ORLANDO – Apixaban outperformed both rivaroxaban and dabigatran in a retrospective, observational study of real-world prescribing of direct oral anticoagulants in nearly 163,000 U.S. patients with nonvalvular atrial fibrillation, Steven B. Deitelzweig, MD, reported at the annual meeting of the American College of Cardiology.

This ongoing study, known as ARISTOPHANES (Anticoagulants for Reduction in Stroke: Observational Pooled Analysis on Health Outcomes and Experience of Patients), is the largest real-world analysis of direct oral anticoagulants (DOACs) to date. Unlike most of the previous observational studies of DOACs, which used a single insurance claims database, ARISTOPHANES pools data from Medicare and four large U.S. commercial insurance claims databases that collectively cover more than 180 million Americans.

In this way, investigators were able to assemble a study population of 162,707 patients on DOACs for stroke prevention in atrial fibrillation who were propensity score–matched for 30 variables, including demographics and comorbidities, in order to control for potential confounders, explained Dr. Deitelzweig, chair of hospital medicine and vice president of medical affairs at the Oschner Medical Center, New Orleans.

This was a study of real-world prescribing. Unlike in randomized trials, where everyone is on a standard-dose DOAC, lower-dose therapy was common. It was prescribed for 21% of patients on apixaban, 15% on dabigatran, and 24% on rivaroxaban.

ARISTOPHANES results

The biggest difference in outcome was between patients on apixaban and those on dabigatran. The incidence of stroke/systemic embolism in 27,096 patients on apixaban (Eliquis) was 1.01% in 360 days, for a statistically significant 31% reduction in risk relative to the 1.42% incidence rate in 27,096 extensively matched patients on dabigatran (Pradaxa). The 360-day incidence of major bleeding was 2.7% in the apixaban group and 3.3% in those on dabigatran, for a 23% relative risk reduction. In the apixaban/rivaroxaban comparison, which included 62,619 patients in each group, the incidence of stroke/systemic embolism was 1.21% with apixaban and 1.42% with rivaroxaban, for a 27% relative risk reduction. Major bleeding occurred in 3.1% of the apixaban group and 5.3% of those on rivaroxaban, for a 46% reduction in risk favoring apixaban.

In a comparison of 27,538 patients on dabigatran and an equal number of rivaroxaban, the 360-day cumulative incidence of stroke/systemic embolism was 1.40% with dabigatran versus 1.23% with rivaroxaban, while the major bleeding rate was 3.28% in the dabigatran group and 4.76% with rivaroxaban.

Dr. Deitelzweig was quick to acknowledge the major limitation of ARISTOPHANES.

“Only associations can be drawn from a nonrandomized, retrospective, observational study, not conclusions regarding causality, even though the cohorts were matched using propensity scoring,” he emphasized.

The critique

Session cochair Jeanne E. Poole, MD, professor of medicine and director of the clinical cardiac electrophysiology program at the University of Washington, Seattle, commented, “The problem, of course, with using these large databases is that you may not be able to find out important information, such as whether rivaroxaban was being taken appropriately with meals, which is frequently not the case. If it wasn’t, that decreases absorption and efficacy by 40%. That’s a limitation.”

Audience member James A. Reiffel, MD, rose to add that, in his view, another significant limitation of all real-world, observational analyses using claims data is that it’s not possible to know why physicians selected a given drug for a given patient. He used as an example a patient with atrial fibrillation and gastroesophageal reflux disease (GERD).

“People with GERD may be less likely to get dabigatran, and if they’re less likely to get dabigatran with GERD, maybe they’re less likely to get a GI bleed. I don’t know,” said Dr. Reiffel, professor of clinical medicine and director of the electrocardiography laboratory at Columbia University Medical Center, New York.

“We have to take all the real-world analyses with a little grain of salt,” he added.

Dr. Deitelzweig replied, “This study is not meant to be the be-all and end-all.”

That being said, he noted that although randomized trials have shown that DOACs are at least as effective and safe as warfarin for stroke prevention in atrial fibrillation, there have been no randomized, head-to-head clinical trials comparing them, nor are any such studies likely to be done for the foreseeable future. Yet physicians and their patients are hungry for comparative effectiveness data, even if it doesn’t rise to the status of level I evidence.

ARISTOPHANES is sponsored by Bristol-Myers Squibb and Pfizer. Dr. Deitelzweig reported serving as a consultant to Pfizer and receiving research grants from Bristol-Myers Squibb and Portola.

bjancin@mdedge.com

SOURCE: Deitelzweig SB. ACC 2018.

Apixaban prevails in study of 163,000 DOAC users