Promising Data on Antidotes to New Anticoagulants

December 20, 2013

New "proof of concept" data on 2 antidotes for the new oral anticoagulants have shown impressive results, with both bringing about immediate reversal of anticoagulant effect with no indication of prothrombotic complications.

The products are andexanet (Portola Pharmaceuticals), an imitation factor Xa without biological properties, which reverses the anticoagulant action of the factor Xa inhibitors (including rivaroxaban and apixaban), and a fragment of an antibody (Fab), under development by Boehringer Ingelheim, which is a specific antidote to dabigatran.

Data on andexanet were presented at the American Society of Hematology (ASH) 55th Annual Meeting. The Fab findings were presented at the American Heart Association American Heart Association (AHA) 2013 Scientific Sessions.

Commenting for Medscape Medical News, Lars Wallentin, MD, University Hospital Uppsala, Sweden, explained that no antidote is available for the new anticoagulants for patients who develop severe life-threatening bleeding.

Agents used at the moment include fresh frozen plasma and prothrombin complex concentrates and activated factor VIIa, but there is not much data on their use and they can be prothrombotic so could cause problems if too much is given, he said.

"Specific antidotes would be better. That is always the best way to reverse bleeding. Everyone would be more comfortable if they were available. We don't have any specific antidotes for antiplatelet drugs and that is always a concern."

Bleeding Risk Low

Dr. Wallentin pointed out that the bleeding risks are lower with the new anticoagulants than with warfarin.

"Even without a specific antidote, the bleeding complications are not very concerning, but of course there are always situations where an antidote would be useful, for example for a patient undergoing emergency surgery."

While Dr. Wallentin welcomes the development of these agents, he said he doesn't think they will be used that frequently.

"I think they will mostly stay on the shelves as serious bleeds with these new agents are not a great problem. But it's a bit like when you're flying. If you know there is a parachute there you feel safer."

The new data come from phase 1/2 studies in human volunteers. Dr. Wallentin noted that while some data in actual patients with bleeding would probably be required, he doesn't think large-scale trials in patients will be conducted.

"Placebo-controlled trials would be very difficult to design, but open-label studies would be possible. These discussions with regulatory authorities are going on now. I would expect these products to be on the market in 2 years time, maybe earlier if the FDA [US Food and Drug Administration] will accept most data from human volunteers."

"We showed a dose-dependent reduction in anticoagulant activity of rivaroxaban with andexanet. It had an almost immediate effect — very similar to what was seen in a previous study with apixaban," John Curnutte, MD, PhD, Portola Pharmaceuticals, told Medscape Medical News.

"The level of reversibility is highly predictable. Each molecule of andexanet will bind 1 molecule of rivaroxaban, so we can calculate how much reversal to expect, based on how much rivaroxaban is on board. We showed a high level of reversal with 4 out of 6 patients achieving 90% reversal at the 720-mg dose and this was prolonged with the additional infusion."

Dose-Dependent Effect

Dr. Curnutte explained that andexanet has been genetically engineered to exactly resemble factor Xa except that the proteolytic site has been deactivated so it cannot cleave thrombin. As a result it is not prothrombotic.

"We now have data on more than 80 human subjects, and have not seen any serious adverse effects. The current study extends our earlier work with apixaban to rivaroxaban as well. It shows that our antidote works with more than just 1 factor Xa inhibitor. It is designed to work with all factor Xa inhibitors," he said.

The current study used the same template as that for the previous apixaban study. Healthy volunteers were premedicated with rivaroxaban, 20 mg, once daily (the recommended dose for atrial fibrillation) for 6 days. Then, when they were at maximum concentrations (3 hours after the last dose), andexanet was given.

Four different dose regimens of andexanet or placebo were studied in a 6:3 ratio (9 patients per dosage group). Andexanet was administered as a slow bolus (30 mg/min) at 210, 420, 600, or 720 mg with the 720-mg bolus immediately followed by a 240-mg infusion (1-hour infusion at 4 mg/min). Pharmacodynamic and safety data were collected through day 48 with pharmacokinetic data through day 10.

Results showed that immediately after completion of andexanet administration, there was a dose-dependent reduction in anti–factor Xa activity, which returned to placebo levels by approximately 2 hours after treatment.

In parallel, the plasma concentrations of unbound rivaroxaban also decreased in a dose-dependent manner.

Table. Andexanet: Major Results

Andexanet Dose Molar Ratio of Andexanet to Rivaroxaban 2 min after Bolus Reduction in Anti–Factor Xa Activity (%) Reduction in Plasma Concentration of Unbound Rivaroxaban (%)
210-mg bolus 0.8 20 32
420-mg bolus 1.2 58 51
600-mg bolus 1.3 70 75
720-mg bolus + 240-mg infusion 1.1 81 70


In addition, rivaroxaban-induced inhibition of thrombin generation and prolongation of both prothrombin time and activated clotting time were also rapidly reversed by administration of andexanet in a dose-dependent manner.

No Evidence of Thrombotic Adverse Effects

Andexanet administration was not associated with increases in prothrombin fragments F1+2, thrombin:antithrombin, or D-dimer (all values were within normal ranges) for doses up to 420 mg and were associated with minimal increases for the 600-mg and 720-mg plus infusion regimens.

