(M1130-11-64) Assessing the Impact of Excipient and Food Intake on Bioequivalence Using PBPK and Virtual Bioequivalence Trial: A Case Example with Acyclovir Immediate Release Tablets

Purpose: Physiologically-based pharmacokinetic (PBPK) modeling is often used in lieu of dedicated DDI studies for new drug development. The application of PBPK in generic drug development areas is being explored and under development in light of the increasing interest from generic industries and regulatory agencies. Using PBPK modeling to assess the impact of excipients on BE for supporting the possible biowaiver of non-Q1 (qualitatively) the same and Q2 (quantitatively) similar Biopharmaceutical Classification System (BCS) Class 3 generic products and to assess the impact of food on BE are the areas of interest in generic drug development. As a proof of concept, in this study, we have utilized an oral PBPK model of acyclovir immediate release (IR) tablet for assessing the impact of excipient and food intake on the BE of generic acyclovir IR tablet using virtual healthy subjects and virtual bioequivalence (VBE) trials.

Methods: GastroPlus^TM (version 9.8, Simulations Plus Inc., CA, USA) software with PKPlus^TM module and Advanced Compartmental Absorption Transit (ACAT^TM) module were used for acyclovir PBPK model development. Once the disposition parameters were estimated using pharmacokinetic (PK) data of 2.5 mg/kg acyclovir IV infusion, oral PBPK model was developed using these parameters along with apparent intestinal permeability value (P­_app) to predict the plasma profile of acyclovir 200 mg, 400 mg & 800 mg IR tablets under the fasted condition. To simulate food intake, GastroPlus^TM fed physiology was used and plasma profile of acyclovir 800 mg IR tablet taken with food was predicted. The model was validated using pharmacokinetic data for oral 400 and 800 mg IR tablets under fasted and fed conditions. VBE studies were carried out under both fasted and fed conditions using virtual population simulations along with incorporation of in vitro dissolution data, obtained from FDA’s approved abbreviated new drug application (ANDA) for test and reference (Zovirax^®) tablet, into PBPK model. To assess the impact of excipient-mediated intestinal permeability changes on the BE, the P­_app value in the PBPK model used to simulate PK profiles of the test product was changed at different percent (test) compared to the P­_app value used in the base PBPK model (reference). Additionally, similar to the fasted condition, the impact of P­_app value changes were also assessed under the fed condition.

Results: The developed acyclovir oral PBPK model was capable of predicting the plasma profile of 200 mg, 400 mg & 800 mg acyclovir IR tablet under both fasted and fed conditions with prediction error < 20%. The model also predicted the observed T_max increment of acyclovir IR tablet when taken with food. Results from VBE trials using this PBPK model showed that the test and reference tablets are BE under fasted and fed conditions when respective in vitro dissolution data were used as model input (Figure 1). When the impact of excipient-mediated intestinal permeability enhancement on the PK of acyclovir was assessed in a single representative virtual subject using this PBPK model, it was found that the ratios of PK parameters (C_max, AUCs) for test and reference were less than 1.25 for up to 85% increment of the P­_app value of the test product compared to the reference product (Figure 2A). However, after the inter-subject variabilities were included in these predictions by using population simulation and VBE, the test and reference were BE up to 30% increase of P­_app value in the test product compared to the reference product’s P­_app value (Figure 2B). The failed BE under both fasted and fed condition could be a consequence of the combined impact of P­_app value change caused by possible formulation differences and presence of inter-subject variabilities in VBE simulations.

Conclusion: This study presents the development and validation of an oral PBPK model for acyclovir IR tablet in an effort to assess the impact of excipient and food intake on the BE assessment using VBE trials. The VBE results suggested that more than 30% change of P­_app value for test product due to presence of certain excipient may result in failed BE of acyclovir 800 mg IR tablet under both fasted and fed conditions. Additionally, the model also predicted the food induced increased T_max, observed in clinical PK studies.

Acknowledgements:This abstract reflects the views of the presenter and should not be construed to represent FDA’s views or policies.


Figure-1: VBE trial outcomes for acyclovir 800 mg IR tablet conducted using PBPK model and virtual healthy subjects under fasted (A) and fed condition (B). The test and reference acyclovir 800 mg IR tablets are BE under both fasted and fed conditions.


Figure-2: (A) Impact of excipient-mediated apparent intestinal permeability (Papp) changes on the PK parameters predicted using single subject simulation and acyclovir oral PBPK model. (B) VBE trials show that the test and reference acyclovir 800 mg IR tablets are BE under fasted and fed condition for up to 30% Papp value increment in the test product.