Dosing the New Oral Anticoagulants in Atrial Fib: Can We Do Better?

August 04, 2014

NEW YORK, NY — "Oral administration, please, but no cumbersome dosing strategies" could have been the tag line of the research community that spent decades looking for a more user-friendly alternative that is at least comparable to warfarin in effectiveness and safety. And there's broad agreement, perhaps not universal, that those conditions were met or exceeded in the pivotal stroke-prevention trials presented to regulators in Europe and North America who approved the thrombin inhibitor dabigatran (Pradaxa, Boehringer Ingelheim) and factor Xa inhibitors rivaroxaban (Xarelto, Bayer) and apixaban (Eliquis, Pfizer/Bristol-Myers Squibb) for patients with nonvalvular atrial fibrillation (AF).

But some who may relish the fixed-dose advantages offered by these new oral anticoagulants (NOACs), at least as they were developed and approved, still see potential roles for assays in tailoring dosing to individual patients. Experts heartwire queried about the issue contend that regular, lifelong monitoring by assays, à la INR monitoring for warfarin, is unlikely to be needed or even useful for the new agents. But possibly, assays for coagulation function or drug serum levels at the outset of NOAC therapy could identify an optimal fixed dosage for individuals that might boost the efficacy and/or safety the drugs demonstrated in the pivotal trials. Meanwhile, periodic assays might potentially be used in selected higher-risk patients.

"A Great Step Forward"

Dr Lars Wallentin

The trials that convinced regulators to approve the NOACs for stroke prevention in AF, along with meta-analyses published since, have concluded that the drugs, in fixed doses and compared with warfarin, were "at least or even more efficacious, and even safer concerning intracranial hemorrhage. And that was a great step forward," said Dr Lars Wallentin (Uppsala Clinical Research Center, Sweden) in an interview.

Would dabigatran, apixaban, or rivaroxaban have appeared even safer or more efficacious in the trials had they been administered differently? "I think that's the question now. In some ways, we are in the post–vitamin-K-antagonist era, and that raises new questions," he said.

Wallentin shared top leadership positions in both the RE-LY trial of dabigatran and the ARISTOTLE trial of apixaban, both central to the drugs' eventual approvals for stroke prevention in AF. The message for researchers and pharma companies, he said, is that while much has been achieved, "you could do better. We need to improve the safety of these agents." That could be done through additional comparative trials. With dabigatran, for example, dosing guided by assays performed "maybe just once or twice at the start or just in higher-risk patients" might lead to better outcomes.

"We all believe there could be potentially improved models for dosing these agents in specific subpopulations, but none have been prospectively evaluated," noted Dr Jonathan L Halperin (Icahn School of Medicine at Mount Sinai, New York, NY). He was on the executive steering committee of the ROCKET-AF trial of rivaroxaban and the executive committee of ENGAGE AF-TIMI 48 , a stroke-prevention trial of the investigational factor Xa inhibitor edoxaban (Lixiana, Daiichi-Sankyo).

"We're steering the ship between the shoals of thromboembolism and bleeding, and there are some pretty narrow waters there in many of these older, sicker patients with, say, prior stroke. The risks are huge if we are not right. There's doubtless going to be room for improvement, but getting from where we are to that ability to customize dosing on an evidence-based foundation is not going to be easy."

Enoxaparin All Over Again

"A very similar scenario played out years ago when enoxaparin came on the scene," observed Dr Deepak Bhatt (Brigham and Women's Hospital, Boston, MA) for heartwire . "One of its advantages, most would argue, was that [unlike unfractionated heparin] it didn't need [activated clotting time] ACT monitoring. But there were certainly physicians saying if we could monitor enoxaparin, we would use it more," he said.

Dr Deepak Bhatt

"I think, other than the cost issue, there's really little downside to being on the NOACs, and that's without any specific monitoring." Still, "the ability to monitor levels and further titrate the novel oral anticoagulants might further improve outcomes. That's a possibility. I think that might be worthy of prospective study. But I'd predict that it would be a bit of a trade-off. In real life, monitoring adds complexity and cost to care."

Adding such complexity increases the risks of "all sorts of medical errors" and may discourage adherence to a more complex monitoring scheme, according to Bhatt. Monitoring-guided NOAC dosing might elevate efficacy or safety in some individuals, but "in populations of patients, for the number that you help, you might end up unintentionally harming more."

Dr Alexander Turpie

If physicians want to go down that road, Bhatt said, probably "it would require a whole new set of well-thought-out large studies to see if that kind of tailored approach improves outcomes." Bhatt served on the executive committee of ATLAS-2 and the steering committee of APPRAISE-2 , trials of rivaroxaban and apixaban, respectively, in the setting of acute coronary syndromes.

If future trials suggest that assay-guided dosing can improve NOAC therapy in AF, it would probably apply more to dabigatran than apixaban or rivaroxaban, Dr Alexander Turpie (McMaster University, Hamilton, ON) told heartwire .

