A combined oral contraceptive affects mucosal SHIV susceptibility factors in a pigtail macaque (Macaca nemestrina) model

Abstract

Background: Injectable hormonal contraception may increase women's risk of HIV acquisition and can affect biological risk factors in animal models of HIV. We established, for the first time, a model to investigate whether combined oral contraceptives (COC) alter SHIV susceptibility in macaques.

Methods: Seven pigtail macaques were administered a monophasic levonorgestrel (LNG)/ethinyl estradiol (EE) COC at 33% or 66% of the human dose for 60 days. Menstrual cycling, vaginal epithelial thickness, and other SHIV susceptibility factors were monitored for a mean of 18 weeks.

Results: Mean vaginal epithelial thicknesses were 290.8 μm at baseline and 186.2 μm during COC (P = 0.0141, Mann-Whitney U-test). Vaginal pH decreased from 8.5 during treatment to 6.5 post-treatment (0.0176 two-tailed t-test). Measured microflora was unchanged.

Conclusions: COC caused thinning of the vaginal epithelium and vaginal pH changes, which may increase SHIV susceptibility. 0.033 mg LNG + .0066 mg EE appeared effective in suppressing ovulation.

Keywords: SHIV; contraception; non-human primate (NHP); susceptibility; transmission risk.

Figure 1. Menstrual cycling following COC use

For 96 days prior to baseline sampling, animals were observed for menstrual bleeding and their perineal sex skin swellings were graded. The sex skin swellings were graded on a 1-4 scale, with 1 being no swelling up to 4 being the most swollen. The sex skin swellings are denoted as the columns on the x-axis. Menstrual bleeding is recorded as black dots. Baseline samples included progesterone levels and vaginal biopsies. Biopsies are shown as black triangles and progesterone levels are the solid black line. Progesterone levels were continued for four weeks after COC’s were discontinued. 2-3 biopsies were taken as baselines, three were taken during COC administration and three were taken after COC’s were discontinued. Study day zero is the first day of COC administration. COC’s were given for at least 62 days and no longer than 74 days. The first and last doses are shown on the time line as a dotted line extending from the x-axis. The weights for the animals were: BB173 (7.18kg), BB537 (6.79 kg), BB4800 (5.59kg), BB552 (7.21 kg), BB 269 (6.65 kg), BB4870 (7.54 kg), and BB967 (5.56 kg).

Figure 2. Drug Levels

LNG and EE were measured one week prior to COC, during COC administration and one week after withdrawal. The low dose group is denoted with black dots and the high dose group is black squares. Ethinyl Estradiol levels in both groups were not detectable. Limit of detection was 0.1 (ng/ml) for LNG and (0.006 ng/ml) for EE. No synthetic hormones were detected in the baseline samples. The median amount of LNG measured for the low dose and high dose groups were 0.3763ng/ml LNG + 0.5919ng/ml respectively.

Figure 3. Vaginal epithelial changes following COC administration

Hematoxylin and eosin stained vaginal biopsies from BB967 during follicular (day 8), luteal (day 29), and first and second medicated phases. The follicular phase biopsy shows an overall thicker vaginal epithelial thickness, with a markedly thickened non-nucleated layer compared to the luteal phase biopsy. Both biosies taken during COC administration show a nucleated layer that is thinner than both follicular and luteal phase specimens, and complete absence of a non-nucleated layer.

Figure 4. Quantitative evaluation of vaginal epithelial thickness changes following COC use

Mean epithelial thickness measurements for each animal in the low dose (A) and high dose (B) groups and summary of the mean measurements for both groups (C). The low dose group showed more consistent thinning of the vaginal epithelium during drug administration, with almost complete elimination of the non-nucleated layer, as compared to the high dose group.

Figure 5. Vaginal pH and H₂0₂+ producing microflora following COC use

A. Vaginal pH The vaginal pH was taken on each animal weekly during the all three phases of the study. The median pH of each animal is shown as black dots. They are displayed separately in high and low dose groups. B. Peroxide producing bacteria Colonies of lactobacilli and viridans streptococci H₂O₂ producing bacteria are shown from baseline samples and both cycles of COC administration. The lactobacilli are shown as black dots and the viridans streptococci are shown as black triangles. Since the axis is logarithmic, only values greater than zero can be plotted. For this graph, five values were zero or negative, so are not included on the graph.