Apixaban: the (almost) undisputed champion anticoagulant
After COBRRA trial, price
After their arrival in the early 2010s, novel anticoagulants progressively displaced warfarin as first-line agents for long-term anticoagulation for stroke prevention from atrial fibrillation, and later, treatment and prevention of venous thromboembolism (pulmonary embolism and deep venous thrombosis).
The new drugs were equally or more effective than warfarin but had lower rates of bleeding in randomized trials, especially dreaded intracranial hemorrhages.
(Within the decade, makers of two of the blockbuster drugs had paid over $1.4 billion to settle multidistrict lawsuits alleging that they didn’t warn patients about the still-significant bleeding risks.)
Because the drugs were compared individually against warfarin, not each other, their comparative efficacy and bleeding risk could only be extrapolated from their individual noninferiority trials.
While rivaroxaban held the early lead in sales for several years, observational data suggested that apixaban had a lower bleeding risk with comparable efficacy. Clinicians began to choose apixaban far more often; since 2020, its sales have exceeded those of all its major competitors combined (~$14 billion for apixaban in 2025).
Apixaban’s marketers (BMS and Pfizer) have protected their patents and massive revenues, but generic versions of competitors dabigatran and rivaroxaban have become available in various markets.
The Canadian government had a financial interest in learning whether pricey apixaban was truly superior, and funded a large randomized trial.
The COBRRA Trial
At 32 centers in Canada, Australasia, and Ireland, 2,760 patients with acute venous thromboembolism were randomized to treatment with either apixaban or rivaroxaban at usual doses.
Within 3 months, clinically relevant bleeding had occurred less than half as often with apixaban as with rivaroxaban (3.3% vs. 7.1%).
Major bleeding occurred in 32 patients (2.4%) taking rivaroxaban, vs. 5 patients with apixaban.
Major bleeding was defined as in a critical site, associated with a decrease of at least 2 g/dL in hemoglobin, transfusion of two or more units of packed red cells, or contributing to death.
There was no difference in recurrent venous thromboembolism (~1% in both groups).
What Now
Rivaroxaban still retains patent protection in multiple markets, with over $5 billion in sales in 2024.
However, it’s hard to see clinicians voluntarily choosing rivaroxaban over apixaban after the COBRRA trial results, at least for acute VTE treatment.
Some clinicians will be compelled to prescribe rivaroxaban due to preexisting negotiated contracts (manifesting on the clinical side as pharmacy formulary decisions). From a patient safety standpoint, renewing such deals will be difficult to defend, however.
Dabigatran was already running a distant third to apixaban and rivaroxaban in sales. It requires several days of initial injections of enoxaparin (other agents don’t), and has fewer approved indications. A major RCT showed higher rates of GI bleeding with dabigatran than with warfarin, and its renal clearance is slower than with other agents.
Edoxaban is used in Asia and Europe at a lower cost than apixaban. In the absence of proof of higher bleeding rates, its use might still be defended. In the U.S., edoxaban carries a black box warning for patients with high-normal renal function (CrCl >95 mL/min); this subgroup taking edoxaban had higher stroke rates in a major RCT.
In other words, after 16 years of the NOAC wars, apixaban wins.
Expect rivaroxaban to go on sale, especially in the secret discounting deals the makers cut with massive health systems and pharmacies.
This will set up some awkward conversations on pharmacy formulary meetings, in which executive leadership pressure will require middle management to defend the financial decision to use a less-safe anticoagulant. “No margin, no mission” will be said a lot.
In the U.S., Eliquis™ (apixaban) retains patent protection until 2031, according to reports. Although BMS-Pfizer only projects sales out ~2 years, continuing at 2025’s levels would produce an additional $70 billion in revenue for the clot-and-competitor-busting little pill. BMS-Pfizer also have options to extend Eliquis’s patent protection beyond 2031.
Reference
Castellucci LA, Chen VM, Kovacs MJ, et al. Bleeding Risk with Apixaban vs. Rivaroxaban in Acute Venous Thromboembolism. The New England Journal of Medicine. 2026;394(11):1051-1060. doi:https://doi.org/10.1056/NEJMoa2510703