vMICROb: Long-Acting Progestin Contraception and the Vaginal Microbiome
Study Details
Study Description
Brief Summary
Despite many years of research, controversy persists as to whether hormonal contraception promotes HIV acquisition. A number of observational studies on depot medroxyprogesterone acetate (DMPA) injection showed an increase in HIV risk and no evidence of increased risk with oral contraceptive pills. There are no human studies currently published on the impact of the levonorgestrel intrauterine device (LNG IUD) on HIV transmission risk and minimal data on the effects of the etonogestrel subdermal implant (ESI) on risk of HIV acquisition. Establishing whether any of these highly effective contraceptives increases the risk of HIV infection would have far-reaching public health implications, particularly in areas of high HIV prevalence such as sub-Saharan Africa, where injectable contraception accounts for nearly half of contraceptive use.
Perturbations in the normal vaginal microbiota, or community of microorganisms inhabiting the vaginal body niche, have long been known to affect the risk of transmission of HIV. Studies have shown altered vaginal microbiota with DMPA injection and preserved vaginal microbiota with the LNG IUD, but no studies have compared these methods head-to-head or used culture-independent sequencing methodology. The investigators propose a prospective pilot study to evaluate the impact of different long-acting progestin contraceptive formulations on the vaginal microbiome. Specifically, the investigators aim to identify and compare metagenomics profiles associated with DMPA, LNG IUD, and ESI contraceptive use by community analysis of vaginal swab samples from women collected longitudinally after contraceptive method initiation. The investigators hypothesize that DMPA will increase community diversity in the vaginal microbiota, whereas the LNG IUD and ESI will not affect the balance of microorganisms in the vagina. Women who are planning to initiate DMPA, LNG IUD, and ESI contraception as well as controls not seeking contraception will be recruited for the study from Boston Medical Center (BMC), a tertiary care center with a racially and socioeconomically diverse patient population. Women will have longitudinal follow-up with self-sampling of the vagina for sexually transmitted infection testing and metagenomics analysis at method initiation, 2-3 months, and 6 months. Establishing the safest long-acting progestin contraceptive alternative will promote effective contraception use and lower rates of HIV acquisition worldwide.
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
|
Detailed Description
Women between the ages of 18-45 initiating long-acting progestin contraception will be enrolled at a routine office visit at the family planning clinics of each institution. Enrollment will begin in April 2016 and will stop after 6 months to allow for longitudinal follow-up and analyses. For this pilot study, a total of 30 long-acting progestin contraceptive initiators with 10 subjects in each method group (DMPA, LNG IUD, ESI) and 5 age-matched, controls seeking tubal sterilization will be selected for longitudinal STI screening and vaginal sampling for microbiome analysis.
After consent is obtained, trained staff will interview subjects using a brief sociodemographic/habits questionnaire that includes questions on age, race, body mass index, smoking status, history of sexually transmitted infections and yeast vaginitis, weekly frequency of coitus, and number of lifetime sexual partners. Subjects will be instructed to self-sample the vagina at three timepoints: 1) at method initiation, 2) 2-3 months later, and 3) 6 months later. Self-collected vaginal swab specimens will be assayed for Trichomonas vaginalis, Chlamydia trachomatis, and Neisseria gonorrhoeae using DNA amplification in the clinical lab. Microbiome samples collected at each timepoint will be analyzed and compared to the initiation sample to detect longitudinal differences.
Study Design
Arms and Interventions
| Arm | Intervention/Treatment |
|---|---|
| Depo Provera Ten women ages 18-40 who elect to start a using Depo Provera for contraception. |
|
| Mirena IUD Ten women ages 18-40 who elect to start a using a Mirena IUD for contraception. |
|
| Nexplanon Ten women ages 18-40 who elect to start a using Nexplanon for contraception. |
|
| Control: Tubal Sterilization Five women ages 18-40 who elect to have a tubal sterilization. |
Outcome Measures
Primary Outcome Measures
- Biomarker discovery [1 year]
Use of the following two approaches will provide preliminary data for future larger studies from this limited pilot study. If the hypothesis of a vaginal microbial shift with Depo Provera (DMPA) use is confirmed (our PRIMARY analysis), these tools will enable visualization of the differences. The tables of relative abundance of either taxonomic groups or metabolic pathways will be used to identify differential bacterial biomarkers present among patient and sample classes. Two bioinformatics tools particularly useful for biomarker discovery are LEfSe and MaAsLin (Multivariate Association with Linear Models, http://huttenhower.sph.harvard.edu/maaslin).
Secondary Outcome Measures
- Correlation of disease state and metadata factors with microbiome structure [1 Year]
Effects of different sources of variation (e.g. smoking status, presence of STI) on community structure will be assessed in MANOVA models, and for OTUs of interest, ANOVA. Analyses will use either frequencies of OTUs or PCA values from major axes. Effects of various sources of variation on community diversity will be analyzed by ANOVA on diversity indices, obtained from QIIME.
Other Outcome Measures
- Site based approach to the analysis of the impact of disease status and metadata [1 Year]
The Dirichlet-multinomial distribution allows the analyst to perform tests of hypotheses (e.g., compare microbiomes across groups) and to estimate parameters describing microbiome properties. The Dirichlet-multinomial distribution prevents Type I error inflation by taking into account the overdispersion in count data in the microbiome. This model allows comparison of microbiome populations between more than two groups of subjects (i.e. Mirena IUD users, DMPA users, Nexplanon implant users). This test is analogous to an analysis-of-variance test in classical statistics. We will access biostatisticians experienced with new technology-derived datasets at Case Western and the Lifespan/Tufts/Brown/BU CFAR Biostatistical Core.
Eligibility Criteria
Criteria
Inclusion criteria
-
Women between the ages of 18-40 who elect to start a long-acting progestin contraceptive or who opt for tubal sterilization.
-
Our site will enroll 18 patients total: 5 DMPA; 5 Mirena IUD, 5 Nexplanon, 2-3 tubal ligations/Essures (controls)
Exclusion criteria
-
Non-English-speaking without translator
-
Current or recent (within past 3 months) use of hormonal contraception
-
Currently menstruating
-
Vaginal intercourse within 48 hours of visit
-
Known or suspected pregnancy, or pregnancy within the past 6 weeks.
-
Use of progestin method for primary indication other than contraception (e.g. pelvic pain, menorrhagia)
-
Current STI or vaginitis (yeast or BV)
-
Tampon usage
-
Regular douching
-
Chronic antibiotic use within past 4 weeks
-
HIV positive
-
Immunosuppressive therapy (organ transplant, chemotherapy)
Contacts and Locations
Locations
| Site | City | State | Country | Postal Code | |
|---|---|---|---|---|---|
| 1 | Boston Medical Center | Boston | Massachusetts | United States | 02118 |
| 2 | Women and Infants Hospital | Providence | Rhode Island | United States | 02905 |
Sponsors and Collaborators
- Boston Medical Center
Investigators
- Principal Investigator: Wendy Kuohung, MD, Boston University
Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- H-34707