Subscribe Now
Forgot Password?

Weekly News, Analysis, and Commentary

Legal & Regulatory Feature

Photo creation by RRY Publications, LLC

Computer-Based Modeling, Pathway to Faster FDA Approvals

Walter Eisner • Thu, April 13th, 2017

Print this article

If the U.S. Senate gives its blessing, the FDA is about to get a new leader in Scott Gottlieb, M.D. Gottlieb has been an advocate of speeding the approval process through shorter trials.

On April 6, 2017, perhaps sensing the changing tide, the medical device scientists at the FDA, led by Jeff Shuren, M.D. and Janet Woodcock, M.D., wrote an article published in the New England Journal of Medicine, entitled, "An FDA Viewpoint on Unique Considerations for Medical-Device Clinical Trials." In the article the scientists show how the agency is currently looking at faster ways to evaluate and get medical devices to patients.

To be fair, agency leaders have been talking about these issues for a few years now.

Computer-Based Modeling

The scientists write that in the future computer-based modeling may change the way the agency thinks about device validation "in other ways, allowing for much smaller clinical trials, or may change the way we think about running trials, in that some 'clinical' information may be derived from simulations."

They note that device manufacturers are increasingly developing stochastic (randomly determined) engineering models that may have the capability to simulate clinical outcomes for “virtual patients” by modeling a relationship between bench outcomes and clinical end points. "If it can be shown that these virtual patients are similar, in a precisely defined way, to real patients, future trials may be able to rely partially on virtual-patient information, thus lessening the burden of enrolling additional real patients."

This is already happening with The Medical Device Innovation Consortium, a public–private partnership among industry, nonprofit organizations, and federal agencies (including the FDA). The Consortium is exploring ways to facilitate how computer modeling and simulation may become a validated and accepted part of a clinical trial. "For devices for which some uncertainty remains after such premarketing studies, postapproval registries may be a helpful tool to provide additional confirmation of device performance," add the scientists.

Registries and Electronic Health Records Can Eliminate Control Groups

The scientists also point to registry studies, which they believe have the potential to play a critical role in device-approval decisions as a source of evidence from clinical experience ("real-world" clinical data).

"High-quality registries and electronic health records hold enormous potential to serve as efficient tools to answer premarketing and postmarketing questions.”

For example, registry data have been used as historical controls and have also allowed the FDA to reduce or replace postapproval study requirements.

Registry data may also serve as the basis for statistically derived objective performance criteria, write the scientists. Future devices may be compared with those criteria to show acceptable clinical performance, thereby eliminating the need for a control group.

“In the longer term, we expect to see randomized studies embedded within registries, leveraging that existing infrastructure and minimizing the burden for study start-up and execution. For example, a next-generation unapproved device that requires a clinical trial to support approval could be studied under a protocol designed to have its data requirements aligned with an existing registry that gathers high-quality outcomes data for approved devices of that type. Investigational sites already participating in the registry could leverage their experience and existing processes to easily participate in the premarketing trial. Such an approach could simplify data gathering, auditing, and analysis by relying on methods already in place for the registry.”

Scott Gottlieb has not arrived yet at the FDA. But apparently, his handwriting is already on the wall.

Send to a Friend

The article link will be sent to the email address you provide

Your Name (required)

Your Email (required)

Friend's Email (required)


Leave a Reply


Email Address (will not be published)