A target-mediated drug disposition population pharmacokinetic model of GC1118, a novel anti-EGFR antibody, in patients with solid tumors

Abstract

GC1118 is a monoclonal antibody for epidermal growth factor receptor (EGFR) that is currently under clinical development to treat patients with solid tumors. In this study, the pharmacokinetics (PKs) of GC1118 were modeled in solid tumor patients who received a 2-h intravenous infusion of GC1118 at 0.3, 1, 3, 5, or 4 mg/kg once-weekly (Q1W) on days 1, 8, 15, and 22 or 8 mg/kg every other week on days 1 and 15. A target-mediated drug disposition population PK model adequately described the concentration-time profiles of GC1118. Monte-Carlo simulation experiments of the PK profiles and EGFR occupancies (ROs) by GC1118 based on the final model showed that Q1W at 4 or 5 mg/kg will produce a better antitumor effect than Q2W at 8 mg/kg. Because GC1118 was safer at 4 mg/kg than 5 mg/kg in the phase I study, we suggest to test the 4 mg/kg Q1W regimen in further clinical trials with GC1118. Study Highlights WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC? GC1118, a fully human IgG₁ monoclonal antibody (mAb) for epidermal growth factor receptor (EGFR), showed a nonlinear pharmacokinetic (PK) profile in monkeys and humans. The total clearance of GC1118 decreased as the dose was increased up to 3-4 mg/kg in humans, beyond which it remained stable. The recommended phase II dose for GC1118 was 4 mg/kg intravenously infused over 2 h once weekly. WHAT QUESTION DID THIS STUDY ADDRESS? We developed a target-mediated drug disposition (TMDD) population PK model that described the nonlinear PK profile of GC1118 in patients with solid tumors. We also simulated the PK profiles and receptor occupancies for different dosage regimens. WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE? The TMDD population PK model adequately described the nonlinear and multiphasic PK profiles of GC1118 in humans. The simulation experiment showed that once-weekly GC1118 at 4-5 mg/kg could be more efficacious than the biweekly regimen at 8 mg/kg. HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE? The pharmacometrics analysis could support better informed drug development decisions for GC1118, particularly for determining an optimal dosage regimen.

Figure 1

Mean serum GC1118 concentration‐time profiles after multiple i.v. infusions (left: linear scale, right: semilogarithmic scale). Observed serum GC1118 concentrations (µg/ml) after once weekly 0.3 mg/kg (●), 1 mg/kg (○), 3 mg/kg (▲), 5 m/kg (■), 4 mg/kg (△) i.v. infusion (cohorts 1 to 5, respectively) over 2 h and biweekly 8 mg/kg (□) i.v. infusion (cohort 6) over 2 h are shown

Figure 2

Target‐mediated drug disposition model for GC1118. A_C, unbound GC1118 amount in the central compartment; A_P, unbound GC1118 amount in the peripheral compartment; CL_A, clearance (CL) of the unbound GC1118; CL_B, CL of the EGFR; CL_C, CL of the GC1118‐EGFR complex; CPLX_P, amount of GC1118‐EGFR complex in the peripheral compartment; EGFR, epidermal growth factor receptor; K_D, dissociation constant between GC1118 and EGFR; Q, intercompartmental CL of unbound GC1118 between central and peripheral compartment; R_B, EGFR production rate; V_C, volume of distribution in the central compartment; V_P, volume of distribution in the peripheral compartment

Figure 3

General goodness‐of‐fit plots for the final population pharmacokinetic model of GC1118: observed vs. individual predicted serum GC1118 concentrations (upper left); observed vs. population predicted serum GC1118 concentrations (upper right); conditional weighted residuals (CWRES) vs. population predicted serum GC1118 concentrations (lower left); CWRES vs. time since last dose (lower right)

Figure 4

Visual predictive check (VPC) of the final population pharmacokinetic model of serum GC1118. The empty circles (○) represent the observed serum GC1118 concentrations, the solid (—) and dashed (‐ ‐ ‐) red lines denote the median values of the observed and predicted serum GC1118 concentrations, respectively, the solid (—) and dashed (‐ ‐ ‐) blue lines are the lower 2.5th and upper 97.5th values of the observed and predicted serum GC1118 concentrations, respectively, and the shaded areas indicate the 95% confidence intervals around the lower 2.5th, median, and upper 97.5th predicted concentrations, respectively. The y‐axis of the VPC plot is drawn on the semilogarithmic scale

Figure 5

Simulated pharmacokinetic (PK) profiles of GC1118 (plots a and b) and epidermal growth factor receptor (EGFR) occupancies (RO, plots c and d) at different dosage regimens after first intravenous infusion (plots a and c) and after reaching the steady‐state (plots b and d). Each solid line (━) in plots aand b represents mean concentration at a different dosage regimen in semi‐logarithmic scale; 3 mg/kg once weekly (Q1W) (black) 4 mg/kg Q1W (red), 5 mg/kg Q1W (green), 8 mg/kg biweekly (Q2W) (blue), and 10 mg/kg Q2W (dark red). Observed concentrations at 8 mg/kg Q2W from 8 patients (●) were overlaid to the same plot to evaluate the predictive performance of the simulation. Solid (—) and dashed line (‐ ‐ ‐) in plots C and D represents mean RO and 5th and 95th percentiles of RO, respectively, at a different dosage regimen in linear scale; 3 mg/kg Q1W (black) 4 mg/kg Q1W (red), 5 mg/kg Q1W (green), 8 mg/kg Q2W (blue), and 10 mg/kg Q2W (dark red)