Andexanet was well tolerated, and there were no thrombotic, serious, or severe adverse events. Adverse events that occurred in more than 10% of patients and were deemed by the investigator to be related to andexanet were infusion-related reaction and postural dizziness.

Dr. Curnutte explained that the infusion given after the 720-mg dose was administered at a rate of 4 mg/min because that was the rate used in the apixaban study.

"In the apixaban study that infusion rate kept the reversal at the same high level achieved with the bolus, but with rivaroxaban a slight reduction was seen over the infusion period. So some fine tuning is needed with a slightly higher infusion rate for rivaroxaban," he said.

He noted that there were some differences in the apixaban study with the bolus dose, too.

"It took about double the dose of andexanet to reverse rivaroxaban compared to apixaban. We got up to 90% reversal of apixaban with a 400-mg dose, but we needed to go up 720 mg to see a similar effect with rivaroxaban."

Dr. Curnette explained that if the loading dose were given alone, the reversal activity would gradually decrease over the next hour, which again is faster with rivaroxaban than with apixaban.

"The length of reversal required will depend on the bleed. If a patient was undergoing surgery, they would probably need a loading dose, then an infusion for the length of the surgery. In a medical situation such as a gastric bleed, the bolus could be given first and the infusion added if more of an effect was required.

"By just giving the bolus we seem to giving enough time to allow the coagulant system to come back enough to reduce bleeding," Dr. Curnette continued. "This hiatus may be all that is required in most cases. But in more serious bleeds then the infusion can be used as well. There doesn't seem to be any downside of giving the infusion. There is no evidence of any thrombotic side effects."

Accelerated Approval Pathway

In collaboration with the FDA, Dr. Curnette said Portola has agreed to conduct phase 1 and 2 studies in healthy volunteers, and possibly phase 3 studies.

"The number of variables in bleeding patients are enormous. The FDA has designated an accelerated approval pathway for andexanet. While the agency wants some data in bleeding patients in a clinical setting this may not necessarily be part of a study. We are in discussion with the FDA as to how to proceed. We anticipate phase 3 will start in 2014 with an approval application likely in 2015. "

The data on the dabigatran Fab, presented at the AHA meeting in November, came from a phase 1/2 study that was the first investigation of whether equimolar doses of Fab would reverse the anticoagulant effects of dabigatran in humans.

"The Fab binds to dabigatran with high affinity, but because it lacks the Fc receptor it does not activate the immune system. It has no off-target toxicity, no prothrombotic activity, and it has a shorter half-life than a full monoclonal antibody — hours versus days — which allows a quicker return of anticoagulation. IV [intravenous] administration gives an immediate action," Stephan Glund, PhD, Boehringer-Ingelheim, told Medscape Medical News.

The study involved a total of 145 healthy male volunteers. In part I, participants received single rising IV doses of up to 8 g Fab. In part 2, Fab doses of 1, 2, and 4 g were administered as 5-minute IV infusion in the presence of dabigatran (220 mg twice daily for 4 days). The dose of dabigatran used was higher than given to patients with atrial fibrillation to allow for the better renal function of healthy volunteers, so similar maximum concentrations were reached. Nine participants received Fab and 3 received placebo for each of the 3 doses tested.

Results showed that all doses of Fab were safe and well tolerated. Pharmacokinetic measurements of unbound dabigatran indicated Fab binding and reversal of the anticoagulant effects of dabigatran directly after infusion.

Prolongation of clotting times induced by dabigatran was reversed to baseline at the end of the 5-minute infusion of the antidote, as consistently demonstrated by all clotting assays. There was a dose-dependent reversal with increasing doses of antidote.

Complete reversal lasted for about 30 minutes after administration of 1 g Fab. Reversal was complete and sustained in 7 of 9 participants administered 2 g and in all participants administered 4 g. In these participants, thrombin time, the most sensitive coagulation parameter, was reversed from a ratio of up to 14-fold over baseline to less than 2-fold.

The half-life of the Fab is between 14 and 17 hours, Dr. Glund noted, with clotting values returning to baseline levels after 24 hours. "With the higher doses tested, one 5-minute infusion is all that is required for complete reversal. There is no need for a longer infusion."

Prothrombotic effects were evaluated by measuring endogenous thrombin potential, predose and 15 minutes after the end of the Fab infusion. No difference in this marker was observed, showing no prothrombotic effect, Dr. Glund reported.

Safety data showed no clinically relevant drug-related adverse events and no immunogenic reactions.

Dr. Curnette is employed by Portola Pharmaceuticals and Dr. Glund is employed by Boehringer-Ingelheim. Dr. Wallentin reports that he has received consulting fees/honoraria from Bristol-Myers Squibb/Pfizer, Athera Biotechnologies, Boehringer-Ingelheim, Regado Biosciences, AstraZeneca,  Abbott, and GlaxoSmithKline.  He also reports that he has received research grants from Boehringer-Ingelheim, GlaxoSmithKline, Merck/Schering-Plough, AstraZeneca, and Bristol-Myers Squibb/Pfizer and other significant financial benefit from AstraZeneca, Bristol-Myers Squibb/Pfizer, Boehinger-Ingelheim, and GlaxoSmithKline.

American Society of Hematology (ASH) 55th Annual Meeting. Presented December 9, 2013. Abstract 336.

American Heart Association (AHA) 2013 Scientific Sessions. Presented November 18, 2013. Abstract 17765.  


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.