Safety and Pharmacokinetics of Quick-Dissolving Polymeric Vaginal Films Delivering the Antiretroviral IQP-0528 for Preexposure Prophylaxis

Abstract

For human immunodeficiency virus (HIV) prevention, microbicides or drugs delivered as quick-dissolving films may be more acceptable to women than gels because of their compact size, minimal waste, lack of an applicator, and easier storage and transport. This has the potential to improve adherence to promising products for preexposure prophylaxis. Vaginal films containing IQP-0528, a nonnucleoside reverse transcriptase inhibitor, were evaluated for their pharmacokinetics in pigtailed macaques. Polymeric films (22 by 44 by 0.1 mm; providing 75% of a human dose) containing IQP-0528 (1.5%, wt/wt) with and without poly(lactic-co-glycolic acid) (PLGA) nanoparticle encapsulation were inserted vaginally into pigtailed macaques in a crossover study design (n = 6). With unencapsulated drug, the median (range) vaginal fluid concentrations of IQP-0528 were 160.97 (2.73 to 2,104), 181.79 (1.86 to 15,800), and 484.50 (8.26 to 4,045) μg/ml at 1, 4, and 24 h after film application, respectively. Median vaginal tissue IQP-0528 concentrations at 24 h were 3.10 (0.03 to 222.58) μg/g. The values were similar at locations proximal, medial, and distal to the cervix. The IQP-0528 nanoparticle-formulated films delivered IQP-0528 in vaginal tissue and secretions at levels similar to those obtained with the unencapsulated formulation. A single application of either formulation did not disturb the vaginal microflora or the pH (7.24 ± 0.84 [mean ± standard deviation]). The high mucosal IQP-0528 levels delivered by both vaginal film formulations were between 1 and 5 log higher than the in vitro 90% inhibitory concentration (IC₉₀) of 0.146 μg/ml. The excellent coverage and high mucosal levels of IQP-0528, well above the IC₉₀, suggest that the films may be protective and warrant further evaluation in a vaginal repeated low dose simian-human immunodeficiency virus (SHIV) transmission study in macaques and clinically in women.

INTRODUCTION

In the absence of an effective vaccine, preexposure prophylaxis (PrEP) with antiretroviral (ARV) drugs is currently a very promising biomedical intervention to control the human immunodeficiency virus (HIV) epidemic (1, 2). PrEP involves the use of ARV by high-risk HIV-negative individuals to prevent HIV acquisition. Many ARVs that have traditionally been used for HIV treatment have advanced as PrEP agents due to their safety and potency profiles (1, 3–5). Of the available ARVs, tenofovir disoproxil fumarate (TDF), a nucleoside reverse transcriptase inhibitor (NRTI) used widely in HIV treatment, has been studied extensively in animals and humans as a PrEP agent (3, 5). The success with oral TDF and Truvada, TDF in combination with emtricitabine (FTC), in inhibiting HIV infection in nonhuman primates led to the clinical trials in humans with TDF and Truvada (1, 6–11). In stark contrast to the success of TDF and Truvada in clinical trials in men who have sex with men and heterosexual HIV-discordant couples, two large clinical trials in women with oral Truvada, VOICE and FEM-PrEP, have demonstrated no protection due to a lack of adherence (12, 13).

Providing optimal dosage forms of potent HIV PrEP agents for women has the potential to improve product adherence (14, 15). Topical dosage forms can be effective in reducing HIV infection, as demonstrated by the CAPRISA 004 vaginal 1% tenofovir (TFV) gel trial, where per-protocol use of the gel before and after sex with no more than 2 doses in 24 h (BAT-24) reduced HIV acquisition by 39% (2). As observed in the oral PrEP trials, greater efficacy was reported among the high adherers in CAPRISA 004 (54% versus 28% in the low adherers) (2). Subsequent trials with 1% TFV gel, such as FACTS 001, employing the BAT-24 regimen used in CAPRISA-004, and the VOICE trial, with a daily application of 1% TFV gel, were not effective in protecting women against HIV acquisition (12, 16). It was estimated based on the returned used applicators that only 13% of the FACTS 001 trial participants were able to use the gel in ≥80% of sex acts per month (16). Neither the noncoital daily application of the TFV gel with VOICE nor the pericoital vaginal dosing regimen with FACTS 001 demonstrated effectiveness in comparison to the placebo group against HIV acquisition (12, 16). The above-described studies indicate the need for highly efficacious ARVs and delivery platforms that would improve and enhance user compliance, particularly in women.

Woman-initiated HIV prevention strategies are limited, and negotiation of safe-sex practices by women is not readily acceptable in many high-HIV-incidence settings (17). As with hormonal contraception, regional differences in what is acceptable can vary widely, and women may prefer to choose between multiple dosage forms based on their socioeconomic status and individual preferences (18, 19). Multiple topical PrEP delivery platforms, such as gels, intravaginal rings (IVRs), films, vaginal inserts, and soft-gel capsules are currently being developed (20–24). A product that is destined for topical use should be safe, widely acceptable to promote and enhance adherence to its use, cost-effective, and deliver high mucosal concentrations of the drug to prevent HIV infection. Microbicide delivery through solid dosage forms, such as quick-dissolving vaginal films, may be more acceptable to women than gels because of their small size, lack of a need for an applicator, easier storage and transport, and increased product stability.

