In vitro profiling of the vaginal permeation potential of anti-HIV microbicides and the influence of formulation excipients Carolien Grammen, Patrick Augustijns, Joachim Brouwers ⇑ Laboratory for Pharmacotechnology and Biopharmacy, KU Leuven, Belgium article info Article history: Received 15 July 2012 Revised 31 August 2012 Accepted 12 September 2012 Available online 20 September 2012 Keywords: HIV microbicide Vaginal gel Solubility Permeability abstract In the search for an effective anti-HIV microbicidal gel, limited drug penetration into the vaginal submu- cosa is a possible reason for failed protection against HIV transmission. To address this issue in early development, we here describe a simple in vitro strategy to predict the tissue permeation potential of vaginally applied drugs, based on solubility, permeability and flux assessment. We demonstrated this approach for four model microbicides (tenofovir, darunavir, saquinavir mesylate and dapivirine) and additionally examined the influence of formulation excipients on the permeation potential. When formu- lated in an aqueous-based HEC gel, high flux values across an HEC-1A cell layer were reached by tenofo- vir, as a result of its high aqueous solubility. In contrast, saquinavir and dapivirine fluxes remained low due to poor permeability and solubility, respectively. These low fluxes suggest limited in vivo tissue pen- etration, possibly leading to lack of efficacy. Dapivirine fluxes, however, could be enhanced up to 30-fold, by including formulation excipients such as polyethylene glycol 1000 (20%) or cyclodextrins (5%) in the HEC gels. Alternative formulations, i.e. emulsions or silicone elastomer gels, were less effective in flux enhancement compared to cyclodextrin-HEC gels. In conclusion, implementing the proposed solubility and permeability profiling in early microbicide development may contribute to the successful selection of promising microbicide candidates and appropriate formulations. Ó 2012 Elsevier B.V. All rights reserved. 1. Introduction With about 7000 new HIV infections each day, half of them occurring in women, there is an urgent need for a female con- trolled HIV prevention method such as anti-HIV microbicides (WHO, 2012). These are compounds that are applied vaginally or rectally in order to prevent the transmission of HIV. Since non-spe- cific first generation microbicide candidates (nonoxynol-9, Savy Ò , Carraguard Ò , cellulose sulfate, PRO-2000, BufferGel Ò ) all failed in phase 3 clinical trials, a shift has occurred towards more specific anti-retroviral drugs (ARV) such as (non-)nucleoside reverse trans- criptase inhibitors ((N)NRTIs), protease inhibitors (PIs) and entry- inhibitors. The CAPRISA 004 trial, testing a gel of the NRTI tenofovir (1%) in African women, provided proof of concept that vaginally applied anti-HIV microbicides can reduce transmission of the virus (Abdool Karim et al., 2010). For small molecular ARV drugs, it is imperative that they are taken up into the tissue where HIV trans- mission takes place, i.e. the vaginal submucosa containing most HIV target cells (Adams and Kashuba, 2012; Hladik and Doncel, 2010). In addition to activity or toxicity issues, poor tissue uptake of microbicides can thus be a possible cause of failed protection against HIV transmission. The importance of the pharmacokinetic disposition of microbi- cides upon their application in the vaginal or rectal lumen has long been underrated in microbicide research. Nowadays however, sev- eral in vitro and in vivo model systems are being used to evaluate key issues in microbicide disposition. Drug release properties of vaginal formulations are often studied in in vitro set-ups using large amounts of isopropanol/water mixtures. Obviously, these high volumes of non-aqueous liquid are not representative for the human vaginal fluid, limiting the in vivo relevance of these sys- tems (Kiser et al., 2012; Woolfson et al., 2010). To study actual tis- sue uptake, several model systems are available such as the MatTec EpiVaginal™ tissue model, human ectocervical explants and ani- mal models (primarily rabbit, sheep or macaque) (Ayehunie et al., 2006; Clark and Friend, 2012; Dobard et al., 2012; Kiser et al., 2012). While very useful in the final stages of preclinical microbicide development, these techniques are expensive, time consuming, not readily available (in case of ectocervical explants) and/or labor intensive. Consequently, neither of these approaches is suitable as a screening tool to evaluate the tissue permeation po- tential of numerous microbicide candidates in early development, or to select appropriate formulations. 0166-3542/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.antiviral.2012.09.011 ⇑ Corresponding author. Address: Laboratory for Pharmacotechnology and Bio- pharmacy, KU Leuven, Gasthuisberg O&N 2 – Herestraat 49, Box 921, 3000 Leuven, Belgium. Tel.: +32 16 330310; fax: +32 16 330305. E-mail address: joachim.brouwers@pharm.kuleuven.be (J. Brouwers). Antiviral Research 96 (2012) 226–233 Contents lists available at SciVerse ScienceDirect Antiviral Research journal homepage: www.elsevier.com/locate/antiviral