Population Pharmacokinetic Comparability of CT-P16 Versus EU-Avastin® and US-Avastin® in Healthy Subjects and Patients with Non-Squamous Non-Small Cell Lung Cancer
摘要
CT-P16 (Vegzelma®) is a bevacizumab biosimilar approved for indications including non-squamous non-small cell lung cancer. We developed an integrated population pharmacokinetic model pooling data from two phase I studies in healthy subjects and one phase III study in patients with non-small cell lung cancer to (i) quantify the influence of the drug product on pharmacokinetic parameters within a single unified framework simultaneously evaluating CT-P16 against both EU-Avastin® and US-Avastin®; (ii) characterise covariate-driven exposure variability, including disease-status effects, across healthy subjects and patients with non-small cell lung cancer, and compare these effects with previously reported analyses of reference bevacizumab and other bevacizumab biosimilars; and (iii) contextualise simulated steady-state exposure under the approved 15 mg/kg every-3-weeks regime against a published exposure–response benchmark, thereby supporting a biosimilar assessment within the streamlined regulatory paradigm.
MethodsConcentration data from one phase III trial (NCT03676192) and two phase I trials (NCT03247673; CT-P16 1.2) were pooled. The population pharmacokinetic analysis used nonlinear mixed-effects modelling in NONMEM® (Version 7.4) with first-order conditional estimation with interaction. Drug product (CT-P16, EU-Avastin®, US-Avastin®) and clinical/demographic covariates were evaluated by stepwise selection. Model adequacy was assessed by goodness-of-fit diagnostics, non-parametric bootstrap resampling (1000 replicates), and a visual predictive check. Steady-state exposure under 15 mg/kg every 3 weeks was simulated using both typical population-predicted profiles and the observed phase III patient covariate distribution with inter-individual variability.
ResultsA total of 8058 serum concentrations from 834 subjects (phase I: 187 healthy male volunteers; phase III: 649 patients with non-small cell lung cancer) were analyzed. A two-compartment model with first-order elimination adequately described the pooled data. Body weight, sex, and disease status were retained as covariates on clearance; sex, baseline serum albumin, and disease status were retained on central volume of distribution. Drug product was not retained as a significant covariate, indicating no clinically meaningful difference between CT-P16 and either EU-Avastin® or US-Avastin®. The estimated body-weight exponent on clearance (0.369) was numerically indistinguishable from previously published values for reference bevacizumab (0.368 [
Within a unified population pharmacokinetic framework, CT-P16 demonstrated comparable pharmacokinetics to both EU-Avastin® and US-Avastin®, with no clinically meaningful drug-product effect on clearance or central volume of distribution. Covariate effects mirrored those reported for reference bevacizumab, and simulated steady-state exposures were generally consistent with a published progression-free-survival-based exposure benchmark at the median level across covariate-defined patient subgroups. These findings extend the totality of evidence supporting CT-P16 biosimilarity and illustrate how population pharmacokinetic modelling can quantitatively inform a streamlined biosimilar assessment in the era of reduced reliance on comparative clinical efficacy trials.
Clinical Trial RegistrationNCT03247673; NCT03676192.