Table 3 Published studies evaluating interventions for the reversal of new oral anticoagulants

In vitro studies and in vivo studies in animal models

Escolar et al. [50]

Apixaban

rFVIIa (270 μg/kg)

aPCC (75 IU/kg)

PCC (50 IU/kg)

In vitro study in health donor samples spiked with apixaban at a concentration of 200 ng/mL assessing TG and TE. Additional studies assessed hemostasis under flow conditions.

TG parameters were variably improved in order of efficacy by PCC, aPCC, and rFVIIa. TE parameters were corrected in order of efficacy by rFVIIa, aPCC, and PCC. Flow studies (resembling more closely a bleeding situation) showed improvements in order of efficacy by rFVIIa, PCC, and aPCC.

Martin et al. [34–36]

Apixaban

rFVIIA, PCC, fibrinogen concentrate. Doses not reported.

Rabbit in vivo bleeding model (hepatosplenic section) assessing BT, PT, TE, TG, and blood loss. Apixaban dose was not reported.

All agents improved laboratory parameters but did not reduce blood loss.

Chan et al. [23]

Dabigatran

rFVIIa (150 μg/kg)

aPCC (FEIBA VH, 100 IU/kg)

In vitro study using healthy donor samples spiked with dabigatran 150 ng/mL assessing TE

Both agents improved TE clot initiation time. aPCC achieved a more potent correction than rFVIIa.

Hoffman et al. [55]

Dabigatran

PCC (Beriplex) 0.25-2.0 IU/mL)

In vitro studies in human plasma spiked with dabigatran 189-944 ng/mL assessing TG

PCC corrected some parameters in the TG assay.

Lange et al. [51]

Dabigatran

Activated charcoal perfusion and hemodialysis

In vitro and in vivo (pig model) assessment of Absorba 300C activated charcoal filter or Polyflux 140H hemodialysis filter. Dabigatran tested in porcine whole blood at a concentration of 1,000 ng/mL.

Activated charcoal perfusion resulted in near-complete removal of dabigatran but showed saturation (maximum binding capacity ~30 mg). Hemodialysis with flows achieved similar results.

Pragst et al. [40]

Dabigatran

PCC (Beriplex P/N) 20, 35, or 50 IU/kg

Rabbit in vivo bleeding model (renal incision) assessing blood loss and time to hemostasis after dabigatran 0.4 mg/kg

Dose-dependent reduction of blood loss and time to hemostasis. Blood loss normalized at a dose of 50 IU/kg.

Toth et al. [48]

Dabigatran

Anti-dabigatran humanized Fab (30, 90, 175 mg/kg)

In vivo model in Rhesus monkeys receiving 12 mg/kg per day × 4 days. Coagulation activity was assessed by dilute TT.

Complete correction of TT at all 3 doses.

Van Ryn et al. (a) [41, 43, 44]

Dabigatran

PCC (Beriplex 35 IU/kg)

PCC (Octaplex 40 IU/kg)

Activated PCC (FEIBA 100 IU/kg)

rFVIIa (Novoseven 0.5 μg/kg)

Rat in vivo bleeding model (tail cut) assessing time to hemostasis, aPTT, TT, PT, ECT, diluted TT, and Hemoclot. Dabigatran was administered at 30 mg/kg.

All agents tested completely corrected prolongation of BT up to 2 hours.

TT, aPTT, and ECT remained prolonged after all four agents.

PT was reversed to baseline with all agents.

Van Ryn et al. (b) [13, 45, 54]

Dabigatran

Anti-dabigatran antibody and Fab fragment

In vitro study in human plasma and rats ex vivo using diluted TT. In vivo rat tail bleeding model. Dabigatran was administered at 30 mg/kg.

In vitro studies showed rapid inhibition of dabigatran. BT in rat tail bleeding model was fully corrected.

Van Ryn et al. (c) [20, 42]

Dabigatran

Activated charcoal hemoperfusion

In vitro assessment of Adsorba cartridge. Dabigatran tested in bovine whole blood at a concentration of 1,000 ng/mL.

Activated charcoal hemoperfusion resulted in near-complete removal of dabigatran.

Zhou et al. [46]

Dabigatran

PCC (Beriplex P/N; 50 and 100 IU/kg)

rFVIIa (NovoSeven; 8.0 mg/kg)

FFP (murine, 200 μL/kg)

In vivo and ex vivo study in mice receiving dabigatran at 2.25, 4.5, and 9.0 mg/kg assessing tail vein bleeding time and intracerebral hemorrhage volume and expansion (intrastriatal collagenase injection model)

At 9.0 mg/kg of dabigatran PCC (50 and 100 IU/kg) but not rFVIIa reduced intracerebral hematoma growth and mice mortality. FFP had an inconsistent beneficial effect on hematoma size reduction but not on mortality at lower dabigatran doses. PCC reduced tail vein BT more effectively at a dose of 100 IU/kg.

Godier et al. [27–29]

Rivaroxaban

rFVIIa (150 μg/kg)

PCC (Kaskadil 40 IU/kg)

Rabbit in vivo bleeding model (hepatosplenic section) assessing BT, TE, TG, and blood loss. Rivaroxaban was administered at 5 mg/kg.

rFVIIa corrected BT but not blood loss. PCC did not correct BT or blood loss. PCC and rFVIIa decreased aPTT and TE clotting time.

Gruber et al. [30, 31]

Rivaroxaban

rFVIIa (NovoSeven; 210 μg/kg)

aPCC (FEIBA; 50 IU/kg)

In vivo study in baboons receiving rivaroxaban (IV bolus 0.6 mg/kg followed by continued infusion 0.6 mg/kg per hour) assessing PT, TE, and BT

aPCC reduced PT and normalized BT. fVIIa reduced but not fully corrected BT and PT.

