Pharmacokinetic evaluation of tenofovir disoproxil fumarate released from an intravaginal ring in pigtailed macaques after 6 months of continuous use.

Although some pre-exposure prophylaxis modalities, in particular Truvada™ (oral TDF, the prodrug of tenofovir, in combination with emtrictabine) and topical tenofovir (TFV) gels, have been shown to be effective in reducing acquisition of HIV, it is pivotal to develop different delivery platforms to increase effectiveness by providing sustained topical drug delivery to overcome poor adherence. [ 1 , 3 , 10 , 25 , 26 ]. Intravaginal rings (IVR) have been used as a vaginal delivery device for contraception, hormone replacement therapy and treatment of endometriosis, dysmenorrhea, and polycystic ovarian syndrome [ 4 , 5 , 8 , 20 ]. The capacity of IVRs to provide sustained and controlled release of drugs over extended periods of time have recently been exploited to deliver antiretrovirals (ARV) [ 7 , 12 , 13 , 18 , 27 ]. The high costs associated with large-scale phase II/III clinical trials require us to fully evaluate the performance of the drug delivery systems in animal models for pharmacokinetics, safety and potential for prevention of mucosal immunodeficiency virus transmission. Here the nonhuman primate model has been the model of choice [ 14 , 17 , 19 , 24 ]. Although nonhuman primate efficacy studies against vaginal virus transmission using an IVR are limited [ 2 , 9 , 22 ] we recently demonstrated complete protection against 16 once-weekly vaginal exposures of SHIV162p3 in normally cycling female pigtailed macaques with a TDF reservoir IVR [ 23 ]. Because of sampling restrictions during the efficacy phase of the study, the study was extended for an additional 8 weeks after the last virus exposure to further define the pharmacokinetics and safety of the TDF IVR during 6 months of continuous use.

The animals were housed under biosafety level 2 containment conditions according to the Guide for the Care and Use of Laboratory Animals (National Research Council of the National Academies, 2010) at the Centers for Disease Control and Prevention (CDC). The CDC Animal Care and Use Committee approved all procedures outlined in the study. TDF reservoir IVR (130 ± 10 mg TDF and 20 ± 2 mg NaCl per IVR) were inserted into normal cycling pig-tailed macaques (n=6) at week 0. The TDF IVR macaques and 12 untreated controls were exposed once weekly to SHIV162p3 for 16 weeks [ 23 ]. The IVRs were replaced monthly, two days after virus exposure, at weeks 4, 8, 12, 16 and 20, with the final IVR removed at week 24. Vaginal secretions were obtained with Weck-Cel sponges during the challenge phase at the monthly IVR changes (weeks 4–16) and once weekly with a multiswab device for the follow-up period (weeks 19–24). Cervicovaginal lavages (CVL) of the genital tract were collected during the follow-up period. Vaginal biopsies were obtained at weeks 20, 21, and 23. TDF and TFV were quantitated by liquid chromatography-mass spectrometry in vaginal secretions and tissues [ 16 , 19 ]. Analysis of mucosal inflammation was done by the measurement of cytokines using a Milliplex™ MAP (Millipore) fluorescent multiplexed bead-based assay as previously described [ 14 , 21 ]. Change in cytokines and chemokines over time were examined using Friedman tests of log transformed values followed by Wilcoxon signed-rank tests with false discovery rate (FDR) adjusted p -values for post-hoc pairwise comparisons between each of 8 time points (wk 19 – 24.5) and wk −1 and wk 0, respectively.

The TDF-IVR macaques (n=6) remained both viral RNA negative and seronegative after 16 weekly SHIV exposures while 11/12 control macaques were infected after a median of four exposures [ 23 ]. Mucosal TDF concentrations were highly variable in the follow-up phase of the study (data not shown), most likely due to the instability of TDF in an aqueous environment, but were consistent with levels previously described [ 23 ]. High median mucosal concentrations of TFV, 1.8 × 10 5 (range, 1.1 × 10 4 – 6.6 × 10 5 ) ng/mL , were found in vaginal fluids collected at the monthly IVR exchange. During the follow-up phase, median TFV concentrations were 1.2 × 10 5 (7.5 × 10 0 – 2.1 × 10 6 ) ng/mL ( Fig 1B ). Although the distribution of drug in the vaginal vault was consistent (10 4 – 10 6 ng/mL), TFV levels at the upper end of the range were more frequently detected proximally than distally to the placement of the IVR during the challenge and the follow up phase. A CVL of the macaque genital tract with 5 mL of PBS is approximately a 1:10 dilution of vaginal secretions [ 6 ]. TFV concentrations in CVL ( Fig 1C ) appeared very stable during the follow up phase, 6.4 × 10 3 ng/mL (1.9× 10 2 – 1 × 10 5 ) and were found to be above the 1000 ng/mL that was effective in preventing HIV acquisition in women using a 1% TFV gel [ 15 ]. Although a modest decline in TFV concentrations in CVL was observed 3 days after ring removal, these remained above the 1000 ng/mL protective threshold ( Fig 1C ). The vaginal biopsy concentrations of TFV were also high with median of 1.3 × 10 4 (8.1× 10 2 – 9.0 × 10 4 ) ng/g tissue ( Figure 1D ). These levels are substantially higher than 113 ng/mL in vaginal tissue reported in the CAPRISA-004 trial. The tissue TFV concentrations and the cumulative in vivo release of 64 ± 11 mg in 28 days (determined from residual drug in IVR post- removal ) mirrored our previous observations [ 23 ]. The induction of mucosal inflammation due to IVR presence continuously for 6 months was monitored by measurement of 18 vaginal cytokines and chemokines ( Table 1 ). Although there were some differences in cytokines pre-ring placement compared to post-ring, when using FDR adjusted p - values these differences reflected variability between time points (wk 19–24.5), suggesting that the effects are not due to ring placement, and in another study we found that there was no difference between TDF and placebo rings in vaginal cytokine and chemokine concentrations (manuscript in preparation). These findings show that extended presence of the IVR does not alter the cytokine/chemokine patterns in the local mucosa, although other subclinical toxic effects due to the drug or the IVR cannot be deciphered by the above data [ 11 , 14 , 21 ] This is the first study to report the extended presence of an IVR in macaques and shows that it can safely deliver TDF mucosally at protective concentrations in this model. A phase 1 study was initiated in December, 2013 (28).