Reversal Strategies for Non-vitamin K Antagonist Oral Anticoagulants: A Critical Appraisal of Available Evidence and Recommendations for Clinical Management

A Joint Position Paper of the European Society of Cardiology Working Group on Cardiovascular Pharmacotherapy and European Society of Cardiology Working Group on Thrombosis

Alexander Niessner; Juan Tamargo; João Morais; Lorenz Koller; Sven Wassmann; Steen Elkjær Husted; Christian Torp-Pedersen; Keld Kjeldsen; Basil S. Lewis; Heinz Drexel; Juan Carlos Kaski; Dan Atar; Robert F. Storey; Gregory Y. H. Lip; Freek W. A. Verheugt; Stefan Agewall


Eur Heart J. 2017;38(22):1710-1716. 

In This Article

The Potential Benefit of Specific Reversal Agents for Non-vitamin K Antagonist Oral Anticoagulants

A number of new reversal agents are under development, including monoclonal antibodies, decoy receptors, and small molecules with high affinity for NOACs (Table 2). These agents aim to specifically, completely, and rapidly reverse the effect of NOACs (Figure 2). Another perceived advantage of these reversal agents is the lack of a pro-thrombotic effect of the agents themselves. However, the sole interruption of anticoagulation therapy may be associated with an increased risk of thromboembolic events. Accordingly, in Phase III trials for AF, an increased thromboembolic risk has been observed at the end of the study after study treatment has been stopped.[7] Despite the assumed high efficiency of specific reversal agents, their effectiveness in preventing clinical events needs to be determined. Specifically, life-threatening situations with impairment of renal and liver function will constitute a challenge for improvement of survival by specific reversal of NOAC effects.

Figure 2.

Mechanism of non-vitamin K antagonist oral anticoagulants and their antidotes adapted from Enriquez et al.63 The prothrombinase complex, consisting of FXa and FIIa, catalyses the conversion of pro-thrombin (II) to thrombin (IIa), leading to fibrin generation and clot formation. The effect of the thrombin inhibitor dabigatran (dab) may be reversed by idarucizumab, a humanized antibody fragment against dabigatran OR ciraparantag, a small synthetic molecule competitively binding the non-vitamin K antagonist oral anticoagulants (A). The effect of the FXa inhibitors (a-Xa) apixaban, edoxaban and rivaroxaban may be reversed by andexanet alpha, a modified inactive recombinant FXa competitively binding circulating FXa inhibitors OR ciraparantag (B).

Idarucizumab, an Antibody Fragment Targeting Dabigatran

Idarucizumab (BI 655075) is a humanized, highly selective, and specific monoclonal antibody fragment (fab) which binds dabigatran with high affinity (dissociation constant, KD, 2.1 pM), i.e. ~350 times stronger than for thrombin. It rapidly (<5 min) and completely inhibited the anticoagulant activity of dabigatran in ex vivo human clotting tests (half maximal inhibitory concentration [IC50] of 2–5 nM) as well as in in vivo models, exerting a sustained effect for up to 6 h after intravenous (i.v.) injection.[40] However, idarucizumab has no direct activity on coagulation or platelet aggregation.[40,41] In Phase I placebo-controlled studies, ecarin clotting time (ECT) and diluted thrombin time (dTT) were significantly prolonged with dabigatran and reversed to control levels after i.v. dosing with idarucizumab (1, 2, 4, and 5 + 2.5 g) in 47 men.[42] It was found to be safe and effective, in that dTT was almost immediately corrected.[41,42] For the 1 g dose, there was partial return of dabigatran-induced anticoagulation accompanied by an increase of unbound dabigatran concentrations ~2–4 h after i.v. infusion.[42,43] Additionally, idarucizumab reversed dabigatran-induced inhibition of wound site fibrin formation.[44] In another 46 male and female volunteers, 2.5–5 g of idarucizumab resulted in immediate, complete, and sustained reversal of dabigatran-induced anticoagulation. The reversal effect was consistent for elderly and renally impaired subjects [creatinine clearance (CrCl) 30 to <90 mL/min).[43] Dabigatran anticoagulation could be re-established 24 h after idarucizumab dosing.

