CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

reproducibility in terms of physical properties including size and surface charge. Initial screening was conducted to determine apparent permeability (Papp) across Caco-2, as well as triple culture (Caco-2, HT29-MTX and Raji B cells) monolayers that can more accurately capture particle-specific mechanisms. The single-dose in vivo oral PK and tissue distribution of lead candidates were evaluated in male rats, followed by an assessment of steady-state PK for the best performing SDN. In vivo experiments were conducted at 20 mg / kg doses and compared with an aqueous pre-clinical formulation (20% cremaphor in water). Statistical analysis was conducted using an unpaired t-test. Results: Three SDNs demonstrated improved Papp providing 115%, 152%, and 200% increases compared to an aqueous DRV/r preparation across Caco-2 monolayers. Augmented single-dose PK was observed in rats for all three formulations. The lead SDN demonstrated a 3-fold increase in AUCτ (28,036 versus 8516 ng.hr/ml; P = 0.001) and a 2-fold increase in Cmin (3626 versus 1655 ng/ml; P = 0.0005) values at steady-state. Steady-state studies using a loading dose followed by a 50% lower dose maintenance demonstrated the ability for dose reduction while not compromising PK exposure. Conclusion: These data provide preclinical demonstration of a DRV/r SDN formulation with the potential for dose reduction whilst maintaining PK exposure. The approach utilises accepted excipients, which significantly de-risks future translation. The lead formulation will now be translated to spray-dry manufacturing to achieve scale necessary for first in human clinical evaluation. 481 CHARM-03:SAFETY, PHARMACOKINETICS, AND PHARMACODYNAMICS OF ORAL AND TOPICAL MARAVIROC Ian McGowan 1 , Rhonda M. Brand 1 , Jarret C. Engstrom 2 , Aaron Siegel 2 , Ashley Myerski 2 , Cindy Jacobson 2 , Mark A. Marzinke 3 , Craig W. Hendrix 3 , Alex Rinehart 4 , John Steytler 5 , Charlene S. Dezzutti 1 , Nicola Richardson-Harman 6 , Hans M. Spiegel 7 , Ross Cranston 1 1 University of Pittsburgh, Pittsburgh, PA, USA, 2 Magee–Womens Research Institute, Pittsburgh, PA, USA, 3 Johns Hopkins Hospital, Baltimore, MD, USA, 4 ViiV Healthcare, Research Triangle Park, NC, USA, 5 International Partnership for Microbicides, Paarl, South Africa, 6 Alpha StatConsult, LLC, Damascus, MD, USA, 7 NIAID, Rockville, MD, USA Background: Maraviroc (MVC) is being evaluated as a HIV-1 PrEP agent. In the non-human primate model, oral MVC did not provide protection from SHIV162p3 rectal challenge whereas topical rectal MVC was protective. Phase 1/2 studies have not demonstrated significant viral inhibition in the ex vivo / in vitro colorectal explant challenge model. The purpose of the CHARM-03 study was to characterize the safety, pharmacokinetic (PK), and pharmacodynamic profile of MVC following oral, vaginal, and rectal administration. Methods: Healthy HIV-1 uninfected men and women with homozygous wildtype CCR5 genotype were enrolled into a randomized open label crossover trial in which they received 300 mg oral MVC daily for 8 consecutive days, 1%MVC rectal gel for 8 consecutive days, and female participants received daily vaginally administered MVC 1% gel for 8 consecutive days. Each product administration was followed by a washout period of 14 to 21 days. Blood and tissue (cervical and rectal) were collected prior to and +2, +24 and +48 hours after the final dose of study product. One biopsy was snap frozen for PK and a second biopsy was incubated in culture medium for 2 hours before being snap frozen for PK. Tissue biopsies were challenged with HIV-1BaL as previously described (McGowan I et al. Lancet HIV 2016). Results: Twenty participants (11 male) were enrolled in the study. Twenty-five adverse events (24 Grade 1 and one Grade 2) were reported in 11 participants. Rectal PK at Day 8 of dosing (mean ± SD) was: oral tablet (7.9 ± 7.2 ng/mg), rectal gel (29.2 ± 36.5 ng/mg), and vaginal gel (0.1 ± 0.1 ng/mg). HIV-1 p24 inhibition was seen only in rectal tissue after oral dosing and only 1-2 hours post dose (p<0.05); inhibition was not detected in cervicovaginal tissue after oral dosing, nor in either tissue site with rectal or vaginal dosing. Pairwise comparison of snap frozen versus cultured tissue showed significant loss of MVC during 2 hours incubation (median Log 10 ± IQR) in rectal tissue following both oral tablet (1.16 ± 0.7) and rectal gel (0.96 ±1.04) and in cervical tissue following vaginal gel (1.34 ± 0.79) (p<0.0001) Figure 1. Conclusion: Oral and topical MVC were safe and well tolerated and the use of oral MVC was associated with colorectal explant viral inhibition. MVC significantly disassociated from explant tissue during incubation which may account for the absence of viral suppression after rectal and vaginal gel exposure.

