ֱ

Xarelto Flops for COVID-19 Clot Prevention Outside ICU

— ACTION trial adds nuance to the story from large platform trials

MedpageToday

Extended therapeutic-level prophylactic anticoagulation with rivaroxaban (Xarelto) for COVID-19 patients with elevated D-dimer levels wasn't better and carried more bleeding risk than lower hospital-only doses, the ACTION trial showed.

Standard prophylactic-level dosing actually "won" on mortality, duration of hospitalization, and duration of oxygen use through 30 days more often compared with the more intensive regimen, reported Renato Lopes, MD, PhD, of the Duke Clinical Research Institute in Durham, North Carolina, at the virtual American College of Cardiology (ACC) meeting.

The trial randomized 615 patients to a standard approach of in-hospital prophylactic-dose anticoagulation or to a strategy with rivaroxaban (20 mg daily in hospital) if stable or enoxaparin (1 mg/kg twice daily) if unstable while in the hospital then rivaroxaban continued through day 30 whether still in the hospital or discharged home.

In the hierarchical comparison between outcomes among pairs of patients stratified by those categories, extended therapeutic-dose anticoagulation won 34.8% of the time, while the standard approach won 41.3% of the time (win ratio 0.86, 95% CI 0.59-1.22).

Among the secondary endpoints, the therapeutic-dose group had fewer thromboembolic events (7.4% vs 9.9%), but a relative 49% more deaths from any cause (11.3% vs 7.6%), although neither was significant.

Major and other clinically relevant bleeding by International Society on Thrombosis and Haemostasis criteria were a relative 3.64-fold elevated with the higher, longer dose strategy (8.4% vs 2.3%). The elevated risk of any bleeding was likewise statistically significant.

Combining the potential benefit in thromboembolic events and potential harm in mortality and bleeding left the net clinical benefit a toss-up.

ACC session study discussant Robert Harrington, MD, of Stanford University in Palo Alto, California, questioned whether the study had enough patients in the unstable group to make reliable conclusions for them. Lopes agreed that what it adds to the evidence really centers on the potential of an oral anticoagulant in stable patients.

Overall, just 7.4% of patients in the therapeutic-dose group and 5.3% of the standard group were assessed as unstable at baseline. The proportions were similar for severe COVID-19. Results didn't differ significantly for those groups or any other subgroups.

Patients with very high bleeding risk, kidney problems, or a formal indication for therapeutic-dose anticoagulation at baseline were excluded.

While the population was largely similar to those in the non-ICU COVID-19 arms of the REMAP-CAP, ATTACC, and ACTIV-4a trials that recently reported that full-dose prophylactic anticoagulation held substantial benefit regardless of initial D-dimer level, Lopes cautioned about making comparisons.

Rather than being contradictory, he told ֱ, the differences in design were likely at play. Those trials, which were released on the preprint server without peer review this week, stopped therapeutic-dose prophylaxis at discharge and largely used heparin.

The intervention may seem similar in but is "completely different in terms of agent and treatment duration," Lopes noted. "The outcome assessed was also different."

A number of other trials of antithrombotic therapy in COVID-19 are expected to read out over the next several months, he pointed out. "ACTION doesn't give the full amount of answers to this very complex field but clearly adds one more piece to the puzzle about rivaroxaban in therapeutic dose."

Although not very common in the U.S., in many parts of the world rivaroxaban use in this setting had become routine, he noted.

For the best dose, best agent, and best strategy for COVID-19 prophylactic anticoagulation, Lopes said, "I think we're going to have to wait a few more months."

Disclosures

The ACTION trial was funded by an unrestricted research grant from Bayer.

Lopes disclosed relationships with Amgen, Bristol Myers Squibb, GlaxoSmithKline, Bayer, Medtronic, Pfizer, Sanofi-Aventis, Boehringer Ingelheim, Daiichi Sankyo, and Portola.

Harrington disclosed relationships with WebMD, SignalPath (Evidint), and Element Science.

Primary Source

American College of Cardiology

Lopes R "Anticoagulation in patients hospitalized with COVID-19" ACC 2021.

Secondary Source

medRxiv

Lawler PR, et al "Therapeutic anticoagulation in non-critically ill patients with Covid-19" medRxiv 2021; DOI: 10.1101/2021.05.13.21256846.