The Vaginal Microbiome Matters: A Wake-Up Call for PrEP Research

Tenofovir is one of the more powerful HIV medications on the market today. So powerful, in fact, that it comprises part of Truvada (tenofovir/FTC), the only U.S. Food and Drug Administration-approved pre-exposure prophylaxis (PrEP), an HIV prevention tool that has grown increasingly popular in the United States.

In Africa, where more than one million women are newly infected with HIV every year, doctors and scientists are looking for new formulations of tenofovir that are just as effective as a pill. To date, a gel containing tenofovir has been seen as one of the more promising new formulations.

Yet, prior studies of tenofovir gel in African women have had disappointing results. In one study, called FACTS, a tenofovir-containing gel didn't appear to impact a woman's chance of contracting HIV at all.

Why would the gel work well in some trials but not in others? Recently, microbiologist Adam Burgener, Ph.D., and his colleagues hypothesized that the discrepancy might have something to do with the microbiome: the invisible colonies of bacteria that live on our bodies and interact with our immune systems.

"The female genital tract is the first point of contact for HIV when a woman has sexual intercourse," Burgener, head of the Proteomics Group, National HIV and Retrovirology Labs, Public Health Agency of Canada, told "Lots of studies over the past years have showed that ... a woman's risk of infection can depend on the bacterial composition of the vaginal microbiome," he said.

To explore this hypothesis, Burgener and his colleagues took a closer look at a previously published study called CAPRISA 004, which found tenofovir gel reduced HIV infections by 39% on average. The trial sponsor, the Centre for the AIDS Program of Research in South Africa, had preserved vaginal bacteria samples from women taking part in the study.

When Burgener looked at the samples from 688 women, he found that more than half (59.2%) had a vaginal microbiome dominated by a type of "good" bacteria called Lactobacillus, which is known to be beneficial for vaginal health. The rest of the women (40.8%), had vaginal microbiomes comprised predominantly of Gardnerella vaginalis, a type of bacteria that is correlated with negative reproductive health outcomes.

"In women who had Lactobacillus, the tenofovir worked really well," said Burgener. "It was 78% protective in women who used it properly, which means that women who were taking the gel with the drug were at 78% lower risk of becoming infected with HIV."

However, the drug only provided protection to 26% of the women with microbiomes dominated by Gardnerella vaginalis -- even among the group of women who were most compliant with the medication directions.

The stark difference was also evident among women who did not adhere to medication, with a protective rate of 17% and 4% for women with Lactobacillus and Gardnerella vaginalis, respectively.

As it turns out, the medical community dramatically underestimated the power of the vaginal microbiome -- a biological factor that could make or break the effectiveness of PrEP and vaccine research efforts.

For women with Lactobacillus, tenofovir gel provided "high level of protection ... comparable to the effectiveness of Truvada" in some studies, said Burgener. But for women with Gardnerella vaginalis, it was almost as if "the drug just didn't work," he said.

After Burgener identified the association between Gardnerella vaginalis and a lack of effect, he took the experiment one step further. He sent samples to a laboratory to identify precisely how and why the Gardnerella vaginalis bacteria was preventing tenofovir from working.

"We saw that, in the Gardnerella vaginalis samples, the drug was depleted 50% within four hours. My jaw dropped," he said. These tests showed that anaerobic bacteria such as Gardnerella vaginalis are able to metabolize the drug directly. In other words, Gardnerella vaginalis didn't just stop tenofovir from working; it actively gobbled it up.

"It's one of those discoveries that in hindsight seems obvious, but going into it, I did not expect metabolism to be mechanism," said Burgener. "When we first presented the results, there was a lot of shock, but bacteria can do really strange things. They can eat oil; they can eat rocks. They can degrade anything in nature. The fact that bac in our bodies can consume [synthetic] drugs is not all that surprising."

Burgener hopes his findings will have an immediate impact on ongoing prevention research. Lactobacillus, the "good" bacteria, creates an acidic environment in which other bacteria, viruses and yeast are less likely to thrive. In theory, health workers could administer a relatively inexpensive pH test to screen for women who are most likely to benefit from tenofovir-containing gel, Burgener explained.

Broadly, however, the recent paper, published in Science in June, highlights why more research is needed to fully understand the relationship between the vaginal microbiome, the immune system and effective HIV prevention medications.

For example, among women who have Lactobacillus-dominant vaginal microbiomes, a condition called bacterial vaginosis (BV) can flare up when the population of Lactobacillus is repressed. In 1995, researchers compared the rates of HIV among sex workers in Thailand and found that those with BV were four times as likely to contract HIV. More recently, in studies of women from Zimbabwe and Uganda, researchers discovered that BV or pre-BV might account for 29% of new HIV infections.

In general, Burgener hopes his paper will be a "wake up call to the research community that we need to do more research on the microbiome," he said, adding that the HIV prevention field has largely neglected the biology of the female genital tract.

"Now that the microbiome is showing to be so important for infection risk and showing to interact with HIV drugs, we need to do more work on studying how the microbiome affects the effectiveness of drugs," Burgener said.