|
(437) Using a biology-based model (DEBtox) to analyse toxicity tests : interest and recommandations. Pery, Alexandre1, Flammarion, Patrick1, Bedaux, Jacques2, Kooijman, Bas2, Garric, Jeanne1, 1 2 ABSTRACT- The conventional statistical analysis of bioassays results often neglects biological significance. However, mathematical equations with few parameters but realistic from a biological point of view do exist to describe toxicokinetics and effects on test organisms of chemical compounds. Here we studied the relevance of a biology-based model to analyse bioassays. This model, proposed by Bas Kooijman and Jacques Bedaux in 1996, is called DEBtox. It provides an estimate of a No Effect Concentration (NEC), as an alternative to NOEC. We showed that the basic model can be successfully extended to incorporate problems usually faced during bioassays like time-dependence of the parameters estimates, time-varying exposure concentrations, unsuited choice of initial concentrations or influence of food. Computer simulations as well as experimental works with survival toxicity tests were performed to confirm the relevance of the extensions. These works also proved the capacity of the model to predict the effects of discontinuous pollution events. We concluded that DEBtox model can largely contribute to improve the use of bioassays in ecological risk assessment. Finally we gave some recommendations to optimize the use of the model and proposed an experimental test to check its capacity to describe the effects of a given compound on a test organism. Key words: No Effect Concentration, Model, Bioassay, Risk assessment |
