A Subcutaneous Implant of Tenofovir Alafenamide Fumarate Causes Local Inflammation and Tissue Necrosis in Rabbits and Macaques.

We describe the in vitro and in vivo evaluation of a subcutaneous reservoir implant delivering tenofovir alafenamide hemifumarate (TAF) for the prevention of HIV infection. These long-acting reservoir implants were able to deliver antiretroviral drug for over 90 days in vitro and in vivo We evaluated the implants for implantation site histopathology and pharmacokinetics in plasma and tissues for up to 12 weeks in New Zealand White rabbit and rhesus macaque models. A dose-ranging study in rabbits demonstrated dose-dependent pharmacokinetics and local inflammation up to severe necrosis around the active implants. The matched placebos showed normal wound healing and fibrous tissue encapsulation of the implant. We designed a second implant with a lower release rate and flux of TAF and achieved a median cellular level of tenofovir diphosphate of 42 fmol per 106 rhesus macaque peripheral blood mononuclear cells at a TAF dose of 10 µg/kg/day. This dose and flux of TAF also resulted in adverse local inflammation and necrosis near the implant in rhesus macaques. The level of inflammation in the primates was markedly lower in the placebo group than in the active-implant group. The histological inflammatory response to the TAF implant at 4 and 12 weeks in primates was graded as a severe reaction. Thus, while we were able to achieve a sustained target dose, we observed an unacceptable inflammatory response locally at the implant tissue interface.

Design of a Drug-Eluting Subcutaneous Implant of the Antiretroviral Tenofovir Alafenamide Fumarate.

PURPOSE: Sexual transmission of HIV has been clinically proven to be preventable with a once-daily oral tablet; however, missed doses dramatically increase the risk of HIV infection. Long-acting subcutaneous implants do not allow the user to miss a dose. A desirable long-acting drug-eluting implant can deliver a constant amount of drug, adjust the delivered dose, and be readily manufactured. We present a long-acting, subcutaneous implant design composed of tenofovir alafenamide hemifumarate (TAF) pellets loaded in a sealed polyether urethane tube for the prevention of HIV transmission. METHODS: Implants were prepared with pressed drug pellets and extruded polyurethane tubing. In vitro release rate of implants using different pellet formulations, rate-controlling membranes, and geometries were measured. RESULTS: Tenofovir alafenamide release appeared to be governed by a pseudo-steady state and followed a mass transport model of release from a cylindrical drug reservoir. Implant seal integrity was tested and confirmed using mechanical testing. The inclusion of sodium chloride in the pellet increased the release rate and reduced initial lag. The release was sustained for 100 days. CONCLUSIONS: The release rate of tenofovir alafenamide mechanistically varied with geometry and rate controlling membrane composition. The polyether urethane implant presented herein is modular and tunable to adjust the release rate and duration of the TAF release.

A Dose Ranging Pharmacokinetic Evaluation of IQP-0528 Released from Intravaginal Rings in Non-Human Primates.

PURPOSE: Design of intravaginal rings (IVRs) for delivery of antiretrovirals is often guided by in vitro release under sink conditions, based on the assumption that in vivo release will follow a similar release profile. METHODS: We conducted a dose-ranging study in the female reproductive tract of pigtail macaques using matrix IVRs containing IQP-0528, a poorly soluble but highly potent antiretroviral drug with an IC90 of 146 ng/mL. These IVRs consisted of drug-loaded segments, 15.6% IQP-0528 in Tecoflex 85A, comprising either all, half, or a quarter of the entire ring. RESULTS: In vitro release under sink conditions demonstrates loading-proportional release, with a cumulative 30-day release of 48.5 ± 2.2 mg for our 100% loaded ring, 24.8 ± .36 mg from our 50% loaded ring, and 13.99 ± 1.58 mg from our 25% loaded ring. In vivo, while drug concentration in vaginal fluid is well in excess of IQP-0528's EC90, we find no statistical difference between the different ring loadings in either swab drug levels or drug released from our rings. CONCLUSIONS: We show that in vitro release may not accurately reflect in vivo release, particularly for poorly soluble drugs. All tested loadings of our IVRs are capable of delivering IQP-0528 well in excess of the IC90.

