To the Editor: Concizumab and marstacimab have recently been introduced as humanized monoclonal antibodies that reduce tissue factor pathway inhibitor (TFPI) levels in the blood and enhance coagulation in patients with haemophilia, both with and without inhibitors (PwH and PwH-inh). Some clinical trials have demonstrated that subcutaneous prophylactic treatment with anti-TFPI antibodies significantly reduces annualized bleeding rates compared to BPAs 1, 2. Bleeding episodes, however, have been reported in some patients receiving anti-TFPI antibody therapies 1, 2. In such cases, additional haemostatic treatment with concomitant replacement clotting factors or BPAs has been shown to be effective 1. Thromboembolic episodes have been reported in some patients with haemophilia A (PwHA) receiving concomitant concizumab and rFVIIa 1. These observations underscore the need for careful monitoring of BPAs and non-factor product therapies, including anti-TFPI antibodies, in clinical practice. A plasma-derived product comprising a mixture of FVIIa and FX (pd-FVIIa/FX) at a protein weight ratio of 1:10 has been approved for congenital PwH-inh and acquired haemophilia in Japan 3, but efficacy and safety need to be clarified in PwH under low TFPI condition. The present study was designed to investigate the haemostasis mechanisms associated with the supplementation of rFVIIa, pd-FVIIa/FX, or FX in TFPI-reduced PwH model plasma using TGA measurements. Blood samples were obtained from one PwHA with a high-titer FVIII inhibitor. An anti-C2 domain alloantibody (alloAb) was obtained from him as previously reported 4. rFVIIa (NovoSeven) and pd-FVIIa/FX (Byclot) were obtained from Novo Nordisk A/S (Bagsværd, Denmark), and KM Biologics Co., Ltd. (Kumamoto, Japan), respectively. Purified plasma-derived human FX, full-length TFPI, and an anti-FIX polyclonal antibody (polyAb) were sourced from KM Biologics. Recombinant human tissue factor (rTF; Innovin, Dade, Marburg, Germany), TFPI-def plasma (BioMedica Diagnostics, Inc., Windsor, NS, Canada), thrombin-specific fluorogenic substrate (Z-Gly-Gly-Arg-AMC, Bachem, Bubendorf, Switzerland), thrombin-specific fluorogenic substrate FluCa kit and thrombin calibrator (Thrombinoscope BV, Maastricht, Netherlands) were purchased from the indicated vendors. Phospholipid (PL) vesicles containing 10% phosphatidylserine, 60% phosphatidylcholine, and 30% phosphatidylethanolamine were prepared. The TFPI-reduced PwH-inh plasma model was established in vitro as follows: TFPI-def plasma was preincubated with anti-FVIII polyAb or anti-FIX polyAb to completely neutralize FVIII activity (termed ‘FVIII-depleted’) or FIX activity (termed ‘FIX-depleted’) as previously described 4. Full-length TFPI (40 pM, corresponding to 10% of normal TFPI) was then added to simulate the TFPI-reduced PwH plasma, reflecting the reduction of mean plasma free-TFPI concentration to 10% of baseline observed with concizumab prophylaxis in PwH 1. Based on recommended BPAs doses 1, rFVIIa (2.2 µg/mL, corresponding to approximately 44 nM or 90 µg/kg administration) was added to FVIII-depleted TFPI-def plasma or FIX-depleted TFPI-def plasma. Pd-FVIIa/FX was added to FVIII-depleted TFPI-deficient or FIX-depleted TFPI-deficient plasma at FVIIa concentrations of 0.187, 0.375, and 1.5 µg/mL, corresponding to approximately 3.7, 7.5, and 30 nM, respectively (equivalent to 7.5, 15, and 60 µg/kg), and at the corresponding FX concentrations of 1.87, 3.75, and 15 µg/mL (equivalent to approximately 32, 64, and 254 nM), respectively. In addition, FX alone was added at concentrations of 65, 130, and 520 nM (corresponding to 0.5, 1, and 4 IU/mL), respectively. TGA was performed as previously described 4. Briefly, plasma samples (80 µL) were preincubated for 10 min with 20 µL of a trigger reagent containing TF and PL vesicles (final concentration; f.c. 0.5 pM, and 4 µM, respectively). Following the addition of 20 µL of a reagent containing CaCl2 and fluorogenic substrate (f.c. 16.7 mM and 2.5 mM, respectively), fluorescent signal development was monitored using Fluoroskan Ascent microplate reader (Thermo Fisher Scientific, Waltham, MA, USA). The peakTh value obtained from 20 healthy individuals were 181 ± 48 nM as previously reported 5, although these samples were not analysed simultaneously with the experimental samples. Given that concizumab prophylaxis in PwH reduces mean plasma free-TFPI concentration by 10% of baseline 1, thrombin generation potential was validated