From the Journals

Rivaroxaban trends toward higher thrombotic risk than vitamin K antagonists in APS

View on the News

Another APS study fails to show noninferiority of rivaroxaban

The recent trial by Ordi-Ros et al. revealed similar findings to a previous trial, TRAPS, by Pengo et al., which compared rivaroxaban with warfarin among patients with thrombotic antiphospholipid syndrome and triple positivity for antiphospholipid antibodies. Despite the caveat that TRAPS was prematurely terminated, in both studies, a higher proportion of patients in the rivaroxaban group than the vitamin K antagonist group had thrombotic events, most of which were arterial, whether considering MI or stroke. Furthermore, both studies did not show noninferiority of rivaroxaban versus dose-adjusted vitamin K antagonists.

The reasons for this failure of noninferiority remain unclear.

Denis Wahl, MD, PhD, and Virginie Dufrost, MD, are with the University of Lorraine, Nancy, France, and the Centre Hospitalier Universitaire de Nancy. No conflicts of interest were reported. His remarks are adapted from an accompanying editorial (Ann Intern Med. 2019 Oct 15. doi: 10.7326/M19-2815).



Patients with thrombotic antiphospholipid syndrome (APS) may have a greater risk of recurrent thrombosis when taking rivaroxaban instead of a vitamin K antagonist, suggests a recent trial conducted in Spain.

Stroke was also more common among those taking rivaroxaban, while major bleeding was slightly less common, reported lead author Josep Ordi-Ros, MD, PhD, of Vall d’Hebrón University Hospital Research Institute in Barcelona, and colleagues in Annals of Internal Medicine.

“Two randomized, controlled trials comparing rivaroxaban with warfarin suggested that rivaroxaban may be efficacious in patients with previous venous thromboembolism who are receiving standard-intensity anticoagulation but showed an increased thrombotic risk in those with triple-positive antiphospholipid antibodies,” the investigators wrote. However, they also noted that these findings required a cautious interpretation because of study limitations, such as premature termination caused by an excess of study events and the use of a laboratory surrogate marker as a primary outcome.

To learn more, the investigators performed an open-label, phase 3 trial involving 190 patients with thrombotic APS. Patients were randomized in a 1:1 ratio to receive either rivaroxaban (20 mg per day, or 15 mg per day for patients with a creatinine clearance of 30-49 mL/min per 1.73 m2) or an adjusted dosage of vitamin K antagonists (target international normalized ratio of 2.0-3.0, or 3.1-4.0 for those with a history of recurrent thrombosis).

Patients underwent evaluations every month for the first 3 months and then every 3 months thereafter, each of which involved a variety of laboratory diagnostics such as checks for antinuclear antibodies and lupus anticoagulant, among others. Statistical analyses aimed to determine if rivaroxaban was noninferior to therapy with vitamin K antagonists based on parameters drawn from previous meta-analyses, as no studies had compared the two types of treatment when the present study was designed.

After 3 years of follow-up, almost twice as many patients in the rivaroxaban group had experienced recurrent thrombosis (11.6% vs. 6.3%), although this finding lacked statistical significance for both noninferiority of rivaroxaban (P = .29) and superiority of vitamin K antagonists (P = .20). Still, supporting a similar trend toward differences in efficacy, stroke was more common in the rivaroxaban group, in which nine events occurred, compared with none in the vitamin K antagonist group. In contrast, major bleeding was slightly less common with rivaroxaban than vitamin K antagonists (6.3% vs. 7.4%).

“In conclusion, rivaroxaban did not demonstrate noninferiority to dose-adjusted vitamin K antagonists for secondary thromboprophylaxis in patients with thrombotic APS,” the investigators wrote. “Instead, our results indicate a recurrent thrombotic rate that is nearly double, albeit without statistical significance.”

The study was funded by Bayer Hispania. One coauthor reported additional relationships with Pfizer, Lilly, Janssen, and others.

SOURCE: Ordi-Ros J et al. Ann Intern Med. 2019 Oct 15. doi: 10.7326/M19-0291.

Next Article: