(M1130-10-57) Development and Verification of Physiologically-Based Pharmacokinetic (PBPK) Model for Vaginal Administration of Dapivirine Ring and Gel Formulation

Purpose: The vaginal route of administration enables both local and systemic drug delivery for the treatment of vaginal infections, sexually transmitted diseases, cervical cancer and hormonal contraception. A great interest is focused on the vaginal delivery of microbicides to inhibit sexual transmission of Human Immunodeficiency Virus (HIV) and Herpes Simplex Virus 1 & 2 (HSV). Vaginal route presents challenges for PBPK model development due to limited number of studies available for mechanistic understanding of drug transport through the vaginal tissue. The purpose of this work is to develop and verify a robust modelling framework that can be applied to vaginally administered drugs. The present case study employs Mechanistic Vaginal Absorption & Metabolism (MechVAM) model implemented in Simcyp with Dapivirine as an example to simulate plasma concentrations after the administration of vaginal gel and ring formulation.

Methods: The physicochemical properties of Dapivirine are taken from the PubChem database while blood-to-plasma and unbound fraction predicted in Simcyp. A full PBPK model was developed by predicting volume of distribution using Roger and Rowland method while clearance value was manually optimized to recover the observed data due to lack of any oral/intravenous literature data. MechVAM model implemented in Simcyp incorporates physiological data relating to the female reproductive system. The model structure divides vagina into luminal fluid, vaginal epithelium, stroma and cervix along with the specific vaginal site of drug administration to avoid the hepatic first-pass effect. The drug released from the formulation in the luminal fluid of vaginal cavity enters systemic circulation either through the cervix or crossing the vaginal epithelium and stroma. The model is connected to all distribution and clearance models available in the Simcyp simulator. Input parameters include fraction unbound in each compartment, partition coefficients and permeabilities which determine the vaginal absorption and uptake into the systemic circulation. Taking the drug physicochemical parameters and formulation characteristics/in vitro release profile as input parameters, the model simulates drug transport through vaginal fluid and tissue. The partition coefficient across vaginal fluid and tissue layers are calculated or optimized (1). The fraction unbound in various layers and diffusion coefficients are obtained from the literature or calculated (2). Simulations presented herein have been carried out in Simcyp Version 21. Case study 1. The clinical pharmacokinetic data by Nel et al. (3) was simulated to predict plasma concentrations employing MechVAM model. The authors measured the plasma concentration after administration of 0.05% Dapivirine vaginal gel to 12 healthy female volunteers (3). Case study 2. The clinical pharmacokinetic data by Nel et al. (4) for 25 mg Dapivirine reservoir and matrix vaginal ring was simulated to predict plasma concentrations. The authors administered both the rings separately to 25 healthy female volunteers and measured plasma concentrations for 33 days (4). The zero order release rate constant for reservoir ring and first order release rate constant for matrix ring was calculated based on the amount of residual drug remaining in the ring after study completion were used as model input.

Results: Case study 1. Figure 1 compares the simulated systemic concentration to the observed profiles from Nel et al. (3). The model well captures the observed data for the 0.05% dapivirine vaginal gel. Case study 2. Figures 2A and B compare the simulated and observed plasma concentration profiles for reservoir ring administered for 1 and 33 days, respectively. Figures 2C and D compare the simulated and observed plasma concentration profiles for matrix ring administered for 1 and 33 days, respectively. Tables 1 and 2 show that, all the predicted primary PK parameters are well within the two-fold error of observed data for gel and ring respectively.

Conclusion: The MechVAM model predictions are in good agreement with the reported clinical data both for vaginal ring and gel formulations of dapivirine. Although systemic concentrations were used to verify the model, it can be leveraged further to predict local concentrations in the vaginal tissue. The prediction of local concentrations along with simulation of extended systemic effects will potentially allow for a full exploration of bioequivalence. The model can provide an insight in a systematic way regarding the impact of the drug and formulation characteristics on local and systemic exposure ultimately leading to informed decisions regarding clinical trial design. This can potentially reduce the risk associated with clinical trials and potentially reduce the number of trials required.

References: 1. Halwes, M.E., Steinbach-Rankins, J.M., Frieboes, H.B., Pharmacokinetic modeling of a gel-delivered dapivirine microbicide in humans. Eur J Pharm Sci, 2016. 93: p. 410-8.
2. Gao, Y., Katz, D.F., Multicompartmental pharmacokinetic model of tenofovir delivery by a vaginal gel. PLoS One, 2013. 8(9): p: e74404.
3. Nel, A.M., et al., Pharmacokinetics of 2 dapivirine vaginal microbicide gels and their safety vs. Hydroxyethyl cellulose-based universal placebo gel. J Acquir Immune Defic Syndr, 2010. 55(2): p. 161-9.
4. Nel, A.M., et al., Safety and pharmacokinetics of dapivirine delivery from matrix and reservoir intravaginal rings to HIV-negative women. J Acquir Immune Defic Syndr, 2009. 51(4): p. 416-23.


Figure 1: Observed versus Predicted plasma concentration of Dapivirine from 0.05% vaginal gel using Nel et al. 2010 data.


Figure 2: Observed versus Predicted plasma concentration of Dapivirine from reservoir ring for 1 and 33 days (A & B) respectively; matrix ring for 1 and 33 days (C & D) respectively using Nel et al. 2009 data.


Table 1: Observed versus Predicted PK parameters of Dapivirine following application of 0.05% Dapivirine gel, Table 2: Observed versus Predicted PK parameters of Dapivirine following application of 25 mg reservoir and matrix-type Dapivirine ring