Authors: Emily Shor, PharmD; Alex Meyr, PharmDPGY-1 Pharmacy Practice Residents
Mentor: Davina Dell-Steinbeck, PharmD, BCPSSSM Health St. Mary’s Hospital – St. Louis
Program Number:Approval Dates:
Approved Contact Hours: One (1) CE(s) per LIVE session.
BackgroundWarfarin, a vitamin K antagonist (VKA) discovered in the 1920s, has been commonly used for prophylaxis and treatment of venous thromboembolism (VTE).1 The Food and Drug Administration (FDA) approved the first direct oral anticoagulant (DOAC), dabigatran, in 2010. This was followed by rivaroxaban in 2011, apixaban in 2012, and edoxaban in 2015.2-5 While warfarin is dosed to achieve target International Normalized Ratio (INR), the doses for DOACs for VTE treatment, VTE prophylaxis, and stroke/systemic embolism prophylaxis in nonvalvular atrial fibrillation (NVAF) are shown in Table 1.1-5
The pharmacokinetic and pharmacodynamic profiles of each oral anticoagulant differ and should be considered when selecting an oral anticoagulant (Table 2). To varying extents, all DOACs are excreted renally. Dabigatran has the lowest protein binding and is the only dialyzable DOAC. As renal function decreases, each DOAC’s half-life and anticoagulant effects increase, resulting in increased risk for bleeding.2-5
A patient’s risk for bleeding while receiving an oral anticoagulant can be assessed using the HEMORR2HAGES or HAS-BLED scores.7 Patients’ bleeding risk can vary based on the type of surgery; major orthopedic, cardiac, vascular, neurosurgical, cancer, and urologic surgeries are associated with an increased bleeding risk whereas minor surgeries are associated with low bleeding risk. Therefore, depending on the procedure, patients may need to hold their oral anticoagulant to avoid an increased risk of bleeding. Other key factors that are associated with increased risk of bleeding with concurrent oral anticoagulants include history of stroke/bleeding, trauma, nonadherence, concurrent use of antiplatelet or nonsteroidal anti-inflammatory drugs, advanced age, impaired renal or liver function, alcohol abuse, malignancy, and rheumatic heart disease.8
Warfarin is routinely monitored with INR levels; however, DOAC therapy does not require routine drug monitoring. Routine coagulation assays, such as prothrombin time (PT), activated partial thromboplastin time (aPTT), or anti-Xa levels, have not been used for therapeutic monitoring of DOACs because results of these tests have not been reliably correlated to therapeutic outcomes. However, some studies show that PT, aPTT, and INR can be used to determine excessive DOAC plasma concentrations. Because the results of these tests are not sensitive, in patients with elevated levels, other etiologies should be assessed alongside the patient’s anticoagulation history.9
In patients with high risk of bleeding, reversal of anticoagulation may be necessary. Some approaches to managing acute bleeding may include mechanical intervention or administration of antidotes. This review will assess the role of possible antidotes for oral anticoagulants, including warfarin, dabigatran, and Factor Xa inhibitors.
Warfarin ReversalThe risk of bleeding with warfarin administration increases significantly when the INR exceeds 4.5. Currently there are three options for the urgent reversal of vitamin K antagonists, including vitamin K, four-factor prothrombin complex concentrate (PCC), and fresh frozen plasma (FFP). For patients with VKA-associated major bleeding, ACCP/CHEST guidelines recommend rapid reversal of anticoagulation with PCC rather than FFP (Grade 2C). Vitamin K 5 to 10 mg may also be administered by slow IV injection rather than reversal with coagulation factors alone (Grade 2C).10 ACCP/CHEST guidelines provide the following recommendations for warfarin reversal based on INR and evidence of bleeding:
The time to effect of oral and intravenous vitamin K, respectively, is 24 hours and 8-12 hours. Both formulations’ effects can last for several days. In contrast, FFP and PCC have an immediate effect, which typically lasts 12 to 24 hours. The risk of thrombosis with vitamin K or FFP is insignificant; however, the use of PCC is associated with an increased risk of thrombosis.1,11
While vitamin K has historically been used as warfarin’s primary reversal agent, in 2013, the FDA approved PCC (KCentra®) for the urgent reversal of VKA-associated acute major bleeding or need for urgent surgery or invasive procedure. PCC increases the levels of vitamin K-dependent coagulation factors (II, VII, IX, and X) and proteins C and S. PCC is contraindicated in patients with disseminated intravascular coagulation or known heparin-induced thrombocytopenia. Additionally, PCC has boxed warnings for arterial and venous thromboembolic complications. Both fatal and on-fatal arterial and venous thromboembolic complications have been reported with PCC in clinical trials and post-marketing surveillance.11
