How to avoid unnecessary testing on animals
Under REACH, testing on vertebrate animals (e.g. rats, other mammals or fish) can only be used as a last resort to fulfil information requirements for registration. For companies manufacturing or importing between 1 to 100 tonnes per year, there are multiple ways to avoid unnecessary animal testing and to reduce the number of animal tests.
For each individual information requirement you should consider the following:
- Gather and share existing data. You may get access to published literature that is sufficient for fulfilling the information requirement. If a result of a valid animal test is available in the SIEF, it must be shared with co-registrants. The owner of the test must be compensated according to pre-agreed rules.
- Data waiving or adaptations: rules for adaptations are part of the legal text. They can be either specific (under column 2 of each endpoint) or general (under Annex XI).
To use the general rules, you can waive data or use an adaptation based on the following scientific arguments:
- Weight of evidence. You have sufficient information from several independent sources that lead to the conclusion that your substance has (or does not have) a particular property.
- QSAR models. Many properties of your substance can be predicted from structurally similar substances by using computer models.
- In vitro methods. Tests performed with isolated tissues, organs or cells instead of a whole organism can be adequate to conclude on an information requirement.
- Grouping and read-across. If you can show that the way your substance behaves for a certain property is similar to how another substance behaves, existing results for that property can be "read across" to your substance.
If you decide to use one of the possibilities, you are claiming an adaptation.
- Prepare a well-documented and valid scientific justification if you adapt the standard information requirements and submit it in your registration dossier.
- The approach you choose must deliver reliable information that is comparable to that from the standard test. If not, then you need to run the test as required.
- The approach you choose must allow you to classify your substance. When you have classified and labelled your substance, additional testing may not be needed.
Note that you should carefully consider whether to rely on the worst-case classification just to avoid more testing. Over-classification may, for example, trigger risk management measures under occupational health and safety legislation or lead to prioritisation for REACH regulatory risk management measures.
More in this section
REACH 2018 phases
Help with terminology