Vaginal films are polymeric, thin, solid dosage forms which when applied mucosally undergo hydrolysis to release the active pharmaceutical ingredient (API) incorporated in their matrix. A consumer product preference study conducted among 526 sexually active women in Burkina Faso, Tanzania, and Zambia reported that vaginal films and soft-gels were preferred over vaginal tablets due to the ease of insertion and faster dissolving time (15). Other promising attributes, such as no leakage, lack of feeling the product inside, likelihood to use the product in the future, and acceptability among their male partners were also described for the films (15). A vaginal contraceptive film (VCF) containing the spermicide nonoxynol-9 was the first commercially available film (25). Other feminine hygiene products, including a VCF lubricating film and a vaginal cleansing film, are also marketed. Vaginal films to treat bacterial vaginosis and vaginal candidiasis have also been developed (26, 27). A number of promising ARVs and/or microbicides, such as dapivirine, TFV, dapivirine in combination with TFV and maraviroc, retrocyclin-101, and cellulose acetate phthalate, have been formulated into films (28–34). A few studies have evaluated the safety, pharmacokinetics (PK) and pharmacodynamics of films in macaques and humans (29, 35).

Films are limited by the amount of API that can be incorporated into them, and the API generally does not exceed 50% of the of the product's dry weight (36, 37). Therefore, the API in films needs to be very potent to overcome this limitation. Pyrimidinediones, an important class of nonnucleoside reverse transcriptase inhibitor (NNRTI), are potent topical PrEP candidates (38, 39). Of the available analogs, IQP-0528 [1-(3-cyclopropyl)methyl-6-(3,5-dimethylbenzoyl)-5-isopropyl-2,4(1H,3H)-pyrimidinedione], has been formulated for topical PrEP due to its desirable characteristics, such as subnanomolar NNRTI activity, high therapeutic index, chemical stability, and ability to be synthesized from readily available precursors, thus lowering production costs (40–44). IQP-0528 acts by binding to a hydrophobic pocket of the reverse transcriptase enzyme, and a high mucosal concentration is required to prevent local human immunodeficiency virus type 1 (HIV-1) replication. IVRs delivered IQP-0528 to macaque mucosal fluid and tissue at concentrations that were several log greater than the in vitro 90% inhibitory concentration (IC₉₀) (40). The half-maximal effective dose (ED₅₀) concentration of IQP-0528 at 3 nM is 700-fold lower than that of TFV (2,000 nM) (42). Unlike TDF, explant tissue pretreated with IQP-0528 provided significant protection to cocultured T cells against HIV, indicating the ability of IQP-0528 and not of TDF to protect newly recruited immune cells (45). The above-described features of high potency and the ability of IQP-0528 to move in and out of cells favor its formulation into polymeric films. In efforts to broaden the microbicide pipeline and to provide women with as many HIV-preventative options as possible, IQP-0528 has been developed into a vaginal film. Initial formulations demonstrated its in vitro ability to effectively and rapidly deliver IQP-0528 into vaginal tissue and target cells. The physicochemical properties, safety evaluation in organotypic in vitro tissue models, and anti-HIV activity of vaginal films containing IQP-0528 have demonstrated its potential as an effective microbicide (46). Additionally, by integrating IQP-0528 into dissolvable nanoparticles (IQP-0528NP), the films offer the potential to confer long-term drug delivery (47).

Nonhuman primates, in particular pigtailed macaques, have been pivotal in evaluating the PK, safety, and efficacy of a number of gel- and IVR-based microbicide candidates and informing clinical trial designs due to their great similarity to women in vaginal architecture, menstrual cycling, and microfloral composition (48–57). Here, we further evaluate IQP-0528 released from a quick-dissolving film with and without nanoparticle formulation and report the in vivo safety and bioavailability in pigtailed macaques in a multiday PK study with repeated applications of the film.

MATERIALS AND METHODS

We previously reported the development and optimization of a polymeric, quick-dissolving, stable, nontoxic film as a solid dosage form for the vaginal delivery of IQP-0528 with high efficacy in vitro against HIV (46). These films contained 0.1% (wt/wt) IQP-0528 and dissolved at pH 7.

IQP-0528 nanoparticle formulation for controlled release drug delivery.

Nanoparticles containing the NNRTI IQP-0528 were encapsulated into a combination PLGA-Eudragit S 100 formulation via a double emulsion process (58). Nanoparticles with Eudragit S 100 were incorporated 1:1 into 5,000- to 20,000-Da monomethoxy(polyethylene glycol) (mPEG)-PLGA formulations. The nanoparticles were 434 ± 46 nm (mean ± standard deviation) in diameter, with a zeta potential of −3.26 mV (47).

Quick-dissolving vaginal films containing IQP-0528.

