Skip to main content

New IHI project hopes to dramatically cut animal use in standard safety studies

VICT3R will use existing data and artificial intelligence to create virtual control groups that could replace real animals in drug and chemical safety studies.

04 October 2024
A close-up of a white mouse, seen from the front, against a black background. Image by Janson George via Shutterstock
Image by Janson George via Shutterstock

Contributing to the '3Rs'

The EU has long had the goal of replacing, reducing and refining the use of animals in research (the ‘3Rs’). However, studies involving animals are still required by legislation relating to the evaluation of medicines and chemicals, for example.

The aim of VICT3R is to dramatically reduce the numbers of animals used in safety testing by replacing actual control groups with ‘Virtual Control Groups’ (VCGs). Control groups (which do not receive the substance under study) account for around 25% of the animals used in drug and chemical safety studies, so replacing them with virtual animals would have a significant impact on the numbers of animals used.

Introducing the Virtual Control Group concept

The Virtual Control Group concept was born out of the IMI2 project eTRANSAFE, which set out to improve the quality of the toxicity tests that are carried out on potential medicines before they are tested in humans. Over the years, pharmaceutical companies and other organisations have amassed a wealth of data on control groups in toxicity tests, and eTRANSAFE realised that this data could be harnessed to create virtual control groups. Tests demonstrated the feasibility of the VGC concept, but a lot more work is needed before it can be used in practice.

VICT3R aims to address the challenges hampering the implementation of VGCs and achieve regulatory acceptance of the VGC concept.

It will do this by building on the work started in eTRANSAFE to develop a database comprising high quality, standardised data from past toxicity studies. Operational procedures and computational tools will help to support the characterisation and quality control of data collected. The VICT3R project will then draw on statistics and artificial intelligence (AI) to create workflows and computational tools to generate VGCs based on the data in the database. They will assess the validity, reproducibility and robustness of the VGCs by comparing their performance with that of real control groups.

"Data reuse and data sharing, as well as the application of AI approaches, engender terrific opportunities for implementing the 3R principles in the use of animals in preclinical research,” said project coordinator Ferran Sanz of Universidad Pompeu Fabra.

Collaboration - the key to success

Ultimately, the project hopes to achieve regulatory acceptance of the VGC concept. Furthermore, while VICT3R focuses on specific toxicity studies, the team will also explore how the concept could be applied in other types of studies involving animals.

“The endeavour of changing a regulatory paradigm in nonclinical safety testing cannot be undertaken by a single company or institution,” explained VICT3R’s industry lead, Thomas Steger-Hartmann of Bayer AG. “It definitively requires a broad collaborative approach where only IHI offers a unique and ideal umbrella.”