"Dabigatran has only about 6% absorption in the gastrointestinal tract, and thus even small [dosing] deviations can make a big difference in the amount of drug in the blood. The factor Xa inhibitors, on the other hand, are up to 100% absorbed, so there's much greater latitude for minor deviations," Turpie explained.

"So far I don't see any specific reason to suggest that monitoring would be helpful with factor Xa inhibitors across the board."

RE-LY Intrigue Reported in the BMJ

Indeed, dabigatran was the primary focus of recommendations in one of four BMJ reports published in late July calling for:

  • Identification of a dabigatran therapeutic plasma-level range.

  • Regulators on both continents to "recommend initial dose adjustment based on plasma measurements."

  • Regulators to strike "does not in general require routine anticoagulant monitoring" from the drug's labeling.

Those recommendations followed accounts of alleged shortfalls or improprieties by regulators, industry, and clinical trials in how they ushered dabigatran to market, as described in the media earlier in 2014 and disclosed in the four BMJ reports. As covered by heartwire previously, the four reports argue that the actual bleeding risk associated with the thrombin inhibitor in RE-LY remains unclear. And they question whether the investigators and Boehringer Ingelheim inadequately shared pharmacokinetic data from RE-LY dabigatran recipients that suggested the drug's bleeding risk could be improved by preventing excessively high plasma concentrations.

Dr Stuart J Connolly

"That's just patently false," RE-LY co–principal investigator Dr Stuart J Connolly (McMaster University and Hamilton Health Sciences, ON) said to heartwire . "All of the pharmacokinetic data were submitted to the FDA, which did its own analysis on it and would have been able to see this relationship itself and in fact did see it."

And although some have questioned the completeness of the trial's original bleeding-event adjudication, Wallentin observed, the claimed discrepancies were divided between patients on dabigatran and those on warfarin.

"We think that despite the additional bleeding events, the results are extremely robust for reduction in ischemic strokes with the higher 150-mg [twice-daily] dose, with reduction in intracranial hemorrhage with both [the 150-mg and 110-mg twice daily] dosages, and with reduction in bleeding with the lower dose, with improvement in cardiovascular mortality."

The BMJ reports, Wallentin said, "from my perspective are really arguing that all of the companies working on these new agents should also now take the next steps to further improve a very good treatment."

Dr Gregory YH Lip (University of Birmingham, UK) had a more contrary take on the BMJ reports: most of what's in them isn't news, and "I think it's disappointing how this has been a bit hyped up again. At the end of the day, it's just going to scare people away from effective stroke prevention."

Lip also said he would "have to be convinced" that assays for assessing dabigatran serum levels or anticoagulant activity would be practical or clinically worth it, unless they turned out to be the kind of fast-turnaround point-of-care tests clinicians would like for patients presenting urgently with bleeding or those needing emergency surgery.

No Warfarin Redux

Long-term repeated monitoring of coagulation or serum levels is unlikely for the NOACS the way it's come to be done for warfarin; the older drug differs from the newer ones in a number of important ways. It has a narrow therapeutic range, as reflected in INR readings; corresponding therapeutic ranges for the NOACs haven't been pinned down but are believed to be relatively forgiving. Warfarin's activity, notoriously, is slow to rise and fall and is affected by dietary vitamin-K intake, alcohol intake, and interactions with some antibiotics and a number of other agents. All that unpredictability makes it important to keep track of clotting function and need for dosing adjustments.

Dr Gregory YH Lip

In contrast, dabigatran and the factor Xa inhibitors aren't affected by diet and have moderately rapid onsets of action, Lip observed, and their effectiveness and safety at fixed dosages vs optimized warfarin have already been shown in clinical trials. And he disputes what some argue, that regular monitoring would help keep patients compliant with NOAC therapy.

Patients can easily be noncompliant for weeks before coming to the clinic for an assay, he said. "Even if I had [access to an appropriate assay], and the patient comes to see me in the clinic at, say, mid-morning, they can just take their drug earlier in the morning and it's going to show the drug in the therapeutic range. It will appear that they have been compliant." In contrast, Lip noted, INR reflects warfarin activity across a number of weeks prior to the measurement.

What About Limited Assay Guidance?

It might be sufficient to limit assays for serum levels or anticoagulant activity to once or twice at the outset of NOAC therapy to "determine the response," said Turpie. "Once you establish a dose, I think it would be reasonable not to have regular monitoring." Individual responses are likely to remain fairly stable along with body weight, renal function, "and the genetic metabolism of the drug," he said. "But for patients with deterioration of renal function, as with aging or from a disease, you might need additional measurements."

"It's even conceivable that [initial monitoring-guided dosing] might be beneficial in a much larger segment of patients," Bhatt said. "But before going there, I'd want to see some prospective studies showing that, in real life, that type of monitoring actually reduces bleeding or reduces ischemic stroke."

. . . Or Elimination of Assays Altogether?

