When preventing HIV, bacteria in the vagina matter

Bacteria in the vagina affect whether a drug stops an HIV infection or is itself stopped cold.

A vaginal gel containing tenofovir, an antiretroviral drug used to treat HIV infection, was three times as effective at preventing HIV in women who had healthy vaginal bacterial communities as it was in women with a less beneficial mix. The finding may help explain why the effectiveness of these gels has varied in trials, researchers report in the June 2 Science.

“The vaginal microbiota is yet another variable that we have to take into account when we are thinking about why one intervention does or doesn’t work,” says clinical scientist Khalil Ghanem of Johns Hopkins University School of Medicine, who coauthored a commentary accompanying the study.

For women, one strategy to prevent HIV infection is to apply medicated vaginal gels before and after sex. But results have been mixed regarding how well the gels work. The hit-or-miss effectiveness can partly be explained by some patients not taking the medication as prescribed. But study coauthor Adam Burgener, a microbiologist at the Public Health Agency of Canada in Winnipeg, wondered if there might also be a biological explanation.

The main residents of a healthy vaginal microbial community, or microbiota, are Lactobacillus species. The bacteria produce lactic acid, making the vaginal tract more acidic and possibly “less hospitable for potential pathogenic organisms,” Ghanem says.

To examine the effect of the vaginal microbiota on tenofovir, Burgener and colleagues turned to a previous trial of South African women, which showed that the drug reduced HIV infections by 39 percent. During that trial, samples of vaginal mucus were taken. In the new study, the researchers measured bacterial proteins in 688 of those samples to determine the bacteria in the women’s vaginas when the samples were collected.

Just over 400 women’s vaginal microbiota mainly had Lactobacillus species; the microbiota of the other 281 women were dominated by non-Lactobacillus species, such as Gardnerella vaginalis. Within those two groups were women who had used tenofovir vaginal gel and those who had used a non-medicated gel as a placebo.

In the Lactobacillus-dominant group, the incidence of HIV was 61 percent lower in women using the medicated gel compared with those using the placebo gel. But in the non-Lactobacillus dominant group, it was only 18 percent lower. There was no appreciable difference in the consistency of the gel’s reported use between the two groups, the researchers note.

“Women with Lactobacillus had three times more protection offered by the gel,” Burgener says. “That’s a pretty remarkable difference in the efficacy of a drug.”

Looking at a random subset of 270 of the samples, the researchers found that the vaginal gel drug levels were lower in the mucus from the non-Lactobacillus group. So, in a test tube, they mixed a laboratory strain of G. vaginalis with tenofovir. After four hours, the amount of tenofovir in the tube had decreased by 50 percent. In a similar experiment with two Lactobacillus species, the amount of the drug remained about the same. It appears that the G. vaginalis bacteria “gobbled up the drug and depleted it,” Burgener says.

It’s known that microbes in the gut can impact the metabolism of medications, says clinical scientist and commentary coauthor Susan Tuddenham of the Johns Hopkins University School of Medicine. “This study tells us that when we are thinking about vaginally delivered medications, we may need to think about the vaginal microbiome as well.”

The work also shows that women closely following directions for vaginal medications “could be doing everything right and still not getting the full benefit of that medication,” Ghanem says.