The HPTN 052 results were headline news at IAS 2011. Principal Investigator Myron Cohen, MD, talked with the IAS about the years of work that went into the trial, the research team’s emotions upon first learning the results, and the weeks that followed. Dr. Cohen is Associate Vice Chancellor for Global Health and Director of the Institute for Global Health & Infectious Diseases at the University of North Carolina at Chapel Hill.
By Scott Sanders
As Myron Cohen and his colleagues finished presenting the results of the HPTN 052 trial to a packed room at IAS 2011, the audience rose to its feet in a standing ovation. It was a rare occurrence at a scientific meeting, reflecting the significance of the results and the tremendous efforts behind them.
The emotions of Cohen and his team members that July afternoon in Rome were a far cry from those they had felt less than three months before. Heading back to the Washington, DC, airport after a regularly scheduled meeting with the study’s Data and Safety Monitoring Board (DSMB) on 28 April 2011, the researchers feared the study might be about to come to an end. The research team was completely blinded to the study results data and some of the unusual actions of the DSMB members in the meeting that day had them fearing the worst.
The weight of their fears reflected the hopes that they and so many others had put into the success of HPTN 052, a large Phase III randomized trial examining the effectiveness of two antiretroviral treatment strategies in preventing sexual transmission of HIV. The study had enrolled over 1,750 HIV-serodiscordant couples at 13 sites around the world, representing a huge investment of time, resources and effort by them, their entire research team and the study subjects.
The 20-Year Lead Up to the Results
While the study began enrolling subjects four years earlier, the road to the fateful DSMB meeting began more than 20 years earlier, even before the advent of antiretroviral therapy. In the late 1980s, Cohen’s research group at the University of North Carolina at Chapel Hill was studying the transmission of sexually transmitted diseases and had come to believe that concentrations of virus, including HIV, in genital secretions would predict transmission. Once AZT was available, they began to study its impact on the concentrations of HIV in genital secretions. Over the next decade, Cohen’s group and many others studied this concept and by 2000, according to Cohen, “we thought that the biological plausibility was strong enough that we should try and prove that people taking antiretrovirals (ARVs) are rendered less contagious. That was the genesis of HPTN 052.”
The road from concept to launch took another seven years. In 2000, the team received permission from the trial’s sponsor, the National Institutes of Health (NIH), to design the study, but it wasn’t until 2002 to 2003 that they received the go-ahead. Next they had to obtain donations of study drugs for use during the entire study. The team went to many different drug companies and had to secure the drugs from each independently. “It was millions of dollars worth of drugs, so each drug we secured for the trial seemed to us like a huge triumph. Each company that agreed to come on board was critical to allowing us to move forward.”
The study started in a successful pilot with approximately 90 couples in 2005, with enrollment into the full trial beginning in earnest in 2007. “It was really a 20-year effort. It wasn’t like I woke up one day and said we will do a study to determine whether treatment will prevent transmission. It was a very long and arduous effort.”
Cohen and his team persisted over the years despite the skepticism of many. “After we got started people asked, ‘Why are you bothering to do this? Everyone knows it works.’ But I would argue that we should not be doing a clinical trial without that criticism, because there needs to be enough biological plausibility to justify all the effort. We should not just be guessing what’s going to happen.”
A Fateful Meeting
The April 2011 meeting was the team’s eleventh meeting with the DSMB, and they had no reason to believe it would be any different than previous meetings. At the meeting, the DSMB had a recommendation, as they usually did, which up to that point had been to continue the study. This time, they said they had a recommendation, but they could not tell the team what it was. Instead, the DSMB needed to tell the trial sponsor, NIH, which would then decide whether they were going to accept or reject the recommendation.
“Then they told us to leave. So we got in a car to the airport. One of the people who had been working on the trial from the beginning was understandably upset because she thought that the study had been stopped or it was going to be recommended by the DSMB that it should be stopped because of futility. I was thinking, ‘Ok, we did the very best job we could possibly do.’ We were mostly just dumbstruck.”
On their way to the airport, a senior NIH official called Cohen on his mobile phone and asked the team to return to the NIH. By that time, key players from the NIH had gathered in a room. The research team went in and sat down and initially eye contact with those in the room was minimal. It had been a similar lack of eye contact at the DSMB meeting that led the team to believe that the trial might be ending.
The most senior NIH person in the room said, “Congratulations. You knocked it out of the park.” The DSMB’s recommendation to the researchers was brief and twofold: One was congratulating them for a well-conducted study, and the other was requesting the team make the results public as soon as possible, given the magnitude of the prevention outcome.
Then they handed the team copies of the 400-page report of the study data that they had never seen. “We went in another room and all of us poured through the data. People were really euphoric. It was a pretty big shock having gone from believing that the study might have failed to seeing that we had virtually abrogated HIV transmission with ART. It was really a very emotional moment. I think everybody in the room was a bit tearful; we felt like we had made a real contribution.”
