Larger Absolute Rivaroxaban Benefit in Diabetes: COMPASS

March 29, 2020

In the COMPASS trial of patients with stable coronary or peripheral artery disease (PAD), the combination of aspirin plus rivaroxaban, 2.5 mg twice daily, provided a larger absolute benefit on cardiovascular endpoints — including a threefold greater reduction in all-cause mortality — in patients with diabetes compared with the overall population.

The results of the diabetes subset of the COMPASS trial were presented by Deepak Bhatt, MD, Brigham and Women's Hospital Heart & Vascular Center, Boston, Massachusetts, on March 28 at the "virtual" American College of Cardiology 2020 Scientific Session (ACC.20)/World Congress of Cardiology (WCC). They were also simultaneously published online in Circulation.

"Use of dual pathway inhibition with low-dose rivaroxaban plus aspirin is particularly attractive in high-risk patients such as those with diabetes," Bhatt concluded.

The COMPASS trial was first reported in 2017 and showed a new low dose of rivaroxaban  (2.5-mg twice-daily; Xarelto, Bayer/Janssen Pharmaceuticals) plus aspirin, 100 mg once daily, was associated with a reduction in ischemic events and mortality and a superior net clinical benefit, balancing ischemic benefit with severe bleeding, compared with aspirin alone for secondary prevention in patients with stable atherosclerotic vascular disease. 

But clinicians have been slow to prescribe rivaroxaban in this new and very large population.

"It's been more than 2 years now since main COMPASS results, and there isn't a sense that this therapy has really caught on," chair of the current ACC session at which the diabetes subgroup results were presented, Hadley Wilson, MD, Sanger Heart and Vascular Institute, Charlotte, North Carolina, commented:

He asked Bhatt whether the diabetes subgroup may be "the tipping point that will make people aware of rivaroxaban and then that may trickle down to other patients."

Bhatt said that he hoped that would be the case. "We as a steering committee of this trial could say the results were positive so rivaroxaban should now be used in everyone with stable coronary or peripheral arterial disease, but that is impractical and as you out point out it hasn't happened," he replied. 

"In PAD/vascular medicine we have embraced this new therapy. In the broader cardiology world there are a lot of patients with stable coronary arterial disease at high ischemic risk who could take rivaroxaban, but its use is bound to be limited by it being a branded drug and the fact that there is a bleeding risk," Bhatt explained.  

"I think we need to identify patients the highest ischemic risk and focus drugs such as these with a financial cost and a bleeding risk on those who most likely will derive the greatest absolute reduction in risk," he said. "The PAD subgroup is one group where this is the case, and now we have shown the diabetes subgroup is another. And there is no incremental bleeding risk in this group over the whole population, so they get a much greater benefit without a greater risk. I hope this helps get rivaroxaban at the new lower dose used much more often."

A total of 18,278 patients were randomly assigned to the combination of rivaroxaban and aspirin or aspirin alone in the COMPASS trial. Of these, 6922 had diabetes mellitus at baseline and 11,356 did not have diabetes.

Results from the current analysis show a consistent and similar relative risk reduction for benefit of rivaroxaban plus aspirin vs placebo plus aspirin in patients both with and without diabetes for the primary efficacy endpoint, a composite of cardiovascular death, myocardial infarction (MI), or stroke, with a hazard ratio of 0.74 for patients with diabetes and 0.77 for those without diabetes, the researchers report.

Because of the higher baseline risk in the diabetes subgroup, these patients had numerically larger absolute risk reductions with rivaroxaban than those without diabetes for the primary efficacy endpoint at 3 years (2.3% vs 1.4%) and for all-cause mortality (1.9% vs 0.6%).

These results translate into a number needed to treat (NNT) with rivaroxaban for 3 years to prevent one CV death, MI, or stroke of 44 for the diabetes group vs 73 for the nondiabetes group; the NNT to prevent one all-cause death was 54 for the diabetes group vs 167 for the nondiabetes group, the authors write.  