in FVIII-depleted- or FIX-depleted TFPI-def plasmas supplemented with full-length TFPI at concentrations of 0, 40, and 400 pM (corresponding to 0%, 10%, and 100% of normal levels, respectively). PeakTh values were found to be inversely proportional to TFPI concentrations in these plasmas (data not shown), consistent with previous report 6. Coagulation potential was evaluated in FVIII-depleted or FIX-depleted TFPI-reduced plasma supplemented with bypassing agents (BPAs) and FX preparations. PeakTh was significantly enhanced to normal levels with the addition of the recommended dose of rFVIIa 1. The ETP in FVIII-depleted TFPI-reduced plasma with 10% TFPI spiked with rFVIIa did not differ from that without rFVIIa, but that in FIX-depleted TFPI-reduced plasma with 10% TFPI mixed with rFVIIa was significantly greater than that without rFVIIa (Figure 1, Table 1). In contrast, supplementation with the recommended dose of pd-FVIIa/FX (1.5 µg/mL) significantly elevated both PeakTh and ETP beyond normal levels in these model plasma samples compared to unsupplemented controls (Figure 1, Table 1). Based on these findings, we speculated that the induced FVIIa/TF in FVIII-depleted or FIX-depleted TFPI-reduced plasmas converted the supplemented FX from pd-FVIIa/FX to FXa and might lead to further augment thrombin generation, possibly indicating the enhancing coagulant effect of FX supplementation. Previous studies have shown that FX (520 nM) combined with emicizumab improves coagulation potential in plasma from PwHA-inh, comparable to the effect of rFVIIa at 2.2 µg/mL 7. Accordingly, FX (520 nM) was added to FVIII- or FIX-depleted TFPI-reduced plasma at a TFPI of 10% to assess its coagulant effects. TGA revealed that FX alone augmented PeakTh beyond normal levels (Figure 1, Table 1). To determine the optimal dose of pd-FVIIa/FX or FX alone in FVIII-depleted or FIX-depleted TFPI-reduced plasmas, pd-FVIIa/FX (0.187 or 0.375 µg/mL) or FX (65 or 130 nM) at low concentration was supplemented to FVIII-depleted or FIX-depleted TFPI-reduced plasmas with TFPI 10%. The TGA revealed that the low dose of pd-FVIIa/FX (0.187 or 0.375 µg/mL) or FX (65 or 130 nM) augmented the TG potential to normal level (Figure 1, Table 1). Thrombin generation mediated by rFVIIa or pd-FVIIa/FX or FX in FVIII-depleted or FIX-depleted TFPI-deficient plasma in the presence of exogenous TFPI. FVIII- or FIX-depleted TFPI-deficient plasma samples were prepared to represent patients with PwH receiving anti-TFPI antibody therapy. The TF-triggered thrombin generation was estimated in the presence of 10% full-length TFPI. The grey bar represents the reference range obtained from normal plasma. Thrombin generation was monitored after addition of exogenous TFPI together with either rFVIIa (2.2 µg/mL, approximately 44 nM, equivalent to 90 µg/kg), pd-FVIIa/FX FVIIa: 1.5 µg/mL (approximately 30 nM; equivalent to 60 µg/kg) and FX: 15 µg/mL (approximately 254 nM), or FX alone (65, 130, or 520 nM). Experiments were performed 3–4 times and representative thrombin generation curves are shown (black, rFVIIa 2.2 µg/mL; red, pd-FVIIa/FX 1.5 µg/mL; blue, TFPI 10% alone; purple, FX 520 nM; green, FX 130 nM; orange, FX 65 nM). HA, haemophilia A; HB, haemophilia B; TFPI, tissue factor pathway inhibitor. FVIII-depleted TFPI-def plasma (HA model) FIX-depleted TFPI-def plasma (HB model) To evaluate the potential thrombotic risk, we performed additional experiments in the 10% TFPI plasma without FVIII/FIX depletion. The PeakTh in 10% TFPI plasma was already above the normal range. Additional rFVIIa or pd-FVIIa/FX did not significantly increase thrombin generation, whereas additional FX modestly enhanced PeakTh under this condition (Figure S1). These data indicate that additional FX may also contribute to the enhancement of coagulation potential in 10% TFPI plasma compared with additional rFVIIa or pd-FVIIa/FX. The coagulation effects of FX or pd-FVIIa/FX were more clearly understood when compared with those of rFVIIa. rFVIIa (2.2 µg/mL) significantly increased the PeakTh values in FVIII-depleted or FIX-depleted TFPI reduced plasmas to normal levels, indicating its effectiveness as a haemostasis regimen for breakthrough bleeding in PwH receiving anti-TFPI antibody prophylaxis. On the other hand, pd-FVIIa/FX (1.5 µg/mL) significantly increased PeakTh values above normal levels. Additionally, even the addition of low dose of pd-FVIIa/FX (0.187 or 0.375 µg/mL) or FX (65 or 130 nM) to FVIII-depleted or