Adaptive Clinical Trial Designs

"It does not rain in Seattle.” That’s what Shiu-Lok Hu of the Washington National Primate Research Center said just a few days ago in New Orleans when he announced that the next Annual Symposium on Nonhuman Primate Models for AIDS will be in Seattle next year. 

At least today, that seemed to be true. It was not raining at the opening night here in Seattle of the Keystone Symposium on Immunological Mechanisms of Vaccination, held from October 27-November 1. The opening session was held as a joint session with the Grand Challenges in Global Health Meeting. Grand Challenges in Global health is a grant program funded by the Bill & Melinda Gates Foundation, and the Foundation’s head of global health, Tachi Yamada, was the first Keynote speaker tonight. 

“Our highest priority is vaccines,” Yamada told the attendees. “We believe vaccines to be the most cost effective tool for reducing health care costs.” At the same time, he added, “we really don’t know how to make vaccines in a predictable way. It’s still a little bit of black magic." 

Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, said in the second keynote address that there are three diseases that are “begging for immunological knowledge on our road to a vaccine: Influenza, HIV and Malaria.” 

As for HIV, he said the RV144 Thai trial showed the first signal of efficacy of an HIV vaccine. However, he said, the HIV vaccine field needs to get trial results more quickly than once every seven or eight years. “We don’t have a quarter of a century more to get a vaccine,” Fauci said. “We need to get it more quickly than that.”

One way to get trial results more quickly is an adaptive trial design, Fauci said. Such a design could involve several Phase II trials that are done at the same time. Then researchers look very early if there is any effect or signal in any of these trials. That allows them to then adapt the other trials accordingly. “You look very early,” Fauci said, “and if you see a signal [in one trial] and you see other [trials] that don’t look like they have any signal, you just take the one that has the signal and then adapt the other trials [by] either [combining] the controls or [using] the people in one trial to roll over to the other trial.” 

Currently, such adaptive trial designs are being discussed for follow-up trials to RV144, he said.