Preclinical Evaluation of NXT007 Using in Vitro and In Vivo Models of Hemostasis and Thrombosis

Introduction NXT007 is a next generation factor (F)VIIIa-mimetic bispecific antibody that bridges FIXa and FX to restore intrinsic tenase activity in people with hemophilia A (PwHA), currently in Phase 1/2 trials. The molecule was developed by optimizing the framework of emicizumab to increase FVIIIa-mimetic activity and half-life, with the goal of improving potency, efficacy and dosing convenience. In preclinical studies, NXT007 reached a non-hemophilic level of thrombin generation (TG) in hemophilia A (HA) plasma (comparable to 40-100% FVIII) and was more potent in a non-human primate model of HA, compared with historical emicizumab data. The current study evaluated NXT007 compared with emicizumab, using additional in vitro and in vivo preclinical models of hemostasis and thrombosis. Results NXT007 and emicizumab increased tissue factor (TF)-triggered peak height TG when spiked into platelet-rich and platelet-poor HA-like (FVIII-neutralized) plasma, with bell-shaped dose-response profiles. NXT007 was more potent than emicizumab, peaking at lower concentrations and with higher maximum effect (comparable to 100% porcine FVIII [pcFVIII]). Maximum peak height TG by emicizumab was achieved at ~30-fold higher concentration (similar to 40% pcFVIII) compared with NXT007. Shorter tissue plasminogen activator-mediated fibrinolysis times were observed in congenital HA plasma compared with normal plasma, indicating, as expected, that unstable clots are susceptible to breakdown. NXT007 and emicizumab delayed fibrinolysis in a dose-dependent manner, with NXT007 having ~20-fold more potent antifibrinolytic effect. Fibrinolysis times correlated with peak height TG in the same congenital HA plasma, and both parameters had similar dose-response profiles, suggesting a functional interaction between NXT007-driven TG and resistance to fibrinolysis. NXT007 and emicizumab corrected clotting times and kinetics of HA-like (FVIII-neutralized) blood, measured by rotational thromboelastometry (ROTEM), with NXT007 being ~23-fold more potent. In collagen/TF-coated flow chambers perfused with HA-like blood at arterial shear rates (750s-1), NXT007 and emicizumab increased fibrin deposition without increasing platelet adherence (thrombus size), assessed by confocal microscopy. Maximum effect of NXT007 occurred at 10 µg/mL (similar to 100% pcFVIII), after which the procoagulant effect plateaued. For emicizumab, maximum effect was achieved at three times higher concentration (similar to 15% pcFVIII). HA mice (supplemented with human FIX/FX) required high doses of emicizumab (10-20 mg/kg, plasma concentrations ~100-200 µg/mL) to achieve significant reduction in total bleeding time following tail vein transection. NXT007 was more potent in controlling bleeding, with lower doses (1-5 mg/kg, plasma concentrations ~10-50 µg/mL) significantly reducing total bleeding time and/or blood loss. Bleeding pattern analysis showed that mice given NXT007 had shorter times to bleeding arrest and lower rates of spontaneous rebleeding following bleeding rechallenge, compared with emicizumab. In the ferric chloride carotid injury model of arterial thrombosis, administration of 150 IU/kg recombinant FVIII (corresponding to 300% plasma FVIII:C) to HA mice (supplemented with human FIX/FX) increased vessel occlusion by 75%, indicating the model was sensitive to hypercoagulation. Doses of NXT007 and emicizumab of 2-20 mg/kg (plasma concentrations ~20-200 µg/mL) had no effect on maximum blood flow reduction, indicating that they do not present a prothrombotic profile in this model. Conclusions In this preclinical study, a range of in vitro and in vivo assays with complementary endpoints were used to evaluate the activity of NXT007. Assays were benchmarked against emicizumab, the safety and efficacy of which has been established through a comprehensive clinical development program. The increased FVIIIa-mimetic activity of NXT007 translated to significantly increased potency in TG, clotting, fibrinolysis, and bleeding assays. Further, there was no evidence of hypercoagulability or prothrombotic effects at anticipated therapeutic or supratherapeutic concentrations. Overall, our preclinical data support the ongoing clinical evaluation of NXT007 and suggest that it has the potential to bring even greater benefit to PwHA.