An uncontrolled Phase III study (RE-VERSE AD) will assess for the reversal of the anticoagulant effects of dabigatran by the i.v. administration of 5 g of idarucizumab in dabigatran-treated patients, who have uncontrolled bleeding or require emergency surgery or procedures (NCT0210494).[45] Up to 300 dabigatran-treated patients with uncontrollable overt bleeding in need of reversal as judged by the attending physician (group A), OR in need of emergency surgery or procedures within the first 4 h (group B) will be enrolled. Patients with bleeding manageable conservatively (group A) or requiring surgery where the risk of uncontrolled or unmanageable bleeding is low (group B) are excluded. The primary endpoint is the maximum reversal of anticoagulant effect of dabigatran in the first 4 h measured with a specific coagulation test (dTT or ECT). A first interim analysis has been performed in 90 patients (A: n = 51, B: n = 39, median age 76.5 years, median CrCl 58 mL/min). The most common bleeding events in group A were GI bleeding (n = 20) and ICH (n = 18). Overall, 22 patients were excluded from further analysis as they already had normal coagulation tests at baseline. Idarucizumab rapidly and completely reversed the anticoagulant activity of dabigatran in 88–98% of patients.[46] In parallel, the concentration of unbound dabigatran was reduced to a level at or near the lower limit of quantification. In few patients, a rebound of the effect of idarucizumab was observed after 12 h (n = 6) and 24 h (n = 16), probably due to redistribution of extravascular dabigatran into the intravascular compartment. These results confirm successful reversal of the specific effect of dabigatran in critically ill patients who suffer uncontrollable bleeding or who require emergency surgery while on dabigatran. With regard to clinical bleeding, haemostasis was restored after a median of 11.4 h in group A. In group B, normal intraoperative haemostasis was observed in 92% of patients. However, this normalization of haemostasis did not prevent 18 deaths. One thrombotic event was observed within 72 h after administration of idarucizumab and another four events after 72 h. A controlled design against placebo or CFCs would perhaps have been better, to assess the impact of the reversal on the clinical course of patients. But, given the clinical type of the RE-VERSE AD trial patients, such a design presents major ethical challenges. The interim analysis of the RE-VERSE AD trial led to the approval of idarucizumab by the European Medicines Agency and the U.S. Food and Drug Administration.

Andexanet (PRT064445), a Truncated Human Recombinant FXa

Andexanet alpha (AnXa; PRT064445), a truncated human recombinant FXa, is catalytically inactive and lacks the membrane-binding domain of native FXa but retains the ability of native FXa to bind direct FXa inhibitors as well as low-molecular-weight heparin-activated antithrombin III with high-affinity (mean KD 0.5–1.5 nM for direct FXa inhibitors).[47] It acts as a decoy receptor and competes with native FXa, thereby reversing the anticoagulant effects of all FXa inhibitors. AnXa dose-dependently reversed the inhibition of FXa activity by FXa inhibitors and corrected clot formation ex vivo.[47] In animal bleeding models, it reduced blood loss and restored haemostasis after pretreatment with rivaroxaban as well as indirect FXa inhibitors (low-molecular-weight heparins and fondaparinux). The effect of AnXa correlated with the reduction in free FXa inhibitor plasma levels and peak anti-FXa activity. However, AnXa did not interfere with normal FXa function in haemostasis and did not have anticoagulant activity. In a Phase I study involving 32 healthy volunteers not on anticoagulant therapy, AnXa proved to be safe and well tolerated.[48]

A placebo-controlled Phase II trial (NCT01758432) investigated the reversal effect of AnXa after dosing to steady state with direct and indirect FXa inhibitors in healthy volunteers. Boluses of AnXa ranging from 210 to 800 mg were used for reversal of FXa inhibitors.[49,50] A dose-dependent reversal of the anticoagulant effect was observed when measuring anti-FXa activity with a maximum decrease of >90%.[50] A subsequent infusion of 4–8 mg AnXa/min for 1–2 h preserved the reversal effect. Anti-FXa activity returned to placebo levels 2 h after the end of AnXa infusion. In parallel, plasma concentrations of unbound FXa inhibitors decreased immediately after AnXa bolus. Thrombus generation and clotting time were also dose-dependently reversed by AnXa. No thrombotic or serious adverse events were observed. However, tissue factor pathway inhibitor activity also decreased due to its binding to AnXa. These results suggest that AnXa has the potential to be a universal antidote for reversal of anticoagulation produced by direct and indirect FXa inhibitors.

