Concerns have been raised that the implementation of the EU’s Chemicals Strategy for Sustainability would increase the use of animal testing. There are also calls for speeding up the replacement of animal tests with so-called NAMs (New Approach Methodologies).

The reluctance to test chemicals on animals is naturally shared by most people, including us at ChemSec. However, the aim of laboratory testing of chemicals on animals is to avoid harmful effects on people and wildlife outside of the laboratory. So, we need to find a balance between the safety of people and wildlife outside of the laboratory and the welfare of animals inside the laboratory.

The aim of this paper is to provide an overview of why and how animal testing is part of the EU’s chemicals legislation and to discuss various possibilities and pitfalls when moving towards a reduction in animal testing.

The current situation

Animal tests are used for many purposes
Historically, animal tests have been the only viable way to obtain information about the possible harmful effects of chemicals used in pharmaceuticals, pesticides, consumer products and industrial processes. In 2019, tests were conducted on about 10.6 million animals in the EU. The main purpose for these tests was scientific research (72%) while 17% of all tests were conducted to satisfy legislative requirements. Within these, medicinal products accounted for 61%, veterinary medicine for 17% and industrial chemicals for 8.6%. In summary, about 1.4% of all animal tests in the EU are used for testing “industrial chemicals”, which include tests performed under the REACH and CLP regulations.

Information requirements
Globally, the production of man-made chemicals has grown 50-fold since 1950, and it is set to triple again by 2050 compared to 2010. New chemicals are being introduced at a faster pace than we are able to assess basic hazard properties. Scientists have therefore concluded that we have exceeded the planetary boundaries for chemicals and plastics. The number of chemicals on the EU market alone is estimated at 100,000 and for many of these the available information on hazards and exposure has been characterised as poor.

The CLP regulation is the basic EU regulation for classifying hazardous chemicals. The aim is to ensure that all chemicals on the market are classified, labelled and packaged according to their hazardous properties, including toxic effects on humans and the environment.

The REACH regulation, introduced in 2007, is based on the recognition that a lack of basic data on chemicals has created unprecedented risks for human health and the environment. Many chemicals had not been properly tested before they were placed on the market and were therefore able to spread widely in the environment. One of the main purposes of REACH was to introduce the principle of “No data – No market” to ensure that at least a basic set of hazard data is available for chemicals marketed in the EU. The larger the volume of a chemical that is produced, the more information is required.

The approach to animal tests in the REACH and CLP regulations
For chemicals marketed in higher volumes (>10 ton/yr), the data required to comply with REACH have historically involved testing on fish, mice, rats or guinea pigs. There are no requirements to use larger mammals.

Animal welfare was already in focus when REACH was adopted and the regulation includes several mechanisms to limit animal tests, which may explain the relatively low number of tests compared to other sectors. These mechanisms include:

• An obligation for all companies that register the same chemical to share information so that no test is performed twice.
• That registrants needing data from a test on vertebrate animals must first send their testing proposal to the European Chemicals Agency, so the Agency can assess if the test is necessary.
• The option for companies to waive data requirements, if they can provide evidence that a given test is not necessary.

The EU General Directive for protection of animals used for scientific purposes also applies to both the REACH and the CLP regulations. According to CLP, companies must classify chemicals based on existing information and only if all other means of getting information have been exhausted may testing on animals be done.

Recent science has shown that there are still huge data gaps, especially regarding information such as a chemical’s potential to disrupt our hormone system or impair brain development. Acknowledging the serious shortcomings of the current chemicals policy, the EU’s Chemicals Strategy for Sustainability includes plans to revise the regulations and, among other things, require better information. It has been predicted that new data requirements would lead to many more animal tests. However, it is not possible to predict how many as this will also depend on whether data can be obtained by other means than animal tests.

It is worth noting that high estimates of the need for future animal tests also reflect the huge number of poorly tested chemicals that are still on the market.

Which non-animal methods are available today?
Scientists are working to develop new approaches towards assessing chemicals and new test methods that can replace animal tests. These include cell cultures and tests on human tissue or human volunteers. A rapidly growing field is the various “omics” technologies that make it possible to get information on tens of thousands of endpoints from one sample, such as exposed cell cultures (or samples from tested animals). Computer-based models are also becoming more and more effective in predicting the toxicity of chemicals based on their chemical structure.