Hollenbach et al. [53]

Rivaroxaban

r-Antidote (PRT064445) (76 mg/rabbit)

rFVIIa (150 μg/kg)

In vivo rabbit liver laceration model assessing blood loss, anti-fXa activity, PT, and aPTT. Rivaroxaban was given at 1 mg/kg.

r-Antidote (PRT064445) reduced blood loss anti-fXa activity, PT, and aPTT. rFVIIa improved PT and aPTT but did not decrease blood loss.

Keller et al. [32]

Rivaroxaban

rFVIIa 90 and 180 μg/kg

In vitro study using healthy donor samples spiked with rivaroxaban 800, 2,000, or 5,000 ng/mL assessing TE

Both dosages improved TE parameters at all levels of rivaroxaban.

Lloyd et al. [47]

Rivaroxaban

High-purity factor × concentrate (1, 2.5, and 5 IU/mL)

In vitro study using commercial rivaroxaban calibration plasmas spiked with fX assessing PT

fX reduced but not completely corrected PT. No dose-response was observed.

Lu et al. [11, 12]

Rivaroxaban

r-Antidote (PRT064445) (0.96 μmg/mouse)

In vitro and ex vivo studies in mouse, rat, and human plasma assessing anti-fXa activity and TG. Blood loss assessed in a mouse tail transection model. Rivaroxaban was administered at 50 mg/kg.

r-Antidote (PRT064445) decreased blood loss, anti-fXa activity and whole blood INR.

Olesen et al. [37]

Rivaroxaban

rFVIIa (1.0 and 2.0 μg/mL)

PCC (0.29 and 0.58 IU/mL)

fIX/fX concentrate (0.29 and 0.58 IU/mL)

In vitro study in health donor samples spiked with rivaroxaban at a concentration of 1.33 μg/mL assessing TE

rFVIIa, PCC, and fIX/fX concentrate improved TE parameters but no complete reversal was obtained.

Perzborn et al. [38, 39]

Rivaroxaban

PCC (Beriplex) 25 or 50 IU/kg

Rat in vivo bleeding model (mesenteric) assessing bleeding time, PT, and TAT after rivaroxaban 2 mg/kg

PCC at 50 IU/kg nearly normalized BT, improved PT, and decreased TAT levels. PCC at 25 IU/kg had no effect.

Studies in humans receiving new oral anticoagulants

Eerenberg et al. [24–26]

Rivaroxaban

Dabigatran

PCC (Cofact; 50 IU/kg)

Randomized controlled trial in healthy volunteers receiving either dabigatran (150 mg bid po) or rivaroxaban (20 mg bid po) × 2.5 days after which PCC (or placebo) was given by IV infusion. Rivaroxaban was assessed by PT and ETP. Dabigatran was assessed by aPTT, ETP, TT, and ECT.

For rivaroxaban, PCC infusion normalized PT and ETP.

For dabigatran, PCC did not correct aPTT, ETP lag time, TT, or ECT.

Galan et al. [49]

Dabigatran

Rivaroxaban

rFVIIa (270 μg/kg)

aPCC (75 IU/kg)

PCC (50 IU/kg)

Ex vivo studies using samples from healthy individuals receiving dabigatran (150 mg po bid) or rivaroxaban (20 mg po). Samples spiked with reversal agents were tested for TG, TE, PT, INR, and aPTT. Additional studies assessed hemostasis under flow conditions.

rFVIIa and aPCC corrected the effect of rivaroxaban on PT, INR and aPTT. aPCC corrected rivaroxaban induced abnormalities in TG whereas PCC and rFVIIa had modest or no effects, respectively.

aPCC corrected the effect of dabigatran on aPTT. Flow studies (resembling more closely a bleeding situation) showed that alteration induced by rivaroxaban were variably reversed by all three agents whereas only aPCCs reversed the effects of dabigatran.

Marlu et al. [33]

Dabigatran

Rivaroxaban

rFVIIa (NovoSeven; 20, 60, and 120 μg/kg)

aPCC (FEIBA; 20, 40, and 80 IU/kg)

PCC (Kanokad; 12.5, 25, and 50 IU/kg)

Crossover randomized ex vivo study in healthy donors receiving either dabigatran (150 mg po) or rivaroxaban (20 mg po) × 1 dose assessing TG, aPTT, and PT

For rivaroxaban PCC strongly corrected ETP-AUC and modestly corrected ETP-Peak. No correction of lag time. rFVIIa corrected only kinetic parameters. aPCC corrected all parameters.

For dabigatran, PCC improved ETP-AUC but had no effect on kinetic parameters. rFVIIa and aPCC corrected lag time.

Pilliteri et al. [52]

Dabigatran

Rivaroxaban

PCC (Beriplex; 12.5, 25, 50, and 100 IU/kg

PCC (Cofact; 12.5, 25, 50, and 100 IU/kg)

aPCC (FEIBA; 20, 40, 80, and 160 IU/kg)

rFVIIa (NovoSeven; ~75, 125 250 and 500 μg/kg)

Crossover randomized ex vivo study in health donors receiving either dabigatran (150 mg po) or rivaroxaban (20 mg po) × 1 dose assessing TG, aPTT, and PT, TT, rivaroxaban, and dabigatran levels and ECT.

aPCC and rFVIIa showed reversal of both anticoagulants assessed by TG. Cofact showed improvement on TG but Beriplex did not.