New York City-based tech company Inductive Bio has secured up to $21m from the US Advanced Research Projects Agency for Health (ARPA-H) to lead the development of AI drug toxicity prediction models, aimed at improving drug safety assessment and minimising reliance on animal testing.
The Digital Acceleration of Toxicity Assessment with Mechanistic and AI-driven Predictions (DATAMAP) project, in collaboration with Baylor College of Medicine, Amgen, Torch Bio and Cincinnati Children’s Hospital Medical Center, will leverage data from advanced human model systems.
The project team will generate biological data from organoids, ex-vivo human tissue, and microphysiological systems.
These models are designed to better reflect human biology compared to traditional approaches. The team aims to minimise late-stage drug development failures and accelerate delivery of safer therapeutics to patients.
The award is secured from the ARPA-H Computational ADME-Tox and Physiology Analysis for Safer Therapeutics (CATALYST) programme, which seeks to transform preclinical drug safety prediction using human-based models.
Almost 90% of clinical-stage drug candidates currently do not reach the market, with one quarter failing due to safety concerns that preclinical animal tests did not predict.
The project will initially focus on cardiotoxicity and drug-induced liver injury (DILI), responsible for almost 40% of drug withdrawals due to safety issues.
Inductive plans to develop AI models using the generated data and will collaborate with the US Food and Drug Administration (FDA) to confirm their suitability for regulatory purposes. Amgen will provide guidance on development, the practical context of drug development and possible uses for new therapies.
Inductive Bio co-founder and the project team principal investigator Ben Birnbaum stated: “While animal tests have historically been the gold standard for assessing drug safety pre-clinically, the reality is that they often fail to capture how humans will actually respond to new therapeutic candidates.
“We’re excited to bring together Inductive’s datasets and AI models with organoids, ex-vivo human tissue and microphysiological systems to better predict drug safety issues that animal testing misses. Our biopharmaceutical collaborator will demonstrate these technologies’ real-world impact by submitting an investigational new drug application to the FDA using the technology.”
