The role of testing in CLP
For classification and labelling purposes, manufacturers, importers or downstream users of a substance or mixture are required to gather and assess any existing available information related to the hazardous properties of a substance or mixture.
If no data are available, ecotoxicological and toxicological tests should be performed that comply with the requirements under the REACH Regulation, the OECD principles of good laboratory practice (GLP) and any internationally recognised methods validated according to international procedures to ensure that the data are of high quality and reliable.
If animal tests are being performed, they should comply with the requirements for the protection of laboratory animals (Directive 2010/63/EU).
Alternative methods to animal testing
A number of alternative methods have been developed to replace the use of animals with non-animal systems, reduce the number of animals in a test, or refine the procedures to make them less painful or stressful to the animals under study (3Rs principle).
ECHA, stakeholders and many other regulatory authorities have endorsed the 3Rs principle. Directive 2010/63/EU on the protection of animals used for scientific purposes includes an explicit reference to the 3Rs principle.
Alternative approaches may also take into account chemical properties, (Q)SAR predictions and models, and in vitro tests on cells or tissues with current or new technologies including genomics and proteomics. Furthermore, toxicological properties of substances may be predicted using information from test data on analogues by the ‘read-across’ approach or for a group of substances using the ‘category’ approach. Sufficient information and proper justification must be presented to support these predictions.
ECHA promotes alternatives to animal testing that meet regulatory needs by assessing the risks of substances to human health and the environment, while avoiding unnecessary animal testing.
Testing for human health and environmental hazards
CLP does not normally require new testing regarding health and environmental hazards of substances or mixtures for the purposes of classification and labelling.
Only if all other means of generating information have been exhausted and data of adequate reliability and quality are not available, may testing be undertaken as a means of generating new information in compliance with the test methods referred to in Article 13(3) of REACH. Testing on humans and non-human primates is prohibited.
Information to be assessed before considering in vivo testing includes:
- data from tests,
- whether or not they were carried out according to the principles of GLP,
- historical human data,
- data from in vitro methods, and
- the use of relevant information from analogous substances or mixtures to predict hazardous properties for the ‘target’ substances or mixtures under consideration (read-across).
For human health hazards, no in vitro tests or Q(SAR) predictions could currently fully replace toxicology studies performed to characterise the health effects of chemicals for a number of endpoints, including long-term exposure or exposure over multiple generations.
Data should be generated for relevant routes of exposure (oral, dermal and inhalation) and on the forms or physical states in which the substance is placed on the market and in which it can reasonably be expected to be used.
When the available test data are only supportive, contradictory, equivocal or cannot easily be compared with the CLP criteria, a weight-of-evidence approach using expert judgement should be carried out. Relevant data obtained from other sources (e.g. scientifically valid clinical or epidemiological case studies) may be used for classification and labelling.
For environmental hazards, data used for classification (aquatic acute and chronic) are principally based on establishing toxic hazard endpoints in three different aquatic trophic levels using highly standardised test protocols. Fish, crustacea and algae (or other plants) are used as surrogates representing a range of species and taxa within each trophic level. Environmental fate information (degradation and bioaccumulation) of a substance or mixture is used in combination with toxicity data to determine chronic toxicity and M-factors.
Testing for physical hazards
For physical hazards, there is an obligation for the manufacturer, importer or downstream user of a substance or mixture to generate new data, unless there is adequate and reliable information already available.
Existing data for physical hazard determination should be evaluated in terms of their suitability and the quality of the tests used.
New tests must be performed in compliance with a recognised quality system or by accredited laboratories (e.g. with EN ISO/IEC 17025) and should be based on the methods or standards referred to in Part 2 of Annex I to CLP.
The tests should be conducted with the appropriate physical state and form of the substance and mixture that is put on the market. If, for example, for the purposes of supply or transport, the same chemical is to be presented in a physical form different from that which was tested and which is considered likely to materially alter its performance in a classification test, the substance has to also be tested in the new form.
Therefore, parameters such as concentration, shape, particle size, density, etc. may also influence the outcome of the test and should then be reported.