Differential Mechanisms of Tenofovir and Tenofovir Disoproxil Fumarate Cellular Transport and Implications for Topical Preexposure Prophylaxis.

Intravaginal rings releasing tenofovir (TFV) or its prodrug, tenofovir disoproxil fumarate (TDF), are being evaluated for HIV and herpes simplex virus (HSV) prevention. The current studies were designed to determine the mechanisms of drug accumulation in human vaginal and immune cells. The exposure of vaginal epithelial or T cells to equimolar concentrations of radiolabeled TDF resulted in over 10-fold higher intracellular drug levels than exposure to TFV. Permeability studies demonstrated that TDF, but not TFV, entered cells by passive diffusion. TDF uptake was energy independent but its accumulation followed nonlinear kinetics, and excess unlabeled TDF inhibited radiolabeled TDF uptake in competition studies. The carboxylesterase inhibitor bis-nitrophenyl phosphate reduced TDF uptake, suggesting saturability of intracellular carboxylesterases. In contrast, although TFV uptake was energy dependent, no competition between unlabeled and radiolabeled TFV was observed, and the previously identified transporters, organic anion transporters (OATs) 1 and 3, were not expressed in human vaginal or T cells. The intracellular accumulation of TFV was reduced by the addition of endocytosis inhibitors, and this resulted in the loss of TFV antiviral activity. Kinetics of drug transport and metabolism were monitored by quantifying the parent drugs and their metabolites by high-performance liquid chromatography tandem mass spectrometry (HPLC-MS/MS). Results were consistent with the identified mechanisms of transport, and the exposure of vaginal epithelial cells to equimolar concentrations of TDF compared to TFV resulted in ∼40-fold higher levels of the active metabolite, tenofovir diphosphate. Together, these findings indicate that substantially lower concentrations of TDF than TFV are needed to protect cells from HIV and HSV-2.

Mass transfer in the dissolution of a multicomponent liquid droplet in an immiscible liquid environment.

The Epstein-Plesset equation has recently been shown to predict accurately the dissolution of a pure liquid microdroplet into a second immiscible solvent, such as oil into water. Here, we present a series of new experiments and a modification to this equation to model the dissolution of a two-component oil-mixture microdroplet into a second immiscible solvent in which the two materials of the droplet have different solubilities. The model is based on a reduced surface area approximation and the assumption of ideal homogeneous mixing [mass flux d(m(i))/dt = A(frac(i))D(i)(c(i) - c(s)){(1/R) + (1/(πD(i)t)(1/2)}] where A(frac(i)) is the area fraction of component i, c(i) and c(s) are the initial and saturation concentrations of the droplet material in the surrounding medium, R is the radius of the droplet, t is time, and D(i) is the coefficient of diffusion of component i in the surrounding medium. This new model has been tested by the use of a two-chamber micropipet-based method, which measured the dissolution of single individual microdroplets of mutually miscible liquid mixtures (ethyl acetate/butyl acetate and butyl acetate/amyl acetate) in water. We additionally measured the diffusion coefficient of the pure materials-ethyl acetate, butyl acetate, and amyl acetate-in water at 22 °C. Diffusion coefficients for the pure acetates in water were 8.65 × 10(-6), 7.61 × 10(-6), and 9.14 × 10(-6) cm(2)/s, respectively. This model accurately predicts the dissolution of microdroplets for the ethyl acetate/butyl acetate and butyl acetate/amyl acetate systems given the solubility and diffusion coefficients of each of the individual components in water as well as the initial droplet radius. The average mean squared error was 8.96%. The dissolution of a spherical ideally mixed multicomponent droplet closely follows the modified Epstein-Plesset model presented here.

Design and Testing of a Cabotegravir Implant for HIV Prevention.

