Pre-Analysis for Hemostatic Effect of Bypassing Agents in Hemophilia a Patients with Inhibitors Under the Emicizumab Prophylaxis By Global Assays

Introduction: Emicizumab (Emi) is an anti-factor (F)IXa/FX bispecific antibody that mimics FVIIIa cofactor function. The clinical trials on Emi prophylaxis for hemophilia A patients with inhibitor (HA-Inh) demonstrated marked reduction in bleeding rates. However, Emi at clinically relevant plasma concentrations (~50 mg/ml) corresponding to ~15% of FVIII equivalent activity may not provide enough hemostasis at severe bleeding or major surgical intervention. In these circumstances, additional infusions of bypassing agents (BPAs) may be needed for hemostatic management. In the HAVEN 1 study, thrombotic events/ thrombotic microangiopathy occurred in Emi-treated HA-Inh receiving consecutive infusions of aPCC in higher doses. After these reports, guidance was provided to avoid the use of aPCC as first choice during Emi prophylaxis. If aPCC is the only available BPA, use of the lowest dose expected to achieve hemostasis with the initial dose ≦50 U/kg was recommended. To further support these recommendations, more investigation is required for the use of BPAs under Emi prophylaxis. Objective: We examined the coagulation effects by spiking BPAs to whole blood or plasma samples from Emi-treated HA. Methods: Samples from eight Emi-treated HA-Inh and two hemophilia A patients without inhibitors (HA-Inh(-)) in phase 1/2 and HAVEN 1 were tested. The Emi doses were 0.3-3.0 mg/kg/w or 1.5 mg/kg/w, respectively. Either aPCC or rFVIIa was spiked to patients' whole bloods or plasmas, and assessed by rotational thromboelastometry (ROTEM; Ca²⁺ trigger) and clot waveform analysis (CWA; ellagic acid/tissue factor-mixed trigger; Nogami, JTH2018). Furthermore, samples in Emi-treated HA-Inh of phase 1/2 (n=1) or HAVEN 2 (n=2) receiving BPAs infusion for hemostatic management were assessed by both assays. Results and Discussion: The baseline levels, before spiking BPAs, in ROTEM parameters, such as clot time (CT) and clot formation time (CFT), of Emi-treated HA were 1,380/521sec(s) (reference range 762-1,127/207-511s), respectively. Spiking aPCC markedly shortened CT/CFT to 167/81.5s and 148/78.0s at 0.65 and 1.3 U/mL corresponding to 50 and 100 U/kg infusion, respectively. Even at 0.13 U/mL (10 U/kg), CT/CFT shortened to 290/92.6s, which was shorter than that of reference range. Its effect was dose-dependent. Spiking rFVIIa also improved CT/CFT to 726/179s, 551/141s at 32.3 and 112 μg/mL (90 and 270 μg/kg). In CWA, the baseline levels of adjusted-|min1| (ad|min1|), indicative of maximum coagulation velocity, in Emi-treated HA were 5.53 (normal control 7.98±0.24). Ad|min1| was improved to 6.48 and 8.04 by spiking aPCC (10 and 100 U/kg), or 7.11 and 7.12 by spiking rFVIIa (90 and 270 μg/kg), respectively. For rFVIIa its effect was not dose-dependent. According to the investigation for in vivo effect by BPAs infusion in a total of 9 treatment events of Emi-treated HA-Inh, CT/CFT (1,812/553s) was improved to 1,151/346s, of which was within normal range. Ad|min1| (5.7) was improved to 6.6 30 min(m) after aPCC infusion (44-74 U/kg). By rFVIIa infusion (90-119 μg/kg), CT/CFT (1,962/758s) improved to 912/207s, and ad|min1| (4.9) also improved to 6.5 30m after infusion, of which improvement remained within normal range. The hemostasis effects were clinically satisfying and no thrombosis occurred in all cases. The ROTEM parameters by spiking aPCC were apparently hypercoagulant relative to those by infusion at same dose, supporting our previous report (Furukawa JTH2015). The shortening effect on the ROTEM parameters by spiking aPCC 100 U/kg in Emi-untreated HA seemed to be similar to that by spiking aPCC 10 U/kg in Emi-treated HA. In addition, ad|min1| by spiking aPCC 10 U/kg with Emi also improved as well as ~100 U/kg without Emi in our previous report (Nogami JTH2018). Taken together, it is suggested that the hemostatic effect of aPCC infusion (10 U/kg) to Emi-treated HA could be comparable to that of 100 U/kg to Emi-untreated HA-Inh, though there is a limitation to predicting the hemostatic effect of aPCC in a clinical setting from these ex vivo results. Conclusion: Spiking tests in Emi-treated HA-Inh by ROTEM and CWA are useful to evaluate clinical coagulation potentials and dose finding of BPAs in Emi-treated HA receiving BPAs. Additionally, under Emi prophylaxis, a lower dose of aPCC infusion might be a possible option to treat breakthrough bleed in Emi-treated HA-Inh.