The MEIC (multicentre evaluation of in vitro cytotoxicity) programme is a five-year programme to validate in vitro tests for general toxicity, and is organised by the Scandinavian Society for Cell Toxicology. Interested laboratories are invited, on an international basis, to test 50 published reference chemicals in their respective assays. Submitted results will then be evaluated yearly by the MEIC Committee for their relevance to various types of human toxicity, including an evaluation for the same chemicals of the prediction by animal tests of human toxicity. To establish the validation methods, a preliminary validation cycle is being performed in 1989/90 with submitted results for the first ten MEIC chemicals. The present paper is the very first step of this preliminary validation process. The prediction of human toxicity by five cytotoxicity assays (altogether 14 different cell systems/endpoints) has been evaluated, and also compared with the predictive value of rodent LD50 tests. Mouse LD50 prediction of human lethal dosage for these substances was good, while rat LD50 prediction was less satisfactory. The collective predictions by all 14 cell systems/endpoints of human toxicity in the form of a multivariate PLS (partial least squares) model of human acute lethal blood concentrations, as well as the corresponding prediction by a HeLa cell assay, were comparable to the efficiency of mouse LD50 prediction of human lethal dosage. When combined with simple toxicokinetic data (absorption of chemicals in the intestine and distribution volumes), the PLS model and the HeLa assay were able to predict human lethal dosage of the ten chemicals as accurately as the mouse LD50 value. The small number of chemicals studied to date means that general conclusions cannot be drawn from these results. Further validation of more chemicals with the in vitro methods is essential and promises to be worthwhile.
Metabolic characterization of cell systems used in in vitro toxicology testing: Lung cell system BEAS-2B as a working example
