Anti-Emicizumab Antibodies Do Not Cross-React with Mim8 in Vitro

Background

Patients with severe hemophilia A may develop inhibitors against factor VIII (FVIII) in around 30% of cases. Recently, the introduction of non-replacement therapies such as emicizumab, a FVIII-mimicking agent administered as a subcutaneous injection, has revolutionized the treatment of patients with inhibitors. However, although rarely, some patients may develop antibodies against this drug. If neutralizing, these antibodies interfere with the activity of the drug, making it ineffective. Mim8 (Novo Nordisk®) is a novel experimental FVIII-mimetic human bispecific antibody that has a similar function as emicizumab, by bridging activated FIX (FIXa) and FX to activate FX, although with a different molecular structure compared to emicizumab. It is currently in phase II clinical trial for subcutaneous treatment of patients with hemophilia A with or without FVIII inhibitors (1, 2).

It is currently unknown whether the antibodies developed against emicizumab by patients with hemophilia A could cross-react with Mim8.

Aim

Our aim was to study the cross-reactivity of anti-emicizumab antibodies developed by patients with hemophilia A against Mim8 with an in-house detection method.

Methods

We studied the serum of three patients who developed anti-emicizumab antibodies. Plasma from one patient with persistent inhibiting antibodies was collected both during the treatment (thus also containing emicizumab at steady-state levels) and two years after treatment discontinuation due to inefficacy (neutralizing persistent antibodies). Plasma from two patients who developed transient antibodies against emicizumab were also tested in the course of treatment with emicizumab (non-neutralizing transient antibodies). The plate was coated both with emicizumab and with Mim8 provided by the pharmaceutical companies for research purposes. Plasma samples, diluted 1/20, were loaded into the coated wells and incubated 90 min at 37°C. The cross-reactivity to Mim8 was evaluated also by using the affinity purified anti-emicizumab IgG, which was loaded at 5 ug/mL. A properly adapted ELISA method already described (3) was used as reference assay. Then, biotinylated-emicizumab (1.5 ug/mL) or biotinylated-Mim8 (at 2 and 4 ug/mL) were added and the plate incubated 1 hour at 37°C. Moreover, a competition test was performed by using a mixture of biotinylated-emicizumab (1.5 ug/mL) and an excess of Mim8 (at 4, 8 and 40 ug/mL) in the detection phase.

Results

The Mim8 molecule - either alone, or matched with emicizumab - used both in the capture phase and in the detection phase did not bind to neither patient's plasma antibodies nor to anti-emicizumab purified IgG, which were instead revealed with the reference assay. The binding of the anti-emicizumab antibodies to biotinylated-emicizumab was not inhibited by the addition of Mim8, even at 40 ug/mL.

Conclusions

Our in-house method showed that anti-emicizumab antibodies do not react with Mim8 in vitro. Observational studies to test whether Mim8 can be used safely in patients with anti-emicizumab antibodies are needed to confirm our findings in vivo as well.

References

1. Østergaard et al. A FVIIIa-mimetic bispecific antibody (Mim8) ameliorates bleeding upon severe vascular challenge in hemophilia A mice. Blood. 2021;blood.2020010331. doi:10.1182/blood.2020010331.

2. Valsecchi C et al. J Thromb Haemost. 2021;19(3):711-718.