By Deborah Borfitz
March 24, 2008 | The vast majority of adaptive clinical trials (ACTs) that have been done to date appear to involve sample size re-estimation, says Jay Herson, PhD, senior associate in biostatistics at Johns Hopkins University. The “adaptive” methodology, which Herson helped develop in the mid-1990s, requires neither Bayesian analysis nor special accommodating software. It is well understood and no longer incites controversy within the industry or red flags at the FDA.
Regulatory acceptance of a seamlessly combined Phase IIB/III trial may be quite another matter, says Herson. “The big problem is that companies aren’t getting the dose right even when they do many Phase II trials in the traditional fashion. Now they want to learn a lot in one trial and there’s a lot of room for error and a lot of room for the FDA to not approve a drug because [of recent product recalls and public safety concerns].”
At Wyeth, a champion of the adaptive “Learn and Confirm” model of drug development, six adaptive dose-ranging studies involving “dynamic termination rules” were started in 2007 and another eight are expected to begin this year, according to Michael Krams, MD, assistant vice president, adaptive trials, clinical development. “If we decide not to go ahead with an adaptive design, it’s mostly an internal [rather than an FDA-imposed] decision.”
, meanwhile, is operating more than a dozen ACTs across all clinical trial phases, reports Trevor Mundel, MD, PhD, global director of the company’s immunology and infectious disease business franchise.
Interest in ACTs is currently focused on Phase III confirmatory trials using methods other than group sequential analysis, sample size re-estimation, and dose ranging, says Herson. But companies are only now beginning to approach the FDA with proposals for late-phase ACTs using novel ACT methodologies. Excluding Pfizer’s oft-mentioned Phase II stroke study, the medical literature is devoid of reports on large, pivotal pharmaceutical ACTs and likely will be for at least another two years.
According to Crystal Rice, spokesperson for the FDA’s Center for Drug Evaluation and Research, the agency doesn’t have “exact numbers” on the adaptively designed protocols it has evaluated to date. “We consider interim looks, step-down analyses of endpoints, re-estimation of sample sizes, and such to be adaptive – although they’re not the controversial ones.” She adds that “most all of the medical areas are receiving some protocols that have some aspect of the novel adaptive design associated with them.”
These novel designs are currently being handled on a “case-by-case basis,” giving the agency “experience and definitions that will be useful for later advice,” says Rice. “The guidance we are working on, when out for comment, should provide a better structure to deal with these designs.”
The kinds of adaptive methodologies being talked about today may not even be suitable for 21st-century drugs based on nanotechnology, genomics, and tailored, targeted treatment modalities that are now under development, notes Herson. “How far behind will the FDA be then [with its guidance]?”
ACTs are “just a component of a broad array” of innovative statistical methodologies currently being piloted, adds Mundel. The list includes a model that considers a drug’s random effects and data analysis based on recurrent patterns, such as binge drinking in alcoholics or seizures in epileptics.
Adaptive Clinical Trials: More Talk Than Action
Ten Years After: Learn and Confirm
Biting the Adaptive Trials Bullet
Change Agents for Adaptive Design
This story first appeared in eCliniqua,one of Bio-IT World’s free e-newsletters. Subscribe here.