COMPASS Trial

Manesh Patel, MD: Chris, can I ask you about the COMPASS trial and how you think about the results? Can you remind us about that?

Christopher Granger, MD:COMPASS was an amazing trial, it was a home run. It was 27,000 patients who had vascular disease, coronary disease, peripheral disease, and that case was lumped to carotid disease, symptomatic leg disease, and coronary disease with an ABI [ankle-brachial index] less than 0.9. So asymptomatic coronary disease plus some degree of asymptomatic vascular disease. There were 3 arms, but if we focus on the 2.5 mg of rivaroxaban twice a day plus aspirin arm, that had a 24% advantage in terms of MACE [major adverse cardiac event] compared to aspirin. The remarkable thing was for the peripheral arterial disease [PAD] population, for the first time with a medical therapy, there was a greater relative risk reduction in MACE, a 28% relative risk reduction in MACE, and almost a 50% reduction in MALE, major adverse limb events, including amputation and revascularization. There was a 42% reduction in stroke, and an even bigger reduction in stroke in the PAD population.

Manesh Patel, MD: They included people with carotid disease.

Christopher Granger, MD:Carotid disease was included, and there was an 18% reduction in all-cause mortality. They were impressive overall results, and we can get to the specifics of the PAD population. This was impressive especially in the PAD population for 3 reasons, the risk is higher, the relative risk reduction was greater, the point estimate for the relative risk reduction. And there’s nothing else that’s been shown to have such a dramatic effect in reducing vascular events, the PAD events.

Manesh Patel, MD: No. If you would have said to me before the COMPASS trial started, is it going to be positive? I would have said I’m not sure, because it’s a big swing. It’s a swing at all of vascular disease, coronary plus PAD. The doses were all studied through the ATLAS trial and other things. I wasn’t involved in the trial, but I remember I was thinking, “That’s a hard study.” I told John Eikelboom, [MD,] “Good luck buddy, that won’t be easy.” And then, of course, I had to buy him a beer later when they stopped it early for a mortality reduction. I was like, “Wow, maybe I need to rethink my understanding of biology.” The trial produced powerful findings, but those are chronic vascular patients, and I agree with you, the patients with PAD in the COMPASS trial are unique. They all have coronary disease, so they’re patients with polyvascular PAD; they’re the highest risk patients with PAD. Some even have heart failure and renal dysfunction, so it was probably the highest risk. Larry, how did you take the findings? Did anything dramatically stick out at you from the trial? It’s been difficult to get people to change an antiplatelet life of aspirin only to go to low-dose rivaroxaban plus aspirin. How do you think about it?

Larry Allen, MD: Like you and Chris said, it’s a home run study, it’s quite impressive. I do wonder, there are some trials that are clearly different, but like in the heart failure community, the COMMANDER trial is done looking at rivaroxaban, a high-risk population as well. It’s very different, not polyvascular, but many patients with significant heart failure, coming out of the hospital, you’d be worried. And in that study, you see tiny differences in ischemic events, maybe offset by bleeding, and no difference in mortality. I don’t know whether it’s differences in patient population or what you’re treating, sometimes one study within the confidence intervals was one way and one was the other. There’s some uncertainty in the degree of benefit, but yes, it’s hard. The other thing is, in the heart failure community, if you have somebody who has heart failure, you’re already talking about a lot of drugs, especially in the low EF [ejection fraction] population. You consider how they have a little vascular disease, you’re treating them with a beta-blocker and sacubitril/valsartan and spironolactone, and now we have an SGLT2 inhibitor, do I add additional anticoagulant therapy to their regimen? It’s easier if they have AF [atrial fibrillation], which is in 40% of the population, but it gets complicated to try to weigh all these things. Will you do them all, and when do you start, and how do you sequence them? And how does COMPASS weigh into that when it’s a different population than I’m treating at that point?

Manesh Patel, MD: That’s right. The COMPASS population is better aimed at that patient with polyvascular PAD. Like you said, what’s the phenotype that I’m going to think that’s the therapy I want to give? When I think about trials and the ones we’ve discussed, it’s the population, the biology I’m trying to interrupt, and does that biology lead to a modifiable end point? Does that modifiable end point then lead to something that can have meaningful things? Like in COMMANDER, it’s hard to modify some of those heart failure end points in those patients. Anticoagulation at any level is going to be tough, unless there are other issues. In vascular disease, whether it’s lower extremity vascular diseases, the patient’s phenotypically predominantly presenting with vascular disease in the lower extremity or others.

There are 2 points to your question, which is great. One is, how sick are the patients in COMPASS who got the therapy? You’re right in that the PAD was very evident because they either got revascularization or had symptomatic claudication. Although one could argue in most vascular clinics, that’s most of them, so I agree with you, we shouldn’t be screening. There were some patients with CAD [coronary artery disease] who got in with the risk enhancement criteria that was an ABI that was low. It’s probably a sicker patient because they’re polyvascular, coronary, and peripheral.

Christopher Granger, MD: That was 1500 patients, and the point estimate for the benefit was equal in that group. But I agree with you, that the place we should concentrate on using this therapy is in patients who have symptomatic peripheral artery disease. The carotid disease group was fascinating. The point estimate was the most impressive for patients with carotid disease, but that’s another set of issues when we get into the stroke prevention area.

Manesh Patel, MD: I want to be clear, I wasn’t advocating screening for PAD because there are enough data to say, therefore, you should be using it. But there are enough patients with highly symptomatic PAD with claudication or patients undergoing revascularization that we should start there because they’re the highest risk. However, knowing vascular risk for a patient might be different, but that’s the first conversation. Larry and I were pushing and pulling a little, because you’re right, screening anything is hard to do. It’s based on vascular disease, and we tend not to care about it as much as cancer, for example, where we’ll screen for diseases differently than we will for vascular disease.

This transcript has been edited for clarity.