Disclaimer: The views presented in the Newsletter do not necessarily represent the official position of the European Chemicals Agency. All content and links are up to date at the time of publication. Users are informed that any reliance on the accuracy and/or completeness of any content posted by other users on this website is done at their own risk. The European Chemicals Agency (ECHA) accepts no liability in respect of any content posted by users.
Editor-in-chief: Maurizio Roncaccia
Editors: Paul Trouth and Päivi Jokiniemi
Article related to: REACH
Want to know about… read-across?
If you are preparing a registration, read-across could well be an option for filling missing information. With it, you can use information from a known substance to predict the properties of another one. When used correctly, it can help you avoid unnecessary testing on vertebrate animals.
Specific expertise required
Read-across is a complex process and you need specific (eco)toxicological expertise to do it. It is mostly used for endpoints such as reproductive toxicity, repeated dose toxicity and long-term toxicity testing on aquatic organisms and is therefore mostly relevant for high tonnage registrations. However, for the 2018 deadline, you can also use read-across for registrations above 10 tonnes for:
screening studies on repeated/developmental toxicity;
a repeated dose toxicity 28-day study; or
short-term toxicity testing on aquatic organisms.
If used correctly, read-across can also reduce your costs as there is no need to test every target substance. However, developing scientifically robust predictions based on read-across takes work and ECHA has rejected the majority of cases so far because registrants have not provided sufficient justification for their predictions or enough scientific evidence to support them.
Explain, explain, explain
You should always explain why and how it is possible to predict the target substances’ properties. The starting point should be the structural differences and similarities between the source and the target substance.
Generic information on substances is not sufficient to make a read-across case. For example, just stating that two substances are structurally similar is not enough to predict their toxicological properties – these need to be scientifically proven.
Show your working
You should adequately document the scientific reasoning for any read-across. This should cover, among other things, the assumptions made and the conclusions drawn. Key data should be identified and it should include references to the substance dataset. The information needs to be substance-specific.
You need to include the supporting scientific evidence in the dossier and provide robust study summaries where possible. This is necessary so that ECHA can verify your read-across hypothesis.
This framework can help you to assess the quality of your own read-across. It covers REACH information requirements concerning human health, environmental fate and environmental hazards. It presents the scientific aspects that ECHA considers crucial when evaluating read-across. Hopefully, this will enable you to improve your explanations of why and how read-across can be used in your dossier.
ECHA developed the RAAF based on the most frequent types of read-across approaches used. These are written as scenarios. Each scenario is characterised by a number of scientific considerations, which are crucial to assessing read-across.
These assessment elements include questions, possible outcomes and examples. Answering these questions helps to determine whether the read-across approach is scientifically acceptable or not.
The framework does not replace the official guidance on read-across for registrants. It complements it.
Another document on read-across involving substances of unknown or variable composition, complex reaction products or biological materials (UVCBs) and multi-constituent substances is currently being drafted and will be published later this spring.
10 tips for a good read-across
Make use of the QSAR Toolbox or similar tools to look for structurally similar substances.
Provide substance identity information on all substances included in the read-across. Also, consider impurities and potentially different substance compositions when developing your argument.
Give a hypothesis-driven justification for why the data from one substance can be used to fill the data gap for another substance. Do that for each property.
Show how structural similarity and dissimilarity between the substances justifies your prediction.
Create a data matrix that indicates consistency over the properties and highlights potential trends within the category.
Analyse experimental data to confirm your hypothesis and check whether there are any contradictions.
Justify read-across adequately and provide supporting and credible information.
Verify that the source information you want to use for read-across complies with the REACH information requirements.
Provide toxicokinetic information on the substances under consideration to make the read-across hypothesis more robust.
Use the RAAF to check how robust your read-across adaptation is.