies that have come out. Do you want to give aframework of the timeline of the last 5 years in the various trials?

Hans-Christoph Diener, MD, PhD: Yes.Fifteen years ago, the whole thing started with the ximelagatran trial programcalled SPORTIF; this was the first direct thrombin inhibitor studied in nonvalvular atrial fibrillation (NVAF). And the trial showed that the new drug was noninferior to warfarin and had a tendency for a lower bleeding risk. So wewere very happy that we could possibly offer something new to our patients.Unfortunately, the development of ximelagatran had to be terminated because of liver-associated safety problems.

The next drug to enter phase 3 development was dabigatran. The interesting point here was in the trial called RE-LY; therewere 2 doses of the drug compared with warfarin -- the lower dose at 110 mg,and the higher dose at 150 mg. The expectation was that the efficacy of thehigher dose of dabigatran would most probably be equivalent to warfarin, withthe lower dose possibly having inferior efficacy, but with a better safetyprofile in terms of bleeding. The big surprise was that it turned out that thehigher dose of dabigatran, 150 mg twice daily, was clearly superior to warfarinfor stroke prevention, and the low dose had a very good safety profile.

The next drugs that went into development were rivaroxaban in the ROCKET AF study, with a higher risk population; thenapixaban had 2 studies: ARISTOTLE, where apixaban was compared to warfarin, and AVERROES, where apixaban was compared to aspirin.

Then the final study was ENGAGE AF whichstudied 2 main doses of edoxaban -- 30 and 60 mg, once daily -- and a 15 mg lowdose option. However, edoxaban is not yet approved for the prevention of strokein NVAF.

Now, I think the most important issue here is the consistency of the results across all of these trials, because we alwaysbasically have the same result, in terms of efficacy.

And we have this great safety profile, inparticular, when it comes to intracranial bleeds. It’s very reassuring thatthis result was seen over and over again.

Dr Cannon: It is. I have noticed that now,with all of these trials we’ve gotten over the hurdle of consistency. Now thisis something new: How are we going to use these new agents? Well, let’s juststart by anchoring in the data; Were the RE-LY results a surprise when you presented them here 5 years ago?

Michael P. Ezekowitz, MBChB, DPhil, FRCP:Chris, they were a total surprise. They exceeded our expectations, particularlyfor the 150 mg twice daily dose. The 110 mg dose actually met expectations because it was noninferior and safer, so we were surprised. We realized that wewere rewriting the medical text book, which meant that those who were notdirectly involved had to be educated. All physicians -- not only cardiologists,but all physicians -- needed to be educated, as well as the patients.

One point I would like to make is that,while these drugs are far better than the conventional drug, warfarin, they'renot perfect. Any effective anticoagulant is associated with an increased bleeding risk -- albeit at a low rate -- compared to no anticoagulant at all.So we all need to be educated that we need to expect some bleeding, and we needto be able to deal with it. We also have to recognize, if we were to see the landscape in aggregate, that the bleeding rates would be much less than would have been the case with warfarin.

Dr Cannon: That’s an excellent point.Certainly, in acute coronary syndromes, where we’re adding more agents on topthat balance of preventing thrombotic events and bleeding is always there. It’salso interesting that patients often have a different perspective on how much bleeding they would tolerate to avoid a stroke, where they would really want toavoid a stroke as a key thing.

Well, of the other trials, as they playedout, what has been your perspective on the data across the four other trials?

Dr Connolly: Probably the biggest point I’dlike to make would be that they’ve all consistently shown fairly largereductions in ICH and subdural hemorrhage. In a way, that was the biggest surprise of RE-LY -- that you could more or less get rid of this problem. It’snot a massive problem with warfarin, but somewhere around three-quarters of a percent of patients per year will have an ICH of some sort, and they're pretty bad.

This is a limitation of warfarin. The factthat the 3 more recently evaluated agents get rid of this problem, perhaps notquite as well as dabigatran does, but certainly to a very substantial extent,tells us that this is actually a problem with warfarin. This is warfarin’sproblem, which these NOACs don’t have. Warfarin may have some advantages,perhaps, in valve disease, but it’s got this other big disadvantage. That would probably be a big take away from what came out of the other trials.

I guess the other thing, when we werewaiting for ROCKET AF to come out was: Would it be just dabigatran that provided these benefits? Or would the other agents be able to do the same sorts of things? And they have; —however, I would say they are not perfectly consistent with one another. It’s not like they're all in the same class; there are some real differences. But there's a general sense that the NOACs arebetter than warfarin. And that’s driven by the reduction in intracranial bleeding.

Dr Cannon: There are different doses testedin RE-LY, as well as in ENGAGE AF, and then obviously, there was a lower dosethat had to be selected in the other 2 trials in certain circumstances. Do you think dosing is a factor in what we see across the different trials?

