Determination of the starting dose for acute oral toxicity (LD50) testing in the up and down procedure (UDP) from cytotoxicity data

Citation
H. Spielmann et al., Determination of the starting dose for acute oral toxicity (LD50) testing in the up and down procedure (UDP) from cytotoxicity data, ATLA-ALT L, 27(6), 1999, pp. 957-966
Citations number
21
Categorie Soggetti
Animal & Plant Sciences
Journal title
ATLA-ALTERNATIVES TO LABORATORY ANIMALS
ISSN journal
02611929 → ACNP
Volume
27
Issue
6
Year of publication
1999
Pages
957 - 966
Database
ISI
SICI code
0261-1929(199911/12)27:6<957:DOTSDF>2.0.ZU;2-W
Abstract
To reduce the number of animals used in acute oral toxicity testing, cytoto xicity data (IC50) can be used to determine the starting dose for in, vivo testing by applying the standard regression between IC50 and acute oral LD5 0 values in the Register of Cytotoxicity (RC). In the RC, the correlation b etween cytotoxicity, represented by the mean IC50 (IC50x), and the acute or al LD50 of rats and/or mice has been determined for 347 chemicals by applyi ng the linear regression model for log-transformed pairs of IC50 versus ora l LD50. The standard regression line of the two toxicity parameters is char acterised by an intercept a = 0.625 and regression coefficient b = 0.435, a nd 252 of 347 chemicals (72.6%) are located within a dose-range differing b y not more than 0.699 (factor F-G less than or equal to log 5) from the sta ndard regression line. In the present study, we have used the RC and its IC 50/LD50 regression model to predict the LD50 values from cytotoxicity data for nine chemicals which were tested in an evaluation study of the Up and D own Procedure (UDP). For seven of the nine chemicals, LD50 values (mg/kg) p redicted from the RC were in the same dose-range as LD50 values determined in vivo, while the dose-range differed by more than one order of magnitude for the two remaining chemicals. Thus, the prediction of LD50 values from c ytotoxicity data was promising in this limited data set. It is proposed tha t a tiered in vitro/in vivo testing approach will reduce animal use in the UDP method. As the first step, the in vitro cytotoxicity of a new chemical is determined. By applying the RC regression and adapting it to the sensiti vity of a specific cell line, the LD50 value (mg/kg) can be predicted from the IC50 value. The predicted LD50 dose is then used as the starting dose i n the UDP. In the RC model, the precision of the prediction increases with decreasing toxic potential, and the majority of industrial chemicals (aroun d 90%) are not toxic according to EU classification criteria.