PCC’s package insert provides the following recommendations for dosing based on the patient’s pre-treatment INR:11
While PCC’s package insert recommends variable dosing, literature has supported the use of fixed-dose PCC as well.11 A randomized, plasma-controlled, phase IIIb study evaluated the safety and efficacy of 4FPCC for urgent VKA reversal. Adult patients (n=98) with an INR of at least 2.0 within three hours prior to study treatment and presenting with an active major bleed (life-threatening or potentially life threatening, acute bleed with a decrease in hemoglobin of at least 2 g/dL, or bleeding requiring transfusion of a blood product) were included. These patients received PCC, which was dosed based on the package insert’s recommended dosing strategy. 44.9%, 27.6%, and 72.5% of patients were reported to have a primary rating of excellent, good, and effective response, respectively. Overall, 62% of patients achieved an INR of less than or equal to 1.3 within 30 minutes post infusion with a median dose of 2475 IU. Four of the eight thromboembolic events reported were determined to be related to PCC.12
An observational cohort pilot study assessed the safety and efficacy of PCC (Cofact®) for the reversal of VKA treatment in the setting of major or clinically relevant non-cranial bleeding or an emergency invasive procedure. In this study, the indication for PCC use and target INR was defined by the provider. Patients either received fixed dose PCC (1040 IU for major bleeding, 530 IU for emergency invasive procedure) or doses based on presenting INR, target INR, and body weight (pre-specified by study protocol). Ultimately, the study did not find any statistically significant differences in outcomes. Target INR was achieved in 70% and 81% of patients receiving fixed dose (n=35) or variable dose (n=32), respectively. The median dose utilized in the fixed dose group was 1040 IU and 1560 IU in the variable dose group (p=0.001). 28.8% of patients in the fixed dose group received 530 IU for an emergency invasive procedure and 62.8% received 1040 IU for major bleeding. One patient in the fixed dose group experienced a thromboembolic event compared to two patients in the variable dose group. This study suggests that a fixed dose of 1040 IU of PCC may be efficacious in rapidly reversing VKA therapy. If fixed dose PCC is shown to be as efficacious and safe as variable dosing, then this may result in significant cost savings. However, many studies reporting fixed dose PCC results have varying study methodologies, making it challenging to extrapolate this data.13
For example, a retrospective cohort study assessed the safety, efficacy, and cost of a fixed dose PCC (Kcentra®) protocol. This study included adult patients receiving 1500 IU of PCC per the hospital’s protocol for any clinical indication for emergent VKA reversal and who were on chronic VKA therapy (mean presenting INR 3.3). Of the 39 included patients, 71.8% received PCC for an intracranial bleed. Their mean age and weight were 70 years and 79.5 kg respectively. With a single dose of 1500 IU PCC, 92.3% of patients’ INRs decreased to less than 2.0, and 71.8% of patients’ INRs successfully decreased to less than or equal to 1.5. One patient required a second dose of PCC, and no patients experienced a thromboembolic event within seven days. This study concluded that 1500 IU PCC was safe and efficacious for emergent warfarin reversal. While this study identified cost-savings, they did not directly compare these results to a variable dosing strategy. Additionally, patients with a presenting INR of less than 2.0 (n=4) and patients receiving FFP (n=11) were included, which may confound results.14 Because the mean presenting INR in this study was 3.3, it is also difficult to assess PCC’s efficacy and safety in patients with significantly higher INRs at baseline.
Few studies have evaluated the safety and efficacy of fixed-dose PCC in comparison to variable dose PCC. However, reported studies show promising utility. The reported fixed-dose PCC studies have shown to be effective and offer similar outcomes as the landmark trials approving variable dose PCC. However, studies assessing fixed dose PCC included varying populations, making extrapolation of results challenging. For example, studies have not assessed the use of fixed dosing in the extremes of weight where the fixed dose may vary significantly when compared to variable dosing strategies. Because fixed dose PCC for the reversal of major bleeding has not been FDA approved, this may pose as an ethical dilemma if the fixed dose chosen is significantly higher or lower compared to a patient’s ultimate variable dose. Additionally, studies assessing the safety and efficacy of PCC for intracranial hemorrhages utilize varying fixed doses, so this niche population needs further study.
