Selinski, Silvia Quinke, Barbara Golka, Klaus Bolt,Hermann M. Urfer, Wolfgang
Technical Report, SFB 475: Komplexitätsreduktion in Multivariaten Datenstrukturen, Universität Dortmund 2002,45
The determination of toxicokinetic parameters is an essential component in the risk assessment of potential harmful chemicals. It is a key step to analyse the processes involved in the formation of DNA adducts which are connected with the development of chemical-induced cancer. A general problem is the extrapolation of toxicological data from experimental animals to the human organism. Therefore a valid characterisation of the relevant processes for the whole species is required, i.e., of population mean parameters instead of sets of parameters for different individuals. These, again, may vary between repeated experiments at the same or at different administered doses. Nevertheless, these differences are of great importance in obtaining a more precise insight into the variability structure of process investigated within the test animal population, so that a valid basis for further research is the final result. The theory of hierarchical models, particularly the work of Racine-Poon (1985) and Racine-Poon and Smith (1990), provides a procedure which incorporates both, modelling of the variability structure and estimation of population mean parameter vectors. The present study was designed to elucidate interindividual and interoccasion variability of toxicokinetic parameters relevant for the biological transformation of one of the basic petrochemical industrial compounds, ethylene 2 (ethene), which is also a physiological body constituent, to its metabolite, ethylene oxide, which is a proven carcinogen.
Ethylene ethylene oxide toxicokinetics population model repeated measurements EM algorithm interindividual variability interoccasion variability