Universal Test and Treat Won't Stop
HIV Epidemic

Heather Boerner

July 17, 2019

Universal test-and-treat strategies resulted in "modest to no reductions" in new HIV transmissions in three large population-based studies, writes Salim Abdool-Karim, MD, ChB, PhD, of the Centre for the AIDS Programme of Research in South Africa, in a commentary published online today in the New England Journal of Medicine.

But that doesn't necessarily mean that expanding diagnoses and offering immediate access to HIV medicines is ineffective at stopping HIV transmission, said Myron Cohen, MD, of the University of North Carolina School of Medicine in Chapel Hill.

"We were never going to treat our way out of the epidemic," Cohen, who was not involved in the trials, told Medscape Medical News. "The take-home message is that this is the right thing at the right time, but it's not enough."

Cohen has been at the center of this conversation since 2011, when the study for which he was the lead investigator, HIV Prevention Trials Network 052 (HPTN-052), showed that treating a person living with HIV prevented that person from transmitting HIV to others.

By the time HPTN-052 reported final results in 2016, studies to test the concept that treatment prevents transmission were already underway. And health ministries in multiple countries were tracking the roll out of universal testing and treatment and its impact on the number of new infections. In 2017, the Kingdom of eSwatini (formerly Swaziland) announced that a drastic scale-up of treatment led to a 44% decrease in new infections, as reported by Medscape Medical News.

At the time, American Ambassador Deborah Birx, MD, told Medscape Medical News that the data is "real, documented evidence that you can translate science into programs."

Today, the New England Journal of Medicine published three studies that build on those data. The studies had been presented separately at the Conference on Retroviruses and Opportunistic Infections 2019 and the International AIDS Conference 2018.

"Debate ensued when mathematical models predicted that universal testing and treatment could achieve epidemic control within a few years in high burden settings," writes Abdool-Karim in today's commentary. Unfortunately, the models' predictions haven't panned out.

"In this issue of the Journal, three trials of universal testing and treatment show varying outcomes, from modest to no reduction in HIV incidence," Abdool-Karim writes.

Table. HIV Incidence Reduction by Universal Test and Treat (UTT) Trial



N =

% change, viral suppression

 % change, HIV incidence

Ya Tsie trial, Botswana

UTT  + home testing + engagement support + focus on men & <25 + VMMC




PopART-A (HPTN 071)

South Africa Zamia

UTT + home testing + CHW engagement support




PopART-B (HPTN 071) South Africa Zamia

Tx by guidelines + CHW engagement support 




SEARCH, Kenya Uganda

UTT + NCD Tx +engagement support + PCC




VMMC=voluntary medical male circumcision; NCD Tx=Non-communicable disease treatment; PCC=patient-centered care; CHW=community health worker; * Listed as "not significant"; ** at 24 months


Even study authors, such as the authors of PopART (HPTN 071), write that they were "surprised" that the more intensive treatment in one group of the trial resulted in almost no difference between controls and the intervention group. And even the studies that did show a substantial drop in HIV incidence, such as the Ya Tsie trial, found that it wasn't statistically significant by the end of the 3-year trial.

So what does all this mean? To Abdool-Salim, it means "programs need to consider going beyond universal testing and treatment to universal testing, treatment, and prophylaxis to achieve HIV epidemic control."

In addition, he points to four insights from the trials. One is that universal test and treat was effective in increasing viral suppression for people living with HIV. He adds that the lack of statistical significance could be related to what was happening in the control groups — and in the countries — while the trials were ongoing.

In the Ya Tsie trial, for instance, standard care at the start of the trial was to initiate antiretroviral treatment (ART) once CD4 counts dropped below 350 cells/mm3 or if viral load topped 10,000 copies/mL. But in 2015, the country began treating everyone — including those in the control group — when CD4 levels were higher, at 500 CD4 cells/mm3. By the end of the trial, universal test and treat was the law of the land. The same was true by the end of the PopART and SEARCH trials.

This, Cohen said, "raises the bar" for both viral suppression and reduction in HIV incidence in the trials. For the interventions to have had the impact that people expected, background HIV incidence would have needed to be relatively stable.

But "the standard of care is not stable," Cohen said. "It would be unethical and unconscionable to leave the standard of care alone [at 350 cells/mm3 or above] just because there are trials going on."

The second point Abdool-Karim makes is that this is real life. In randomized controlled trials like HPTN-052, they were able to prove that treatment prevented transmission because they could match the virus in the person living with HIV to the person who acquired it during the trial. If people acquired HIV from elsewhere, they were excluded from the calculations. Abdool-Karim points to a 2018 trial conducted in South Africa, where 35% of new HIV infections "were phylogenetically identified to emanate from outside the trial area."

So the new infections may not have come from people being treated in the trials.

Third, Abdool-Karim questions whether the testing and treatment in the trials captured people early enough in their infection to stop spread. Identifying and treating people with long-term infection, which is less virulent, may not stop spread in hot spots.

Finally, he asks whether the people who reach viral suppression in the trials were the people who were more likely to engage in behaviors that could transmit the virus. For instance, SEARCH specifically reached out to people under age 25 and men. But even still, consistently, men and young people were less likely to be virally suppressed than their older or female counterparts.

Cohen agreed.

"It's not that the people treated didn't need treatment," he said. "Unsurprisingly, many young people have a much higher rate of acute, early infection so their viral load is higher, [and] they're more sexually active."

The bottom line, Cohen said, is that the universal test-and-treat model is already out of the bag.

"We're not going to stop doing universal test and treat," he said. "So the question now is, how do we get more bang for our buck?"

The Ya Tsie trial and PopART trial were both funded by the National Institute of Allergy and Infectious Diseases (NIAID), the National Institute of Mental Health (NIMH), the National Institute of Drug Abuse (NIDA), the US President's Emergency Plan for AIDS Relief (PEPFAR), the International Initiative for Impact Evaluation, and the Bill and Melinda Gates Foundation. The SEARCH trial was funded by PEPFAR, and the National Institutes of Health. Cohen reports receiving advisory board honoraria or travel reimbursement from Janssen Global Services, Roche Molecular Systems, and Merck Research. Abdool-Karim has disclosed no relevant financial relationships.

NEJM. Published online July 17, 2019. Ya Tsie abstract, PopART abstract, SEARCH abstract, Editorial

For more news, follow Medscape on FacebookTwitter, Instagram, and YouTube

Follow Heather Boerner on Twitter @HeatherBoerner


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.