Strategic considerations in industry's use of in vitro toxicology

Evaluation of the EYTEX® System as a Screening Method for the Ocular Irritancy of Chemical Products

The EYTEX® method is an in vitro test used to predict ocular irritation, based on alterations in a protein matrix. We have evaluated this method with the aim of using it to screen chemicals. One hundred and forty-two products (commodities and specialities), having a wide range of chemical properties and ocular irritation potential in the rabbit, were tested with either the standard, MPA, AMA, KMA or UMA protocols. The results were compared with in vivo data obtained previously for each chemical and with the EEC labelling of eye irritation for dangerous substances. Intralaboratory repeatability and interlaboratory reproducibility were evaluated with seven other laboratories.

Ninety-three per cent of the chemicals tested were qualified with the EYTEX method. The coefficients of the linear correlation between the EYTEX score and the maximal Draize score on the one hand and the maximal corneal score on the other hand, were 0.69 and 0.65, respectively. Compared to the EEC classification, the labelling of dangerous substances and for the prediction of severe irritants, sensitivity, specificity and equivalence were 94%, 89% and 78%, respectively. We observed nine false positives (22%) and two false negatives (2%) for the identification of R41-, R34- and R35-labelled products. The predictive values, for identifying R41, R34 and R35 products and non-irritants or R36 products were 78% and 97%, respectively. Repeatability (6.3) and reproducibility (8.9) were quite satisfactory.

The EYTEX system exhibits the characteristics of a good screening method: compatibility with a large range of chemicals; a simple and rapid procedure; good intralaboratory and interlaboratory reproducibility; cost effectiveness; high sensitivity, specificity and predictive value; and a low incidence of false negative and false positive results. Based on these results, we consider the EYTEX method to be a valuable tool for the screening of eye irritancy.

Contributions of Physicochemical Properties to the Evaluation of Ocular Irritation

International guidelines for the evaluation of ocular irritation advise that all the available information, structural and physicochemical, should be taken into account in order to minimise the testing of substances under conditions that are likely to produce severe reactions in laboratory animals.

In order to determine whether the pH and the acidic/alkaline reserve would be suitable parameters for the prediction of ocular irritation, we have measured, under standard conditions, the pH and the acidic/alkaline reserve of 166 chemicals (90 non-irritant, 22 irritant, and 54 severely irritant for the eye, according to the EEC classification and labelling system).

The combination of these parameters permitted the correct identification of 74% of the severely irritant chemicals and 97% of the non-severely irritant (non-irritant or irritant) chemicals. The predictive value for a severely irritant potential was 93%, and 88% for non-severely irritant chemicals.

This physicochemical approach, used as a first line screen for ocular irritation, is very simple and inexpensive.

Research and Development of In Vitro Pharmacotoxicology: A European Perspective

The experts taking part in the Workshop were: E. Rodríguez-Farré ( Coordinator); G.N. Fracchia, (Secretary); M. Adolphe, École des Hautes Études, Paris, France); P.H. Bach (University of East London, UK); M. Baeder (Hoechst Ltd, Hattersheira, Germany); R. Bass (BGA, Berlin, Germany); H.G. Baumgarten (Frei Universität, Berlin, Germany); H. Bazin (DGXII, CEC, Brussels, Belgium); P. Bentley (Ciba-Geigy, Basle, Switzerland); A. Boobis (Royal Postgraduate Medical School, London, UK); J. Castell (Hospital La Fé, Valencia, Spain); J.P. Contzen (DGXII, CEC, Brussels, Belgium); A. Cordier (Sandoz Pharma Ltd, Basle, Switzerland); J. Diezi (Université de Lausanne, Switzerland); L. Dubertret (INSERM U-312, Creteil, France); P.M. Fasella (DGXII, CEC, Brussels, Belgium); J.H. Fentem (FRAME, Nottingham, UK); A. Guillouzo (INSERM U-49, Rennes, France); I. Kimber (Zeneca, Macclesfield, UK); T. Krieg (Universität zu Koln, Germany); A. Mantovani (Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy); K. Miller (BIBRA, Carshalton, UK); J.P. Morin (INSERM U-295, Rouen, France); D. Paul (Fraunhofer Institut für Toxikologie und Aerosolforschung, Hannover, Germany); P.W.J. Peters (Riijkinstituut voor Volksgezondheid en Milieuhygiene, Bilthoven, The Netherlands); J. Picard (Faculté des Sciences, Louvain la Neuve, Belgium); D. Poggiolini (Ministry of Health, Rome, Italy); C.M. Regan (University College, Dublin, Ireland); C.A. Reinhardt (SIAT, Zurich, Switzerland); B. Robaire (McGill University, Montreal, Canada); M. Roberfroid (Université Catholique de Louvain, Brussels, Belgium); V. Rogiers (Vrije Universiteit Brussels, Belgium); J. Rueff (Istituto de Higiene e Medicina Tropical, Lisbon, Portugal); H. Spielmann (ZEBET, Berlin, Germany); H. Stolte (Medizinische Hochschule, Hannover, Germany); J. van Noordwijk (European Pharmacopeia Commission, Bosch en Duin, The Netherlands); E. Walum (University of Stockholm, Sweden); D.C. Williams (Trinity College, Dublin, Ireland); and M. Yaniv (Institut Pasteur, Paris, France), and their contributions are gratefully acknowledged.

Multivariate analysis of the first 10 MEIC chemicals

Nonlinear mapping coupled to powerful graphical tools was used to compare the toxicological responses of 32 in vivo and in vitro test systems to the first 10 MEIC chemicals. The obtained results clearly underline the usefulness of our methodological approach for the comparison of the different endpoints and the selection of a battery of in vitro toxicity tests allowing to estimate the possible harmful effects of chemicals in vivo.