An alternative strategy, equally untested in prospective trials, would be to prescribe higher or lower dosages based the patient's clinical presentation and perhaps tools for stratifying risk of bleeding and stroke, such as HAS-BLED or CHA2DS2-VASC scores, respectively. That prospect has been much discussed by clinicians in the field. "I think the first step is to start using the lower dose of dabigatran in the more vulnerable populations, like the elderly," Connolly said. "But first, we have to convince ourselves that it will have an impact."

Bhatt agreed that it's another option, "and I guess it resonates a bit more with me as a clinician than just focusing overly on pharmacokinetics. Having flexibility of doses does allow one to titrate medications based on renal function, based on perceived or calculated bleeding risk and calculated thrombotic risk. Sometimes a blood level tells you only part of the story, and sometimes the eyeball test tells you much more," he said.

"The only way to know for sure is to test the regimens prospectively," Bhatt again emphasized. "At least, a number of the bleeding complications I've seen with novel anticoagulation have been in a setting where patients have borderline renal function and then it got worse, but there was no change in their anticoagulation regimen. There, it's important to monitor their renal function and make sure there doesn't need to be some sort of dosage adjustment."

How the Trials Collectively Tested the Idea

There's already been something of an experiment with this kind of approach to NOAC dosing in the drugs' stroke-prevention trials. "Each successive study of the target-specific oral anticoagulants has used a progressively more sophisticated dosing strategy," Halperin pointed out.

RE-LY randomized patients to fixed dosages that weren't altered based on age, renal function, or any other features. In ROCKET-AF, patients were randomized to different rivaroxaban dosages based on creatinine clearance; prospective analysis indicated comparable outcomes for the patients with vs without renal dysfunction, according to Halperin.

ARISTOTLE randomized patients to 5-mg twice-daily apixaban unless they met two of three criteria for increased risk (>80 years, weight <60 kg, and serum creatinine >1.5 mg/dL), in which case they received 2.5 mg twice daily.

And in ENGAGE AF-TIMI 48, there were both high-dose and low-dose randomization groups, but either edoxaban dosage could be cut by half if patients showed slowed creatinine clearance or low body weight or were receiving any of a list of drugs known to affect the drug's activity.

"So it's clear that we're moving in a direction of a customized dosing strategy," Halperin said.

Lip described how it works in his clinic, which is at a large referral center. "The approach we have is to fit the drug to the patient and vice versa," he said. "Let's say a patient I see has a high risk of bleeding. I would then fit the NOAC with the better bleeding profile, for example apixaban, or maybe use dabigatran 110 [mg twice daily]. If the patient had renal function that was slightly dubious, I'll be inclined to choose a NOAC less dependent on renal excretion; in the UK, that would mean either apixaban or rivaroxaban. If my patient is having recurrent stroke despite being on really well-controlled warfarin, I would go for the highest-potency drug on the assumption that [the patient] is not at excessively high bleeding risk, and that would be dabigatran 150 [mg twice daily]."

Currently, one impediment to such a strategy in the US is the FDA's controversial decision not to approve the 110-mg twice-daily dabigatran dosage that fared well in RE-LY. It instead approved the trial's 150-mg twice-daily dosage and a 75-mg twice-daily dosage. The latter had not been explored in any prospective randomized trial, but according to the agency it was supposed to give clinicians a low-dose option in elderly patients. But instead of providing dosing flexibility, it restricted available evidence-based options.

Indeed, the reports in BMJ last month urged the FDA to green-light the 110-mg twice-daily dabigatran dosage "that is approved in the European Union, Canada, New Zealand, and Australia to permit dose adjustment in high-risk patients."

Eventually, Wallentin said, clinicians everywhere might have all three dosages at their disposal, which "might be the ideal solution for the future. Maybe there should be even more dosage levels available, even for the other NOACs. That might be the next step, but it needs to be proven in new prospective trials."

Connolly discloses receiving research grant support and honoraria for consulting and lectures from Boehringer Ingelheim. Bhatt discloses receiving research grants from Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, Roche, Sanofi, and the Medicines Company. Halperin has previously disclosed consulting fees from Astellas Pharma, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Pfizer, and Sanofi and chairing a data safety monitoring committee for a trial sponsored by AstraZeneca. Lip reports serving as a consultant for Bayer, Astellas, Merck, Sanofi, Bristol-Myers Squibb/Pfizer, Daiichi-Sankyo, Biotronik, Medtronic, Portola, and Boehringer Ingelheim; and serving on speakers' bureaus for Bayer, Bristol-Myers Squibb/Pfizer, Boehringer Ingelheim, Daiichi-Sankyo, Medtronic, and Sanofi. Turpie has previously disclosed consulting for Bayer and serving on speakers' bureaus for Johnson & Johnson and Boehringer Ingelheim. Wallentin reports receiving research grants from AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, GlaxoSmithKline, MedCo, Merck, Pfizer, and Schering-Plough.


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.