In December 2007, Elizabeth Mataka, the United Nations Special Envoy for HIV/AIDS in Africa, highlighted one of many stark realities of the HIV epidemic: “61% of all those living with HIV in Africa are women,” she wrote. “And 26 years into the epidemic, we know that underpinning this terrible statistic is gender inequality.”
Condoms work. But in light of the tremendous and persistent burden of HIV infection, researchers have worked to develop alternative approaches to reduce HIV transmission, particularly methods that might allow women increased options, and with that, more control over their own health. Whether or not pre-exposure prophylaxis (PrEP)– a daily anti-viral regimen to prevent HIV infection – is an important tool in the HIV-prevention armamentarium has been hotly debated, both in the scientific literature and in the popular press. Prior trials have yielded equivocal results, with the 2012 FEM-PrEP trial failing to show efficacy in reducing HIV acquisition among women.
To better evaluate PrEP’s potential for women, researchers designed the VOICE trial, published in NEJM this week, to assess the efficacy of 3 different PrEP regimens. 5,000 sexually-active women 18 to 45 years of age from South Africa, Uganda, and Zimbabwe were randomized to one of 5 treatment arms: oral tenofovir disoproxil fumarate (TDF), oral TDF plus emtricitabine (TDF-FTC), oral placebo, tenofovir (TFV) vaginal gel, or placebo vaginal gel.
Study participants were counseled to use study products daily and were provided HIV risk-reduction counseling, condoms, and hepatitis B immunizations. The mean age of participants was 25.3 years; 21% of women were unmarried, and 71% were using an injectable hormonal contraception. HIV-1 testing was performed monthly and plasma tenofovir levels checked quarterly.
The results were disappointing: during the trial period, 312 incident HIV-1 infections occurred, with no difference in the rates of seroconversions between treatment arms. 52 occurred in the TDF arm (HR 1.49), 61 in TDF-FTC (HR 1.04), 61 in TFV gel (HR 0.85), 60 in oral placebo, and 70 in gel placebo. In fact, the Data Safety Monitoring Board stopped both the oral TDF and TFV gel arms early for futility.
Importantly, adherence to study drugs was low, and the discordance between participant self-reporting and measured plasma tenofovir levels is striking: in interviews, participants across treatment arms reported mean adherence of 90%. However, in a random sample, tenofovir was detected on average in no more than 30% of quarterly plasma samples in any treatment arm. This finding underscores the importance of objective markers of adherence in clinical trials.
In terms of adverse events: more serum creatinine elevations were seen in participants randomized to oral TDF-FTC relative to oral placebo (1.3% vs. 0.2%; P=0.004). There were no significant differences in other adverse events between treatment arms.
NEJM Deputy Editor Dr. Lindsey Baden describes where we go from here: “Developing effective, deployable HIV prevention strategies remains a high global priority. These strategies must account for human behavior and limit potential toxicity in otherwise healthy persons to be successful.”
So, is PrEP a promising strategy to reduce rates of HIV-1 infection among women in sub-Saharan Africa? Not as demonstrated in this trial. Perhaps the low adherence tells the whole story, but for now, this study adds to the disappointing data regarding PrEP and the effort to prevent HIV in women.