Vaginal films containing the IQP-0528 nanoparticles (IQP-0528NP) and vaginal films containing unencapsulated IQP-0528 (IQP-0528) were manufactured through the solvent casting and evaporation method as previously described (34, 46). Briefly, the polymeric film excipients were mixed in a water solution, followed by the addition of 1.5% (film, wt/wt) IQP-0528 or the equivalent drug concentration of IQP-0528 nanoparticles. The polymeric solution was cast into thin sheets on an Elcometer 4340 automatic film applicator (Elcometer, Rochester Hills, MI) to evaporate the solvents, resulting in the final quick-dissolving polymeric film. The physicochemical properties of the films were determined as previously described and had the following characteristics: area, 1,200 mm²; mass, 235 ± 4 mg; opaque white appearance and smooth and pliable texture; thickness, 0.22 ± 0.01 mm; visual disintegration time, 8.6 ± 1.5 min; water content, 4.635% ± 0.049%; tensile strength, 9.675 ± 0.330 N; puncture strength, 6.710 ± 0.454 N; dissolves at pH 7 (34). The dissolution of the films was pH independent and controlled by the vaginal fluid volume. The IQP-0528 content of the films was determined by high-performance liquid chromatography (HPLC) stability indicating analytical methods previously developed (41). Both the nanoparticle (IQP-0528NP) and the unencapsulated (IQP-0528) films contained a measured IQP-0528 drug content of 1.50% ± 0.08% (wt/wt). The increase in drug loading from previously published products did not result in any negative effects in stability or performance (data not shown) (46). For the nonhuman primate studies, the human-sized vaginal films were scaled down proportionally to the size of the primate vagina: a 25% reduction of the film surface area (22 by 44 by 0.1 mm; providing 75% of a human dose).

In vitro release of IQP-0528 from vaginal films.

The in vitro release of IQP-0528 (1.5%, wt/wt) from the quick-dissolving films and the nanoparticle-encapsulated (1.5%, wt/wt drug equivalent) quick-dissolving films was evaluated in a USP 4 apparatus (Sotax CP 7) as previously detailed (34, 41). The flow rate was set to 15 ml/min at a temperature of 37°C. A mixture of 10% acetonitrile in Dulbecco's phosphate-buffered saline solution (pH 7.0) was the dissolution medium. IQP-0528 content was determined by HPLC (41). The release of IQP-0528 from the film was found to be 90.25% ± 2.32% by 30 min, with complete drug recovery by 1 h. Conversely, the in vitro release of IQP-0528 from the nanoparticles resulted in a burst release of 37.72% ± 8.45% available drug at 10 h and 51.65% ± 7.22% available drug at 24 h.

Nonhuman primate studies.

All macaques were housed at the Centers for Disease Control and Prevention according to the protocols in the Guide for the Care and Use of Laboratory Animals (59). All the study procedures were approved by the Centers for Disease Control and Prevention Institutional Animal Care and Use Committee. The sample collection and the timelines are listed below and employed published protocols (60, 61).

Twenty-four-hour PK analyses.

Six sexually mature female pigtailed macaques (Macaca nemestrina) were enrolled in a two-arm crossover PK study (IQP-0528NP and IQP-0528 film formulations) with a 2-week washout in between the crossovers. The crossover PK study was repeated 2 months later to obtain a larger number of samples for the study. Three pigtailed macaques received the IQP-0528NP film (1.5%, wt/wt); the other three pigtailed macaques received the IQP-0528 film (1.5%, wt/wt). The vaginal films were inserted at time zero in the upper half of the posterior vagina (halfway between the cervix and introitus). The vaginal film disintegration was visually inspected at 1 h after film application. Visual examination for abrasions or inflammation was also performed. Vaginal secretions were obtained proximal (ectocervix) and distal (introitus) to the cervix at 0, 1 (Weck-Cel spears; Beaver Visitec), 4, and 24 h (eight Ultracell surgical sponges, 3.5 by 4 mm). Three vaginal punch (3 mm; proximal, medial, and distal to the cervix) and two rectal punch biopsy specimens were collected at 24 h with Miltex Townsend number 30-1445 biopsy forceps. Variations of the vaginal pH were monitored from vaginal secretions obtained at 0, 1, 4, and 24 h by rolling a Dacron swab with the secretion onto a pH colorimetric indicator strip (Millipore Billerica, MA, USA). The vaginal pH values of animals that were undergoing menstruation were not readable due to the bleeding and exudation. A second swab sample was collected to evaluate changes in vaginal microflora at baseline (time zero) and 24 h after film application (one crossover PK only). These were placed individually in Port-a-Cul (Becton Dickinson, Franklin Lakes, NJ) tubes and transported to Magee Women's Research Institute (Pittsburgh, PA) within 24 h of collection for microbial analysis. The swabs were characterized for the presence of aerobic and anaerobic bacteria by semiquantitative culture as described previously (40, 60). Hydrogen peroxide production by lactobacilli and viridans group streptococci was tested qualitatively on tetramethylbenzidine agar plates (62). Based on the absence of significant differences in the vaginal microflora of pigtailed macaques following the presence of IQP-0528 vaginal rings for 28 days, it was demonstrated that IQP-0528 was not detrimental to the native vaginal microflora even when present for a long time (40). Hence, in this study, acute changes in the vaginal microbial milieu were only monitored for the 24-h PK studies.

Repeated-dose PK.

