Results from a major five-year trial in South Africa show that rolling-out universal test and treat policy without innovation to better link people to treatment is unlikely to reduce the number of new HIV infections.

The Treatment as Prevention (TasP) trail was run by the Africa Health Research Institute (AHRI) in partnership with the French Agency for Research on AIDS and Viral Hepatitis (ANRS) in the Hlabisa area of northern KwaZulu-Natal between March 2012 and April 2016. The trial offered repeat home testing and immediate antiretroviral treatment in trial clinics for those who tested positive. The results, published in scientific journal The Lancet HIV on 1 December, show while there is good take-up of the offer of HIV screening at home – very few of those diagnosed with HIV sought treatment. Because their entry into the healthcare system was infrequent and slow, HIV transmission in the population was not reduced.

The results are crucial for governments to consider as they implement or scale up their ‘treat all’ programmes. Key AHRI research done in rural KZN populations previously has demonstrated that roll-out of antiretroviral treatment (ART) has significantly reduced death. Several other studies done in this population and elsewhere have shown that ART also has the capacity to reduce transmission to other people. However, the strategy of widespread treatment to reduce transmission at the population level had not before been tested in a formal trial setting. This is one of four international randomised trials which seek to assess the effectiveness of the universal test and treat strategy in reducing HIV transmission, and the first to report its results.

Trial design

The trial area was divided into 22 geographic zones, or clusters, each of about 1 280 people aged 16 or over. The clusters were randomly split into two equal groups: an intervention group and a control group. 13 239 people were included in the intervention arm and 14 916 in the control arm. Every six months, trial participants were offered rapid HIV screening in their homes. In the intervention group, people who tested HIV positive were offered immediate ART, whatever their CD4 count. In the control group, treatment was initiated according to the previous indications recommended by South Africa’s Department of Health (originally a CD4 count of less than 350, and from January 2015 a CD4 count of less than 500). Mobile treatment clinics were sent to each area to facilitate access to care for people who tested positive. They could also use local health services at any time.

Findings:

  1. Screening at home is well accepted. HIV status was identified at least once for 88% of the people contacted.
  1. There was no significant population-level impact of universal ART on the HIV infection rate. The annual incidence of HIV infection was an estimated 2.13% (2.13 individuals infected out of 100 in a year) in the intervention group and 2.27% in the control group, figures that are not statistically different.
  2. Linkage to HIV care was both slow and poor, leading to a lower than anticipated increase of population ART coverage. The rate of linkage-to-care was similar in both arms, with only 30% of individuals registering at the trial clinic within six months of home HIV diagnosis, considerably lower than the expected 70%.
  3. The researchers estimate that only 4 out of 10 HIV positive people living in the region had an undetectable viral load. Yet UNAIDS estimates that to meet its targets designed to end the epidemic by 2030, 7 out of 10 HIV positive people must have an undetectable viral load by 2020 (90-90-90 target).
  4. Despite the lack of impact on transmission the trial provides further evidence regarding the individual benefits of universal ART, with high viral suppression observed at 12 months. There was a higher retention rate in the intervention than the control arm, with participants on ART more likely to be retained in care than those pre-ART. These are strong arguments to roll out universal test and treat without any restriction.

Despite the offer of treatment after an HIV positive diagnosis, there was clearly a reluctance on the part of many to be treated immediately. Researchers said that stigma could still play a large role in a participant’s decision not to link to care. They also commented on the increasing mobility among rural South African populations, meaning that many people enrolled in the trial wouldn’t have been in the area to take advantage of treatment.

“Despite being a negative finding, this is a very important finding in terms of HIV policy. AHRI has previously demonstrated the potential benefit of ‘treat all’, but this trial has now shown that implementing the policy is not enough. Our next steps are to investigate if initiating treatment at home might increase people’s entry into care. We are also looking at potential solutions including financial incentives and mobile and technology options, some of which could be gender specific,” said AHRI Director and UCL Professor Deenan Pillay, who co-led the study.

“We are proud to have co-led the TASP trial with our South African colleagues with a multidisciplinary approach. This is the first ever of a series of community based trials to release its final findings. We should not be distracted to promote the universal test and treat strategy, and look for more efficient solutions to reach this goal of test and treat all as quickly as possible,” said ANRS Director, Professor François Dabis.

*ANRS 12249 TasP was coordinated by Professor François Dabis (ANRS Director), Professor Marie-Louise Newell (University of Southampton, United Kingdom) and Professor Deenan Pillay (AHRI Director, and University College London Professor). The trial was performed in partnership with the KwaZulu-Natal Department of Health, South Africa. Funding was provided by ANRS (France REcherche Nord&sud Sida-hiv Hépatites), GTZ (German Agency for Technical Cooperation), the Bill & Melinda Gates Foundation, the International Initiative for Impact Evaluation (3ie), and the Wellcome Trust (United Kingdom). The trial was conducted with support from Merck & Co. Inc, and Gilead Sciences, which provided Atripla®.

Top image: Blood is taken for a rapid HIV test. Photo: Ben Gilbert, Wellcome.