PII-095 - ASSESSMENT OF THE IMPACT OF NON-COMPARABLE IN VITRO ALCOHOL DOSE DUMPING RELEASE ON SYSTEMIC EXPOSURE USING PBPK ABSORPTION MODELING FOR TOPIRAMATE EXTENDED-RELEASE CAPSULES.

Y. Cheng¹, F. Wu¹, Y. Mousa¹, L. Zhao¹, L. Fang¹; ¹U.S. Food and Drug Administration, Silver Spring, MD, USA.

Background: Due to concerns of alcohol dose dumping (ADD), additional dissolution testing using various ethanol concentrations in pH 1.1 medium is recommended for topiramate extended-release (ER) capsules. This study used a physiologically based pharmacokinetic (PBPK) absorption model to assess whether the non-comparable in vitro dissolution results observed at pH 1.1 with 20% ethanol, as opposed to 5% or 40% ethanol, would impact bioequivalence (BE) determination between test product and reference listed drug (RLD).

Methods: Physicochemical properties and parameters obtained from the literature, PK data, and GastroPlusTM software (Version 9.8.2, Simulations Plus Inc., CA, USA) were used to develop the PBPK absorption model. Dissolution data at pH 1.1, pH 6.8, and pH 7.5 and developed bi-phasic dissolution profiles (i.e., drug release for 1 hour at pH 1.1 and release at pH 6.8 or pH 7.5 afterwards) were incorporated in the PBPK model to predict PK profiles following administration with test product and RLD of 200 mg topiramate ER capsules under fasted conditions. Dissolution data and PK data of RLD of other strengths (25, 50, and 100 mg) were used to further validate the PBPK model. PBPK modeling and virtual BE (VBE) simulations were conducted to evaluate the impact of the non-comparable in vitro ADD release on the predicted systemic exposure, and overall BE determination.

Results: The oral PBPK model incorporating in vitro dissolution data at pH 6.8 and pH 7.5, accurately predicted plasma profiles for 200 mg topiramate ER capsules under fasted conditions with prediction errors (PE) of < 20% (Figure 1). Although the 200 mg topiramate ER capsule test product showed non-comparable (faster) in vitro drug release at pH 1.1 with 20% ethanol than the RLD, the PBPK model predicted that the test product is bioequivalent to the RLD in vivo (Figure 1). The observed differences in in vitro ADD release do not affect the predicted in vivo systemic exposure, indicating a low risk of bioinequivalence (Figure 1).

Conclusion: The PBPK absorption model adequately predicted various topiramate PK profiles. Moreover, PBPK modeling and VBE trial results suggest that the test product and RLD have comparable PK parameters although faster release of test product was observed in the in vitro ADD study with 20% ethanol.