Activity Transcript

Christian T. Ruff, MD, MPH: Hello, I'm Christian Ruff, Director of General Cardiology at Brigham and Women's Hospital, investigator at the TIMI Study Group, and an assistant professor of medicine at Harvard Medical School.

Welcome to our program titled, Stroke Prevention in Atrial Fibrillation: Latest Guidelines and Recommendations.

I'd like to welcome my colleagues today. First, joining me today is Craig Coleman, Professor of Pharmacy Practice at the University of Connecticut. Welcome Craig.

Craig I. Coleman, PharmD: Thank you, Christian.

Dr Ruff: And as well, my colleague Peter Rossing, who is the Head of Research and the Chief Physician at the Steno Diabetes Center in Denmark. Welcome Peter.

Peter Rossing, MD, DMSc: Thank you very much.

Dr Ruff: And a special welcome to our audience. It's a pleasure to speak with you today about what I hope you'll find an exciting topic, very important in the practice of medicine today regarding anticoagulation.

We all know that our patients are at risk for stroke and these strokes are associated with significant morbidity and mortality. In fact, they're the most devastating ischemic strokes that we encounter from practice.

For over half a century, anticoagulation therapy has really been the cornerstone of stroke prevention, and in the last 10 to 15 years, there's been a paradigm shift, where we've moved to NOACs or DOACs, which have become the preferred therapy over VKAs because they're as effective and vastly safer with respect to serious or life threatening bleeding. So if we could just start off, Craig, can you remind us what the official US and EU guidelines are for NOACs compared to warfarin and other VKAs?

Dr Coleman: Both the US and European guidelines actually currently have a class 1A recommendation, so the highest possible recommendation that they'll issue, for the use of NOACs, that includes dabigatran, rivaroxaban, apixaban, or edoxaban. Any of those NOACs are recommended over the use of warfarin in eligible patients with AF.

When I say eligible patients, we're not talking about patients who have, for instance, moderate-to-severe mitral stenosis or a true mechanical heart valve. But otherwise based in randomized controlled trial evidence, current guidelines, both US and European do recommend the use of NOACs preferentially.

Dr Ruff: Great, thanks Craig. And that really applies to most patients who have AF. There are actually, as you mentioned, very few patients would fall into a category that they would not be eligible for a NOAC over warfarin.

Peter, but we need to be careful in a specific subset of vulnerable patients. And I'm talking about those with renal impairment. All the NOACs are renally cleared, and we know warfarin is not a renally cleared drug. Can you talk to us a little bit about the renal clearance and how we approach patients with impaired renal function?

Dr Rossing: We indeed need to be careful because they have a difference in renal elimination. Dabigatran and edoxaban have the highest renal clearance with 85% and 50% renal elimination, whereas rivaroxaban and apixaban have around 30% renal elimination. That means with reduced renal function, we need to be careful and make those adjustments in the dosing of these NOACs.

Here the different drugs differ a little bit, in the sense that dabigatran, rivaroxaban, and edoxaban dosing is dependent on the CrCl, where you reduce the dose with a CrCl around 50mL/min, and you stop around 15.

For apixaban it's a little bit different, because here it's dependent on the presence of either you're above 80 years, or you weigh less than 60 kilos, or you have at elevated creatinine above 1.5 mg/dL or 133 µmol/L. And you need 2 out of 3 of these criteria in order to reduce the dose of apixaban.

Dr Ruff: I think you hit on a critical point that all of these drugs we need to consider the correct dose adjustment in patients, particularly with renal impairment, because you want to make sure you're giving the right dose to the right patient, so we really need to pay attention to all of the differences in dose adjustment between the agents. And that's really probably the most important distinguishing characteristic amongst the drugs.

Craig, we're talking about using these drugs in patients with AF. As we know, AF is predominantly a disease of the elderly. Just by the nature of their age, as well as their other significant medical comorbidities, many, if not most of these patients will have renal impairment. So just how common is renal impairment in patients with AF?

Dr Coleman: It's very much a common thing, again, based on the fact, again, that these patients are elderly and we know that as you age we tend to have a natural decline in our GFR, our renal function, in addition to all the potential comorbidities these patients have that impact their renal function, such as hypertension, diabetes.

Estimates say about 1 in every 3 patients with AF in fact have CKD. And when I say CKD, we're really talking about patients who have an eGFR of less than 60, which is right around where many of these drugs or are at least going to have to start to consider whether or not we're going to want to make those adjustments.

