ANNEXA-4

Published

Connolly SJ, et al. "Andexanet Alfa for Acute Major Bleeding Associated with Factor Xa Inhibitors". The New England Journal of Medicine. 2016. 375(12):1131-1141.
PubMed • Full text • PDF • ClinicalTrials.gov

Clinical Question

In adults with potentially life-threatening acute major bleeding, what is the efficacy and safety of andexanet alfa for reversing the anticoagulant effects of the factor Xa inhibitors apixaban, rivaroxaban, edoxaban, or enoxaparin?

Bottom Line

In this single-arm trial, andexanet alfa administration decreased anti-Xa activity and achieved hemostasis in a majority of patients, and was associated with a high rate of thrombosis and death.

Major Points

The direct oral anticoagulants (DOACs) including dabigatran, apixaban, and rivaroxaban are safe and effective alternatives to warfarin among patients with thromboembolism and nonvalvular atrial fibrillation, as demonstrated in RE-LY, AMPLIFY, ROCKET AF, and other large trials. While more convenient due to fixed dosing rather than requiring dose adjustments based on INR, the DOACs have been limited by the lack of a highly efficacious method of reversal of their anticoagulant effect. RE-VERSE AD established the monoclonal antibody idarucizumab in reversing the anticoagulant effect of the direct thrombin inhibitor dabigatran. Andexanet alfa is a recombinant, modified human factor Xa decoy protein developed to reverse the effects of factor Xa inhibitors including enoxaparin, apixaban, and rivaroxaban. On the basis of two studies in healthy volunteers in which andexanet alfa reversed the anticoagulant effects of apixaban and rivaroxaban (ANNEXA-A and ANNEXA-R), the FDA granted accelerated approval to andexanet alfa in May 2018 for the treatment of life-threatening or uncontrolled bleeding in patients receiving either of these two factor Xa inhibitors. Prospective studies were needed to demonstrate the efficacy of andexanet alfa in the management of patients with hemorrhage associated with factor Xa inhibitors. Additional studies are ongoing in a variety of patient populations, and the present study represents an interim analysis of one such study.

The Andexanet Alfa, a Novel Antidote to the Anticoagulation Effects of Factor Xa Inhibitors-4 (ANNEXA-4) study was a multicenter, prospective, single-arm observational trial with target accrual of 350 patients presenting with acute major bleeding within 18 hours of a dose of factor Xa inhibitor. Patients were treated with andexanet alfa as a bolus followed by an infusion, and outcomes includes changes in anti-Xa activity and clinical hemostasis. This report by Connolly and colleagues presents interim results of the first 67 patients enrolled in the study. Excellent or good hemostasis was achieved in 79% of patients with similar rates of hemostasis among patients presenting with GI or CNS hemorrhage. Pharmacokinetically, the anti-Xa activity was decreased in most patients evaluated. In a subset of patients for whom anti-Xa levels did not decrease, clinical hemostasis was adequate. Adverse events were frequent: mortality was 15% and thromboembolic events occurred in 18%, and one third of these thromboembolic events occurred within 3 days of andexanet alfa exposure.

This study is the first to outline a real-world experience with andexanet alfa. Mature results were published in February 2019, confirming the results of this interim analysis. Despite the favorable results in this trial, enthusiasm about andexanet alfa is limited by the single-arm nature of this study's design, the high rate of death and thromboembolism after drug administration, and the fact that patients with severe neurologic compromise or those requiring urgent surgery — two groups in which factor Xa inhibitor reversal is desirable in clinical practice — were excluded by this trial's eligibility criteria. Because of these and other factors, widespread use of andexanet alfa will likely be limited until efficacy and safety against standard-of-care maneuvers (eg, prothrombin complex concentrates) are established. As a condition of accelerated approval, the FDA has required that the manufacturer conduct a randomized study of andexanet alfa versus standard of care in patients with intracranial hemorrhage related to factor Xa inhibitor use; this so-called Study 18-513 (NCT03661528) is currently recruiting and is expected to be completed October 2022. Even if safety and efficacy are demonstrated, cost is likely to remain a major limitation of this therapy. The wholesale price of a single 100-mg vial of andexanet alfa is $3,300 and in the most optimistic dosing scenario a single reversal would cost $29,700. By comparison, at a cost of approximately $1.50 per unit, 4-factor prothrombin complex concentrate would cost approximately $7,500 to reverse a 100-kg patient.

