Basic
Emerging
Kevin Snyder, PhD
Certara
Gaithersburg, Maryland, United States
Nigel Greene, PhD
Vice President of Toxicology
Recursion Pharmaceuticals
Salt Lake City, Utah, United States
Michael DeNieu, PhD
Supervisor, Data Management
Labcorp
Madison, Wisconsin, United States
Lennart Anger, PhD, MSc, ERT
Senior Principal Scientist/Computational Toxicologist
Translational Safety
Genentech
South San Francisco, California, United States
Falgun Shah, PhD
Director
Computational Toxicology
Merck
West Point, Pennsylvania, United States
The requirement by FDA for the generation and submission of standardized CDISC-SEND-formatted toxicology study data has enabled the construction of large databases of toxicology study data that can be used to build predictive models. The Nonclinical Topics Working Group of the Pharmaceutical Users Software Exchange (PHUSE) has initiated a project to facilitate collaboration among regulators, pharmaceutical companies, contract research organizations, and software vendors to collaboratively develop open source software solutions to improve the fidelity and accessibility of these methods. More specifically, supervised machine learning models will be trained to detect and characterize patterns in toxicology study endpoints that are associated with the documented conclusions of expert toxicologists, e.g. target organs of toxicity, and then applied to streamline the interpretation of newly generated toxicology study data. Additional study interpretations, e.g. adversity of findings, NOAEL determination, clinical translatability, structure activity relationship – will be explored for development of predictive models. This symposium will provide an update on the progress of this project as well as perspective on the applicability of its deliverables from a diverse set of stakeholders.