Reversal of the anticoagulant effect of dabigatran: idarucizumab

Drug action

Idarucizumab is a humanised monoclonal antibody fragment (Fab) with a very high affinity to dabigatran. It potently and specifically binds to dabigatran and its metabolites, thereby preventing dabigatran from exerting its anticoagulant effect. Its action is specific to dabigatran etexilate and so it will not reverse the action of other anticoagulants (idarucizumab summary of product characteristics).

Licensed therapeutic indication

Idarucizumab (Praxbind, Boehringer Ingelheim Limited) was launched in the UK in December 2015. It is licensed for use in adults treated with dabigatran etexilate (Pradaxa, Boehringer Ingelheim Limited) when rapid reversal of its anticoagulant effects is required:

Idarucizumab is for hospital use only. It can be used in conjunction with standard supportive measures, which should be considered as medically appropriate.

Course and cost

The recommended dose of idarucizumab is 5 g given intravenously as 2 consecutive infusions of 2.5 g/50 ml over 5 to 10 minutes each or as 2 consecutive 2.5 g bolus injections (idarucizumab summary of product characteristics).

A second 5 g dose of idarucizumab may be considered in the following situations:

The cost per 5 g (2×2.5 g/50 ml) dose of idarucizumab is £2,400 excluding VAT (MIMS, April 2016).

Pollack et al. 2015

Table 1 Summary of Pollack et al. 2015

Clinical effectiveness

At the time of the interim analysis of the RE-VERSE AD study, a total of 90 people had been enrolled into the study, 51 people in group A (serious bleeding), and 39 people in group B (urgent procedure). At baseline, a total of 22 people were found to have normal dilute thrombin times, and 9 of these were also found to have normal ecarin clotting times. These people had better renal function (median creatinine clearance 67 ml/minute compared with 48 ml/minute) and a longer patient‑reported time since the last dose of dabigatran etexilate (median 30.3 hours compared with 12.8 hours) than those people with abnormal clotting times at baseline. These 22 people were enrolled into the study and given idarucizumab because they met the inclusion criteria; however, they were excluded from the efficacy analysis because of their normal clotting parameters.

In the 68 people with elevated dilute thrombin time, and 81 people with elevated ecarin clotting time at baseline, the median maximum reversal of the anticoagulant effect of dabigatran etexilate was 100% (95% confidence interval [CI] 100 to 100%).

After a 5 g dose of idarucizumab, the dilute thrombin time returned to normal in 98% of people in group A who could be evaluated and 93% of people in group B who could be evaluated. The ecarin clotting time returned to normal in 89% of people in group A who could be evaluated and 88% of people in group B who could be evaluated. However, by 12 and 24 hours after idarucizumab was given, both the dilute thrombin time and ecarin clotting time started to increase in some people. At 12 and 24 hours, the dilute thrombin time was above the upper limit of normal in 10% of people in group A who could be evaluated and 19% in group B. The ecarin clotting time was above the upper limit of normal in 28% of people in group A who could be evaluated and 46% of people in group B.

The median plasma concentration of unbound dabigatran etexilate at baseline was 84 nanograms per ml (range 3 to 641 nanograms per ml) in group A, and 76 nanograms per ml (range 4 to 2880 nanograms per ml) in group B. After the first of the two 2.5 g doses of idarucizumab, the plasma concentration of unbound dabigatran etexilate was less than 20 nanograms per ml in all but 1 person, a level that the study authors (Pollack et al. 2015) report produces little or no anticoagulant effect. At 24 hours after idarucizumab was given, median plasma levels of unbound dabigatran etexilate had risen in some people and were above 20 nanograms per ml in 16/78 (21%) people who were evaluable.

In the 35/51 people in group A who could be evaluated, the median investigator‑reported time to cessation of bleeding was 11.4 hours. Three people had no bleeding assessment at baseline, and the time to cessation of bleeding could not be ascertained in 13 people with intracranial haemorrhage, gastrointestinal bleeding, intramuscular bleeding, pericardial bleeding or retroperitoneal bleeding.

In group B, administration of idarucizumab removed the need for emergency dialysis in 1 person who had taken a large overdose of dabigatran etexilate. Another 2 people remained unstable and so were unsuitable for surgical procedures despite receiving idarucizumab. In the remaining 36 people in group B who underwent urgent surgical procedures, normal intraoperative haemostasis was reported in 33 people (92%). Mildly abnormal haemostasis was reported in 2 people and moderately abnormal haemostasis in 1 person.

