Direct oral anticoagulants (DOACs) revolutionized oral anticoagulation. Among other favorable characteristics, the most compelling advantage of DOACs over warfarin is a reduced risk of serious bleeding. Owing to their short half-lives, temporary discontinuation of DOACs alone may be sufficient for many instances of bleeding, or semiurgent or low-bleed-risk invasive procedures. However, emergencies such as major bleeding carry substantial short-term mortality (up to 14% within 30 days), emphasizing the need for agents that improve clinical outcomes by normalizing hemostasis rapidly and effectively.2
Ciraparantag binds to heparins and the DOACs through noncovalent charge-charge interactions. Binding to other drugs and blood proteins (including coagulation factors) has not been demonstrated. Studies in human volunteer subjects receiving 15 to 300 mg of ciraparantag showed dose-proportional pharmacokinetics with maximum serum concentrations achieved within 5 to 9 minutes and half-life of 12 to 19 minutes.3 Ciraparantag is hydrolyzed by serum peptidases into nonactive metabolites and is renally excreted. In animal bleeding models, ciraparantag reduced blood loss after tail transection in rats treated with edoxaban, dabigatran, apixaban, rivaroxaban, unfractionated heparin, and enoxaparin. Similarly, ciraparantag (at doses of 20 mg/kg and 30 mg/kg) reduced bleeding time following liver laceration in edoxaban-treated rats. Within 60 minutes of ciraparantag administration, fibrin strands were seen in blood pellets examined by electron microscopy. In a randomized double-blind trial, human volunteer subjects were treated with edoxaban (60 mg oral dose) followed by ciraparantag (100 to 300 mg) or placebo.4 Ciraparantag rapidly (within 10 to 30 minutes) corrected prolonged whole blood clotting time (WBCT) from 37% above baseline to 10% above baseline, an effect that lasted for 24 hours. Adverse events were mild with no evidence of procoagulant activity. Similar findings were seen after administration of ciraparantag (100 to 300 mg) or placebo in healthy volunteer subjects treated with enoxaparin (1.5 mg/kg). There was no difference in WBCT between the ciraparantag and placebo groups at 12 to 15 hours after administration, consistent with anticoagulant clearance.
The findings summarized by Ansell and colleagues, including rapid onset of action, reduction in bleeding in animal models, prolonged pharmacodynamic effects, and tolerability, support the potential of ciraparantag as an anticoagulant reversal agent. Similar to the clinical development of other anticoagulant reversal agents idarucizumab (for dabigatran) and andexanet alfa (for factor Xa inhibitors), there are important considerations for designing and conducting clinical trials to establish the efficacy and safety of reversal agents.
The challenges of measuring DOAC drug levels and interpreting their clinical significance in patients being considered for urgent reversal are well described.2 Routine coagulation tests are not accurate or reliable for determining DOAC levels, and specific DOAC assays are not widely available. Furthermore, there are no established DOAC therapeutic ranges, and the plasma concentrations corresponding to a clinically significant hemostatic defect are unknown. As a result, decisions to administer reversal agents are often made using the timing of the last dose to inform the likelihood of clinically significant drug levels. For example, in the ANNEXA-4 study, which evaluated andexanet alfa, patients with acute major bleeding on factor Xa inhibitors were eligible if the last dose was within the previous 18 hours5 ; drug levels were below the cutoff for inclusion in the efficacy analysis in up to 25% of patients, suggesting they may not have had clinically significant anticoagulant levels, although the threshold for "clinically significant" is uncertain. This has important implications for determining net clinical benefit for individual patients, and for health care resource utilization and costs. An additional complexity of laboratory testing with ciraparantag is its binding to anionic additives in blood collection tubes (eg, sodium citrate, EDTA) precluding the use of routine coagulation tests until it is cleared. As a result of this characteristic, the WBCT was used for laboratory assessment of anticoagulant effect in animal models and human volunteer studies. Although validated for use in these research studies, this test is not routinely available for clinical use, requires expertise, and may be subject to interobserver variability.
Ultimately, the efficacy and safety of reversal agents can only be established with randomized trials measuring clinically relevant patient-important outcomes. There is uncertainty regarding the incremental benefits and harms of idarucizumab and andexanet alfa, which were studied in open-label prospective cohort studies lacking control groups.5,6 Cohort studies evaluating 4-factor prothrombin complex concentrate (4-factor PCC) in patients with bleeding on factor Xa inhibitors similarly lacked control groups.7 The ANNEXA-I randomized controlled trial is evaluating andexanet alfa vs usual care for intracranial hemorrhage in patients receiving factor Xa inhibitors (#NCT03661528). Although the results of ANNEXA-I are highly anticipated, the role of reversal remains uncertain in patients with gastrointestinal bleeding, the most frequent single site of bleeding. Among patients with major gastrointestinal bleeding in REVERSE-AD and ANNEXA-4, the 30-day mortality rate was 11%.8,9
Another important consideration is the reliable measurement of bleeding cessation. Ascertainment of outcomes defined using subjective criteria is vulnerable to bias. In REVERSE-AD, clinical hemostatic efficacy was assessed by treating physicians, whereas in ANNEXA-4 it, was assessed by independent adjudicators using prespecified definitions.5 To standardize outcome ascertainment and reporting, the International Society on Thrombosis and Haemostasis published consensus definitions of hemostatic efficacy.10 Importantly, various definitions of hemostatic efficacy have not been prospectively validated.
In REVERSE-AD, ANNEXA-4, and 4-factor PCC studies, high 30-day mortality rates were seen after major bleeding (13% to 32%) despite moderate to high rates of clinical hemostasis, which is not entirely explained by index bleed site.5,6 These findings suggest that existing definitions of hemostatic efficacy may not adequately capture ongoing bleeding, or that major bleeding triggers a series of adverse events leading to death despite cessation of bleeding. In addition to concerns about thrombotic events after anticoagulant reversal, these findings raise additional uncertainty about overall net clinical benefit. The results of clinical trials designed to establish the effect of anticoagulant reversal agents (including ciraparantag) on clinically relevant and patient-important outcomes are eagerly awaited.
Conflict-of-interest disclosure: The author has received honoraria from Aspen Pharma, Bayer, BMS-Pfizer, Leo Pharma, Novartis, Portola Pharmaceuticals, and Servier.