Investigational New Drug (IND) Application aims to establish a safety profile and the first human dose for a drug candidate to enter the Phase 1 clinical trial. The IND submission consists of a broad range of data to be reviewed, e.g., Various organ toxicity such as hepatotoxicity and Cancer potential such as mutagenicity and carcinogenicity. It has a 30-day review clock for the FDA to provide a decision. Due to a tight review clock, a process has been established based on the existing information about the drug candidate; among them is the availability of its safety profile. The FDA promotes alternative methods, including artificial intelligence (AI), to support drug discovery and development. In this presentation, I will introduce a SafetAI framework that consists of a suite of AI models for various safety endpoints critical to the IND review. Specifically, I will employ DeepDILI and DeepCarc models to exemplify how the SafetAI framework works and the potential in the IND review process. Meanwhile, I will share lessons learned from the communication with FDA ISTAND (Innovative Science and Technology Approaches for New Drugs) program, which aims to encourage tools that leverage digital health technologies, such as AI-based algorithms to evaluate patients, develop novel endpoints, or inform study design in regulatory applications. Finally, I will provide the future work of SafetAI and the remaining challenge of translating AI solutions into real-world applications.

SafetAI – an AI framework to facilitate the CDER IND review

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.