Now, it's also important to remember that patients with AF who also have renal impairment are at a higher risk for bleeding and a higher risk for stroke. In fact, patients who have a CrCl less than 60 mL/min, have a similar stroke risk at 3 years as a patient with AF who's already had a stroke that has normal renal function. So again, very profound impact of CKD on a patient's risk for both stroke and the risk of bleeding.

Dr Ruff: People have argued that actually renal function is a great integrator of risk, and that as you said, renal function on top of CHA₂DS₂-VASc or HAS-BLED dramatically increases your ability to predict who will have a stroke or bleeding events, so these are certainly very high risk patients.

Craig, when we're talking about the evidence base that informs our use of these agents in patients with renal impairment, can you talk a little bit about the pivotal phase 3 trials of NOACs in AF? How many patients with renal impairment were in those trials? Can we be confident that there is a wealth of evidence from randomized clinical trials that these drugs are effective and safe in patients with renal dysfunction?

Dr Coleman: There were four phase 3 RCTs, one for each of the various NOACs, rivaroxaban, apixaban, dabigatran and edoxaban.

Each of these trials were very large randomized phase 3 controlled trials with at least 14,000 patients enrolled. It is important to point out that of all the 4 clinical trials, only ROCKET AF, or rivaroxaban, actually used a specific renal dose for dose reduction, whereas the other medications, renal function could certainly have been part of the dose reduction strategy, but was not specific or the only criteria that would require dose reduction.

In terms of the proportion of patients with moderate renal impairment, and it's important to also remember that none of these clinical trials ever enrolled patients with CrCl less than 25 to 30 mL/min, of the proportion of patients with moderate renal impairment varied somewhat between the studies, but was generally between 15% and 20% of the total study population.

If you want to think about, which is also important, the number of patients who received the reduced dose, and in particular, the number of patients who received the reduced dose of various NOAC because of the renal impairment, what we find is it was about 20% in ROCKET AF, or rivaroxaban, about 20% for edoxaban and ENGAGE AF, about 10% in the RE-LY patients actually received a reduced dose because of renal impairment.

But when it comes to ARISTOTLE or apixaban, it's always kind of a little bit interesting to point out that only 1.6% of the apixaban patients actually received the reduced dose of apixaban, the 2.5 mg twice daily dose because they had impaired renal function.

Dr Ruff: I think that's a very important point. As we know, we see a lot more use of that low dose in apixaban in routine practice. We know one thing that comes up, as you've nicely said, we have a wealth of data in patients who have renal impairment for most of the drugs, showing that the drug can be safely lowered and provide adequate protection in those patients. But some ask, and I'm sure it's come up in your work as well, that well, shouldn't I just use warfarin? Warfarin is not a renally cleared drug. I know that there are these dose adjustments as your CrCl falls, but shouldn't I just stick with the VKAs in patients who have declining renal function? Because I don't really have to worry about dose adjustment and I just feel more comfortable with warfarin in those patients. Is that something that's reasonable?

Dr Coleman: You're right, warfarin is certainly not predominantly or mainly cleared renally, but that doesn't necessarily make it a better choice in patients with renal insufficiency. First and foremost, we need to remember that a common trend of all the RCTs, of course, was that the NOACs, in comparison to a VKA, were all able to reduce the patient's risk for ICH by about 50%. So certainly we don't want to give up that benefit.

But we also have to realize that warfarin itself, and we've known this for quite some time, is not necessarily friendly to the kidneys. So it's been hypothesized that warfarin, due to its inhibition of vitamin K-dependent proteins, such as a matrix Gla-protein, that it can in fact increase or speed up a patient's renal decline.

We often call this a VKA nephropathy, or a warfarin nephropathy. And so warfarin is in fact associated with a faster increase in renal decline. And part of that comes from the fact that through that mechanism I just mentioned, that VKAs such as warfarin, contribute to vascular calcification.

There's been studies that have done CT angiographic analysis of patients with AF, and what they've demonstrated is that patients who receive warfarin in comparison to those patients who don't receive any anticoagulant, have greater calcification at various different sites, including the ascending aorta, descending aorta, and as well as the aortic valve.

Certainly, we have some evidence to suggest that warfarin patients are in fact going to have an increased or an accelerated trajectory of renal decline. And so again, warfarin is not necessarily the friendliest of agents to the kidney because of that risk for VKA nephropathy.

