In Vitro Spiking Comparison of Coagulation Potential Between Emicizumab and Mim8 in Whole Blood and Plasma From a Single Patient With Severe Hemophilia A Receiving FVIII Prophylaxis and Warfarin

Hemophilia A (HA) is a common hereditary bleeding disorder caused by deficiency or dysfunction of factor (F)VIII. Regular prophylaxis with FVIII products has markedly improved the quality of life for people with severe HA (PwHA) [1]. Recent studies reported that approximately 15% of PwH in the United States have cardiovascular diseases (CVDs) [2], and the prevalence of atrial fibrillation (AF) in Europe increases from 0.84% to 3.4% in those aged over 60 years [3]. Direct oral anticoagulants (DOACs), such as dabigatran and FXa inhibitors, are widely used for thromboembolic disorders and are generally safer and more effective than warfarin [4]. However, vitamin K antagonists (VKAs) remain essential for patients with mechanical heart valves or certain pediatric cases [5]. Current guidelines recommend maintaining FVIII or FIX activity (FVIII:C or FIX:C) above 20%–30% in PwH on anticoagulant therapy [6]. Emicizumab (Hemlibra), a bispecific antibody bridging FIX(a) and FX, mimics FVIIIa cofactor function. Its trough plasma concentration (∼50 µg/mL) corresponds to 10–15 IU/dL FVIII:C previously identified using an acquired HA primate model [7]. Recently, Mim8 (Novo Nordisk A/S, Bagsværd, Denmark), which exhibits higher potency in preclinical studies, was developed as the next-generation bispecific antibody [8], and is currently undergoing clinical trials. We previously demonstrated that the hemostatic activity of emicizumab is substantially reduced during warfarin treatment [9]. However, the coagulation potential in PwHA receiving Mim8 under warfarin therapy remains to be investigated. We hypothesized that Mim8 may provide sufficient hemostatic function even under warfarin therapy. Therefore, we investigated the impact of warfarin on Mim8-mediated coagulation using rotational thromboelastometry (ROTEM) and thrombin generation assay (TGA). One severe PwHA (FVIII:C <1 IU/dL) without FVIII inhibitor found severe stenosis in all three major coronary arteries by coronary angiography, and he received coronary artery bypass grafting under supplementation of recombinant FVIII product. For several months after the surgery, he continued taking warfarin orally. He received daily prophylactic short half-life FVIII concentrate at a dose of 40 IU/kg to prevent bleeding during warfarin therapy. No bleeding events occurred during this period. His whole blood samples were collected at the international normalized ratio (INR) values of 1.6, 2.4, and 3.1 either during the perioperative period or outpatient visits. Each whole blood sample containing FVIII preparations was preincubated with a polyclonal anti-FVIII inhibitor antibody (10 BU/mL) to completely neutralize FVIII:C (termed ‘FVIII-depleted’). These samples were incubated with emicizumab (50 µg/mL; therapeutic concentration) or Mim8 (5 µg/mL; expected therapeutic concentration) for 15 min at room temperature. ROTEM triggered by CaCl2 and TGA triggered by tissue factor (TF) and ellagic acid (Elg) were performed as previously described [9, 10]. In 20 healthy individuals, the mean± 1.96 standard deviation (SD) of clotting time (CT) and clot formation time (CFT) in ROTEM were 938±244 s and 1,269±377 s, respectively. In Mix (TF/Elg)-triggered TGA, those of peak thrombin (PeakTh) and endogenous thrombin potential (ETP) were 430±139 nM and 2,394±624 nM×min, respectively. FX:C, FIX:C, FVIII:C, FVII:C, and FII:C were measured by one-stage clotting assay. ROTEM data at different INR values are shown in Figure 1. In samples with INR 1.6 (FVIII:C 87 IU/dL), the FX:C, FIX:C, FVII:C, and FII:C were 43, 55, 40, and 58 IU/dL, respectively. The CT and CT+CFT in the raw sample of INR was 1,085 sec and 1,745 sec, respectively. In FVIII-depleted these samples, CT value was beyond 8,000 s (data not shown). When emicizumab was added to FVIII-depleted samples, the CT (3,232 sec) and CT+CFT (6,550 sec) were longer than those observed in raw sample. In contrast, the CT (1,348 sec) and CT+CFT (2,340 sec) in FVIII-depleted samples supplemented with Mim8 were equivalent to those of the raw sample and shorter than those