US FDA Piloting Use of AI for ‘Real-Time’ Clinical Trials

The U.S. Food and Drug Administration will test out real-time clinical trials using artificial intelligence tools and data science. The goal is to accelerate the development of promising new drug therapies, which the agency said are slowed by data and procedural bottlenecks.

See Also: Reduce Cloud Risk in Healthcare with Security by Default

The agency said Wednesday a real-time clinical trial has already involved “successful initiation” of two proof-of-concepts – one by pharmaceutical giant AstraZeneca and the other by Amgen.

AstraZeneca is conducting a phase 2 multi-site trial called Traverse involving patients with mantle cell lymphoma who haven’t yet received treatment. That trial involves participation from the University of Texas MD Anderson Cancer Center and University of Pennsylvania.

Amgen is conducting a Phase 1b trial called Stream-SCLC with patients with limited-stage small cell lung carcinoma. Final site selection is in process.

The FDA met with sponsors of both trials over the establishment of criteria for reporting key data signals – such as efficacy outcomes and safety concerns, the agency said.

The agency plans to expand these two proofs-of-concept into a broader pilot program. In a request for information published Thursday, the agency sought views on how AI-enabled technologies “can improve efficiency, speed and quality of decision-making in early phase clinical trials.” The RFI seeks input on potential pilot program design and implementation, as well as evaluation metrics and success criteria.

“For 60 years, we’ve been conducting clinical trials in the same way, where key data signals can take years to reach the FDA. The lag time can delay regulatory decisions unnecessarily and slow down the drug development timeline,” said FDA Commissioner Marty Makary in a statement.

“We are boldly advancing a modern approach whereby FDA scientists can view safety signals and endpoints in real time as a trial progresses. This will help us accelerate promising therapies, and build toward our ultimate goal of running real-time, continuous trials across all phases of drug development.”

The FDA did not detail the AI and data science technologies used in the AstraZeneca and Amgen proof-of-concept projects.

In its RFI, which is open for public comment until May 29, the FDA asked for comments on the types of trials that might benefit most from the application of AI, the type of infrastructure that is needed and how the pilot can accommodate participants with varying levels of AI capacity.

Most clinical development occurs in discrete phases. “Because each defined phase of clinical development is run according to a protocol and typically as a separate study, there is generally a hiatus in the development program after one phase ends and the next begins. This slows the pace of product development,” the agency said.

The agency said it intends to disseminate final selection criteria for the wider pilot program in July and complete pilot selections in August.

Some experts said the FDA’s adoption of AI in the clinical trial process is an important and promising move.

“AI can compress timelines by automating tasks the agency has historically done manually: adverse-event signal detection, protocol-deviation monitoring, document review and pattern recognition across the millions of pages submitted,” said Jim Foote, founder and CEO of First Ascent Biomedical, a biotech firm that is using AI-enabled tools in its work involving advanced oncology treatment discovery.