FDA Accepts AI-Based Digital Liver Model into ISTAND Program to Predict Drug-Induced Liver Injury
Key Insights
The FDA's CDER has accepted a letter of intent for an AI-driven digital liver model under the ISTAND pilot program, marking the first step toward qualification as a Drug Development Tool.
The tool predicts drug-induced liver injury (search)—a leading cause of clinical trial failures—by comparing new small-molecule chemical structures with drugs of known safety profiles.
If fully qualified, drugmakers could use the model in regulatory submissions to improve early safety assessments, reduce reliance on animal testing, and support informed decisions before human trials.
The U.S. FDA's Center for Drug Evaluation and Research (CDER) announced on Wednesday that it has accepted a letter of intent for an artificial intelligence-based Drug Development Tool designed to predict drug-induced liver injury (search) (DILI)—one of the most significant causes of clinical trial failures. The tool, an AI-driven digital liver model, has been admitted to the agency's Innovative Science and Technology Approaches for New Drugs (ISTAND) pilot programme.
Drug-induced liver damage remains a persistent challenge in pharmaceutical development, and current methods do not reliably predict human risk. The FDA noted that this new tool could potentially help improve early safety assessments, reduce reliance on animal testing, and support more informed decisions before human clinical trials begin.
How the AI Digital Liver Model Works
The model is designed to assess the hepatotoxicity risk of new small-molecule drugs by comparing their chemical structures with existing medicines that have established safety profiles. Rather than replacing current evaluation methods, the tool complements them and aids decision-making in early development stages. By flagging compounds with structural similarities to known hepatotoxic agents, the model offers an additional layer of safety intelligence before candidates advance to human testing.
ISTAND Program and Qualification Pathway
The acceptance of the letter of intent represents the first step in a multi-stage qualification process under the FDA's Drug Development Tool qualification programme. This programme works collaboratively with developers to guide tools toward a specific context of use. The letter of intent is the initial submission stage used by the FDA to determine whether a tool will be accepted into the programme.
If the digital liver model successfully completes all stages of qualification, drug manufacturers would be permitted to use the tool in their regulatory submissions, including marketing applications.
Leadership Perspective
"New technologies are showing incredible promise in helping improve and streamline drug development, with the ultimate goal of enhancing patient care," said Michael Davis, CDER's acting director. His remarks underscore the agency's growing openness to computational and AI-driven approaches that can address longstanding bottlenecks in drug safety evaluation.
The review process will now proceed through the ISTAND framework, with the FDA working alongside the tool's developers to validate its predictive performance and define the specific context in which it can be reliably applied.