You are here:
A FLEXIBLE APPROACH FOR EVALUATING FIXED RATIO MIXTURES OF FULL AND PARTIAL AGONISTS FOR MIXTURES OF MANY CHEMICALS.
Citation:
Gennings, C., W. H. Carter, R. A. Carchman, M J. DeVito, J E. Simmons, AND K M. Crofton. A FLEXIBLE APPROACH FOR EVALUATING FIXED RATIO MIXTURES OF FULL AND PARTIAL AGONISTS FOR MIXTURES OF MANY CHEMICALS. Presented at Society of Toxicology, New Orleans, LA, March 06 - 10, 2005.
Description:
Detecting interaction in chemical mixtures can be complicated by differences in the shapes of the dose-response curves of the individual components (e.g. mixtures of full and partial agonists with differing response maxima). We present an analysis scheme where flexible single chemical dose-response curves are combined in an 'additivity model' with prediction for the mixture based on a constraint of additivity. Iterative algorithms are used to estimate mean responses for this model at observed mixture combinations using only single chemical parameters. A 'full model' is also fit to the data with additional mixture parameters included. A likelihood-ratio test is used to test the hypothesis of additivity by comparing the full and additivity model predictions in the likelihood functions. When interaction is detected, an interaction threshold model may be estimated along the mixture ray. This model has two components: an implicit additivity region and an explicit part that describes the departure from additivity; the interaction threshold is the boundary between the two regions. The methods are illustrated with resulting data from a study of a mixture of 18 PHAHs in female Long Evans rats exposed by oral gavage for four consecutive days with serum thyroxine (T4) as the response variable. The mixing ratio was selected based on the ratios found in breast milk, fish and other sources of human exposure. The methods demonstrated three plateau levels for the maximum effects of the single chemicals and allowed for dose threshold effects for each single chemical and the fixed-ratio mixture. The likelihood-ratio test of additivity was rejected due to evidence of greater than additive interaction in the high dose region. Estimation of the interaction threshold within the observed experimental region suggested evidence of additivity in the low dose region. (This abstract does not reflect U.S. EPA policy.)