European researchers are at the forefront of a major shift in chemical safety testing, leveraging artificial intelligence (AI) and advanced human-based models to move away from traditional animal experiments. This initiative aims to address the ethical concerns, high costs, and scientific limitations associated with current methods, which see approximately 150,000 rats and mice used annually in the European Union.
The Imperative for Change in Safety Testing
For decades, animal testing has been the standard for assessing chemical safety. However, this practice presents significant challenges. It is both expensive and time-consuming; testing a single substance for chronic toxicity can cost up to €15 million and take three to seven years. More critically, the scientific community, including Professor Mathieu Vinken of Vrije Universiteit Brussel, notes that animal test results may only be about 50% representative of human reactions.
The EU's REACH chemical-safety framework already promotes the “three Rs” principle: replace animal testing where possible, reduce the number of animals used, and refine procedures to minimize suffering. The difficulty in studying long-term effects, such as cancer or reproductive issues, in animals with much shorter lifespans than humans further underscores the need for alternative approaches.
AI and New Approach Methodologies (NAMs) Lead the Way
Scientists are developing New Approach Methodologies (NAMs), a diverse set of techniques designed to reduce or replace animal testing. AI is a pivotal component of this evolution. Instead of observing animal reactions, researchers collect biological data from human cells and employ computational models to identify toxicity patterns and predict potential effects.
This approach becomes particularly powerful when AI is integrated with laboratory systems that mimic human biology. One notable example is the “organ-on-a-chip” technology. These miniature devices contain human cells engineered to reproduce specific organ functions. Researchers expose these chips to chemicals and observe human biological responses, with AI then interpreting the complex data to predict impacts on organs like the liver, kidney, or brain.
Collaborative European Projects Drive Innovation
Several EU-backed projects within the €60 million ASPIS research cluster are advancing these non-animal testing methods:
- PrecisionTox: This project compares how approximately 200 chemicals affect various organisms, including fruit flies, roundworms, and human cells. Its goal is to identify shared biological mechanisms, shifting the focus from merely knowing if a chemical causes harm to understanding why it does, thereby building more human-relevant models.
- RISK-HUNT3R: This initiative combines non-animal test results with real-world human exposure data to comprehensively assess chemical risks. By testing 60 chemicals, it aims to produce safety assessments that regulators can utilize, moving beyond simple lab replacements to a holistic risk prediction system.
Towards a Future of Reduced Animal Testing
While researchers acknowledge that completely eliminating animal testing is not yet feasible, these advancements offer significant potential to reduce animal use and enhance testing accuracy. Europe has a precedent for such shifts; animal testing for cosmetics was phased out over two decades, culminating in a full EU marketing ban in 2013, which spurred the development of alternatives like reconstructed human skin models.
The ultimate goal is to establish a new paradigm where human biology, sophisticated laboratory models, real-world exposure data, and AI collaboratively predict chemical risks. This transition promises a more humane, efficient, and scientifically robust approach to ensuring chemical safety for people.