The Validation of Alternative Test Methods

As non-animal toxicity tests began to be developed in the late 1980s, the need for their validation, i. e. the independent assessment of their relevance and reliability for particular purposes, was recognised, and international discussions on the principles of validation and their practical application took place, notably, in Europe and the USA. This came to involve recognition of a prevalidation stage to ensure that a method was ready for formal validation and of a prediction model, an unambiguous algorithm for converting test data into a prediction of a relevant in vivo pharmacotoxicological endpoint. By the mid-1990s, validated non-animal methods could be proposed for the development of test guidelines and for regulatory acceptance. Meanwhile, the validation process itself evolved as a result of practical experience, and processes such as catch-up validation, weight-of-evidence validation, QSAR model validation and integrated testing strategies came to be seen as important. Major roles in these developments were played by ECVAM in Europe, ICCVAM in the USA, JaCVAM in Japan, and the OECD, but validation centres now operate in a number of other countries.