Dr Connolly: Dosing is, I think,underappreciated as an important factor in how these drugs work. And that’sbeen really nicely demonstrated by both RE-LY and ENGAGE AF, where we have 2 separate full evaluations of the 2 doses. And although the doses are reallyvery close to one another, in pharmacologic terms, a dose difference is usually considered to be a doubling or quadrupling of a dose. Here we’re talking about 30 percent increases. So the dose changes are small, yet the effects are quite different.

So dose is very important. Getting theright dose in this condition has been extremely hard because you can't reallydo large pilot studies against an outcome like stroke, so it’s been achallenge. There's more work to be done in the future. I think we shoul drealize that 5 years is just a drop in the bucket. But we had warfarin for 50 years, and we continue to improve warfarin therapy decade after decade.Warfarin treatment got better and better; it still has problems, but it isbetter. We’re going to do the same with the use of NOACs. We’re going to learna lot more about areas that we don’t know -- dosing optimization, monitoring,perhaps, and all kinds of other issues that we haven't even begun to address.

Dr Cannon: Well, from your perspective – asthe "recipient" of some of the downsides when patients actually go onto have either an ICH or a stroke -- how have you seen the last 5 years evolve,in terms of treatment and its consequences?

Dr Diener: Yes, I have seen the negative from the outset. Five years ago, we would see that about 90 percent of allpatients who came to our stroke unit with a cardioembolic stroke were either untreated, or they were on aspirin, or, if they were on warfarin, the INR was below 2. So we knew that we could do much, much better.

Now, when dabigatran was approved, it wasused relatively quickly by both cardiologists and neurologists, because we hadthis clear view -- our highest priority is to prevent strokes. The uptake by general practitioners (GPs) was much lower; I think that always happens withnew drugs, because GPs are initially a little bit uncomfortable with new drugs.They had to learn how to handle it in their patients. But now the uptake has also changed with GPs.

Now, from the stand point of neurologists,this is a big step forward. And I’ll give you one example. When we had apatient with a mild stroke and AF in the past, and we started on warfarin, wecouldn’t discharge this patient on a Friday, because no one would take the INR on the weekend. And it took about 7 days until the patient was fully anticoagulated.

Early on, we see an increase in stroke risk in patients put on warfarin, but there is a long-term benefit in stroke prevention. This is not a problem anymore, because people are put on one of the NOACs Friday morning, and discharged Friday afternoon, and these patients arefully anticoagulated by that evening. The availability of the NOACs has madeour lives much easier.

Dr Cannon: The issue of how long itactually takes to get anticoagulated is something that, in the recent trialswas evident. In ENGAGE AF, it took approximately two weeks, which was the median time that people took to transition at the end of the trial to achievetherapeutic INRs with the VKAs. That’s within a clinical trial. As you know,that’s a very high risk period for the patient; then, in clinical practice, therisk is probably even worse. That’s eliminated, as you say, with the use ofNOACs.

So, when using a novel agent, you'reanticoagulated. We know one of the other perspectives in the United States has been the insurance companies blocking our ability to use these agents. That’sbeen one of the more frustrating things for me: We have great therapies; then there's paperwork to fill out, and worries and questions. I’d say I have seenin the last year, with the various agents, that the barriers seem to be lessening. Now, we’re able to really focus on the patient and choose things appropriately.

Speaking of that, what kind of patient factors go into choosing novel or different drugs?

Dr Ezekowitz: Well, Chris, cost is still anissue. I agree that it’s better than it used to be. The United States, as we know, is a very complex medical system. Regionally complex, but complexity iscreated in terms of the different insurance companies, and also depends on the policies that the patients have. Even with Medicare, if patients enter theso-called donut hole, where they're running out of their supply of money for medication, their ability to obtain a critical therapy such as an anticoagulant may be compromised.

Beside costs, we need to be cognizant ofother factors. Obviously, patients with mechanical heart valves cannot be anticoagulated with dabigatran or any of the other novel agents. Also, in termsof renal function, all these drugs, to a greater or lesser extent, are clearedby the kidney. In the U.S., we have a 75 mg twice daily dose of dabigatran forthe renally impaired, up to a creatinine clearance of 15. I see that as anadvantage.

In my own practice group, what I see isthat individuals over the age of 80 generally are treated off-label with the 75mg twice daily dose; obviously, we don’t have systematic data on those patients, but I suspect they do quite well. Renal function is therefore acritical issue with respect to these newer agents. Warfarin remains the onlydrug that we can use in the severely renally impaired patients.

Dr Cannon: Well Stuart, let me ask you. Arethere characteristics of patients or situations where you are able to chooseamong the different novel agents that may give an advantage, in your opinion,of one over another drug?

