In Vivo Studies of Activated Porcine Factor VIII

A novel approach to assess the spontaneous gastrointestinal bleeding risk of antithrombotic agents using Apc min/+ mice

Assessment of the bleeding risk of antithrombotic agents is usually performed in healthy animals with some form of vascular injury to peripheral organs to induce bleeding. However, bleeding observed in patients with currently marketed antithrombotic drugs is typically spontaneous in nature such as intracranial haemorrhage (ICH) and gastrointestinal (GI) bleeding, which happens most frequently on top of preexisting pathologies such as GI ulcerations and polyps. Apc min/+ mice are reported to develop multiple adenomas through the entire intestinal tract and display progressive anaemia. In this study, we evaluated the potential utility of Apc min/+ mice as a model for assessing spontaneous GI bleeding with antithrombotic agents. Apc min/+ mice exhibited progressive blood loss starting at the age of nine weeks. Despite the increase in bleeding, Apc min/+ mice were in a hypercoagulable state and displayed an age-dependent increase in thrombin generation and circulating fibrinogen as well as a significant decrease in clotting times. We evaluated the effect of warfarin, dabigatran etexilate, apixaban and clopidogrel in this model by administering them in diet or in the drinking water to mice for 1–4 weeks. All of these marketed drugs significantly increased GI bleeding in Apc min/+ mice, but not in wild-type mice. Although different exposure profiles of these antithrombotic agents make it challenging to compare the bleeding risk of compounds, our results indicate that the Apc min/+ mouse may be a sensitive preclinical model for assessing the spontaneous GI bleeding risk of novel antithrombotic agents.

Impact of inhibitor epitope profile on the neutralizing effect against plasma-derived and recombinant factor VIII concentrates in vitro

The inhibitory capacity of plasma samples from 24 patients with severe haemophilia A and high-responding inhibitors were evaluated in a concentrate-based assay using two plasma-derived (Haemate and Monoclate-P) and three recombinant (Helixate, Recombinate and ReFacto) factor VIII concentrates and correlated with the corresponding epitope profile. In most, but not all, inhibitor plasmas with a relatively low reactivity against the von Willebrand-containing product Haemate, the main epitopes were located in the FVIII light chain. The reactivities within the group of recombinant products varied in that the reactivity against the B-domain deleted ReFacto was in general higher than that against Recombinate and Helixate. This difference did not correlate with any particular epitope profile and indicates that the B-domain, type of formulation and/or purification procedures may have an impact on the inhibitor reactivity in vitro. The ratio between the inhibitor titres in the concentrate-based assay and the Bethesda assay was dependent on the inhibitor plasma and concentrate used. Taken together, our results show that the reactivity of inhibitor plasmas varies considerably between different FVIII concentrates and that it does not fully correlate with the epitope profile. Potential clinical implications of the observed differences in inhibitor reactivity are discussed.

Influence of von Willebrand factor on the reactivity of human factor VIII inhibitors with factor VIII

In order to determine the difference in reactivity of factor (F) VIII inhibitors against the FVIII/von Willebrand factor (vWF) complex and against vWF-deficient FVIII, we investigated a panel of 10 antibodies to FVIII from multitransfused individuals with severe haemophilia A and other pathologies. Immunoblotting of purified FVIII and purified thrombin-cleaved FVIII revealed that in all cases inhibitor epitopes could be localized in the heavy chain (A2 subunit) while in four cases they were also present in the light chain. One of the FVIII inhibitors remained unclassified. The effect on FVIII:C of purified IgG from inhibitor plasmas was tested against a high purity FVIII/vWF concentrate and a monoclonally purified FVIII concentrate with only trace contents of vWF, by two different functional assays. Our results suggest that for those inhibitors showing A2 plus light chain (LC) reactivity, the IgG concentration required to inhibit 50% of FVIII activity in vitro is higher for the FVIII/vWF complex than for the vWF-deficient FVIII. We conclude that there might be a protective role of vWF (at least in vitro) against FVIII inhibitors with A2 and LC subunit specificity.