Jaeseong Oh (1), Doo H. Lee (2), Sunyoung Cho (2), Su-jin Rhee (1), Kyungho Jang (3), Seonghae Yoon (1), Kyung-Sang Yu (1), In-Jin Jang (1), Jae-Yong Chung (4)
(1) Department of Clinical Pharmacology and Therapeutics, Seoul National University College of Medicine and Hospital, Seoul, Republic of Korea; (2) Vivizon, Inc. Seoul, Korea; (3) Department of Clinical Pharmacology and Therapeutics, CHA Bundang Medical Center, CHA University, Seongnam, Gyeonggido, Republic of Korea; (4) Department of Clinical Pharmacology and Therapeutics, Seoul National University Bundang Hospital, Seongnam, Gyeonggido, Republic of Korea
Objectives: VVZ-149, a dual antagonist on GlyT2 and 5HT2A, is an investigational analgesic product targeted for post-operative pain. The clinical therapeutic concentration range of VVZ-149 including its active metabolite (VVZ-368) was expected to be 600 – 1900 μg/L, according to the preclinical study results. The aim of this study was to develop a population PK model for VVZ-149 and VVZ-368.
Methods: A randomized, double-blind, single ascending-dose (SAD) and multiple ascending-dose (MAD), placebo-controlled clinical trial was conducted in 67 healthy volunteers (NCT01905410). Subjects intravenously received 0.25-8 mg/kg of VVZ-149 in SAD study and 8-14 mg/kg/day for three days in MAD study. Serial blood samples were collected to analyze plasma concentration of VVZ-149 and VVZ-368. A total of 2,042 plasma concentrations were pooled to develop a population PK model using the nonlinear mixed-effects method in NONMEM (ver. 7.3). The First-Order Conditional Estimation with Interaction estimation method was implemented, which was followed by model qualification using bootstrapping and visual predictive checks (VPCs).
Results: The final model consisted of 3 compartments: central (1) and peripheral (2) for VVZ-149 and central for VVZ-368 (3). All transfers were described with first-order processes. Inter-subject variabilities were modelled by exponential error models and proportional error models were used to account for intra-subject variabilities of both VVZ-149 and VVZ-368. The typical values of population pharmacokinetic parameters were; clearance of VVZ-149 (CL, 48.8 ∙ (creatinine clearance/111)0.51 L/h), central volume of distribution (V1, 19.0 L), inter-compartmental clearance (Q, 88.5 L/h), peripheral volume of distribution (V2, 78.5 L), metabolic clearance of VVZ-149 converted into VVZ-368 (CLpm, 1.01 L/h), clearance of VVZ-368 (CLm, 2.73 L/h). The inter-individual variability (CV%) of CL, V1, V2 and CLpm were 18.6%, 34.7%, 22.1% and 46.8%, respectively. Model evaluation by goodness of fit plots, bootstrapping and VPCs suggested that the proposed model was adequate and robust with good precision. For the therapeutic purpose, a loading-maintenance dose of VVZ-149 (loading dose: 2.6 mg/kg/h for 0.5 hour; maintenance dose: 0.9 mg/kg/h) were predicted to be proper dosage regimen to achieve the effective concentration range.
Conclusions: The final model adequately described the observed plasma concentrations of VVZ-149 and VVZ-368 in healthy male volunteers. The simulated dose regimen can be applied in further drug development.
Reference: PAGE 25 (2016) Abstr 5845 [www.page-meeting.org/?abstract=5845]
Poster: Drug/Disease modeling - Other topics