Dr Connolly: Well, right now we have 3 different agents, and we actually have multiple doses of those agents to choosefrom. We don’t have comparative head-to-head trials to help us make those choices, so we do have to kind of make some assumptions and some guesses. Butwe do so, because every day in practice, we have to choose one of the drugs that we’re going to give to our patients.

So there are some directions, I would say,without being absolutely rigid about it, that I tend to follow. One is that, inyounger patients, I favor the use of dabigatran 150 mg twice daily, which ishighly effective for stroke prevention, both ischemic and hemorrhagic, and inpeople under the age of 70, has demonstrated quite a significant reduction inmajor hemorrhage. So it’s really a very favorable drug in those patients.

In older patients, on the other hand --especially the fragile elderly, in their 80s and 90s -- I tend to favor apixaban, which has very nice data in very aged patients. We also have nicedata from AVERROES in that population. So I tend to favor it there.

Then, there's the whole issue of adherence to chronic medical therapy, which is a real issue, and tends to beunderappreciated by cardiologists who write the prescription and assume that everyone does what they’ve been told to do. We know that doesn’t happen. Sothere are patients where a once-a-day regimen is clearly favorable. Patientssometimes ask for once-a-day regimen. And I'm very comfortable prescribingrivaroxaban to those patients. So I'm prescribing these drugs in this way; Ialso use the different doses of dabigatran, both the high dose and the low dose, in different patients.

Dr Cannon: Terrific. Well, there have beensome nice real-world data released in the last few months. Do you want tobriefly cover the Mini-Sentinel and other information from the Medicaredatabase?

Dr Ezekowitz: Well, Chris, one thing you’vegot to be impressed by is the consistency of the data that is coming out fromthe clinical trials, and now real-world data. The US Food and Drug Administration (FDA) is clearly independent, and they have what we call a Mini-Sentinel program. The bottom line is the data in the Mini-Sentinel programbasically duplicated the results that we got from the RE-LY trial, which is extremely reassuring to everyone. In fact, it helps validate the drug, so that is quite impressive.

We also have data from a trial called RELY-ABLE, which is the long-term, continuation of RE-LY. If you look at thesepatients in the most recent analysis presented at the American HeartAssociation (AHA), the patients that were randomly assigned to RE-LY, werefollowed for what was then the maximum of 6.7 years, the Kaplan-Meier curvefollowed the same trajectory for all the endpoints as they actually did forRE-LY.

The expectation is that the results of RE-LY are most likely going to continue over the long term, which is extremely important, because anticoagulation for AF is long-term therapy.

Dr Connolly: I just want to expand on the Mini-Sentinel, and just point out that the validity of that very large databasevery carefully analyzed and adjusted for patient differences in over 100,000 patients, and showed not only the reduction in stroke and the reduction in ICH,but also showed a small increase in GI bleeding. So it wasn’t like it was awhite wash. It showed almost, point for point, the results of RE-LY, which really tells you that it’s very likely to be true. This consistency suggests that this data is telling us that the results of RE-LY seen in an 18,000-patient trial in fact are now materializing in patients in clinical practice. We are actually reducing mortality and reducing stroke in real world practice.

Dr Ezekowitz: Absolutely. To emphasize that point, just one technical observation is that the ischemic stroke reduction wasclose to 20 percent, which was exactly the same as what we saw in RE-LY.

Dr Diener: And with regards to myocardialinfarction (MI), there was no difference between warfarin and dabigatran. Therewas a trend, as you remember, in the RE-LY trial. But it didn’t show up in any of the registries. So we can be really confident that dabigatran does not increase risk of MI.

Dr Cannon: I don’t think we can go all the way to being really confident, but it reduces the concern.

Dr Diener: Right.

Dr Cannon: How has the uptake been inEurope? Has it been accepted? I know it’s hard to speak for all of Europe, buthave there been data on how this class is doing?

Dr Diener: Yes, usually there are twofactors that drive it. One is the adoption of evidence-based medicine, which isalways much better in Scandinavia and the UK. And then it trickles down to thesouth of Europe. And the other one is reimbursement. In particular, dabigatranand apixaban, underwent cost-effectiveness analyses by the National Institutefor Health and Care Excellence (NICE) in the UK, Scandinavia, and Germany, and they have determined that their use is cost-effective. That’s why their use isnow fully reimbursed.

Since it is fully reimbursed, theprescription numbers have gone up dramatically. Now France, Spain, and theother countries have to follow. Here, it’s not the adoption of science; it’sreally a reimbursement issue.

Dr Cannon: I mean it is interesting howthis becomes a factor in how we try and offer treatments to patients. You haveto calculate that in. But it is nice to have these formal regulatory reviews tosay, “Yes, this is a cost-effective therapy we should offer.