(W0930-04-25) Physiologically Based Pharmacokinetic (PBPK) Oral Absorption Model to Predict Mucosal Permeability of Oral Cavity Drug Products

Purpose: Buccal drug administration presents multiple benefits: improved compliance, and bypassing hepatic first-pass elimination, among others. These advantages led to the development of specific drug products designed for certain populations (e.g., elderly patients) or certain active pharmaceutical ingredients (APIs) (e.g., APIs that have high hepatic extraction). Following the intraoral administration of such a product, the API dissolves in the saliva and is absorbed across the oral cavity mucosal tissue. To assess buccal absorption of APIs formulated as oral cavity drug products, in vitro permeability assays were previously conducted using the organotypic EpiOral™ tissue model (ORL-200, MatTek Corp., Ashland, MA) presenting a close resemblance to the human buccal mucosa. As part of this research project, a mechanistic in silico model was developed to facilitate the translation of the in vitro measurements into predictions of in vivo buccal absorption. The developed physiologically based pharmacokinetic (PBPK) model may be utilized to predict the clinical pharmacokinetics (PKs) of APIs following oral cavity administration and may support the development of new and generic oral cavity drug products.

Methods: An oral absorption PBPK model, representing the in vitro EpiOral tissue permeability assay was developed and implemented in MembranePlus™ software (beta version, Simulations Plus Inc., Lancaster, CA). The model structure integrated three compartments: a donor side, a multi-layer buccal tissue, and a receiver side. The model described the partitioning of the API from the donor compartment to the top tissue layer, diffusion of the API through the different tissue layers, and partitioning of the API from the last tissue layer into the receiver compartment. Additionally, the model can account for a possible non-specific loss of the API that may occur during the experiment (e.g., evaporation, API binding to the plasticware). The model also considers the experimental protocol of API depletion through sampling and the replacement of the media (HBSS buffer pH 7.4) in the received side. The model was used to deconvolute the diffusivity (D_m) and fraction unbound (f_ut) in the oral mucosa tissue for acyclovir, apomorphine, asenapine, buprenorphine, fentanyl, naloxone, sufentanil, and zolpidem using concentrations in the donor, tissue, and receiver compartments measured in the EpiOral system.

Results: A total of 8 APIs, approved in the US. market for oral cavity administration, were used to develop the oral absorption PBPK model. For each API, using the fitted D_m (ranging from 1.02X10^-7 to 5.16X10^-7 cm²/s) and f_ut (ranging from 2.5 to 22%) parameters allowed a reasonable description of the observed APIs’ concentrations in the donor, tissue, and receiver compartments measured during and at the end of the EpiOral assay. The model also adequately captured the initial lag in the API appearance in the receiver compartment applicable to some APIs. Parameter sensitivity analysis was used to investigate possible sources of variability in the measured concentration-time profiles in the studied matrices. Physiological variability of the tissue layer thickness in the EpiOral system was identified as one of the major contributors to the observed variability. Using the buccal tissue layer thickness values within the expected physiological range (90-140 um) resulted in a satisfactory description of all measured concentration-time profiles in the receiver compartment for each investigated API.

Conclusion: The knowledge of the API physicochemical properties impacting its diffusion through the oral mucosa is necessary for effective development of oral cavity drug product. To support this effort, a pragmatic modeling approach, using an oral absorption PBPK model, was developed. The buccal mucosa diffusivity and binding of 8 selected APIs was model-predicted based on state-of-the-art in vitro assay data. The model also described the impact of tissue thickness variability on the apparent API permeability in the in vitro system. Future work will use these results, to parameterize PBPK models to predict in vivo absorption of those APIs, administered as oral cavity drug products to humans. Ultimately, this model-based framework may be able to support the model-informed drug development paradigm of new and generic oral cavity drug products.

Acknowledgements: Funding Source and Disclaimer: The project was funded by a contract from the U.S. Food and Drug Administration (contract # 75F40120C00150). The views expressed here do not reflect official policies of the US FDA or the Department of Health and Human Services, nor does any mention of trade names imply endorsement by the US Government.