Human biology is different from that of rats, mice, or rabbits, and there is no guarantee that the results of animal studies will apply to humans. Although great strides have been made in the development of more decent alternatives in recent years, their regulatory acceptance remains a problem.
Humans often metabolize chemicals differently than animals, further reducing the relevance of animal models for exposure. “Every now and then the body metabolizes something that activates and makes it worse,” says Steve Gutsell, computer scientist and team leader at Unilever’s Safety and Environmental Assurance Center (SEAC). He believes that many cases where animal and human effects are different have their roots in metabolism.
If you speak to the leading scientists in the field, animal models are archaic today
Steve Gutsell, computer scientist and team leader, Unilever
Much work has been put into the development of human-relevant cell models and approaches that enable a more precise assessment of human exposure and great progress has been made. “If you talk to the leading scientists in the field, animal models are archaic today,” says Gutsell. “The standard rat toxicity study has not been carried out for more than half a century.”
Unilever began looking seriously at animal testing with cell-based assays about 40 years ago. Computer modeling now also plays a major role, and the SEAC team includes over 30 mathematical modelers, computational chemists, and bioinformaticians.
However, there are very few specific examples in which a single animal experiment has been directly replaced by an animal-free alternative. Rather, a successful replacement relies on a completely different approach.
Exposure driven risk assessments can be used to make safety decisions. “You can determine the exposure and then the hazards to consider and thus the ultimate risk,” Gutsell says. “We believe it should be done that way. But you can do it the other way around, looking at all possible dangers and gathering information to ensure safety. ‘
Several external collaborations, notably one with the US Environmental Protection Agency (EPA) and another with chemists at Cambridge University, have contributed to Unilever’s progress. The company is also working closely with other companies and institutions as part of the EU ToxRisk research program to find ways to free the science of safety from reliance on animal testing.
the number of chemical receptor interactions that are now cataloged and available in databases represents a real change
Take the collaboration with Jonathan Goodman’s group in Cambridge. Adverse outcome pathways (AOPs) assume that toxicology is a logical sequence of events that can be mapped. “When you talk about an adverse event that occurs due to exposure to a chemical, there are two anchor points,” Gutsell says. “There are the negative consequences that happen to a person – they can develop a rash, change their behavior, or even die. But the first anchor point is a chemical interaction, a “Molecular Initiating Event” or MIE. This is chemistry, not biology. ‘
If a chemical causes an MIE associated with an AOP, it may (or may not) lead to the adverse event. But, critically, if it doesn’t cause this MIE, it can’t. “For example, if it doesn’t bind to the estrogen receptor, it won’t have endocrine effects through an estrogenic mechanism,” he says.
Goodman realized that an atlas of MIEs would be a valuable tool, not to mention a map. “The number of chemical receptor interactions that are now cataloged and available in databases is a real change,” says Gutsell. “I started out by making quantitative predictions of structure-activity relationships using models with sets of maybe 10 or 100 data points. Now each model typically has 2,000 to 10,000 data points – and the Cambridge group has developed predictive models for over 100 MIEs. ‘ It is so successful that the 2020 Atlas Project was awarded the Lush Science Prize for the replacement of animal experiments.
Crucially, Unilever now has an in-house software tool that uses artificial intelligence and neural networks to generate predictions. “You give it a chemical structure, it checks it against 100 potential MIEs and tells you whether the interaction is likely to occur,” says Gutsell. “This broad approach gives some assurance that the broad biological field of toxicology is covered and can serve as a guide for further in vitro testing.”
The second major collaboration with the EPA began in 2015. The EPA’s Toxcast program used in vitro assays to prioritize the chemicals most likely to be toxic for further animal testing. Unilever wondered if safety decisions could be made based on the in vitro data by adding to the exposure scenario.
“You’re looking for large ranges of exposure – big differences between the level of exposure and the level at which you start to see a biological effect,” says Paul Russell, director of chemistry at Unilever SEAC. ‘It worked really well. The assay kit evolved from what was commercially available back then and is now quite targeted and more relevant to humans. It examines phenotypic effects in cells, receptor interactions and also transcriptional changes. ‘
While there is a desire to move away from animal testing from both an ethical and a scientific point of view, it can still be required by regulators. In the EU, the Scientific Committee on Consumer Safety oversees the safety of non-food consumer products and Unilever has discussed with them how developments in “next generation” safety science can be applied to cosmetics. “They like to view exposure-based risk assessment as a way to make safety decisions about products rather than asking for animal data,” Gutsell says.
However, there is a dichotomy – the ingredients of cosmetic products have to be registered under REACH (registration, evaluation, approval and restriction of chemicals) when they are manufactured in the EU or when they are imported into the EU. And the registration process invariably requires data from animal experiments.
“The way Reach is enforced by the European Chemicals Agency (ECHA) essentially requires that you conduct animal testing,” he says. Submissions with all possible alternatives have been tried by the industry but the dossiers are routinely rejected, with ECHA requiring tests such as 90-day rodent feeding studies instead. Many chemical suppliers feel that there is no alternative to animal testing.
We have to get better at articulating this; ECHA needs to understand them better.
However, there are in vitro methods validated by the OECD for some well-defined endpoints, such as skin sensitization, and their value is clear. A rethink is required for more complex systemic effects. “You will never find a direct replacement for the classic 90-day rodent feeding study, which can have many systemic effects,” says Gutsell. “The idea that there could be a drop-in replacement like a cell-based assay or a computer model is unrealistic.”
A real mindset shift is required. The amount of an ingredient in a product can be very low and exposure low when used by a consumer. However, ECHA is also concerned about the occupational safety of workers who manufacture and use chemicals in large quantities, rather than just small quantities contained in a single consumer product. Even then, the exposure potential in a factory setting can be very low in reality, despite the large quantities.
Gutsell gives the example of a chemical that Unilever uses in a reaction to make a soap product. “The substance is practically harmless: if it comes into contact with your hands for a long time, it is slightly irritating,” he says. “There is virtually no chance of occupational exposure: the only time operators come into contact with it is when maintenance protocols are not properly followed. Even then, the danger and risk are very low. However, ECHA says testing is needed as there is potential for some exposure, although unlikely. “
We need to provide free education and training materials around the world that have been translated into many languages so that people can understand how to use the new risk assessment tools
Paul Russell, Head of Chemical Science, Unilever
The way Reach looks at toxic and non-toxic in a binary way isn’t scientific, he adds. “It’s a sliding scale and the exposures are quite acceptable. It would be better to say that a substance is safe when used with these controls or below these concentrations. If we had invested even 10% of our efforts in generating animal data to understand the actual exposure of workers in the factory, we could show that the tests are not necessary. We have to get better at articulating this; ECHA needs to understand them better. “
Russell agrees. “We need to provide education and training materials around the world on cruelty-free approaches to free safety assessment, translated into many languages so that people can understand how to use the new risk assessment tools,” he says. “An exposure-based approach is key to being able to do this without animals.” Programs like the Animal Free Safety Assessment Collaboration and the Save Ralph Campaign with Humane Society International are important in raising awareness, and many other industry partners are involved.
He adds that Unilever is also a strong advocate for global regulatory change and is backing down ECHA’s demands for more chemicals to be tested on animals, regardless of how they are used. “We need to share our science and make everything we do in our research program publicly available,” he says. “We need to share case study examples using non-animal methods to show that we don’t need to use animals. The science is already there. We have to use it. ‘