of the emicizumab-added samples. At INR 2.4 (FVIII:C 23 IU/dL), the FX:C, FIX:C, FVII:C, and FII:C were 19, 43, 37, and 30 IU/dL, respectively. The CT and CT+CFT of the raw sample was 5,505 sec and 7,168 sec, respectively. The CT of FVIII-depleted samples supplemented with emicizumab exceeded 8,000 sec. In contrast, the CT (2,580 sec) and CT+CFT (6,102 sec) in FVIII-depleted samples supplemented with Mim8 were mildly shorter than those of the raw sample. At INR 3.1 (FVIII:C 27 IU/dL), the FX:C, FIX:C, FVII:C, and FII:C were 13, 32, 18 and 24 IU/dL, respectively. All of CT or CT+CFT were beyond 8,000 sec. Rotational thromboelastometry (ROTEM) data in FVIII-depleted samples supplemented with Mim8 or emicizumab at different INR conditions. The coagulation function of whole blood samples from one PwHA receiving oral warfarin therapy and FVIII product supplementation was measured at INR 1.6, 2.4, and 3.1. FVIII-depleted samples obtained by the addition of anti-FVIII polyclonal antibody were incubated with either Mim8 (5 µg/mL) or emicizumab (50 µg/mL) for 15 min at room temperature. The CT and CFT were recorded using Ca2+-triggered ROTEM (NATEM mode). CT; clotting time, CFT; clot formation time, INR; International normalized ratio, Emi; Emicizumab. TF/Elg-triggered TGA results in the various INR are presented in Figure 2. In samples with INR 1.6, the addition of Mim8 in FVIII-depleted sample exhibited lower PeakTh (86 nM) and ETP (1,127 nM×min) than raw sample contained FVIII:C (87 IU/dL) (131 nM and 1,509 nM×min, respectively), but higher PeakTh and ETP than emicizumab (33 nM and 756 nM×min, respectively). The TG potential of the raw sample, in which the FVIII inhibitor titer (Bethesda) was adjusted to approximately 2 BU/mL (corresponding FVIII:C 26 IU/dL), was also lower than that observed with Mim8. In samples with INR 2.4, the addition of Mim8 in FVIII-depleted sample showed similar PeakTh (49 nM) and ETP (841 nM×min) to raw sample contained FVIII:C 23 IU/dL (48 nM and 536 nM×min, respectively) and higher coagulation potential than emicizumab (29 nM and 503 nM×min, respectively). By contrast, at INR 3.1, FVIII-depleted samples mixed with Mim8 exhibited lower PeakTh (23 nM) and ETP (462 nM×min) than raw sample contained FVIII:C 27 IU/dL (44 nM and 610 nM×min, respectively) and similar coagulation potential to emicizumab (17 nM and 331 nM×min, respectively). Overall, these findings indicated that the coagulation potential of Mim8 was greater than that of emicizumab, and Mim8 may maintain sufficient coagulation potential equivalent to FVIII activity >20% even under oral warfarin therapy (INR 1.6–2.4). Thrombin generation assay (TGA) in FVIII-depleted samples with Mim8 or emicizumab under various INR conditions. The coagulation function of plasma samples from a PwHA receiving oral warfarin therapy and FVIII product supplementation was analyzed at INR 1.6, 2.4, and 3.1 by TF and ellagic acid-triggered TGA. FVIII-depleted samples obtained by the addition of anti-FVIII polyclonal antibody were incubated with either Mim8 (5 µg/mL) or emicizumab (50 µg/mL) for 15 min at room temperature. In sample with INR 1.6, the TG potential of the raw sample, in which the FVIII inhibitor titer (Bethesda) was adjusted to approximately 2 BU/mL (corresponding FVIII:C 26 IU/dL), was also evaluated. Thrombin generation curves are shown (red; Raw sample, green; Mim8 5 µg/mL, blue; Emi 50 µg/mL, Black; Raw sample spiked with anti-FVIII inhibitor at 2 BU/mL). INR; International normalized ratio, Emi; Emicizumab. In the present study, we demonstrated that Mim8 could maintain a higher coagulation potential under warfarin treatment, possibly due to its enhanced FIXa-mediated activation of FX [10], whereas emicizumab-induced coagulation was markedly reduced. To our knowledge, this is the first report evaluating Mim8-mediated coagulation under VKAs therapy. Warfarin inhibits the γ-carboxylation of VK-dependent coagulation factors (FII, FVII, FIX, and FX), leading to a decreased prothrombinase activity and thrombin generation. Because the cofactor activity of emicizumab depends on the availability of FIX(a) and FX, its hemostatic potential is markedly influenced by warfarin-induced depletion. In contrast, Mim8 appears to compensate for this reduction