Background: Certolizumab pegol is an effective biologic for patients with Crohn's disease (CD). Individual differences in certolizumab pegol apparent clearance (CL/F) affect exposure and possibly efficacy. A previously developed population pharmacokinetic (PK) model did not account for dynamic changes in clinical parameters during therapy.
Objective: The aim of this study was to refine the existing PK model to capture the time-varying influence of covariates.
Methods: Data collected from 2157 Crohn's disease patients in nine studies were analyzed using nonlinear mixed-effects modeling software (NONMEM). Certolizumab pegol concentration-time data were described by a one-compartment PK model with first-order absorption, and one-compartment disposition with linear, time-dependent elimination using antidrug antibody (ADAb) concentration as a continuous variable.
Results: The final dataset consisted of 12,926 analyzable records. Parameter estimates were absorption rate constant 1.83/day, CL/F 0.527 L/day, and apparent volume of distribution (V/F) 8.33 L. ADAb concentration (2.5-214 units/mL) increased the median CL/F by 142-174%. For a typical patient, body weight (46.8-100.5 kg) increased the median CL/F and V/F from 82 to 120%. Albumin (32-48 g/L) decreased and C-reactive protein (0.5-54.0 mg/L) increased the median CL/F from 123 to 85% and from 83 to 113%, respectively. Between-patient variability of CL/F was 19.6%.
Conclusions: By incorporating time-varying covariates, this population PK model reduces between-patient variability on CL/F estimates, and the relative influence of ADAb can now be assessed. As Crohn's disease patient covariates are often time-dependent, this model is more reflective of patient drug exposure with sustained treatment.