The current regulatory system is largely based on the information obtained from animal tests and many of the new test methods do not provide the same type of information. In order for the new methodologies to be used more widely to regulate chemicals, the tests would first need to be developed and then validated to determine how well they predict the possibility of harmful effects. This is a substantial and expensive effort which is to a large extent in progress. But it will take time before every animal test can be replaced by non-animal tests which enable at least the same level of protection.

In summary, it is not currently possible to replace most animal tests. Even if given the highest priority, it will take many years before it is possible to fully replace animal tests with New Approach Methodologies.

Possibilities to already now reduce animal testing

Fortunately, there are already many possibilities to reduce and avoid many animal tests.

• Make better use of the precautionary principle and avoid “paralysis by analysis”

When it comes to discussions about restricting a chemical or identifying it as a Substance of Very High Concern (SVHC), there is a tendency amongst industry representatives to argue that more data is needed before a regulatory decision can be made. One important contribution towards reducing animal tests could be for animal welfare organisations and industry stakeholders to support authorities in acting quicker on available data. This would be in line with the precautionary principle and lead to more effective control of chemicals while avoiding further animal tests.

• NAMs can fill knowledge gaps

If used in the right way, NAMs can contribute to improving safety. Non-animal methods have the potential of being cheaper and quicker than many animal tests. As such, they can be used to evaluate a large number of chemicals for hazardous properties and fill data gaps. Depending on the endpoint, NAMs may already provide enough evidence to prove harm, which is essential in order to regulate substances. When it comes to providing evidence of safety, as for REACH registration, NAMs are however in many cases not sufficient at the present moment.

• Improved data sharing would enhance information and limit tests

REACH requires that companies producing the same chemical must share their results from mandatory tests. If companies and scientists would also share non-mandatory data, as well as data that might be relevant for the evaluation of similar substances, then further animal tests would be avoided.

• Improved study design and better use of independent academic data
• Grouping of chemicals would improve safety and limit animal tests

In recent years, regulators have been working on the concept of grouping similar substances for hazard assessment and restriction. This practice should be used much more to speed up substitution of harmful substances and avoid further hazard testing of chemicals in the same group. Sharing of data for different substances in the same group would also provide useful information that could limit the need for animal tests.

Pitfalls to watch out for

Some proposals to limit animal tests will inevitably lower the protection of human health and environment. In our view, this is not acceptable.

• The “No Data – No Market” principle must prevail

In recent years, we have seen how untested and unregulated chemicals have created huge problems for both the environment and human health. If we are to continue producing chemicals at the same rate as today, we need to raise the level of information in the future rather than to lower it. However, according to the EU Commission, the REACH requirement to only use animal testing as a last resort has already been implemented at the expense of information that could be relevant for the protection of human health and the environment. This is the most important pitfall to avoid. It is clear that as greater priority is given to avoiding animal testing, the more important it will be to keep in mind the basic “No Data – No Market” principle of the REACH regulation.

• It is not enough to know the exposure

Industry stakeholders and others have suggested to focus more on exposure in risk assessments to avoid testing chemicals when the exposure is low. However, it is not possible to assess the risks of using chemicals without knowing the integrated hazards of the chemical. It is also often difficult to obtain information on real-life use and combined exposure of chemicals used in consumer products. Finally, an exposure-based approach would have little value for the most harmful chemicals, especially those for which there is no minimum concentration for effect and no safe exposure level.

• Are animal tests unreliable?

Animal welfare organisations sometimes denounce animal tests as “unreliable”. However, there are no independent sources that show that NAMs are more reliable than animal tests, or vice versa. In some cases, NAMs may outperform animal tests with regards to reproducibility. Animal tests certainly have weaknesses and uncertainties, but their advantage is that their use is well-established among scientists and regulators. If a new testing regime based on NAMs is developed it is important to have a broad understanding of the useability and the uncertainties associated with the new approach. This is another reason to ensure a prudent transition.

Summary and way forward

In summary, we find there are several possibilities for finding common ground between concerns for animal welfare and concerns for chemical safety. But we also see some pitfalls that must be avoided.

REACH and CLP account for less than 1.4% of all animal tests done in the EU today and are mainly conducted on fish, mice, rats or guinea pigs. There is no verified data available to show how many more animal tests would be needed with the revision of the CLP and REACH regulations.

In the near future, there are unfortunately no alternative methods available to replace all testing on animals. There are, however, ways to minimise the use of animal tests without compromising safety. And with this paper, we hope we have helped contribute to this.