The six macaques utilized for the 24-h PK study were enrolled in a multiday PK study with repeated applications of the film in a 23-day period, following a 2-month washout between the 24-h and the repeated-dose PK studies. As described above, three pigtailed macaques received the IQP-0528NP film and the other three animals received the IQP-0528 film on day 0. The sample collection was identical to that in the 24-h PK study. Repeated applications of the film were carried out on days 7, 10, 14, 17, and 20 to determine the duration of IQP-0528 delivery. Samples were obtained on days 7, 14, and 23 to quantitate the bioavailability of IQP-0528 7, 4, and 3 days following the last film application, respectively. The sample collection schedule is listed in Table 1. All samples on days 7 and 14 were collected prior to new film insertion. We monitored for mucosal inflammation by the measurement of cytokines in vaginal secretions obtained with Weck-Cel spears (0 and 1 h after film application) and Ultracell surgical sponges (4 h after film application on days 7, 14, and 23) using a Milliplex MAP (Millipore, Billerica, MA, USA) fluorescent bead-based multiplex assay as previously described (40, 61).

TABLE 1.

Repeated-dose PK and safety study sample collection schedule^a

Procedure or sample type	Time of film insertion or sampling
	0 minb	60 min	4 h	24 h	Day 7b	Day 10	Day 14b	Day 17	Day 20	Day 23
Insert film	X				X	X	X	X	X
Bloodc	X	X	X	X	X	—	X	—	—	X
Vaginal fluidd	X	X	—	—	—	—		—	—
Vaginal fluide	—	—	X	X	X	—	X	—	—	X
Punch biopsyf	—	—	—	X	—	—	—	—	—	X

^aFemale pigtailed macaques received IQP-0528NP film (n = 3) or IQP-0528 film (n = 3).

^bSamples were collected immediately prior to film application.

^cSamples were collected in CPT tubes, and plasma was used for drug analysis.

^dSamples were collected with Weck-Cel spears from locations proximal and distal to the cervix.

^eSamples were collected with Ultracell surgical sponges (8 plugs: 4 proximal and 4 distal to the cervix).

^fVaginal punch biopsy specimens were taken from locations proximal, medial, and distal to the cervix, and two rectal punch biopsy specimens were taken.

The 24-h data from the two crossover PK studies (n = 12 for each arm, IQP-0528NP film and IQP-0528 film) were combined with the data obtained in the first 24 h from the repeated-dose PK study (n = 3 for each arm, IQP-0528NP film and IQP-0528 film) to obtain a total of 15 sets of data for each arm.

Quantitation of pyrimidinedione in plasma, tissue, vaginal lavage fluid, and secretion samples.

IQP-0528 in plasma and vaginal fluid samples collected at 0, 1, 4, and 24 h and days 7, 14, and 23, and in vaginal and rectal biopsy specimens was quantitated by liquid chromatography-tandem mass spectrometry (LC-MS/MS) as described elsewhere (40, 63). Briefly each sample was extracted with 0.2% acetonitrile and analyzed using a Phenomenex phenyl hexyl column with a 0.2% formic acid–acetonitrile mobile phase. The linear gradient of this assay was 20% to 98%. IQP-0528 was determined to be stable in freeze-thawed plasma and vaginal fluid samples. The lower limit of quantification (LLOQ) was determined to be 5 ng/sample of matrix for tissue and vaginal fluid obtained with spears. The LLOQ for vaginal fluid obtained with the Ultracell surgical sponges was 25 ng/sample of matrix after extraction, and values were corrected for the volume collected on the sponge. The LLOQ was 5 ng/ml for plasma. The vaginal fluid and tissue densities were assumed to be 1.0 g/ml to convert weight/weight concentrations of IQP-0528 (nanograms of IQP-0528 per gram of vaginal fluid or tissue) to molarity (μM).

Data analysis.

The changes in cytokines and chemokines of the macaques receiving IQP-0528NP and IQP-0528 film in the repeated-dose PK studies were monitored by Friedman tests of the log-transformed values. These were followed by Wilcoxon signed-rank test with false discovery rate (FDR)-adjusted P values for post hoc pairwise comparisons between each of 5 time points (1 and 4 h and days 7, 14, and 23) and 0 h (57). Wilcoxon matched-pairs signed-rank test was used to compare the proximal and distal drug measurements between the IQP-0528NP and the IQP-0528 film groups and within groups as well.

RESULTS

Safety.

We analyzed the in vivo safety of the IQP-0528NP and IQP-0528 film formulations through visual examination and vaginal microflora, pH, and cytokine and chemokine measurements. Upon routine visual examinations while obtaining mucosal samples, no vaginal abnormalities such as tissue abrasion or inflammation were noted with either film application in the 24-h PK studies or following repeated dosings and at all time points throughout the study, including day 23. To evaluate whether the IQP-0528 film formulations triggered detrimental changes to the normal vaginal microflora, vaginal fluid at baseline (time zero) and 24 h after film application was characterized for changes in the occurrence of aerobic and anaerobic microorganisms. The prevalences of normal aerobic microorganisms, such as H₂O₂-producing and non-H₂O₂-producing lactobacilli and viridans group streptococci and diphtheroids, and anaerobic species, such as anaerobic Gram-negative rods (black and nonpigmented), were found to be generally stable with sporadic variations 24 h after the application of either film (Fig. 1). A similar stable distribution was seen among 16 other vaginal microbial species (data not shown). Transient and sporadic variations were noted for some species, such as H₂O₂-producing viridans group streptococci, and the same has been observed in pigtailed macaques previously (40, 64). The prevalences of the H₂O₂-producing and non-H₂O₂-producing lactobacilli and viridans group streptococci and diphtheroids and anaerobic Gram-negative rods (black and nonpigmented) were found to be similar in the IQP-0528NP and the IQP-0528 film groups at 24 h (data not shown).

