Toward implementing virtual control groups in nonclinical safety studies.

Area of Science:

• Toxicology
• Animal Welfare
• Regulatory Science

Background:

• Historical control data in animal toxicity studies are primarily used for comparative analysis.
• The controlled nature of these studies suggests potential for using historical data to form virtual control groups.
• Implementing virtual control groups could significantly reduce the number of animals used in safety testing.

Purpose of the Study:

• To explore the feasibility and prerequisites for utilizing historical control data to establish virtual control groups.
• To identify necessary data collection, curation, statistical evaluation, and validation strategies.
• To assess the potential of virtual control groups to replace concurrent controls and reduce animal usage.

Main Methods:

• A workshop convened stakeholders from pharmaceutical, chemical industries, academia, regulatory bodies (FDA), CROs, and NGOs.
• Summarized European initiatives for centralized animal control data sharing and curation.
• Explored initial approaches for matching criteria between virtual and treatment groups and qualification strategies.

Main Results:

• The report outlines current efforts in data centralization and curation for animal control data.
• Identified initial methods for matching virtual controls to study treatment arms.
• Began reflecting on qualification procedures and potential challenges associated with virtual control groups.

Conclusions:

• Virtual control groups offer a promising avenue for reducing animal use in toxicity studies.
• Careful consideration of data management, statistical methods, and validation is crucial for successful implementation.
• Ongoing collaboration and development are needed to refine the concept and ensure its reliability for regulatory use.