ATLA::Alternatives to Laboratory Animals

Volume 25, Number 5

The validation of toxicological prediction models.

ATLA 25, 505-516, September/October 1997

Graeme Archer,¹ Michael Balls,¹ Leon H. Bruner,² Rodger D. Curren,³ Julia H. Fentem,¹ Hermann-Georg Holzhütter,⁴ Manfred Liebsch,⁵ David P. Lovell and Jacqueline A. Southee⁷

¹ECVAM, JRC Environment Institute, 21020 Ispra (VA), Italy; ²The Procter & Gamble Company, Health and Beauty Care Europe, Egham, Surrey TW20 9NW, UK;³Institute for In Vitro Sciences Inc., Suite 220, 21 Firstfield Road, Gaithersburg, MD 20878, USA; ⁴Humboldt-Universität zu Berlin, Bereich Medizin (Charité), Insitute für gesundheitlichen Verbraucherschutz und Veterinärmedizin (BgVV), Diedersdorfer Weg 1, 12277 Berlin, Verbraucherschutz und Veterinärmedizin (BgVV), Diedersdorfer Weg 1, 12277 Berlin, Germany; ⁶BIBRA International, Woodmansterne Road, Carshalton, Surrey SM5 4DS, UK; ⁷Microbiological Associates Ltd, Stirling University Innovation Park, Stirling FK9 4NF, UK

SUMMARY

An alternative method is shown to consist of two parts: the test system itself; and a prediction model for converting in vitro endpoints into predictions of in vivo toxicity. For the alternative method to be relevant and reliable, it is important that its prediction model component is of high predictive power and is sufficiently robust against sources of data variability. In other words, the prediction model must be subjected to criticism, leading successful models to the state of confirmation. It is shown that there are certain circumstances in which a new prediction model may be introduced without the necessity to generate new test system data.

Keywords: alternative method, model criticism, prediction model, validation