Development of a decision support system for the introduction of alternative methods into local irritancy/corrosivity testing strategies. Creation of fundamental rules for a decision support system.

Abstract

The notification procedure of the European Union (EU) for new chemicals requires the application of protocols on physicochemical and toxicological tests for the evaluation of physicochemical properties and probable toxic effects of each notified substance. A computerised database was developed from data sets and toxicological test protocols relating to substance properties responsible for skin and eye irritation/corrosion. To develop specific structure-activity relationship (SAR) models and to find rules for a decision support system (DSS) to predict local irritation/corrosion, physical property data, chemical structure data and toxicological data for approximately 1300 chemicals, each having a purity of 95% or more, were evaluated. The evaluation demonstrated that the lipid solubility and aqueous solubility of a chemical are relevant to, or - in some cases - responsible for, the observed local effects of a substance on the skins and eyes of rabbits. The octanol/water partition coefficient and the measured value of the surface tension of a saturated aqueous solution of the substance give additional information that permits the definition of detailed SAR algorithms that use measured solubility values. Data on melting points and vapour pressure can be used to assess the intensity and duration of local contact with a chemical. Considerations relating to the reactivity of a pure chemical can be based on molecular weight and the nature of the heteroatoms present. With respect to local lesions produced following contact with the skin and eyes of rabbits, the data evaluation revealed that no general "local irritation/corrosion potential" of a chemical can be defined. A variety of mechanisms are responsible for the formation of local lesions on the skin or in the eyes: serious lesions are produced by mechanisms different from those that cause moderate irritation in these organs. In order to develop a DSS that uses the information extracted from the database, chemical main groups were categorised on the basis of their empirical formulae, and rules were defined of the type IF (physicochemical property) A, THEN not (toxic) effect B, based on correlations between specific local effects and measured physicochemical values. Other rules of the type IF substructure A, THEN effect B were developed based on correlations between specific local effects and the submitted structural formulae. Reactive chemical substructures relevant to the formation of local lesions and rules for the prediction of the absence of any skin irritation potential were identified. Proposals are made relating to the development of alternatives to eye irritation testing with rabbits.