A Population Pharmacokinetic Model for Concizumab Based on Phase 1 and Phase 2 Trial Data

Introduction Concizumab is a high-affinity anti-tissue factor pathway inhibitor (TFPI) monoclonal antibody in clinical investigation for the subcutaneous (SC) treatment of patients with hemophilia. The data generated from phase 1 and 2 concizumab trials have been used to develop a population pharmacokinetic (PK) model with the aim of supporting dose selection for phase 3 trials.
Methods The objective of this study was to develop a model to describe the PK of concizumab across administration routes in various groups of patients with hemophilia to generate a generally applicable population PK model of concizumab. The model was developed based on available PK data from four phase 1 trials (for both intravenous [IV] and SC concizumab administration) and two phase 2 trials (for SC concizumab administration). Trial populations in the phase 1 trials included both healthy subjects and patients with hemophilia, whilst the phase 2 trials enrolled patients with hemophilia A or B with inhibitors and patients with hemophilia A without inhibitors. A structural population PK model was first developed based on phase 1 data and the final population PK model was then estimated using data from both phase 1 and phase 2 trials. Simulations were performed for phase 3 concizumab exposure using a full parametric simulation (n=10,000), including both inter‑individual and intra-individual variability for the selected population. Randomly sampled body weights from a normal distribution with mean and variance corresponding to body weight distribution from phase 2 trials were used to simulate patient profiles.
Results The population PK dataset used for the model comprised 1,504 observations from 119 subjects (89 patients and 30 healthy individuals), with a mean age of 35 years (range: 18–65 years) and mean body weight of 74.4 kg (range: 47.1–130 kg). The PK model parameters were first estimated based on phase 1 data alone, and after fixing the majority in order to ensure robustness of the model only a few parameters were re-estimated based on phase 1 and 2 data combined. The PK model (Figure 1) was evaluated by standard goodness-of-fit plots and qualification assessments. Using visual predictive checks, it was shown that the model was able to reproduce the median and the 5^th and 95^th percentiles of the observed concizumab concentrations from phase 1 and 2 trials, and so it was deemed suitable for simulation purposes. The PK model suggested a target‑mediated drug disposition following concizumab binding to TFPI at the endothelium, and subsequent elimination of the complex to account for the non-linear elimination.
Conclusions The developed model accurately described the PK of concizumab delivered at a wide dose range by either SC or IV administration to both healthy subjects and patients with hemophilia A or B with and without inhibitors. The model was used for simulations to select the dosing regimen for subsequent phase 3 studies.

Figure 1. Concizumab pharmacokinetic model. Structure of the final concizumab PK model for SC and IV dosing with target-mediated drug disposition via the endothelial TFPI. CL, clearance; dose_iv, intravenous dose; dose_sc, subcutaneous dose; IV, intravenous; k_a, absorption rate constant; k_com, elimination rate constant of the concizumab-TFPI complex; k_on and k_off, rate constants for binding of concizumab to the endothelial TFPI; k_tr, rate constant from the transit compartment; Q, inter-compartmental clearance; R_tot, amount of endothelial TFPI available for concizumab binding; SC, subcutaneous; TFPI, tissue factor pathway inhibitor; V, volume.