A total of 18/90 (20%) people died; 9 in group A and 9 in group B. Ten of the deaths were because of vascular causes including 5 fatal bleeding events. Half of the deaths occurred within 96 hours of treatment with idarucizumab and appeared to be related to the index event (3 people had an intracranial haemorrhage, 2 people had septic shock, 1 had multiorgan failure, 1 had haemodynamic collapse, 1 had respiratory failure, and 1 had cardiac arrest). The remaining deaths occurred later than this and appeared to be associated with existing comorbidities.

Blood products (such as packed red cells, fresh frozen plasma, platelets, volume expanders and pro‑haemostatic agents) were given to 33/51 (65%) people in group A and 17/39 (44%) people in group B.

Safety and tolerability

The summary of product characteristics states that idarucizumab binds specifically to dabigatran and its metabolites and therefore will not reverse the effects of any other anticoagulant. It also states that people being treated with dabigatran etexilate have underlying disease states that put them at higher risk of thromboembolic events. Reversing the anticoagulant effect of dabigatran etexilate exposes people to the thrombotic risk of their underlying disease. Therefore restarting anticoagulant therapy should be considered as soon as is medically appropriate to reduce this risk.

In the interim analysis of the RE-VERSE AD study which included 90 people, 5 people had a thrombotic event during study follow‑up (3 people in group A and 2 people in group B). Only 1 of these people had a thrombotic event within 72 hours of receiving idarucizumab; a deep‑vein thrombosis (DVT) and pulmonary embolism (PE) 2 days after treatment. Additionally 1 person had a DVT, PE and left atrium thrombus 9 days after treatment, 1 person had a DVT 7 days after treatment, 1 person had a non‑ST segment elevation myocardial infarction 13 days after treatment and 1 person had an ischaemic stroke 26 days after treatment. None of these people were receiving antithrombotic treatment when the event occurred.

After administration of idarucizumab, treatment with dabigatran etexilate can be restarted after 24 hours if the person is clinically stable and adequate haemostasis has been achieved. Other antithrombotic therapy, for example low‑molecular weight heparin, can be started at any time after administration of idarucizumab, if the patient is clinically stable and adequate haemostasis has been achieved (summary of product characteristics for idarucizumab).

According to the summary of product characteristics for idarucizumab, no adverse reactions to idarucizumab were identified when safety was evaluated in 224 healthy people as well as 123 people in the ongoing RE‑VERSE AD study. Of these 123 people, 26 people died but each of the deaths that occurred could be attributed either as a complication of the index event or associated with comorbidities. Thrombotic events were reported in 5 people (as discussed above) and each could be attributed to the person's underlying medical condition.

In the interim analysis of the RE‑VERSE AD study, Pollack et al. report that a total of 21 people experienced a serious adverse event. These included the 5 thrombotic events and 18 deaths discussed above as well as gastrointestinal haemorrhage (n=2), postoperative wound infection (n=1), delirium (n=1), right ventricular failure (n=1) and pulmonary oedema (n=1). Some people had more than 1 event.

In the European public assessment report for idarucizumab, it is highlighted that the amount of sorbitol contained in each dose of idarucizumab is very high. This is reflected in the summary of product characteristics which advises that treatment with idarucizumab must be weighed against the potential benefit of such an emergency treatment in people with hereditary fructose intolerance.

Evidence strengths and limitations

The strengths of the RE-VERSE AD study include its broad inclusion criteria, enrolling people with acute trauma and those who were expected to die within 3 days, or to require surgery within 24 hours. The RE‑VERSE AD study also included older people (mean age 76.7 years) and a high proportion of people with intracranial bleeding (20%). However the interim analysis was based on a limited data set of only 90 people, of which only 68 to 81 people were included in the efficacy analyses.

The study has inherent limitations due to its uncontrolled design and, as discussed in an editorial by Bauer (2015), the absence of a control group makes it difficult to assess the clinical benefit of idarucizumab in people with (or at risk of) bleeding associated with dabigatran etexilate. The uncontrolled design was agreed by the European Medicines Agency because the use of a comparator arm would have been unethical. There is no licensed alternative to idarucizumab and the target treatment population were treated in emergency situations where all standard supportive care treatments should be allowed (European public assessment report for idarucizumab).

The mortality rate in the RE‑VERSE AD study was high at 20%. However, participants included people with conditions such as intracranial bleeding, in whom the mortality rate is expected to be high. Half of the deaths occurred after 96 hours and appeared to be related to existing comorbidities rather than the index event. The RE‑VERSE AD study was not designed to assess the impact of idarucizumab on mortality and it is not possible to say what effect idarucizumab has on mortality compared with existing standard supportive measures alone in these high risk populations.

