(M1530-11-63) A Population Pharmacokinetic Modeling Approach in Determining the Absolute Bioavailability of Sibeprenlimab, an IgG2 Monoclonal Antibody, from Two Single Ascending Dose Trials in Healthy Subjects

Purpose: Two separate single ascending dose (SAD) healthy subject trials have been conducted with the subcutaneous and intravenous route of administration of sibeprenlimab, respectively. The aim of this study was to develop a preliminary population pharmacokinetic (PK) model using the data from these two trials and estimate the absolute bioavailability of sibeprenlimab in lieu of noncompartmental PK analysis.

Methods: Data from two SAD trials were included in this analysis, in which sibeprenlimab was either administered intravenously (IV; 0.5 - 12 mg/kg) or subcutaneously (SC; 200 mg [1x 1 mL injection], 400 mg [1x 2 mL injection], 400 mg [2x 1 mL injection], and 600 mg [1x 2 mL injection + 1x 1 mL injection]). A preliminary population PK model was developed, using non-linear mixed effects modelling, to describe both sibeprenlimab PK following IV and SC administrations (NONMEM version 7.4.1). The absolute bioavailability of sibeprenlimab following SC administration was introduced as a PK parameter in the population PK model. Covariate analysis was conducted to account for the between-subject variability in sibeprenlimab PK. The covariates of interest included body weight, age, sex, ethnicity (Japanese versus non-Japanese), creatinine clearance (CRCL), serum albumin, baseline ‘A Proliferation-Inducing Ligand’ (APRIL) levels, antidrug antibody (ADA) status, dose, SC cohort (1 injection vs 2 injections), and study. Covariates that showed statistically significant improvement on model fitting were retained in the model. The model performance was evaluated by goodness-of-fit plots as well as visual predictive check (VPC).

Results: A total of 928 plasma concentrations of sibeprenlimab from 86 subjects (38 from the IV SAD trial and 48 from the SC SAD trial; 31 male and 55 female; age: 18-54 years old; body weight: 39.7-103 kg) were included in the development of the preliminary population PK model. A two-compartment model with parallel linear and non-linear elimination, and first order absorption with a lag time for SC absorption adequately described both sibeprenlimab PK following IV and SC administrations. Clearance and volume of distribution were allometrically scaled by body weight with fixed exponents (0.75 for clearance, and 1 for volume of distribution); for SC administration with 2 injections, values of the absorption parameters were estimated separately for the 2 injections. Typical values of PK parameters are generally predicted with good precision as the relative standard errors (RSE%) are ≤ 28.5% except for the absorption delay of the 2nd SC injection (RSE%: 47.2%), which is not unexpected. Between-subject variability was described with an exponential error model applied to relevant PK parameters using blocked diagonal structure. None of the other covariates were included in the final model based upon statistical significance, except body weight inclusion into the final model based upon a typical monoclonal antibody PK characteristic. Model evaluation suggested acceptable goodness of fit and predictive performance. The population PK model estimated absolute bioavailability of sibeprenlimab following SC administration is 74.8%.

Conclusion: A preliminary population PK model adequately described sibeprenlimab PK following either IV or SC administration. The model estimated absolute bioavailability of sibeprenlimab following SC administration is 74.8%.

Acknowledgments: Conflict of Interest: Yanlin Wang and Xiaoyan Zhang are full-time employees of Otsuka Pharmaceutical Development & Commercialization, Inc., Princeton, NJ, United States.
Frank Engler received consulting fee from Visterra Inc. for the work done on this project.
Asher D. Schachter is a full-time employee of Visterra Inc., Waltham, MA, United States.