Data safety monitoring boards (DSMBs) have played a big role in helping the most effective and safe HIV therapies advance.

In 2020, lost amidst the news of the coronavirus vaccine trials, there were some amazing news in HIV prevention — thanks to DSMB work, says Mitchell Warren, executive director of AVAC, a global HIV prevention organization in New York.

A pre-exposure prophylaxis (PrEP) trial, with a bimonthly injectable intervention, had excellent efficacy data. So a DSMB recommended stopping randomization in the large trials, Warren explains.

The injectable was incredibly effective, and the DSMB recommended that everyone be told that it worked better than the PrEP pill to which it was compared, he adds.

“So, they were told both products worked, and they had to choose which one to use, but the trial didn’t stop because there still was a lot of safety data the board wanted,” Warren says.

The interim results, reviewed by the DSMB in May found that the PrEP regimen with long-acting cabotegravir (CAB LA) injected once every eight weeks was superior to daily oral tenofovir/emtricitabine (TDF/FTC) for HIV prevention.1

Then, on Nov. 9, a PrEP regimen containing cabotegravir injected every eight weeks in women from southern and east Africa also worked well at preventing HIV infection. The injectable PrEP had superior efficacy to Truvada at preventing HIV. The DSMB found no safety concerns and recommended that the National Institute of Allergy and Infectious Diseases stop the blinded phase of the trial, which originally was set to continue until 2022, and share the results.2

From the perspective of people at risk of HIV infection, this early clinical trial news means there soon might be a drug that can be injected just every other month and it will be more effective than taking a daily PrEP pill.

“Even though we got excited about the news of the women’s trial, it came out on Nov. 9 — the same day Pfizer put out its first data about the coronavirus vaccine,” Warren says. “If that had not happened, this would have been front-page news.”

DSMBs sometimes have to stop trials when the study product does not work well. They also might stop trials of interventions that work better than anticipated.

For example, the circumcision trials, performed in South Africa, Kenya, and Uganda about 15 years ago, showed high levels of safety and efficacy. Men who were circumcised as part of the clinical trial had significantly fewer HIV infections. The interim research showed a protective rate of at least 60% among circumcised male participants who had sex with women.3

The DSMB found that the circumcision intervention was ethical and safe, and the DSMB wanted to look at interim data more frequently, Warren says.

“Then, nine months later, the DSMB saw data from both trials and said it is time to stop the trial,” he explains. “The DSMB said it was unethical to continue the trials because these two trials showed that circumcision is a safe and effective tool.”

In 2007, the World Health Organization began to recommend circumcision programs at least two years before the trials had been scheduled to end, Warren says.4

By 2018, more than 18 million men in priority countries had been medically circumcised as part of the voluntary medical male circumcision program.5

These trials and the DSMB’s intervention before they ended demonstrate the importance of having independent boards review interim results, Warren notes.

“When you are looking for efficacy, you want very solid data,” he adds. “When it comes to harm, even some slight serious harm is enough to stop a trial.”

There also have been HIV intervention trials in which the studies were stopped because the DSMB found in checking interim data that the active product arm had an infection rate similar to placebo. The DSMB said it was not ethical, nor a good use of resources, to continue the trials as there was no hope the product would show benefit, Warren explains.

A Ghana microbicide trial, called SAVVY, had drastically lower HIV incidence rates than what statisticians had expected when the trial was designed, showing that participants were benefiting from prevention messages and services. The trial would be unable to answer the question of whether the microbicide worked, and the DSMB recommended the trial be stopped.6

DSMBs sometimes halt trials because of potential harm to participants. For instance, two clinical trials in Africa of a vaginal microbicide, Ushercell, were halted in 2007 because an interim data review from one of the trials suggested a possible increased risk of HIV infection among female participants.6

“HIV has been at the front end of a lot of DSMB innovations,” Warren says. “With HIV in the 1980s, we began to see an engagement of communities pushing for involvement in research and designing protocols.”

Having an advocate like Warren on a DSMB was pioneered by the HIV community.

“Developing trust in the research process and research enterprise is really challenging,” Warren says.

Including advocates on a DSMB is a way to involve the affected community in research and build trust and confidence, he adds.

“Those pauses by DSMBs give us confidence that there is someone looking at safety, not just researchers moving ahead,” Warren says. “DSMBs monitor safety of people in trials, and they also help people gain confidence in the research enterprise, and that’s critically important today — more than ever.”


  1. HIV Prevention Trials Network. HPTN 083 study demonstrates superiority of cabotegravir for the prevention of HIV. July 7, 2020.
  2. National Institutes of Health. NIH study finds long-acting injectable drug prevents HIV acquisition in cisgender women. Nov. 9, 2020.
  3. Johnson KE, Quinn TC. Update on male circumcision: Prevention success and challenges ahead. Curr Infect Dis Rep 2008;10:243-251.
  4. World Health Organization. Statement on Kenyan and Ugandan trial findings regarding male circumcision and HIV. Dec. 13, 2006.
  5. Avert. Voluntary medical male circumcision for HIV prevention. April 23, 2020.
  6. Global Advocacy for HIV Prevention. Fact sheet: Data safety monitoring boards.