Draft Guidance Helps Designers Modify Ongoing Clinical Trials

Draft Guidance Helps Designers Modify Ongoing Clinical Trials

Sponsors considering whether to use an adaptive clinical trial design should select a number of realistic scenarios and determine how likely each is to succeed with a particular adaptive design, the FDA says.

This exercise should be conducted as the sponsor plans a pivotal study, according to draft guidance released May 15 on adaptive designs for medical device clinical trials.

The guidance is intended to make it easier for sponsors to adjust ongoing trials in the face of new information or unexpected outcomes without undermining the trial’s integrity or validity. Adaptive designs can also enhance researchers’ understanding of a product’s benefits and risks and ease the transition from premarket investigation to postmarket follow-up, the FDA says.

The idea has gained traction in recent years as a way to reduce trial costs and increase the odds of a study’s success.

Typically, modifications should be identified before a trial begins and described in the protocol. But modifications made during a trial may be scientifically valid if designers don’t know the outcomes, FDA says. The guidance applies to all trial stages and all device applications.

Sponsors that choose to use adaptive designs need to focus on two key goals — controlling the chance of erroneous conclusions and minimizing operational bias, the FDA says. This means understanding the chances of false positive and false negative conclusions and the effect of multiplicity on results — e.g., analysis of multiple endpoints or subgroups.

To avoid the risk of false results, the FDA recommends sponsors use analytical methods and simulation studies that make it easier to spot errors. To minimize bias, it is critical that sponsors and investigators don’t have access to unblinded results during an ongoing trial, the agency adds.

The guidance describes 11 types of study adaptations. Among these are:

  • Group sequential design, in which an interim analysis of treatment outcomes is done so the study can be stopped for success or futility, if necessary;
  • Sample size adaptation. While adding more patients to improve the results of a failed trial is not scientifically valid, adaptive design techniques can be used to allow for an extension that minimizes the chance of errors, the FDA says;
  • Dropped treatment arm based on poor effectiveness;
  • Revised randomization ratio. This can improve study efficiency and protect patients by directing more subjects to a treatment that has been found to have better outcomes;
  • Unplanned changes to the device or endpoint, which are common during early feasibility studies.

The FDA recommends against the use of adaptive designs in trials that enroll patients so quickly that all the results are obtained at about the same time, giving designers little chance to make changes based on interim data. The agency also discourages them in complex studies that have multiple primary or secondary endpoints.

Sponsors submitting an adaptive trial design should explain the proposed adaptation, including what, when, how and why the adaptation will be performed.

Comments are due Aug. 17 to docket no. FDA-2015-D-1439. View the guidance at www.fdanews.com/05-18-15-adaptive.pdf. — Elizabeth Orr

$745 $545

Bonus

Subscribe to Clinical Trials Advisor NOW and save $200 off the regular one-year price of $645 — plus receive a FREE copy of our webinar, Reporting Failed Trial Data: New Rules for ClinicalTrials.gov – a $347 value!

Key Benefits

LINKS TO KEY DOCUMENTS — Need to read regulatory references, guidances, enforcement actions, and other documents? Simply click on the source documents in your electronic issue.

ONLINE ACCESS — Consider our newsletter archive your personal library! Search your current issue — and hundreds of past issues — by keyword and relevancy.