Will OCEANIC-AF Sink Factor XI Inhibitors?

This transcript has been edited for clarity. 
Michelle L. O’Donoghue, MD, MPH: Hi. This is Dr Michelle O’Donoghue, reporting for Medscape. I’m here at the European Society of Cardiology (ESC) Congress in London, and one of the more hotly anticipated presentations is that of OCEANIC-AF. This is a trial of factor XI inhibition in patients with atrial fibrillation. 
Joining me to discuss the results is Dr Manesh Patel. He is division chief at Duke Department of Medicine. Thanks for joining me, Manesh. 
Manesh R. Patel, MD: Thanks for having me. 
O’Donoghue: Let’s start by talking about the theory behind factor XI inhibition. 
Patel: Our hope always is the holy grail of trying to get to stopping clots and not bleeding, or what we’ve discussed as uncoupling thrombosis from hemostasis.
The history of our practice here in the coagulation cascade is for conditions like atrial fibrillation; we used warfarin, which affected many spots along the coagulation cascade. The good news is it certainly stopped clots, but there’s a lot of bleeding. There were significant advances, I’ll say, with direct thrombin or factor X inhibitors, like edoxaban, rivaroxaban, and apixaban, that inhibited that and improved it.
There are a few types of factor XI inhibitors. OCEANIC-AF used something called asundexian, a small molecule affecting factor XIa inhibition — so activated factor XI. What happens in the coagulation cascade is you have a common pathway, and then you have a contact pathway, which is thrombin amplification.
The idea is that if you inhibit factor XI, you might still have some normal hemostasis and clot formation that you need to stop bleeding, but you would stop the propagation of thrombus to get to that pathologic thrombus. That was the biologic concept. I often say it’s our understanding of the current cartoon of thrombosis and hemostasis, but it’s an important biologic process. That was the hope.
O’Donoghue: As you say, in many ways, this has been the holy grail because the direct-acting oral anticoagulants (DOACs) were, of course, a big advance in the field of atrial fibrillation, anticoagulation management, and anticoagulation in general. Bleeding is still a significant concern, and certainly gastrointestinal (GI) bleeding has been one of those.
The appeal of this class has been that idea that potentially you could have your cake and eat it too — that you could prevent clot formation without having the excess in bleeding. 
Patel: Yeah, that’s right. That’s the way almost all of our antithrombotic steps forward have been. If warfarin was leading to more bleeding, then the DOACs actually beat warfarin by leading to fewer strokes in atrial fibrillation, but much of it was hemorrhagic stroke and less bleeding. 
The concept for factor XIa inhibition with asundexian was first done in healthy volunteers, but then it was studied in some phase 2 studies to get ready for the atrial fibrillation study.
The phase 2 studies included PACIFIC-AF, where we aimed to understand how it looked in relation to apixaban in patients with atrial fibrillation. PACIFIC-AF was a short-term study looking at bleeding with two doses of asundexian, 20 mg and 50 mg, and looking at a factor XI activity assay to say, did you inhibit factor XI activity as you intended a nd then was there a difference in bleeding compared with apixaban? 
What was previously presented and published with PACIFIC-AF showed that at trough, about 92% of factor XI activity was inhibited with the 50-mg dose, and a little less with the 20-mg dose. Overall, bleeding was significantly less than with apixaban. The 50-mg dose was chosen to go into a large phase 3 study that we have to do to see if we can reduce stroke, systemic embolism, and the combination of that benefit for bleeding. 
O’Donoghue: I think you’re highlighting the real challenge, which is about dose selection, ultimately, for a pathway that is not yet tested in terms of knowing its clinical efficacy.
As you said, you had the factor XIa activity data, suggesting 92% inhibition of XIa at trough, as well as 94% at peak?
Patel: Yeah. The thinking is that we’ve got a reasonably high inhibition. There wasn’t a huge amount of difference, and then the idea that there’s significantly less bleeding. Not only is dose selection important, we were trying to do it in a population that we would be testing against a comparator that we thought we would be using in phase 3.
There are no perfect ways to pick doses. I often say, I wish our bodies behaved like we learned in medical school with pharmacokinetics and pharmacodynamics (PK/PD) and we’d never have to do these trials. As we take them out into 74-year-old patients with atrial fibrillation and not normal kidney function, etc., that’s sort of the action we’re trying to do. 
O’Donoghue: Let’s talk about OCEANIC-AF.
Patel: OCEANIC-AF, unfortunately, was stopped early for inferiority. I’ll go through the trial design and talk about what we learned. It’s a trial that was designed to evaluate asundexian, a small molecule for factor XIa inhibition, at this 50-mg dose, compared with apixaban in atrial fibrillation patients at risk for stroke or systemic embolism. 
We defined that as a CHA2DS2-VASc score in men ≥ 3 or in women ≥ 4. If it was men with a score of 2 or women with a score of 3, based on the guidelines at that time, you could have other predefined risk factors, such as age, renal dysfunction, prior GI bleeding, or antiplatelet use, or being anticoagulant naive (ie, not having had an anticoagulant for 6 weeks).
That was the design. It was anticipated to be a 1:1 randomization, double-blind study, looking for the primary efficacy endpoint of stroke or systemic embolism and being noninferior to apixaban at least, and then also looking to see if we have enough power, which we designed to see if it was superior for bleeding or net clinical benefit.