FIG 1.

Vaginal microbiological composition in macaques from one crossover 24-h PK study that received the IQP-0528NP (n = 6) and IQP-0528 (n = 6) films: prevalence (number of macaques that were positive) of a set of microorganisms representative of the normal vaginal microflora in macaques with IQP-0528NP film (a) or IQP-0528 film (b) at baseline (0 h) and 24 h after film application. Lacto, lactobacilli; Viridans, viridans group streptococci; GNR, Gram-negative rods; S. aureus, Staphylococcus aureus.

The vaginal pH was monitored at baseline and 1, 4, and 24 h after film application. Neither film produced a significant change in the pH at any of the time points tested (Fig. 2). All macaques maintained their pH within the range of 4.5 to 8 that we have typically seen in pigtailed macaques (61). The mean pH and standard deviation at baseline for all of the animals was 7.36 ± 0.91 (IQP-0528NP film) and 7.20 ± 0.70 (IQP-0528 film). Friedman's test found no significant change in the pH after product application (1, 4, and 24 h) compared to 0 h for the IQP-0528NP film and the IQP-0528 film (P = 0.0923 and P = 0.4753, respectively) (Fig. 2). Though 20% of the animals showed a 1- to 2-log change in pH from baseline to 24 h after application, it was not statistically significant (Wilcoxon matched-pairs signed-rank test, P = 0.0585). The pH range noted in these macaques over the course of the study is normally seen in pigtailed macaques (48, 61).

FIG 2.

Vaginal pH values in macaques that received the IQP-0528NP and IQP-0528 films. (a) Vaginal pH was measured prior to film application (0 min) and 1, 4, and 24 h after film application. Each symbol represents the mean of 15 measurements, and the standard deviations are indicated by error bars. (b) Differences in vaginal pH between the baseline (0 min) and 1, 4, and 24 h after IQP-0528 film application. The symbols represent the mean pH difference from the baseline, and the error bars indicate the standard deviations.

The induction of mucosal inflammation, if any, due to the presence of the IQP-0528NP or IQP-0528 film was monitored by analysis of 18 cytokine and chemokine measurements from vaginal fluid samples obtained from the macaques at 0, 1, and 4 h and at days 7, 14, and 23. Friedman tests found differences over time in the IQP-0528NP film group in the levels of macrophage inflammatory protein 1β (MIP-1β) and interleukin 18 (IL-18) and in the IQP-0528 film macaques in granulocyte colony-stimulating factor (G-CSF), IL-15, IL-1 receptor antagonist (IL-1Ra), and IL-13 (see Tables S1 and S2 in the supplemental material). However, when the analysis included Wilcoxon signed-rank test confirmed with false discovery rate (FDR)-adjusted P values for post hoc pairwise comparisons between each of 5 time points (1 and 4 h and days 7, 14, and 23) and 0 h, these changes were found to arise due to differences between time points (1 and 4 h and days 7, 14, and 23), suggesting that these changes were not due to delivery device use, as shown previously (57). No statistically significant differences were noted in any of the other cytokines or chemokines measured in this study, including IL-1β, which showed increased levels on days 7 and 14 in the IQP-0528NP film macaques.

Pharmacokinetic drug measurements.

The film was found to be completely dissolved upon visual examination at 1 h after film application, indicating a disintegration time of less than 60 min for both the IQP-0528NP and IQP-0528 film. Tables 2 and 3 summarize the PK parameters of IQP-0528 among both groups across many mucosal compartments at various time points. Drug measurements were below the LLOQ (5 ng/ml) in all plasma samples tested. Drug was detected in all vaginal fluid samples obtained within the first 24 h (Table 2; Fig. 3A and B). The median vaginal fluid drug levels were similar among samples obtained from locations proximal and distal to the cervix, with the exception of the proximal samples obtained at 4 h (IQP-0528 film group) and 24 h (IQP-0528NP film group), which showed statistically significant increases (P ≤ 0.05 and 0.001, respectively) in comparison to the levels in the distal samples. The data in Table 3 show that the maximum concentrations (C_max) obtained proximal to the cervix were similar among the IQP-0528NP and the IQP-0528 film macaques (1.06 × 10⁴ and 1.09 × 10⁴ μg/ml, respectively). The C_max values in vaginal fluid samples obtained distally were higher in the IQP-0528 film group than in the IQP-0528NP film group (1.58 × 10⁴ and 3.72 × 10³ μg/ml, respectively). Twofold-higher proximal and threefold-higher distal area under the concentration-time curve (AUC_0–24) values were noted for the IQP-0528 film group than for the IQP-0528NP film group. Vaginal fluid samples obtained on days 7, 14, and 23 to quantitate the bioavailability of IQP-0528 7, 4, and 3 days following the last film application were highly variable with regard to drug detection (range 0.13 to 238.92 μg/ml), with only 25% above the LLOQ with the IQP-0528NP films and 31% above the LLOQ with IQP-0528 films (data not shown).

TABLE 2.