The last thing to remember, I think, when we talk about warfarin in renal insufficiency, is if we go as far as talking about warfarin in patients with ESRD, so in patients are on dialysis who have CrCl of less than 15 mL/min, or eGFRs less than 15, it's always important there to remember that, one, there is no RCT with warfarin, just like there aren't any for NOACs in this particular space, but also that warfarin did not behave or perform particularly well in these subset of patients. So when we were looking at dialysis patients, those who received warfarin vs not receiving warfarin, patients receiving warfarin didn't do any better in terms of stroke. In fact, had an increased risk of bleeding.

Dr Ruff: That that's very powerful data. In fact, there are a lot of dialysis guidelines throughout the world that don't even routinely recommend anticoagulation because warfarin performs so poorly in that group, even though, as you mentioned, it's not predominantly renally cleared. So it certainly seems like we could do better.

You've expanded a little bit about warfarin actually may be worse to use in patients with renal dysfunction, not predominantly renally cleared, but may accelerate renal decline and renal injury. What about NOACs in these patients? What data do we have as far as declining renal function and NOAC use?

Dr Coleman: If we go back to the same mechanism, Christian, and we talk about how warfarin increases vascular calcification, there's certainly data suggests that NOACs do not. In fact, they may even improve vascular calcification, but certainly they aren't worsening vascular calcification.

If you look at those same studies I talked about before showing warfarin or other VKAs increase your risk for calcification at various sites, we see that the NOACs compared to no anticoagulation do not. And when you kind of think of the transitive property here of comparing the two or when you look at NOACs vs VKAs, there's certainly a reasonable trend towards a reduction in vascular calcification in patients receiving NOACs vs VKAs.

Probably more importantly though, is we have some great real world evidence that supports the fact that NOACs are not associated with renal decline in comparison to VKAs. In some cases, NOACs, particularly dabigatran or rivaroxaban, may even result in a slower progression of renal decline, a slower development of ESRD in comparison to warfarin. And that was shown in a very nice study using EHR data, looking at actual CrCls, GFRs and SCr values by a group at the Mayo Clinic and published in JACC in 2017.

I think probably the most important thing to point out is that, in fact, the current ACC/AHA/HRS guidelines update in 2019, suggested that over time NOACs, and they a single out dabigatran and rivaroxaban, but I think we can generally say that NOACs may be associated with a lower risk for adverse renal outcomes compared to warfarin in patients with AF. Definitely something we should keep in mind.

Finally, again just to draw a comparison. Compared to warfarin, we said we didn't have great data suggesting it did particularly well in patients with renal insufficiency, particularly ESRD. We know from many of the RCTs, for example, for rivaroxaban with ROCKET AF, that there really was no statistical interaction based upon renal insufficiency or renal function. So patients who had a renal insufficiency, particularly moderate renal insufficiency, did very similar in comparison to warfarin as compared to those people who had normal renal function in comparison to warfarin.

So again, the efficacy and safety of rivaroxaban, at least in comparison to warfarin, seems to be maintained in patients with moderate renal insufficiency. We have real world evidence that shows the same thing. If we look at the lower or reduced dose of rivaroxaban, 15 mg once daily, we again can see that in comparison to warfarin in real world data, for instance, there was a great study done by a Danish research group and published in BMJ in 2017, showed very similar results to what we saw in the RCTs.

Most importantly, or certainly equally importantly, if we're talking about patients, it's with ESRD, now we have a number of real world studies looking at patients receiving NOACs in comparison to VKAs in patients with ESRD, or who are on hemodialysis.

Again, there are no RCTs, but now we have a trial looking at apixaban vs warfarin and a trial looking at rivaroxaban vs warfarin, both suggesting that using a NOAC may result in a reduction in major bleeding in comparison to warfarin, although no difference in stroke and SE.

More recently, there is a newer real world evidence study directly comparing rivaroxaban to warfarin. Again, all these are EHR or claims database studies. And this third study directly comparing rivaroxaban to apixaban, there appears to be no difference for any of the key endpoints, including stroke or SE, major bleeding, GIB, or ICH between the two NOACs that were compared head to head in this real world evidence.