The trial was designed to enroll 18,000 patients. Actually, we rapidly enrolled over 14,000 patients in 11 months, from December 2022 to November 2023. It tells you about the desire among people to be on a better anticoagulant or a different one, or about the concerns that still exist in clinical practice around bleeding.
Unfortunately, the data monitoring committee contacted us on November 19, 2023, and said, we see a hazard with asundexian increasing the rate of stroke or systemic embolism. Therefore, the trial was stopped for inferiority. 
O’Donoghue: How great was the excess in stroke and systolic embolic events? 
Patel: There are a couple of important findings on that I want to make sure are clear. You know, anytime a trial is stopped early — this was a median exposure of about 160 days — it’s hard to know what the final effects would have been. It can sometimes overemphasize or underemphasize findings. The cumulative incidence for stroke was 1.3% with asundexian and 0.4% with apixaban. The hazard ratio was around 3.7 for asundexian compared with apixaban. 
We presented the trial and described two key sets of data that we looked at to understand that. The first is that we did have a planned PK/PD analysis. We’re still doing analysis on that. It shows that at trough, in OCEANIC-AF as in PACIFIC-AF, that level of 92% inhibition was still present. 
We don’t have many patients with events at 30 days because the trial was stopped early. There were 10 patients who had a stroke or systemic embolism in the control group, and those rates seem to be similar between those two groups.
Then we did a subgroup analysis to see if there were subgroups that might explain this or other things that we saw. The one group we had prespecified that had an interaction was whether patients were previously treated or not. For those who were naive to an oral anticoagulant, the curves look much closer together. The hazard ratio there was 1.45. 
This is prespecified, exploratory, and hypothesis generating — all those words we use to say, don’t overinterpret this. What it probably tells us is there’s some effect at this dose of factor XI activity compared with apixaban, but not enough. Those are the main conclusions: It’s inferior to apixaban for stroke or systemic embolism.
In all, 83% of these patients were previously treated and well treated. The rates of stroke or systemic embolism were lower than we anticipated on apixaban. That’s something we have to work on. I think it moves the field. I want to be clear. I think it moves the field forward. 
O’Donoghue: How so? 
Patel: I think what we learned first is that this dose of this drug is inferior to apixaban — no question about that. It may be that you need to have near-complete inhibition of factor XI activity. That is a dose or pathway conversation, but it’s likely to say you might need more. 
The second is that in the population we studied, there’s a price to pay if you don’t enroll more anticoagulant-naive patients or patients that are not chronically on an anticoagulant and stable on that. It might be that there’s a price to pay. These are well-treated patients who have low event rates, so you want to at least make sure you enroll some patients that are not chronically stable on the therapy. 
Lastly, at this dose or others, whenever you change therapies, there’s a risk and you have to be careful about that.
O’Donoghue: I think the dose question is an important one. I’m sure after your presentation, people will be grappling with the question, is this the end of the factor XI/XIa story, or what can we really learn from this? 
I think what I’m grappling with is, as you say, we see the 92%-94% suppression of XIa activity. Yet, when I look at the excess in risk for ischemic stroke, I’m struck that it’s almost as though it was behaving as a placebo. I would have thought at first blush with that degree of inhibition, perhaps it’s not quite enough, agreed. I still would not have expected to see the excess that was ultimately observed.
Now, who knows, of course, had the trial been conducted to completion, whether that would have been to some extent attenuated. What are your thoughts there? 
Patel: I think the story is that it’s just the beginning. I think there is activity. The reason that the excess is better than placebo — which is, I think, the question you’re asking — is when we look at those around 2400 patients that were not previously treated with an anticoagulant, the curves are closer together. It’s still not sufficient, but that hazard ratio of 1.45 at least makes me know that there’s activity and therapeutic opportunity. 
The second, and I want to be very clear, is the new anticoagulants in atrial fibrillation rarely are significantly better at reducing ischemic stroke [compared with warfarin]. They’re better at reducing bleeding in the brain and the overall complexity of what happens to patients, not just with stroke but also with staying on therapy and not stopping the drug because of bleeding.
I think the pathway for factor XI is going to be a net clinical benefit pathway, where when you get the right dose in this group and the right population, you get to a place where the hazard ratio is near 1 for stroke and the bleeding reduction is so significant that patients stay on it and get a benefit.
I think that is a pathway that is possible. In this trial, all we can speak to is what we saw with this dose in this total population. It’s clearly inferior. Whether it’s like what placebo would have done, we don’t know, because there wasn’t a placebo arm. We all speculate what we think the rate would be, but we wouldn’t have speculated some of those rates even in the apixaban group.
I think there’s a mixture of not just the rate we see, but how well the control groups also do. 
O’Donoghue: It’s great to see that ultimately there are still ongoing trials in this space. Those will be ultimately, of course, the true test.
Patel: They’re well-run trials that I think will move the field forward. 
O’Donoghue: Congratulations on a well-run trial. I think it still informs importantly upon the space in general.
Patel: Thanks for having me. 
O’Donoghue: Signing off for Medscape, this is Dr Michelle O’Donoghue.
Michelle O’Donoghue is a cardiologist at Brigham and Women’s Hospital and senior investigator with the TIMI Study Group. A strong believer in evidence-based medicine, she relishes discussions about the published literature. A native Canadian, Michelle loves spending time outdoors with her family but admits with shame that she’s never strapped on hockey skates.