meic

The MEIC (Multicentre Evaluation of In Vitro Cytotoxicity) study is an international programme to evaluate the relevance and reliability of in vitro tests for predicting acute systemic toxicity. The study was initiated by Dr. Björn Ekwall in 1989 and have been organised by a committee elected by the Scandinavian Society for Cell Toxicology. During 1989-96, 97 laboratories world-wide have voluntarily participated to test the same reference chemicals in their own in vitro systems.The reference chemicals were selected by experts at the Swedish Poison Information Centre at the beginning of the project and represent different classes of chemicals with varied human toxicity. They all have reasonably good human toxicity data and LD50-values for both rats and mice. In other respects they were randomly selected.

The submitted in vitro results have centrally been evaluated by a systematic comparison with appropriate human acute toxicity data (toxic and lethal blood and tissue concentrations, toxicokinetics, etc.). In parallel, the animal LD50-values have been compared with analogous human toxicity, i.e. human lethal dosage, to be able to compare prediction of the standard toxicity tests and the in vitro toxicity tests. The ultimate goals of this evaluation have been to evaluate the relevance for human acute systemic toxicity of in vitro cytotoxicity tests, and to try to select the best combination of tests (battery) for future in vitro acute toxicity testing.

At the closure of the project, in January 1996, 29 laboratories had tested all 50 reference chemicals in 61 cytotoxicity assays. The results obtained from the in vitro-in vivo evaluations have been published in a series of eight articles in ATLA.

Part I of this series presents the methodology of 68 in vitro toxicity assays used to test the first 30 chemicals. (MEIC Evaluation of Acute Systemic Toxicity: Part I: C. Clemedson et al, 1996, ATLA 24, 249-272).

Part II presents the results from the 68 methods together with an in vitro-in vitro comparison of the obtained results. (MEIC Evaluation of Acute Systemic Toxicity: Part II: C. Clemedson et al, 1996, ATLA 24, 273-311).

Part III presents results of 16 additional in vitro methods used to test the first 30 reference chemicals together with a new comparative analysis of the accumulated results from 84 methods. (MEIC Evaluation of Acute Systemic Toxicity: Part III: C. Clemedson et al, 1998, ATLA 26, 93-129).

Part IV presents results from tests of reference chemicals 31-50 in 67 in vitro assays together with an in vitro-in vitro comparison of these results. (MEIC Evaluation of Acute Systemic Toxicity: Part IV: C. Clemedson et al, 1998, ATLA 26, 131-183).

Part V presents the human and animal toxicity databases which are used as the "gold standard" for the in vitro-in vivo comparisons in the final evaluation. (MEIC Evaluation of Acute Systemic Toxicity: Part V: B. Ekwall et al, 1998, ATLA, 26, 571-616).

Part VI presents the first steps of the final in vitro - in vivo evaluation (i) a comparison of rat and mouse oral LD50 with human acute lethal doses for the all 50 reference chemicals, (ii) a display of the correlation between IC50-values from all assays that have been used to test all 50 chemicals and three independent sets of human acutely lethal blood concentration, and (iii) a series comparisons between average IC50-values from ten human cell line 24h assays and human lethal blood concentrations. (MEIC Evaluation of Acute Systemic Toxicity: Part VI: B. Ekwall et al, 1998, ATLA 26, 617-658).

Part VII presents the prediction of human toxicity by results from testing of the first 30 reference chemicals with 27 further in vitro essays. (MEIC Evaluation of Acute Systemic Toxicity: Part VII, C. Clemedson et al, 2000, ATLA 28, 161-200).

PART VIII presents a multivariate partial least squares evaluation, including the selection of battery of cell line tests with good prediction of human acute lethal peak blood concentrations for the MEIC reference chemicals (MEIC Evaluation of Acute Systemic Toxicity: Part VIII, B. Ekwall et al, 2000, ATLA 28, 201-234).

The eight articles can be ordered free of charge from NICA. The results of the MEIC project showed that an average IC50 of all ten 24 hour exposure tests with human cell lines predicted human peak concentrations from LC50 curves better (R2 = 0.74) than the prediction of human lethal doses by LD50 for rats and mice (R2 = 0.60-0.66), for the 50 reference chemicals. When some known human toxicokinetic data (knowledge of the passage across the blood-brain barrier and the timing of the lethal action) was used together with the cytotoxic concentration to predict human lethal concentrations, prediction increased considerably. The cell line assay only underestimated the human toxicity for seven chemicals, i.e. paracetamol, digoxin, malathion, nicotine, cyanide, paraquat, and atropine, all known to lead to lethality by actions on specific receptors or cells. As a practical result of MEIC, a battery of three 24h human cell line assays could be selected with use of multivariate PLS-analysis which had a good direct prediction (R2= 0.77) of human lethal peak concentration. The prediction increased markedly (R2=0.83) when a simple algorithm based on knowledge of blood-brain barrier passage was used to adapt in vitro to in vivo concentrations.

The MEIC results demonstrate a high relevance of human cell tests (basal cytotoxicity) for human acute toxicity of chemicals, but at the same time show that (A) other important toxic mechanisms exist, which only might be measured by supplementary in vitro toxicity tests, and (B) modelling of human toxicity is improved by additional toxicokinetic data which probably can be obtained by new in vitro kinetic tests. These results encouraged Dr. Björn Ekwall to initiate the EDIT-project (Evaluation-guided development of New In vitro Tests) in 1998.