Emicizumab levels in patients with acquired and congenital Hemophilia A

Abstract Introduction Emicizumab is a bispecific antibody that mimics the action of FVIII and has emerged as an effective hemostatic therapy in patients with congenital hemophilia A (CHA), including CHA with inhibitors (CHAWI). Due to its hemostatic efficacy emicizumab has also become widely used in patients with acquired hemophilia A (AHA) although it is not licensed in this indication in most countries. Emicizumab levels reported from a study in patients with AHA were apparently lower than those reported in the HAVEN trials in patients with CHA including CHAWI. This single center retrospective analysis aimed at comparing emicizumab levels measured during routine clinical care of AHA, CHA, and CHAWI patients. Methods Hospital records were searched to identify all patients treated with emicizumab. Their electronic and paper files were screened and data extracted for analysis. All patients with AHA, CHA, and CHAWI were included, who received emicizumab and had at least one emicizumab level determined. Only levels were considered that were drawn during steady state, defined as ≥4 weeks after the first dose of emicizumab and not after the last dose of emicizumab. Emicizumab dosing was according to institutional guidelines for AHA (loading dose of 6 mg/kg and 3 mg/kg on days 1 and 2, respectively, followed by maintenance dosing of 1.5 mg/kg once weekly) and CHA/CHAWI (loading doses of 3 mg/kg weekly for 4 weeks, followed by 1.5 mg/kg once weekly or 3 mg/kg every 2 weeks). Emicizumab levels were measured using a modified one-stage clot assay and calibrators provided by R2 diagnostics. The limit of quantification (LOQ) was 10 µg/ml. Levels below LOQ were replaced by LOQ/2. Baseline characteristics recorded included diagnosis, age, sex, ethnicity, FVIII activity, inhibitor titer, body height, body length, body surface area, and serum albumin. Generalized mixed linear modeling (GMLM) was used to assess baseline characteristics as covariates, emicizumab levels as response variable, and patient as random factor. Effect estimates were reported together with 95% confidence intervals. Statistical significance was assumed for a p value <0.05. Results Fifty patients, all of Caucasian descent, were identified, who received emicizumab and at least one emicizumab level measured. Forty-six of 50 (92%) had emicizumab levels measured during steady state and were enrolled for this analysis. Patients with AHA (n=13) had a mean age of 75 years and a balanced sex distribution (46% female, 54% male), while patients with CHA (n=29, mean 42 years) and CHAWI (n=4, mean 18 years) were younger and always male. The median (interquartile range, IQR) of emicizumab plasma concentrations was 35 (28-41), 38 (30-49), and 39 (37-42) µg/mL in patients with AHA, CHA and CHAWI, respectively. Univariate GMLM did not reveal significant effects of age, sex, body length, body weight, or diagnosis on emicizumab levels. There was a trend towards increasing emicizumab levels with increasing serum albumin that did not reach statistical significance. The effect estimate suggested that emicizumab levels rose 0.746 µg/mL for every increase in serum albumin of 1 g/L. However, confidence intervals for this estimate were large – including zero – because of high inter-patient variation. The mean emicizumab level at steady state was similar across patient groups of hemophilia diagnosis, age, and sex. Albumin explained approximately 10% of the variation of emicizumab levels. Patients had between one and 41 emicizumab levels measured during steady state. The median and IQR of measured steady-state levels for each patient suggested moderate variation in emicizumab levels within patients. Another linear mixed effect model showed that the first measured emicizumab level explained 45% (marginal R2) of the variance in emicizumab levels. Conclusions Emicizumab levels measured in steady state were similar for patients with AHA, CHA and CHAWI and did not significantly change with age or sex. Emicizumab levels tended to increase with serum albumin. The first measured emicizumab level explained ~45% of the variation within patients suggesting that frequent measurements of emicizumab levels are not required.