Ki Soon Kim, Eben Jung, Seongmee Jeong, Woo-Yong Oh, Yoonsook Lee
National Institute of Food and Drug Safety Evaluation
Objectives: In 2013, the recommended initial dose of zolpidem changed to 5 mg for female in FDA[1]. The concern is that increased driving impairment caused by Cmax and AUC of zolpidem, which are higher in female than in male. Unlike FDA’s decision, EMA highlighted the risk of re-administration due to lower dose during the same night which affects the increase of the next day’s driving impairment[2]. Therefore, we aimed to develop a population PK-PD model of zolpidem to evaluate the PK/PD difference in male and female Korean subjects and to examine the need for dose adjustment for Koreans.
Methods: 15 Male and 15 female healthy subjects were administered a single oral dose of zolpidem 10 mg. Blood samples were collected up to 12 hours post-dose to determine plasma concentration of zolpidem. Digital Symbol Substitution Test (DSST) and Visual Analogue Scale (VAS) of sleepiness were used to evaluate PD of zolpidem (baseline, post-dose 0.5, 1, 1.5, 2, 3, 4, 6, 8 and 12h). Population PK-PD analysis was used a sequential approach which was as follows: plasma concentrations of zolpidem were fitted first to a PK model and subsequently the relationships between DSST, VAS and zolpidem concentration were fitted to PD model. All models were developed using NONMEM (version 7.3). Simulations were performed using the final PK and PD model to evaluate the effect of change in doses, sex difference, need for dose adjustment in Koreans.
Results: The PK of zolpidem was adequately described by a 2-compartment model with erlang-type absorption with lag time followed by zero-order absorption with lag time. Body weight was identified as a covariate showing significant influence on the central volume (V2/F). The simulated PK profiles of zolpidem 10 mg was not shown significantly different between male and female profiles. About 6% of 2,500 females who administered exceeded the criteria (50 ng/ml, at high risk of traffic accidents) set by FDA after 8 hours, which was 0.4 times that of FDA results (15%).
The PK-PD relationship between DSST, VAS and plasma concentration of zolpidem was described by the direct effect model using sigmoid Emax model. After dosing, the maximum effect of DSST and VAS was shown at 32 out of 140 and 6 out of 10, respectively. There was no significant difference between male and female with respect to the simulated PK-PD profiles for DSST and VAS. When taking the regimen of 5mg BID in female group, the cognitive decline and sleepiness lasted two to three hours longer than when taking that of 10mg QD.
Conclusions: The PK and PK-PD modeling of zolpidem were successfully developed and applied to analyze PK and PD data with regard to the effect of sex difference in Koreans. The results of modeling and simulation show that it is not necessary to reduce the dose of zolpidem in female from 10 mg to 5 mg and the recommended dose of zolpidem in MFDS is reasonable.
References:
[1] https://www.fda.gov/files/drugs/published/Drug-Safety-Communication–Risk-of-next-morning-impairment-after-use-of-insomnia-drugs–FDA-requires-lower-recommended-doses-for-certain-drugs-containing-zolpidem-%28Ambien–Ambien-CR–Edluar–and-Zolpimist%29.pdf
[2] https://www.ema.europa.eu/en/medicines/human/referrals/zolpidem-containing-medicines
Reference: PAGE () Abstr 9357 [www.page-meeting.org/?abstract=9357]
Poster: Drug/Disease Modelling - Safety