Five people experienced thrombotic events during the interim analysis of the RE‑VERSE AD study. However, only one of these occurred within 72 hours of receiving treatment with idarucizumab and none of the people who experienced a thrombotic event were receiving antithrombotic therapy at the time of the event. The European public assessment report for idarucizumab states that the occurrence of these events after reversal of the anticoagulant effect was not unexpected, and was not linked to the action of idarucizumab. The summary of product characteristics for idarucizumab states that if the patient is clinically stable and adequate haemostasis has been achieved, dabigatran etexilate treatment can be re‑initiated 24 hours after administration of idarucizumab. Other antithrombotic therapy (such as low‑molecular weight heparin) can be started at any time, if the patient is clinically stable and adequate haemostasis has been achieved.

The European public assessment report for idarucizumab states that the chosen end points in the RE‑VERSE AD study to demonstrate the ability of idarucizumab to reverse the effect of dabigatran etexilate were suitable. However, the primary outcome, reversal of the anticoagulant effect, is a surrogate for clinical efficacy. The clinical benefit of this outcome depends on the individual patient, clinical situation, disease or bleeding severity and location of the bleeding. Secondary clinical outcomes reported in the study were median investigator‑reported time to cessation of bleeding and intraoperative haemostasis. The European public assessment report for idarucizumab reports that cessation of bleeding was difficult to assess in some people (such as those with intracranial bleeding) because the investigators could not visualise or identify the bleeding site. Also cessation of bleeding was subjective and based upon whatever the investigator could visualise or measure.

Other planned secondary clinical outcomes, such as severity and extent of bleeding, and for those with intracranial haemorrhage, comparison of the baseline score on the modified Rankin scale with the score at 90 days, were not reported in the interim analysis because of the small number of people with available follow‑up data.

Nearly a quarter of people enrolled into the RE‑VERSE AD study (who were given idarucizumab because they met the inclusion criteria) were subsequently found to have a normal dilute thrombin time at baseline and 10% were found to have a normal ecarin clotting time. These people were excluded from the efficacy analyses, further limiting the data set on which the efficacy analyses were based. An editorial by Bauer (2015) suggests that this group of people would have little or no circulating anticoagulant in their blood and would not be expected to benefit from idarucizumab. The editorial suggests that it may be useful to have procedures in place to ensure activity measurements for dabigatran etexilate are evaluated where the clinical situation allows before idarucizumab is given. However, the summary of product characteristics for idarucizumab has no requirement for clotting parameters to be measured.

Blood products (such as packed red cells, fresh frozen plasma, platelets, volume expanders and pro‑haemostatic agents) were given to 65% of people in the serious bleeding group and 44% of the urgent procedure group. Therefore people who are given idarucizumab may still require other treatments to manage their bleeding.

Although after a 5 g dose of idarucizumab the dilute thrombin time and ecarin clotting time returned to normal in about 90% or more of people who could be evaluated, in some people these values had deteriorated by 12 and 24 hours. At these time points, the dilute thrombin time was above the upper limit of normal in 10–19% of people who could be evaluated and the ecarin clotting time was above the upper limit of normal in 28–46% of people who could be evaluated. Additionally, the median plasma concentration of unbound dabigatran, which had fallen to a level that produces little or no anticoagulant effect after the first of the two 2.5 g doses idarucizumab, had risen again in some people at 24 hours after idarucizumab. Pollack et al. suggest that these findings may reflect the redistribution of extravascular dabigatran into the intravascular compartment. The summary of product characteristics for idarucizumab states the clinical situations where consideration of an additional 5 g dose of idarucizumab may be considered in people with prolonged clotting times.

Alternative treatments

There are currently no other licensed agents to reverse the anticoagulant effect of dabigatran etexilate (or any other NOAC).

The dabigatran etexilate summary of product characteristics states that in the event of haemorrhagic complications, treatment must be stopped and general supportive measures such as surgical haemostasis and blood volume replacement carried out. In people undergoing surgery or invasive procedures, temporarily stopping dabigatran etexilate may be required. The decision to stop dabigatran etexilate and when to stop it will depend on the person's risk of having a thromboembolic event, their renal function and the bleeding risk associated with the procedure (see the dabigatran etexilate summary of product characteristics for more information).

The dabigatran etexilate summary of product characteristics now also states that in situations when rapid reversal of the anticoagulant effects of dabigatran are required (for emergency surgery, urgent procedures, or life‑threatening or uncontrolled bleeding) the specific reversal agent idarucizumab is available.

Costs of alternative treatments

There are no other licensed agents to reverse the anticoagulant effect of dabigatran etexilate. Managing haemorrhagic complications and reversing anticoagulant effects is multifactorial and various agents are used. The cost of alternative treatments will depend on the preparation chosen, and it is not possible to provide a comprehensive list of costs for alternative treatments in this evidence summary. The summary of product characteristics for idarucizumab states that idarucizumab can be used in conjunction with standard supportive measures, which should be considered as medically appropriate.

