A Phase 1 Randomized, Blinded Comparison of the Pharmacokinetics and Colonic Distribution of Three Candidate Rectal Microbicide Formulations of Tenofovir 1% Gel with Simulated Unprotected Sex (CHARM-02)

Abstract

CHARM-02 is a crossover, double-blind, randomized trial to compare the safety and pharmacokinetics of three rectally applied tenofovir 1% gel candidate rectal microbicides of varying osmolalities: vaginal formulation (VF) (3111 mOsmol/kg), the reduced glycerin vaginal formulation (RGVF) (836 mOsmol/kg), and an isoosmolal rectal-specific formulation (RF) (479 mOsmol/kg). Participants (n = 9) received a single, 4 ml, radiolabeled dose of each gel twice, once with and once without simulated unprotected receptive anal intercourse (RAI). The safety, plasma tenofovir pharmacokinetics, colonic small molecule permeability, and SPECT/CT imaging of lower gastrointestinal distribution of drug and virus surrogate were assessed. There were no Grade 3 or 4 adverse events reported for any of the products. Overall, there were more Grade 2 adverse events in the VF group compared to RF (p = 0.006) and RGVF (p = 0.048). In the absence of simulated unprotected RAI, VF had up to 3.8-fold greater systemic tenofovir exposure, 26- to 234-fold higher colonic permeability of the drug surrogate, and 1.5- to 2-fold greater proximal migration in the colonic lumen, when compared to RF and RGVF. Similar trends were observed with simulated unprotected RAI, but most did not reach statistical significance. SPECT analysis showed 86% (standard deviation 19%) of the drug surrogate colocalized with the virus surrogate in the colonic lumen. There were no significant differences between the RGVF and RF formulation, with the exception of a higher plasma tenofovir concentration of RGVF in the absence of simulated unprotected RAI. VF had the most adverse events, highest plasma tenofovir concentrations, greater mucosal permeability of the drug surrogate, and most proximal colonic luminal migration compared to RF and RGVF formulations. There were no major differences between RF and RGVF formulations. Simultaneous assessment of toxicity, systemic and luminal pharmacokinetics, and colocalization of drug and viral surrogates substantially informs rectal microbicide product development.

Trial registration: ClinicalTrials.gov NCT01575418.

Figures

FIG. 1.

CHARM-02 study flowchart. Pharmacokinetic (PK) analysis included eight participants who maintained study eligibility, received all three study products, and completed all three sampling periods. The safety analysis included all nine research participants enrolled regardless of eligibility throughout the study.

FIG. 2.

Number of overall Grade 2 adverse events (AEs) and subset of AEs (Grade 1 and 2) deemed related to study gels by product.

FIG. 3.

Median (interquartile range) plasma tenofovir (TFV) concentration (log10) for each time point by product without (a) and with (b) simulated unprotected receptive anal intercourse. Circle, rectal formulation (RF); diamond, reduced glycerin vaginal formulation (RGVF); triangle, vaginal formulation (VF).

FIG. 4.

Single photon emission computed tomography with transmission computed tomography (SPECT/CT) scan of radiolabel distribution (SPECT, color scale) in the distal gastrointestinal tract with anatomic reference to pelvis and spine (CT, amber scale). (a)Left panel is 10 ml of candidate “microbicide” aqueous gel labeled with 111In-diethylenetriaminepentaacetic acid (111In-DTPA), administered with syringe with luer lock tip applicator 2 h prior to imaging. Distribution is shown from the rectum to the splenic flexure. (b)Right panel is 2.5 ml of autologous semen labeled with 99mTc-sulfur colloid (100 nm HIV-sized particle in colloidal suspension) as “HIV” surrogate, administered 1 h after dosing of the candidate microbicide gel via simulated urethra in phallic device (coital dynamics simulator) and following 5 min of simulated receptive anal intercourse; imaging is 1 h after simulated sex. (c) Signal intensity (“concentration”) vs. distance plots resulting from three-dimensional tube fitting of scans shown in (a) and (b). Distance is relative to the coccygeal plane. PK-distance parameters (Dmax, DCmax, Dave, and Dmin) are derived from this concentration–distance data. Signal intensities are relative within each isotope scan and are not directly comparable between isotopes, therefore, HIV surrogate signal intensity is scaled to fit within the height of the microbicide intensity.

FIG. 5.

Correlation between radiolabeled small molecule surrogate, 111In-DTPA peak concentration in blood plasma (a), and percent urine recovery of 111In-DTPA (b) vs. tenofovir Cmax in plasma. Circle, rectal formulation (RF); diamond, reduced glycerin vaginal formulation (RGVF); triangle, vaginal formulation (VF).