FIX-depleted TFPI reduced plasmas could lead PeakTh values within the normal range and those parameters were equivalent those of rFVIIa (2.2 µg/mL). The half-life of FX in plasma are ∼40 h 8, indicating that the FX alone would be useful for enhancing coagulant effect in TFPI-reduced condition. This effect may be linked to reduced FVII and FX levels in TFPI-reduced plasma due to FVIIa and FXa activation, respectively. Previous reports have indicated poor responsiveness to rFVIIa in plasma from PwHA-inh with reduced FX levels due to consumption 9. In contrast, the present results suggest that additional FX could efficiently augment the coagulation potential by enhancing the generation of FXa mediated by FVIIa/TF in the presence of pd-FVIIa/FX or FX alone. Overall, even low doses of pd-FVIIa/FX (0.187 or 0.375 µg/mL) or FX (65 or 130 nM) may be appropriate as a haemostatic treatment regimen for breakthrough bleeding in PwH receiving anti-TFPI antibody prophylaxis. Supplementation of aPCC (1.3 IU/mL) to FVIII-depleted or FIX-depleted TFPI-def plasmas was followed by TGA measurement. However, the increase in PeakTh in FVIII-depleted or FIX-depleted TFPI-def plasma spiked with aPCC was modest (Figure S2), contradicting a previous report 6. This discrepancy may stem from the presence of TFPI in aPCC. We previously demonstrated that several batches of aPCC contained total TFPI (approximately 12 ng/unit) and most of this was free full-length TFPI and that the amount of it corresponded to almost 40 % of normal free full-length TFPI level 10. In this context, the assessment of coagulation potential in 10 % TFPI plasma spiked with aPCC was not feasible because the addition of aPCC to TFPI-deficient plasma increases the TFPI level to approximately 40 %. Overall, the aPCC data in Figure S2 should be interpreted with caution because aPCC itself contains TFPI and may partially restore TFPI levels in this model. The present study has some limitations. The coagulation effect of aPCC could not be adequately assessed using the TFPI-reduced model. Additionally, anti-TFPI antibodies such as concizumab and marstacimab were not utilized due to ongoing clinical trials for marstacimab and the absence of PwH on concizumab prophylaxis in this study. Nevertheless, supplementation with FX preparation and pd-FVIIa/FX augmented thrombin generation in FVIII-depleted or FIX-depleted TFPI-reduced plasma by promoting FVIIa/TF-induced FX activation. Consequently, both pd-FVIIa/FX and FX alone may be useful for treating breakthrough bleeding in patients with PwH receiving anti-TFPI antibody prophylaxis. Careful monitoring of coagulation potential is essential when using BPAs for breakthrough bleeding in patients with PwH undergoing rebalancing therapy with anti-TFPI antibodies. Shigeharu Oh: performed the experiments, analysed the data, created the figures, and wrote the paper. Yuto Nakajima: designed all the experiments, interpreted the data, prepared the figures, wrote and edited the manuscript, and approved the final version for publication. Eisuke Takami, Hirotoshi Nakano: supervised the study. Keiji Nogami: designed the study, supported and interpreted the data, wrote the paper, and edited the manuscript. This research was partly supported 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), as well as funding from KM Biologics Co., Ltd. This study was conducted in accordance with the principles of the Declaration of Helsinki. The study protocol was approved by the Medical Research Ethics Committee of Nara Medical University (approval No. 2503). Written informed consent was obtained from the patient. Patient confidentiality was protected throughout the study. Shigeharu Oh has received grants from KM Biologics Co., Ltd. Yuto Nakajima has received grants from Takeda Pharmaceutical Co. Eisuke Takami and Hirotoshi Nakano are employees of KM Biologics Co., Ltd. Keiji Nogami has received grants, personal fees, and non-financial support from Chugai Pharmaceutical Co., Ltd.; personal fees from F. Hoffmann-La Roche Ltd.; grants and personal fees from Sysmex Co., SEKISUI MEDICAL, Takeda Pharmaceutical Co., Sanofi S.A., CSL Behring Co., KM Biologics Co., Novo Nordisk A/S, Bayer AG, Fujimoto Seiyaku; and is an inventor of patents relating to emicizumab. The datasets generated and/or analysed in the current study are available from the corresponding author upon reasonable request. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Oh et al. (Fri,) studied this question.