The HPTN 052 study

It was 2011 that finally saw publication of conclusive evidence of the efficacy of HIV treatment as prevention, from a randomised controlled trial that specifically aimed to measure its effect.  

The HPTN 052 trial was originally designed to report its findings in 2015 but was stopped in 2011 when it became obvious that the reduction in transmissions seen was so great that it would be unethical to continue recruiting study participants to the less protected arm.  

The HPTN 052 study found that the efficacy of treatment as prevention was 96% – in other words HIV-positive people taking ARVs were more than 20 times less likely to infect their partners than untreated people.

The study, conducted in 18 sites in eight countries in three continents across the world, began enrolling participants in 2005, recruiting 1736 serodiscordant couples. Most were heterosexual but there were 38 male/male couples (3%).

To enrol, the HIV-positive partner had to have a CD4 count between 350 and 550 cells/mm³ at baseline. The trial then randomised them so that 50% started taking ARV drugs immediately and 50% waited until their CD4 count had fallen below 250 cells/mm³, unless they developed an AIDS-defining illness first. (At the time the trial protocol was designed, WHO’s recommended CD4 threshold was 200 cells/mm³.  This was revised to 350 cells/mm³ in 2009).

When recruited, the average age of participants was 33, just over half (52%) of the HIV-positive participants were male, and their average baseline CD4 count was 438 cells/mm³.

The investigators stopped the trial in May 2011, four years before its planned ending date, when the trial’s Data and Safety Monitoring Board found that out of 28 HIV infections observed in the HIV-negative partners where the positive partner was clearly the source, only one occurred in a couple where the HIV-positive partner was taking ARVs. This equates to a 96% (over twentyfold) risk reduction.

There was also a health advantage to starting treatment early for the HIV-positive partner: there were only three cases of TB in the people who started ARVs immediately and 17 in those who delayed treatment: an 82% reduction in TB cases.

There was one case of transmission from an HIV-positive partner who had been taking ARVs. At the time of writing, we do not have viral load data for this individual, as detailed data from the study have not yet been published.

There were also eleven transmissions where phylogenetic analysis showed that the virus could not have come from the primary partner who was taking ARVs. This proportion of transmission from ‘extramarital’ partners (28%) was almost exactly the same as the proportion of extramarital transmissions seen in the Partners in Prevention study.  

The conclusive nature of the HPTN 052 result is leading to demands for a more rational, evidence-based approach to HIV prevention and treatment provision worldwide, with a greater emphasis on the provision of antiretrovirals (see later in this section).

To summarise: none of these studies have documented a proven case of heterosexual transmission where the HIV-positive partner had an undetectable viral load, though statistical uncertainty means that the possibility cannot be ruled out. They do document that providing ARV treatment to the positive partner in a heterosexual couple reduces the chances of HIV infection by 90% or more.