The FDA has qualified its first artificial intelligence tool to help drug developers evaluate fatty liver disease in their clinical trials.
The agency gave a green light to the AI-Based Histologic Measurement of NASH (AIM-NASH), a cloud-based technology from PathAI designed to assess disease activity for metabolic dysfunction-associated steatohepatitis (MASH), according to a Dec. 8 release.
More specifically, AIM-NASH is meant to help score the different stages of liver biopsy, including fat infiltration, inflammation and scarring.
The freshly qualified system uses AI algorithms to analyze images of liver biopsies based on an industry-standard scoring system. Pathologists will then review the images and either accept or reject the AI-generated scores. The program has not yet received an FDA clearance for its use as a diagnostic outside of a clinical trial.
The FDA’s move was based on “comprehensive validation studies” that found the AIM-NASH-powered comparisons to expert scores were similar to the comparisons between individual pathologists and expert findings, according to the agency release.
Related
MASH is a severe liver disease that impacts millions of Americans. The condition can worsen liver function and lead to liver cancer, liver transplant and death.
Historically, the indication has carried a reputation as one of biopharma’s most stubborn challenges. This is due, in part, to the complexity of the disease and the invasive nature of liver biopsies.
Currently, MASH studies require several experts to independently evaluate liver histology, a process that can be both time-consuming and produce variable results. The new AIM-NASH program may help standardize these assessments and cut down on how long the process takes.
Earlier this year, the FDA said it was considering using a noninvasive liver measurement as a surrogate endpoint for certain MASH patients, a move that could speed up drug development.