479 DESIGN & SYSTEMS PHARMACOLOGY ANALYSIS OF ORAL LONG-ACTING ANTIRETROVIRAL THERAPY Morgan L. Craig 1 , Alison L. Hill 1 , Christopher D. Bruno 1 , Shirley S. Mo 1 , Ameya Kirtane 2 , Martin A. Nowak 1 , Robert Langer 2 , Carlo G. Traverso 2 1 Harvard University, Cambridge, MA, USA, 2 MIT, Cambridge, MA, USA Background: Universal coverage of antiretroviral therapy (ART) for prevention and treatment is complicated by high pill burdens and suboptimal adherence. Investigational long-acting (LA) ART injections/implants may be impractical in resource-limited settings. Here we develop an oral LA-ART delivery system, together with a systems pharmacology framework for pre-clinical prediction of drug kinetics and treatment outcomes. Methods: Our device consists of drug-loaded polymer arms with an elastic core. A 1cm capsule expands for gastric retention and slow release. Dolutegravir (DTG), rilpivirine (RPV) and cabotegravir (CAB) were incorporated based on lowmass. Drug kinetics were compared to immediate release (IR) forms in pigs. A physiologically-based pharmacokinetic (PBPK) model simulates drug absorption, distribution, metabolism, and excretion to get whole-body kinetics from physiochemical data. ART as treatment was simulated with an existing calibrated multi-strain viral dynamics model. ART as PrEP was simulated with a new calibrated mechanistic model of heterosexual transmission. Dose-response curves were parameterized from ex vivo assays and clinical trials. Fitness effects of known resistance pathways were assembled from the literature. Results: Prototype oral LA-ART devices in pigs reached Cmax after ~1 d (mean DTG 800, RPV 27, CAB 450 ng/mL) and continued to release drug for 7 d (mean Ctrough DTG 700, RPV 20, CAB 350 ng/mL), while the IR form reached comparable peaks (mean Cmax DTG 500, RPV 22, CAB 1300 ng/mL) but was undetectable by 2 d. The PBPK model recreated human IR kinetics (Cmax 3900, 200, 16200 ng/mL; tmax 2.5, 4.5, 1 hr for DTG, RPV, CAB) and was used to predict LA kinetics for varying doses (20-200% daily IR mass) and adherence. Models 1 yr maintenance monotherapy under simulated kinetics predicted similar rates of treatment failure for IR vs LA-ART in populations with realistically heterogeneous adherence. % of failures due to resistance increased with LA-ART for RPV only. Predicted changes in adherence patterns due to LA-ART could allow for 60% dose reductions. Modeling transmission recreated the observed efficacy of tenofovir-based PrEP, and predicts that LA-DTG reduces 1-year transmission risk by ~98% right after dosing, >50% for 10 days after dose, and may be superior to tenofovir. Conclusion: Orally-delivered LA-ART devices are feasible to engineer, effective in large animal models, and predicted by pharmacometric modeling to treat/ prevent HIV in humans. 480 PRECLINICAL EVALUATION OF A REDUCED DOSE DARUNAVIR/RITONAVIR NANOPARTICLE FORMULATION Helen J. Box 1 , Joanne Sharp 1 , Megan Neary 1 , Darren M. Moss 2 , Lee M. Tatham 1 , Hannah Kinvig 1 , Alison C. Savage 1 , Samantha Ashcroft 1 , David Back 1 , Steve Rannard 1 , Andrew Owen 1 1 University of Liverpool, Liverpool, UK, 2 Keele University, Keele, UK Background: Enhanced oral pharmacokinetics (PK) has previously been demonstrated for antiretroviral drugs by formation of solid drug nanoparticles (SDNs) using an emulsion-templated freeze-drying approach. Darunavir (DRV) has poor aqueous solubility, low bioavailability, and even with ritonavir- boosting (DRV/r) requires an 800mg daily dose. The current study describes the preclinical development of DRV/r SDNs that provide augmented PK exposure, potentially enabling dose reduction to reduce costs and make better use of available manufacturing capacity. If successful, the approach may also result in

Poster Abstracts

smaller oral dosage forms, facilitating future co-formulation strategies. Methods: A library of DRV/r SDN formulations were manufactured with acceptable excipients (FDA CDER-listed) and characterised to ensure

CROI 2018 172