Direct Thrombin Inhibitor Reversal
Idarucizumab (Praxbind®) is a humanized monoclonal antibody approved in 2015 for the reversal of dabigatran in patients undergoing emergency surgery or urgent procedure, as well as for life-threatening or uncontrolled bleeding.15 Idarucizumab binds to dabigatran and its metabolites with greater affinity than dabigatran binds to thrombin. Specifically, in vitro studies have shown that its affinity toward dabigatran is approximately 350 times stronger than dabigatran’s affinity for thrombin. It has also been demonstrated that idarucizumab does not bind to thrombin substrates, nor does it impact platelet aggregation.16
In terms of safety, idarucizumab was studied by Glund and colleagues in a trial that included 110 healthy male volunteers, in a 3:1 ratio (idarucizumab to placebo) to receive doses of idarucizumab ranging from 20 mg to 8 g, given either as 1-hour infusions or 5-minute infusions.17 Drug-related adverse events seen in this study were infrequent and of mild intensity. Additionally, this study demonstrated that idarucizumab had no effect on thrombin or other coagulation factors when given to patients not taking dabigatran.17
In another study by Glund and colleagues, both the safety and efficacy of idarucizumab was assessed in a phase 1, randomized, placebo-controlled trial in Belgium. This study included healthy male volunteers from the age of 18 to 45 who were given dabigatran 220 mg twice daily for three days and once on day four and then randomly assigned them to four different idarucizumab dose groups (1 g, 2 g, 4 g, or 5 g plus 2.5 g).18 Twelve patients were included in each of the first three groups, and 11 patients were included in the last group. Within each group, patients were assigned in a 3:1 ratio of idarucizumab to placebo. Overall, this study found idarucizumab to be well-tolerated as all drug-related adverse events reported (~15% of patients) were of mild intensity, ranging from infusion site erythema to epistaxis and hematuria. Regarding efficacy, idarucizumab’s ability to reverse dabigatran was rapid, complete, and dose-dependent (reduction in diluted thrombin time for each of the four dosing strategies mentioned above was 74%, 94%, 98%, and 99%, respectively).18
Given the demonstrated safety and efficacy of idarucizumab in healthy individuals, Pollack and colleagues subsequently published a prospective cohort study that aimed to assess the effects of idarucizumab in patients on dabigatran who either had serious bleeding (group A) or required an urgent procedure (group B).19 In the interim analysis of this trial (RE-VERSE AD trial), 90 patients received 5 g of idarucizumab given as two, 2.5 g bolus infusions no more than 15 minutes apart. The primary efficacy endpoint of this study was the maximum reversal effect of dabigatran from the end of the first idarucizumab infusion to four hours after the second infusion. The maximum reversal effect was determined by dilute thrombin time or ecarin clotting time (ECT) and was calculated as a percentage. Extent and severity of bleeding, hemodynamic stability, and adverse events made up most of the secondary endpoints.19
Over 90% of patients included in the RE-VERSE AD trial had atrial fibrillation and were on dabigatran for stroke prevention. The average age of this group of patients was 76.5 years, the median creatinine clearance was 58 mL/min, and the median time since last dabigatran dose was 15.4 hours.19 The primary efficacy outcome (reversal effect of idarucizumab) was assessed in 68 of the 90 included patients because the other 22 patients had dilute thrombin times that were within normal limits at study entry. The primary outcome was also assessed based on the ECT test in 81 of the 90 patients. Overall, idarucizumab was efficacious for patients in both groups (serious bleeding and urgent procedure), as the median maximum percent reversal was calculated to be 100% (95% CI: 100 to 100).19 Additionally, the dilute thrombin time was normalized in 98% and 93% of patients in group A and B respectively, while the ECT was normalized in 89% and 88% of patients, respectively. The median time to cessation of bleeding was estimated to be approximately 11.4 hours in a subset of patients in group A. In terms of pharmacokinetics, this study found that concentrations of unbound dabigatran were less than 20 ng/mL, which indicated minimal or no anticoagulant activity in 93% of patients after 12 hours and 79% of patients after 24 hours of idarucizumab administration.19 With regard to safety outcomes, there were 21 serious adverse events, including 18 deaths, five thrombotic events, and two cases of gastrointestinal hemorrhage; however, few were directly related to the use of idarucizumab (one thrombotic event occurred within 72 hours of its use).19
An update to this study was published about a year after the interim analysis of RE-VERSE AD and included an additional 43 patients. Thus, a total of 123 patients (66 in group A and 57 in group B) were included in the analysis.20 Baseline characteristics remained about the same as in the interim analysis, and 95% of patients received dabigatran for the indication of stroke/systemic embolism prophylaxis in NVAF. With regard to the primary outcome, Pollack and colleagues found that complete reversal of dabigatran occurred in over 89% of patients, and the median time to cessation of bleeding in 48 of the group A patients was 9.8 hours.20 Additionally, mean time to surgery in group B patients was approximately 1.7 hours after infusion of idarucizumab with no major bleeding occurring post-surgery. Thrombotic events occurred in five patients over a 24-day post-infusion period; however, none of those patients were anticoagulated during that time. Death occurred in 21% of patients, but these were not found to be directly related to use of idarucizumab.20 Overall, this study continued to show that idarucizumab can rapidly and completely reverse dabigatran, reduce time to surgery or an urgent procedure, and achieve hemostasis within approximately ten hours.