Based upon that, the current AHA/ACC/HRS guidelines, again, the update in 2019, now suggests that we could potentially consider NOACs. They particularly call out apixaban based upon the evidence that existed at the time, but certainly I would personally think that other NOACs, including rivaroxaban, as it's also not predominantly renally cleared, that we can consider using these NOACs potentially as alternative agents to warfarin in patients with significant renal insufficiency or who have ESRD. And for the ACC/AHA guidelines, this is a IIB, Class B recommendation based on nonrandomized research.

Dr Ruff: This is an incredibly important topic. Obviously, there was a lot of concern when these drugs were first studied that because they were renally cleared they would be less safe in patients with renal dysfunction because they had elevated drug levels potentially and increase the risk of bleeding. The exact opposite turned out to be true, as you mentioned, not only when appropriately dose adjusted, these agents are vastly safer across the entire spectrum of renal dysfunction.

And obviously from the clinical trial data, we did not have data in severe CKD, ESRD, and it's been very difficult to do clinical trials in that area, they're desperately needed.

But real world evidence gives us a huge advantage that we've accumulated data that not only are the drug levels quite low in these patients when they've done pharmacokinetic studies, but when that translates into outcomes, then it does appear that the benefits of NOACS over warfarin are preserved, even in those patients who have stage renal disease who are on dialysis.

Obviously, we need more information. We would love randomized data. And this varies throughout the world, as you mentioned Craig, at least in the United States the guidelines have at least allowed you to consider some of these agents because the data for warfarin in ESRD, as you pointed out, is so poor.

Now, Peter, Craig gave us a very nice overview of how do we approach patients with renal dysfunction and renal decline. Can you just talk a little bit about how we tailor stroke prevention and AF therapy to optimize patients with declining renal function?

Dr Rossing: I think you said it yourself earlier, it's about the right dose for the right patient. It means you need to assess renal function and adjust the dose when you start treating, and you also need to monitor renal function with time. That means you need regularly to measure if renal function is declining, and initially rather frequent, and then less often depending on the slope of renal function, depending on the level of renal function. You need to do it more or less often if you have completely stable renal function. So that's very important to follow the dose adjustments to the decline in renal function.

And of course, be aware of drug adherence, which is important as with any other drug. So in that case, if you use the right dose you'll get the optimal effect and the lowest possible side effects. But you need to monitor this and follow it carefully.

Dr Ruff: And Peter, just from a practical perspective, how frequently do you check renal function in your patients? What about if they are normal renal function to begin with, how often do you check it? And in a patient who already has impaired renal function, how frequently would you check their renal function in routine practice?

Dr Rossing: Our guidelines would say initially you follow renal function every 3 months if you have impaired renal function, otherwise at least once a year. And then there's a rule of thumb that you can take CrCl and divide it by 10. That gives you the number of months for screening. So if you have a clearance of 50, then 5 months to make measurements, but if it's lower, then more frequent.

Dr Ruff: I think that's a great handy trick, once a year on everybody, but in patients with impaired renal function, potentially it's going to be every 3 to 6 months. And that seems like that would probably cover most patients. And as we mentioned, that getting the dose right in these patients is critical because we need to use these drugs as they were tested in the clinical trials.

And so, if I could just summarize for our audience, I know I've learned a lot of important lessons today. One, that renal dysfunction is a major problem in patients with AF because these patients are generally elderly with comorbidities that predispose them to renal decline. Having renal dysfunction is a potent risk factor for both having a stroke and bleeding, so these are some of our most vulnerable patients. There's a lot of hesitancy to anticoagulate these patients due to the fear of bleeding, but we have a wealth of randomized data that using these agents is effective and safe, even in patients with renal dysfunction down to a CrCl of 25 to 30.

We've had really exciting real world data that has helped fill in some of the important gaps in our understanding of treating patients with renal dysfunction. Perhaps most importantly, that warfarin, although not a predominantly renally cleared drug, can actually accelerate renal decline. So you would preferentially actually want to use a NOAC in patients with renal dysfunction, because you want to preserve their renal function and slow their progression to ESRD.

And in some very exciting data, it appears that using these agents, even in patients with severe CKD, those with ESRD on dialysis, that the Xa inhibitors do appear to be safe and effective in those patients, although we desperately need randomized trials in this area.

And so I'd like to thank my colleagues, Craig and Peter, for what I hope was a very enlightening discussion. I learned a lot. And I would like to thank our audience for participating in this activity, and please continue on to answer the questions that follow and complete their evaluation.

This transcript has been edited for clarity