Effects of hormonal contraceptives on the female genital tract microbiota in South African adolescents

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Background Young women in sub-Saharan Africa are disproportionally affected by HIV and often rely on injectable hormonal contraception (HC) to prevent unintended pregnancies. However, HC might affect HIV-1 risk through changes in the female genital tract (FGT) microbiota. We examined the impact of three different HC methods on the adolescent female genital tract microbiota and related cytokine and HIV target cell levels at the cervical mucosa in a randomized, crossover trial. Study design and methods 131 adolescent females aged 15 to 19 from Cape Town were enrolled into a randomized, crossover study. The participants were randomized into three study arms: 1. progestin-only injectable norethisterone enanthate (Net-En), 2. combined oral contraceptive pills (COCPs) or 3. combined contraceptive vaginal ring (CCVR) for 16 weeks. Participants then switched to one of the other HC options for a final four months. Vaginal samples were collected at baseline, crossover and exit. STI testing and Nugent scoring were performed at all study visits. Vaginal microbiota was characterized by 16S rRNA gene amplicon sequencing, cytokine concentrations were measured by Luminex and CD4+ T cells analysed by flow cytometry. Results Using fuzzy clustering, three major female genital tract bacterial community types were identified. Two of these were dominated by Lactobacillus species (L. crispatus and L. iners, respectively) and the third was comprised of a diverse group of anaerobic bacteria associated with bacterial vaginosis (BV). In an intention-to-treat analysis at crossover, participants randomized to COCP had a significantly less diverse vaginal microbiota compared to participants randomized to either Net-En or CCVR. The same was observed in an according to protocol analysis at crossover. Using differential abundance testing and random forest analyses, we found that species associated with BV and risk of HIV were significantly more abundant in, and predictive of, participants on Net-En (e.g. Prevotella, Sneathia and Dialister) or CCVR (e.g. Prevotella, Mycoplasma and Parvimonas) compared to COCP while L. iners was more common in the COCP group. Cytokine concentrations were positively associated with a diverse vaginal community and with specific bacterial taxa associated with BV and increased risk of HIV including species enriched in participants on Net-En and NuvaRing. In contrast, there were no association of the frequencies of CD4+ T cells expressing CCR5+ with the vaginal community or BV status. There was likewise no significant association with BV or diversity with Th17 cell frequency, yet BVassociated bacteria were more abundant in participants with higher frequencies of Th17 cells. Conclusions Our data generated from a randomized study suggests that COCPs use may exert a positive influence on genital health through an increase in lactobacilli and a decrease in BV-associated bacterial taxa with an accompanying decrease in overall bacterial diversity, vaginal pH and cytokine levels. In contrast, the vaginal microbiota of participants on Net-En and NuvaRing have increased levels of bacteria associated with BV and HIV risk and increased cytokine levels. We did not observe any association of the frequencies of CD4+ T cells expressing CCR5 or Th17-like cells with the vaginal community, BV status or HC use.