For A-fib/PCI Combo, Edoxaban Matches VKA but Not Superior: ENTRUST-AF PCI

PARIS, France—Edoxaban (Savaysa; Daiichi Sankyo) plus a P2Y12 inhibitor is noninferior to vitamin K antagonist (VKA)-based triple therapy in terms of bleeding—with no difference in ischemic events—when used in patients with A-fib after successful PCI, the ENTRUST-AF PCI trial shows.

Through 1 year, a composite of major or clinically relevant nonmajor bleeding defined by ISTH criteria occurred in 17% of patients taking the edoxaban-based regimen and 20% of those taking the VKA-based regimen, a difference that met criteria for noninferiority (P = 0.001) but not superiority (P = 0.1154).

The findings were consistent across subgroups and other bleeding definitions, Andreas Goette, MD (St. Vincenz-Hospital, Paderborn, Germany), reported here at the European Society of Cardiology Congress 2019. The findings were published simultaneously online in the Lancet.

“The ENTRUST-AF PCI showed that in patients with atrial fibrillation who underwent successful PCI, a full dose of edoxaban plus once-daily P2Y12 inhibitor is noninferior to triple therapy and that the dual therapeutic regimen as compared to [the] triple therapy-based regimen showed a similar rate with regards to ischemic endpoints,” Goette concluded at a press conference.

Renato Lopes, MD, PhD (Duke Clinical Research Institute, Durham, NC), who served as a discussant following Goette’s presentation, told TCTMD that over the last 5 years the lack of randomized data to guide clinicians in the management of A-fib patients post-PCI has been addressed by trials of three other non-VKA oral anticoagulants (NOACs): PIONEER-AF PCI, RE-DUAL PCI, and AUGUSTUS. ENTRUST-AF PCI is “closing this loop in terms of NOACs in this particular clinical setting,” Lopes said.

He noted that comparisons between the various agents need to be carefully done because the trials in this setting have differed in size, inclusion criteria, and design. In June, his team published a network meta-analysis that included apixaban (Eliquis; Bristol-Myers Squibb), dabigatran (Pradaxa; Boehringer Ingelheim), and rivaroxaban (Xarelto; Bayer/Janssen), showing that NOAC-based dual therapy reduced bleeding without increasing ischemic events. He updated the meta-analysis with the results of ENTRUST-AF PCI and the results were largely unchanged.

The data as a whole do support the concept that dropping aspirin as soon as possible and using a NOAC is the treatment of choice for patients with A-fib undergoing PCI. Dominick Angiolillo

The message that can be taken away is that “today, for these patients, after the first few days of treatment when they present with an ACS or PCI, they’re going to use aspirin for a few days and after those few days I think that the best regimen that we have for bleeding is a NOAC plus a P2Y12 inhibitor without aspirin,” Lopes said. “That’s going to be the regimen that gives us the safest results for bleeding, for intracranial bleeding primarily, without a significant cost in ischemic events in terms of MACE or stroke or stent thrombosis or myocardial infarction.”

Dominick Angiolillo, MD, PhD (University of Florida College of Medicine – Jacksonville), agreed. “The data as a whole do support the concept that dropping aspirin as soon as possible and using a NOAC is the treatment of choice for patients with A-fib undergoing PCI,” he commented to TCTMD, adding that clopidogrel should be the P2Y12 inhibitor of choice. “I think that the overall guidelines on the two sides of the Atlantic are now pretty consistent from this standpoint. You need to use a NOAC and you need to use a NOAC at the stroke prevention dose. That’s very important. And overall, we know that with newer-generation drug-eluting stents for the most part you can drop aspirin rather soon.”

Making Sense of Bleeding

About 15% of patients with A-fib will require stent implantation, creating a tricky scenario in which patients may require both oral anticoagulation and dual antiplatelet therapy with a P2Y12 inhibitor and aspirin, Goette noted. This type of triple therapy has been associated with increased bleeding.

The three prior NOAC trials have evaluated the use of standard or reduced doses of the direct oral anticoagulants combined with a P2Y12 inhibitor in this setting, and together they showed that dropping aspirin significantly improves the safety of antithrombotic therapy without increasing ischemic events.

ENTRUST-AF PCI, conducted at 186 sites in 18 countries, rounds out the group of four trials. Investigators enrolled 1,506 patients (mean age 70 years; 26% women) who had A-fib requiring oral anticoagulation and who had undergone successful PCI for stable CAD or ACS 4 hours to 5 days earlier. Patients were randomized to:

• Edoxaban 60 mg once daily plus a P2Y12 inhibitor (the default was clopidogrel but patients could be switched to prasugrel or ticagrelor) for 12 months. A half dose of edoxaban was used in patients with a creatinine clearance of 15 to 50 mL/min, a body weight of 60 kg or less, or concomitant use of certain P-glycoprotein inhibitors.
• VKA plus a P2Y12 inhibitor for 12 months, as well as aspirin 100 mg once daily for 1 to 12 months.

Clopidogrel was the P2Y12 inhibitor used by more than 90% of patients in both arms. In the VKA group, triple therapy was taken for a median of 66 days.