Long-acting antiretroviral implants could help protect high-risk individuals from HIV infection. We describe the design and testing of a long-acting reservoir subcutaneous implant capable of releasing cabotegravir for several months. We compressed cabotegravir and excipients into cylindrical pellets and heat-sealed them in tubing composed of hydrophilic poly(ether-urethane) -. The implants have a 47 mm lumen length, 3.6 mm outer diameter, and 200 µm wall thickness. Four cabotegravir pellets were sealed in the membrane, with a total drug loading of 274 ± 3 mg. In vivo, the implants released 348 ± 107 µg/day (median value per implant, N = 41) of cabotegravir in rhesus macaques. Five implants generated an average cabotegravir plasma concentration of 373 ng/ml in rhesus macaques. The non-human primates tolerated the implant without gross pathology or microscopic signs of histopathology compared to placebo implants. Cabotegravir plasma levels in macaques dropped below detectable levels within two weeks after the removal of the implants.

The effect of hydrogen bonding on the diffusion of water in n-alkanes and n-alcohols measured with a novel single microdroplet method.

While the Stokes-Einstein (SE) equation predicts that the diffusion coefficient of a solute will be inversely proportional to the viscosity of the solvent, this relation is commonly known to fail for solutes, which are the same size or smaller than the solvent. Multiple researchers have reported that for small solutes, the diffusion coefficient is inversely proportional to the viscosity to a fractional power, and that solutes actually diffuse faster than SE predicts. For other solvent systems, attractive solute-solvent interactions, such as hydrogen bonding, are known to retard the diffusion of a solute. Some researchers have interpreted the slower diffusion due to hydrogen bonding as resulting from the effective diffusion of a larger complex of a solute and solvent molecules. We have developed and used a novel micropipette technique, which can form and hold a single microdroplet of water while it dissolves in a diffusion controlled environment into the solvent. This method has been used to examine the diffusion of water in both n-alkanes and n-alcohols. It was found that the polar solute water, diffusing in a solvent with which it cannot hydrogen bond, closely resembles small nonpolar solutes such as xenon and krypton diffusing in n-alkanes, with diffusion coefficients ranging from 12.5x10(-5) cm(2)/s for water in n-pentane to 1.15x10(-5) cm(2)/s for water in hexadecane. Diffusion coefficients were found to be inversely proportional to viscosity to a fractional power, and diffusion coefficients were faster than SE predicts. For water diffusing in a solvent (n-alcohols) with which it can hydrogen bond, diffusion coefficient values ranged from 1.75x10(-5) cm(2)/s in n-methanol to 0.364x10(-5) cm(2)/s in n-octanol, and diffusion was slower than an alkane of corresponding viscosity. We find no evidence for solute-solvent complex diffusion. Rather, it is possible that the small solute water may be retarded by relatively longer residence times (compared to non-H-bonding solvents) as it moves through the liquid.

Enzyme dehydration using Microglassification™ preserves the protein's structure and function.

Controlled enzyme dehydration using a new processing technique of Microglassification™ has been investigated. Aqueous solution microdroplets of lysozyme, α-chymotrypsin, catalase, and horseradish peroxidase were dehydrated in n-pentanol, n-octanol, n-decanol, triacetin, or butyl lactate, and changes in their structure and function were analyzed upon rehydration. Water solubility and microdroplet dissolution rate in each solvent decreased in the order: butyl lactate > n-pentanol > triacetin > n-octanol > n-decanol. Enzymes Microglassified™ in n-pentanol retained higher activity (93%-98%) than n-octanol (78%-85%) or n-decanol (75%-89%), whereas those Microglassified™ in triacetin (36%-75%) and butyl lactate (48%-79%) retained markedly lower activity. FTIR spectroscopy analyses showed α-helix to ß-sheet transformation for all enzymes upon Microglassification™, reflecting a loss of bound water in the dried state; however, the enzymes reverted to native-like conformation upon rehydration. Accelerated stressed-storage tests using Microglassified™ lysozyme showed a significant (p < 0.01) decrease in enzymatic activity from 46,560 ± 2736 to 31,060 ± 4327 units/mg after 3 months of incubation; however, it was comparable to the activity of the lyophilized formulation throughout the test period. These results establish Microglassification™ as a viable technique for enzyme preservation without affecting its structure or function.