Going Public
There was no rest after the excitement. The team had a call the next day to determine how to go forward. They decided to announce the results at a press conference on 12 May 2011 and concluded immediately that they would use available drugs to offer all the HIV-infected subjects in the delayed arm the opportunity to start therapy. To do so, they needed the permission of all the industry partners to use the drugs outside of the original study design. They all agreed. Each of the 13 study sites then had a deadline to contact their IRBs, their governments and organizations, and their study subjects to tell them what was going on before the results were released to the public.
“Obviously, it was really exciting to have the first call with the sites. Nothing was more fun than to tell the investigators about their success.” The sites had a script of what to say to their study subjects, making clear that they would be offered ART.
On 12 May, the team was in for more shocks. “I really had no idea how much news this would generate because I had been through all these years with people telling me that ‘everybody already knows it’. So then we had the press conference and seeing the tremendous interest in this trial worldwide was another shock to my system. When I saw that The Economist had a cover story about this, that was the biggest shock of all.”
Each of the sites also became a mini hotbed of publicity. All became a source of pride to the research community in those countries and to the government. Every site had a series of articles on the front page of all their newspapers.
Presenting the Science
While the results had been secret between 28 April and 12 May, Cohen had started working on how to present the results to his scientific colleagues. He received permission from the NIH to talk with the IAS about presenting the results at IAS 2011 and with the New England Journal of Medicine (NEJM) about publishing them. The wheels were in motion and between 12 May and July, the worldwide research team undertook a remarkable effort to prepare for IAS 2011 and write the journal article.
Having not seen any data until 28 April, the team quickly developed and submitted four late breaker abstracts to IAS 2011 for consideration; all scored highly and were accepted for presentation. In addition to preparing the presentations, the team wrote the manuscript for the NEJM, securing an agreement to publish it at the hour of the presentation at the IAS meeting. According to Cohen, a team of people from all over the world was getting up at 3:00 in the morning seven days a week and working 18 hours a day to meet the deadlines.
Recognizing the significance of the results, the IAS 2011 leadership worked to ensure that the research community was able to take full advantage of the opportunity presented by the conference. Upon learning of the results, Prof. Quarraisha Abdool Karim, co-chair of the conference’s prevention science track and a co-chair of HPTN, thought it was important that the data be shared in the next public forum of peers–IAS 2011– and that it be published simultaneously in a peer-reviewed publication as soon as possible. “It is remarkable that Cohen was able to achieve it in record time. It was no easy feat.”
“It was important not to just have a presentation in Rome. We also needed sufficient time during the conference to provide as much detail as possible. People needed to look at all facets of the data and the nuances, and we needed to give peers the time to review and discuss the information,” said Abdool Karim. “We anticipated that other symposia at the conference would be thinking about the results in terms of treatment access and prevention.” Ultimately, the planning committee devoted a full session to the trial results on the first day of the conference, unusual for late breaker abstracts, which are traditionally presented on the final day of the conference. (The session’s content was expanded a few days before the conference to also include the results of CDC TDF2 and the Partners PrEP Study).
Cohen and his team were determined to take full advantage of the opportunity, which would be the first time any of their peers would see the data. They scrupulously reviewed and confirmed the genetic data linking the new infections to partners participating in the trial, worked hard to prepare thorough and clear presentations and rehearsed “a million times”. “The science drives the reaction. We felt it was important that we present unassailable science. Any little aberration would be disastrous.” Many in audience were furiously taking notes during the presentations and when Cohen completed the final presentation, the audience was on its feet with applause – a reaction he attributed to the momentum the results gave to the treatment-as-prevention paradigm, and the audience’s recognition of both the duration of the effort and the team’s commitment to presenting the data clearly.
“We gave proof of a concept that we had worked on – and believed in– for a very long time and the magnitude and the care with which this was done helps blow a strong wind behind the concept.” Cohen also notes there is still much research to be done to prove that this strategy can work under real-world conditions. “I have not changed my opinion one bit about test-and-treat. This is the floor for test-and-treat. I like the strategy of treatment as prevention very much, but if this [trial] had not worked the strategy would probably be impossible. Even with these results, it does not prove the strategy can work. There are a lot of other variables that will help define whether the strategy works or doesn’t work, not just the 052 success.”
Looking Forward
Many clinical trials built on this science and aimed at proving it is possible to reduce HIV incidence with ART are now getting underway. They will focus on the feasibility of finding infected people and putting them in treatment. “How important is acute infection?” asks Cohen. “If acute infection is very important then we are going to see failure because we can’t find the people with acute infection, at least not easily. So, linking people to care, making sure they take their pills, dealing with acute infection, those are all things that will allow treatment as prevention to succeed or fail.”
HPTN 052 is also ongoing, and the hope is to monitor the study subjects until 2015 to assess the durability of the intervention. The trial is no longer randomizing subjects to different trial arms and the majority of delayed arm subjects, but not all are choosing to go on therapy now. The initial results showed that the magnitude of effectiveness is near 100%, if not 100%, but examining the durability of intervention allows the team to continue following the subjects for many years.
Online Resources
The abstracts and audio presentations with slides from the session at IAS 2011 are available here.
Videos from the press conference’s preceding the session is available here.