Adaptive trials are very much on the minds of the clinical trial community, with a few big sponsors like Novartis and Merck plunging into the systems necessary to support them. Many other companies are wondering about just how much infrastructure (technologically) or process change (organizationally) will be necessary to implement adaptive designs.

At last month’s annual Regulatory Affairs Professionals (RAPS) conference, there was a session on the regulatory nuances of adaptive approaches. The FDA’s Stephen Wilson, deputy director, division of biometrics, was on the RAPS agenda. Wilson told a few jokes and gave a short history lesson about the FDA prior to user fees. Appropriately, Wilson declined to declare any new policy from the podium.

Wilson’s Wisdom

Right about then, just as we were pondering lunch, Wilson began saying a few things that caused us keep our laptop open. Industry is knocking on the FDA’s door to show the agency its adaptive game plans, he says, and FDA appreciates that. “We are having a lot of discussions internally on adaptive designs,” says WIlson. “We are having companies come in and not discuss products. That is very helpful to us.”

He encouraged the sponsor community to use electronic, not paper, techniques. “Make it all electronic, please,” Wilson said. “The Gateway is open. You don’t have to use Federal Express.”

Advance Work

Within FDA, Wilson said, there is an ongoing dialogue about how to guide industry on statistical analysis plans, a relatively new element of what the agency inspects prior to a trial. “We’re trying to figure out how to communicate about the statistical analysis plan. With adaptive designs, it becomes even more important.”

In the recent past, he said, biostatisticians could look at the data at the end of the trial in a special way, hum a certain incantation, and all would be right with the numbers. That era, Wilson believes, is ending. “We’re getting into an environment where nobody knows how those studies are coming out,” Wilson says. “Statisticians cannot fix it at the end. You have to design it well up front.”

Critical Path

To laughter in the audience, Wilson said he’s frequently asked how many patients are needed if the budget is $2 million. “That is not a realistic way to plan a trial,” he said.

Referring to the Critical Path Agenda, which as far as we can tell has been mostly a rhetorical push, Wilson assured the audience that drafts of the promised official FDA guidance documents would be emerging on schedule—as in next year and 2009. Said Wilson: “These guidances are going to be coming down the line in the year they’re supposed to. As regulatory professionals, you have to read this 5 year plan. It may change what you do.”

You can put “adaptive” into the search box of our site, or check out this story about United Biosource’s views on the topic or our own list of questions for adaptive studies.