IQP-0528 drug concentrations across mucosal sites in macaques that received IQP-0528NP film and IQP-0528 film

Sample type, location relative to cervix	Value (μg/ml [vaginal fluid] or μg/g [vaginal tissue]) at indicated time after film application fora:
	IQP-0528NP film at:									IQP-0528 film at:
	1 h			4 h			24 h			1 h			4 h			24 h
	No. of samples	% above LLOQa	Median (range)	No. of samples	% above LLOQ	Median (range)	No. of samples	% above LLOQ	Median (range)	No. of samples	% above LLOQ	Median (range)	No. of samples	% above LLOQ	Median (range)	No. of samples	% above LLOQ	Median (range)
Vaginal fluid
Proximal	15	100	112.39 (13.13–1,213)	30	100	482.48 (8.68–10,560)	30	100	70.81 (1.13–1,100)	15	100	58 (5.60–2,104)	30	100	340.14 (5.06–10,940)	30	100	931.80 (8.42–4,045)
Distal	15	100	170.95 (0.54–726.79)	30	100	52.73 (4.15–3,722)	30	100	32.49 (1.1–546.74)	15	100	185.29 (2.73–2,027)	30	100	117.49 (1.86–15,800)	30	100	260.45 (8.26–3,700)
Vaginal tissue
Proximal	ND3	ND	ND	ND	ND	ND	15	67	1.08 (0.03–78.88)	ND	ND	ND	ND	ND	ND	15	80	1.87 (0.03–222.58)
Medial	ND	ND	ND	ND	ND	ND	15	73	1.32 (0.03–38.75)	ND	ND	ND	ND	ND	ND	15	80	3.82 (0.03–174.6)
Distal	ND	ND	ND	ND	ND	ND	15	67	0.64 (0.03–30.25)	ND	ND	ND	ND	ND	ND	15	73	3.09 (0.03–55.55)

^aFor each arm (IQP-0528NP film and IQP-0528 film), n = 15. LLOQ, lower limit of quantification; ND, not done.

TABLE 3.

Summary of vital PK parameters^a

Sample type, film formulation, sampling location relative to cervix	Cmax (μg/ml)	Tmax (h)	AUC0–24 h (h · μg/ml)	V (liters)	CL (ml/day)
Vaginal fluid
IQP-0528NP
Proximal	10,561	4	6,481	0.0032	0.0694
Distal	3,722	4	1,273	0.0072	0.3534
IQP-0528
Proximal	10,939	4	13,346	0.0013	0.0337
Distal	15,795	4	4,326	0.0026	0.1040
Vaginal tissue
IQP-0528NP
Proximal	78.87	NA	12.99	0.42	34.65
Medial	38.75	NA	15.82	0.34	28.44
Distal	30.25	NA	7.63	0.71	58.95
IQP-0528
Proximal	222.58	NA	3.44	0.24	130.96
Medial	174.60	NA	45.79	0.12	9.83
Distal	55.55	NA	37.06	0.15	12.14

^aData from the two-arm crossover 24-h PK studies (n = 12) were combined with the data from the first 24 h of the repeated-dose PK study for each arm to generate the PK parameters (n = 15). C_max, maximum concentration; T_max, time to C_max; AUC, area under the concentration-time curve; AUC_0–24, AUC from the time of film placement to the 24-h sample collection; V, volume of distribution; CL, clearance; NA, not applicable.

FIG 3.

IQP-0528 concentrations in macaques that received the IQP-0528NP and IQP-0528 films: IQP-0528 concentrations were measured at 1, 4, and 24 h after film application in vaginal fluid samples obtained at locations proximal (A) and distal (B) to the cervix and at 24 h in vaginal tissue specimens obtained at locations proximal, medial, and distal to the cervix (C). Samples with drug concentrations below the LLOQ were assigned one-half of the LLOQ (i.e., 25 ng/ml). The horizontal lines represent the median concentrations. ***, P ≤ 0.0001; *, P ≤ 0.05.

Median vaginal tissue concentrations of 1.08, 1.32, and 0.64 μg/g were detected in tissue samples obtained from locations proximal, medial, and distal to the cervix, respectively, in the IQP-0528NP film group (Fig. 3C; Table 2). The proximal, medial, and distal IQP-0528 concentrations in the IQP-0528 film macaques were 1.87, 3.82, and 3.09 μg/g, respectively. Fourteen vaginal tissue samples in the IQP-0528NP film group and 10 in the IQP-0528 film group were found to be below the LLOQ. A statistically significant increase in the amount of IQP-0528 detected in vaginal tissue samples obtained proximal to the cervix was only noted in the IQP-0528 film group compared to the IQP-0528NP film group (Wilcoxon matched-pair signed-rank test, P = 0.0494), though a trend toward an increased concentration of IQP-0528 was noted for samples obtained from locations medial and distal to the cervix in the IQP-0528 film group. All vaginal tissue specimens obtained on day 23 (3 days after last film application) were below the LLOQ for IQP-0528 (data not shown). Drug levels were below the LLOQ in all rectal tissue samples tested.

DISCUSSION

This study demonstrates that quick-dissolving IQP-0528 vaginal film formulations exhibit promising safety and pharmacokinetics. Using the well-defined pigtailed macaque model, these films were found to exhibit preliminary safety and deliver high mucosal levels of IQP-0528, between 1 and 5 log greater than the in vitro IC₉₀ of 0.146 μg/ml, as early as 1 h after film application and even 24 h later, suggesting that these polymeric quick-dissolving vaginal films may deliver IQP-0528 at concentrations sufficient to prevent HIV infection.