Guidelines

As of February 2019, no guidelines have been published that reflect the results of this trial.

Design

• Multicenter, prospective, open label, single-group trial
• N=114 patients with acute major bleeding while receiving a factor Xa inhibitor
• Setting: 22 centers in the US, UK, and Canada
• Follow-up: 30 days
• Enrollment: 2015-2022 planned
• Analysis: Descriptive
• Primary Outcome: Hemostasis within 12 hours after infusion

Population

Inclusion Criteria

• Age ≥18 age
• Receipt of factor Xa inhibitor in prior 18 hours (apixaban, rivaroxaban, edoxaban, or enoxaparin only)
• Acute major bleeding, defined as:
  • Potentially life-threatening bleeding with hemodynamic compromise, or
  • Acute overt bleeding with ≥2-g/dL hemoglobin drop, hemoglobin ≤8 g/dL if no baseline is available, or investigator's opinion that hemoglobin is likely to fall to ≤8 g/dL
  • Acute symptomatic bleeding in a critical area/organ
• Only patients with anti-Xa activity ≥75 ng/mL were included in the efficacy analysis

Exclusion Criteria

• Surgery within 12 hours after presentation
• Intracranial hemorrhage with GCS <7
• Estimated intracerebral hematoma volume >60 mL
• Expected survival <1 month
• Major thrombotic event within prior 2 weeks
• One of the following within 7 days prior to screening: vitamin K antagonist, dabigatran, prothrombin complex concentrate, or whole blood or plasma

Baseline Characteristics

From the safety population.

• Demographics: mean age 77 years, 48% female, 81% Caucasian
• Time to administration: presentation to drug 4.8 h, time from consent to drug 1.7 h
• Anthropomorphics: mean BMI 28.1
• eGFR, mL/min: <30 9%, 30-60 46%, ≥60 39%
• Anticoagulation indication: Atrial fibrillation 70%, VTE 22%, both 7%
• Medical history: MI 19%, stroke 25%, DVT 30%, PE 9%, atrial fibrillation 73%, HF 34%, DM 34%
• Agent being reversed: apixaban 46%, edoxaban 0%, enoxaparin 6%, rivaroxaban 48%
• Site of bleed: GI 49%, intracerebral 21%, subdural 16%, subarachnoid 4%, other 9%

Interventions

• Treatment was administered as an initial bolus over 15-30 min followed by a 2-hour infusion
• Doses based on agent being reversed
  • Apixaban or rivaroxaban >7 hours before andexanet alfa: bolus 400 mg and infusion 480 mg
  • Enoxaparin, edoxaban, or rivaroxaban ≤7 hours before administration or at an unknown time: bolus 800 mg and infusion 960 mg

Outcomes

Primary Outcomes

From the efficacy population (n=47).

Excellent or good hemostasis 12 hours after andexanet infusion

79% (95% CI 64-89)

Secondary Outcomes

Median anti-Xa level

Agent	n	Baseline	Post-bolus	End of infusion	4-h post-infusion
Apixaban	20	149.7	10.3	12.5	103.0
Enoxaparin	1	0.61	0.15	0.19	0.46
Rivaroxaban	26	277.0	16.8	30.6	177.7

Modified Rankin scale in patients with intracranial bleed (n=28)

Baseline 2.2±1.9 to 2.0±2.0 at 30 days

Subgroup Analysis

Excellent or good hemostasis for GI bleed (n=47)

84%

Excellent or good hemostasis for intracranial bleed (n=47)

80%

Adverse Events

From the safety population (n=67).

Thrombotic events

18% (1 MI, 5 stroke, 7 DVT, 1 PE)

Within 3 days of andexanet alfa: 30% patients

Between 4-30 days of andexanet alfa: 70% patients

Mortality

15%

Criticisms

• This is an interim analysis and thus mature results are needed before establishment of activity.
• No patient in the population had been exposed to edoxaban which is not commonly used in the US.
• Single-group cohort so no comparison can be made. (Separate RCT is ongoing.)
• Patient with severe bleeding were excluded from analysis.

Funding

• Portola Pharmaceuticals, Bayer, Bristol-Myers Squibb, Johnson & Johnson, and Pfizer