Jae-Yong Chung – Professor, Clinical pharmacology and therapeutics, Seoul National University, College of Medicine; SoHyeon Lim – Student, Clinical pharmacology and therapeutics, Seoul National University, College of Medicine

Objectives: VVZ-149 (opiranserin, INN) is a benzamide-derived intravenous analgesic for moderate to severe postoperative pain. Because central nervous system (CNS) exposure may contribute to its pharmacologic effects, this study aimed to develop a physiologically based pharmacokinetic (PBPK) model of VVZ-149 and apply it to predict plasma and cerebrospinal fluid (CSF) exposure.

Methods: A whole-body PBPK model for VVZ-149 was developed using PK-Sim? version 12 based on clinical trial data in healthy adults. Physicochemical properties were incorporated into a small-molecule model, tissue distribution was described by the Rodgers–Rowland method. Hepatic metabolism was implemented as a CYP3A4-mediated intrinsic clearance process. A brain PBPK model was then developed using the default PK-Sim brain subcompartments—blood, plasma, interstitial fluid (ISF), and intracellular space—and brain ISF exposure was used as a surrogate for CSF exposure, which is expected to equilibrate rapidly with lumbar CSF for small lipophilic molecules. Sensitivity analysis and parameter identification were performed using observed CSF data from a previously conducted clinical study in healthy adults (n=4), lumbar CSF sampling, and model performance was evaluated against observed concentration-time profiles. External validation was performed using independent plasma concentration-time data from phase 1 clinical studies (n=6). The final model was used to simulate 1,000 virtual subjects aged 18–81 years receiving the approved regimen of a 160-mg loading infusion over 0.5 h followed by an 840-mg maintenance infusion over 9.5 h.

Results: The final PBPK model adequately described observed plasma concentration-time profiles of VVZ-149 in healthy adults. Sensitivity analysis identified the plasma-to-brain ISF partition coefficient (PCCSF) as the most influential brain-specific parameter for CSF exposure, and parameter identification using observed CSF data estimated an optimal PCCSF value of 0.10 using the Levenberg–Marquardt algorithm. External validation showed that the model adequately reproduced observed plasma concentration-time profiles under a loading/maintenance dosing regimen, with a mean prediction error of −8.64% and 86.7% of simulated concentrations within a two-fold range of the observed data. Using the PBPK model, the plasma and CSF concentration–time profile was simulated. Simulations indicated a median CSF-to-plasma maximum concentration ratio of 0.116 and a median area under the concentration-time curve–based penetration ratio of 0.107.

Conclusions: A mechanistic whole-body PBPK model of VVZ-149 was developed to characterize plasma and CSF exposure and externally validated using independent clinical data. This PBPK model provides a quantitative framework for characterizing systemic and CNS exposure despite limited CSF sampling.

Citations:

Keywords: Opiranserin, Physiologically based pharmacokinetics modeling, Central nervous system exposure

Æ÷½ºÅÍ Á¦¸ñÀ¸·Î µÑ·¯º¸±â