A Population Pharmacokinetic Analysis of Emicizumab in Patients with Acquired Hemophilia a Enrolled in a Phase III Study (AGEHA)

Abstract

INTRODUCTION: Emicizumab is a bispecific antibody mimicking the cofactor function of activated factor VIII. A population pharmacokinetic (PK) analysis was performed to characterize the PK of emicizumab in patients with acquired hemophilia A (PwAHA) leveraging a model that had been developed in patients with congenital hemophilia A (PwCHA) [1]. METHODS: Plasma emicizumab concentration data from 12 PwAHA, who received emicizumab subcutaneously in a phase III study (AGEHA) and were included in its primary analysis [2], were analyzed by a nonlinear mixed-effect modeling approach. A linear one-compartment model with first-order absorption and elimination was employed for the structural model. Covariate investigation was limited to minimal given the possibly poor power for identifying right covariates and their right effects due to the much smaller sample size than in PwCHA (N = 381). A full covariate modeling approach was applied to include and infer the covariate effects that had been identified to be statistically significant in PwCHA, i.e., body weight (BW) on the apparent clearance (CL/F) and apparent volume of distribution (V/F), age on the bioavailability (parameterized as relative bioavailability [Frel] to a typical age), and albumin level (ALB) on the CL/F, except for race on the V/F because all patients in this analysis were Asian, without statistical testing for inclusion or exclusion. Because the age and ALB (50 to 92 years and 20 to 40 g/L, respectively) included some values out of the ranges that had been evaluated in PwCHA (1.22 to 77 years and 33 to 57 g/L, respectively), these covariate effects were separately estimated for ≤ 77 and > 77 years and for < 33 and ≥ 33 g/L, respectively, without statistical testing for separation or integration. Other covariate effects that had been identified not to be statistically significant in PwCHA were not repeatedly explored in this analysis. A stepwise covariate modeling approach was applied to explore the effects of a potential new covariate of sex, which had not been tested in PwCHA because all patients were male, on the CL/F, V/F, and absorption rate constant (KA) with statistical testing for inclusion and exclusion. Effects of renal and hepatic functions (normal to severely impaired and normal to moderately impaired, respectively; no patients had severe hepatic impairment) on the CL/F, V/F, and KA were explored graphically in a post hoc manner. Parameter estimates were corrected as appropriate when compared with PwCHA. RESULTS: In PwAHA, all prior identified covariate effects within the ranges previously evaluated in PwCHA exhibited consistent directions (positive or negative) in the point estimates with those in PwCHA, and their 95% confidence intervals (CIs) involved the point estimates in PwCHA. The effects of age > 77 years and ALB < 33 g/L had 95% CIs involving not only the point estimates for ≤ 77 years and ≥ 33 g/L, respectively, but also the null value. Effects of sex were not statistically significant. No consistent effects were observed with the severity of renal or hepatic impairment. The 95% CIs of the CL/F/Frel and V/F/Frel in PwAHA involved the point estimates in PwCHA regardless of the presence or absence of covariate effect correction, indicating no clear effects of being a PwAHA itself. The 95% CI of the KA in PwAHA fell below the point estimate in PwCHA, indicating a possibly slower absorption than in PwCHA; however, its impact on the steady-state exposure was predicted to be limited. Simulations validated the observed achievement of the efficacious mean steady-state trough concentration of > 30 μg/mL with the tested maintenance dose of 1.5 mg/kg once weekly in AGEHA as in PwCHA [2,3], supporting the appropriateness of applying the maintenance dose for PwAHA. CONCLUSIONS: No remarkable differences from PwCHA in the covariates or their effects on the PK of emicizumab were indicated in PwAHA. This finding would warrant application of the same covariate-based dose adjustment as in PwCHA (i.e., only BW is considered) for PwAHA. REFERENCES: 1. Clin Pharmacokinet 2020;59:1611-25. 2. ISTH 2022 Congress. Abstract OC 40.2. 3. Clin Pharmacokinet 2021;60:931-41.