A comparison of statistical approaches to the derivation of EC3 values from local lymph node assay dose responses

Effective risk assessment and management of allergic contact dermatitis req uire three key factors: adequate hazard identification, measurement of the relative potency of identified hazards and an understanding of the nature, extent and duration of exposure. Suitable methods for hazard identification , such as the murine local lymph node assay (LLNA) and the guinea-pig maxim ization test, are well established and conditions of human exposure normall y can be well anticipated. Thus, the need is for a robust and quantitative method for the estimation of relative skin sensitizing potency, One possibl e approach is via the analysis of LLNA dose-response data, In the LLNA, con tact allergens are defined currently as those chemicals that cause a threef old or greater increase in lymph node cell proliferative activity compared with concurrent vehicle-treated controls. It is possible to estimate the co ncentration of a sensitizer required to generate a threefold stimulation of proliferation in draining lymph nodes; such a concentration is known as th e EC3 value. Using a variety of statistical approaches to derive EC3 values from LLNA dose-response data for 10 chemicals, it has been demonstrated th at simple linear interpolation between the values either side of the threef old stimulation index provides a robust assessment of the EC3 value without the need for recourse to more sophisticated statistical techniques. Provid ed that the appropriate concentrations of test chemical have been selected, EC3 values obtained in this way are reproducible both within and between l aboratories and form the basis for examination of the utility of this appro ach for the estimation of relative skin sensitizing potency.