Films have traditionally been found to be safe in humans for oral and vaginal use. They are used to orally deliver pain medications, vitamins, and supplements (65–67). Microbicide drugs, such as dapivirine, retrocyclin-101, cellulose acetate phthalate, and dapivirine in combination with TFV, have been formulated into films for vaginal use, and VCF containing nonoxynol-9, a VCF lubricating film, and a vaginal cleansing film (Apothecus Pharmaceuticals) are all available commercially (25, 28–32). It is important to evaluate the safety of topical PrEP products, as any breach to the integral mucosal barrier and the associated inflammation promote the recruitment of local CD4 T cells and facilitate HIV infection (68). The PrEP product should not alter the innate vaginal microbial milieu, as altered microbial states may increase the risk of HIV acquisition (69–71). Vaginal facultative and anaerobic bacteria remained stable throughout the study with both the IQP-0528NP and IQP-0528 film formulations (Fig. 1). A lactobacillus-dominated vaginal microbiome is indicative of a healthy vaginal microenvironment in women and protects against pathogenic microorganisms (72). Pigtailed macaques have a similar frequency of lactobacillus colonization in their vaginal tract (48). A single application of either film formulation did not disturb the vaginal lactobacilli or the pH (Fig. 2). The presence of an IQP-0528 IVR for 28 days in pigtailed macaques did not cause a significant change in the vaginal microfloral composition (40). It was demonstrated that IQP-0528 was not detrimental to the native vaginal microflora even when present for a long time. Therefore, the vaginal microbial milieu in this study was only monitored at the time of insertion and 24 h later. The mucosal cytokine and chemokine concentrations, in particular the proinflammatory cytokines, remained stable throughout the study (see Tables S1 and S2 in the supplemental material). Changes in a few cytokines were noticed; however, these were found to arise due to variability between time points, according to statistical analysis with FDR-adjusted P values. This observation is typical and has been reported previously in pigtailed macaques (57). Overall, the findings suggest that both film formulations of IQP-0528 are safe and do not perturb the vaginal environment.

The recently completed phase I trial in women with dapivirine film and gels demonstrated a reduction in the innate anti-HIV activity of cervicovaginal lavage (CVL) fluid samples obtained from the gel users and not the film users (35). It was reported that dilution, due to the greater volume used with gels than with films, could likely account for the reduction in the innate anti-HIV activity. The reduction in innate anti-HIV activity was also observed 3 weeks after the last gel application. Therefore, the use of IQP-0528 films would likely not alter the innate anti-HIV activity in CVL of women.

IQP-0528 was chosen as the lead pyrimidinedione candidate for topical PrEP and has been formulated as a single entity or as a combination microbicide into IVRs and gels (40–44). IQP-0528 formulated in films showed subnanomolar efficacy in CEM-SS cells and human peripheral blood mononuclear cells (PBMCs) against HIV-1_IIIB and clinical strains of HIV-1, respectively (46). These films may offer additional advantages over other dosage forms, such as lower production costs due to their small size. The solid dosage form may increase the stability of a drug by preventing its precipitation and degradation by hydrolysis or oxidation. Low residence time has been reported with traditional vaginal dosage forms, such as suppositories and liquid formulations, due to the self-cleaning action of the vaginal tract (73, 74). The bioadhesive properties of films may increase the local retention time and improve clinical performance (65–67). The median vaginal fluid concentrations of IQP-0528 obtained 60 min after film application were 3 log greater than the in vitro IC₉₀ of 0.146 μg/ml (45), with a time to maximum concentration of drug in serum (T_max) of 15,795.08 μg/ml at 4 h. The median vaginal fluid level obtained 24 h after application was 3.3 log higher than the in vitro IC₉₀ (Fig. 3A and B). The dynamic range in vaginal fluid concentrations of IQP-0528 might be a result of IQP-0528's characteristic of moving in and out of cells and the in vivo variability in the volume of vaginal secretions in macaques, as the films, when applied to the mucosa, undergo hydrolysis to release the API. Though no studies to date have quantitated the cervicovaginal fluid volume in macaques, the median volume in women was estimated to be 0.51 ml, with an interquartile range of 0.33 to 0.69 ml (75). The median vaginal tissue concentration obtained at 24 h after film application was 21 times greater than the in vitro IC₉₀ of 0.146 μg/ml (Fig. 3C). High concentrations of IQP-0528 were detected in vaginal fluid and tissue samples obtained from locations proximal, medial, and distal to the cervix. This observation suggests that a uniform distribution of IQP-0528 is obtained in the vaginal cavity with these polymeric films. IQP-0528's activity is not affected by the presence of semen (6.25%, vol/vol), suggesting that the NNRTI would remain effective during or after coitus (42). Along with coitus, the uniform distribution could potentially minimize areas in the vaginal tract that are more vulnerable to HIV infection due to a concentration gradient (68). The low clearance (CL), which signifies higher steady-state concentrations of IQP-0528, and the low volume of distribution (V), the ratio between the dose given and the concentration in the vaginal fluid or tissue, indicate that a large amount of active drug is available at the site of action. We observed that plasma concentrations of IQP-0528 were below the analytical LLOQ, a unique advantage observed with topical PrEP, as it could limit the emergence of drug resistance and systemic side effects. The rapid disintegration and the high mucosal fluid and tissue concentrations suggest that these films may provide a quick, viable option against HIV-1 transmission.

