New PrEP Results Exceed HIV Researchers' High Expectations

Executive Editor

Two newly presented studies on HIV pre-exposure prophylaxis (PrEP) have demonstrated efficacy that surpassed the high expectations of many scientists. The findings, presented on Feb. 24 at CROI 2015 in Seattle, Washington, support assertions that PrEP use in "real world" scenarios has the power to dramatically reduce HIV infection even among very high-risk men who have sex with men (MSM).

Here are the headline grabbers from the first full unveiling of results from two studies known as PROUD and IPERGAY:

  • Both studies found that PrEP -- whether used daily or "on demand" -- reduced HIV risk by 86% among very sexually active MSM.
  • IPERGAY found that for every 18 French or Canadian MSM who took PrEP before and after sex, one HIV infection was averted.
  • PROUD fared even better: It found that for every 13 British MSM who took PrEP daily, one HIV infection was averted.
  • Both studies found no significant evidence that volunteers taking PrEP had begun having more, or riskier, sex as a result.

There's a good bit more to each of these studies than those highlights; for instance, PROUD and IPERGAY significantly differed in their construction and execution. In fact, those differences make the similarities between their findings all the more striking.

PROUD: Daily PrEP Use in a "Real" Setting

Sheena McCormack presenting data from the PROUD study at CROI 2015
Sheena McCormack presenting data from the PROUD study at CROI 2015

PROUD is a pilot study of 545 HIV-negative MSM age 18 or older who reported having unprotected anal sex within the past 90 days (or intent to do so within the next 90 days). They were recruited through sexual health clinics in a number of British cities. PROUD's intent is to offer a glimpse into how PrEP might fare in a "real-world" setting of gay men who are aware of their HIV risk and are willing to take a daily pill to potentially protect themselves.

The initial plan was to randomize the participants so that roughly half would immediately begin receiving daily PrEP as tenofovir/emtricitabine (Truvada), while the other half would not begin PrEP for 12 months (though all participants received follow-up safe-sex counseling and testing for HIV and other sexually transmitted infections [STIs]). However, interim results looked so good for PrEP that last October all 139 still in the "deferred" group were offered PrEP before their year of deferral was over.

The presentation by Sheena McCormack at CROI 2015 offered the clinical community its first solid glimpse at the data that led to that decision.

Overall, 22 HIV infections occurred during the study, three of which occurred in the immediate-PrEP arm and 19 of which occurred in the deferred-PrEP arm. (The number of infections in the deferred-therapy arm was roughly three times what the researchers expected based on prior clinic data, suggesting this was a very high-risk group of volunteers that had stepped forward to take part in the study.)

Compared to deferred PrEP, immediate PrEP appeared to reduce HIV incidence by 86% (P=0.0002).

The researchers calculated that the "rate difference" between the two groups was 7.6, which meant that the "number needed to treat" was 13 -- in other words, roughly 13 HIV-negative MSM needed to receive PrEP in order to avert one HIV infection, a number that McCormack noted was the "lowest reported to date."

There was no significant difference in non-HIV STI rates between volunteers in the immediate arm vs. the deferred arm. However, STIs of any kind were common in both groups: 57% in the immediate group and 50% in the deferred group. Rectal STI rates were very similar between the arms, at 35% for immediate-PrEP volunteers versus 32% for deferred-PrEP volunteers. (More STI screenings took place among the immediate group than the deferred group, McCormack noted, suggesting that may at least in part account for the greater number of diagnoses among those who began PrEP immediately.)

In the end, "our concerns that PrEP would be less effective in the real world were unfounded," McCormack said. Referring to preliminary survey data on the potential effect of PrEP use on volunteers' sexual behaviors, she added, "Gay men incorporated PrEP into their existing risk-reduction strategies, and these continued to include condom use."

McCormack said that cost-effectiveness analyses were underway to determine the economic value of PrEP rollout in this population.

IPERGAY: Similar Value for "On Demand" PrEP

Jean-Michel Molina presenting data from the IPERGAY study at CROI 2015
Jean-Michel Molina presenting data from the IPERGAY study at CROI 2015

At nearly the same time last year that the PROUD study's deferred-PrEP arm was being dismantled, another major PrEP study announced that it, too, would eliminate its non-PrEP arm. That study was IPERGAY, a randomized, placebo-controlled, double-blinded trial designed to assess the efficacy of a unique strategy of "on demand" PrEP specifically among MSM. (The strategy effectively involves a person taking a total of four PrEP doses in the 48 hours around a sexual act.)

An interim review of the study data led the study's monitoring board to decide that the PrEP-receiving arm was so clearly superior that the study should be unblinded, and all volunteers currently receiving a placebo should be offered tenofovir/emtricitabine.

At the time it was unblinded, IPERGAY had enrolled 414 volunteers, all of whom were HIV-negative MSM in France and Canada who reported having anal sex without a condom with at least two partners in the past six months.

Here at CROI, we got our first look at some of the numbers that factored into the decision to unblind the study. Jean-Michel Molina presented the results.

Over a mean follow-up period of 13 months, 16 HIV infections occurred -- two in the PrEP arm (or 0.9 infections per 100 person-years) versus 14 in the placebo arm (or 6.6 infections per 100 person-years). Similar to the PROUD study, IPERGAY researchers were surprised by the high rate of infection in the non-PrEP arm, which suggested an unusually high amount of risk behavior among the volunteer group.

The difference amounted to an 86% reduction in HIV risk for PrEP users compared to placebo users -- an identical reduction rate to that observed in the PROUD study, and far beyond the 50% target the researchers had designed the study to find.

The "number needed to treat" for one year to avert one HIV infection was 18 people -- slightly higher than the 13 found by the PROUD researchers, but still quite low.

Due to the unique, "on demand" nature of the dosing, IPERGAY assumed that adherence rates among its participants would be all over the map -- and they were. On average, both PrEP and placebo users took roughly 16 doses of their study drug per month.

Based on an analysis of self-reports, less than half of the PrEP-using participants took their pills correctly for every sexual encounter, as per the study protocol; 29% took them suboptimally and 28% did not take them at all. Yet risk reduction still reached 86%. Molina suggested this was because "participants were able to select the type of sexual partners with whom they would want to use PrEP," and felt that overall PrEP adherence was "good."

Over the course of the study, Molina noted that markers of sexual behavior remained consistent among both groups, suggesting no evidence of risk compensation among PrEP takers. For instance, roughly 70% of volunteers reported recent unprotected anal sex at baseline, a percentage that was virtually unchanged 24 weeks into the study. Median number of sexual acts (10) and partners (8) also did not appear to change significantly.

Myles Helfand is the editorial director of and

Follow Myles on Twitter: @MylesatTheBody.