Adaptive clinical trials in drug development are moving towards broader acceptance and feasibility with the upcoming publication of an industry white paper and an FDA workshop on the subject.

Adaptive clinical trials have been touted by industry as a way to streamline the clinical development process, decrease time and length of trials, expose fewer patients to experimental treatments and provide richer data sets. However, use of such designs has been limited partly due to a lack of regulatory guidelines.

A Pharmaceutical Research & Manufacturers of America working group - established last year to "investigate and facilitate opportunities for wider acceptance and usage of adaptive designs" published an executive summary of their white paper in the Journal of Biopharmaceutical Statistics in May.

Wyeth Assistant VP-Clinical Development Michael Krams, co-chair of the PhRMA working group, noted at the recent Drug Information Association annual meeting that the task of addressing adaptive clinical trials initially appeared quite daunting, and the group wasted a whole day squabbling over the definition of adaptive designs.

"When the PhRMA working group first came together it was very clear that talking about adaptive designs was a little bit like going to a restaurant requesting carbohydrates. There's a lot that can been done with carbohydrates," he said.

In the executive summary, the working group cites three applications of adaptive clinical trials where it believes "it is feasible to overcome the challenges and implement these approaches more broadly and with maximal impact."

The first application is seamless Phase II/III trials, which hold the potential to significantly reduce study times, the number of patients necessary to reach the same quality of evidence and provide greater longer-term data.

Many companies are moving away from the three-phase paradigm to more fluid designs, such as the seamless Phase IIb/III model. Wyeth is adopting a "learn and confirm" paradigm for clinical development to focus on better understanding the medical value of a molecule up-front (¹ Also see "Wyeth Replacing Three-Phase Development With “Learn And Confirm” Model" - Pink Sheet, 3 July, 2006., p. 19).

For the seamless Phase IIb/III design to be implemented effectively, the design of the confirmatory phase of the trial - "apart from the selection decision to be made at the end of the learning phase" - needs to be envisioned at the beginning, the group advises.

Furthermore, the group recommends that "because these trials provide confirmatory data, it is critical that the process for data review, decision making and implementation be carefully specified and adhered to."

The PhRMA working group recommends sponsors "routinely" consider adaptive dose-finding trials for drug development and then seamlessly move them into a confirmatory trial where appropriate. Adaptive approaches allow the sponsor to "optimize dose assignments for future patients to best answer the research question of interest."

Additionally, "ineffective therapies could be discontinued earlier with more confidence, and the late-stage attrition rate, currently estimated at 45% in Phase III, could be reduced by facilitating the right dose or doses being taken forward in confirmatory trials."

A model-based approach should be used for such trials and sponsors should "prospectively define the dose assignment mechanism, regardless of the methodology used," to "minimize bias, enhance the integrity of the trial, and maintain the validity of inferences," the summary says.

Sample size re-estimation is recommended as a way to ensure that trials are neither too large nor too limited. The PhRMA group again advocates up-front planning; the protocol should specify the extent to which the sample size will be re-evaluated.

"Sample size re-estimation should never be a substitute for adequate up-front planning; rather, it is an acknowledgement of potential limitations of the information available at the time of trial design."

Although it is noted that "methods exist for sample size re-estimation based on updated information on treatment effects obtained at interim analyses," such methods need to be used with caution due to their potential to create operational biases."

"As an alternative, it should be strongly considered whether an appropriate group sequential scheme as a method of sample size determination could better meet the study objectives, as operational concerns are minimized and statistical behavior is often superior."

The summary addresses the logistical, procedural and statistical issues associated with adaptive clinical trial designs. It notes that rapid data capture, a logistical concern, is necessary to optimize on the efficiency of adaptive design.

The preservation of the Type I error rate is identified as a key statistical issue by the working group, which advises that it can be controlled when companies "specify the scope of possible adaptations and decisions prospectively."

The working group suggests that interim results be reviewed by a data monitoring committee that is adequately "firewalled" from personnel involved in the trial. However, the authors note that instances may arise when sponsor personnel need to be involved in the process due to the nature of adaptive trial designs.

In order to preserve the integrity of the trial, the group recommends that for each situation requiring sponsor representation, "a rationale or justification should be able to be described by which sponsor representation...is necessary."

Guidelines should also ensure that the sponsor be "adequately distanced from all trial activities" and that their "access to interim data is 'minimal' to meet the needs" of the sponsor's involvement.

The PhRMA group will release the complete white paper in November. It will be comprised of six articles, appearing in the Drug Information Journal and will cover the same topics as the executive summary, but in more detail. FDA and PhRMA have an adaptive clinical trial workshop planned for Nov. 13-14 to discuss issues related to adaptive trial design.

Working group co-chair Krams made a plea to industry to seize the opportunity to broaden use of adaptive clinical design. "If we are reckless in the way we approach this, if we come with crazy ideas to regulators and taint the notion of adaptive design with nonsensical proposals," then we will lose this opportunity, he said.

Developing criteria for use of innovative trial designs was one component of FDA's Critical Path opportunities list (² Also see "FDA To Provide Updates On Critical Path Opportunities List Projects" - Pink Sheet, 20 March, 2006., p. 5).

At the Conference on Adaptive Clinical Trial Design in Washington, D.C. on July 10, FDA Deputy Commissioner for Medical & Scientific Affairs Scott Gottlieb announced that the agency will issue five guidance documents to "help articulate the pathway for developing adaptive approaches in clinical trials."

Earlier this year, Gottlieb had said that the move towards more mechanism-based therapies may call for a transition from the traditional three-phase regime (³ Also see "The Next Phase: FDA Looks Beyond Three-Stage Drug Development Model" - Pink Sheet, 13 February, 2006., p. 10).

The first guidance document will discuss the use of multiple endpoints in the same trial and is expected to be discussed "as soon as January."

The other four guidance documents will cover enrichment designs, non-inferiority trial design, adaptive designs and missing clinical data.

Additionally, Gottlieb reported that FDA has established teams to provide consultation to divisions considering adaptive trial designs. "The good news is we are already seeing a lot of interest in adaptive approaches, especially in early-stage clinical trials," he said.

"While we don't have accumulated statistics, it is also clear that adaptive designs are becoming a more common element in special protocol assessments."

Gottlieb said that additional public meetings will help FDA develop the guidance documents. The deputy commissioner recently noted an increased desire at the agency to seek public input on emerging scientific issues (⁴ Also see "FDA Needs Bigger Window To The Public, Say Deputies" - Pink Sheet, 17 July, 2006., p. 22).

While Gottlieb acknowledged that FDA was to blame for some trepidation surrounding use of adaptive designs, another reason "is the marketplace itself," he told a Marwood Group panel discussion in Washington, D.C. July 12.

"Companies like the traditional three phases so that they have clear stopping points to make management decisions and they also have clear points at which to be able to speak publicly about interim results and an adaptive trial may change that whole paradigm," Gottlieb said.

"It will be interesting to see how the market reacts [to adaptive trials], because I think that's going to be a driver of whether a company is still confidant to pursue these types of things...and I think the marketplace has shown a willingness to reimburse for efficacy."

Adaptive Trials Going Mainstream With Industry White Paper, FDA Guidance

More from Archive

More from Pink Sheet