Next-Generation PrEP Trials: The Age of "Active Controls"

A medicine you get only every two months to reduce HIV risk sounds like a great deal for some people. This could be an option in the future, but only if two big efficacy trials of long-acting injectable cabotegravir (CAB-LA) find that one shot of this long-acting antiretroviral in the buttocks every eight weeks shields trial participants from HIV.

The first of two efficacy trials, HPTN 083, launched in December 2016. It tests injectable CAB-LA as PrEP among men who have sex with men (MSM) and transgender women. A companion study among women—HPTN 084—is due to start later this year. Both of these trials have a design that’s known as “double-dummy double-blind”. Each is designed to compare CAB-LA to daily oral PrEP. HPTN 083 is a “non-inferiority trial” and HPTN 084 is a “superiority” trial. Do these terms sound familiar? They might not. (See the centerspread for a lexicon and illustrations.) The advent of daily oral PrEP as a WHO-recommended prevention strategy has propelled changes in trials of other prevention strategies.


Why? While the comparison isn’t exact—and the history is controversial—consider research on prevention of vertical transmission of HIV. Once a lengthy regimen of AZT showed efficacy, another trial that sought to test a simpler strategy—single doses of nevirapine for the mother and newborn—came under intense scrutiny for a design that included a placebo arm, an arm with proven efficacy (the AZT regimen) and an arm with a regimen with unproven efficacy (short-course nevirapine). Many stakeholders felt that a placebo arm was unethical, and it was ultimately dropped. The trial did test an unproven and proven strategy head-to-head, even as some stakeholders also raised concerns about asking some women to use an unproven strategy when a proven one existed. The rationale was, in part, that less complex strategies were needed. The trial went on and ultimately found efficacy, providing an additional, valuable option for prevention of vertical transmission, as well as lessons about the difficulties of post-placebo trial design.

Similar issues are in play with PrEP today. Daily oral PrEP is effective when taken correctly and consistently. But additional options are needed, such as an injection. Both oral and injectable PrEP are designed to be used on their own. So a trial with “background” oral PrEP given to all participants isn’t a great option, as injectable-oral PrEP combos are not in the works.

HPTN 084, the planned trial of CAB-LA in African women, is an example of the complexities of trials in the post-placebo era. It was originally designed as an open-label study (see centerfold). In this design, some women would have been randomly assigned to receive daily oral PrEP, others to receive the injection. With this design, researchers hoped that women randomized to receive oral PrEP would use it more consistently than women did in some of the PrEP efficacy trials. In some of those trials, women’s adherence was quite low, perhaps because they did not know whether they were receiving an active product, or whether that product worked. (However, there is now evidence from PrEP projects that women can and will take daily oral PrEP consistently.) Each group would have known what they were receiving and been counseled accordingly.

Regulatory authorities raised concerns about the open-label design and the possibility that it would introduce bias into the research. When participants aren’t blinded and know what they are receiving, they may change their behaviors in ways that impact the validity of the results (e.g., if women receiving oral PrEP who understood that it was a proven tool increased their risk behaviors or women receiving the experimental injection increased condom usage). In both open-label and blinded trials, people are assigned to study arms by chance. The difference is that in blinded trials participants don’t find out what they are receiving. The argument for this design is that it offers a more fair comparison of two options versus unblinded trials.

Regulators’ concerns were discussed within the scientific community and in a community consultation that AVAC helped to organize with the HPTN 084 trial team. There was rich discussion in these meetings about the trial design. Ultimately, the HPTN 084 design changed to the double-dummy double-blind design (same as HPTN 083). The DISCOVER trial of a different drug being tested for oral PrEP, F/TAF, is also using this design. Time will tell whether future trials of PrEP strategies can utilize such a design.

It’s complex territory, and AVAC will continue to work with our partners in research and civil society to ensure that the trial designs are ethical, the goals well understood, and the outcomes on track to achieve the ultimate goal—a sustained end to epidemic levels of new HIV infections worldwide.