The cost per 5 g (2×2.5 g/50 ml) dose of idarucizumab is £2,400 excluding VAT (MIMS, April 2016).

Likely place in therapy

The interim analysis of the RE‑VERSE AD study (Pollack et al. 2015), demonstrated that idarucizumab completely reversed the anticoagulant effect of dabigatran etexilate. Idarucizumab is licensed for use in adults when rapid reversal of dabigatran etexilate's anticoagulant effects is required for emergency surgery or urgent procedures, or in life‑threatening or uncontrolled bleeding.

Idarucizumab is likely to be an important treatment option for people with life‑threatening bleeding and those who need urgent surgery associated with a bleeding risk. The decision on whether to treat people with idarucizumab will need to be made on an individual patient basis, taking into account factors such as the location, size and severity of the bleeding (or the risk of bleeding associated with the surgical intervention or invasive procedure), the thrombotic risk of their underlying disease, and any comorbidities the person has.

Dabigatran etexilate has a half‑life of 12–14 hours in people with normal renal function, but this is increased in people with renal impairment. Therefore the timing of the last dose of dabigatran etexilate is likely to be a factor in whether reversal of its anticoagulant effect is needed. The bleeding risk from elective surgery can generally be managed by stopping dabigatran etexilate between 24 and 96 hours before a surgical procedure, with timing dependent on the person's risk of having a thromboembolic event, their renal function and the bleeding risk associated with the procedure. Idarucizumab is not indicated for elective surgery, only for emergency surgery or urgent procedures where the surgery cannot be delayed (see the summaries of product characteristics for dabigatran etexilate and idarucizumab for more information).

People who are treated with idarucizumab may still need other supportive measures, for example blood products, to manage their bleeding. The summary of product characteristics for idarucizumab states that idarucizumab can be used in conjunction with standard supportive measures, which should be considered as medically appropriate.

The summary of product characteristics states that a second dose of idarucizumab may be considered in people with a recurrence of clinically relevant bleeding together with prolonged clotting times, in people in whom potential re‑bleeding would be life‑threatening who also have prolonged clotting times, or in people requiring a second emergency procedure who have prolonged clotting times. This would incur an additional cost of £2,400 excluding VAT per 5 g dose (MIMS, April 2016).

Estimated usage

The manufacturer of idarucizumab (Praxbind, Boehringer Ingelheim Limited) estimates that approximately 0.98% of people per year may need emergency surgery whilst they are being treated with dabigatran etexilate, and therefore may be eligible for treatment with idarucizumab. Similarly they estimate that 1.34% of people per year may experience life‑threatening or severe bleeding during treatment with dabigatran etexilate that may warrant the use of idarucizumab. These figures are based on estimates from the RE‑LY study (Connolly et al. 2009) which compared dabigatran etexilate with warfarin in people with atrial fibrillation.

Expert advisers

Dr Joe Martins, Consultant Cardiologist, Medical Service Head for Cardiology, The Dudley Group NHS Foundation Trust

Professor Martin R Cowie, Professor of Cardiology, Imperial College London

Dr Will Lester, Consultant Haematologist, University Hospital Birmingham, Queen Elizabeth

Daniel Horner, Consultant in Emergency Medicine and Intensive Care, Salford Royal NHS Foundation Trust

Declarations of interest

Joe Martins has received speaker honoraria, payments for consultancy on an advisory board and educational sponsorships for presenting at conferences from Bayer.

Martin Cowie has provided consultancy advice, and been paid honoraria for educational events for Boehringer Ingelheim, Bayer, Pfizer/BMS and Daiichi‑Sankyo. He is on the medical advisory committee of the Atrial Fibrillation Association (non‑fee paying) and chaired groups for NICE on the patient decision aid for the NICE atrial fibrillation guidelines, and the implementation collaborative on NOACs (both non‑remunerative).

Will Lester has received speaker honoraria from Boehringer Ingelheim, Bayer, Bristol Myers Squibb, and Pfizer. He has been on advisory boards for Boehringer Ingelheim, Bayer, Bristol Myers Squibb, Daiichi Sankyo and Pfizer. He has also received support from Boehringer Ingelheim to attend a scientific meeting.

Daniel Horner has no relevant interests to declare.

Copyright

© National Institute for Health and Care Excellence, 2016. All rights reserved. NICE copyright material can be downloaded for private research and study, and may be reproduced for educational and not‑for‑profit purposes. No reproduction by or for commercial organisations, or for commercial purposes, is allowed without the written permission of NICE.

ISBN: 978‑1‑4731‑1862‑1