Edoxaban-based double therapy was noninferior but not superior to VKA-based triple therapy. The investigators noticed, however, that bleeding rates were unexpectedly low in the VKA arm very early on, so they performed a post-hoc landmark analysis at 14 days. That showed that patients on VKA-based triple therapy had lower rates of bleeding compared with those treated with edoxaban-based double therapy in that initial 2-week period, after which bleeding rates remained lower in the NOAC arm.

Digging in deeper to find an explanation, the researchers found that in the first day after PCI, the vast majority of patients in the VKA arm (94%) had subtherapeutic international normalized ratios (INRs). INRs below 2 were seen in 69% of patients from day 2 to 7 and in 42% during the second week. Nevertheless, overall time in therapeutic range was high at 63%, Goette said.

Lack of Superiority a Problem?

Lopes stressed that even though a common message can be taken away from the four NOAC trials, ENTRUST-AF PCI stands out from rest in that edoxaban-based double therapy did not prove superior to VKA-based triple therapy in terms of bleeding, unlike the other NOAC-based regimens. “You want to be superior. You have to be superior because you’re comparing two drugs versus three drugs, and [if] you’re not superior then I think your results are less meaningful, less strong if you will,” he said.

Asked whether that will matter to clinicians, he said it remains to be seen. “How much this is going to influence decisions is difficult to say because the [ENGAGE AF-TIMI 48 trial], which is the pivotal A-fib trial, showed tremendous safety of edoxaban compared to warfarin,” he said.

Lopes said the authors tried to explain the lack of superiority on the primary bleeding endpoint with poor control of warfarin early on, adding that he doesn’t think that explains it fully because after first week, VKA therapy was reasonably well controlled and there were still low rates of bleeding.

The relative lack of bleeding in the VKA arm could have been the play of chance, but the failure to reach superiority will likely contribute to physician decision-making when it comes to selecting a NOAC for patients with A-fib after PCI, along with considerations about the rigor of the various trial designs and the doses used, Lopes said.

Angiolillo said that for patients already taking edoxaban, “there are no concerns with continuing with the strategy if, for example, they were to undergo PCI. Perhaps it would be less exciting for a physician to start treatment with edoxaban if their diagnosis of A-fib is also at the time of PCI simply because the trial did not have as robust findings as some of the other ones.”

However, because uptake of edoxaban has lagged behind that of the others NOACs in the United States—due to its entry behind the other three onto the market—"I don’t think it will have a huge impact for US practice,” he predicted.

For his part, Goette didn’t see the lack of superiority having an impact on choice of NOAC. “Actually, I don’t think so, because we can explain the mechanism,” he said during the press conference.

How Soon Is Too Soon to Drop Aspirin?

The rate of overall ischemic events—a composite of CV death, stroke, MI, definite stent thrombosis, or systemic embolic events—was not significantly different between groups (7% in the edoxaban arm and 6% in the VKA arm; HR 1.06; 95% CI 0.71-1.69). The individual components occurred a similar rates as well.

Goette pointed out, however, that there were numerically increased rates of MI and stent thrombosis with early withdrawal of aspirin, as has also been seen in the A-fib/PCI trials of the other NOACs.

That consistent result “raises concerns,” the investigators write in their Lancet paper. “This finding might be attributed to response variability and efficacy of clopidogrel as a single antiplatelet drug when the carryover effect from the periprocedural antiplatelet therapy (including aspirin) is over.”

Though analyses are ongoing, Goette said, “this really suggests that stopping just after single dose or single shot of aspirin, that that is definitely too early to stop aspirin. At least from the data we have [preliminarily], it really suggests that in everybody, a longer treatment for 1 or 2 weeks of aspirin may be very appropriate.”

It is possible that in selected patients with selected types of procedures, we might need to keep aspirin a little bit longer. We just don’t know yet how much longer. Renato Lopes

Lopes told TCTMD he’s working on an analysis that addresses the question of how soon aspirin can be stopped in patients with A-fib after PCI. But what can be said now, he stated, is the benefit of continuing aspirin to prevent rare events like stent thrombosis is outweighed by the bleeding hazard. Even so, there may be a subset of patients in whom the ischemic risk is high enough to warrant use of aspirin for a little bit longer, and research into identifying those patients and determining the appropriate duration of aspirin use is ongoing, Lopes said.

“Again, I think the overall message is that for most patients it’s okay to drop aspirin and the risk of bleeding is much higher than the risk of ischemic events,” he said. “But it is possible that in selected patients with selected types of procedures, we might need to keep aspirin a little bit longer. We just don’t know yet how much longer.”

Angiolillo said the data from the NOAC trials need to be digested further to help answer questions about aspirin duration, and in particular, analyses looking at the timing of ischemic events are needed.

An editorial co-authored by Angiolillo and Davide Capodanno, MD, PhD (AOU Policlinico-Vittorio Emanuele, Catania, Italy), offers a suggestion.

“Use of dual antithrombotic therapy does not imply abandoning aspirin in the peri-PCI period, and a weaning period should be considered as was done in the aforementioned trials (ie, between discharge up to 2 weeks). For patients at higher thrombotic potential, such as those undergoing complex PCI or with an acute coronary syndrome, prolonging the duration of aspirin or considering more potent P2Y12 inhibition can be envisaged on the basis of careful weighing of bleeding and ischemic risk at the individual level,” they advise.