Factor Xa-Inhibitor Reversal
Unlike idarucizumab, which is a monoclonal antibody with high affinity toward dabigatran, andexanet alfa (Andexxa®) is a recombinant modified human Factor Xa (FXa) protein that exhibits a procoagulant effect through binding and sequestering FXa inhibitors.21 Additionally, andexanet alfa has demonstrated an ability to inhibit Tissue Factor Pathway Inhibitor (TFPI), which can lead to increased thrombin generation.21
The safety and efficacy of andexanet alfa was assessed in two trials, where healthy volunteers were given either apixaban 5 mg twice daily (ANNEXA-A trial) or rivaroxaban 20 mg daily (ANNEXA-R trial), followed by various regimens of andexanet alfa.22 Specifically, part 1 of each trial assessed the effects of an andexanet alfa bolus, while part 2 of each trial studied the effects of an andexanet alfa bolus followed by a 2-hour continuous infusion. The primary outcome of these studies was the average percent change in anti-factor Xa activity. Between the two studies, 101 participants were included (48 in ANNEXA-A and 53 in ANNEXA-R).22 In part 1 of each trial, Siegal and colleagues found anti-factor Xa activity to be rapidly reversed (within 5 minutes) significantly more with a bolus administration of andexanet alfa compared to placebo (ANNEXA-A: mean reduction, 94±2% vs. 21±9%; p<0.001; and ANNEXA-R: 92±11% vs. 18±15%, p<0.001).22 However, given the short half-life of andexanet alpha (approximately 1 hour), its anti-factor Xa activity only lasted for about two hours and then slowly returned to levels seen in the placebo arms. Similarly, in part 2 of these trials (bolus plus infusion), use of andexanet alfa continued to provide a significantly greater reduction of anti-factor Xa activity compared to placebo, and this activity was retained for about one to two hours after the infusion ended. Furthermore, both studies showed that patients who received andexanet alfa had significantly greater restoration of thrombin generation compared to patients who received placebo. Additionally, no serious adverse events (including thrombotic events) were reported, and antibodies to FXa did not develop in any participant.22
Given the promising results of the ANNEXA-A and ANNEXA-R trials, a subsequent study is being conducted by Connolly and colleagues (ANNEXA-4) to assess the safety and efficacy of andexanet alfa in patients who had acute bleeding within 18 hours of receiving a FXa inhibitor.23 In the interim analysis, 67 included patients received a bolus of andexanet alfa followed by a 2-hour infusion. The average age of the patients included was 77 years, and most patients had either gastrointestinal or intracranial bleeding. Additionally, the majority of patients were receiving either rivaroxaban or apixaban (less than 6% received enoxaparin prior to the study, and none were taking edoxaban).23 This study had several exclusion criteria, including (but not limited to): surgery scheduled within 12 hours of presentation; intracranial hemorrhage and Glasgow Coma Scale score of less than 7 or intracerebral hematoma volume of greater than 60 ml; survival expected to be less than 1 month; major thrombotic event in the past 2 weeks; receipt of warfarin, dabigatran, prothrombin complex concentrate, or whole blood or plasma within the past week. Additionally, while each patient was given a bolus dose of andexanet alfa followed by a continuous infusion, the actual amount given varied depending on the FXa inhibitor and the timing of its last administration (Table 5).23
In terms of the primary outcome, only 47 of the 67 patients were analyzed because 20 patients had baseline anti-factor Xa levels that were too low or missing. Overall, patients who were taking either rivaroxaban or apixaban before receiving andexanet alfa experienced a decrease in anti-factor Xa activity of 89% and 93% (respectively).23 However, four hours after the end of the infusion, this decrease was only 39% in the rivaroxaban arm and 30% in the apixaban arm. Clinical hemostasis was considered “excellent” or “good” (assessed by an adjudicator) in 79% of the patients in the efficacy subgroup (n=47) 12 hours after the infusion. Furthermore, in terms of the safety population (n=67), 12 patients (18%) experienced thrombotic events during the 30-day follow-up, and 10 patients died (15%).23 Although the results of this study make andexanet alfa an interesting option for reversal of FXa inhibitors, a major limitation is the exclusion of many surgical patients. Most patients were determined to achieve hemostasis 12 hours after the infusion, but given its short half-life, it is unclear if surgical patients will have adequate anticoagulant reversal prior to a procedure. This study is ongoing, and more studies are necessary to determine the most appropriate indications for andexanet alfa.
