Andexanet Alfa, a Universal Antidote Under Development for Factor Xa Inhibitors, Reverses Rivaroxaban-Induced Inhibition of Thrombin Generation Initiated By the Intrinsic Coagulation Pathway Independent of TFPI

Background: Andexanet alfa (AnXa) is a modified, recombinant human fXa molecule being developed to reverse the anticoagulant activity of fXa inhibitors in patients during episodes of major bleeding. As a modified fXa, AnXa retained high binding affinity to fXa inhibitors and had no significant interaction with major plasma coagulation proteins, except for tissue factor pathway inhibitor (TFPI), an endogenous fXa inhibitor with sub-nanomolar affinity to both fXa and AnXa. In previous in vitro and clinical studies in healthy volunteers, AnXa has shown dose-dependent and complete reversal of both direct and indirect fXa inhibitors in tissue factor (TF)-initiated thrombin generation (TG). To delineate the contribution of AnXa-TFPI interactions to TG, we compared rivaroxaban (Riva)-induced inhibition of TG initiated via the extrinsic pathway (TF) versus the intrinsic pathway (non-TF). The differential effect of AnXa on TG and clot formation via the two pathways was further studied in the context of recombinant tissue plasminogen activator (rtPA)-induced fibrinolysis using thromboelastography (TEG).

Methods: TF-initiated TG in human plasma was measured using a calibrated automated thrombogram (CAT) and the PPP-reagent (5 pM TF). Non-TF-initiated TG was measured using CAT and Actin FS, an aPTT reagent. Anti-fXa activity was measured using a modified anti-fXa chromogenic assay with reagents from the Coamatic Heparin kit (DiaPharma). Riva was used as the standard. Pooled human plasma was spiked with Riva (0 - 2 µM) or Riva (1 µM)+AnXa (0 - 4 µM), and TG and anti-fXa activity were measured. Clot formation with or without AnXa was measured in plasma using a TEG 5000 analyzer. The functional fibrinogen (TF) and Kaolin reagents (both from Haemonetics) were used according to the manufacturer's instruction. For low TF-initiated clot formation, the PPP-reagent (5 pM TF) was diluted in phospholipid (4.0 µM) followed by measuring the TEG profiles with 0.25, 0.5 and 1.0 pM TF.

Results: The potential contribution of AnXa-TFPI interaction to TG was studied under similar conditions using TF or Actin FS in human plasma containing AnXa, Riva or Riva+AnXa. AnXa alone had minimal effect on the endogenous thrombin potential (ETP) in either assay. AnXa was able to fully reverse Riva-induced anticoagulation in the Actin FS TG assay, independent of AnXa-TFPI interaction that primarily modulates TF activity. The modulation of TF activity could be assessed by correlating ETP vs. the anti-fXa activity in samples containing Riva or Riva+AnXa. Riva dose-dependently inhibited TF-initiated TG as anti-fXa activity increased. At similar anti-fXa levels (i.e., similar free Riva concentration), Riva+AnXa had higher ETP than Riva alone. As expected, this difference was not observed in the Actin FS TG assay.

To further investigate the role of AnXa-TFPI interaction on coagulation and fibrinolytic pathways, the profile of clot formation was studied in human plasma using TEG without Riva. AnXa (4 µM) had no effect on the TEG parameters in the kaolin assay or the functional fibrinogen assay containing high TF, with or without rtPA (150 ng/mL). When low TF (0.25, 0.5, and 1.0 pM) was used to initiate clot formation in the absence of rtPA, AnXa reduced the TEG-R parameter (lag time equivalent to clotting time), but had no effect on maximum amplitude (MA). The fibrin clot formed under each condition (±AnXa) was lysed slowly at low rtPA (75 ng/mL, ~1 nM), resulting in well-segregated processes of coagulation and fibrinolysis. However, rtPA at 150 ng/mL (~2 nM) dramatically changed the lysis profiles that overlapped the fibrin formation, possibly due to the higher rtPA activity ratio, relative to the major plasma inhibitor PAI-1 (<1 nM). With the optimal rtPA, fibrin clot formed at each TF concentration (±AnXa) was compensated by the fibrinolytic activity of rtPA resulting in an increased fibrin degradation product.

Conclusions: In the absence of a fXa inhibitor, AnXa had minimal effect on TF or Actin FS-initiated TG with no direct interaction or effect on rtPA function. AnXa dose-dependently and completely reversed Riva-induced inhibition of TG initiated by either the intrinsic or extrinsic pathway, but had different effect on ETP due to the AnXa-TFPI interaction that mainly modulates TF function. AnXa-TFPI interaction may enhance TF-initiated TG at low TF conditions with increased fibrin degradation product in the presence of rtPA.