more effectively because of its enhanced FIXa-dependent activity compared with emicizumab. In this context, Mim8 may maintain sufficient coagulation potential equivalent to FVIII:C >20% even under warfarin therapy (INR 1.6–2.4). However, at INR 3.1, the coagulation function of Mim8 was comparable to that of emicizumab and lower than that of the raw sample (FVIII:C 27 IU/dL). At such high INR, the characteristic advantage of Mim8 - its strong FIXa-dependent activity—may not be fully exerted because of the markedly reduced FIX levels. Consequently, Mim8-induced coagulation potential may be reduced compared with that observed with FVIII:C of 27 IU/dL. Therefore, the hemostatic advantage of Mim8 might not be maintained beyond INR 3.0. From a clinical perspective, these findings have potential implications for the management of PwHA who require long-term anticoagulation. As life expectancy in PwHA continues to increase, the coexistence of hemophilia and cardiovascular comorbidities, such as AF and ischemic heart disease, is expected to become more common. In such settings, achieving an appropriate balance between bleeding risk and prevention of thromboembolic events remains a major clinical challenge. Although current recommendations suggest maintaining FVIII activity above 20%–30% during anticoagulant therapy, evidence supporting these thresholds in patients receiving non-factor therapies is limited. In this single-case observation, Mim8 was associated with a higher coagulation potential than emicizumab under warfarin therapy, particularly within the INR range of 1.6–2.4. This observation suggests that Mim8-treated patients may maintain a relatively preserved global coagulation potential under vitamin K antagonist therapy without additional factor supplementation in selected clinical situations. However, considering the limited data, careful clinical and laboratory monitoring remains essential, especially at higher INR levels. Several limitations should be addressed. First, the present data were derived from a single PwHA; therefore, firm conclusions cannot be drawn. Future studies with larger numbers of patients and various anticoagulant treatments, including DOACs, are needed to clarify safe and effective antithrombotic management for PwHA receiving next-generation FVIII-mimetic therapies. Furthermore, some data discrepancies were observed between ROTEM and TGA results. Nevertheless, both assays provide novel insight into the potential use of Mim8 in PwHA under warfarin therapy. Further clinical evaluation is warranted to determine whether Mim8 could provide sufficient coagulation function under antithrombotic management. Yuto Nakajima designed all experiments, interpreted the data, prepared figures, wrote and edited the manuscript and approved the final version to be published. Tomofumi Yoshimoto and Hiroaki Yaoi and Kenichi Ogiwara supported clinically. Keiji Nogami designed the research, supported the data, interpreted the data, wrote the paper, and edited the manuscript. This research was supported by research funding from Novo Nordisk, and partly by a Grant-in-Aid for Scientific Research (KAKENHI) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) to KN (24K10935) and YN (25K19236). This study was approved by the Medical Research Ethics Committee of Nara Medical University (No. 4071). Written informed consent was obtained from the participant in accordance with institutional ethical guidelines and the Declaration of Helsinki. Yuto Nakajima has received a grant from Takeda Pharmaceutical Co and Novo Nordisk A/S. Tomofumi Yoshimoto has no conflict of interest to disclose. Hiroaki Yaoi has received a grant from Novo Nordisk A/S. Kenichi Ogiwara taught a course endowed by CSL Behring and has received a grant from Novo Nordisk A/S and personal fees from Chugai Pharmaceutical Co., Ltd. and CSL Behring. Keiji Nogami has received grants, personal fees, and non-financial support from Chugai Pharmaceutical Co., Ltd.; personal fees from F. Hoffmann-La Roche Ltd.; and grants and personal fees from Sysmex Co., Takeda Pharmaceutical Co., Sanofi S. A., CSL Behring Co., KM Biologics Co., Novo Nordisk A/S, Bayer AG and is an inventor of patents relating to emicizumab. The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.