WASHINGTON — FDA Commissioner Scott Gottlieb, MD, described new efforts to leverage real-world evidence in clinical trials during a briefing at the Bipartisan Policy Center.
At the Monday briefing, Gottlieb also announced a pilot program in partnership with the country’s top academic centers designed to strengthen and maintain an expert workforce.
Gottlieb noted that public input would be critical to harnessing real-world data. The FDA plans to convene a stakeholder meeting sometime later this year to assess how digital systems, including “remotely monitoring data trends, accrual, and integrity over the course of a trial,” can be used for oversight in the “accrual” and “integrity” of new data in clinical trials, he said.
On-site monitoring accounts for about one-third of the costs of clinical trials, and still does not guarantee quality, he said. In some cases, remote and risk-based monitoring, which is less expensive can actually provide better oversight, Gottlieb added.
He also said he wants to use digital technology to bring clinical trials to the patient, through “decentralized trials,” as they are known, which leverage resources such as electronic health records and wearable devices. These trials will reach more diverse populations, better reflect real-world use of products, and allow providers and patients to make more informed treatment decisions, he said.
Gottlieb noted that the agency has convened a working group and is developing guidance around these types of trials.
The agency is also focused on helping reviewers better understand the impact of labeling changes on prescribers’ decisions and patient outcomes. For example, the FDA’s Information Exchange and Data Transformation is exploring the impact of labeling changes to two immune checkpoint inhibitors for cancer treatment from weight-based to flat dosing in community practices.
Finally, the FDA is piloting a 2-year “competitive fellowship” program focused on artificial intelligence (AI) and machine learning. Post-doctoral fellows from top universities will be invited to work alongside mentors from the FDA’s medical product center to design “high-impact AI-based regulatory science tools,” Gottlieb said.
Following Gottlieb’s remarks, former FDA commissioners explored the benefits and challenges of transforming data into evidence.
Andrew von Eschenbach, MD, president of Samaritan Health Initiatives, likened these new digital systems in health to those that keep planes from crashing into one another at night.
He noted that there are sensors in every plane that continuously log and collect data about how a plane functions. That same type of continuous monitoring could impact healthcare, he noted.
“In both real-time and retrospectively, we’re going to be able to prevent problems … to save lives,” said von Eschenbach, who is also a former director of the National Cancer Institute.
Robert Califf, MD, PhD, a professor of medicine and cardiology at the Duke School of Medicine in Durham, North Carolina, described the potential for data on a more personal level — Califf’s mother is in her 90s, has multiple myeloma, and is currently in hospice.
“There must be 200,000 people just like her. … If we had all the information compiled [on those patients], you would know which decisions are right and which are wrong, but because of our current views or privacy and the way we deal with it, it keeps that information from being useful for my mom, or whomever is in that situation,” he said.
Califf said he wasn’t advocating for privacy to be totally ignored, but “we need a set of principles to guide it, not exactly like the ones we have now.”
Mark McClellan, MD, PhD, founding director of the Duke-Margolis Center for Health Policy, cautioned that when it comes to “turning data into evidence” the health system is “a long way from getting there.”
He pointed out that noticing a relationship between a treatment and an outcome doesn’t mean there’s causation. McClellan, who is also a former administrator for the Centers for Medicare & Medicaid Services, added that integrating randomization into studies involving real-world evidence is critical, and that methods used to analyze such data must be sound.
Von Eschenbach said the FDA is best suited to offer ethical guidance on the use of real-world data because “it’s the honest broker” and has access to data on all FDA-approved drugs and therapies.
Califf disagreed: “The FDA deals with medical products. It doesn’t deal with healthcare. … We need both.”