Integrated Testing Strategies for Use with Respect to the Requirements of the EU REACH Legislation

Abstract

Integrated testing strategies have been proposed to facilitate the process of chemicals risk assessment to fulfil the requirements of the proposed EU REACH system. Here, we present individual, decision-tree style, strategies for the eleven major toxicity endpoints of the REACH system, including human health effects and ecotoxicity. These strategies make maximum use of non-animal approaches to hazard identification, before resorting to traditional animal test methods. Each scheme: a) comprises a mixture of validated and non-validated assays (distinguished in the schemes); and b) decision points at key stages to allow the cessation of further testing, should it be possible to use the available information to classify and label and/or undertake risk assessment. The rationale and scientific justification for each of the schemes, with respect to the validation status of the tests involved and their individual advantages and limitations, will be discussed in detail in a series of future publications.

Using In Silico Tools in a Weight of Evidence Approach to Aid Toxicological Assessment

“…The conclusion of a positive prediction for skin sensitisation from WoE could be used along with reactive chemistry and in vitro data to allow the classification and labelling or risk assessment to be performed on a compound, without the need for in vivo testing 15. When a negative outcome is given, in silico predictions are not considered sufficient on their own,39 and therefore must be backed up with the other testing data mentioned. Concurrent data at this stage though may still be considered sufficient for classification and labelling or risk assessment.…”

“…Integrated testing strategies are an integral part of using alternative methods for toxicity testing for REACH,4–15 and can formally be defined as ‘any approach to the evaluation of toxicity which serves to reduce, refine or replace an existing animal procedure, and which is based on the use of two or more of the following: physicochemical data, in vitro data, human data, animal data (where unavoidable), computational methods (such as (Q)SARs ((quantitative) structure‐activity relationships)) and biokinetic models’ 4…”

“…The development of integrated testing strategies has been discussed previously for many different endpoints 4–15. The initial stages of these decision trees firstly examine any existing data.…”

“…Although the proposed integrated testing strategies within the literature4–15 all state that obtaining information from in silico methods is an important step, only some give examples of programs and models which could be used for the specific endpoint(s) discussed. Also, none of the schemes make it clear how the outcomes or predictions from these sources should be used within a WoE argument.…”

Sequential Testing of Chemicals When Costs Matter: A Value of Information Approach

“…Sequential testing strategies have been suggested for several toxicological endpoints (Van Leeuwen et al 2007;Blaauboer 2008;Combes et al 2008). Furthermore, they have been widely recognized and recommended as tools for optimizing regulatory decision-making on chemicals' hazards and risks (e.g., under the new European chemicals legislation REACH) (ECHA 2008;Grindon et al 2008). However, the key general principles of sequential testing that can guide risk managers and toxicologists have not been fully explored yet: Which test should be performed first and when should testing stop?…”