Evaluating Chemical Interaction Studies for Mixture Risk Assessment

Abstract

We describe a set of criteria to evaluate the quality of data and interpretations in chemical interaction studies. These criteria reflect the consensus of the literature on interaction analysis developed over decades of research in pharmacology, toxicology, and biometry; address common pitfalls in published interaction studies; and can be easily applied to common methods of interaction analysis. The criteria apply broadly to interaction data for drugs, pesticides, industrial chemicals, food additives, and natural products and are intended to assist risk assessors who must evaluate interaction studies for use in component-based mixture risk assessments. The criteria may also assist researchers interested in conducting interaction studies to inform mixture risk assessment. The criteria are also intended to serve larger scientific goals, including increasing the repeatability of results obtained in chemical interaction studies, enhancing the reliability of conclusions drawn from interaction data, providing greater consistency of interpretations among various analysts, and decreasing uncertainty in using interaction data in risk assessments. We describe the basis for each criterion and demonstrate their utility by using them to evaluate interaction studies from the recent toxicological and pharmacological literature, which serve as examples of different types of data sets that the risk assessor may encounter.

A critical review of methods for comparing estrogenic activity of endogenous and exogenous chemicals in human milk and infant formula.

“…Generating dose-response data for a whole mixture presents additional difficulties, because concentrating or diluting a mixture to produce a range of concentrations changes the dose-equivalent ratios of mixture components. Because mixture effects vary with both concentrations and ratios of the constituents (Berenbaum 1989;Borgert et al 2001;Greco et al 1995), an elaborate toxicologic assessment of individual whole mixtures is usually not practically or technically feasible.…”

The Toxicology of Chemical Mixtures Risk Assessment for Human and Ecological Receptors

“…Published binary studies sometimes fail to perform statistics, omit statistical procedures from the paper, do not define "no interaction" or end up comparing binary toxicity to individual component toxicity. Without predicting what dose additive toxicity should have been, it is impossible to determine if interactions took place (Borgert et al, 2001;Simmons, 1995).…”

“…Multiple assays should be involved. In vitro studies can detect pharmacodynamic interactions but often in vivo studies are necessary to identify kinetic interactions (Borgert et al, 2001;Teuschler et al, 2002). The route of exposure should be environmentally relevant.…”

“…The route of exposure should be environmentally relevant. "No interaction" should be defined as either dose addition or response addition and results compared using appropriate statistical tests to determine significant biological variation from the expected results (Borgert et a., 2001;Hertzberg and MacDonell, 2002). Doses need to be selected carefully.…”

“…Doses need to be selected carefully. They should reflect environmental exposure concentrations, include the individual component NOAELs and LOAELs, and also be spaced to develop a dose-response curve (Borgert et al, 2001;Cassee et al, 1998;Hertzberg and MacDonell, 2002;Teuschler et al, 2002).…”