Because the bleeding hazards were similar among patients with and without diabetes, the absolute net clinical benefit (MI, stroke, cardiovascular death, or bleeding leading to death or symptomatic bleeding into a critical organ) for rivaroxaban was "particularly favorable" in the diabetes group (2.7% fewer events in the diabetes group vs 1.0% fewer events in the nondiabetes group), they add.

Panelist at the ACC Featured Clinical Research session at which these results were presented, Jennifer Robinson, MD, University of Iowa College of Public Health, Iowa City, asked Bhatt how clinicians were supposed to decide which of the many new agents now becoming available for patients with stable coronary artery disease to prescribe first.

"We often forget about rivaroxaban when we're thinking about what to add next for our secondary prevention patients," she said. "You also led the REDUCE-IT trial showing benefit of icosapent ethyl, icosapent ethyl icosapent ethyl icosapent ethyl and there is also ezetimibe, PCSK9 inhibitors and SGLT2 inhibitors. For your patients with coronary disease who are already on a high dose statin which one of these would you add next?"

"That is what physicians need to ponder all the time," Bhatt replied. "And when a patient has several risk factors that are not well controlled, I guess it's all important. I go through a checklist with my patients and try and figure what they're not on that could further reduce their risk."

"In the COMPASS trial there was an overall positive result with rivaroxaban in the whole population. And now we have shown that patients with diabetes had an even greater absolute risk reduction. That pattern has also been seen with other classes of agents including the statins, PCSK9 inhibitors, and icosapent ethyl," Bhatt noted.

"In patients with diabetes, I will usually target whatever is standing out most at that time. If their glycemic control is completely out of whack, then that is what I would focus on first, and these days often with a SGLT2 inhibitor or GLP-1 agonist. If the LDL was out of control, I would add ezetimibe or a PCSK9 inhibitor. If the triglycerides were high, I would add icosapent ethyl. If multiple things were out of control, I would usually focus on the number most out of kilter first and try not to forget about everything else."

But Bhatt noted that the challenge with rivaroxaban is that there is no test of thrombosis risk that would prompt the physician to take action. "Basically, the doctor just has to remember to do it. In that regard I would consider whether patients are at low bleeding risk and are they still at high ischemic risk despite controlling other risk factors and, if so, then I would add this low dose of rivaroxaban."

Another panel member, Sekar Kathiresan, MD, asked Bhatt whether he recommended using available scores to assess the bleeding/thrombosis risk/benefits of adding an antithrombotic.

Bhatt replied: "That's a terrific question. I guess the right answer is that we should be doing that, but in reality I have to concede that I don't use these scores. They have shown appropriate C statistics in populations, but they are not fantastic in individual patients."

"I have to confess that I use the eyeball test. There is nothing as good at predicting future bleeding as past bleeding. So if a patient has had bleeding problems on aspirin alone I wouldn't add rivaroxaban.  But if a patient hasn't bled before, especially if they had some experience of dual antiplatelet therapy, then they would be good candidates for a low vascular dose of rivaroxaban," he said.  

"It is not as easy as with other drugs as there is always a bleeding trade-off with an antithrombotic. There is no such thing as a free lunch. So patients need careful assessment when considering prescribing rivaroxaban and regular reassessment over time to check if they have had any bleeding," he added.

The COMPASS study was funded by Bayer. Bhatt reports honoraria from Bayer via the Population Health Research Institute for his role on the COMPASS trial and other research funding from Bayer to the Brigham & Women's Hospital.  

American College of Cardiology 2020 Scientific Session (ACC.20)/World Congress of Cardiology (WCC). Abstract 20-LB-20544-ACC. Presented March 28, 2020.

Circulation. Published online March 28, 2020.  Full text 

For more from | Medscape Cardiology, follow us on Twitter and Facebook.


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.