Although vaginal films deliver inherently lower drug-dosing levels than other delivery methods, the measurable drug concentrations are similar to those observed with gels and intravaginal rings. IQP-0528 delivery through IVR, gels, and films has been evaluated in the macaque nonhuman primate model. The 14% (wt/wt) IVRs provided a sustained release of 0.5 mg/day IQP-0528 over 28 days (40), whereas gels delivered 15 mg of IQP-0528 in a single application (76). In contrast, the quick-dissolving vaginal films described here delivered only 450 μg IQP-0528 upon application. However, the median vaginal fluid IQP-0528 concentration obtained with the films at 4 h (1.82 × 10⁵ ng/ml) was similar to those obtained with IVRs (6.08 × 10⁵ ng/ml at day 3) and gels (4.56 × 10⁵ ng/ml at 4 h). Likewise, the median vaginal tissue concentrations were comparable among the films (3.10 μg/g), IVRs (2.97 μg/g), and gels (18.8 μg/g). The measurable drug concentrations obtained with the films illustrate that it is likely that the IQP-0528 film could deliver the drug at concentrations similar to those obtained with other delivery modalities, such as IVRs and gels, up to 24 h after application.

Nanoparticles are an attractive and efficient mucosal drug delivery system that provides sustained and controlled release of a variety of biologically active agents, such as small molecules and proteins, with desirable features such as increased solubility, targeted delivery, enhanced cellular uptake, protection of the active ingredient from degradation, deep tissue penetration, and lowered risk of systemic toxicity by reducing the dose of drug needed for therapeutic purposes (77–82). PLGA-based encapsulation of nanoparticles is widely used for drug delivery because of its biodegradability, allowing its reabsorption by the body and, hence, lowering toxicity (83). PLGA nanoparticles have been demonstrated to intravaginally deliver small interfering RNA (siRNA) and Rantes (58, 81). Therefore, the incorporation of nanoparticle encapsulation into a microbicide formulation has the potential to confer long-term protection from a single dose.

In vaginal tissue and secretion samples, the data for the IQP-0528NP film formulation were similar to the data for the IQP-0528 film formulation. In in vitro drug release studies, the release profile of the IQP-0528NP films was orders of magnitude longer than that of the IQP-0528 films. The IQP-0528 films released all of the incorporated drug within an hour. The IQP-0528NP films produced a small burst release of IQP-0528 from the nanoparticles in a similar time frame, with only 37.72% ± 8.45% of the loaded drug being available after 10 h of dissolution. Following that burst release, IQP-0528 was constantly and slowly released from the circulating nanoparticles for over 10 days. These data supported that the IQP-0528NP could provide long-term drug delivery. However, in vivo, the nanoparticle encapsulation did not provide any greater advantage for IQP-0528 distribution. Limited drug detection in vaginal fluid and no detection in vaginal tissues 24 h after film application even upon repeated film application suggests that the nanoparticles do not have an extended residence time in vaginal tissue. The IQP-0528 nanoparticles appear to have a clearance rate in vaginal tissue similar to that of the IQP-0528 molecule itself. The IQP-0528NP may not be able to provide controlled long-term release of IQP-0528 as observed during in vitro testing, where clearance is not an issue. Therefore, there may be no in vivo long-term pharmacokinetic benefits of delivering IQP-0528 in the current formulation.

A limitation of the current study is that no preclinical or clinical data exist that define the pharmacologically relevant concentrations of IQP-0528 that are needed for protection against HIV-1. IQP-0528 has been evaluated against a panel of NRTI- and NNRTI-resistant viruses. The compound retains full activity against some NNRTI-resistant viruses but does lose some potency against others. Against the strains where some resistance is observed, the compound exhibits 50% effective concentrations (EC₅₀s) that range between 100 and 900 nM. In addition, one macaque in the IQP-0528NP film group menstruated during the analysis. Higher median proximal vaginal fluid IQP-0528 concentrations were obtained at 1, 4, and 24 h (126.6, 534.59, and 84.72 μg/ml, respectively) while excluding the menstruating animal than while including the menstruating animal (Table 2). The distal vaginal fluid concentrations (179.57, 50.23, and 36.06 μg/ml at 1, 4, and 24 h, respectively) were similar among the two groups. A greater number of animals would be required to determine the influence of menstruation on drug absorption. Despite these limitations, this study demonstrates that quick-dissolving IQP-0528 vaginal film formulations exhibit promising safety and pharmacokinetic profiles. The IQP-0528 films demonstrated that relatively low drug-dosing levels are able to reach above IC₉₀ values. This formulation compares well with other film formulation developments, where a dapivirine film showed high PK values in vaginal tissue. Concentrations were detected at levels that can protect against ex vivo HIV infection (35).

The next step in developing this formulation is the performance of additional PK studies in macaques, followed by simian-human immunodeficiency virus (SHIV) challenge studies to determine the pharmacodynamics. Subsequent to these studies, PK studies in human tissue will further inform the development of IQP-0528 vaginal films. The excellent coverage and mucosal levels of IQP-0528, well above the IC₉₀, suggest that these films hold potential to serve as a quick, viable, woman-initiated prevention option in high-HIV-incidence settings.