Andexanet alfa’s package insert defines low and high doses of andexanet alfa consistently with the dosing strategy utilized in the ANNEXA-4 trial (Table 5).21,23 The average wholesale price for 100 mg of andexanet alfa is $3,300. The cost of the low dose regimen would be approximately $29,040, which would include patients receiving less than or equal to 10 mg doses of rivaroxaban or less than or equal to 5 mg doses of apixaban. The cost of the high dose regimen would be approximately $58,080, which would include patients receiving greater than 10 mg doses of rivaroxaban, greater than 5 mg doses of apixaban, or if the previously administered dose of rivaroxaban/apixaban is unknown.21
Because there is limited data supporting the use of andexanet alfa in patients who require surgical intervention, along with the high cost associated with its use, many institutions may be reluctant to include it on their formulary. However, guidelines by the European Heart Rhythm Association and the American Heart Association/American Stroke Association state that in the event of a life-threatening bleed caused by a FXa inhibitor, PCC can be used even though it is not currently FDA approved for this indication.24-25
CiraparantagCiraparantag (PER977) is a new, investigational drug that is being developed for use as a potential reversal agent for both FXa inhibitors and factor IIa inhibitors.26 The exact mechanism of ciraparantag is unknown but in vitro studies have shown that it binds to anticoagulants via noncovalent hydrogen bonds. A study by Ansell and colleagues assessed the safety and efficacy of this agent when given as monotherapy and after a 60 mg dose of edoxaban in 80 healthy volunteers.26 In this study, whole-blood clotting time was used as means to determine the anticoagulant effect of edoxaban as well as assess the efficacy of ciraparantag. The patients were given a dose of edoxaban followed by a single intravenous dose of ciraparantag (ranging from 25 mg to 300 mg) or placebo three hours later. Patients who received ciraparantag 100 mg or 300 mg demonstrated a statistically significant decrease in whole-blood clotting time to within 10% of the baseline value, which occurred within 10 minutes. Conversely, it took patients in the placebo arm roughly 12 to 15 hours to reach a similar decrease in whole-blood clotting time.26 Furthermore, patients maintained the reduction in whole-blood clotting time for 24 hours after the administration of one dose of ciraparantag. Procoagulant activity of ciraparantag was not evident based on D-dimer levels, TFPI levels, and whole-blood clotting time. Overall adverse events were transient, mild, and/or not related to ciraparantag.26 There are more studies to come involving this new agent, which will determine if its reversal ability is similar with the other direct oral anticoagulants.
ConclusionsLots of patients require anticoagulation for a variety of indications which puts these patients at an increased risk of bleeding. Educating patients on the appropriate use of their anticoagulant therapy, as well as encouraging adherence, is key in order to help improve safety and efficacy. Specifically, providers need to be aware of other concomitant bleeding risk factors that their patients may have so that modifiable risk factors can be mitigated, and the most appropriate anticoagulant regimen is selected. Warfarin has therapeutic monitoring, which can help identify patients who are at a high risk of bleeding. Additionally, there are well-studied reversal strategies, including vitamin K and PCC. However, warfarin reversal strategies are continuing to be evaluated regarding fixed versus variable dosing of PCC. Unlike warfarin, DOACs do not require therapeutic monitoring, so their use continues to increase; however, because of the lack of sensitive therapeutic monitoring, correlating a patient’s bleed to the use of a DOAC can be more challenging. Thus, appropriate supportive care and treatment options need to be available for patients in the event of bleeding directly related to a DOAC. The current FDA-approved reversal agents for DOACs include idarucizumab and andexanet alfa. Idarucizumab is only indicated for the reversal of dabigatran, and it has been shown to be safe and efficacious in various patient populations, including those who require urgent surgical intervention. Conversely, andexanet alfa is indicated for the reversal of rivaroxaban and apixaban, and studies have shown that it can quickly reverse levels of these FXa inhibitors; however, because of its short half-life, it has not been well studied in patients requiring acute surgical intervention. Given that the aforementioned reversal agents are relatively new to the market, more data is needed in order to determine their optimal use. Currently, more patients are being enrolled in studies assessing the use of these agents, and newer agents are being developed (i.e. ciraparantag) which may potentially provide even more options available for providers to use in the setting of acute bleeding caused by DOACs.
Submit for CE