In a next step, AnXa was tested in older healthy volunteers measuring anti-FXa activity. The first ANNEXA-A trial (NCT02207725) tested the reversal of apixaban 5 mg twice daily given for 4 days to 33 healthy volunteers aged 50–70 years. AnXa 400 mg i.v. bolus showed a 94% complete reversal rate, effectively and rapidly reversing the anticoagulant effect of apixaban.[51] No serious adverse events or thrombotic events were observed. In the second ANNEXA-A trial, AnXa will be administered as 400 mg i.v. bolus followed by a continuous infusion of 4 mg/min for 120 min. In the ANNEXA-R trial (NCT02220725), the reversal of rivaroxaban will be tested in older healthy volunteers.[52] ANNEXA-E for reversal of edoxaban is planned.

A Phase III trial will evaluate the safety and ability of AnXa in reversing the effect of Xa inhibitors in patients with an acute major bleeding episode (NCT02329327). Like RE-VERSE AD testing idarucizumab, this will be an uncontrolled, open-label study. The primary endpoint will be the proportion of patients with excellent or good haemostasis.

Ciraparantag (PER977), a Small-Molecule Antagonist

Ciraparantag (PER977) is a synthetic, peptide-like, small molecule designed for the reversal of oral direct FXa and FIIa inhibitors, low-molecular-weight heparins, unfractionated heparin, and fondaparinux but not VKA. Given the molecular structure of PER977, immunogenic reactions are unlikely.

In ex vivo studies, PER977 completely reversed the anticoagulant activity of rivaroxaban and apixaban measured with anti-FXa activity in human plasma.[53] No pro-coagulant effects were observed. In animals treated with high doses of edoxaban, PER977 reduced blood loss in bleeding models (rat tail transection).[54] Similar results were observed with rivaroxaban, apixaban, and dabigatran.[55,56] In parallel, PER97 restored global coagulation tests to baseline levels within 20 min. Additionally, no pro-coagulant effects were observed.[55] In a liver laceration model, the highest investigated dose of PER977 and AnXa showed a similar reduction of blood loss in rabbits pretreated with rivaroxaban.[57] However, PER977 did not affect anti-FXa activity in this model.

A Phase I trial (NCT01826266) evaluated the safety, tolerability, and plasma and urinary pharmacokinetics of a wide range of single i.v. doses of PER977 (5–300 mg) administered either alone or following a single dose of edoxaban.[58] PER977 >100 mg reversed edoxaban 60 mg within 10 min and restored clot formation without rebound signals for 24 h.[59] No pro-coagulant effects were observed. In this first trial, PER977 was safe and well tolerated.

An ongoing clinical trial (NCT02207257) evaluates the safety, tolerability, and effect on whole blood clotting time of escalating intravenous doses of PER977 (25, 50, 100, 300, and 600 mg) administered after 60 mg edoxaban as a 'rescue' medication in healthy volunteers and repeated for a second day to investigate any effects of PER977 on the re-anticoagulation with edoxaban and second reversal with PER977 (NCT02207257). Phase III clinical trial results need to be awaited to assess the full clinical therapeutic potential of this promising new NOAC antidote.

Areas of Future Research

While traditional controlled Phase III trials may not be feasible in this area of research in future, an accumulating body of evidence from observational real-world studies will be even more important.[60] Areas of uncertainty to be addressed in future studies include (i) the risk of thromboembolic events in patients with immediate reversal of anticoagulation (with reversal agents without a pro-thrombotic effect per se); (ii) the net clinical benefit of specific antidotes, particularly in patients with severe but not life-threatening bleeding or urgent surgery; (iii) the effect of specific reversal agents on clinical events with specific focus on ICH; and (iv) the effect of specific reversal agents in patients with spontaneous and provoked bleeding.