The ECVAM international validation study on in vitro embryotoxicity tests: Results of the definitive phase and Evaluation of prediction models

From 1996 to 2000, ZEBET (Centre for Documentation and Evaluation of Alternative Methods to Animal Experiments at the BgVV, Berlin, Germany) coordinated the European Centre for the Validation of Alternative Methods (ECVAM) prevalidation and validation study on three embryotoxicity tests: a) a test employing embryonic stem cell lines (EST); b) the micromass (MM) test; and c) the postimplantation rat whole-embryo culture assay (WEC test). The main objectives of the study were to assess the performance of these three in vitro tests in discriminating between non-embryotoxic, weakly embryotoxic and strongly embryotoxic compounds. Phase I of the study (1997) was designed as a prevalidation phase, for test protocol optimisation, and for the establishment of a comprehensive database of in vivo and in vitro data on embryotoxic compounds. Phase II (1998–2000) involved a formal validation trial, conducted under blind conditions on 20 test compounds selected from the database, which were coded and distributed to the participating laboratories. In the preliminary phase of the validation study, six chemicals out of the 20, which showed embryotoxic potential, were tested. These results were used to define new biostatistically based prediction models (PMs) for the MM and WEC tests, and to evaluate those developed previously for the EST. As a next step, the PMs were evaluated by using the results for the remaining 14 chemicals of the definitive phase of the validation study. The three in vitro embryotoxicity tests proved to be applicable to testing a diverse group of chemicals with different embryotoxic potentials (non-embryotoxic, weakly embryotoxic, and strongly embryotoxic). The reproducibility of the three in vitro embryotoxicity tests were acceptable according to the acceptance criteria defined by the Management Team. The concordances between the embryotoxic potentials derived from the in vitro data and from the in vivo data were good for the EST and the WEC (PM2) test, and sufficient for the MM test and the WEC (PM1) tests according to the performance criteria defined by the Management Team before the formal validation study. When applying the PM of the EST to the in vitro data obtained in the definitive phase of the formal validation study, chemicals were classified correctly in 78% of the experiments. For the MM and the WEC tests, the PMs provided 70% and 80% (PM2) correct classifications, respectively. And, very importantly, an excellent predictivity (100%, except